ESMO Guidelines

clinical practice guidelines Annals of Oncology 25 (Supplement 3): iii10–iii20, 2014
doi:10.1093/annonc/mdu159
Published online 6 July 2014
Anal cancer: ESMO-ESSO-ESTRO Clinical Practice

Guidelines for diagnosis, treatment and follow-up†
R. Glynne-Jones1, P. J. Nilsson2, C. Aschele3, V. Goh4, D. Peiffert5, A. Cervantes6 & D. Arnold7*
1
Centre for Cancer Treatment, Mount Vernon Hospital, Northwood, Middlesex, UK; 2Department of Molecular Medicine and Surgery, Karolinska Instituet and Center for
Surgical Gastroenterology, Karolinska University Hospital, Stockholm, Sweden; 3Medical Oncology and Hematology, Felettino Hospital, La Spezia, Italy; 4Division of
Imaging Sciences and Biomedical Engineering, King’s College London, London, UK; 5Department of Radiotherapy, Centre Alexis Vautrin, Vandoeuvre-lès-Nancy, France;
6
Department of Hematology and Medical Oncology, INCLIVA, University of Valencia, Valencia, Spain; 7Klinik für Tumorbiologie, Freiburg, Germany
aetiology incidence and epidemiology

Squamous cell carcinoma of the anus (SCCA) is strongly asso- Epidermoid anal cancer is a rare disease accounting for 1%–2%
ciated with human papillomavirus (HPV) infection which of digestive tract tumours and 2%–4% of colon, rectal and anal
represents the causative agent in 80%–85% of patients (usually tumours. The annual incidence is 1 in 100 000, is higher in
from HPV16 or HPV18 subtypes in Europe) as is its precursor women and is increasing [1]. In Europe, ∼2000 males and
lesion anal intra-epithelial neoplasia (AIN) 3. Factors increasing ∼2300 females are diagnosed with anal cancer every year. Five-
the risk of HPV infection and/or modulating host response and year survival has changed little in the last two decades. In the
the persistence of this infection appear to affect the epidemi- United States, the overall 5-year survival rates for 1994–2000
clinical practice
ology of this tumour. Anal intercourse and a high lifetime were 60% for men and 78% for women (SEER data). In Europe,
guidelines
number of sexual partners increase the risk of persistent HPV 5-year survival varied between 66% (Central Europe) and 44%
infection in men and women, leading eventually to malignancy. (Eastern Europe).
Other important risk factors include human immunodeficiency
virus (HIV), immune suppression in transplant recipients, use
of immunosuppressants such as long-term corticosteroids, a diagnosis
history of other HPV-related cancers, autoimmune disorders, SCCA commonly presents with bleeding; hence, diagnosis is
social deprivation and cigarette smoking. Cigarette smoking often delayed because bleeding is attributed to haemorrhoids.
may also be important in the modulation/persistence of HPV SCCA may also present with any combination of a mass, non-
infection and, hence, outcomes from treatment. Herpes simplex healing ulcer, pain, bleeding, itching, discharge, faecal incontin-
virus may play a secondary role in disease progression. Dietary ence and fistulae. Not uncommonly lesions are palpated first by
habits, chronic inflammatory diseases and the presence of hae- the patient. The diagnosis of anal cancer is made on biopsy-
morrhoids do not appear to predispose to epidermoid anal proven histology. Small, early cancers are sometimes diagnosed
cancer. Among men who have sex with men (MSM), the inci- serendipitously following the removal of anal tags. More advanced
dence of anal cancer is in the region of 35 per 100 000. In men lesions in the distal anal canal may extend on to the skin at the
who are HIV seropositive, the incidence increases to 75–135 per anal margin. Rarely patients present with inguinal lymphaden-
100 000. The incidence is also higher among HIV seropositive opathy.
women. Prolonged survival with highly active anti-retroviral
treatment (HAART) is likely to lead to further increase in inci-
dence among HIV-positive subjects. screening and prevention
The existence of an identified viral aetiological agent and the
ability to detect pre-neoplastic lesions may allow the develop-
ment of screening and prevention programmes. Vaccination of
*Correspondence to: ESMO Guidelines Working Group, ESMO Head Office, Via L. Taddei girls against oncogenic HPV is now being recommended for the
4, CH-6962 Viganello-Lugano, Switzerland. prevention of cervical cancer, and a recent report indicated that
E-mail: clinicalguidelines@esmo.org
up to 80% of anal cancers could also be avoided with prophylac-
†
These Guidelines were developed by the European Society for Medical Oncology tic quadrivalent HPV vaccine (against HPV types 6, 11, 16 and
(ESMO), the European Society of Surgical Oncology (ESSO) and the European Society of 18). But currently vaccination has no role when SCCA is actually
Radiotherapy and Oncology (ESTRO) and are published jointly in the Annals of Oncology,
present [2].
the European Journal of Surgical Oncology and Radiotherapy & Oncology. The three so-
cieties nominated authors to write the guidelines as well as reviewers to comment on Screening programmes using anal cytology and high-resolution
them. anoscopy have been proposed for high-risk populations (MSM
© The Author 2014. Published by Oxford University Press on behalf of the European Society for Medical Oncology.
All rights reserved. For permissions, please email: journals.permissions@oup.com.
Downloaded from https://academic.oup.com/annonc/article-abstract/25/suppl_3/iii10/1740790

by guest
on 05 February 2018
Annals of Oncology clinical practice guidelines
and HIV– women with a history of anal intercourse or other stromal tumours, poorly differentiated neuroendocrine tumours
HPV-related anogenital malignancies) based on the achievements and lymphoma.
obtained in cervical cytology screening. However, no randomised Tumours of the anal margin are generally well differentiated
control study has yet demonstrated the advantage of screening in and often occur in men, in contrast to canal tumours which are
these high-risk populations. normally poorly differentiated and more common in women.
Histological grading is subject to inter-observer variability, and
considerable heterogeneity is seen in larger tumours. High-grade
pathology/molecular biology tumours have been thought to have a worse prognosis, but this
has not been confirmed in multivariate analysis. Histological sub-
anal intra-epithelial neoplasia classifications of basaloid, transitional, spheroidal and cloacogenic
Anal cancer may arise from a precursor dysplastic lesion—AIN cell cancers have no additional confirmed bearing on manage-
—also known as anal squamous intra-epithelial lesions. The ment. Some authors report that a basaloid histological subtype
prevalence of AIN in the general population is low, but higher has a higher risk of developing metastatic disease. The biology
in 30%–40% of MSMs. Progression from AIN 1 and AIN 2 to and prognosis of keratinising and non-keratinising tumours of
AIN 3 is uncommon, as is progression from AIN 3 to invasive the anal canal also appear to be similar. Verrucous carcinomas
malignancy in immunocompetent patients, but appears more are a variant and are sometimes described as giant condylomas or
likely in immunosuppressed patients, and is influenced by HIV Buschke–Lowenstein tumours, which may have a better prognosis
seropositivity, low CD4 count and serotype of HPV infection. than SCCA, for whom some consider surgery the best option.
HPV-associated tumours usually retain wild-type P53, and this
explains why patients with HPV-associated tumours respond well
to concurrent chemoradiotherapy. Data on the interaction with
anatomy and lymphatic drainage
wild-type P53 in current or former cigarette smokers, as in head Lymph node involvement at diagnosis is observed in 30%–40%
and neck cancer, are lacking. of cases while systemic spread is uncommon with distant extra-
Both synchronous and metachronous HPV-related vaginal pelvic metastases recorded in 5%–8% at onset.
and cervical intra-epithelial and malignant squamous lesions The anal canal extends from the anorectal junction to the
are frequent and should be screened for in younger women. anal margin (see Figure 1).
Useful palpable landmarks are the puborectal sling and the
inter-sphincteric groove, respectively. The columnar, or cylind-
histology ric, epithelium of the rectum extends to about 1 cm above the
Histological confirmation is mandatory as other histologies are dentate line where the anal transitional zone begins. Below the
possible including adenocarcinoma, melanoma, gastrointestinal dentate line, the epithelium is all squamous. The anal margin is
Levator ani m.
Puborectalis m.
External sphincter
Dentate line
Anal canal
Internal sphincter
Anal margin
Figure 1. Anatomy of the anal region.
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu159 | iii

by guest
on 05 February 2018
clinical practice guidelines Annals of Oncology
the pigmented skin immediately surrounding the anal orifice, Positron emission tomography (PET)/CT with [18F]fluoro-
extending laterally to a radius of ∼5 cm. In practice, at diagnosis, deoxyglucose (FDG–PET/CT) has a high sensitivity in identify-
the precise point of origin is often uncertain, and the distinction ing involved lymph nodes, as the majority of anal carcinomas
between anal canal and anal margin tumour is often difficult, if are FDG-avid. Several studies have shown that FDG–PET/CT
not impossible. Hence, some have classified into three distinct can alter staging in ∼20% of cases, with a trend towards up-
regions—i.e. intra-anal, perianal (visualised with gentle traction staging, and can alter treatment intent in ∼3%–5% of cases. The
on the buttocks) and skin tumours (beyond a 5 cm radius from main impact of FDG–PET/CT on therapy stems from its high
the anal opening). sensitivity in identifying involved lymph nodes and influencing
Proximally lymphatic drainage is to perirectal nodes along the radiation therapy planning by defining sites of metabolically
inferior mesenteric artery. Immediately above the dentate line, active tumour. Therefore, FDG–PET has been recommended in
drainage is to internal pudendal nodes, and to the internal iliac the current USA National Comprehensive Cancer Network
system. Infra-dentate and perianal skin drains to the inguinal, treatment recommendations.
femoral and external iliac nodes. The tumour–node–metastasis (TNM) clinical staging system
is based on accurate assessment of size (T stage), regional lymph
node involvement (N) and metastatic spread (M). Nodal status
clinical assessment is based on the distance from the primary site rather than the
A relevant history is required to elicit symptoms, other relevant number of nodes involved, see Table 1. Nodal involvement of
medical conditions, current medications and predisposing anal canal lesions differs from that of anal margin tumours.
factors, which should be documented. Examination should Biopsy by needle aspiration biopsy is usually only carried out
include digital rectal examination (DRE) to examine the anal for clinically palpable inguinal nodes or those enlarged >10 mm
lesion and perirectal nodal involvement and, in women ( par- on CT or MRI. Sentinel lymph node biopsy can reveal microme-
ticularly with low anteriorly placed tumours), a vaginal examin- tastatic spread of disease, and may be more accurate than diag-
ation to determine the site and size of the primary tumour, nostic imaging, but has not been properly evaluated.
vaginal/vaginal septal involvement, mucosal involvement and Squamous cell carcinoma antigen (SCCAg) is a serum tumour
exophytic or ulcerative tumour or the presence of a fistula marker expressed by carcinomas of the anal canal, and may be
(Table 3). Vaginal involvement may require the prophylatic related to tumour stage and/or nodal status, but its clinical utility
siting of a de-functioning stoma because of the risk of an ano- in diagnosis and follow-up remains controversial.
rectal-vaginal fistula. However, since only 50% of initial colos-
tomies are reversed, this decision should be weighed carefully.
Palpation of the inguinal nodes is important, particularly those
risk assessment
superficial inguinal nodes, medial and close to the pubis. Anal cancers occur rarely, and factors influencing outcome and
Proctoscopy by a specialist surgeon or radiation oncologist long-term survival have proved difficult to study with multivari-
and, if painful, examination under anaesthesia (EUA) may be ate analysis. Several factors are relevant to initial decision-
appropriate to facilitate biopsy (Table 3). It is also easier to de- making (Table 2). The European Organisation for Research and
termine anatomical relations to surrounding structures and to Treatment of Cancer (EORTC)-22861 study demonstrated that
allow accurate clinical staging. It is advantageous if the treating skin ulceration, nodal involvement and male sex were independ-
radiation oncologist is present during this EUA to document ent variables associated with locoregional failure rate (LRF) and
precise measurements, as these are often critical for later target overall survival (OS) [5]. The Radiation Therapy Oncology
volume delineation in treatment planning. Group (RTOG) 9811 analysis supported the EORTC previously
Colonoscopy is not required to assess pathology in the prox- reported factors (clinically involved nodes and male sex), and
imal bowel, because synchronous lesions are not reported for also established tumour diameter >5 cm as an independent vari-
SCCA. able predicting disease-free survival (DFS) and OS. The ACT I
analysis showed that palpable inguinal nodal status and gender
are independently prognostic for OS, LRF and anal cancer death
staging (ACD). In addition, after adjusting for sex and nodal status, pre-
With an indolent natural history and a low rate of distant metas- senting haemoglobin was a further prognostic factor for ACD.
tases, anal cancer is usually amenable to locoregional treatment. HIV testing is recommended in any patient with a lifestyle that
Imaging should include magnetic resonance imaging (MRI) puts them at risk of contracting HIV infections. Some argue that
(Table 3) of the pelvis or, if not available, endo-anal ultrasound all patients with anal cancer should be screened for HIV [6].
(EUS). Distant metastases can be assessed with computed tom- Before the widespread use of HAART, HIV-positive patients
ography (CT) of the thorax and abdomen. MRI provides excel- were considered to have enhanced toxicity from chemoradiation
lent contrast and spatial resolution, providing information on (CRT), particularly in patients with low CD4 counts <200/mm3
tumour size, local extent and spread, and invasion of adjacent which may impact on compliance [7]. Such patients were
organs and more accurate nodal involvement [3]. excluded from the randomised trials. More recent evidence sug-
Accurate assessment of tumour size and depth of mural inva- gests similar outcomes in HIV-positive patients treated with
sion is possible with EUS, due to excellent spatial detail but is HAART in terms of complete response and survival to HIV-
best reserved for small T1 lesions as the small field-of-view negative patients [8, 9]. However, we have no randomised data
limits assessment of regional lymph nodes and infiltration of to guide best practice in immuno-compromised and HIV-posi-
structures beyond the anal canal. tive patients.
iii | Glynne-Jones et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
Table 1. TNM staging. American Joint Committee on Cancer/ Table 2. Factors to consider in treatment decision-making for anal
Union for International Cancer Control (AJCC/UICC) seventh cancer
edition TNM clinical and pathological classification of anal cancer
Disease-related Patient-related factors Other
Primary tumour (T) factors
TX Primary tumour cannot be assessed. Clinical and Patient preferences Local expertise
T0 No evidence of primary tumour. radiological TNM (brachytherapy
Tis Carcinoma in situ (i.e. Bowen disease, high- stage etc.)
grade squamous intra-epithelial lesion, and
Site of tumour Biological age/renal Geriatricians with
anal intra-epithelial neoplasia II–III.)
(margin, canal, function/Charlson interest in
T1 Tumour ≤2 cm in greatest dimension.
rectal) geriatric oncology
T2 Tumour >2 cm but ≤5 cm in greatest
assessment
dimension.
T3 Tumour >5 cm in greatest dimension. Extent of tumour, i.e. Co-morbidities/
T4 Tumour of any size invades adjacent organ(s), involvement of current
e.g. vagina, urethra, and bladder. vagina (risk of medications and
Regional lymph nodes (N) fistulation) in performance status
NX Regional lymph nodes cannot be assessed. addition to size
N0 No regional lymph node metastasis. Response to Socioeconomic and
N1 Metastases in perirectal lymph node(s). treatment (early psychological
N2 Metastases in unilateral internal iliac and/or and at 26 weeks) factors/social
inguinal lymph node(s). support
N3 Metastases in perirectal and inguinal lymph
nodes and/or bilateral internal iliac and/or Need for symptom Severity of initial Specialist palliative
inguinal lymph nodes. control symptoms care
Distant metastasis (M)
TNM, tumour–node–metastasis.
M0 No distant metastasis.
M1 Distant metastasis.
Anatomic stage/prognostic groups
Stage T N M
0 Tis N0 M0
possible quality of life. Treatment dramatically differs from
I T1 N0 M0 adenocarcinomas of the lower rectum.
II T2 N0 M0 Combinations of 5-fluorouracil (5-FU)-based CRT and other
T3 N0 M0 cytotoxic agents [mainly mitomycin C (MMC)] have been
IIIA T1 N1 M0 established as the standard of care, leading to complete tumour
T2 N1 M0 regression in 80%–90% of patients, with locoregional failures of
T3 N1 M0 ∼15%. A multidisciplinary approach is mandatory, involving ra-
T4 N0 M0 diation therapists, medical oncologists, surgeons, radiologists
IIIB T4 N1 M0 and pathologists. The role of surgery as a salvage treatment is
Any T N2 M0 accepted. Assessment and treatment should be carried out in
Any T N3 M0 specialised centres treating a high number of patients as early as
IV Any T Any N M1 possible in the clinical diagnosis (Table 3). To date, the limited
evidence from only six randomised trials [5, 10–14], the rarity of
In ref. [4], used with the permission of the American Joint the cancer and the different behaviour/natural history depend-
Committee on Cancer (AJCC), Chicago, IL, USA. The original ing on the predominant site of origin (the anal margin, anal
source for this material is the AJCC Cancer Staging Handbook, canal or above the dentate line) provide limited direction for any
Seventh Edition (2010) published by Springer Science and Business
individual oncologist (Table 4). An example of treatment of
Media LLC, www.springer.com.
anal cancer is shown in Figure 2.
Every effort should be made to ensure patients stop smoking surgery as primary treatment
before therapy, because smoking may worsen acute toxicity Until the mid-1980s, radical surgery was the cornerstone of
during treatment and enhance late toxicity. treatment. However, following publications from the 1970s on
combined modality therapy, surgery as the primary therapeutic
initial management of local option has generally been abandoned.
Still today, smaller lesions (<2 cm in diameter), involving the
and locoregional disease anal margin and not poorly differentiated may be treated by
The primary aim of treatment is to achieve cure with locoregio- primary surgery in the form of a local excision provided ad-
nal control and preservation of anal function, with the best equate margins (>5 mm) can be obtained without
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu159 | iii

by guest
on 05 February 2018
Table 3. Diagnostic work-up
Mandatory Optional but often recommended Optional
Biopsy to confirm diagnosis

Full medical history HIV test
Full clinical body exam Needle aspiration groin nodes
Digital rectal examination Examination under anaesthesia
Blood tests including renal function
Proctoscopy Colonoscopy
Pelvic MRI Endo-anal ultrasound
CT thorax/abdomen PET/CT
Assessment of genital tract in females for CIN/VIN Assessment by geriatrician
CIN, cervical intra-epithelial neoplasia; VIN, vulval intra-epithelial neoplasia; MRI, magnetic resonance imaging; HIV, human immunodeficiency
virus; CT, computed tomography; PET, positron emission tomography.
Table 4. Stage and site-based treatment
Anal canal
Surgery (radical or local excision) generally contra-indicated as primary treatment option
Stage I Standard-dose RT, infused 5-FU and mitomycin C (stage group under-represented in randomised studies)
Low-dose RT, infused FU, and mitomycin C (no data from randomised studies)
Stage II–III Standard-dose RT, infused FU, and mitomycin C (evidence from multiple randomised studies)
Stage IV 5-FU and cisplatin, carboplatin/taxol, or possibly irinotecan/cetuximab
Anal margin
Stage I, well differentiated: Local excision (re-excision or chemoradiation if involved/close margins)
Stage II–III Standard-dose RT, infused 5-FU and mitomycin C
Stage IV 5-FU and cisplatin, or carboplatin/taxol
5-FU, 5-fluorouracil; RT, radiotherapy.
compromising sphincter function [IV, C]. Local excision has primary APE may be offered to patients previously irradiated in
not been shown to be efficacious for small tumours in the anal the pelvic region.
canal and is contra-indicated. Although more extensive and
poorly differentiated lesions have a greater risk of being lymph
node positive, it is important to do proper clinical and radio-
logical staging also of smaller lesions in order to rule out the
chemoradiation
presence of positive nodes as this is a contra-indication to local Evidence supporting the effectiveness of CRT as a radical treat-
excision. Piecemeal resections render assessment of resection ment has been provided by multiple phase II and case-series
margins in the specimen impossible and should not be carried studies. Subsequent randomised trials have established the
out. optimal regimen, although no individual randomised study has
In case of inadequate margins or R1 resection (occurs some- directly compared surgery versus CRT. Recommendations are
times after a resection of ‘anal tags’ or ‘haemorrhoids’), a further based on the results of the phase II and six randomised phase III
local excision may be considered after adequate staging, and clin- trials (EORTC 22861, UKCCCR ACT I, RTOG 87-04, RTOG
ical assessment provided R0 resection can be achieved. However, 98-11, ACCORD-03, CRUK ACT II). 5-FU with MMC com-
it is recommended that all patients having undergone a local re- bined with radiotherapy are generally recommended, rather
section, irrespective of resection margin, should be discussed by than 5-FU and cisplatin, MMC and cisplatin, any single drug or
an appropriate multidisciplinary team (MDT) to facilitate deci- any combination of three drugs [I, A].
sions regarding re-excision or definitive CRT. Stage I patients represent only 10%–15% in the majority of
Until the introduction of definitive CRT, abdomino-perineal randomised CRT trials, hence application of overall data to T1
excision (APE) was recommended for all other tumours (except tumours is limited. However, for small tumours (T1), some
those amenable to local excision). Primary APE was associated investigators have used external beam radiotherapy alone, fol-
with local failure in up to half of cases, and 5-year survival rates lowed by a small volume boost either with photons, electrons or
in the region of 50%–70% were reported [IV, C]. Today, interstitial implantation.
iii | Glynne-Jones et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
Chemoradiation schedule and assessment used in the ACT II trial
Chemoradiation
• 5FU 1000 mg/m2 days 1–4 (week 1) and 29–32 (week 5) by continuous 24 h IV infusion.
• MITOMYCIN 12 mg/m2 IV bolus on day 1 (maximum single dose 20 mg)
• RADIOTHERAPY*: Total dose 50.4 Gy delivered in 28 daily fractions starting on Day 1.
Assessment of tumour response

• Digital examination at 11, 18 and 26 weeks from the start of the treatment.
• Abdominopelvic CT at week 26.
• Confirm residual or recurrent disease by biopsy (routine biopsies not recommended).
Complete response Persistent / recurrent disease
Follow-up Surgery
Figure 2. Chemoradiation schedule used as a control arm in the ACT II trial as a working example of treatment of anal cancer [14]. *Note that the radiother-
apy schedule should not be considered standard of care.
In contrast, early investigators [15, 16] reported that CRT, with (3) Additional maintenance/consolidation chemotherapy follow-
the addition of MMC to 5-FU, demonstrated excellent local ing CRT has not impacted on local control, DFS or OS [11].
control in small tumours (<4 cm). Sequential phase II studies
with CRT have shown the efficacy of relatively low total radiation The 2-month treatment gap used in early trials, which aimed
doses (30–50 Gy) in combination with 5-FU and MMC. to allow time for tumour shrinkage and recovery of acute pelvic
Randomised, controlled studies in Europe have demonstrated toxicity, has now been abandoned (consensus of experts).
that synchronous CRT, as the primary modality, is superior to Although randomised trials have not been carried out, the
radiotherapy alone. The RTOG phase III study compared 5-FU evidence from phase II studies and data extrapolated from ran-
with 5-FU and MMC in combination with radiotherapy (median domised trials in rectal cancer suggest that capecitabine might
dose 48 Gy), and did not use a planned gap, but boosted poor be considered as an alternative to infused 5-FU.
responders with a further 9 Gy. This study confirmed the super-
iority of the combination of MMC and 5-FU.
It remains unclear whether increasing the radiation dose to radiotherapy technique and treatment
>50 Gy in patients with locally advanced anal cancer receiving fields
combined modality therapy will improve the results—particu-
larly in good responders. The patient is usually treated in the supine position, although
The second generation of randomised studies investigated the there are some exceptions where prone positioning for very exo-
role of cisplatin as a replacement of MMC in combination with phytic tumours may be better with bolus applied.
5-FU and radiation. In these studies, cisplatin and FU were also Uninterrupted treatment, avoiding a gap, is considered radio-
used before or after CRT as neo-adjuvant or maintenance treat- biologically the most effective treatment [III]. Doses of at least
ment, respectively. 45–50 Gy without a treatment gap are recommended for T1–2
The results of these studies indicate that: N0. Higher doses may be required for more advanced tumours,
particularly if a planned treatment gap is used. Boost doses to
(1) Cisplatin in combination with infused 5-FU and radiation the primary tumour have usually ranged from 15 to 25 Gy, with
does not improve either complete response rates or local higher doses applied for observed poor response. Hence, cur-
control compared with MMC and does not reduce overall rently, it is not possible to make a definitive recommendation
toxicity (but results in less myelotoxicity); (based on inter-trial comparisons of differing dose fractiona-
(2) Neo-adjuvant chemotherapy before CRT has not improved tions with or without a treatment gap) on the requirement for,
either locoregional or distant control, and colostomy-free the form (external beam or brachytherapy) or the appropriate
survival (CFS) is significantly worse [5, 10, 12]. More doses for a boost after 50 Gy.
mature data suggest that local control and DFS are also Dogmatic definition of treatment fields is also beyond the
worse [12]. Neo-adjuvant chemotherapy should not be scope of this article. There are significant differences in ap-
given outside clinical trials [I]. proach within Europe but, in general, treatment should aim to
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu159 | iii

by guest
on 05 February 2018
encompass the primary tumour and any sites of likely nodal in- normal tissues and the contralateral mucosa and sphincter.
volvement within the high-dose volume. Expertise for iridium-192 interstitial implantation is limited to a
Delivery of radiotherapy in anal cancer is complex because of few European institutions. Low-dose rate, high-dose rate (HDR)
the varying size and shape of the target volume, and the proxim- and pulsed dose rate brachytherapy have been tested in clinical
ity to dose-sensitive critical structures, such as small bowel, practice. There are currently limited data on the use of HDR
rectum, bladder femoral heads, perineum and external genitalia. brachytherapy in anal cancer and lack of consensus on the optimal
These structures often received high doses of radiation with con- fractionation schedule. Curative brachytherapy as a single modal-
ventional parallel opposed techniques. The first randomised ity is not recommended, but may be applicable as a boost follow-
trials have mainly relied on 2D-based radiation therapy plan- ing response to CRT. Double-plane, or volume implants may be
ning in which anatomic (bony) landmarks were used to guide necessary, depending on the extent of the tumour, but risk late
field design using orthogonal X-ray images. More recently, con- necrosis and radiation proctitis. Computerised 3D image-based
formal (CT-guided or 3D) radiotherapy-based treatments have treatment planning should allow optimal dose distribution.
been used, which allow the radiation oncologist to identify normal
as well as target soft-tissue structures on axial CT images, and have
treatment of the elderly
led to improved treatment accuracy and delivery.
Recent randomised trials [5, 11, 12] have shown good local Although some have recommended dose reductions, omission
control in early-stage tumours. However, radiotherapy techni- of chemotherapy or reduction of irradiated volumes for elderly
ques that have relied on anterior–posterior/posterior–anterior and frail patients, current data suggest that elderly patients
(APPA) fields may be associated with severe acute toxicity should be treated similarly to their younger counterparts. The
causing excessive breaks in treatment leading to treatment physiological fitness of elderly patients is increasing with longer
failure, and also late radiation morbidity. predicted life expectancy (based upon actuarial tables).
Overall grade 3 and 4 acute toxicity during CRT in ACT II and Consequently, this group of patients are at risk of significant
RTOG 9811 was similar in both arms, i.e. 72% and 74%, respect- under-treatment if treatment choices are based purely upon age.
ively. The most common grade 3/4 adverse events were; skin, A good collaboration between geriatricians, clinical nurse spe-
haematological and gastrointestinal [5, 11]. More conformal cialists and radiation and medical oncologists will facilitate the
treatment strategies such as intensity-modulated radiotherapy delivery of radical treatment.
(IMRT) spare organs at risk, reduce toxicity and may allow full or
even escalated doses to be achieved within a shorter overall treat- post-operative chemoradiation
ment time. Hence, IMRT or volumetric modulated arc therapy is
currently recommended for the treatment of anal cancer, setting Post-operative CRT should be considered in patients who have
strict radiation dose constraints to normal organs. undergone excision of perianal skin tags where piecemeal histo-
Several ‘proof-of-principle’ studies of IMRT in anal canal car- logical assessment means that completeness of excision cannot be
cinoma have reported significant reduction in the doses deliv- guaranteed, or in the case of narrow margins, when re-excision is
ered to the bowel, bladder and genitalia/perineal skin. not feasible, and for patients considered at risk of pelvic node in-
Prospective phase II multicentre studies (RTOG 0529) have volvement. Similar indications as for skin cancers are relevant, i.e.
shown that IMRT is deliverable in a multicentre setting [17, 18], depth of invasion, size of tumour and the extent of the surgical
with a reduction in toxicity when contrasted with the best arm margin. Other indications are local excision of anal canal lesions
of the RTOG 9811 trial. The efficacy of doses <1.8 Gy per day (which are not recommended), and in the rare cases when radical
are assumed, but data are inadequate. surgery has been carried out as primary treatment but the resec-
Australasian planning guidelines interpret CT definitions and tion margin is involved. It is recommended that all such patients
provide a high-resolution atlas for contouring gross disease and should be discussed by an appropriate MDT to facilitate decisions
organs at risk [19], which complements the existing RTOG elect- regarding re-excision or post-operative CRT.
ive nodal anorectal atlas [20]. The descriptions of the elective
target volumes or compartments are useful and reproducible. toxicity and supportive care during
The inguinal nodes should be formally included in the radi-
ation fields in the majority of cases, even in the absence of
radiotherapy
clearly demonstrable involvement. The incidence of nodal in- Patients should be assessed, and full blood counts checked
volvement increases with increasing primary tumour size and is weekly if MMC is used, as CRT is associated with high risks of
at least 20% in patients with T3 disease. However, some clini- higher grades of haematological toxicity. Patients should be
cians may treat clinically uninvolved inguinal nodes only in informed of the negative effect of smoking before CRT. Smoking
certain circumstances [e.g. T3–4 primary disease, location of may worsen acute toxicity during treatment and lead to a poorer
primary tumour within the canal (below the dentate line), ≤1 outcome in terms of DFS and CFS. Every effort should be made
cm from the anal orifice or if there is involvement of pelvic to ensure patients quit smoking before therapy.
lymph nodes (on CT or MRI criteria)]. Tolerance to treatment can be maximised with antibiotics, anti-
fungals, anti-emetics, analgesia, skin care, advice regarding nutri-
tion to prevent or correct weight loss and psychological support.
brachytherapy The recommendation for post-treatment use of vaginal dila-
Brachytherapy is a highly conformal treatment which is able to tors in sexually active females is controversial. Pre-menopausal
deliver a high dose to the primary tumour, sparing surrounding women should be informed that menopause will ensue and
iii | Glynne-Jones et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
fertility will be lost, unless the ovaries are moved out of the radi- there remains considerable heterogeneity in terms of outcomes,
ation field. Hormone replacement therapy may be appropriate particularly for more advanced stages. Biomarkers to provide
in those in whom an early menopause is induced. Sperm predictive and prognostic information, and to inform individua-
banking should be discussed before the commencement of treat- lised therapies, would be helpful. Only a limited number of
ment with male patients who wish to preserve fertility. markers in small numbers have been analysed to date, with a
variety of treatment regimens. Cytogenetic, immunohistochem-
ical and molecular markers provide information on cancer
response evaluation pathogenesis, but are not sufficiently robust to guide prognosis
Anal cancers tend to regress slowly after completion of CRT or select treatment. Further work in randomised studies is
treatment. DRE is the mainstay of determining complete re- required.
sponse after treatment—defined as the absence of tumour and/ The cell cycle regulator p16 is overexpressed in high-risk
or ulceration. Examination may be more informative when HPV-related cervical cancers, which may represent a simple sur-
carried out under general anaesthesia if pain persists or response rogate biomarker for identifying squamous cell carcinomas har-
is difficult to quantify. Careful clinical inspection of the inguinal bouring HPV DNA. Patients with moderate/strong p16 staining
regions in addition to radiographic evaluation (with pelvic MRI may achieve better response to CRT and have a lower risk of
and CT scans, or as a comparison of PET-CTs, if available) is relapse than patients with absent or weak staining.
also necessary. Oedema, residual fibrosis or scar tissue can be A recent systematic review examined 29 different biomarkers
difficult to distinguish from persistent active disease. However, [23]. Tumour suppressor genes p53 and p21 were the only bio-
biopsies of persistent clinically suspicious lesions 8–12 weeks markers which were prognostic in more than one study.
after CRT completion are not routinely recommended. Molecular biomarkers associated with HPV deregulation (i.e.
Treatment-related effects may confound the pathological inter- p16, Ki67, MCM7, K17, K7, K2 and HPV E4) may be relevant.
pretation of post-treatment biopsies. Good radiographic partial In an analysis of 240 patients randomised in the UKCCR ACT I
regression can be managed by close follow-up, to confirm that anal cancer trial, the presence of mutated p53 predicted for a
(a delayed) complete regression takes place, which may take 6 poorer cause-specific survival. In summary, there are no current
months. A decision regarding salvage surgery should be de- biomarkers that consistently predict sensitivity to CRT, and
ferred safely in these circumstances. Hazard ratios from the more research is required to identify molecular markers.
ACT II data indicate that assessment at 26 weeks is the most dis-
criminating end point with the most significant effect on
outcome, and is therefore the optimum time point for definitive
follow-up and surveillance and long-term
assessment with a view to salvage surgery. Residual or ‘recur- implications
rent’ tumour must be confirmed histologically before consider- follow-up
ing proceeding to radical surgery.
MRI complements clinical assessment, and acts as a useful Patients in complete remission at 8 weeks should be evaluated
baseline. MRI can capture and document response, but no indi- every 3–6 months for a period of 2 years, and 6–12 monthly
vidual MRI feature appears predictive of eventual outcome [21]. until 5 years, with clinical examination including DRE and pal-
EUS is controversial as oedema and scar tissue may be difficult pation of the inguinal lymph nodes. Anoscopy or proctoscopy is
to distinguish from persistent tumour. an additional option, but is sometimes poorly tolerated and too
To date, few FDG–PET/CT studies have assessed treatment painful following CRT. Some recommend MRI on a 6-monthly
response, and the timing of assessment is controversial. The basis for 3 years. Suspicious progressing lesions should be biop-
benefit of PET is rather to detect residual sub-clinical pelvic or sied. Data from ACT II suggest very few (<1%) relapses occur
extra-pelvic/para-aortic node involvement. after 3 years, so extended imaging surveillance after this time is
not recommended.
Patients tend to relapse locoregionally rather than at distant
prognostic factors sites. Regular CT scans for metastatic surveillance outside trials
Prognostic factors for survival (and CFS) in anal cancer include remains controversial, as the evidence for benefit of resection of
male sex, positive lymph nodes particularly positive inguinal metastases as carried out in colorectal cancer is lacking, al-
lymph nodes and primary tumour size >5 cm. The original though anecdotally, salvage is achieved in some cases.
EORTC 22861 trial also found that skin ulceration was prognos-
tic for worse survival and local control. Recent multivariate ana- management of advanced/metastatic disease
lysis from the ACT I trial validated against the ACT II dataset Approximately 10%–20% of patients suffer distant relapse. The
showed that positive inguinal nodes and the male sex were prog- most common sites of metastatic spread are to the para-aortic
nostic for local regional failure, ACD and OS, as was haemoglo- nodes, liver, lungs and skin, which usually appear relatively late
bin [22]. Even in the context of HIV, patients with anal cancer and in the context of local persistence or recurrence of disease
who smoke also appear to have a worse OS than non-smokers. following treatment. The prognosis in this group is poor with
only 10% of patients with distant metastases surviving 2 years or
more, but long-term survivors are described. Patients with small
personalised medicine volume or isolated metastatic disease should be further dis-
Despite intriguing developments in our understanding of the cussed by an appropriate MDT, in case there are surgical or
molecular biology and processes which lead to anal cancer, CRT options.
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu159 | iii

by guest
on 05 February 2018
There is no consensus on the standard chemotherapy treat- done in rectal cancer, and surgery is carried out in a heavily irra-
ment. The choice of chemotherapy is often influenced by previ- diated area, the risk of post-operative complications, in particular
ously used agents in the initial CRT regimen, but regimens with involving the perineal wound, are substantial. Perineal recon-
good documented activity are limited and generally have pro- struction with musculocutaneous flaps is generally recommended
duced unsatisfactory results. and appears to reduce complication rates [26]. Persistent or pro-
Otherwise fit patients with symptomatic metastatic or recurrent gressive disease in the inguinal lymph nodes should be considered
disease not amenable to surgery should be considered for chemo- for surgery, i.e. radical groin dissection and pre- or post-operative
therapy, usually with a combination of cisplatin and 5-FU. irradiation discussed—depending on the dose distribution from
Activity is also reported for carboplatin, doxorubicin, taxanes and the definitive CRT. Flap reconstruction may be needed in some
irinotecan ± cetuximab—or combinations of these agents. These instances when recurrence in an irradiated groin is subject to sur-
options will be influenced by the disease-free interval, and the gical salvage. Salvage surgery can achieve local pelvic control in
patient’s preferences and performance status. Responses are rarely ∼60% of cases, and a 5-year survival rate of 30%–60%.
complete and usually of short duration. Currently, the internation-
al rare cancers initiative, which is a consortium of international
investigators from the UK, US, Europe and Australia, has devel- palliative care
oped a multicentre international trial testing the role of carbopla- Pain due to recurrent pelvic tumour can be extreme, and requires
tin/paclitaxel against the common standard 5-FU/cisplatin. expertise in combinations of opiate and non-opiate pain relief,
sedatives and anxiolytics. Nerve blocks and re-irradiation may be
quality of life feasible. Fistula from the bladder or rectum is not uncommon
Data on long-term quality of life are sparse, but appear to show and demands meticulous skin care and, rarely, surgical diversion
that patients are satisfied despite objective impairment of
sphincter function. Continence and quality of life appear Table 5. Summary of recommendations
impaired in many patients [24, 25]. Efforts should be made to
Patients with anal cancer should be managed, from diagnosis
document quality of life and late effects. Population data suggest through initial treatment and subsequent surveillance, by an
that function is poor, particularly if patients continue to smoke. experienced and specialised multidisciplinary team.
Sexual and urinary function may also be compromised. In the Locoregional control with good quality of life and the avoidance of a
RTOG 9811 trial, the rate of severe long-term toxic effects was permanent stoma is the primary aim of treatment.
similar in both arms, 11% versus 10%, but only 5% required a The optimal total dose of radiation is unknown. Chemoradiation
colostomy for treatment-related late effects. Adverse late effects (CRT) with at least 45 Gy, infused FU and MMC remains the
appear to relate mostly to total radiation dose received in multi- standard treatment of stage II or higher anal canal tumours and a
variate analysis rather than the type of chemotherapy. boost with 15–20 Gy may be applicable, especially if
Information regarding treatment side-effects should be pro- chemotherapy cannot be safely delivered, leading to cure in the
vided clearly, particularly on sexual functioning as substantial majority of patients.
numbers of patients describe difficulty with their sex lives, with Less intensive treatment programmes with lower doses of irradiation
specific concerns regarding loss of libido, inability to enjoy sex may be successfully used in smaller tumours or fragile patients
and erectile dysfunction. Follow-up of issues relating to sexual although evidence from randomised trials is not available.
dysfunction has been sub-optimal, particularly for female Neo-adjuvant and adjuvant chemotherapy using cisplatin has not been
patients who have undergone radical pelvic radiotherapy. shown to improve outcomes (progression-free survival or OS).
There is increasing support in the literature towards the devel- CRT with 5-FU and cisplatin had similar complete response and
opment of nurse-led, late effects/survivorship clinics for patients overall toxicity when compared with 5-FU and MMC but less
who have received pelvic radiotherapy. There are reports detailing haematological toxicity. Any marginal benefit for cisplatin in
terms of haematological toxicity is likely to be outweighed by the
the effectiveness of pelvic floor exercises and/or biofeedback train-
extra resources needed to administer cisplatin—two courses of i.v.
ing in patients who experience faecal urgency and incontinence.
treatment with hydration over several hours, compared with only
a single dose of MMC.
salvage surgical treatment Response should be assessed from 6 weeks, but data suggest that the
Patients with locally persistent, progressive or recurrent disease optimal time to assess complete response may be 26 weeks, rather
should be considered for surgical salvage [I, A]. A very small pro- than 11 weeks, if surgical salvage is discussed.
portion of patients may be treated by local resection. At least an Surveillance/follow-up after completion of CRT treatment has not
abdomino-perineal excision is required in the majority of patients been rigorously examined, but should focus on salvage of local
since achievement of a negative resection margin appears crucial. failure (<10% will recur after the first 3 years following completion
For some patients, a posterior or total pelvic exenteration is of chemoradiation treatment).
The techniques for surgical salvage in anal cancer are different from
required and surgery should preferably be carried out in institu-
that carried out for rectal cancer, are associated with high
tions with experience of multi-visceral resections. In addition to a
morbidity, and require input from multiple surgical teams (e.g.
positive biopsy, pre-operative local staging is mandatory and MRI
urology and plastic reconstruction).
provides an excellent alternative. Although distant metastases are
unusual, CT scan of thorax and abdomen (or PET/CT) is advised OS, overall survival; 5-FU, 5-fluorouracil; MMC, mitomycin C.
to rule out the occurrence of such. Because the salvage operation
for anal cancer involves a wider perineal resection than what is
iii | Glynne-Jones et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
procedures in patients with reasonable life expectancy. Palliation Recent reviews and guidelines are available as listed in the lit-
of the dying patient with anal cancer is often difficult. The symp- erature.
toms are often odorous and may make emotional support for the A summary of recommendations is provided in Table 5. Levels
patient and family members very challenging. of evidence and grades of recommendation have been applied
using the system shown in Table 6. Statements without grading
were considered justified standard clinical practice by the expert
conclusion authors and the reviewers. The present guidelines have been formu-
An MDT approach is essential for the optimal management of lated with the assistance of the United Kingdom National Cancer
anal cancer. Despite the results of four randomised phase III Research Institute (NCRI) multidisciplinary Anal Cancer Group.
trials in anal cancer, the paradigm of external beam radiation
therapy with concurrent 5-FU and MMC developed over 30
years ago by Norman Nigro remains the standard of care. conflict of interest
As anal cancer is a rare tumour, the authors strongly believe RG-J has reported that, in the last 3 years, he has received hon-
that it is in the interest of all patients to be offered participation oraria for lectures from Roche, Merck-Serono, Pfizer and Sanofi
in a clinical trial. National and international trials in this disease and that he has received honoraria for advisory boards from
site are on-going throughout Europe. Roche, Merck-Serono, Eli Lilly and Sanofi. He also has received
funding and free cetuximab from Merck-Serono for a phase I/II
study integrating cetuximab into CRT in rectal cancer. He
note
receives a supply of bevacizumab from Roche for 3 months for 60
PubMed and Medline were searched for articles published patients in a current randomised phase II study as neo-adjuvant
between 1990 and June 2013. The search terms included squa- chemotherapy in rectal cancer. He has finally received support for
mous cell carcinoma, anal cancer, anal canal carcinoma, anal attending international meetings in GI cancer from Roche,
margin cancer, survival, diagnosis, recurrence, surgery, chemo- Merck-Serono and Sanofi. The other authors have reported no
therapy, radiotherapy, CRT and chemoradiotherapy. potential conflicts of interest.
Table 6. Levels of evidence and grades of recommendation

(adapted from the Infectious Diseases Society of America-United references
States Public Health Service Grading Systema) 1. Jemal A, Simard EP, Dorell C et al. Annual Report to the Nation on the Status of
Cancer, 1975–2009, featuring the burden and trends in human papillomavirus
Levels of evidence (HPV)-associated cancers and HPV vaccination coverage levels. J Natl Cancer Inst
I Evidence from at least one large randomised, controlled 2013; 105: 175–201.
trial of good methodological quality (low potential for 2. Palefsky JM, Giuliano AR, Goldstone S et al. HPV vaccine against anal HPV
bias) or meta-analyses of well-conducted randomised infection and anal intraepithelial neoplasia. N Engl J Med 2011; 365:
1576–1585.
trials without heterogeneity
3. Goh V, Gollub FK, Liaw J et al. Magnetic resonance imaging assessment of
II Small randomised trials or large randomised trials with a
squamous cell carcinoma of the anal canal before and after chemoradiation: can
suspicion of bias (lower methodological quality) or
MRI predict for eventual clinical outcome? Int J Radiat Oncol Biol Phys 2010; 78:
meta-analyses of such trials or of trials with 715–721.
demonstrated heterogeneity 4. Edge SB, Byrd DR, Compton CC. eds. AJCC Cancer Staging Handbook, 7th
III Prospective cohort studies edition. New York, NY: Springer 2010.
IV Retrospective cohort studies or case–control studies 5. Bartelink H, Roelofsen F, Eschwege F et al. Concomitant radiotherapy and
V Studies without control group, case reports, experts chemotherapy is superior to radiotherapy alone in the treatment of locally
opinions advanced anal cancer: results of a phase III randomized trial of the European
Organization for Research and Treatment of Cancer Radiotherapy and
Grades of recommendation Gastrointestinal Cooperative Groups. J Clin Oncol 1997; 15: 2040–2049.
A Strong evidence for efficacy with a substantial clinical 6. British HIV Association, British Association of Sexual Health and HIV, British
Infection Society. UK National Guidelines for HIV Testing 2008. www.bhiva.org/
benefit, strongly recommended
hivtesting2008.aspx (14 April 2014, date last accessed).
B Strong or moderate evidence for efficacy but with a limited
7. Oehler-Jänne C, Huguet F, Provencher S et al. HIV-specific differences in outcome
clinical benefit, generally recommended
of squamous cell carcinoma of the anal canal: a multicentric cohort study of HIV-
C Insufficient evidence for efficacy or benefit does not positive patients receiving highly active antiretroviral therapy. J Clin Oncol 2008;
outweigh the risk or the disadvantages (adverse events, 26: 2550–2557.
costs, … ), optional 8. Wexler A, Berson AM, Goldstone SE et al. Invasive anal squamous-cell carcinoma
D Moderate evidence against efficacy or for adverse outcome, in the HIV-positive patient: outcome in the era of highly active antiretroviral therapy.
generally not recommended Dis Colon Rectum 2008; 51: 73–81.
E Strong evidence against efficacy or for adverse outcome, 9. Fraunholz I, Rabeneck D, Gerstein J et al. Concurrent chemoradiotherapy with
never recommended 5-fluorouracil and mitomycin C for anal carcinoma: are there differences between
HIV-positive and HIV-negative patients in the era of highly active antiretroviral
a therapy? Radiother Oncol 2011; 98: 99–104.
By permission of the Infectious Diseases Society of America [27].
10. Epidermoid anal cancer: results from the UKCCCR randomised trial of radiotherapy
alone versus radiotherapy, 5-fluorouracil, and mitomycin. UKCCCR Anal Cancer
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu159 | iii

by guest
on 05 February 2018
Trial Working Party. UK Co-ordinating Committee on Cancer Research. Lancet 19. Ng M, Leong T, Chander S et al. Australasian Gastrointestinal Trials Group (AGITG)
1996; 348: 1049–1054. contouring atlas and planning guidelines for intensity-modulated radiotherapy in
11. Flam M, John M, Pajak TF et al. Role of mitomycin in combination with fluorouracil anal cancer. Int J Radiat Oncol Biol Phys 2012; 83: 1455–1462.
and radiotherapy, and of salvage chemoradiation in the definitive nonsurgical 20. Myerson RJ, Garofalo MC, El Naqa I et al. Elective clinical target volumes for
treatment of epidermoid carcinoma of the anal canal: results of a phase III conformal therapy in anorectal cancer: a radiation therapy oncology group
randomized intergroup study. J Clin Oncol 1996; 14: 2527–2539. consensus panel contouring atlas. Int J Radiat Oncol Biol Phys 2009; 74: 824–830.
12. Ajani JA, Winter KA, Gunderson LL et al. Fluorouracil, mitomycin, and radiotherapy 21. Gunderson LL, Winter KA, Ajani JA et al. Long-term update of US GI intergroup
vs fluorouracil, cisplatin, and radiotherapy for carcinoma of the anal canal: a RTOG 98-11 phase III trial for anal carcinoma: survival, relapse, and colostomy
randomized controlled trial. JAMA 2008; 299: 1914–1921. failure with concurrent chemoradiation involving fluorouracil/mitomycin versus
13. Peiffert D, Tournier-Rangeard L, Gérard JP et al. Induction chemotherapy and dose fluorouracil/cisplatin. J Clin Oncol 2012; 30: 4344–4351.
intensification of the radiation boost in locally advanced anal canal carcinoma: final 22. Glynne-Jones R, James R, Meadows H et al. Optimum time to assess complete
analysis of the randomized UNICANCER ACCORD 03 trial. J Clin Oncol 2012; 30: clinical response (CR) following chemoradiation (CRT) using mitomycin (MMC)
1941–1948. or cisplatin (CisP), with or without maintenance CisP/5FU in squamous cell
14. James RD, Glynne-Jones R, Meadows HM et al. Mitomycin or cisplatin carcinoma of the anus: results of ACT II. 2012 ASCO Annual Meeting. J Clin Oncol
chemoradiation with or without maintenance chemotherapy for treatment of 2012; 30(suppl): abstr 4004.
squamous-cell carcinoma of the anus (ACT II): a randomised, phase 3, open-label, 23. Lampejo T, Kavanagh D, Clark J et al. Prognostic biomarkers in squamous cell
2 × 2 factorial trial. Lancet Oncol 2013; 14: 516–524. carcinoma of the anus: a systematic review. Br J Cancer 2010; 103: 1858–1869.
15. Nigro ND, Vaitkevicius VK, Considine B, Jr. Combined therapy for cancer of the 24. Bentzen AG, Balteskard L, Wanderås EH et al. Impaired health-related quality of
anal canal: a preliminary report. Dis Colon Rectum 1974; 17: 354–356. life after chemoradiotherapy for anal cancer: late effects in a national cohort of 128
16. Cummings BJ, Keane TJ, O’Sullivan B et al. Epidermoid anal cancer: treatment by survivors. Acta Oncol 2013; 52: 736–744.
radiation alone or by radiation and 5-fluorouracil with and without mitomycin C. Int 25. Bentzen AG, Guren MG, Vonen B et al. Faecal incontinence after
J Radiat Oncol Biol Phys 1991; 21: 1115–1125. chemoradiotherapy in anal cancer survivors: long-term results of a national cohort.
17. Kachnic LA, Tsai HK, Coen JJ et al. Dose-painted intensity-modulated radiation Radiother Oncol 2013; 108: 55–60.
therapy for anal cancer: a multi-institutional report of acute toxicity and response 26. Sunesen KG, Buntzen S, Tei T et al. Perineal healing and survival after anal cancer
to therapy. Int J Radiat Oncol Biol Phys 2012; 82: 153–158. salvage surgery: 10-year experience with primary perineal reconstruction using the
18. Kachnic LA, Winter K, Myerson RJ et al. RTOG 0529: a phase 2 evaluation of vertical rectus abdominis myocutaneous (VRAM) flap. Ann Surg Oncol 2009; 16:
dose-painted intensity modulated radiation therapy in combination with 68–77.
5-fluorouracil and mitomycin-C for the reduction of acute morbidity in carcinoma 27. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections among
of the anal canal. Int J Radiat Oncol Biol Phys 2013; 86: 27–33. hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33: 139–144.
iii | Glynne-Jones et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
clinical practice guidelines Annals of Oncology 27 (Supplement 5): v50–v57, 2016
doi:10.1093/annonc/mdw329
Oesophageal cancer: ESMO Clinical Practice Guidelines

for diagnosis, treatment and follow-up†
F. Lordick1, C. Mariette2, K. Haustermans3, R. Obermannová4 & D. Arnold5 on behalf of the ESMO
Guidelines Committee*
1
University Cancer Centre Leipzig, University Hospital Leipzig, Leipzig, Germany; 2Department of Digestive and Oncological Surgery, University Hospital Claude Huriez, Lille,
France; 3Department of Radiation Oncology, Leuven Cancer Institute, University Hospitals Leuven, Leuven, Belgium; 4Clinic of Comprehensive Cancer Care, Masaryk
Memorial Cancer Institute and Faculty of Medicine, Masaryk University, Brno, Czech Republic; 5Instituto CUF de Oncologia, Lisbon, Portugal
incidence and epidemiology diagnosis and pathology/molecular

Oesophageal cancer is the 19th most common cancer in the biology
European Union (EU), with ∼45 900 new cases diagnosed in Screening for Barrett’s oesophagus, endoscopic surveillance and
2012 (1% of the total). In the EU, the highest age-standardised ablation of precursor lesions are not in the focus of this guide-
incidence rates for oesophageal cancer are in the Netherlands line. We recommend to follow the recently updated guidelines
clinical practice
for men and the UK for women [1]. Variation between coun- of the American College of Gastroenterology [5].
guidelines
tries is high and may reflect different prevalence of risk factors, All patients with new dysphagia, gastrointestinal bleeding,
use of screening and diagnostic methods. recurrent aspiration or emesis, weight loss and/or loss of appe-
Between 2000–04 and 2005–09, oesophageal cancer mortality tite should undergo an upper intestinal endoscopy [III, A].
declined by 7% (from 5.34 to 4.99/100 000) in EU men, and by Approximately three-quarters of all ACs are found in the distal
3% (from 1.12 to 1.09/100 000) in EU women. Predictions to oesophagus, whereas SCCs occur more frequently in the
2015 show persistent declines in mortality rates for men in the proximal to middle oesophagus [3]. Biopsies should be taken
EU overall and stable rates for EU women, with rates for 2015 of from all suspect areas. The minimal recommended number of
4.5/100 000 men (∼22 300 deaths) and 1.1/100 000 women biopsies is not defined. The diagnosis should be made from an
(∼7400 deaths). endoscopic biopsy with the histology classified according to
Oesophageal cancer has two main subtypes—oesophageal the World Health Organization (WHO) criteria [6]. The differ-
squamous cell carcinoma (SCC) and oesophageal adenocarcin- entiation between SCC and AC is of prognostic and clinical
oma (AC). Although SCC accounts for ∼90% of cases of oe- relevance.
sophageal cancer worldwide, mortality rates associated with AC Immunohistochemical stainings are recommended in poorly
are rising and have surpassed those of SCC in several regions in and undifferentiated cancers (G 3/4) according to WHO to
the EU [2]. differentiate between SCC and AC [V, B]. Additionally, small
Oesophageal carcinoma is rare in young people and increases in cell carcinoma and other rare histologies (endocrine tumours,
incidence with age, peaking in the seventh and eighth decades of lymphoma, mesenchymal tumours, secondary tumours and
life. AC is three to four times as common in men as it is in women, melanoma) must be identified separately from SCC and AC and
whereas the sex distribution is more equal for SCC [3]. should be treated accordingly.
The main risk factors for SCC in Western countries are
smoking and alcohol consumption, whereas AC predominantly
occurs in patients with chronic gastro-oesophageal reflux staging and risk assessment
disease and their risk is correlated with the patient’s body mass Decisions on the initial treatment approach of oesophageal
index with a higher risk for obese persons [3, 4]. cancer are taken on the basis of clinical staging, which should be
done with the highest degree of accuracy possible. Staging
should include a complete clinical examination and a computed
*Correspondence to: ESMO Guidelines Committee, ESMO Head Office, Via L. Taddei 4,
6962 Viganello-Lugano, Switzerland. tomography (CT) scan of the neck, chest and abdomen [III, A].
E-mail: clinicalguidelines@esmo.org Ultrasound of the abdomen can be carried out initially as a
†
simple and inexpensive test to exclude stage 4 liver metastases.
Approved by the ESMO Guidelines Committee: August 2003, last update August 2016.
This publication supersedes the previously published version—Ann Oncol 2013; 24 In candidates for surgical resection, endoscopic ultrasound
(Suppl. 6): vi51–vi56. (EUS) should be carried out to evaluate the T and N tumour
Downloaded from https://academic.oup.com/annonc/article-abstract/27/suppl_5/v50/1741562

by guest
on 05 February 2018
categories [III, B]. The sensitivity and specificity of EUS for the Table 1. TNM staging for oesophageal cancer (UICC/AJCC, 7th
correct evaluation of the T category are 81%–92% and 94%– edition) [8, with permission]
97%, respectively. It is lower for the N category [7]. 18F-fluoro- Definition of TNM (2009)
deoxyglucose-positron emission tomography (FDG-PET; today Primary tumour (T)
mostly done as PET-CT) is particularly helpful to identify other-
wise undetected distant metastases. 18F-FDG-PET should, TX Primary tumour cannot be assessed
therefore, be carried out in patients who are candidates for oeso- T0 No evidence of primary tumour
phagectomy [III, B], as the finding of otherwise unknown Tis Carcinoma in situ/high-grade dysplasia
distant metastases may prevent patients from futile surgery. T1 Tumour invades lamina propria or submucosa
T1a Tumour invades mucosa or lamina propria or muscularis
However, the availability of PET-CT differs among countries
mucosae
and centres.
T1b Tumour invades submucosa
A tracheobronchoscopy should be carried out in the case of
T2 Tumour invades muscularis propria
tumours at or above the tracheal bifurcation to exclude tracheal
T3 Tumour invades adventitia
invasion. In the case of oesophageal SCC due to chronic tobacco
T4 Tumour invades adjacent structures
and alcohol consumption, meticulous investigation of the oral T4a Tumour invades pleura, pericardium, diaphragm or adjacent
cavity, oropharynx and hypopharynx by an ear, nose and throat peritoneum
specialist, as well as trachea-bronchoscopy to exclude a syn- T4b Tumour invades other adjacent structures such as aorta,
chronous second cancer in the aerodigestive tract, should be vertebral body or trachea
carried out [IV, B].
Regional lymph nodes (N)
In locally advanced (T3/T4) ACs of the oesophago-gastric
junction (OGJ) infiltrating the anatomic cardia, laparoscopy can NX Regional lymph nodes cannot be assessed
be done to rule out peritoneal metastases, which are found in N0 No regional lymph node metastasis
∼15% of patients. [IV, C]. The finding of otherwise unknown N1 Metastasis in 1–2 regional lymph nodes
peritoneal metastases may prevent patients from futile surgery. N2 Metastasis in 3–6 regional lymph nodes
The stage is to be given according to the American Joint N3 Metastasis in 7 or more regional lymph nodes
Committee on Cancer (AJCC)/Union for International Cancer Distant metastasis
Control (UICC) TNM staging system (7th edition) (Table 1)
[8]. Anatomic staging should be complemented by medical risk MX Distant metastasis cannot be assessed
assessment, especially in patients who are scheduled for multi- M0 No distant metastasis
M1 Distant metastasis
modal therapy and/or surgery. Medical risk assessment should
comprise a differential blood count as well as liver, pulmonary, Stage grouping
cardiac and renal function tests.
Carcinomas of the oesophagus and gastro-oesophageal junction
The nutritional status and history of weight loss should be
Stage 0 Tis N0 M0
assessed according to The European Society for Clinical Stage IA T1 N0 M0
Nutrition and Metabolism (ESPEN) guidelines [III, A] [9]. More Stage IB T2 N0 M0
than half of patients lose >5% of their body weight before admis- Stage IIA T3 N0 M0
sion to oesophagectomy, and 40% lose >10%. Independent from Stage IIB T1, T2 N1 M0
the body mass index, weight loss confers an increased operative Stage IIIA T4a N0 M0
risk, worsens a patient’s quality of life and is associated with poor T3 N1 M0
survival in advanced disease. Therefore, nutritional support T1, T2 N2 M0
according to the ESPEN guidelines [10] is an integral part of the Stage IIIB T3 N2 M0
medical care for patients with oesophageal cancer in the curative Stage IIIC T4a N1, N2 M0
and in the palliative setting [II, A]. T4b Any N M0
Any T N3 M0
Stage IV Any T Any N M1
The regional lymph nodes, irrespective of the site of the primary tumour,
management of local/locoregional are those in the oesophageal drainage area including coeliac axis nodes and
disease (M0) paraoesophageal nodes in the neck but not supraclavicular nodes.
Edge et al. [8]. Used with the permission of the American Joint Committee
Upfront interdisciplinary planning of the treatment is manda-
on Cancer (AJCC), Chicago, IL, USA. The original source for this material
tory [III, A]. The main factors for selecting primary therapy are
is the AJCC Cancer Staging Handbook, 7th edition (2010) published by
tumour stage and location, histological type, and the patient’s
Springer Science and Business Media LLC, www.springer.com.
performance status (PS) and comorbidities. Nutritional status
matters and should be corrected. Endoscopic stenting should
not be used in locoregional disease in operable patients and al-
ternative routes of feeding (e.g. with needle catheter jejunost- limited disease (cT1–T2 cN0 M0)
omy) should be preferred [II, A] [11]. Patient preferences Surgery is the treatment of choice in limited disease. In patients
should also be assessed and be taken into account. A summary with T1a AC, endoscopic therapy is the preferred therapeutic
of treatment recommendations is shown in Figure 1. approach, being both effective and well tolerated [II, A].
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw329 | v

by guest
on 05 February 2018
Figure 1. Algorithm for the treatment of local/locoregional resectable thoracic oesophageal cancer. EUS, endoscopic ultrasound; FDG-PET, fluorodeoxyglu-
cose-positron emission tomography; MS-CT, multislice-computed tomography; cTNM, clinical tumour, node, metastases classification according to AJCC/
UICC [8]; CRT, chemoradiotherapy; OS, overall survival. 1Criteria for endoscopic instead of surgical resection are specified in the text. 2For patients unable or
unwilling to undergo surgery, combined CRT is superior to radiotherapy alone. 3Evidence suggests that neoadjuvant CRT followed by surgery and definitive
CRT are equally effective with regard to overall survival. Oesophageal surgery should be carried out in experienced (high volume) centres only. For patients not
willing to undergo oesophageal surgery or who are medically unfit for major surgery, definitive chemoradiotherapy should be preferred. Even many experi-
enced centres prefer definitive CRT for oesophageal tumours with a very proximal/cervical location. 4Sufficient evidence supports the use of perioperative
chemotherapy as well as neoadjuvant CRT. Both standards can be recommended with an equal level of evidence/grade of recommendation [I, A]. Several
ongoing studies in Europe are comparing both modalities. Inclusion of patients in one of these studies is encouraged. Some centres prefer neoadjuvant CRT for
tumours of the oesophagus and AEG type I or II according to the Siewert’s classification, while they use perioperative chemotherapy for AEG type III or II, but
this is only a pragmatic solution not currently supported by scientific evidence. 5This is optional in the case of incomplete response to CRT or local relapse.
This should be carried out only in selected patients and experienced centres.
Endoscopic mucosal resection (EMR) and endoscopic sub- bloc resections were shown to be higher with ESD [II, B]. In
mucosal dissection (ESD) are both regarded as effective endo- addition, relapses occurred less often [13].
scopic resection techniques. Similar cure rates compared with Radical and transthoracic oesophagectomy (Ivor-Lewis pro-
surgical resection have been reported in specialised centres [12]. cedure) is the surgical technique of choice [I, B] in localised oe-
Furthermore, in patients with superficial submucosal infiltration sophageal cancer beyond very early stages (T1a N0). A
of an AC, but without further risk criteria ( pT1sm1; <500 μm prospective randomised study showed a strong trend towards
invasion, L0, V0, G1/2, <20 mm diameter, no ulceration), endo- better survival outcomes for this approach in resectable stage I–
scopic resection can be considered as an alternative to oesopha- IV AC and OGJ AC, compared with less radical transhiatal re-
gectomy, but outcomes are still more limited than in mucosal section in AC of the oesophagus [14]. Details concerning endo-
AC [IV, B]. In the case of a high-grade intraepithelial neoplasia scopic and surgical resection techniques are not in the scope of
or a mucosal carcinoma (L0, V0, no ulceration, grading G1/G2, this article but can be found elsewhere [15, 16]. The role of a min-
infiltration grade m1/m2) in the squamous epithelium, an endo- imally invasive approach to the thoracic and/or abdominal cavities
scopic en bloc resection should be carried out [III, A]. ESD is increasing in clinical practice. Recent randomised studies suggest
should be preferred over EMR, especially in lesions >15 mm, as that either thoracoscopic oesophagectomy or Ivor-Lewis procedure
in Japanese studies en bloc resection rate and the rate of R0 en with laparoscopic gastric mobilisation and open right thoracotomy
v | Lordick et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
(called hybrid minimally invasive oesophagectomy) have led to squamous cell carcinoma: Meta-analyses and a recent phase III
significantly lower postoperative complication rates, especially pul- study [20, 22, 23, 31] demonstrate that patients with locally
monary complications. For hybrid minimally invasive oesopha- advanced disease benefit from preoperative chemotherapy or, most
gectomy, it was also demonstrated that short-term oncological likely to a greater extent, from preoperative CRT, with higher rates
outcomes, compared with classical Ivor-Lewis procedure, are not of complete tumour resection and better local tumour control and
deteriorated [17, 18]. Laparoscopic gastric mobilisation is now the survival [I, A]. It was suggested in the past that preoperative CRT
standard procedure, based on the results of two randomised, con- may also increase postoperative mortality rates, but this has not
trolled trials [II, A]. The additional role of thoracoscopic dissection been the case when treatment is carried out in expert centres, with
should be confirmed in additional randomised studies, as well as modern radiation planning techniques, use of adequate radiation
its long-term oncological outcome/safety. If done, the procedure doses and fractionation and a good multidisciplinary cooperation
should be carried out in expert centres for selected patients with and infrastructure. On the basis of the results of the
small tumours. Chemoradiotherapy for Oesophageal Cancer Followed by Surgery
Of note, the results of large, multicentre studies in different Study (CROSS) [31, 32], the weekly administration of carboplatin
health systems provide sufficient evidence to support the cen- (doses titrated to achieve an area under the curve of 2 mg/ml/min)
tralisation of oesophagectomy to high volume centres, with a and paclitaxel (50 mg/m2 of body-surface area) for 5 weeks and
lower rate of morbidity and better infrastructure to deal with concurrent RT (41.4 Gy in 23 fractions, 5 days per week), followed
complications following major surgery, thereby preventing by surgery, can be recommended as a contemporary standard of
further mortality [I, A] [19–21]. care [I, A]. However, only patients with clinical stage T1N1 or T2-
The value of preoperative treatment in limited disease is uncer- 3N0-1 were included in that trial.
tain, as the number of patients who have been included in pro- Two prospective, randomised controlled studies resulted in
spective randomised clinical trials is small [22–25]. A recent equivalent overall survival (OS) outcomes of definitive CRT
randomised study involving 195 patients with stage I and stage II without surgery compared with neoadjuvant CRT followed by
oesophageal cancer showed that compared with surgery alone, surgery, although the non-operative strategy was associated with
neoadjuvant chemoradiotherapy (CRT) with cisplatin plus fluor- higher local tumour recurrence rates [33, 34]. Therefore, neoad-
ouracil did not improve R0 resection rate or survival but enhances juvant CRT with planned surgery or definitive CRT with close
postoperative mortality. The results of this study also suggest that surveillance and salvage surgery for local tumour persistence or
surgery alone should be recommended as the primary treatment progression [30] can be considered to be the recommended de-
approach for cT2N0 oesophageal cancer, despite 50% of patients finitive treatments for locally advanced SCC of the oesophagus
having nodal disease at the time of surgery [II, B] [26, 27]. [II, B] [22]. However, there are currently no data comparing
For patients unable or unwilling to undergo surgery, combined neoadjuvant CRT + surgery versus definitive CRT and salvage
CRT is superior to radiotherapy (RT) alone [II, A] [21]. Four surgery on demand. Definitive CRT is recommended for cervi-
courses of cisplatin/5-fluorouracil (5-FU) combined with radically localised tumours [III, B].
ation doses of 50.4 Gy in fractions of 1.8 Gy are regarded as For patients unable or unwilling to undergo surgery, treat-
standard for definitive CRT. Alternatively, six cycles of oxalipla- ment recommendations from the ‘limited disease’ section may
tin/5-FU/folinic acid (FOLFOX) can be given [I, C] [28]. Recent be adapted.
evolutions in technology with intensity-modulated and volumet-
ric arc RT combined with functional imaging allow for increased adenocarcinoma: On the basis of the recent meta-analyses and
radiation doses up to 60 Gy in fractions of 1.8–2.0 Gy, frequently the largest prospective randomised controlled studies, peri-
using a simultaneously integrated boost. This approach allows for operative chemotherapy with regimens containing a platinum
shortening the overall treatment time, which is advantageous es- and a fluoropyrimidine for a duration of 8–9 weeks in the pre-
pecially in SCC of the oesophagus. There is insufficient evidence operative phase (as well as 8-9 weeks in the postoperative phase,
at this time to state that increased doses of RT improve survival in if feasible) or preoperative CRT (41.4–50.5 Gy) should be con-
oesophageal cancer [29], as the results of randomised studies sidered standard in locally advanced AC of the oesophagus, in-
evaluating the safety and oncological benefits of RT doses higher cluding OGJ cancers [I, A] [22–25]. Direct comparison of
than 50.4 Gy are not yet available. This is of importance if salvage chemotherapy versus CRT is scarce. Smaller randomised studies
oesophagectomy is considered as a therapeutic strategy, since have shown that the addition of RT to neoadjuvant chemother-
doses higher than 55 Gy have shown to be linked with increased apy results in higher histologically complete response rates,
postoperative mortality and morbidity [30]. higher R0 resection rates and a lower frequency of lymph-node
metastases, without significantly affecting survival. In one of two
studies, postoperative mortality was increased after neoadjuvant
locally advanced disease (cT3–T4 or cN1-3 M0) CRT [35, 36].
Surgery alone is not a standard treatment in locally advanced Chemotherapy with cisplatin/5-FU combined with 41.4–
disease, since a complete (R0) tumour resection cannot be achieved 50.4 Gy in fractions of 1.8–2.0 Gy has long been the standard
in ∼30% (T3) to 50% (T4) of cases. Furthermore, even after com- treatment, but two recent randomised trials showed a favourable
plete tumour resection, long-term survival rarely exceeds 20%. Of toxicity profile for (bi)weekly combinations of oxaliplatin/5-FU
note, preoperative treatment (chemotherapy or CRT) has been or carboplatin/paclitaxel with RT [28, 31, 32].
shown to increase R0 resection and survival rates [22–25, 31, 32]. Even after complete tumour response to preoperative chemo
Therefore, preoperative treatment is clearly indicated in operable (radio)therapy, operable patients with AC should proceed to
patients with locally advanced oesophageal cancer [I, A]. surgery [IV, C].
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw329 | v

by guest
on 05 February 2018
management of advanced/metastatic human epidermal growth factor receptor 2 (HER2)-positive AC,

disease (M1) the recommendations of the ESMO gastric cancer guidelines
should be followed [38]. Consequently, HER2-positive metastat-
Patients with metastatic oesophageal cancer can be considered ic AC should be treated with a trastuzumab-containing regimen
for different options of palliative treatment depending on the [II, B]. In contrast, other biologically targeted drugs like the
clinical situation. Single-dose brachytherapy may be a preferred EGFR inhibitor gefitinib were not effective in post-progression
option even after external RT, since it provides better long-term treatment of oesophageal cancer [39].
relief of dysphagia with fewer complications than metal stent Response to neoadjuvant treatment is routinely assessed by
placement [I, B] [37]. the evaluation of tumour-related symptoms, endoscopy and CT
Chemotherapy is indicated for palliative treatment in selected scan. Patients with a curative treatment intention should be re-
patients, particularly for patients with AC who have a good PS ferred to surgery independently of the tumour response, except
[III, B]. Despite scarce evidence, treatment of advanced oe- in the case of metastatic disease. Usually, complete morpho-
sophageal AC is managed mostly according to the recommenda- logical responders should be operated in the case of AC, as the
tions for gastric cancer [38]. Newer regimens based on evidence for a watch-and-wait strategy is sparse for this histo-
oxaliplatin/fluoropyrimidine combinations are an alternative to logical subtype, whereas for SCC, the benefit/risk balance
the ‘classical’ cisplatin/5-FU schedule. Infusional 5-FU may be between surgery and close surveillance should be discussed.
replaced by capecitabine if the swallowing of tablets is not com- Tumour response to chemotherapy may be predicted early by
promised. Taxanes are recommended in first-line combinations FDG-PET in oesophageal and OGJ AC [III, C] [40]. However,
or as monotherapy in second-line therapy. at the present time, changing the therapeutic strategy according
In SCC, the value of palliative chemotherapy is less proved. to early response assessment is investigational. FDG-PET is not
Cisplatin-based combinations showed increased response rates relevant for evaluating tumour response after CRT, as it cannot
but no survival gain compared with monotherapy. Overall, reliably identify complete responders.
results with palliative chemotherapy are inferior to those in AC. A personalised medicine synopsis is given in Table 2.
Therefore, best supportive care (BSC) or palliative monotherapy
should also be considered [II, B].
follow-up, long-term implications and
personalised medicine survivorship
Randomised data with biologically targeted medical therapies Except for those patients who may be potential candidates for
are limited in oesophageal carcinoma. For treating patients with an endoscopic re-intervention or an early ‘salvage surgery’ after
Table 2. Personalised medicine synopsis table for lower oesophageal and gastric cancer
Biomarker Method Use LOE,
GOR
HER2 Immunohistochemistry for HER2 protein expression or ISH Used to select patients with metastatic disease for treatment II, B
for HER2 gene amplification with a trastuzumab-containing regimen
HER2, human epidermal growth factor receptor 2; ISH, in situ hybridisation; LOE, level of evidence; GOR, grade of recommendation
Table 3. Summary of recommendations
Diagnosis and pathology/molecular biology
All patients with new dysphagia, gastrointestinal bleeding, recurrent aspiration or emesis, weight loss and/or loss of appetite should undergo an upper
intestinal endoscopy [III, A].
Immunohistochemical stainings are recommended in poorly and undifferentiated cancers (G 3/4) according to WHO to differentiate between SCC and
AC of the oesophagus [V, B].
Staging and risk assessment
Decisions on the initial treatment approach of oesophageal cancer are taken on the basis of clinical staging, which should be carried out with the highest
degree of accuracy possible. Staging should include a complete clinical examination and a CT scan of the neck, chest and abdomen [III, A].
In candidates for surgical resection, EUS should be carried out to evaluate the T and N tumour categories [III, B].
18
F-FDG-PET should be carried out in patients who are candidates for oesophagectomy [III, B].
In the case of oesophageal SCC due to chronic tobacco and alcohol consumption, meticulous investigation of the oral cavity, oropharynx and
hypopharynx by an ear, nose and throat specialist, as well as trachea-bronchoscopy to exclude synchronous second cancers in the aerodigestive tract,
should be carried out [IV, B].
Continued
v | Lordick et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Table 3. Continued
In locally advanced (T3/T4) ACs of the OGJ infiltrating the anatomic cardia, laparoscopy can be done [IV, C].
The nutritional status and history of weight loss should be assessed according to the ESPEN guidelines [III, A].
Nutritional support according to the ESPEN guidelines is an integral part of the medical care for patients with oesophageal cancer in the curative and in
the palliative setting [II, A].
Management of local/locoregional disease
Upfront interdisciplinary planning of the treatment is mandatory [III, A].

Nutritional status matters and should be corrected. Endoscopic stenting should not be used in locoregional disease in operable patients and alternative
routes of feeding, e.g. with needle catheter jejunostomy, should be preferred [II, A].
Surgery is the treatment of choice in limited disease. In patients with T1a AC, endoscopic therapy is the preferred therapeutic approach, being both
effective and well tolerated [II, A].
In patients with superficial submucosal infiltration of an AC without further risk criteria (pT1sm1; <500 μm invasion, L0, V0, G1/2, <20 mm diameter,
no ulceration), endoscopic resection can be considered as an alternative to oesophagectomy [IV, B].
In the case of a high-grade intraepithelial neoplasia or a mucosal carcinoma (L0, V0, no ulceration, grading G1/G2, infiltration grade m1/m2) in the
squamous epithelium, an endoscopic en bloc resection should be carried out [III, A].
ESD should be preferred over endoscopic mucosa resection, especially in lesions >15 mm [II, B].
In T1/T2 N0 oesophageal cancer, radical and transthoracic oesophagectomy (Ivor-Lewis procedure) should be the surgical technique of choice [I, B].
Oesophagectomy should be done in high volume centres, with a lower rate of morbidity and better infrastructure to deal with complications following
major surgery, thereby preventing further mortality [I, A].
Surgery alone (without neoadjuvant treatment) should be recommended as the primary treatment approach for cT2N0 oesophageal cancer [II, B].
For patients unable or unwilling to undergo surgery, combined CRT is superior to RT alone [II, A].
Four courses of cisplatin/5-FU combined with radiation doses of 50.4 Gy in fractions of 1.8 Gy are regarded as standard for definitive CRT. Alternatively,
six cycles of FOLFOX can be given [I, C].
Preoperative treatment is indicated in operable patients with locally advanced oesophageal cancer (cT3–T4 or cN1–3 M0) [I, A].
Patients with locally advanced SCC benefit from preoperative chemotherapy or, most likely to a greater extent, from preoperative CRT, with higher rates
of complete tumour resection and better local tumour control and survival [I, A].
For patients with squamous cell oesophageal cancer, weekly administration of carboplatin (area under the curve of 2 mg/ml/min) and paclitaxel (50 mg/
m2) for 5 weeks and concurrent RT (41.4 Gy in 23 fractions, 5 days/week), followed by surgery, can be recommended as a contemporary standard of care
[I, A].
Neoadjuvant CRT with planned surgery or definitive CRT with close surveillance and salvage surgery for local tumour persistence or progression can be
considered as a recommended definitive treatment for locally advanced squamous cell cancer of the oesophagus [II, B].
Definitive CRT is recommended for cervically localised tumours [III, B].
For patients with oesophageal AC perioperative chemotherapy with regimens containing a platinum and a fluoropyrimidine for a duration of 8–9 weeks
in the preoperative phase (as well as 8–9 weeks in the postoperative phase, if feasible) or preoperative chemoradiotherapy (41.4–50.5 Gy) should be
considered standard in locally advanced AC of the oesophagus, including OGJ cancers [I, A].
Even after complete tumour response to preoperative chemo(radio)therapy operable patients with AC should proceed to surgery [IV, C].
Management of advanced/metastatic disease
Patients with metastatic oesophageal cancer can be considered for different options of palliative treatment depending on the clinical situation. Single-
dose brachytherapy may be a preferred option even after external RT, since it provides better long-term relief of dysphagia with fewer complications than
metal stent placement [I, B].
Chemotherapy is indicated for palliative treatment in selected patients, particularly for patients with AC who have a good PS [III, B].
In squamous cell oesophageal cancer, the value of palliative combination chemotherapy is less proved. Therefore, BSC or palliative monotherapy should
also be considered [II, B].
Personalised medicine
HER2-positive metastatic AC should be treated with a trastuzumab-containing treatment [II, B].

Tumour response to chemotherapy may be predicted early by 18F-FDG-PET in oesophageal and OGJ AC [III, C].
Follow-up, long-term implications and survivorship
Follow-up visits should be concentrated on symptoms, nutrition and psychosocial support [V, D].
In the case of complete response to CRT and no operation, a 3-month follow-up based on endoscopy, biopsies and CT scan may be recommended to
detect early recurrence leading to a discussion about salvage surgery [IV, B].
WHO, World Health Organization; SCC, squamous cell carcinoma; AC, adenocarcinoma; CT, computed tomography; EUS, endoscopic ultrasound; FDG-
PET, 18F-fluorodeoxyglucose positron emission tomography; OGJ, oesophago-gastric junction; ESPEN, European Society for Clinical Nutrition and
Metabolism; ESD, endoscopic submucosal dissection; CRT, chemoradiotherapy; 5-FU, 5-fluorouracil; FOLFOX, oxaliplatin/5-FU/folinic acid; RT,
radiotherapy; PS, performance status; BSC, best supportive care
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw329 | v

by guest
on 05 February 2018
Table 4. Levels of evidence and grades of recommendation (adapted from the Infectious Diseases Society of America–United States Public Health
Service Grading Systema)
Levels of evidence
I Evidence from at least one large randomised, controlled trial of good methodological quality (low potential for bias) or meta-analyses of well-
conducted randomised trials without heterogeneity
II Small randomised trials or large randomised trials with a suspicion of bias (lower methodological quality) or meta-analyses of such trials or of trials
with demonstrated heterogeneity
III Prospective cohort studies
IV Retrospective cohort studies or case–control studies
V Studies without control group, case reports, experts opinions
Grades of recommendation
A Strong evidence for efficacy with a substantial clinical benefit, strongly recommended
B Strong or moderate evidence for efficacy but with a limited clinical benefit, generally recommended
C Insufficient evidence for efficacy or benefit does not outweigh the risk or the disadvantages (adverse events, costs, ...), optional
D Moderate evidence against efficacy or for adverse outcome, generally not recommended
E Strong evidence against efficacy or for adverse outcome, never recommended
a
(failing) endoscopic resection or definitive CRT, there is no evi- Amgen, Biontech, Bristol-Myers Squibb, Eli Lilly, Ganymed,
dence that regular follow-up after initial therapy has an impact Merck-Serono, MSD, Nordic and Roche; travel support from
on survival outcomes. Amgen, Bayer, Roche and Taiho. CM has reported research
Therefore, follow-up visits should concentrate on symptoms, grants from Nestlé; lecture honoraria from Merck-Serono,
nutrition and psychosocial support [V, D]. Often, during the Nestlé, Roche and Sanofi; travel grants from Ethicon, Bard and
follow-up phase, a multidisciplinary care team is required, coor- Roche. RO has received lecture and advisory honoraria from
dinated by the physician who is seeing the patient on a regular Amgen, Roche, Eli Lilly and Nordic and has received travel
basis. Every patient will develop a variety of needs and problems, support from Merck, Bayer and Roche. DA has reported hon-
which are related to the new condition of life without an oe- oraria/consultancy for Roche, Merck-Serono, Bayer, Lilly and
sophagus or to other treatment sequelae or to psychosocial Servier; research support from Roche. KH has reported no po-
needs. The expertise of a dietician, a radiologist, a gastroenter- tential conflicts of interest.
ologist, a psychologist and a social worker is often needed
during follow-up.
In the case of complete response to CRT and no operation, a references
3-month follow-up based on endoscopy, biopsies and CT scan 1. Ferlay J, Steliarova-Foucher E, Lortet-Tieulent J et al. Cancer incidence and
may be recommended to detect early recurrence leading to a mortality patterns in Europe: estimates for 40 countries in 2012. Eur J Cancer
discussion about salvage surgery [IV, B] [28]. 2013; 49: 1374–1403.
2. Castro C, Bosetti C, Malvezzi M et al. Patterns and trends in esophageal cancer
mortality and incidence in Europe (1980–2011) and predictions to 2015. Ann
methodology Oncol 2014; 25: 283–290.
These clinical practice guidelines were developed in accordance 3. Rustgi AK, El-Serag HB. Esophageal carcinoma. N Engl J Med 2014; 371:
2499–2509.
with the ESMO standard operating procedures for clinical prac-
4. El-Serag HB, Hashmi A, Garcia J et al. Visceral abdominal obesity measured by CT
tice guidelines development, http://www.esmo.org/Guidelines/
scan is associated with an increased risk of Barrett’s oesophagus: a case-control
ESMO-Guidelines-Methodology. The relevant literature has study. Gut 2014; 63: 220–229.
been selected by the expert authors. A summary of recommen- 5. Shaheen NJ, Falk GW, Iyer PG, Gerson LB. ACG clinical guideline: diagnosis and
dations is shown in Table 3, and an overview of these recom- management of Barrett’s esophagus. Am J Gastroenterol 2016; 111: 30–50.
mendations related to therapy is shown in Figure 1. Levels of 6. Hamilton SR, Aaltonen LA (eds). World Health Organization Classification of
evidence and grades of recommendation have been applied Tumours. Pathology and Genetics of Tumours of the Digestive System. Lyon,
using the system shown in Table 4 [41]. Statements without France: IARC Press 2000.
grading were considered justified standard clinical practice by 7. Puli SR, Reddy JB, Bechtold ML et al. Staging accuracy of esophageal cancer by
endoscopic ultrasound: a meta-analysis and systematic review. World J
the experts and the ESMO faculty. This manuscript has been
Gastroenterol 2008; 14: 1479–1490.
subjected to an anonymous peer review process.
8. Edge SB, Byrd DR, Compton CC et al. (eds). AJCC Cancer Staging Manual, 7th
edition. New York, NY: Springer 2010.
conflict of interest 9. Kondrup J, Allison SP, Elia M et al. ESPEN guidelines for nutrition screening 2002.
Clin Nutr 2003; 22: 415–421.
FL has received research support from GlaxoSmithKline and 10. Weimann A, Braga M, Harsanyi L et al. ESPEN guidelines on enteral nutrition:
Fresenius Biotech; lecture and advisory honoraria from surgery including organ transplantation. Clin Nutr 2006; 25: 224–244.
v | Lordick et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
11. Mariette C, Gronnier C, Duhamel A et al. Self-expanding covered metallic stent as 27. Markar SR, Gronnier C, Pasquer A et al. Role of neoadjuvant treatment in clinical
a bridge to surgery in esophageal cancer: impact on oncologic outcomes. J Am T2N0M0 oesophageal cancer: results from a retrospective multi-center European
Coll Surg 2015; 220: 287–296. study. Eur J Cancer 2016; 56: 59–68.
12. Pech O, Bollschweiler E, Manner H et al. Comparison between endoscopic and 28. Conroy T, Galais MP, Raoul JL et al. Definitive chemoradiotherapy with FOLFOX
surgical resection of mucosal esophageal adenocarcinoma in Barrett’s esophagus versus fluorouracil and cisplatin in patients with oesophageal cancer (PRODIGE5/
at two high-volume centers. Ann Surg 2011; 254: 67–72. ACCORD17): final results of a randomised, phase 2/3 trial. Lancet Oncol 2014;
13. Cao Y, Liao C, Tan A et al. Meta-analysis of endoscopic submucosal dissection 15: 305–314.
versus endoscopic mucosal resection for tumors of the gastrointestinal tract. 29. Minsky BD, Pajak TF, Ginsberg RJ et al. INT 0123 (Radiation Therapy Oncology
Endoscopy 2009; 41: 751–757. Group 94–05) phase III trial of combined-modality therapy for esophageal cancer:
14. Hulscher JB, Van Sandick JW, De Boer AG et al. Extended transthoracic resection high-dose versus standard-dose radiation therapy. J Clin Oncol 2002; 20:
compared with limited transhiatal resection for adenocarcinoma of the esophagus. 1167–1174.
N Engl Med 2002; 347: 1662–1669. 30. Markar S, Gronnier C, Duhamel A et al. Salvage surgery after chemoradiotherapy
15. Mariette C, Piessen G, Briez N et al. Oesophagogastric junction adenocarcinoma: in the management of esophageal cancer: is it a viable therapeutic option? J Clin
which therapeutic approach? Lancet Oncol 2011; 12: 296–305. Oncol 2015; 33: 3866–3873.
16. Mariette C, Piessen G. Oesophageal cancer: how radical should surgery be? Eur J 31. van Hagen P, Hulshof MC, van Lanschot JJ et al. Preoperative chemoradiotherapy
Surg Oncol 2012; 38: 210–213. for esophageal or junctional cancer. N Engl J Med 2012; 366: 2074–2084.
17. Biere SS, van Berge Henegouwen MI, Maas KW et al. Minimally invasive versus 32. Shapiro J, van Lanschot JJ, Hulshof MC et al. Neoadjuvant chemoradiotherapy
open oesophagectomy for patients with oesophageal cancer: a multicentre, open- plus surgery versus surgery alone for oesophageal or junctional cancer (CROSS):
label, randomised controlled trial. Lancet 2012; 379: 1887–1892. long-term results of a randomised controlled trial. Lancet Oncol 2015; 16:
18. Mariette C, Meunier B, Pezet D et al. Hybrid minimally invasive versus open 1090–1098.
oesophagectomy for patients with oesophageal cancer: a multicenter, open-label, 33. Stahl M, Stuschke M, Lehmann N et al. Chemoradiation with and without surgery
randomized phase III controlled trial, the MIRO trial. J Clin Oncol 2015; 33 2015; in patients with locally advanced squamous cell carcinoma of the esophagus. J
33 (January 20 Suppl.): abstr 5. Clin Oncol 2005; 23: 2310–2317.
19. Birkmeyer JD, Siewers AE, Finlayson EV et al. Hospital volume and surgical 34. Bedenne L, Michel P, Bouché O et al. Chemoradiation followed by surgery
mortality in the United States. N Engl J Med 2002; 346: 1128–1137. compared with chemoradiation alone in squamous cancer of the esophagus: FFCD
20. Markar SR, Karthikesalingam A, Thrumurthy S, Low DE. Volume-outcome 9102. J Clin Oncol 2007; 25: 1160–1168.
relationship in surgery for esophageal malignancy: systematic review and meta- 35. Stahl M, Walz MK, Stuschke M et al. Phase III comparison of preoperative
analysis 2000–2011. J Gastrointest Surg 2012; 16: 1055–1063. chemotherapy compared with chemoradiotherapy in patients with locally advanced
21. Brusselaers N, Mattsson F, Lagergren J. Hospital and surgeon volume in relation to adenocarcinoma of the esophagogastric junction. J Clin Oncol 2009; 27:
long-term survival after oesophagectomy: systematic review and meta-analysis. 851–856.
Gut 2014; 63: 1393–1400. 36. Klevebro F, Alexandersson von Döbeln G, Wang N et al. A randomized clinical trial
22. Sjoquist KM, Burmeister BH, Smithers BM et al. Survival after neoadjuvant of neoadjuvant chemotherapy versus neoadjuvant chemoradiotherapy for cancer of
chemotherapy or chemoradiotherapy for resectable oesophageal carcinoma: an the oesophagus or gastro-oesophageal junction. Ann Oncol 2016; 27: 660–667.
updated meta-analysis. Lancet Oncol 2011; 12: 681–692. 37. Homs MY, Steyerberg EW, Eijkenboom WM et al. Single-dose brachytherapy
23. Kranzfelder M, Schuster T, Geinitz H et al. Meta-analysis of neoadjuvant treatment versus metal stent placement for the palliation of dysphagia from oesophageal
modalities and definitive non-surgical therapy for oesophageal squamous cell cancer: multicentre randomised trial. Lancet 2004; 364: 1497–1504.
cancer. Br J Surg 2011; 98: 768–783. 38. Smyth EC, Verheij M, Allum W et al. Gastric cancer: ESMO Clinical Practice
24. Ronellenfitsch U, Schwarzbach M, Hofheinz R et al. Preoperative chemo(radio) Guidelines for diagnosis, treatment and follow-up. Ann Oncol 2016; 27 (Suppl 5):
therapy versus primary surgery for gastroesophageal adenocarcinoma: systematic v38–v49.
review with meta-analysis combining individual patient and aggregate data. Eur J 39. Dutton SJ, Ferry DR, Blazeby JM et al. Gefitinib for oesophageal cancer
Cancer 2013; 49: 3149–3158. progressing after chemotherapy (COG): a phase 3, multicentre, double-blind,
25. Allum WH, Stenning SP, Bancewicz J et al. Long-term results of a randomized trial placebo-controlled randomised trial. Lancet Oncol 2014; 15: 894–904.
of surgery with or without preoperative chemotherapy in esophageal cancer. J Clin 40. Lordick F, Ott K, Krause BJ et al. PET to assess early metabolic response and to
Oncol 2009; 27: 5062–5067. guide treatment of adenocarcinoma of the oesophagogastric junction: the
26. Mariette C, Dahan L, Mornex F et al. Surgery alone versus chemoradiotherapy MUNICON phase II trial. Lancet Oncol 2007; 8: 797–805.
followed by surgery for stage I and II esophageal cancer: final analysis of 41. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
randomized controlled phase III trial FFCD 9901. J Clin Oncol 2014; 32: among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
2416–2422. 139–144.
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw329 | v

by guest
on 05 February 2018
Gastric cancer: ESMO Clinical Practice Guidelines for

diagnosis, treatment and follow-up†
E. C. Smyth1, M. Verheij2, W. Allum3, D. Cunningham4, A. Cervantes5 & D. Arnold6 on behalf of the
ESMO Guidelines Committee*
1
Department of Gastrointestinal Oncology, Royal Marsden Hospital, London and Surrey, UK; 2Department of Radiation Oncology, The Netherlands Cancer Institute, Antoni
van Leeuwenhoek Hospital, Amsterdam, The Netherlands; 3Department of Surgery, Royal Marsden Hospital, London and Surrey; 4Department of Medicine, Royal Marsden
Hospital, London and Surrey, UK; 5Medical Oncology Department, INCLIVA University of Valencia, Valencia, Spain; 6Instituto CUF de Oncologia (I.C.O.), Lisbon, Portugal.
incidence and epidemiology suspected, referral to a geneticist for assessment is recom-

mended based on international clinical guidelines [V, B] [7].
Almost one million (951 600) new cases of gastric cancer were
diagnosed globally in 2012, resulting in ∼723 100 deaths [1]. Of
these ∼140 000 cases and ∼107 000 deaths occurred in Europe diagnosis and pathology
[2]. Gastric cancer displays significant global variation in inci- Recommendation: Diagnosis should be made from a gastroscopic
dence; the highest rates are seen in Eastern Asia, Eastern Europe or surgical biopsy reviewed by an experienced pathologist, and
and South America, with lower rates in North America and histology should be reported according to the World Health
clinical practice
Western Europe. A gradual decline in the incidence of gastric Organisation (WHO) criteria [IV, C].
guidelines
cancer has been observed in Western Europe and North Patients in Asian countries are frequently diagnosed with
America over the past 60 years and more recent declines in gastric cancer at an earlier stage than in non-Asian countries.
high-prevalence countries have also become apparent. This is In Japan and Korea, where the incidence of gastric cancer is
epidemiologically distinct from the relative increase in tumours much higher than in Western countries, screening for gastric
of the gastroesophageal junction, which are discussed in a separ- cancer is routine. In patients who develop symptoms from an
ate guideline document. underlying gastric cancer, these commonly include weight loss,
Risk factors for gastric cancer include male gender (incidence dysphagia, dyspepsia, vomiting, early satiety and/or iron defi-
is twice as high), Helicobacter pylori infection, tobacco use, atro- ciency anaemia.
phic gastritis, partial gastrectomy and Ménétrier’s disease [3]. Ninety per cent of gastric cancers are adenocarcinomas
Regional variation in gastric cancer risk factors influences the (ACs), and these are subdivided according to histological
most common anatomical subsites of disease. Distal or antral appearances into diffuse (undifferentiated) and intestinal
gastric cancers that are associated with H. pylori infection, (well-differentiated) types (Lauren classification). Recent large-
alcohol use, high-salt diet, processed meat and low fruit and scale studies in molecular subtyping have defined four sub-
vegetable intake are more common in East Asia. Tumours of the types of gastric cancer across genomic, transcriptomic and
proximal stomach (cardia) are associated with obesity, and proteomic levels; however, these subtypes do not yet have any
tumours of the gastroesophageal junction are associated with impact on treatment [8]. These Clinical Practice Guidelines do
reflux and Barrett’s oesophagus and are more common in non- not apply to rarer gastric malignancies such as gastrointestinal
Asian countries [4]. Gastric cancer demonstrates familial aggre- stromal tumours (GISTs), lymphomas and neuroendocrine
gation in ∼10% of cases, and an inherited genetic predisposition tumours.
is found in a small proportion of cases (∼1%–3%); relevant syn- If a diagnosis of gastric cancer is suspected, diagnosis should
dromes include hereditary non-polyposis colorectal cancer, fa- be made from a gastroscopic or surgical biopsy reviewed by an
milial adenomatous polyposis colorectal cancer, hereditary experienced pathologist, and histology should be reported
diffuse gastric cancer (HDGC), gastric adenocarcinoma and according to the WHO criteria [IV, C].
proximal polyposis of the stomach (GAPPS) and Peutz Jegher’s
syndrome [5, 6]. If a familial cancer syndrome such as HDGC is
staging and risk assessment
Recommendation: Initial staging and risk assessment should
6962 Viganello-Lugano, Switzerland. include physical examination, blood count and differential, liver
E-mail: clinicalguidelines@esmo.org and renal function tests, endoscopy and contrast-enhanced com-
†
Approved by the ESMO Guidelines Committee: August 2016.
puted tomography (CT) scan of the thorax, abdomen ± pelvis
(Table 1) [V, A]. Laparoscopy is recommended for patients with

by guest
on 05 February 2018
Table 1. Diagnostic and staging investigations in gastric cancer The TNM stage should be recorded according to the latest
edition of the American Joint Committee on Cancer (AJCC)/
Procedure Purpose
Union for International Cancer Control (UICC) guidelines and
Full blood count Assess for iron deficiency anaemia staging manual [17, 18] (Tables 2 and 3).
Renal and liver function Assess renal and liver function to
determine appropriate therapeutic
options treatment planning
Endoscopy and biopsy Obtain tissue for diagnosis, histological
Multidisciplinary treatment planning before any treatment deci-
classification and molecular
sion is mandatory. The core membership of the multidisciplin-
biomarkers, e.g. HER2 status
ary team should include surgeons, medical and radiation
CT thorax + abdomen ± pelvis Staging of tumour—to detect local/
distant lymphadenopathy and
oncologists, radiologists and pathologists, with other members
metastatic disease or ascites as available [IV, C].
EUS Accurate assessment of T and N stage
in potentially operable tumours
management of local/locoregional
Determine the proximal and distal
extent of tumour disease
Laparoscopy ± washings Exclude occult metastatic disease Recommendation: Endoscopic resection is appropriate for selected
involving peritoneum/diaphragm very early tumours [III, B]. For stage IB–III gastric cancer, radical
PET, if available May improve detection of occult gastrectomy is indicated and perioperative therapy is recom-
metastatic disease in some cases
mended for these patients [I, A]. Medically fit patients should
undergo D2 resections in high-volume surgical centres [I, B].
CT, computed tomography; EUS, endoscopic ultrasound; PET, positron
emission tomography
surgery
Surgical resection of gastric cancer, specifically at early stages, is
potentially curative. However, the majority of patients still
relapse following resection, and therefore, combined modality
resectable gastric cancer [III, B]. Multidisciplinary treatment therapies are standard for ≥ Stage IB disease.
planning before any treatment is mandatory [IV, C]. The extent of resection is determined by the preoperative
Careful tumour staging is essential to ensure that patients are stage.
appropriately selected for treatment interventions. The recom- Endoscopic resection may be carried out for very early gastric
mended initial staging investigations are detailed in Table 1. cancers (T1a) if they are clearly confined to the mucosa, well-
Identification of malignant lymph nodes on CT: The follow- differentiated, ≤2 cm and non-ulcerated [III, B]. The associated
ing characteristics are frequently demonstrated in malignant lymph node metastatic risk in this group is virtually zero. Two
lymph nodes detected on CT: forms of endoscopic resection are practised; endoscopic
1) Short-axis diameter 6–8 mm in perigastric lymph nodes; mucosal resection (EMR) is acceptable for lesions smaller than
2) round shape; 10 – 15 mm with a very low probability of advanced histology
3) central necrosis and (Paris 0–IIa) [19]. However, European Society of
4) heterogeneous or high enhancement [9–11]. Gastrointestinal Endoscopy Guidelines recommend endoscopic
submucosal dissection (ESD) as the treatment of choice for
However, the sensitivity of CT for lymph node staging is vari- most gastric superficial neoplastic lesions [IV, B] [19].
able (62.5%–91.9% on systematic review), and global consensus T1 tumours that do not meet the above mentioned criteria for
is lacking on specific diagnostic criteria [12]. endoscopic resection require surgery, although less extensive
Endoscopic ultrasound (EUS) is helpful in determining the surgery than other gastric cancers (see below). Lymph node dis-
proximal and distal extent of the tumour and provides further section for T1 tumours may be confined to perigastric lymph
assessment of the T and N stage; however, it is less useful in nodes and include local N2 nodes (D1+, with variation in nodal
antral tumours [III, B]. EUS is more consistently accurate than groups dissected according to the site of cancer). Sentinel lymph
CT for the diagnosis of malignant lymph nodes: patterns asso- node mapping may further modify these approaches.
ciated with malignancy on EUS include hypoechogenicity, For stage IB–III gastric cancer, radical gastrectomy is indicated.
round shape, smooth, distinct margin and size >1 cm [13, 14]. Subtotal gastrectomy may be carried out if a macroscopic prox-
Positron emission tomography (PET)-CT imaging may improve imal margin of 5 cm can be achieved between the tumour and the
staging by detecting involved lymph nodes or metastatic disease. gastroesophageal junction. For diffuse cancers, a margin of 8 cm
However, PET may not be informative in patients with mucin- is advocated. Otherwise, a total gastrectomy is indicated [III, A].
ous or diffuse tumours [III, B]. Perioperative therapy is recommended for these patients.
Laparoscopy ± peritoneal washings for malignant cells is The extent of nodal dissection accompanying radical gastrec-
recommended in all stage IB–III gastric cancers which are con- tomy has been extensively debated. D1 resection implies the
sidered potentially resectable, to exclude radiologically occult removal of the perigastric lymph nodes and D2 implies removal
metastatic disease; the benefit may be greater for patients with of perigastric lymph nodes plus those along the left gastric,
T3/T4 disease [III, B] [15, 16]. common hepatic and splenic arteries and the coeliac axis (see
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw350 | v

by guest
on 05 February 2018
Table 2. TNM staging of gastric cancer as per AJCC, 7th edition [17, 18]
Primary tumour (T) Regional lymph nodes (N) Distant metastasis (M)
TX Primary tumour cannot be assessed NX Regional lymph node(s) M0 No distant metastasis

cannot be assessed
T0 No evidence of primary tumour N0 No regional lymph node M1 Distant metastasis or positive
metastasis peritoneal cytology
Tis Carcinoma in situ: intraepithelial tumour without invasion of N1 Metastasis in 1–2 regional
the lamina propria lymph nodes
T1a Tumour invades the lamina propria or the muscularis mucosae N2 Metastasis in 3–6 regional
lymph nodes
T1b Tumour invades the submucosa N3 Metastasis in 7 or more
regional lymph nodes
T2 Tumour invades the muscularis propria N3a Metastasis in 7–15 regional
lymph nodes
T3 Tumour penetrates the subserosal connective tissue without N3b Metastasis in 16 or more
invasion of the visceral peritoneum or adjacent structuresa regional lymph nodes
T4 Tumour invades the serosa (visceral peritoneum) or adjacent
structuresb
T4a Tumour invades the serosa (visceral peritoneum)
T4b Tumour invades adjacent structuresb
Edge et al. [18]. Used with the permission of the American Joint Committee on Cancer (AJCC), Chicago, IL, USA. The original source for this material is
the AJCC Cancer Staging Handbook, 7th edition (2010) published by Springer Science and Business Media LLC, www.springer.com.
a
T3 tumours also include those extending into the gastrocolic or gastrohepatic ligaments, or into the greater or lesser omentum, without perforation of the
visceral peritoneum covering these structures.
b
Adjacent structures include the spleen, transverse colon, liver, diaphragm, pancreas, abdominal wall, adrenal gland, kidney, small intestine and retro-
peritoneum.
Table 3. Anatomic stage/prognostic groups as per AJCC, 7th

edition [17, 18]
Figure 1) [20]. The current UICC/AJCC TNM (seventh edition)
classification recommends excision of a minimum of 15 lymph
Stage grouping T stage N stage M stage
nodes to allow reliable staging. In Asian countries, experience
Stage 0 Tis N0 M0 from observational and randomised trials demonstrates that D2
Stage IA T1 N0 M0 dissection leads to superior outcomes compared with D1 resec-
Stage IB T2 N0 M0 tion [II, B]. In Western countries, a Dutch [21], an MRC [22]
T1 N1 M0 and a recent Italian [23] trial failed to demonstrate any initial
Stage IIA T3 N0 M0 survival advantage with D2 resection, although the Italian study
T2 N1 M0 suggested a trend towards a benefit in disease-specific survival
T1 N2 M0 for patients with T2–T4 lymph node-positive cancers treated
Stage IIB T4a N0 M0 with D2 resection [23]. Long-term (15-year) follow-up from the
T3 N1 M0 Dutch trial demonstrated fewer locoregional recurrences and
T2 N2 M0 gastric cancer-related deaths with D2 resection; however, this
T1 N3 M0
was offset slightly by an increase in postoperative mortality and
Stage IIIA T4a N1 M0
morbidity [24]. A recent review of the quality of lymph node
T3 N2 M0
dissection in the same study also suggests that non-compliance
T2 N3 M0
in the D2 resection group may have obscured a significant dif-
Stage IIIB T4b N0–1 M0
ference in survival between the randomised groups; this has also
T4a N2 M0
T3a N3 M0
been suggested for the recent Italian study [23].
Stage IIIC T4b N2–3 M0 Consensus opinion is that, in Western countries, medically fit
T4a N3 M0 patients should undergo D2 dissection that is carried out in spe-
Stage IV Any T Any N M1 cialised, high-volume centres with appropriate surgical expertise
and postoperative care [I, B] [25–27]. As a result, perioperative
Edge et al. [18]. Used with the permission of the American Joint outcome has become standardised with morbidity and mortality
Committee on Cancer (AJCC), Chicago, IL, USA. The original source rates of 15% and 3.0%, respectively [23, 28]. The concept of
for this material is the AJCC Cancer Staging Handbook, 7th edition ‘enhanced recovery’ encompasses all aspects of optimal peri-
(2010) published by Springer Science and Business Media LLC, www. operative care for the patient undergoing gastrectomy; guidance
springer.com. is provided by relevant Enhanced Recovery After Surgery
(ERAS®) Society guidelines on this topic [29].
v | Smyth et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Gastric cancer
(Adenorcarcinoma)
Operable Operable Inoperable or

stage T1NO stage >T1 NO metastatic
Preferred pathway
Re-assess
Consider endoscopic/ Preoperative Palliative Best supportive care if
limited resection Surgery chemotherapy unfit for treatment
chemotherapy
Adjuvant Adjuvant HER2-negative: Consider clinical

Surgery HER2-positive:
chemoradiotherapy chemotherapy trials of novel agents
Platinum+
Trastuzumab
fluoropyrimidine-based
+ CF/CX
doublet or triplet regimen
Second-line
Postoperative chemotherapy
chemotherapy
Figure 1. Gastric cancer treatment algorithm.

HER2, human epidermal growth factor receptor 2; CF, cisplatin and 5-fluorouracil; CX, cisplatin and capecitabine
Laparoscopic surgery has the potential benefit of decreased evidence-based for both tumour subsites. An EORTC study in
postoperative morbidity and reduced recovery time. Although which patients were randomised to surgery plus or minus bi-
concerns existed regarding the possibility of a reduced nodal weekly cisplatin (50 mg/m2) and 5-FU in the de Gramont style
harvest with a laparoscopic approach, a recent meta-analysis also increased R0 resection rates in chemotherapy-treated
suggests that lymph node yields are comparable for both patients but closed early due to poor accrual and is not powered
approaches [30]. Trials from the Far East have reported equiva- to show a survival benefit [35]. Perioperative chemotherapy has
lent results to open surgery for distal gastrectomy, but there therefore been widely adopted as a standard of care throughout
remain some technical issues particularly for anastomosis for many parts of Europe [I, A]. Since capecitabine avoids the need
total gastrectomy [31, 32]. Laparoscopic surgery is becoming for an indwelling central venous access device and is non-inferior
one of the recommended options for patients with early gastric to 5-FU in the advanced disease setting [36], capecitabine-con-
cancer; however, it remains to be shown whether laparoscopic taining regimens can also be suggested in the perioperative
surgery can achieve the same results as open surgery in gastric setting (as ECX: epirubicin, cisplatin, capecitabine, in preference
cancers requiring D2 lymphadenectomy. It may be that as tech- to ECF) [IV, C]. Also, other platinum/fluoropyrimidine doublets
niques predicting lymph node involvement develop, those with or triplets may be considered; in particular, oxaliplatin may
negative nodes should be operated laparoscopically, whereas replace cisplatin [as EOX (epirubicin, oxaliplatin, capecitabine)];
those with predicted positive nodes would require open surgery. it is non-inferior to ECX in the metastatic setting [36]).
The effect of dose intensification (e.g. with taxanes) of peri-
operative chemotherapy in gastric cancer remains unclear. In oe-
perioperative chemotherapy sophageal and gastroesophageal junctional AC, intensification of
Recommendation: Perioperative (pre- and postoperative) chemo- preoperative chemotherapy from two cycles of cisplatin and cape-
therapy with a platinum/fluoropyrimidine combination is recom- citabine (CX) to four cycles of ECX resulted in improved patho-
mended for patients with ≥Stage IB resectable gastric cancer [I, A]. logical response rates (secondary end point), but this did not
The UK MRC MAGIC trial demonstrated an improvement in translate into an improvement in overall survival (OS) [37].
5-year survival from 23% to 36% for patients with resectable However, as this trial did not include gastric cancer patients or a
stage II and III gastric cancers treated with six cycles (three pre- postoperative chemotherapy component, direct cross-trial com-
and three postoperative) of perioperative ECF chemotherapy parisons are challenging. A study of the German AIO study group
[epirubicin, cisplatin and 5-fluorouracil (5-FU)] compared with investigating a perioperative FLOT regimen (fluorouracil, leucov-
surgery alone [33]. A subsequent French trial has reported orin, oxaliplatin, docetaxel) versus ECF/X demonstrated higher
similar results with the use of a 28-day regimen of perioperative rates of pathological response for FLOT (15.6% versus 5.8%);
cisplatin and 5-FU [34]. The MAGIC trial recruited predomin- however, correlation with survival outcomes is awaited [38].
antly patients with gastric cancers, whereas the French study was Based on these studies, it may be reasonable to use any fluoro-
composed of a majority of patients with proximal tumours. pyrimidine–platinum doublet or triplet before surgery, although
Therefore, a perioperative treatment approach may be considered the strongest evidence is for cisplatin/fluorouracil ± epirubicin
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw350 | v

by guest
on 05 February 2018
combinations. Recommended treatment duration is 2–3 months. single-agent, adjuvant S1 in a two by two randomised factorial
There is no current evidence to support the use of perioperative trial [52].
trastuzumab therapy or any other biologically targeted drug, in- Historically, a greater benefit has been noted with adjuvant
cluding anti-angiogenic compounds. chemotherapy in Asian studies, and uptake of adjuvant chemo-
therapy in Europe for patients with resected gastric cancer
adjuvant treatment remains limited due to a perceived lack of benefit and routine
use of perioperative chemotherapy. However, a large individual
Recommendation: For patients with ≥Stage IB gastric cancer who patient-level meta-analysis of adjuvant chemotherapy in gastric
have undergone surgery without administration of preoperative cancer has confirmed a 6% absolute benefit for 5-FU-based
chemotherapy (e.g. due to understaging before the initial decision chemotherapy, compared with surgery alone [hazard ratio (HR)
for upfront surgery), postoperative chemoradiotherapy (CRT) or 0.82, 95% confidence interval (CI) 0.76–0.90; P < 0.001] in all
adjuvant chemotherapy is recommended [I, A]. For patients subgroups tested [I, A] [53]. However, as adjuvant chemother-
having undergone preoperative chemotherapy, the addition of apy is also less well tolerated than neoadjuvant chemotherapy, a
postoperative radiotherapy (RT) has no added benefit. perioperative approach is preferred if possible, so that more
patients can benefit from systemic treatment even if the post-
chemoradiotherapy. The North American Intergroup-0116 operative component of treatment is unable to be delivered.
trial demonstrated that adjuvant therapy with 5-FU/leucovorin For adjuvant treatment following preoperative chemotherapy,
(Q28) plus conventionally fractionated RT (45 Gy in 25 the preoperatively chosen regimen should be completed after re-
fractions) resulted in improved OS years compared with surgery section for patients who are fit for treatment, independent from
alone, (50% 3-year survival for patients treated with CRT versus pathohistological findings and considerations. The addition of
41% for those treated with surgery alone [39]). After 10 years of postoperative RT has been shown to not improve survival in
follow-up, this OS improvement remains significant [I, A] [40]. patients receiving chemotherapy before and after curative-intent
Therefore, this treatment approach is currently considered as surgery. Recently, the randomised, phase III CRITICS trial con-
standard therapy in the USA, though it has not gained wide cluded that patients undergoing chemotherapy followed by
acceptance in Europe due to concerns about potential late toxic surgery with curative intent had similar OS and progression-free
effects and the quality of surgery within the trial. Moreover, survival (PFS) regardless of whether they received chemother-
>50% of patients underwent inadequate (less than D1) apy or CRT after surgery [54].
lymphadenectomy, suggesting that postoperative CRT may be
(mainly) compensating for suboptimal surgery [II, B]. This is management of advanced/metastatic
supported by retrospective data from the Dutch D1D2 trial, disease
demonstrating that CRT reduces local recurrence rates following
D1 resection, but provides no benefit in patients who have first-line treatment
undergone D2 resection [41]. However, other randomised and Recommendation: Doublet or triplet platinum/fluoropyrimidine
non-randomised data suggest potential benefits from combinations are recommended for fit patients with advanced
postoperative CRT even after optimal D2 dissection [I, B] [42– gastric cancer [I, A].
45], and this is the subject of ongoing randomised trials. Patients with inoperable locally advanced and/or metastatic
Regarding patients who have had a microscopically incomplete (stage IV) disease should be considered for systemic treatment
resection, a retrospective comparison of the Dutch D1D2 trial (chemotherapy), which has shown improved survival and
has suggested significant improvements in OS and local recur- quality of life compared with best supportive care alone [I, A]
rence rates with use of CRT after an R1 resection, a finding that [55–57]. However, co-morbidities, organ function and perform-
has been confirmed by other retrospective series [IV, B] [41, 46]. ance status (PS) must always be taken into consideration [II, B].
In current postoperative CRT regimens, RT should preferably In general, resection of the primary tumour is not recommended
be given as a concomitant regimen of fluoropyrimidine-based in the palliative setting; however, a small number of patients with
CRT to a total dose of 45 Gy in 25 fractions of 1.8 Gy, 5 frac- initially unresectable locally advanced disease may be deemed op-
tions/week by intensity-modulated RT techniques [IV, A] [47]. erable following a good response to systemic therapy.
The clinical target volume encompasses the gastric bed (with Response to systemic treatments should normally be assessed
stomach remnant when present), anastomoses and draining re- with interval imaging of the chest, abdomen and pelvis, mostly
gional lymph nodes [I, B] [44, 45]. with CT, although alternative imaging techniques may be used
if required to monitor known sites of disease (e.g. magnetic res-
adjuvant chemotherapy. The ACTS-GC trial evaluating adjuvant onance imaging for bone lesions).
chemotherapy with S-1 following D2 resection in Asian patients Doublet combinations of platinum and fluoropyrimidines are
demonstrated an OS benefit for patients treated with adjuvant generally used, and there remains controversy regarding the utility
chemotherapy [I, A] [48, 49]. The CLASSIC trial evaluated a of triplet regimens. However, a meta-analysis has demonstrated sig-
capecitabine–oxaliplatin doublet in a similar population and this nificant benefit from the addition of an anthracycline to a platinum
was associated with significantly improved OS and disease-free and fluoropyrimidine doublet [58]. For anthracycline-based
survival (DFS) [50, 51] compared with surgery alone. triplets, the UK REAL-2 trial demonstrated non-inferiority
However, none of sequential S1-paclitaxel, sequential tegafur between ECF, ECX, EOF (epirubicin, oxaliplatin, 5-FU) and EOX.
and uracil (UFT)-paclitaxel or UFT alone resulted in a superior The EOX regimen was associated with numerically longer median
outcome (DFS as the primary end point) when compared with OS (11.2 versus 9.9 months, HR 0.80, 95% CI 0.66–0.97; P = 0.02)
v | Smyth et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
than ECF, without the need for an indwelling catheter and with chemotherapy as age. Geriatric assessment may be helpful before
reduced rates of thromboembolism [59]. Additionally, a combined initiating treatment of older patients.
analysis has demonstrated that capecitabine is associated with
improved OS, compared with infused 5-FU within doublet and second- and further-line treatment
triplet regimens [I, A] [58].
Triplets containing taxanes are also an evidence-based treat- Recommendation: Second-line chemotherapy with a taxane (doc-
ment choice for first-line chemotherapy [60, 61]. In a phase III etaxel, paclitaxel), or irinotecan, or ramucirumab as single agent
randomised trial, the addition of docetaxel to 5-FU/cisplatin in or in combination with paclitaxel is recommended for patients
a 3-weekly regimen (DCF) was associated with improved OS, who are of PS 0–1 [I, A].
In patients of adequate PS, second-line treatment is associated
but also added significant toxic effects including increased rates
with proven improvements in OS and quality of life compared
of febrile neutropaenia [I, C] [61]. A phase II randomised control
trial of a modified 2-weekly DCF regimen compared with original with best supportive care, with treatment options including iri-
DCF (with growth factor support) halted recruitment in the origin- notecan, docetaxel or paclitaxel, if not used before [I, A] [72–
al DCF arm due to excessive toxicity; however, the study suggested 75]. A randomised phase III trial directly comparing weekly
that a substantially modified DCF regimen was tolerable and effect- paclitaxel with irinotecan has demonstrated similar efficacy for
both regimens [I, A] [76].
ive [II, B] [62]. Several other studies have examined the efficacy of
The anti-VEGFR-2 monoclonal antibody ramucirumab has
docetaxel, fluoropyrimidine and oxaliplatin-containing regimens.
The FLOT regimen (fluorouracil, leucovorin, oxaliplatin and doce- shown activity in two randomised phase III trials. As a single
taxel) resulted in a median PFS of 5.1 months and a median OS of agent it is associated with a survival benefit comparable to cyto-
11 months in a small non-randomised study [63]. An almost iden- toxic chemotherapy in the second-line setting [I, A], whereas
tical regimen used in a randomised phase II trial resulted in en- ramucirumab in addition to paclitaxel is associated with a sur-
couraging median PFS and OS of 7.7 and 14.6 months, vival benefit compared with paclitaxel alone [I, A] [77, 78].
Alternatively, in patients with disease progression >3 months fol-
respectively [II, B] [64]. As an alternative to platinum-based
lowing first-line chemotherapy, it may be appropriate to consider a
therapy, irinotecan plus leucovorin and infusional 5-FU (FOLFIRI)
has been studied in both phase II trials and one phase III rando- rechallenge with the same drug combination as an additional treat-
mised trial in the first-line setting, and may be considered for ment option [IV, C] [79]. In patients with symptomatic, locally
selected patients [60, 65]. advanced or recurrent disease, hypo-fractionated RT is an effective
See the metastasectomy section for surgery in stage IV gastric and well-tolerated treatment modality that may palliate bleeding,
obstructive symptoms or pain [III, B] [80].
cancer.
Treatment options may be used sequentially in second and
third line, but there is no clear evidence for a benefit beyond
elderly patients with gastric cancer second line treatment.
Elderly patients with gastric cancer are under-represented in clin-
ical trials, and there are few randomised data in this setting. personalised medicine and targeted
Regimens that have been specifically addressed in phase II trials
in elderly patients with comparable survival results include cape-
therapy
citabine and oxaliplatin, FOLFOX (leucovorin, 5-FU and oxali- Recommendation: Trastuzumab is recommended in conjunction
platin), single-agent capecitabine and S1 (in Asian patients) [III, with platinum and fluoropyrimidine-based chemotherapy for
B] [66–68]. A phase II trial investigated the miniDOX (docetaxel, patients with HER2-positive advanced gastric cancer [I, A].
oxaliplatin and capecitabine) regimen in primarily older patients Gastric cancers have been demonstrated to be highly molecu-
and also recruited patients with other poor prognostic markers larly diverse and may be driven by a number of different genetic
(PS2, weight loss 10%–25%). This regimen was associated with and epigenetic abnormalities. Four subtypes of gastric cancer
survival comparable to good prognostic groups; however, toxicity have recently been described in The Cancer Genome Atlas;
was also prominent [69]. In addition, a meta-analysis of three these are the Epstein Barr Virus, microsatellite instability (MSI)
phase III trials comparing patients ≥70 years with younger high, genomically stable and chromosomal instability (CIN)
patients demonstrated no differences in response rates or OS subtypes [8]. Each subtype is enriched for selected molecular ab-
between the two patient groups [II, B] [70]. In the perioperative normalities, with some overlap. In particular, the CIN subtype
setting, a German study compared the doublet FLO (infusional is enriched for copy number changes in key receptor tyrosine
5-FU, leucovorin and oxaliplatin) and the FLOT regimens in 44 kinase oncogenes such as HER2, EGFR, FGFR2 and MET.
elderly gastric cancer patients [71]. Although the FLOT regimen These findings have potentially important therapeutic implica-
was associated with a trend towards improved PFS, it was also tions as oncologists attempt to target the key pathways driving
associated with increased toxicity [II, B]. Furthermore, in the the tumour in each individual patient.
MAGIC trial, there was no evidence of heterogeneity of treatment In HER2-positive gastric cancer (10%–15% of cases), the
effect for patients over the age of 70 [33]. However, it must be phase III ToGA trial demonstrated clinically and statistically sig-
considered that all patients included in these analyses were clinic- nificant improvements in response rate, PFS and OS with the
al trial participants, which may not reflect patients treated in a addition of trastuzumab to a cisplatin/fluoropyrimidine doublet
community setting. When making a decision regarding chemo- (median OS 13.8 versus 11.1 months, HR 0.74, 95% CI 0.60–
therapy, the functional age of the patient must also be considered, 0.91; P = 0.0048) [I, A] [81]. The benefits of trastuzumab were
as co-morbidities and PS may have an equal effect on tolerance of even more marked in the traditionally defined HER2-positive
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw350 | v

by guest
on 05 February 2018
subgroup with immunohistochemistry (IHC) 2+/FISH+ advanced peritoneal metastases, data from randomised trials in
tumours, or IHC 3+ tumours. In these patients, the median OS Asia also exist to support the use of cytoreductive surgery plus
was improved from 11.8 to 16.0 months (HR 0.65, 95% CI 0.51– HIPEC in selected patients [94]. However, randomised data are
0.83). Following the ToGA trial results, trastuzumab was lacking for non-Asian patients. A large French series demon-
licensed in Europe for use in HER2-positive disease (IHC3+ or strated a median survival of surgery plus HIPEC of 9.2 months,
2+/FISH+) in combination with capecitabine or 5-FU and cis- with a 5-year survival rate of 13% for all patients and 23% for
platin. This regimen now represents the standard of care for those who had complete cytoreductive surgery [IV, C] [95].
these patients [I, A]. However, recent phase III randomised Currently, this approach cannot be recommended outside the
trials targeting the EGFR and MET-HGF axes have not demon- context of clinical research.
strated improvements in OS for anti-EGFR and anti-MET/HGF
therapies [82–85]. In contrast, emerging data from early phase signet cell tumours
trials suggests that use of immunotherapies such as the PD-1
Gastric AC associated with signet ring cells is associated with a
inhibitors pembrolizumab and nivolumab may result in durable
poor prognosis. Retrospective case series suggest that this gastric
remissions for a proportion of patients with advanced gastric
cancer subtype may be less sensitive to chemotherapy and CRT
cancer [86, 87]. The interaction between immunotherapy for
[IV] [96, 97]. However, evidence is insufficient to support not
gastric cancer and other known biomarkers in gastric cancer
adopting standard chemotherapy or surgical approaches for
such as MSI requires further investigation.
these patients.
A personalised medicine synopsis table is presented in
Table 4.
follow-up, long-term implications and
specific situations survivorship
metastasectomy In the setting of operable gastric cancer, the complexity of treat-
ment frequently induces symptoms that adversely affect health-
In general, patients with metastatic cancer do not benefit from related quality of life. A regular follow-up may allow investiga-
resection of metastases. Uncontrolled case series have demon- tion and treatment of symptoms, psychological support and
strated prolonged survival for selected patients undergoing liver early detection of recurrence, though there is no evidence that it
and lung metastasectomy and surgical removal of Krukenberg improves survival outcomes [III, B] [98, 99].
tumours [V, C] [88–90]. The randomised REGATTA trial estab- Follow-up should be tailored to the individual patient and the
lished (in an Asian patient population) that gastrectomy in stage of the disease [V, B] [100]. Dietary support is recom-
patients with limited metastatic disease does not improve sur- mended for patients on either a radical or a palliative pathway
vival [I, A] [91]. Until further evidence is presented, both with reference to vitamin and mineral deficiencies [V, B] [101,
gastrectomy and metastasectomy should be considered experi- 102].
mental for patients with gastric cancer. New strategies for patient follow-up are currently undergoing
evaluation, including patient-led self-referral and services led by
peritoneal metastases clinical nurse specialists [103].
Several small randomised trials in Asian patients have demon- In the advanced disease setting, identification of patients for
strated a significant survival benefit for adjuvant hyperthermic second-line chemotherapy and clinical trials requires regular
intraperitoneal chemotherapy (HIPEC) in high-risk curatively follow-up to detect symptoms of disease progression before sig-
resected gastric cancer patients; however, these results have not nificant clinical deterioration [IV, B].
been validated in non-Asian patients [92, 93]. For patients with If relapse/disease progression is suspected, then a clinical
history, physical examination and directed blood tests should be
carried out. Radiological investigations should be carried out in
Table 4. Personalised medicine synopsis table for lower patients who are candidates for further chemotherapy or RT
oesophageal and gastric cancer [IV, B].
Biomarker Method Use LOE, The aggressive nature of gastric cancer and historically poor
GOR outcomes even in the setting of operable disease mean that the
concept of survivorship is only now beginning to evolve. Long-
HER2 Immunohistochemistry for Used to select I, A
term implications, late effects of therapy and psychosocial impli-
HER2 protein expression patients with
cations of treatment have been poorly studied to date.
or ISH for HER2 gene metastatic
amplification disease for
treatment with a methodology
trastuzumab-
containing These clinical practice guidelines were developed in accordance
regimen with the ESMO standard operating procedures for clinical prac-
LOE, level of evidence; GOR, grade of recommendation; HER2, human ESMO-Guidelines-Methodology. The relevant literature has
receptor growth factor receptor 2; ISH, in situ hybridisation. been selected by the expert authors. A summary of recommen-
dations is given in Table 1. Levels of evidence and grades of
v | Smyth et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Incidence and epidemiology
• If a familial cancer syndrome such as HDGC is suspected, referral to a geneticist for assessment is recommended based on international clinical
guidelines [V, B]
Diagnosis and pathology
• Diagnosis should be made from a gastroscopic or surgical biopsy reviewed by an experienced pathologist, and histology should be reported according to
the WHO criteria [IV, C]
• Initial staging and risk assessment should include physical examination, blood count and differential, liver and renal function tests, endoscopy and
contrast-enhanced CT scan of the thorax, abdomen ± pelvis [V, A]
• Laparoscopy is recommended for patients with resectable gastric cancer [III, B]
• Multidisciplinary treatment planning before any treatment is mandatory [IV, C]
• EUS is helpful in determining the proximal and distal extent of the tumour and provides further assessment of the T and N stage; however, it is less
useful in antral tumours [III, B]
• PET-CT imaging may improve staging by detecting involved lymph nodes or metastatic disease. However, PET may not be informative in patients with
mucinous or diffuse tumours [III, B]
• Laparoscopy ± peritoneal washings for malignant cells are recommended in all stage IB–III gastric cancers that are considered potentially resectable, to
exclude radiologically occult metastatic disease; the benefit may be greater for patients with T3/T4 disease [III, B]
Treatment planning
• Multidisciplinary treatment planning before any treatment decision is mandatory. The core membership of the multidisciplinary team should include
surgeons, medical and radiation oncologists, radiologists and pathologists, with other members as available [IV, C]
• Endoscopic resection is appropriate for selected early tumours [III, B]

• For stage IB–III gastric cancer, radical gastrectomy is indicated; perioperative therapy is recommended for these patients [I, A]
• Medically fit patients should undergo D2 resections in high-volume surgical centres [I, B]
Surgery
• Endoscopic resection may be carried out for very early gastric cancers (T1a) if they are clearly confined to the mucosa, well-differentiated, ≤2 cm and
non-ulcerated [III, B]
• European Society of Gastrointestinal Endoscopy Guidelines recommend ESD as the treatment of choice for most gastric superficial neoplastic lesions
[IV, B]
• For stage IB–III gastric cancer, radical gastrectomy is indicated. Subtotal gastrectomy may be carried out if a macroscopic proximal margin of 5 cm can
be achieved between the tumour and the gastroesophageal junction. For diffuse cancers, a margin of 8 cm is advocated. Otherwise, a total gastrectomy is
indicated [III, A]. Perioperative therapy is recommended for these patients
• In Asian countries, experience from observational and randomised trials demonstrates that dissection leads to superior outcomes compared with D1
resection [II, B]
• Consensus opinion is that, in Western countries, medically fit patients should undergo D2 dissection that is carried out in specialised, high-volume
centres with appropriate surgical expertise and postoperative care [I, B]
Perioperative chemotherapy
• Perioperative (pre- and postoperative) chemotherapy with a platinum and fluoropyrimidine combination is recommended for patients with ≥stage IB
resectable gastric cancer [I, A]
• Since capecitabine avoids the need for an indwelling central venous access device, and is non-inferior to 5-FU in the advanced disease setting,
capecitabine-containing regimens can also be suggested in the perioperative setting [IV, C]
Adjuvant treatment
• For patients with ≥stage IB gastric cancer who have undergone surgery without administration of preoperative chemotherapy, postoperative CRT or
adjuvant chemotherapy is recommended [I, A]
• Adjuvant therapy with 5-FU/leucovorin plus conventionally fractionated RT resulted in improved OS years compared with surgery alone. After 10 years
of follow-up, this OS improvement remains significant [I, A]
• Postoperative CRT may (mainly) be compensating for suboptimal surgery [II, B]. However, some data suggest potential benefits from postoperative CRT
event after optimal D2 dissection [I, B]
• In patients who have had a microscopically incomplete resection, significant improvements in OS and local recurrence rates with the use of CRT after an
R1 resection have been seen [IV, B]
• RT should preferably be given as a concomitant regimen of fluoropyrimidine-based CRT to a total dose of 45 Gy in 25 fractions of 1.8 Gy, 5 fractions/
week by intensity-modulated RT techniques [IV, A]. The clinical target volume encompasses the gastric bed, anastomoses and draining regional lymph
nodes [I, B]
Continued
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw350 | v

by guest
on 05 February 2018
Table 5. Continued
• OS benefit has been demonstrated for patients treated with adjuvant chemotherapy [I, A]
• The benefit of 5-FU-based chemotherapy has been confirmed compared with surgery alone [I, A]
First-line treatment
• Doublet or triplet platinum/fluoropyrimidine combinations are recommended for fit patients with advanced gastric cancer [I, A]
• Patients with inoperable locally advanced and/or metastatic (stage IV) disease should be considered for systemic treatment (chemotherapy), which has
shown improved survival and quality of life compared with best supportive care alone [I, A]. However, comorbidities, organ function and PS must always
be taken into consideration [II, B]
• Capecitabine is associated with improved OS compared with infused 5-FU within doublet and triplet regimens [I, A]
• DCF in a 3-weekly regimen was associated with improved OS, but also added significant toxic effects including increased rates of febrile neutropaenia
[I, C]
Elderly patients with gastric cancer
• Regimens that have been specifically addressed in phase II trials in elderly patients with comparable survival results include capecitabine and oxaliplatin,
FOLFOX, single-agent capecitabine and S1 (in Asian patients) [III, B]
• The FLOT regimen is associated with a trend towards improved PFS but also with increased toxicity [II, B]
Second- and further-line treatment
• Second-line chemotherapy with a taxane (docetaxel, paclitaxel), or irinotecan, or ramucirumab as a single agent or in combination with paclitaxel is
recommended for patients who are of PS 0–1 [I, A]
• Similar efficacy has been demonstrated for weekly paclitaxel and irinotecan [I, A]
• In patients with disease progression >3 months following first-line chemotherapy, it may be appropriate to consider a rechallenge with the same drug
combination [IV, C]
• In patients with symptomatic locally advanced or recurrent disease, hypofractionated RT is an effective and well-tolerated treatment modality that may
palliate bleeding, obstructive symptoms or pain [III, B]
Personalised medicine and targeted therapy
• Trastuzumab is recommended in conjunction with platinum- and fluoropyrimidine-based chemotherapy for patients with HER2-positive advanced
gastric cancer [I, A]
Specific situations
Metastasectomy
• Gastrectomy in patients with limited metastatic disease does not improve survival [I, A]
Peritoneal metastases
• In patients with peritoneal metastases, the use of cytoreductive surgery plus HIPEC has been studied, but this approach cannot yet be recommended
outside the context of clinical research.
• A regular follow-up may allow investigation and treatment of symptoms, psychological support and early detection of recurrence, though there is no
evidence that it improves survival outcomes [III, B]
• Follow-up should be tailored to the individual patient and the stage of the disease [V, B]
• Dietary support is recommended for patients on either a radical or a palliative pathway, with reference to vitamin and mineral deficiencies [V, B]
• In the advanced disease setting, identification of patients for second-line chemotherapy and clinical trials requires regular follow-up to detect symptoms
of disease progression before significant clinical deterioration [IV, B]
• If relapse/disease progression is suspected, then a clinical history, physical examination and directed blood tests should be carried out. Radiological
investigations should be carried out in patients who are candidates for further chemotherapy or RT [IV, B]
HDGC, hereditary diffuse gastric cancer; WHO, World Health Organisation CT, computed tomography; EUS, endoscopic ultrasound; PET-CT, positron
emission tomography-computed tomography; ESD, endoscopic submucosal dissection; 5-FU, 5-fluorouracil; CRT, chemoradiotherapy; RT, radiotherapy;
OS, overall survival; PS, performance status; DCF, docetaxel, cisplatin, 5-FU; FOLFOX, leucovorin, 5-FU and oxaliplatin; FLOT, fluorouracil, leucovorin,
oxaliplatin and docetaxel; PFS, progression-free survival; HER2, human epidermal growth factor receptor 2; HIPEC, hyperthermic intraperitoneal
chemotherapy; AC, adenocarcinoma.
recommendation have been applied using the system given in conflict of interest
Table 6. Statements without grading were considered justified
standard clinical practice by the experts and the ESMO faculty. ES has reported honoraria from Five Prime Therapeutics for ad-
This manuscript has been subjected to an anonymous peer- visory board participation. DC has reported research support
review process. from Amgen, AstraZeneca, Bayer, Celgene, Medimmune, Merck
v | Smyth et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Table 6. Levels of evidence and grades of recommendation 7. van der Post RS, Vogelaar IP, Carneiro F et al. Hereditary diffuse gastric cancer:
(adapted from the Infectious Diseases Society of America–United updated clinical guidelines with an emphasis on germline CDH1 mutation
carriers. J Med Genet 2015; 52: 361–374.
States Public Health Service Grading Systema)
8. Bass AJ, Thorsson V, Shmulevich I et al. Comprehensive molecular
Levels of evidence characterization of gastric adenocarcinoma. Nature 2014; 513: 202–209.
I Evidence from at least one large randomised, controlled trial of 9. Ba-Ssalamah A, Prokop M, Uffmann M et al. Dedicated multidetector CT of the
stomach: spectrum of diseases. Radiographics 2003; 23: 625–644.
good methodological quality (low potential for bias) or meta-
10. Chen CY, Hsu JS, Wu DC et al. Gastric cancer: preoperative local staging with 3D
analyses of well-conducted randomised trials without
multi–detector row CT—correlation with surgical and histopathologic results.
heterogeneity
Radiology 2007; 242: 472–482.
II Small randomised trials or large randomised trials with a suspicion
11. Kim YN, Choi D, Kim SH et al. Gastric cancer staging at isotropic MDCT including
of bias (lower methodological quality) or meta-analyses of such coronal and sagittal MPR images: endoscopically diagnosed early vs. advanced
trials or of trials with demonstrated heterogeneity gastric cancer. Abdom Imaging 2009; 34: 26–34.
III Prospective cohort studies 12. Kwee RM, Kwee TC. Imaging in assessing lymph node status in gastric cancer.
IV Retrospective cohort studies or case–control studies Gastric Cancer 2009; 12: 6–22.
V Studies without control group, case reports, experts opinions 13. Catalano MF, Sivak MV, Jr, Rice T et al. Endosonographic features predictive of
lymph node metastasis. Gastrointest Endosc 1994; 40: 442–446.
14. Bhutani MS, Hawes RH, Hoffman BJ. A comparison of the accuracy of echo
A Strong evidence for efficacy with a substantial clinical benefit, features during endoscopic ultrasound (EUS) and EUS-guided fine-needle
strongly recommended aspiration for diagnosis of malignant lymph node invasion. Gastrointest Endosc
B Strong or moderate evidence for efficacy but with a limited clinical 1997; 45: 474–479.
benefit, generally recommended 15. Leake PA, Cardoso R, Seevaratnam R et al. A systematic review of the accuracy
and utility of peritoneal cytology in patients with gastric cancer. Gastric Cancer
C Insufficient evidence for efficacy or benefit does not outweigh the
2012; 15 (Suppl. 1): S27–S37.
risk or the disadvantages (adverse events, costs, ...), optional
16. Leake PA, Cardoso R, Seevaratnam R et al. A systematic review of the accuracy
D Moderate evidence against efficacy or for adverse outcome,
and indications for diagnostic laparoscopy before curative-intent resection of
generally not recommended gastric cancer. Gastric Cancer 2012; 15 (Suppl. 1): S38–S47.
E Strong evidence against efficacy or for adverse outcome, never 17. Sobin LH, Gospodarowicz MK, Wittekind C (eds). TNM Classification of Malignant
recommended Tumours, 7th edition. Oxford, UK: Wiley-Blackwell 2009.
18. Edge SB, Byrd DR, Compton CC (eds). AJCC Cancer Staging Manual, 7th edition.
a
By permission of the Infectious Diseases Society of America [104]. New York, NY: Springer 2010.
19. Pimentel-Nunes P, Dinis-Ribeiro M, Ponchon T et al. Endoscopic submucosal
dissection: European Society of Gastrointestinal Endoscopy (ESGE) guideline.
Endoscopy 2015; 47: 829–854.
Serono, Merrimack and Sanofi. WA has reported honoraria for
20. Japanese Gastric Cancer Association. Japanese gastric cancer treatment
presentations at scientific meetings for Taiho, Nestlé and Lilly; guidelines 2010 (ver. 3). Gastric Cancer 2011; 14: 113–123.
honoraria from Nestlé and Lilly for advising on research pro- 21. Bonenkamp JJ, Hermans J, Sasako M et al. Extended lymph-node dissection for
jects. MV has reported research support from Roche. AC has gastric cancer. N Engl J Med 1999; 340: 908–914.
reported consulting and advisory services for and speaking or 22. Cuschieri A, Weeden S, Fielding J et al. Patient survival after D1 and D2
writing engagements and public presentations for Merck resections for gastric cancer: long-term results of the MRC randomized surgical
Serono, Amgen, Servier, Roche, Lilly and Novartis; research trial. Br J Cancer 1999; 79: 1522–1530.
support from Roche, Merck Serono, Servier and Takeda. DA has 23. Degiuli M, Sasako M, Ponti A et al. Randomized clinical trial comparing survival
reported honoraria/consultancy for Roche, Merck Serono, after D1 or D2 gastrectomy for gastric cancer. Br J Surg 2014; 101: 23–31.
Bayer, Lilly and Servier; research support from Roche. 24. Songun I, Putter H, Kranenbarg EM et al. Surgical treatment of gastric cancer:
15-year follow-up results of the randomised nationwide Dutch D1D2 trial. Lancet
Oncol 2010; 11: 439–449.
25. Dikken JL, van Sandick JW, Allum WH et al. Differences in outcomes of
references oesophageal and gastric cancer surgery across Europe. Br J Surg 2013; 100:
1. Torre LA, Bray F, Siegel RL et al. Global cancer statistics, 2012. CA Cancer J Clin 83–94.
2015; 65: 87–108. 26. Begg CB, Cramer LD, Hoskins WJ, Brennan MF. Impact of hospital volume on
2. Arnold M, Karim-Kos HE, Coebergh JW et al. Recent trends in incidence of five operative mortality for major cancer surgery. JAMA 1998; 280: 1747–1751.
common cancers in 26 European countries since 1988: analysis of the European 27. Birkmeyer JD, Siewers AE, Finlayson EV et al. Hospital volume and surgical
Cancer Observatory. Eur J Cancer 2015; 51: 1164–1187. mortality in the United States. N Engl J Med 2002; 346: 1128–1137.
3. Forman D, Burley VJ. Gastric cancer: global pattern of the disease and an 28. National Oesophago-Gastric Cancer Audit 2015. Healthcare Quality Improvement
overview of environmental risk factors. Best Pract Res Clin Gastroenterol 2006; Partnership Ltd, 2015. http://www.hqip.org.uk/resources/national-oesophago-
20: 633–649. gastric-cancer-audit-report-2015/ (8 August 2016, date last accessed).
4. World Cancer Research Fund International/American Institute for Cancer 29. Mortensen K, Nilsson M, Slim K et al. Consensus guidelines for enhanced
Research. Continuous update project report: diet, nutrition, physical activity and recovery after gastrectomy: Enhanced Recovery After Surgery (ERAS®) Society
stomach cancer. 2016. wcrf.org/stomach-cancer-2016 (8 August 2016, date recommendations. Br J Surg 2014; 101: 1209–1229.
last accessed). 30. Quan Y, Huang A, Ye M et al. Comparison of laparoscopic versus open
5. Zanghieri G, Di Gregorio C, Sacchetti C et al. Familial occurrence of gastric gastrectomy for advanced gastric cancer: an updated meta-analysis. Gastric
cancer in the 2-year experience of a population-based registry. Cancer 1990; Cancer 2016; 19: 939–950.
66: 2047–2051. 31. Lee JH, Nam BH, Ryu KW et al. Comparison of outcomes after laparoscopy-
6. Palli D, Galli M, Caporaso NE et al. Family history and risk of stomach cancer in assisted and open total gastrectomy for early gastric cancer. Br J Surg 2015;
Italy. Cancer Epidemiol Biomarkers Prev 1994; 3: 15–18. 102: 1500–1505.
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw350 | v

by guest
on 05 February 2018
32. Kim W, Kim HH, Han SU et al. Decreased morbidity of laparoscopic distal 51. Noh SH, Park SR, Yang HK et al. Adjuvant capecitabine plus oxaliplatin for gastric
gastrectomy compared with open distal gastrectomy for stage I gastric cancer: cancer after D2 gastrectomy (CLASSIC): 5-year follow-up of an open-label,
short-term outcomes from a multicenter randomized controlled trial (KLASS–01). randomised phase 3 trial. Lancet Oncol 2014; 15: 1389–1396.
Ann Surg 2016; 263: 28–35. 52. Tsuburaya A, Yoshida K, Kobayashi M et al. Sequential paclitaxel followed by
33. Cunningham D, Allum WH, Stenning SP et al. Perioperative chemotherapy versus tegafur and uracil (UFT) or S-1 versus UFT or S-1 monotherapy as adjuvant
surgery alone for resectable gastroesophageal cancer. N Engl J Med 2006; 355: chemotherapy for T4a/b gastric cancer (SAMIT): a phase 3 factorial randomised
11–20. controlled trial. Lancet Oncol 2014; 15: 886–893.
34. Ychou M, Boige V, Pignon JP et al. Perioperative chemotherapy compared with 53. Paoletti X, Oba K, Burzykowski T et al. Benefit of adjuvant chemotherapy for
surgery alone for resectable gastroesophageal adenocarcinoma: an FNCLCC and resectable gastric cancer: a meta-analysis. JAMA 2010; 303: 1729–1737.
FFCD multicenter phase III trial. J Clin Oncol 2011; 29: 1715–1721. 54. Verheij M, Jansen EPM, Cats A et al. A multicenter randomized phase III trial of
35. Schuhmacher C, Gretschel S, Lordick F et al. Neoadjuvant chemotherapy neo-adjuvant chemotherapy followed by surgery and chemotherapy or by surgery
compared with surgery alone for locally advanced cancer of the stomach and and chemoradiotherapy in resectable gastric cancer: first results from the
cardia: European Organisation for Research and Treatment of Cancer randomized CRITICS study. J Clin Oncol 2016; 34 (Suppl): abstr 4000.
trial 40954. J Clin Oncol 2010; 28: 5210–5218. 55. Wagner AD, Unverzagt S, Grothe W et al. Chemotherapy for advanced gastric
36. Cunningham D, Starling N, Rao S et al. Capecitabine and oxaliplatin for advanced cancer. Cochrane Database Syst Rev 2010; 3: CD004064.
esophagogastric cancer. N Engl J Med 2008; 358: 36–46. 56. Glimelius B, Ekström K, Hoffman K et al. Randomized comparison between
37. Alderson D, Langley RE, Nankivell MG et al. Neoadjuvant chemotherapy for chemotherapy plus best supportive care with best supportive care in advanced
resectable oesophageal and junctional adenocarcinoma: results from the UK gastric cancer. Ann Oncol 1997; 8: 163–168.
Medical Research Council randomised OEO5 trial (ISRCTN 01852072). J Clin 57. Bouché O, Raoul JL, Bonnetain F et al. Randomized multicenter phase II trial of a
Oncol 2015; 33 (Suppl): abstr 4002. biweekly regimen of fluorouracil and leucovorin (LV5FU2), LV5FU2 plus cisplatin,
38. Pauligk C, Tannapfel A, Meiler J et al. 36LBA: Pathological response to or LV5FU2 plus irinotecan in patients with previously untreated metastatic gastric
neoadjuvant 5-FU, oxaliplatin and docetaxel (FLOT) versus epirubicin, cisplatin cancer: a Federation Francophone de Cancerologie Digestive Group Study—
and 5-FU (ECF) in patients with locally advanced, resectable gastric/ FFCD 9803. J Clin Oncol 2004; 22: 4319–4328.
esophagogastric junction (EGJ) cancer: data from the phase II part of the FLOT4 58. Okines AF, Norman AR, McCloud P et al. Meta-analysis of the REAL-2 and
phase III study of the AIO. Eur J Cancer 2015; 51(Suppl. S3): 756. ML17032 trials: evaluating capecitabine-based combination chemotherapy and
39. Macdonald JS, Smalley SR, Benedetti J et al. Chemoradiotherapy after surgery infused 5-fluorouracil-based combination chemotherapy for the treatment of
compared with surgery alone for adenocarcinoma of the stomach or advanced oesophago-gastric cancer. Ann Oncol 2009; 20: 1529–1534.
gastroesophageal junction. N Engl J Med 2001; 345: 725–730. 59. Starling N, Rao S, Cunningham D et al. Thromboembolism in patients with
40. Smalley SR, Benedetti JK, Haller DG et al. Updated analysis of SWOG-directed advanced gastroesophageal cancer treated with anthracycline, platinum, and
intergroup study 0116: a phase III trial of adjuvant radiochemotherapy versus fluoropyrimidine combination chemotherapy: a report from the UK National
observation after curative gastric cancer resection. J Clin Oncol 2012; 30: Cancer Research Institute Upper Gastrointestinal Clinical Studies Group. J Clin
2327–2333. Oncol 2009; 27: 3786–3793.
41. Dikken JL, Jansen EP, Cats A et al. Impact of the extent of surgery and 60. Dank M, Zaluski J, Barone C et al. Randomized phase III study comparing irinotecan
postoperative chemoradiotherapy on recurrence patterns in gastric cancer. J Clin combined with 5-fluorouracil and folinic acid to cisplatin combined with 5-fluorouracil
Oncol 2010; 28: 2430–2436. in chemotherapy naive patients with advanced adenocarcinoma of the stomach or
42. Kim S, Lim DH, Lee J et al. An observational study suggesting clinical benefit for esophagogastric junction. Ann Oncol 2008; 19: 1450–1457.
adjuvant postoperative chemoradiation in a population of over 500 cases after 61. Van Cutsem E, Moiseyenko VM, Tjulandin S et al. Phase III study of docetaxel and
gastric resection with D2 nodal dissection for adenocarcinoma of the stomach. cisplatin plus fluorouracil compared with cisplatin and fluorouracil as first-line
Int J Radiat Oncol Biol Phys 2005; 63: 1279–1285. therapy for advanced gastric cancer: a report of the V325 Study Group. J Clin
43. Lee J, Lim DH, Kim S et al. Phase III trial comparing capecitabine plus cisplatin Oncol 2006; 24: 4991–4997.
versus capecitabine plus cisplatin with concurrent capecitabine radiotherapy in 62. Shah MA, Janjigian YY, Stoller R et al. Randomized multicenter phase II study of
completely resected gastric cancer with D2 lymph node dissection: the ARTIST modified docetaxel, cisplatin, and fluorouracil (DCF) versus DCF plus growth
trial. J Clin Oncol 2012; 30: 268–273. factor support in patients with metastatic gastric adenocarcinoma: a study of the
44. Zhu WG, Xua DF, Pu J et al. A randomized, controlled, multicenter study US Gastric Cancer Consortium. J Clin Oncol 2015; 33: 3874–3879.
comparing intensity-modulated radiotherapy plus concurrent chemotherapy with 63. Al-Batran SE, Hartmann JT, Hofheinz R et al. Biweekly fluorouracil, leucovorin,
chemotherapy alone in gastric cancer patients with D2 resection. Radiother oxaliplatin, and docetaxel (FLOT) for patients with metastatic adenocarcinoma of the
Oncol 2012; 104: 361–366. stomach or esophagogastric junction: a phase II trial of the Arbeitsgemeinschaft
45. Park SH, Sohn TS, Lee J et al. Phase III trial to compare adjuvant chemotherapy Internistische Onkologie. Ann Oncol 2008; 19: 1882–1887.
with capecitabine and cisplatin versus concurrent chemoradiotherapy in 64. Van Cutsem E, Boni C, Tabernero J et al. Docetaxel plus oxaliplatin with or
gastric cancer: final report of the Adjuvant Chemoradiotherapy in Stomach without fluorouracil or capecitabine in metastatic or locally recurrent gastric
Tumors trial, including survival and subset analyses. J Clin Oncol 2015; 33: cancer: a randomized phase II study. Ann Oncol 2015; 26: 149–156.
3130–3136. 65. Guimbaud R, Louvet C, Ries P et al. Prospective, randomized, multicenter, phase
46. Stiekema J, Trip AK, Jansen EP et al. The prognostic significance of an R1 III study of fluorouracil, leucovorin, and irinotecan versus epirubicin, cisplatin, and
resection in gastric cancer patients treated with adjuvant chemoradiotherapy. capecitabine in advanced gastric adenocarcinoma: A French intergroup
Ann Surg Oncol 2014; 21: 1107–1114. (Fédération Francophone de Cancérologie Digestive, Fédération Nationale des
47. Trip AK, Nijkamp J, van Tinteren H et al. IMRT limits nephrotoxicity after Centres de Lutte Contre le Cancer, and Groupe Coopérateur Multidisciplinaire en
chemoradiotherapy for gastric cancer. Radiother Oncol 2014; 112: 289–294. Oncologie) Study. J Clin Oncol 2014; 32: 3520–3526.
48. Sakuramoto S, Sasako M, Yamaguchi T et al. Adjuvant chemotherapy for gastric 66. Xiang XJ, Zhang L, Qiu F et al. A phase II study of capecitabine plus oxaliplatin as
cancer with S-1, an oral fluoropyrimidine. N Engl J Med 2007; 357: first-line chemotherapy in elderly patients with advanced gastric cancer.
1810–1820. Chemotherapy 2012; 58: 1–7.
49. Sasako M, Sakuramoto S, Katai H et al. Five-year outcomes of a randomized 67. Lee JL, Kang YK, Kang HJ et al. A randomised multicentre phase II trial of
phase III trial comparing adjuvant chemotherapy with S-1 versus surgery alone in capecitabine vs S-1 as first-line treatment in elderly patients with metastatic or
stage II or III gastric cancer. J Clin Oncol 2011; 29: 4387–4393. recurrent unresectable gastric cancer. Br J Cancer 2008; 99: 584–590.
50. Bang YJ, Kim YW, Yang HK et al. Adjuvant capecitabine and oxaliplatin for gastric 68. Catalano V, Bisonni R, Graziano F et al. A phase II study of modified FOLFOX as
cancer after D2 gastrectomy (CLASSIC): a phase 3 open-label, randomised first-line chemotherapy for metastatic gastric cancer in elderly patients with
controlled trial. Lancet 2012; 379: 315–321. associated diseases. Gastric Cancer 2013; 16: 411–419.
v | Smyth et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
69. Rivera F, Massutí B, Salcedo M et al. Phase II trial of miniDOX (reduced MET-positive (MET+) adenocarcinoma of the stomach or gastroesophageal
dose docetaxel-oxaliplatin-capecitabine) in ‘suboptimal’ patients with advanced junction (GEC). J Clin Oncol 2015; 33 (Suppl): abstr 4012.
gastric cancer (AGC). TTD 08-02. Cancer Chemother Pharmacol 2015; 75: 86. Le DT, Bendell JC, Calvo E et al. Safety and activity of nivolumab monotherapy in
319–324. advanced and metastatic (A/M) gastric or gastroesophageal junction cancer (GC/
70. Trumper M, Ross PJ, Cunningham D et al. Efficacy and tolerability of GEC): results from the CheckMate-032 study. J Clin Oncol 2016; 34 (Suppl 45):
chemotherapy in elderly patients with advanced oesophago-gastric cancer: a abstr 6.
pooled analysis of three clinical trials. Eur J Cancer 2006; 42: 827–834. 87. Muro K, Chung HC, Shankaran V et al. Pembrolizumab for patients with PD-L1-
71. Al-Batran SE, Pauligk C, Homann N et al. The feasibility of triple-drug positive advanced gastric cancer (KEYNOTE-012): a multicentre, open-label,
chemotherapy combination in older adult patients with oesophagogastric cancer: phase 1b trial. Lancet Oncol 2016; 17: 717–726.
a randomised trial of the Arbeitsgemeinschaft Internistische Onkologie (FLOT65 88. Markar SR, Mackenzie H, Mikhail S et al. Surgical resection of hepatic
+). Eur J Cancer 2013; 49: 835–842. metastases from gastric cancer: outcomes from national series in England.
72. Thuss-Patience PC, Kretzschmar A, Bichev D et al. Survival advantage for Gastric Cancer 2016. [Epub ahead of print].
irinotecan versus best supportive care as second-line chemotherapy in gastric 89. Shiono S, Sato T, Horio H et al. Outcomes and prognostic factors of survival after
cancer – A randomised phase III study of the Arbeitsgemeinschaft Internistische pulmonary resection for metastatic gastric cancer. Eur J Cardiothorac Surg
Onkologie (AIO). Eur J Cancer 2011; 47: 2306–2314. 2013; 43: e13–e16.
73. Ford HE, Marshall A, Bridgewater JA et al. Docetaxel versus active symptom 90. Rosa F, Marrelli D, Morgagni P et al. Krukenberg tumors of gastric origin: the
control for refractory oesophagogastric adenocarcinoma (COUGAR-02): an open- Rationale of surgical resection and perioperative treatments in a multicenter
label, phase 3 randomised controlled trial. Lancet Oncol 2014; 15: 78–86. western experience. World J Surg 2016; 40: 921–928.
74. Kang JH, Lee SI, Lim DH et al. Salvage chemotherapy for pretreated gastric cancer: a 91. Fujitani K, Yang HK, Mizusawa J et al. Gastrectomy plus chemotherapy versus
randomized phase III trial comparing chemotherapy plus best supportive care with chemotherapy alone for advanced gastric cancer with a single non-curable factor
best supportive care alone. J Clin Oncol 2012; 30: 1513–1518. (REGATTA): a phase 3, randomised controlled trial. Lancet Oncol 2016; 17:
75. Roy AC, Park SR, Cunningham D et al. A randomized phase II study of PEP02 309–318.
(MM-398), irinotecan or docetaxel as a second-line therapy in patients with 92. Fujimoto S, Takahashi M, Mutou T et al. Successful intraperitoneal hyperthermic
locally advanced or metastatic gastric or gastro-oesophageal junction chemoperfusion for the prevention of postoperative peritoneal recurrence in
adenocarcinoma. Ann Oncol 2013; 24: 1567–1573. patients with advanced gastric carcinoma. Cancer 1999; 85: 529–534.
76. Hironaka S, Ueda S, Yasui H et al. Randomized, open-label, phase III study 93. Fujimura T, Yonemura Y, Muraoka K et al. Continuous hyperthermic peritoneal
comparing irinotecan with paclitaxel in patients with advanced gastric cancer perfusion for the prevention of peritoneal recurrence of gastric cancer:
without severe peritoneal metastasis after failure of prior combination randomized controlled study. World J Surg 1994; 18: 150–155.
chemotherapy using fluoropyrimidine plus platinum: WJOG 4007 trial. J Clin 94. Yang XJ, Huang CQ, Suo T et al. Cytoreductive surgery and hyperthermic
Oncol 2013; 31: 4438–4444. intraperitoneal chemotherapy improves survival of patients with peritoneal
77. Wilke H, Muro K, Van Cutsem E et al. Ramucirumab plus paclitaxel versus carcinomatosis from gastric cancer: final results of a phase III randomized clinical
placebo plus paclitaxel in patients with previously treated advanced gastric or trial. Ann Surg Oncol 2011; 18: 1575–1581.
gastro-oesophageal junction adenocarcinoma (RAINBOW): a double-blind, 95. Glehen O, Gilly FN, Arvieux C et al. Peritoneal carcinomatosis from gastric
randomised phase 3 trial. Lancet Oncol 2014; 15: 1224–1235. cancer: a multi-institutional study of 159 patients treated by cytoreductive
78. Fuchs CS, Tomasek J, Yong CJ et al. Ramucirumab monotherapy for previously surgery combined with perioperative intraperitoneal chemotherapy. Ann Surg
treated advanced gastric or gastro-oesophageal junction adenocarcinoma Oncol 2010; 17: 2370–2377.
(REGARD): an international, randomised, multicentre, placebo-controlled, phase 96. Heger U, Blank S, Wiecha C et al. Is preoperative chemotherapy followed by
3 trial. Lancet 2014; 383: 31–39. surgery the appropriate treatment for signet ring cell containing
79. Okines AF, Asghar U, Cunningham D et al. Rechallenge with platinum plus adenocarcinomas of the esophagogastric junction and stomach? Ann Surg Oncol
fluoropyrimidine +/- epirubicin in patients with oesophagogastric cancer. 2014; 21: 1739–1748.
Oncology 2010; 79: 150–158. 97. Charalampakis N, Nogueras González GM, Elimova E et al. The proportion of
80. Tey J, Back MF, Shakespeare TP et al. The role of palliative radiation therapy in signet ring cell component in patients with localized gastric adenocarcinoma
symptomatic locally advanced gastric cancer. Int J Radiat Oncol Biol Phys 2007; correlates with the degree of response to pre-operative chemoradiation. Oncology
67: 385–388. 2016; 90: 239–247.
81. Bang YJ, Van Cutsem E, Feyereislova A et al. Trastuzumab in combination with 98. Allum WH, Blazeby JM, Griffin SM et al. Guidelines for the management of
chemotherapy versus chemotherapy alone for treatment of HER2-positive oesophageal and gastric cancer. Gut 2011; 60: 1449–1472.
advanced gastric or gastro-oesophageal junction cancer (ToGA): a phase 3, 99. D’Ugo D, Biondi A, Tufo A, Persiani R. Follow-up: the evidence. Dig Surg 2013;
open-label, randomised controlled trial. Lancet 2010; 376: 687–697. 30: 159–168.
82. Waddell T, Chau I, Cunningham D et al. Epirubicin, oxaliplatin, and capecitabine 100. Baiocchi GL, D’Ugo D, Coit D et al. Follow-up after gastrectomy for cancer: the
with or without panitumumab for patients with previously untreated advanced Charter Scaligero Consensus Conference. Gastric Cancer 2016; 19: 15–20.
oesophagogastric cancer (REAL3): a randomised, open-label phase 3 trial. 101. Hu Y, Kim HI, Hyung WJ et al. Vitamin B(12) deficiency after gastrectomy for
Lancet Oncol 2013; 14: 481–489. gastric cancer: an analysis of clinical patterns and risk factors. Ann Surg 2013;
83. Lordick F, Kang YK, Chung HC et al. Capecitabine and cisplatin with or without 258: 970–975.
cetuximab for patients with previously untreated advanced gastric cancer (EXPAND): 102. Baek KH, Jeon HM, Lee SS et al. Short-term changes in bone and mineral
a randomised, open-label phase 3 trial. Lancet Oncol 2013; 14: 490–499. metabolism following gastrectomy in gastric cancer patients. Bone 2008; 42:
84. Cunningham D, Tebbutt NC, Davidenko I et al. Phase III, randomized, double- 61–67.
blind, multicenter, placebo (P)-controlled trial of rilotumumab (R) plus epirubicin, 103. Verschuur EM, Steyerberg EW, Tilanus HW et al. Nurse-led follow-up of patients
cisplatin and capecitabine (ECX) as first-line therapy in patients ( pts) with after oesophageal or gastric cardia cancer surgery: a randomised trial. Br J
advanced MET-positive ( pos) gastric or gastroesophageal junction (G/GEJ) Cancer 2009; 100: 70–76.
cancer: RILOMET-1 study. J Clin Oncol 2015; 33 (Suppl): abstr 4000.
104. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
85. Shah MA, Bang YJ, Lordick F et al. METGastric: a phase III study of onartuzumab among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
plus mFOLFOX6 in patients with metastatic HER2-negative (HER2-) and 139–144.
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw350 | v

by guest
on 05 February 2018
doi:10.1093/annonc/mdv295
Cancer of the pancreas: ESMO Clinical Practice

M. Ducreux1,2, A. Sa. Cuhna2,3, C. Caramella4, A. Hollebecque1,5, P. Burtin1, D. Goéré6,
T. Seufferlein7, K. Haustermans8, J. L. Van Laethem9, T. Conroy10 & D. Arnold11, on behalf of the
1
Département de médecine, Gustave Roussy, Villejuif; 2Faculté de Médecine, Université Paris Sud, le Kremlin Bicêtre; 3Département de Chirugie Hépato-biliaire, Hopital
Paul Brousse, Villejuif; 4Département d’imagerie; 5Département d’Innovation Thérapeutique; 6Département de Chirurgie Générale, Gustave Roussy, Villejuif, France;
7
Department of Internal Medicine I, Ulm University Hospital Medical Center, Ulm, Germany; 8Department of Radiation Oncology, Leuven Kankerinstitute, Leuven;
9
Departement of Gastroenterology, Hôpital Erasme, Cliniques Universitaires de Bruxelles, Brussels, Belgium; 10Département de médecine, Institut de Cancérologie de
Lorraine, Vandoeuvre lés Nancy, France; 11Department of Medical Oncology, Tumor Biology Center, Freiburg, Germany
incidence and epidemiology Familial pancreatic cancers, defined as at least two first-degree
A recent study estimating cancer epidemiology in 2014 (within relatives with pancreatic cancer, account for only 5%–10% of all
Europe) showed that pancreatic cancer was the fourth most fatal pancreatic cancer cases. Mutation in BRCA2 is probably the
cancer in men after lung, colorectal, and prostate cancers [1]. most common inherited disorder in familial pancreatic cancer.
Similarly, pancreatic cancer was found to be the fourth most fatal Other familial syndromes linked to pancreatic cancer are: her-
clinical practice
cancer in women after breast, colorectal and lung cancers [1]. editary pancreatitis, hereditary non-polyposis colorectal cancer,
guidelines
With a life expectancy of ∼5% at 5 years, the prognosis of this hereditary breast and ovarian cancers, Peutz–Jeghers syndrome,
cancer has not improved over the past 20 years, and incidence ataxia telangiectasia, familial atypical multiple mole melanoma
and mortality rates are very similar. Death due to pancreatic car- syndrome and Li–Fraumeni syndrome.
cinoma is increasing in Europe with the number rising from The main acquired risk factors for pancreatic cancer are cigarette
75 439 in 2009 to a projected 82 300 deaths in 2014 (+19%) [1]. It smoking (overall relative risk 1.74) and, to a lesser degree, environ-
usually arises in elderly patients with a mean age at onset of 71 mental tobacco smoke. The second most modifiable risk factor of
years for men and 75 years for women. The majority of patients pancreatic cancer is obesity. Tumorigenesis is enhanced by excess
with pancreatic cancer progress to either metastatic or locally adipose tissue, probably through the mechanism of abnormal
advanced disease in the asymptomatic phase. Surgical excision is glucose metabolism. Obesity [body mass index (BMI) > 30 kb/m2]
the definitive treatment with a 5-year survival rate (after resec- is associated with a 20%–40% higher rate of death from pancreatic
tion) of ∼20%, but it is only possible in 15%–20% of the patients. cancer. Meta-analyses have demonstrated associations between
The opportunity to detect pancreatic cancer, while it remains both type 1 and type 2 diabetes mellitus and pancreatic cancer,
curable, depends on the ability to identify and screen high-risk with odds ratios of ∼2.0 and 1.8, respectively [2].
populations before their symptoms arise. Defining the treatment Chronic pancreatitis accounts for ∼5% of pancreatic cancers.
strategy for patients suffering from pancreatic carcinoma requires The most common cause of chronic pancreatitis, in Europe, is
a specialised multidisciplinary team that includes: surgeons, excess alcohol consumption. The causal pathway is not clear,
medical oncologists, gastroenterologists, radiation therapists, radi- however, alcohol consumption by itself is related to an increased
ologists, and supportive and palliative care specialists. risk of pancreatic cancer.
The vast majority (>80%) of pancreatic carcinomas are due Helicobacter pylori, hepatitis B, and human immunodefi-
to sporadically occurring mutations. Only a small proportion ciency virus infection have also been reported to be related to an
(<10%) are due to inherited germline mutations. Germline increase in relative risk of pancreatic cancer, although some con-
mutations in BRCA2, p16, ATM, STK11, PRSS1/PRSS2, SPINK1, founding factors such as cigarette smoking or alcohol consump-
PALB2, and DNA mismatch repair genes are associated with tion have not always been considered [2].
varying degrees of increased risk for pancreatic carcinoma [2]. Dietary factors have been studied extensively, and clearly con-
tribute to the development of pancreatic cancer. Independent of
their role in causing obesity: butter, saturated fat, red meat, and
CH-6962 Viganello-Lugano, Switzerland. processed foods are clearly linked to pancreatic cancer [3].
E-mail: clinicalguidelines@esmo.org Conversely, a high fruit and folate intake could reduce the risk of
† pancreatic cancer [3].
Approved by the ESMO Guidelines Committee: February 2002, last update July 2015.
This publication supersedes the previously published version—Ann Oncol 2013; 24 Different chemical substances have been reported to increase
(Suppl. 6): vi106–vi114. the relative risk of developing pancreatic cancer, among these

by guest
on 05 February 2018
Table 1. Major non-genetic risk factors [5]a key points
Factor Relative risk Attributable fraction • Early symptoms of pancreatic cancer result from a mass effect
Tobacco 2 11%–32% • Common presenting symptoms include jaundice, pain, weight
Helicobacter pylori infection 1.5 4%–25% loss, steatorrhoea
Non-O-blood group 1.4 13%–19%
pathology
Diabetes mellitus 1.4–2.2 1%–16%
Obesity 1.2–1.5 3%–16% Pancreatic cancers arise from both the exocrine and endocrine
Red meat intake 1.1–1.5 2%–9% parenchyma of the gland, however, ∼95% occur within the exo-
Heavy alcohol intake 1.1–1.5 9% crine portion and may arise from ductal epithelium, acinar cells,
Low fruit and folate intake 0.5–1.0 <12% or connective tissue. Only 2% of tumours of the exocrine pan-
creas are benign. The most common pancreatic cancer is a ductal
a
By permission of Oxford University Press on behalf of The adenocarcinoma, which accounts for ∼80% of all pancreatic
International Epidemiological Association. cancers. Microscopically, these neoplasms vary from well-differ-
entiated duct-forming carcinomas (which may be so well differ-
entiated that they mimic non-neoplastic glands) to poorly
differentiated carcinomas, with epithelial differentiation demon-
are: chlorobenzoil, chlorinated hydrocarbon, nickel and nickel strable only on immunolabelling. Ductal adenocarcinomas typic-
compounds, chromium compounds, silica dust, and others [4]. ally elicit an intense stromal reaction which has been postulated to
A recent pooled analysis of 117 meta-analytical studies [5] serve as a barrier to chemotherapy [6]. A number of morphological
assigned a relative risk to a number of these factors improving variants of ductal carcinoma have been characterised, including
our understanding of the respective role of each factor (Table 1). colloid carcinoma and medullary carcinoma. Other variants of
As a summary, it is possible to say that 5%–10% of pancreatic pancreatic cancer, such as adenosquamous carcinoma and undiffer-
cancers are related to a genetic alteration. Among the remaining entiated carcinomas with osteoclast-like giant cells, are important
90%, the major risk factors are tobacco, H. pylori infection and to recognise because they are associated with a poorer prognosis.
factors related to dietary habits (BMI, red meat intake, low fruit On the contrary, acinar cell pancreatic cancers have a slightly
and vegetables intake, diabetes, alcohol intake). Some factors are better prognosis [7]. Neuroendocrine tumours of the pancreas are
difficult to interpret such as chronic pancreatitis, which is known the second most frequent pancreatic cancers, but they have a very
as a risk factor but can be related to alcohol intake. Diabetes is specific pattern that will not be considered in this paper.
also known as a risk factor but this could be the initial symptom Cystic neoplasms represent 10%–15% of cystic lesions of the
of the disease, and can also be found in chronic pancreatitis. pancreas [8]. The most commonly encountered cystic neo-
However, about two thirds of the major risk factors associated plasms include: serous cystadenoma, intraductal papillary mu-
with pancreatic cancer are potentially modifiable, affording a cinous neoplasm (IPMN), and mucinous cystic neoplasm
unique opportunity for preventing one of the deadliest cancers. (either cystadenoma or cystadenocarcinoma). Mucinous lesions
have potential for malignant progression and/or may harbour a
malignancy at the time of diagnosis. The non-mucinous lesions
key points
have no malignant potential.
• 5%–10% of pancreatic cancers are due to genetic alteration
• The main risk factors are tobacco, and factors related to key points
dietary habits (BMI, red meat intake, low fruit and vegetables
intake, diabetes, alcohol intake) • 95% of pancreatic cancers are adenocarcinomas
• Mucinous lesions of the pancreas have potential for malignant
progression
diagnosis and pathology/molecular
biology molecular biology
diagnosis The classical precursor lesions of pancreatic cancer show a ductal
phenotype, suggesting their ductal cell of origin. The most
Early symptoms of pancreatic cancer result from a mass effect. frequent precursors are microscopic pancreatic intraepithelial neo-
Approximately 60%–70% of pancreatic cancer arises in the head plasia (PanIN), followed by IPMN and mucinous cystic neoplasm.
of the pancreas, 20%–25% in the body and the tail, and the PanIN are microscopic (<5 mm) mucinous-papillary lesions,
remaining 10%–20% diffusely involve the pancreas. Tumours which lead to invasive carcinoma through an adenoma-carcinoma
located in the body and the tail are likely to be diagnosed at a sequence [9]. Similarly, IPMN and mucinous cystic neoplasms
more advanced stage than tumours located in the head, as these become neoplastic by stepwise gene alterations.
can develop symptoms related to an obstruction of the common Multiple combinations of genetic mutations are commonly
bile duct and/or the pancreatic duct. Common presenting symp- found in pancreatic cancers and can be classified as follows:
toms of pancreatic cancers include jaundice (for tumours of the
head), abdominal pain, weight loss, steatorrhoea, and new-onset (i) Mutational activation of oncogenes, predominantly KRAS
diabetes. Tumours can grow locally into the duodenum ( prox- found in >90% of pancreatic cancers.
imal for tumour of the head and distal for tumour of the body (ii) Inactivation of tumour suppressor genes such as TP53,
and tail) and result in an upper gastroduodenal obstruction. p16/CDKN2A, and SMAD4.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv295 | v

by guest
on 05 February 2018
(iii) Inactivation of genome maintenance genes, such as hMLH1

and MSH2, which control the repair of DNA damage. Most
of these mutations are somatic aberrations.
Many efforts to understand the genomics of pancreatic cancer
have evolved from the model system and especially cell lines.
However, cell lines are poor models of cancer since solid tumour
samples consist of a variety of tissues which may comprise only a
small proportion of tumour cells. Adjacent to and surrounding
these are the other tissues consisting of stroma and endothelial
cells. This is particularly the case in pancreatic cancer, since the
tumour contains a high content of stromal tissue. An analysis
using whole-genome sequencing and copy number variation ana-
lysis was recently carried out in 100 pancreatic ductal adenocar-
cinomas [10]. Chromosomal rearrangements, leading to gene
disruption were prevalent, affecting genes known to be important
in pancreatic cancer (TP53, SMAD4, CDKN2A, ARID1A, and
ROBO2) and new candidate drivers of pancreatic carcinogenesis
(KDM6A and PREX2). Patterns of structural variation in chro-
mosomes classified pancreatic cancer into four subtypes with po-
tential clinical utility. The subtypes were termed: stable, locally
rearranged, scattered, and unstable.
key points Figure 1. Diagnostic work-up before multidisciplinary decision. CT, com-
puted tomography.
• The most frequent precursors are microscopic PanIN, fol-
lowed by IPMN and mucinous cystic neoplasm
• Multiple combinations of genetic mutations are commonly The great advantage of EUS is its ability to provide tissue
found in pancreatic cancers samples, via fine-needle aspiration, that allow up to 95% diagnostic
• Some of the recent genetic mutations discovered could become accuracy (when carried out by an experienced cytopathologist).
targetable in the near future Aside from allowing the diagnosis of pancreatic adenocarcinoma,
this technique also permits the sampling of atypical lymph nodes
(portocaval especially) to check for tumours with distant metastasis,
staging and risk assessment a finding which would contraindicate radical resection. Incidental
CA 19-9 is not useful for the primary diagnosis of pancreatic hepatic metastases can also be sampled during the same procedure
cancer [I, E]. An increase in serum levels is seen in almost 80% without introducing any major risk [II, A].
of the patients with advanced disease. However, in patients not Radiological studies should include computed tomography
harbouring a functional Lewis enzyme (Lea-b- genotype: 7%–10% (CT) angiography at the pancreatic arterial (40–50 s) and portal
of the population), levels of CA 19-9 are typically undetectable or venous (65–70 s) phases. A consensus statement, describing a
below 1.0 U/ml. Conversely, the level of CA 19-9 is correlated to standardised reporting template, was recently developed to
the level of bilirubin and any cause of cholestasis is able to induce provide a precise reporting of disease staging and to improve the
false-positive results. CA 19-9 has a significant value as a prog- decision-making process for patients with pancreatic cancer
nostic factor and can be used as a marker to measure disease [12]. When assessing vessel involvement, the use of magnetic
burden and potentially guide treatment decisions. A preoperative resonance imaging (MRI) is left to expert discretion. It shows
serum CA 19-9 level ≥500 UI/ml clearly indicates a worse prog- equal benefit to CT scanning with no superiority demonstrated
nosis after surgery [IV, B] (Figure 1). in studies [13]. However, MRI is useful for solving problems
The imaging work-up must determine the tumour size and such as the detection of hepatic lesions that cannot be charac-
precise burden, as well as arterial and venous local involvement. terised by CT [II, A]. MRI and magnetic resonance cholangio-
All these factors are part of the TNM classification (Table 2). In pancratography may also be preferable for cystic neoplasms of
case of jaundice due to an obstructive cancer of the head of the the pancreas and to evaluate biliary anatomy [IV, C].
pancreas, a metal biliary stent should not be placed before initial In the majority of cases, pancreatic adenocarcinoma appears
work-up, because their use is linked to an increase of post- in the pancreatic arterial phase on CT examination, as a hypo-
operative morbidity if it is decided to resect the cancer [III, A]. attenuating homogeneous mass with indistinct margins. The
In case of biliary sepsis, plastic stents should be preferred. interruption (with or without dilatation) of the biliary duct is fun-
Endoscopic ultrasound (EUS) is now largely used in the staging damental to specify tumour extension. The presence of calcifica-
of adenocarcinoma. A recent meta-analysis study showed that EUS tions is very unlikely but a cystic part of the tumour can exist,
had limited value in the detection of all metastatic lymph nodes especially when the tumour originates from a degenerating cystic
[sensitivity (Se) 69%, specificity (Sp) 81%], but was valuable in the pancreatic lesion. Extra-pancreatic local extension has to be deli-
detection of vascular invasion (Se 85%, Sp 91%) and prediction of neated: enlarged lymph nodes (especially in the retroportal space),
resectability (Se 90%, Sp 86%) [11]. hepatic or peritoneal nodules are the main metastatic sites.
v | Ducreux et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Table 2. TNM classification 7th editiona must be described [III, B]. Performance and nutritional status,
as well as medical comorbidities, are important considerations
Primary tumour (T)
for all patients with pancreatic cancer, who are being considered
T0 = No evidence of primary tumour for any major treatment modalities (surgery, chemotherapy, or
Tis = Carcinoma in situ radiation). Advanced age is not a contraindication for any of
T1 = Tumour limited to the pancreas, ≤2 cm in greatest dimension these treatments.
T2 = Tumour limited to the pancreas, >2 cm in greatest dimension Biopsy is indicated for patients requiring a diagnosis, such as
T3 = Tumour extends beyond the pancreas but without involvement of patients initiating chemotherapy or chemoradiation. EUS-guided
the coeliac axis or the superior mesenteric artery
fine-needle aspiration allows preoperative tissue confirmation of
T4 = Tumour involves the coeliac axis or the superior mesenteric artery
malignancy, but fear of tumour cell dissemination along the
(unresectable primary tumour)
needle track has limited its use. A recent study has indicated that
Regional lymph nodes (N) it could be carried out without consequence on efficacy of surgery
NX = Regional lymph nodes cannot be assessed [16]. It must be recommended, especially in doubtful cases.
N0 = No regional lymph node metastasis Percutaneous biopsy of a liver metastasis can be used in metastat-
N1 = Regional lymph node metastasis ic disease, but percutaneous biopsy of the pancreas is contra-indi-
cated in potentially resectable cases [III, B].
(A minimum number of 10 lymph nodes analysed is recommended.)
Positron emission tomography/CT does not currently add
The regional lymph nodes are the peripancreatic nodes which may be much staging information in most patients with resectable
subdivided as follows: disease and cannot be recommended; its role will be clarified by
on-going studies.
Superior Superior to head and body
Endoscopic retrograde cholangiography and pancreatography
Inferior Inferior to head and body
(ERCP) is considered as pathognomonic when it shows a
Anterior Anterior pancreaticoduodenal, pyloric (for tumours of head
only), and proximal mesenteric
double stop on the main bile and pancreatic ducts. However,
Posterior Posterior pancreaticoduodenal, common bile duct, and ERCP had little diagnostic value over CT or MRI for the evalu-
proximal mesenteric ation of patients with pancreatic cancer [III, B].
Splenic Hilum of spleen and tail of pancreas (for tumours of body The additional use of staging laparoscopy to exclude periton-
and tail only) eal metastasis in resectable or borderline resectable patients has
Coeliac For tumours of head only been suggested by some authors, but it is not generally accepted
[14] [IV, C].
M1 Distant metastasis key points

• CA 19-9 is the most useful tumour marker in pancreatic
a cancer [IV, B]
By permission of the American Joint Committee on Cancer (AJCC),
Chicago, Illinois. The original source for this material is the AJCC
• Staging of the patient is initially done by CT scan
Cancer Staging Handbook, Seventh Edition (2010) published by
• EUS provides some complementary information and allows
Springer Science and Business Media LLC, www.springer.com. biopsy of the tumour [II, A]
• MRI should be discussed, especially in cystic lesions [IV, C]
According to the American Hepato-Pancreato-Biliary Association

consensus report, pancreatic ductal adenocarcinoma (when metasta- treatment
ses are absent) is classified as resectable, borderline resectable or At the end of the staging procedures, the tumour can be cate-
unresectable [14]. At the time of diagnosis, pancreatic ductal adeno- gorised as resectable, borderline resectable, locally advanced or
carcinoma is deemed resectable in only 15%–20% of patients. metastatic disease. A treatment decision must be taken in ac-
For arterial vessels, three situations can exist: vessel tumour cordance with these findings, including general and nutritional
contact <180° without deformation, more than 180° without de- status considerations.
formation, or with deformation (i.e. abutment). For venous vessels,
one supplementary situation is described: tear drop deformation at treatment of localised disease
the tumour contact (i.e. distortion). With the use of such criteria,
CT or MRI are able to determine the non-resectability of the Surgical resection is the only potentially curative treatment of pan-
tumour with a high positive predictive value (>90%), but have an creatic adenocarcinoma. However, at diagnosis, <20% of patients
insufficient predictive value to affirm resectability (<50%) [15]. have a resectable tumour. The main goal of surgery is to achieve
Each vessel—superior mesenteric artery (SMA), coeliac axis, negative (R0) resection margins. After radiological evaluation,
and common hepatic artery—has to be assessed individually only patients with a high probability of R0 resection are good can-
with attention paid to local encasement or abutment and a pos- didates for upfront surgery.
sible anatomic variant, as these can be crucial for the surgical
decision making. The portal vein (PV) and the superior mesen- resectability criteria
teric vein (SMV) are the major trunks; any local involvement, An expert consensus group has developed criteria to define
thrombus, or hazy attenuation of the fat surrounding the vessel, tumour resectability, to improve patient selection and the rate of
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv295 | v

by guest
on 05 February 2018
R0 resections [14, 17, 18]. According to the degree of contact • Anteriosuperior group along the common hepatic artery
between the tumour and the vessels (PV or SMV, SMA, coeliac (station 8a)
trunk, and common hepatic artery), tumours are classified as re- • Along the bile duct (station 12b)
sectable, borderline resectable or locally advanced [IV, B]. For • Around the cystic duct (station 12c)
patients with resectable tumours, upfront surgery remains the • On the posterior aspect of the superior (station 13a)
standard of care. Patients with borderline resectable tumours • On the inferior portion of the head of pancreas (station 13a)
have a high probability of R1 resection and, as such, should not • On the right lateral side of SMA (station 14a and 14b)
be considered as good candidates for upfront surgery. Patients • On the anterior surface of the superior (station 17a) and infer-
with locally advanced or metastatic disease have to be considered ior portion of the head of pancreas (station 17b)
as having unresectable tumours. These criteria (which should be
considered when defining resectability) [19] have been adopted For tumours of the body and tail of the pancreas, removal of the
in the National Comprehensive Cancer Network (NCCN) guide- following lymph nodes is recommended:
lines (Table 3) [IV, B]. • At the splenic hilum (station 10)
• Along the splenic artery (station 11)
resection and margins • The inferior margin of pancreas
The location and the size of the tumour determine the type of Standard lymphadenectomy should involve the removal of
surgery. Patients with tumours in the head of the pancreas are ≥15 lymph nodes to allow adequate pathologic staging of the
treated with pancreatoduodenectomy (Whipple procedure). disease. The total number of lymph nodes examined and lymph
Dissection of the right hemi-circumference of the SMA to the nodes ratio (number of involved lymph nodes/number of lymph
right of the coeliac trunk is recommended to obtain a good nodes examined) should be reported in the pathologic analysis
medial clearance and to improve the rate of R0 resection [20]. In [25] [IV, A].
the event of vein involvement, complete venous resection (PV or
SMV) followed by reconstruction to obtain R0 resection is pos- age and pancreatectomy
sible. However, PV or SMV resection is associated with a lower Some authors have proposed a score that accurately predicts the
rate of R0 resection and poor survival, likely due to the inherent risk of perioperative mortality in patients undergoing pancreatic
aggressiveness of the tumour [21] [IV, B]. Arterial resections resection. This surgical outcomes analysis and research (SOAR)
during pancreatoduodenectomy are associated with increased pancreatectomy score is calculated based on preoperative factors
morbidity and mortality, and are not recommended. Frozen sec- (http://www.umassmed.edu/surgery/toolbox/panc_mortality_
tions analysis of pancreatic neck transection and common bile custom/) [26][IV, C].
duct transection margins is recommended [IV, E].
The International Study Group of Pancreatic Surgery [18] has preoperative biliary drainage
recently recommended adhering to the guidelines from British A recent prospective and randomised trial demonstrated an
Royal College of Pathologists (RCpath) for specimen examin- increased complication rate associated with routine preoperative
ation and the R1 definition (margin <1 mm). They advise sur- biliary drainage [27] [I, E]. However, patients in the trial had a
geons to identify the following margins: anterior, posterior, total bilirubin level below 250 µmol/l. Therefore, the correct ap-
medial, or superior mesenteric groove, SMA, pancreatic transec- proach in patients with higher levels remains undefined. If jaun-
tion, bile duct, and enteric. Tumour clearance should be given dice is present at diagnosis of pancreatic carcinoma, endoscopic
for all seven margins [IV, B]. drainage should only be carried out preoperatively in patients
For patients with tumours in the body or tail of the pancreas, with active cholangitis, or in those whom resection for cure
distal pancreatectomy, including the resection of the body and the cannot be scheduled within 2 weeks of diagnosis, and in those
tail of the pancreas and the spleen, is usually carried out. Radical with a bilirubin level below 250 µmol/l.
anterograde modular pancreatosplenectomy, with dissection of the
left hemi-circumference of the SMA, to the left of the coeliac trunk, adjuvant treatment after surgical resection
is recommended to ensure R0 resection [22] [IV, A]. Considering the poor results of surgery alone in pancreatic car-
Some recent studies show that minimally invasive techniques cinoma, many efforts involving chemotherapy, radiotherapy or
(laparoscopy) can reduce the morbidity of pancreatectomies both have been made to improve the 5-year survival of these
without having a negative impact on cancer outcome [23]. patients.
However, data relating to these techniques are insufficient, particu-
larly in relation to oncological results [24]. Therefore, open surgery adjuvant chemotherapy
remains the standard of care [II, C]. Postoperative adjuvant chemotherapy was evaluated in several
randomised trials. In the ESPAC-1 multicentre randomised
lymphadenectomy trial, a 2 × 2 factorial design, 289 patients treated with curative
In pancreatic cancer, extended lymphadenectomy is not recom- resection and complete gross resection received one of four
mended. Standard lymphadenectomy for pancreatoduodenect- therapeutic modalities: exclusive adjuvant chemotherapy [bolus
omy should resect the following lymph nodes: 5-fluorouracil (5-FU) and folinic acid], chemoradiation only
(split course 40 Gy plus 5-FU), or chemoradiation followed by
• Suprapyloric (station 5) chemotherapy or surveillance alone [28]. Patients who received
• Infrapyloric (station 6) chemotherapy had a longer median survival (20.1 versus 15.5
v | Ducreux et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Table 3. Definition of resectability according to NCCN guidelines [19]
Resectability Arterial Venous
status
Resectable No arterial tumour contact [coeliac axis (CA), superior No tumour contact with the superior mesenteric vein (SMV), or
mesenteric artery (SMA), or common hepatic artery (CHA)] portal vein (PV) or <180° contact without vein contour
irregularity
Borderline Pancreatic head/uncinate process

resectable • Solid tumour with CHA without extension to coeliac axis or • Solid tumour contact with the SMV or PV of >180°, contact of
hepatic artery bifurcation allowing for safe and complete <180° with contour irregularity of the vein or thrombosis of the
resection and reconstruction vein but with suitable vessels proximal and distal to the site of
• Solid tumour contact with the SMA <180° involvement allowing for safe and complete resection and vein
• Presence of variant arterial anatomy (e.g. accessory right reconstruction
hepatic artery) and the presence and degree of tumour contact • Solid tumour contact with the inferior vena cava (IVC)
should be noted if present as it may affect surgical planning
Pancreatic body/tail
• Solid tumour contact with the CA of <180°
• Solid tumour contact with the CA of >180° without
involvement of the aorta and with intact and uninvolved
gastroduodenal artery (some members prefer these criteria to
be in the unresectable category)
Unresectable • Distant metastases

Pancreatic head/uncinate process Pancreatic head/uncinate process
• Solid tumour contact with SMA >180° • Unreconstructible SMV/PV due to tumour involvement or
• Solid tumour contact with the CA >180° occlusion (can be due to tumour or bland thrombus)
• Solid tumour contact with the first jejunal SMA branch • Contact with most proximal draining jejunal branch into SMV
Body and tail Body and tail
• Solid tumour contact with the SMA and CA • Unreconstructible SMV/PV due to tumour involvement or
• Solid tumour contact with the CA and aorta occlusion (can be due to tumour or bland thrombus)
months, P = 0.009) compared with patients who did not. The first trial, the EORTC trial compared chemoradiation with simple
results of the CONKO-001 trial comparing gemcitabine to ob- surveillance after pancreaticoduodenectomy. In the subgroup of
servation confirmed the benefit of adjuvant chemotherapy [29]. 114 patients with pancreatic tumour(s), the survival benefit for ad-
Gemcitabine administered for 24 weeks improved disease-free juvant chemoradiation was not significant. The ESPAC-1 trial has
survival (13.4 versus. 6.9 months, P < 0.001) and overall survival even suggested a deleterious effect of adjuvant chemoradiation,
(OS) (22.8 versus 20.2 months, P = 0.005). The ESPAC-3 trial with recurrence-free survival of 10.7 months in the chemoradiation
compared the administration of adjuvant chemotherapy with group versus 15.2 months in its absence (P = 0.04) [28]. Even in
six cycles of either fluorouracil and folinic acid or gemcitabine R1 patients, no benefit was observed with adjuvant chemoradia-
[30]. No difference in OS, recurrence-free quality of life or sur- tion. Thus, no chemoradiation should be given to patients after
vival was observed. Recent information suggests that gemcita- surgery except in clinical trials [I, E].
bine may only be effective in patients with the enzymatic
equipment to transport gemcitabine into the tumour cell recommendations for treatment of localised disease
(hENT1) and metabolically activate it [31]. Unfortunately, there • A multidisciplinary team is necessary
is no sufficiently reliable commercial immunohistochemical • Tumour clearance should be given for all seven margins
antibody allowing the routine use of this test before giving gem- identified by the surgeon [IV, B]
citabine, and this is still not a factor when selecting a chemo- • Standard lymphadenectomy should involve the removal of
therapy regimen. At present, both 5-FU/folinic acid and ≥15 lymph nodes to allow adequate pathologic staging of the
gemcitabine can be considered as a standard of care [I, A]. disease [IV, A]
adjuvant chemoradiation • Adjuvant treatment is done with either gemcitabine or 5-FU
folinic acid [I, A]
Three randomised trials compared the benefits of adjuvant che-
• No chemoradiation should be given to patients after surgery
moradiation after pancreatic resection against surveillance alone. A
except in clinical trials [I, E]
first trial by the GastroIntestinal Tumour Study Group evaluating
chemoradiation (40 Gy + 5-FU) was stopped prematurely after the
treatment of 40 patients. An interim analysis revealed a low rate of
treatment of non-resectable disease
inclusion and a significant difference in survival in favour of the In 30%–40% of patients, while the tumour is confined to the
chemoradiation arm. After the many criticisms made against this pancreatic region, resection is not feasible, mainly due to
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv295 | v

by guest
on 05 February 2018
vascular invasion. The division of this subgroup of patients into the overall median survival of the patients treated with chemo-
two different categories has been shown recently but it is not therapy alone was 16 months. This may be related to more
always easy to define. active treatment of the patients diagnosed with metastasis.
When compared with best supportive care, chemoradiation
borderline resectable lesions showed a benefit in terms of survival in a small phase III trial [37].
Tumours are considered resectable upon good response to neoad- Old trials suggested the superiority of chemoradiation over
juvant treatment including induction chemotherapy, preoperative radiotherapy and chemotherapy alone, which was confirmed by
chemoradiation or a combination of both. Small retrospective meta-analyses [38] [I, B].
studies and two meta-analyses, including patients with both bor- Concerning the comparison with chemotherapy alone, while
derline and resectable lesions, have reported an even better sur- poor-quality randomised trials have suggested a benefit in
vival for these patients than for those with immediately resectable favour of chemoradiation, two recent trials showed opposite
tumours [32, 33]. While the heterogeneity of the trials on neoad- results. In a French trial using an obsolete regimen of chemora-
juvant therapy in borderline resectable pancreatic cancer limits diation (50 Gy + 5-FU cisplatin), the survival was better in the
the power of any conclusion, many individual series demonstrate gemcitabine alone arm (13 versus 8.6 months [39]). In another
improved R0 resection rates and promising survival rates. The trial, comparing chemoradiation with gemcitabine plus radio-
majority of studies used full-dose radiotherapy paired either with therapy versus gemcitabine alone, the OS was significantly
capecitabine, 5-FU or reduced doses of gemcitabine, or even a improved in the chemoradiation arm (11.1 versus 9.2 months)
combination of gemcitabine plus oxaliplatin. It is impossible at yet toxicity also increased by combining the treatment modal-
this time to recommend any chemoradiation treatment other than ities [40] [II, C].
the classical combination of capecitabine and radiotherapy [IV, C]. While many trials have evaluated the best combined regimen
To improve the disease control and to intensify the treatment of of chemotherapy and radiotherapy, no clear definition has been
the systemic disease, it has recently been proposed to begin treat- made of a standard of care. For instance, it has been suggested
ment with chemotherapy before starting chemoradiation. Again, that gemcitabine could be a better sensitiser and chemothera-
due to heterogeneity of the small retrospective series, it is very diffi- peutic agent to combine with radiotherapy than fluoropyrimi-
cult to recommend a specific schedule of treatment, although dines in the treatment of locally advanced pancreatic cancer.
some series have reported better survival using a multimodal strat- However, evidence from one randomised trial favoured capeci-
egy than that observed with upfront surgery in patients with tabine as less toxic and more active than gemcitabine in this
clearly resectable tumours [34]. Recent chemotherapy regimens, setting [41] [I, E].
such as FOLFIRINOX [folinic acid (leucovorin)/5-FU/irinotecan/ In an attempt to combine the advantages of both chemother-
oxaliplatin], have already shown promising results in small series apy and chemoradiation, the use of chemoradiation in patients
of patients with borderline resectable lesions [30%–45% of showing no sign of progressive disease after 3 months of chemo-
objective response rate (ORR)]. A trial, which was stopped therapy alone was suggested to be beneficial. However, no clear
prematurely, reported interesting response rates, median pro- advantage in favour of the chemoradiation was found in a recent
gression-free survival (PFS) and OS for patients treated with large randomised trial investigating this strategy. This trial was
chemotherapy (gemcitabine) followed by chemoradiation (gem- planned to include 722 patients, and was stopped for futility
citabine, 5-FU, cisplatin) versus chemoradiation alone [35]. after the inclusion of 449 patients (only 269 assessable for the
However, this small trial does not allow any definite conclusions main end point, OS). They were treated with 4 months of gemci-
to be drawn. Patients with borderline resectable lesions should tabine ±erlotinib (first randomisation) and then randomised to
be included in clinical trials wherever possible. If this is not feas- receive either two supplementary months of gemcitabine or che-
ible, a period of chemotherapy followed by chemoradiation and moradiation [36]. The median OS showed no improvement in
then surgery appears to be the best option [IV, B]. the chemoradiation group (15.2 versus 16.4 months) even
recommendations for treatment of borderline resectable disease though local tumour control did seem a little bit better in this
group [I, D]. Several small retrospective and prospective studies
• Patients with borderline resectable lesions should be included have suggested that FOLFIRINOX may be able to obtain an inter-
in clinical trials wherever possible esting response rate in this population, and may have rendered
• In routine practice, if the patient is not included in a trial, a some patients with locally advanced cancers resectable. However,
period of chemotherapy (gemcitabine or FOLFIRINOX) fol- it is too early to recommend this treatment and trials are ongoing.
lowed by chemoradiation and then surgery appears to be the Thus, the standard of care for these patients currently remains as
best option [IV, B] 6 months of gemcitabine [I, B].
locally advanced disease

recommendations for treatment of locally advanced disease
When the patient has no metastases and the tumour is not con-
sidered as borderline resectable, the tumour is defined as truly • The standard of care is 6 months of gemcitabine [I, A]
locally advanced (Table 3). Treatment of this group of patients • A minor role of chemoradiation in this subgroup of patients
remains highly controversial. Regardless of the treatment strat- has been observed [I, A]
egy, the average OS for these patients remains low (<1 year) in • It is impossible to recommend any chemoradiation treatment
the oldest studies. However, in the recent LAP07 trial [36], other than the classical combination of capecitabine and
which included only patients with locally advanced disease, radiotherapy [IV, C]
v | Ducreux et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
treatment of advanced/metastatic produced the greatest benefit. However, due to the low level of
evidence, there have been no clear changes in the daily clinical
disease management of these patients. Even with modern agents such as
palliative and supportive care tyrosine kinase inhibitors or monoclonal antibodies against
Before even considering systemic chemotherapy, patients with meta- various targets, the results of large phase III trials evaluating the
static pancreatic cancer may need interventions to provide relief addition of these drugs to gemcitabine have been disappointing.
of biliary and/or duodenal obstruction, malnutrition, and pain. The targeted therapies tested have included bevacizumab, cetuxi-
In the event of a biliary obstruction due to a pancreatic mab, aflibercept, and anti-insulin growth factor agents. The ex-
tumour, the endoscopic placement of a metallic biliary stent is ception is the combination of gemcitabine and the EGFR tyrosine
strongly recommended. The endoscopic method is safer than kinase inhibitor erlotinib, which gained regulatory approval fol-
percutaneous insertion and is as successful as surgical hepatoje- lowing a 12-day improvement in median survival compared with
junostomy [42] [II, B]. gemcitabine alone in a large randomised trial [46]. Arguably,
Duodenal obstruction is preferentially managed by endoscop- however, this duration of survival prolongation is clinically irrele-
ic placement of an expandable metal stent when possible, and is vant for most patients; therefore, erlotinib has not been widely
favoured over surgery [42] [IV, B]. used in this disease. The inefficacy of erlotinib in locally advanced
Pain is also considered to be a major priority in these patients disease which was shown in the LAP07 trial [36] is an additional
and it is observed in almost all patients with advanced pancreat- argument against the use of this drug for this indication.
ic carcinoma. It must be managed aggressively following stand- Major improvements in the treatment of metastatic disease came
ard guidelines on pain treatment, without any major specificity with the demonstration of the efficacy of a 5-FU-based triplet
due to the location of the disease [43]. However, radiotherapy chemotherapy. The FOLFIRINOX regimen has proven superior to
can be used at this stage to control the coeliac pain induced by a gemcitabine alone, in terms of efficacy, despite an increase in tox-
primary pancreatic tumour. Oral supplementation of pancreatic icity [47]. This trial included patients who were selected based on
enzyme has been suggested to help control pain; though this has their ability to receive this aggressive chemotherapy [Eastern
never been proven by a randomised study and should not be Cooperative Oncology Group (ECOG) performance status 0 or 1,
considered as a reason to prescribe such drugs. The input of a and normal or subnormal serum bilirubin level]. The median OS
pain control specialist is often mandatory. was 11.1 months in the FOLFIRINOX group compared with 6.8
Coeliac plexus block (CPB) can lead to pain control and fre- months in the gemcitabine group [hazard ratio (HR) for death,
quently to a decrease in the total amount of systemic drugs and 0.57; 95% confidence interval 0.45–0.73; P < 0.001). Median PFS
their side-effects. The endoscopic method is safer than percu- and ORR were also statistically better with FOLFIRINOX than with
taneous insertion and is as successful as surgical hepatojejunost- gemcitabine. More adverse events were noted in the FOLFIRINOX
omy [42] [II, B]. group; 5.4% of patients in this group had febrile neutropaenia.
While it was classically done percutaneously, one new way to FOLFIRINOX is an option for the treatment of patients with meta-
perform CPB is represented by EUS guidance. Previous studies static pancreatic cancer and good performance status [I, A]. After
have suggested a decrease in success rate when there is evidence of decades of failed randomised trials that have tried to add a drug to
disease outside the pancreas, such as coeliac or portal adenopathy. gemcitabine alone, a recent trial has shown positive results, demon-
CPB should be carried out in the presence of resistant pain and strating that the combination of gemcitabine plus nab-paclitaxel is
only if the clinical condition of the patient is not poor. better than gemcitabine alone in metastatic patients [48]. A total of
861 patients were randomly assigned to receive nab-paclitaxel plus
recommendation for palliative and supportive care in advanced/ gemcitabine or gemcitabine alone. The median OS was 8.5 months
metastatic disease in the nab-paclitaxel–gemcitabine group compared with 6.7
months in the gemcitabine group (HR for death, 0.72). Both median
• Duodenal obstruction is preferably managed by endoscopic PFS and response rate were improved in the nab-paclitaxel–gemcita-
placement of an expandable metal stent when possible, and is bine group compared with gemcitabine alone. In the nab-pacli-
favoured over surgery [42] [IV, B] taxel–gemcitabine group, neuropathy of grade 3 or higher
improved to grade 1 or lower in a median of 29 days [I, A].
oncologic treatment There are no data concerning a direct comparison of
Over the past two decades, the development of improved systemic FOLFIRINOX and Gem-nab-paclitaxel. An indirect comparison
treatments has been a top priority for pancreatic cancer. In 1997, of the two regimens may suggest a slightly greater activity but also
gemcitabine monotherapy was established as the standard of care, higher toxicity of FOLFIRINOX. No specific data favours the use
after being shown to offer greater clinical benefit and a small sur- of one regimen over the other in a defined subgroup of patients.
vival improvement over weekly 5-FU therapy [44] [II, A]. Since Thus, either of these two options can be offered to patients with
then, gemcitabine chemotherapy combinations have been in- serum bilirubin levels less than 1.5× upper limit of normal (ULN)
tensely evaluated. Despite frequently encouraging early-phase and good performance status (ECOG 0-1) [I, A].
data, phase III trials have not given confirmatory results. These In conclusion, it can be considered that there are three
combinations have included cytotoxic agents such as irinotecan, options to treat patients with a metastatic pancreatic cancer
5-FU, cisplatin, oxaliplatin and capecitabine. Three separate according to their general status:
meta-analyses [45] reported a statistically significant survival ad-
vantage of combination therapy compared with gemcitabine 1) For patients with performance status of 3/4, with significant
alone [I, C]. Capecitabine and cisplatin-based combinations have morbidities and a very short life expectancy: only
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv295 | v

by guest
on 05 February 2018
symptomatic treatment can be considered. Even chemother- • If the performance status is 3/4, with significant morbidities
apy with gemcitabine cannot be considered for such and a very short life expectancy: only symptomatic treatment
patients. can be considered
2) In very selected patients with ECOG performance status 2 • In very selected patients with ECOG performance status 2 due
due to heavy tumour load, gemcitabine and nab-paclitaxel to heavy tumour load, gemcitabine and nab-paclitaxel can be
can be considered for best chance of response [II, B]. considered for best chance of response [II, B]
3) For patients with performance status of 2 and/or bilirubin • If the performance status of the patient is 2 and/or the biliru-
level higher than 1.5× ULN: monotherapy with gemcitabine bin level is higher than 1.5× ULN: a monotherapy with gemci-
could be considered [I, A] tabine should be considered [I, A]
4) If the performance status of the patient is 0 or 1 and the biliru- • If the performance status of the patient is 0 or 1 and the biliru-
bin level is below 1.5× ULN two types of combination chemo- bin level is below 1.5× ULN two types of combination chemo-
therapy—the FOLFIRINOX regimen or the combination of therapy—the FOLFIRINOX regimen or the combination of
gemcitabine and nab-paclitaxel—should be considered [I, A] gemcitabine and nab-paclitaxel—should be considered [I, A]
The efficacy of the treatment has to be evaluated every two
months with a comparative CT scan. The treatment has to be personalised medicine
stopped if a RECIST progression is observed and second-line Among the validated drugs for pancreatic cancer, there is cur-
treatment has to be discussed. rently no relevant biomarker used in the medical decision
making and none should be used in clinical practice.
second-line treatment Emerging data from sequencing initiatives in pancreatic cancer
A first randomised trial (168 patients) has shown, in patients are unveiling a vast array of molecular aberrations and also a sig-
with advanced gemcitabine-refractory pancreatic cancer, that nificant inter- and intra-tumoral heterogeneity [52]. This intro-
second-line 5-FU, folinic acid and oxaliplatin, significantly duces major challenges when trying to identify the most relevant
extend the duration of OS when compared with 5-FU, folinic targeted therapy. The most frequent genes harbouring genetic
acid alone [49]. These results have not been confirmed by a aberrations in pancreatic cancer are KRAS, TP53, CDKN2A, and
more recent Canadian trial [45]. Very recently, combination of SMAD4. Among the numerous biomarkers tested in pancreatic
MM-398, a nanoliposomal encapsulation of irinotecan, and 5- cancer, some deserve particular attention.
FU, folinic acid has shown an improvement of OS (6.1 versus Loss of SMAD4 expression in pancreatic cancer has been asso-
4.2 months), PFS and ORR in the intent-to-treat population ciated with a poorer prognosis [53], which could be useful for
over 5-FU/LV alone. Second-line therapy of pancreatic cancer prognostic stratification and therapeutic decision making.
has to be considered in terms of risk benefit for the patient. If BRCA2, PALB2, ATM, or mismatch repair (hMLH1 and
the general status remains correct, considering the conflicting MSH2) gene mutations, which subsequently cause a DNA
results on the use of oxaliplatin, MM-398 when available in all damage repair deficiency, might be more sensitive to platinum
countries may be the best option for second-line treatment of or PARP inhibitors.
these patients [II, B]. Secreted protein acidic and rich in cysteine (SPARC) expression
in the peritumoural stroma has been associated with a longer OS
specific treatment of rare forms in patients receiving nab-paclitaxel + gemcitabine, compared with
those receiving gemcitabine alone. However, this was not con-
Patients with pancreatic cancer related to BRCA1 or BRCA2
firmed upon analysis of the data from the MPACT phase III trial.
mutations have been shown to be more sensitive to platinum salt
STK11 acts as a tumour suppressor gene, germline mutations
treatment. These patients may be good candidates for
of STK11 are associated with Peutz–Jeghers syndrome. In vitro
FOLFIRINOX or even 5-FU cisplatin producing more DNA
data and case reports suggest that mTOR inhibitors demonstrate
adducts [50], while trials with PARP inhibitors are ongoing in
anti-tumour activity [54].
this specific population.
The hedgehog pathway is an early and late mediator of pan-
Pancreatic carcinoma with acinar cells seems to have a better
creatic cancer tumour genesis [55]. Smoothened inhibitor sari-
prognosis. There is no specificity for the treatment of these
degib, which failed to provide any benefit in tests on non-
patients at this time; therefore, the FOLFIRINOX regimen can
selected pancreatic cancer, might be of particular interest
be used. Pancreatic acinar cell carcinoma has recently shown re-
among patients with hedgehog signalling activation. Mutations
current RAF fusions and frequent inactivation of DNA repair
in PTCH are known to activate hedgehog signalling in experi-
genes [51]: these could be targetable in the near future and lead
mental models, and are detected in 2% of patients with pancre-
to new treatment options.
atic cancer.
The human equilibrative nucleoside transporter 1 (hENT1) is
recommendations for oncological treatment of advanced/
metastatic disease responsible for the intracellular uptake of the prodrug gemcita-
bine into tumour cells. While tumour hENT1 expression is
• Biliary stenting: the endoscopic method is safer than percu- thereby presumed to be a predictive biomarker of gemcitabine
taneous insertion and is as successful as surgical hepatojeju- efficacy, contradictory data render its exact role unclear. In the
nostomy [II, B] ESPAC-3 trial, patients with high pancreatic hENT1 expression,
• Pain control is mandatory and frequently needs the help of a treated with gemcitabine, had a longer survival compared
pain specialist with those with a low expression [56]. In the phase III AIO-
v | Ducreux et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Figure 2. Treatment strategy. ChT, chemotherapy; RT, radiotherapy; 5-FU, 5-fluorouracil; LV, leucovorin; PS, performance status; ULN, upper limit of normal.
PK0104 trial, a phase III trial that compared gemcitabine/erloti- • There is no role today for personalised medicine in this
nib followed by capecitabine with capecitabine/erlotinib followed cancer [IV, C]
by gemcitabine, no difference in OS was found with respect to
hENT1 expression [57]. follow-up and long-term implications
In the recent whole-sequencing analysis [10], a significant
proportion of tumours harboured focal amplifications, many of Considering the poor prognosis of the disease upon diagnosis of
which contained druggable oncogenes (ERBB2, MET, FGFR1, a recurrence, there is no evidence that regular follow-up after
CDK6, PIK3R3, and PIK3CA), but at low prevalence among in- initial therapy with curative intent has any impact on the
dividual patients. Genomic instability co-segregated with inacti- outcome. Follow-up visits should concentrate on symptoms, nu-
vation of DNA maintenance genes (BRCA1, BRCA2, or PALB2) trition, and psycho-social support.
and a mutational signature of DNA damage repair deficiency.
Of eight patients who received platinum therapy, four of five recommendations for follow-up
individuals with these measures of defective DNA maintenance • There is no evidence that regular follow-up after initial
responded. therapy with curative intent is useful [IV, D].
key points summary of recommendations

• A few targetable mutations have been identified in pancreatic An overview of recommendations related to therapy is given in
cancer Figure 2 and Table 4.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv295 | v

by guest
on 05 February 2018
Table 4. Summary of key points and recommendations • Pain control is mandatory and frequently needs the help of a pain specialist
Incidence and epidemiology • For patients with performance status of 3/4, with significant
morbidities and a very short life expectancy: only symptomatic
• 5%–10% of pancreatic cancers are due to genetic alteration treatment can be considered
• The main risk factors are tobacco, and factors related to dietary habits • In very selected patients with ECOG performance status 2 due to
(BMI, red meat intake, low fruit and vegetables intake, diabetes, heavy tumour load, gemcitabine and nab-paclitaxel can be considered
alcohol intake) for best chance of response [II, B]
Diagnosis • For patients with performance status of 2 and/or bilirubin level higher
than 1.5× ULN: a monotherapy with gemcitabine could be considered
• Early symptoms of pancreatic cancer result from a mass effect
[I, A]
• Common presenting symptoms include jaundice, pain, weight loss,
• If the performance status of the patient is 0 or 1 and the bilirubin level
steatorrhoea
is below 1.5× ULN two types of combination chemotherapy—the
Pathology FOLFIRINOX regimen or the combination of gemcitabine and nab-
paclitaxel—should be considered [I, A]
• 95% of pancreatic cancers are adenocarcinomas
• Mucinous lesions of the pancreas have potential for malignant Personalised medicine
progression
• A few targetable mutations have been identified in pancreatic cancer
Molecular biology
• There is no role today for personalised medicine in this cancer [IV, C]
• The most frequent precursors are microscopic PanIN, followed by
Follow-up and long-term implications
IPMN and mucinous cystic neoplasm
• Multiple combinations of genetic mutations are commonly found in • There is no evidence that regular follow-up after initial therapy with
pancreatic cancers curative intent is useful [IV, D]
• Some of the recent genetic mutations discovered could become
targetable in the near future
BMI, body mass index; PanIN, pancreatic intraepithelial neoplasia;
Staging IPMN, intraductal papillary mucinous neoplasm; CT, computed
tomography; EUS, endoscopic ultrasound; MRI, magnetic resonance
• CA 19-9 is the most useful tumour marker in pancreatic cancer [IV, B]
imaging; 5-FU, 5-fluorouracil; ULN, upper limit of normal; ECOG,
• Staging of the patient is initially done by CT scan
Eastern Cooperative Oncology Group.
• EUS provides some complementary information and allows biopsy of
the tumour [II, A]
• MRI should be discussed, especially in cystic lesions [IV, C] Table 5. Levels of evidence and grades of recommendation
Treatment of localised disease (adapted from the Infectious Diseases Society of America-United
• A multidisciplinary team is necessary
• Tumour clearance should be given for all seven margins identified by Levels of evidence
the surgeon [IV, B]
I Evidence from at least one large randomised, controlled trial
• Standard lymphadenectomy should involve the removal of ≥15 lymph
of good methodological quality (low potential for bias) or
nodes to allow adequate pathologic staging of the disease [IV, A]
meta-analyses of well-conducted randomised trials without
• Adjuvant treatment is done with either gemcitabine or 5-FU folinic
heterogeneity
acid [I, A]
• No chemoradiation should be given to patients after surgery except in
suspicion of bias (lower methodological quality) or meta-
clinical trials [I, E]
analyses of such trials or of trials with demonstrated
Treatment of non-resectable disease: borderline resectable lesions heterogeneity
• Patients with borderline resectable lesions should be included in
clinical trials wherever possible
V Studies without control group, case reports, expert opinions
• In routine practice, if the patient is not included in a trial, a period of
chemotherapy followed by chemoradiation and then surgery appears Grades of recommendation
to be the best option [IV, B]
A Strong evidence for efficacy with a substantial clinical benefit,
Treatment of non-resectable disease: locally advanced disease strongly recommended
• The standard of care is 6 months of gemcitabine [I, A] B Strong or moderate evidence for efficacy but with a limited
• A minor role of chemoradiation in this subgroup of patients has been clinical benefit, generally recommended
observed [I, A] C Insufficient evidence for efficacy or benefit does not outweigh
• It is impossible to recommend any chemoradiation treatment other the risk or the disadvantages (adverse events, costs, …),
than the classical combination of capecitabine and radiotherapy [IV, C] optional
Treatment of metastatic disease
generally not recommended
• Palliative and supportive care: duodenal obstruction is preferably E Strong evidence against efficacy or for adverse outcome, never
managed by endoscopic placement of an expandable metal stent when recommended
possible, and is favoured over surgery [V, B]
• Biliary stenting: the endoscopic method is safer than percutaneous
a
insertion and is as successful as surgical hepatojejunostomy [II, B]
v | Ducreux et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
methodology 16. Ngamruengphong S, Swanson KM, Shah ND, Wallace MB. Preoperative
endoscopic ultrasound-guided fine needle aspiration does not impair survival of
These clinical practice guidelines were developed in accordance patients with resected pancreatic cancer. Gut 2015; 64: 1105–1110.
with the ESMO standard operating procedures for clinical prac- 17. Abrams RA, Lowy AM, O’Reilly EM et al. Combined modality treatment of
tice guidelines development. The relevant literature has been resectable and borderline resectable pancreas cancer: expert consensus
selected by the expert authors. A summary of recommendations statement. Ann Surg Oncol 2009; 16: 1751–1756.
is given in Table 4. Levels of evidence and grades of recommen- 18. Bockhorn M, Uzunoglu FG, Adham M et al. Borderline resectable pancreatic
cancer: a consensus statement by the International Study Group of Pancreatic
dation have been applied using the system shown in Table 5. Surgery (ISGPS). Surgery 2014; 155: 977–988.
Statements without grading were considered justified standard 19. National Comprehensive Cancer Network. NCCN Guidelines Version 2.2015
clinical practice by the experts and the ESMO faculty. This Pancreatic Adenocarcinoma. http://www.nccn.org.
manuscript has been subjected to an anonymous peer review 20. Delpero JR, Bachellier P, Regenet N et al. Pancreaticoduodenectomy for pancreatic
process. ductal adenocarcinoma: a French multicentre prospective evaluation of resection
margins in 150 evaluable specimens. HPB (Oxford) 2014; 16: 20–33.
conflict of interest 21. Delpero JR, Boher JM, Sauvanet A et al. Pancreatic adenocarcinoma with venous
involvement: is up-front synchronous portal-superior mesenteric vein resection still
MD has reported participation in advisory boards and at sympo- justified? A survey of the association francaise de chirurgie. Ann Surg Oncol 2015;
siums with Celgene. TS has reported participation to advisory 22: 1874–1883.
boards and at symposiums with Celgene and research funding 22. Mitchem JB, Hamilton N, Gao F et al. Long-term results of resection of
adenocarcinoma of the body and tail of the pancreas using radical antegrade
from Celgene. JLVL has reported participation to advisory boards
modular pancreatosplenectomy procedure. J Am Coll Surg 2012; 214: 46–52.
and at symposiums with Celgene and research funding from
23. Kooby DA, Hawkins WG, Schmidt CM et al. A multicenter analysis of distal
Celgene. The other authors have reported no potential conflicts of pancreatectomy for adenocarcinoma: is laparoscopic resection appropriate? J Am
interest. Coll Surg 2010; 210: 779–785.
24. Ricci C, Casadei R, Taffurelli G et al. Laparoscopic versus open distal
references pancreatectomy for ductal adenocarcinoma: a systematic review and meta-
analysis. J Gastrointest Surg 2015; 19: 770–781.
1. Malvezzi M, Bertuccio P, Levi F et al. European cancer mortality predictions for the 25. Tol JA, Gouma DJ, Bassi C et al. Definition of a standard lymphadenectomy in
year 2014. Ann Oncol 2014; 25: 1650–1656. surgery for pancreatic ductal adenocarcinoma: a consensus statement by the
2. Yeo TP. Demographics, epidemiology, and inheritance of pancreatic ductal International Study Group on Pancreatic Surgery (ISGPS). Surgery 2014; 156:
adenocarcinoma. Semin Oncol 2015; 42: 8–18. 591–600.
3. Larsson SC, Wolk A. Red and processed meat consumption and risk of pancreatic 26. Ragulin-Coyne E, Carroll JE, Smith JK et al. Perioperative mortality after
cancer: meta-analysis of prospective studies. Br J Cancer 2012; 106: 603–607. pancreatectomy: a risk score to aid decision-making. Surgery 2012; 152(3 Suppl.
4. Ojajarvi IA, Partanen TJ, Ahlbom A et al. Occupational exposures and pancreatic 1): S120–S127.
cancer: a meta-analysis. Occup Environ Med 2000; 57: 316–324. 27. van der Gaag NA, Rauws EA, van Eijck CH et al. Preoperative biliary drainage for
5. Maisonneuve P, Lowenfels AB. Risk factors for pancreatic cancer: a summary cancer of the head of the pancreas. N Engl J Med 2010; 362: 129–137.
review of meta-analytical studies. Int J Epidemiol 2015; 44: 186–198. 28. Neoptolemos JP, Dunn JA, Stocken DD et al. Adjuvant chemoradiotherapy and
6. Rishi A, Goggins M, Wood LD, Hruban RH. Pathological and molecular evaluation chemotherapy in resectable pancreatic cancer: a randomised controlled trial.
of pancreatic neoplasms. Semin Oncol 2015; 42: 28–39. Lancet 2001; 358: 1576–1585.
7. Wisnoski NC, Townsend CM, Jr, Nealon WH et al. 672 patients with acinar cell 29. Oettle H, Post S, Neuhaus P et al. Adjuvant chemotherapy with gemcitabine vs
carcinoma of the pancreas: a population-based comparison to pancreatic observation in patients undergoing curative-intent resection of pancreatic cancer: a
adenocarcinoma. Surgery 2008; 144: 141–148. randomized controlled trial. JAMA 2007; 297: 267–277.
8. Dudeja V, Allen PJ. Premalignant cystic neoplasms of the pancreas. Semin Oncol 30. Neoptolemos JP, Stocken DD, Bassi C et al. Adjuvant chemotherapy with
2015; 42: 70–85. fluorouracil plus folinic acid vs gemcitabine following pancreatic cancer resection:
9. Esposito I, Konukiewitz B, Schlitter AM, Klöppel G. Pathology of pancreatic ductal a randomized controlled trial. JAMA 2010; 304: 1073–1081.
adenocarcinoma: facts, challenges and future developments. World J 31. Marechal R, Bachet JB, Mackey JR et al. Levels of gemcitabine transport and
Gastroenterol 2014; 20: 13833–13841. metabolism proteins predict survival times of patients treated with gemcitabine for
10. Waddell N, Pajic M, Patch AM et al. Whole genomes redefine the mutational pancreatic adenocarcinoma. Gastroenterology 2012; 143: 664–674.
landscape of pancreatic cancer. Nature 2015; 518: 495–501. 32. Assifi MM, Lu X, Eibl G et al. Neoadjuvant therapy in pancreatic adenocarcinoma:
11. Nawaz H, Fan CY, Kloke J et al. Performance characteristics of endoscopic a meta-analysis of phase II trials. Surgery 2011; 150: 466–473.
ultrasound in the staging of pancreatic cancer: a meta-analysis. JOP 2013; 14: 33. Gillen S, Schuster T, Meyer Zum Buschenfelde C et al. Preoperative/neoadjuvant
484–497. therapy in pancreatic cancer: a systematic review and meta-analysis of response
12. Al-Hawary MM, Francis IR, Chari ST et al. Pancreatic ductal adenocarcinoma and resection percentages. PLoS Med 2010; 7: e1000267.
radiology reporting template: consensus statement of the Society of Abdominal 34. Denost Q, Laurent C, Adam JP et al. Pancreaticoduodenectomy following
Radiology and the American Pancreatic Association. Radiology 2014; 270: chemoradiotherapy for locally advanced adenocarcinoma of the pancreatic head.
248–260. HPB (Oxford) 2013; 15: 716–723.
13. Bipat S, Phoa SS, van Delden OM et al. Ultrasonography, computed tomography 35. Landry J, Catalano PJ, Staley C et al. Randomized phase II study of gemcitabine
and magnetic resonance imaging for diagnosis and determining resectability of plus radiotherapy versus gemcitabine, 5-fluorouracil, and cisplatin followed by
pancreatic adenocarcinoma: a meta-analysis. J Comput Assist Tomogr 2005; 29: radiotherapy and 5-fluorouracil for patients with locally advanced, potentially
438–445. resectable pancreatic adenocarcinoma. J Surg Oncol 2010; 101: 587–592.
14. Callery MP, Chang KJ, Fishman EK et al. Pretreatment assessment of resectable 36. Hammel P, Huguet F, Van Laethem J-L et al. Comparison of chemoradiotherapy
and borderline resectable pancreatic cancer: expert consensus statement. Ann (CRT) and chemotherapy (CT) in patients with a locally advanced pancreatic cancer
Surg Oncol 2009; 16: 1727–1733. (LAPC) controlled after 4 months of gemcitabine with or without erlotinib: Final
15. Wong JC, Lu DS. Staging of pancreatic adenocarcinoma by imaging studies. Clin results of the international phase III LAP 07 study. J Clin Oncol 2013;31(suppl):
Gastroenterol Hepatol 2008; 6: 1301–1308. abstr LBA4003.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv295 | v

by guest
on 05 February 2018
37. Shinchi H, Takao S, Noma H et al. Length and quality of survival after external- 47. Conroy T, Desseigne F, Ychou M et al. FOLFIRINOX versus gemcitabine for
beam radiotherapy with concurrent continuous 5-fluorouracil infusion for metastatic pancreatic cancer. N Engl J Med 2011; 364: 1817–1825.
locally unresectable pancreatic cancer. Int J Radiat Oncol Biol Phys 2002; 53: 48. Von Hoff DD, Goldstein D, Renschler MF. Albumin-bound paclitaxel plus
146–150. gemcitabine in pancreatic cancer. N Engl J Med 2014; 370: 479–480.
38. Sultana A, Tudur Smith C, Cunningham D et al. Systematic review, including meta- 49. Oettle H, Riess H, Stieler JM et al. Second-line oxaliplatin, folinic acid, and fluorouracil
analyses, on the management of locally advanced pancreatic cancer using versus folinic acid and fluorouracil alone for gemcitabine-refractory pancreatic cancer:
radiation/combined modality therapy. Br J Cancer 2007; 96: 1183–1190. outcomes from the CONKO-003 trial. J Clin Oncol 2014; 32: 2423–2429.
39. Chauffert B, Mornex F, Bonnetain F et al. Phase III trial comparing initial 50. Sonnenblick A, Kadouri L, Appelbaum L et al. Complete remission, in BRCA2
chemoradiotherapy (intermittent cisplatin and infusional 5-FU) followed by mutation carrier with metastatic pancreatic adenocarcinoma, treated with cisplatin
gemcitabine vs. gemcitabine alone in patients with locally advanced non- based therapy. Cancer Biol Ther 2011; 12: 165–168.
metastatic pancreatic cancer: a FFCD-SFRO study. J Clin Oncol 2006; 24(suppl): 51. Chmielecki J, Hutchinson KE, Frampton GM et al. Comprehensive genomic
abstr 4008. profiling of pancreatic acinar cell carcinomas identifies recurrent RAF fusions and
40. Loehrer PJ, Sr, Feng Y, Cardenes H et al. Gemcitabine alone versus gemcitabine frequent inactivation of DNA repair genes. Cancer Discov 2014; 4: 1398–1405.
plus radiotherapy in patients with locally advanced pancreatic cancer: an Eastern 52. Yachida S, Jones S, Bozic I et al. Distant metastasis occurs late during the genetic
Cooperative Oncology Group trial. J Clin Oncol 2011; 29: 4105–4112. evolution of pancreatic cancer. Nature 2010; 467: 1114–1117.
41. Mukherjee S, Hurt CN, Bridgewater J et al. Gemcitabine-based or capecitabine- 53. Blackford A, Serrano OK, Wolfgang CL et al. SMAD4 gene mutations are
based chemoradiotherapy for locally advanced pancreatic cancer (SCALOP): a associated with poor prognosis in pancreatic cancer. Clin Cancer Res 2009; 15:
multicentre, randomised, phase 2 trial. Lancet Oncol 2013; 14: 317–326. 4674–4679.
42. Stark A, Hines OJ. Endoscopic and operative palliation strategies for pancreatic 54. Klumpen HJ, Queiroz KC, Spek CA et al. mTOR inhibitor treatment of pancreatic
ductal adenocarcinoma. Semin Oncol 2015; 42: 163–176. cancer in a patient with Peutz-Jeghers syndrome. J Clin Oncol 2011; 29:
43. Ripamonti CI, Santini D, Maranzano E et al. Management of cancer pain: ESMO e150–e153.
Clinical Practice Guidelines. Ann Oncol 2012; 23(Suppl 7): vii139–vii154. 55. Hidalgo M, Maitra A. The hedgehog pathway and pancreatic cancer. N Engl J Med
44. Burris HA, III, Moore MJ, Andersen J et al. Improvements in survival and clinical 2009; 361: 2094–2096.
benefit with gemcitabine as first-line therapy for patients with advanced pancreas 56. Greenhalf W, Ghaneh P, Neoptolemos JP et al. Pancreatic cancer hENT1
cancer: a randomized trial. J Clin Oncol 1997; 15: 2403–2413. expression and survival from gemcitabine in patients from the ESPAC-3 trial. J Natl
45. Ciliberto D, Botta C, Correale P et al. Role of gemcitabine-based combination Cancer Inst 2014; 106: djt347.
therapy in the management of advanced pancreatic cancer: a meta-analysis of 57. Ormanns S, Siveke JT, Heinemann V et al. pERK, pAKT and p53 as tissue
randomised trials. Eur J Cancer 2013; 49: 593–603. biomarkers in erlotinib-treated patients with advanced pancreatic cancer: a
46. Moore MJ, Goldstein D, Hamm J et al. Erlotinib plus gemcitabine compared with translational subgroup analysis from AIO-PK0104. BMC Cancer 2014; 14: 624.
gemcitabine alone in patients with advanced pancreatic cancer: a phase III trial of 58. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
the National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 2007; among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
25: 1960–1966. 139–144.
v | Ducreux et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Biliary cancer: ESMO Clinical Practice Guidelines for

J. W. Valle1, I. Borbath2, S. A. Khan3, F. Huguet4, T. Gruenberger5 & D. Arnold6 On behalf of the
1
Department of Medical Oncology, The Christie NHS Foundation Trust, Institute of Cancer Sciences, University of Manchester, Manchester, UK; 2Department of
Gastroenterology, Cliniques Universitaires Saint-Luc, Brussels, Belgium; 3Hepatology and Gastroenterology Section, Department of Medicine, Imperial College, London,
UK; 4Department of Radiation Oncology, Tenon Hospital, APHP, University Paris VI, Paris, France; 51st Department of Surgery, Rudolfstiftung Hospital, Vienna, Austria;
6
Instituto CUF de Oncologia (I.C.O.), Lisbon, Portugal
incidence and epidemiology 4) the increasing burden of chronic liver disease (e.g. viral hepa-
titis and fatty liver disease as specific risk factors for iCCA
Biliary tract cancer (BTC), comprising <1% of all human [8–10]); and
cancers and ∼10%–15% of all primary liver cancers, presents 5) the potential role of environmental toxins [11, 12].
mostly in the seventh decade with a small male predominance
(male:female ratio of 1.2–1.5:1.0) [1]. They are subclassified as Furthermore, increased laparoscopic cholecystectomy rates
intrahepatic cholangiocarcinoma (iCCA), originating from the over the past few decades have significantly reduced the preva-
clinical practice
biliary tree within the liver, and extrahepatic cholangiocarci- lence of gallstone disease, a stronger risk factor for eCCA than
guidelines
noma (eCCA), outside the liver parenchyma; the latter is further iCCA [13].
subdivided into perihilar cholangiocarcinoma ( pCCA or
Klatskin tumour) and distal cholangiocarcinoma (dCCA) [2], gallbladder cancer
with a frequency of 10%–20% iCCA, 50% pCCA and 30%–40% The incidence of gallbladder cancer (GBC) in Western Europe
eCCA. and the USA is low (1.6–2.0/100 000); however, it is a significant
health problem in Chile, India and Central/Eastern Europe. In
cholangiocarcinoma the Valdivia region of Chile, GBC incidence reaches 24.3/
In Europe, USA and Australasia, the incidence of cholangiocar- 100 000 in females (and 8.6/100 000 in males) [14]. Gallstones
cinoma (CCA) is low (0.3–3.5/100 000); in parts of the world are the strongest risk factor for GBC; other risk factors include
where liver fluke infection is common (e.g. Thailand, China and porcelain gallbladder, gallbladder polyps, primary sclerosing
Korea), rates are much higher [3]. Northeast Thailand has the cholangitis, chronic infection (e.g. salmonella typhi), congenital
highest CCA rate in the world, with an annual incidence of 90/ malformations and obesity.
100 000, accounting for >80% of all primary liver cancers [4].
CCA incidence and mortality rates have increased overall in diagnosis and pathology/molecular
the past few decades in most Western countries; specifically, biology
iCCA rates are rising and eCCA rates falling [5], with the excep-
tion of Denmark, Norway and the Czech Republic, where iCCA When diagnosing a BTC, it is important to distinguish the
rates have fallen. CCA rates in Asia overall have remained static. subtype (iCCA, pCCA, dCCA or GBC) as every subtype has its
These trends may be explained by: own specific characteristics, requiring individual workup. Overall,
however, the best diagnostic tool is magnetic resonance imaging
1) improved diagnostic tools and imaging; (MRI) with magnetic resonance cholangiopancreatography
2) misclassification ( particularly of pCCA during serial updates (MRCP), contrast-enhanced and diffusion-weighted imaging
of the International Classification of Disease [6]); [15]. Computed tomography (CT) is generally less useful.
3) changing migration patterns in the West [7]; Pathology diagnosis should be obtained before any non-surgi-
cal treatment modality (not essential in patients planned for
curative surgery where radiological features are characteristic).
6962 Viganello-Lugano, Switzerland. Endoscopic retrograde cholangiopancreatography (ERCP)-
E-mail: clinicalguidelines@esmo.org guided biopsies are preferred to biliary brush cytology and
†
should be carried out whenever possible [III, A]. Endoscopic
This publication supersedes the previously published version—Ann Oncol 2011; 22 ultrasound (EUS)-guided fine needle aspiration (FNA) is also
(Suppl 6): vi40–vi44. useful for obtaining microbiopsies [16] [II, C] and may be

by guest
on 05 February 2018
considered if ERCP-guided brush cytology or biopsies are nega- management of local/locoregional
tive or inconclusive. Cases of tumour seeding along the FNA disease
needle track have been reported [17]; the exact level of risk is un-
certain, though it appears to be low. Decisions to undertake The therapeutic strategy varies for each type of BTC depending
biopsy should be made in a multidisciplinary setting, particular- on its site of origin. A treatment algorithm is shown in Figure 1.
ly in patients with potentially resectable tumours. Advances in
pathology allow identification of distinct pathological subgroups intrahepatic cholangiocarcinoma
based on immunohistochemistry, reflected in the updated iCCAs usually arise within normal background liver paren-
World Health Organization (WHO) classification [18], high- chyma, and their radiological appearance is, most typically, of a
lighting the importance of tissue acquisition. mass-forming arterially enhancing tumour. Radical surgical
Serum carbohydrate antigen (CA)19-9 (with cut-off >129 removal with clear margins is the only potentially curative
U/ml) has some added value [III, C] [19]. In addition, there is therapy [III, A]. There are well-known prognostic parameters
an increasing understanding of different genomic profiles across that should be taken into account when assessing prognosis in-
BTC [20], and although currently these differences do not direct cluding the presence of lymph node involvement; this has led to
therapy, they may do so in future. the recommendation of routine lymphadenectomy at the level
of the hepato-duodenal ligament during surgery [II, A] [25].
staging and risk assessment Several other factors including size and number of tumours,
grade, the presence of satellite nodules, vascular and/or peri-
Staging needs to take into account the patient’s performance neural invasion should be reported by the pathologist to guide
status [using WHO or European Cooperative Oncology Group decisions regarding adjuvant therapy, although robust evidence
(ECOG) scores], past medical history/co-morbidities and liver for its use is lacking [IV, B].
function tests (LFTs). Imaging consists of MRI (for assessment
of tumour [T]-stage and bile duct involvement), thorax CT (me- perihilar cholangiocarcinoma (Klatskin tumour)
tastases (M)-stage) and EUS (lymph node (N)-stage). Whenever
necessary (e.g. inconclusive MRCP), ERCP or percutaneous Diagnosis of a pCCA and assessment of resectability according
transhepatic cholangiography (PTC) should be carried out to to the Bismuth–Corlette classification can only be determined
improve assessment of T-stage, as this is crucial for surgical de- in a considerable number of patients with surgical exploration.
cision-making [IV, B]. Contrast CT is effective in defining the It is important that, for patients presenting with jaundice, initial
relationship between the tumour and the vasculature ( portal radiological imaging is carried out before an ERCP or PTC is
vein and hepatic artery). The utility of positron emission tomog- undertaken, as the inserted drains/stents obscure the diagnosis
raphy (PET)-CT is controversial and should only be used on a and assessment of the extent of disease [III, A]. The practice of
case-by-case basis. biliary drainage before resection (versus immediate surgery)
Staging is carried out according to the TNM 2010 system and remains controversial and has to be decided on by the treating
is specific for every subtype of BTC (see Table 1) [21]; pCCA team taking into consideration non-tumour related factors, such
may be further subclassified according to the Bismuth–Corlette as performance status and co-morbidities [II, B]. The anatomic-
classification (Table 2). ally longer left hepatic duct before segmental distribution
Risk assessment is different for every CCA subtype. When usually implies that an extended right hemi-hepatectomy is ne-
considering pCCA, patients suffering from primary sclerosing cessary for curative intent and may require portal vein embolisa-
cholangitis (PSC) in the Western world and patients with hepa- tion (including the segment IV branches) to induce
tobiliary flukes or hepatolithiasis in Asian countries are at an hypertrophy of the future liver remnant ( preservable segments
increased risk. Guidelines for surveillance of PSC patients are II and III) [IV, A] [26]. Segment I, which drains into the ductal
available [23]. Cirrhosis and hepatotropic viruses are risk factors bifurcation, where the cancer lies, has to be removed in any
for iCCA, with odds ratios (ORs) of 22.92 [95% confidence curative procedure. Vascular resections at the hilum are possible
interval (CI): 18.24–28.79] for cirrhosis, 4.84 (95% CI: 2.41– but their invasion affects prognosis. Lymphadenectomy should
9.71) for hepatitis C and 5.10 (95% CI: 2.91–8.95) for hepatitis be a standard addition in every CCA surgery.
B, according to a recent meta-analysis [8]. The presence of Liver transplantation in locally unresectable cases has been
iCCA should be considered in cirrhotic patients although the explored in a multidisciplinary approach including a strategy
development of a hepatocellular carcinoma, for which patients consisting of neoadjuvant chemoradiotherapy followed by liver
are already undergoing screening, is more likely. Screening for transplantation by the Mayo Clinic; however, this has not
CCA in newly defined at-risk groups (e.g. with obesity and the become the standard of care [III, C]. The ongoing French phase
metabolic syndrome [8]) has not been established. III TRANSPHIL trial compares this strategy with standard sur-
Patients with premalignant lesions predisposing to GBC gical resection.
warrant surveillance; the risk of malignancy is related to the size
of gallbladder polyps (the most prevalent finding), which are distal cholangiocarcinoma
often found incidentally. Lesions ≥20 mm should be managed In contrast to the other forms of CCA, this type requires the
as GBC after completion of staging investigations. Ultrasound removal of the pancreatic head, usually a partial duodeno-pan-
surveillance is recommended for polyps measuring 6–9 mm (6- createctomy (PDP) with extended bile duct resection up to the
monthly for 1 year, then annually for 5 years), with resection hilum. PDP is a standard procedure that includes draining
only in enlarging polyps (to 10–20 mm in size) [24]. lymph node dissection and reconstruction of the stomach and
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw324 | v

by guest
on 05 February 2018
Table 1. The AJCC/UICC staging of cholangiocarcinoma and gallbladder cancer [21]
clinical practice guidelines

v | Valle et al.
Cholangiocarcinoma Gallbladder cancer

Cholangiocarcinoma - intrahepatic Cholangiocarcinoma - perihilar Cholangiocarcinoma - distal Gallbladder cancer
Primary tumour (T) Primary tumour (T) Primary tumour (T) Primary tumour (T)
TX Primary tumour TX Primary tumour TX Primary tumour TX Primary tumour
cannot be cannot be cannot be cannot be
assessed assessed assessed assessed
T0 No evidence of T0 No evidence of T0 No evidence of T0 No evidence of
primary tumour primary tumour primary tumour primary tumour
Tis Carcinoma in Tis Carcinoma in Tis Carcinoma in Tis Carcinoma in
situ (intraductal situ situ situ
tumour)
T1 Solitary tumour T1 Tumour T1 Tumour T1 Tumour invades
without vascular confined to the confined to the the lamina
invasion bile duct, with bile duct propria or
extension up to histologically muscular layer
the muscle layer
or fibrous tissue
T2a Solitary tumour T2a Tumour invades T2 Tumour invades T1a Tumour invades
with vascular beyond the wall beyond the wall the lamina
invasion of the bile duct of the bile duct propria
to the
surrounding
adipose tissue
T2b Multiple T2b Tumour invades T3 Tumour invades T1b Tumour invades
tumours, with the adjacent the gallbladder, the muscular
or without hepatic pancreas, layer
vascular parenchyma duodenum or
invasion other adjacent
organs without
involvement of
the coeliac axis,
or the superior
Volume 27 | Supplement 5 | September 2016
mesenteric
artery
T3 Tumour T3 Tumour invades T4 Tumour T2 Tumour invades
perforating the unilateral involves the the
visceral branches of the coeliac axis, or perimuscular
peritoneum or portal vein or the superior connective
involving the the hepatic mesenteric tissue; no
local artery artery extension
extrahepatic beyond the
structures by serosa or into
Annals of Oncology
direct invasion the liver
T4 Tumour with T4 Tumour invades T3 Tumour
periductal the main portal perforates the
invasion vein or its serosa (visceral
branches peritoneum)
bilaterally; or the and/or directly
common hepatic invades the liver

by guest
on 05 February 2018
artery; or the and/or one
second-order other adjacent
Annals of Oncology
biliary radicals organ or

bilaterally; or structure, such
unilateral as the stomach,
second-order duodenum,
biliary radicals colon, pancreas,
with omentum or
contralateral extrahepatic bile
portal vein or ducts
hepatic artery
involvement
T4 Tumour invades
the main portal
vein or the
hepatic artery or
invades two or
more
extrahepatic
organs or
structures
Regional lymph nodes (N) Regional lymph nodes (N) Regional lymph nodes (N) Regional lymph nodes (N)
NX Regional lymph NX Regional lymph NX Regional lymph NX Regional lymph
nodes cannot be nodes cannot be nodes cannot be nodes cannot be
assessed assessed assessed assessed
N0 No regional N0 No regional N0 No regional N0 No regional

lymph node lymph node lymph node lymph node
metastasis metastasis metastasis metastasis
N1 Regional lymph N1 Regional lymph N1 Regional lymph N1 Metastases to
node metastasis node metastasis node metastasis nodes along the
present (including nodes cystic duct,
along the cystic common bile
duct, common duct, hepatic
bile duct, artery and/or
hepatic artery portal vein
and portal vein)
N2 Metastasis to N2 Metastases to
periaortic, periaortic,
doi:10.1093/annonc/mdw324 | v
pericaval, pericaval,
superior superior
mesenteric mesenteric
artery and/or artery and/or
coeliac artery coeliac artery
lymph nodes lymph nodes
Distant metastasis (M) Distant metastasis (M) Distant metastasis (M) Distant metastasis (M)
M0 No distant M0 No distant M0 No distant M0 No distant
metastasis metastasis metastasis metastasis
M1 M1 Distant M1 Distant M1 Distant
metastasis metastasis metastasis
Continued

by guest
on 05 February 2018
the remaining pancreas in various ways to achieve macroscopic
Edge et al. [20]. Used with the permission of the American Joint Committee on Cancer (AJCC), Chicago, IL, USA. The original source for this material is the AJCC Cancer Staging Handbook, 7th edition
M0
M0
M0
M0
M0
M0
M0
M1
cure. The prognosis of dCCA is better than that of adenocarcin-
oma of the head of the pancreas [III, A] [27].
Any N
N0–1
Primary tumour (T)
N0
N0
N0
N0
N1
N2
Gallbladder cancer
gallbladder cancer
Stage grouping
GBC has two typical presentations: either (a) incidentally diag-
nosed in the histological workup of simple cholecystectomies or
Any T
Any T
T1–3
Tis (b) as a symptomatic right upper quadrant tumour at an
T1
T2
T3
T4
Gallbladder cancer
advanced stage.
The former requires staging with appropriate imaging (MRI
Stage IIIA
Stage IVA
Stage IVB
or CT) and detailed histopathological analysis to decide whether
Stage IIIB
Stage II
Stage 0
Stage I
further resection is necessary, including T-stage, cystic duct

margin, involvement of resected lymph nodes, grade, perineural
and/or vascular invasion. Every T-stage above T1a and positivity
of any mentioned parameters requires a reoperation where a
M0
M0
M0
M0
M0
M0
M0
M0
M1
segment IVb/V liver resection together with a ligamentary lym-
phadenectomy should be carried out [II, A] [28]. If the gallblad-
der was not removed with a bag during laparoscopic resection
Cholangiocarcinoma - distal
Any N
Any N
or the gallbladder perforated (an adverse prognostic factor), the

Primary tumour (T)
N0
N0
N0
N0
N1
N1
N1
Stage grouping
port sites should also be resected [IV, A].

If GBC is diagnosed during imaging (for symptomatic
patients) or when patients present with jaundice, evaluation of
Any T
Tis
potential resectability is the key factor. Advanced T-stage (in-

T1
T2
T3
T1
T2
T3
T4
cluding T4 tumours) is not a contraindication for resectability

provided they are located in the fundus; these tumours require
Stage IIA
Stage IIB
Stage IV
Stage IA
Stage III
Stage IB
major liver resection with potential resection of the transverse

Stage 0
colon. Achieving a curative resection of an advanced tumour

located at the infundibulum is much more difficult, because it
requires the resection of the bile duct, the duodenal bulb and,
potentially, the pancreatic head together with a major hepatect-
M0
M0
M0
M0
M0
M0
M0
M1
omy, especially if right-sided vessels (right hepatic artery, right

AJCC, American Joint Committee on Cancer; UICC, Union for International Cancer Control.
portal vein) are involved [III, A].

Cholangiocarcinoma - perihilar
Any N
The prognosis of a gallbladder cancer, if not diagnosed at a

N0–1
Primary tumour (T)
N0
N0
N0
N0
N1
N2
very early stage, is somewhat inferior to all other types of CCA.

Stage grouping
(2010) published by Springer Science and Business Media LLC, www.springer.com.
adjuvant treatment. The high rates of local and distant

T2a–b
Any T
Any T
T1–3
recurrence following surgery justify the consideration of an

Tis
T1
T3
T4
adjuvant treatment. The role of adjuvant chemoradiotherapy in

this setting is not well defined because of a lack of data from
Stage IIIA
Stage IVA
Stage IVB
Stage IIIB
randomised trials. Indeed, most of the published studies are

Stage II
Stage 0
Stage I
retrospective, including small numbers of patients with a mix of

gallbladder and bile duct tumours. In a recent meta-analysis of
published data, adjuvant chemotherapy or chemoradiotherapy
appears to be associated with a survival benefit in patients with
M0
M0
M0
M0
M0
M0
M1
BTC with lymph node-positive disease or with microscopically

metastasis
Distant
present
involved margins (R1 resection) [29]. However, major limitations

(including selection bias, heterogeneity of treatments, stage
Any N
Cholangiocarcinoma - intrahepatic
migration over time and variable data quality) preclude definitive

N0
N0
N0
N0
N0
N1
conclusions. Moreover, a differentiation between the two

modalities (chemotherapy versus chemoradiotherapy) have not
been compared in this setting. When employed, the
Any T
Any T
Cholangiocarcinoma
Primary tumour (T)

Table 1. Continued
Tis
recommended dose of radiotherapy is 45 Gy in fractions of 1.8 or

T1
T2
T3
T4
2 Gy with concurrent 5-fluorouracil (5-FU) or capecitabine [IV,

Stage grouping
C]. Recently, two prospective trials have assessed a regimen

Stage IVA
Stage IVB
combining a gemcitabine-based adjuvant chemotherapy followed

Stage III
Stage II
Stage 0
Stage I
by chemoradiotherapy with concurrent 5-FU [30, 31].

The results of two phase III studies that have completed
accrual and are awaiting maturation of data [BilCap
v | Valle et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
(observation versus adjuvant capecitabine; UK, NCT00363584) In the absence of level-1 data, the multidisciplinary team may
and Prodige-12 (observation versus gemcitabine/oxaliplatin; offer adjuvant therapy (radiotherapy, chemoradiotherapy or
France, NCT01313377)] are awaited and are likely to define chemotherapy alone) to patients on the best available evidence
future adjuvant strategies. A further study [ACTICCA-1 (obser- and only after a risk–benefit assessment.
vation versus cisplatin/gemcitabine; Germany, NCT02170090)]
is open and recruiting patients.
management of advanced and/or
metastatic disease
Table 2. The Bismuth–Corlette classification of perihilar
cholangiocarcinoma radiation therapy
Type I Tumour involves the common hepatic duct The role of radiation therapy remains unclear in the treatment
Type II Tumour involves the bifurcation of the common hepatic of locally advanced but non-metastatic BTC.
duct Chemoradiotherapy has been considered a possible option
Type IIIa Tumour involves the right hepatic duct according to non-randomised studies with median survival rates
Type Tumour involves the left hepatic duct between 9 and 14 months. The French FFCD 9902 phase III
IIIb trial compared chemoradiotherapy with chemotherapy alone in
Type IV Tumour involves both the right and left hepatic ducts this setting [32]. Patients were randomised between chemora-
diotherapy (50 Gy with concurrent 5-FU and cisplatin) or
Bismuth et al. [22]. Reprinted with permission. chemotherapy with a combination of gemcitabine and oxalipla-
tin (GemOx). This trial was closed before the completion of
Biliary tract cancer
Early stage Locally-advanced Metastatic
Surgery1 Systemic chemotherapy2

• First-line combination chemotherapy (PS0-1)4
• First-line gemcitabine monotherapy (PS2)
+/–
• Second-line chemotherapy | No standard
• Targeted therapy | No standard
Clinical trials where possible
Adjuvant
chemo-
radiotherapy3
Via MDT
+/– Loco-regional therapy2

• Radiotherapy
Adjuvant • 90Y-radioembolisation (iCCA)
chemotherapy3
Best supportive
Surveillance care
1 Special considerations:
• Need for pre-operative biliary drainage
• Avoid percutaneous biopsy in resectable disease
• Assess Future Liver Remnant
• Assess need for Portal Vein Embolisation
• Neoadjuvant approach (selected cases)
• Completion surgery for incidental gallbladder cancer of T-stage T1b and above
2 Option of salvage surgery should be considered in responding patients with initially inoperable disease
3 Level of recommendation IV,C
4 Cisplatin and gemcitabine [category IA], other gemcitabine-based combination [category IIB]
Figure 1. Algorithm for the management of patients with biliary tract cancer. MDT, multidisciplinary team; PS, performance status; iCCA, intrahepatic cho-
langiocarcinoma.
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw324 | v

by guest
on 05 February 2018
Screening
• Only at-risk populations should be considered for screening, e.g. patients with PSC (screening guidelines in PSC are available)
Diagnosis
• Abdominal ultrasound may be useful for the initial examination (identification of biliary obstruction)
• MRI and MRCP should be carried out before any biliary intervention; CT is less useful
• ERCP allows relief of bile duct obstruction (by stenting); brush cytology and biopsies should be carried out
• For patients deemed suitable for surgery with radical intent, a biopsy is not obligatory (brush cytology may be available). A biopsy should be restricted to
selected cases (e.g. equivocal lesion) and only after discussion at a specialist hepatobiliary MDT; if so, EUS-guided biopsy is preferred and percutaneous
sampling should be avoided
• For patients with advanced/inoperable disease, histological/cytological confirmation is essential; it may be obtained at EUS or metastatic lesions can be
biopsied percutaneously (ultrasound or CT guided)
• FDG-PET imaging has no established role in the diagnosis of BTC
• Baseline CA19-9 should be interpreted with caution and is best used to guide treatment and follow-up; it may have a prognostic value in the absence of
biliary obstruction
Staging
• The established staging system for biliary tract cancer is the one developed by the TNM committee of the AJCC/UICC (currently 7th Edition) with
subclassifications for GBC and iCCA, pCCA and dCCA
• CT (including thorax and pelvis) allows evaluation of distant metastases and vessel involvement
• MRI plus MRCP is useful for T-staging
• EUS helps to clarify N-stage (± nodal biopsy) and adds information regarding vessel involvement
• FDG-PET scan is not routinely recommended for the staging of BTC
• Staging laparoscopy may be considered on an individual basis to exclude the presence of peritoneal metastases if it will influence the decision to proceed
with major resection (e.g. locally advanced GBC)
• Pathology examination and reporting of surgically-resected specimens should follow standardised reporting tools (minimum dataset)
Treatment
Curative
• Radical surgery (with lymphadenectomy) is the only curative treatment of BTC; the exact nature and extent of surgery will depend on tumour subtype/
location and should be agreed at a specialist hepatobiliary multidisciplinary tumour board
• Surgery involving hepatic resection will need to take into account the future liver remnant and may require portal vein embolisation
• For patients with incidentally diagnosed GBC (post-cholecystectomy), reoperation with radical intent should be considered for stage T1b and above (±
resection of port sites)
• Adjuvant therapy (radiotherapy, chemoradiotherapy or chemotherapy alone) may be offered to patients on the understanding that the evidence base is
weak and only after risk–benefit assessment; participation in clinical trials should be encouraged
• Neoadjuvant therapy and liver transplant (Mayo Clinic protocol) in early-stage hilar CCA remains investigational; participation in clinical trials should
be encouraged
• Patients with initially inoperable, non-metastatic disease should be rediscussed at the multidisciplinary tumour board with a view to salvage surgery in
the event of a good response to systemic and/or locoregional treatment, including participation in clinical trials
Palliative
• Systemic chemotherapy is the treatment of choice for patients with locally advanced or inoperable disease; combination chemotherapy for PS 0-1
patients and monotherapy for PS 2 patients
• Cisplatin/gemcitabine is the reference chemotherapy regimen for good PS (0-1) patients; oxaliplatin may be substituted for cisplatin where there is a
concern about renal function
• Gemcitabine monotherapy may be considered for PS 2 patients
• There is no established second-line chemotherapy regimen; patients should be encouraged to participate in clinical trials
• There is no established evidence to support the use of targeted therapies; patients should be encouraged to participate in clinical trials
• Radiotherapy may be considered in patients with localised disease, after first-line chemotherapy; patients should be encouraged to participate in clinical
trials
• Radioembolisation may be considered in patients with inoperable iCCA, usually after first-line chemotherapy; patients should be encouraged to
participate in clinical trials
Best supportive care
• Biliary obstruction is a common occurrence in BTC; establishment of biliary drainage and subsequent stenting should be carried out
• When endoscopic stenting is not possible, percutaneous transhepatic drainage is recommended
Continued
v | Valle et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Table 3. Continued
• In patients with a life expectancy of >3 months, a metal prosthesis is preferred; some patients require repeat stenting on multiple occasions—anticipation
for such an event is required in the planning of stent placement
• Sepsis secondary to biliary obstruction is common and needs to be treated accordingly
• Patients should be advised on the likely duration of stent patency; and of symptoms and signs that are indicative of biliary obstruction or infection and
what they need to do in such an event
• Patients should have full access to palliative care and symptom management (including pain control)
• Patients should have a designated point of contact within the multidisciplinary team for advice and support (e.g. nurse specialist)
Follow-up and personalised medicine
• There is no universal ‘standard’ follow-up schedule; major centres employ a strategy of clinical examination, blood tests (including tumour markers) and
CT scan; the intensity of this schedule should be agreed with the patient depending on the stage of disease
• After potentially curative treatment, follow-up should consist of 3-monthly visits during the first 2 years after therapy including clinical examination,
laboratory investigation (including LFTs and LDH), tumour markers (CEA or CA19-9 if one/both were known to be elevated before surgery) and CT
scan of the thorax, abdomen and pelvis. Visits can be extended to 6-monthly thereafter and prolonged to yearly visits after 5 years of follow-up
• During systemic or locoregional therapy for advanced disease, follow-up should be conducted at a frequency of 8–12 weeks to allow the best assessment
of treatment efficacy or as required for disease-related complications. CA19-9 or CEA may be used to monitor the course of the disease if one/both are
known to be secreted
• Patients should be encouraged to participate in clinical trials and/or tissue biobanking aimed at defining specific disease subgroups with a view to future
risk stratification or selection for specific therapies
• Before participating in clinical trials, patients should have access to all the information required to make an informed decision to give consent, including
potential risks and benefits; as well as support for them and their carers in making such decisions
PSC, primary sclerosing cholangitis; CT, computed tomography; MRCP, magnetic resonance cholangiopancreatography; MRI, magnetic resonance
imaging; ERCP, endoscopic retrograde cholangiopancreatography; MDT, multidisciplinary team; EUS, endoscopic ultrasound; FDG-PET,
fluorodeoxyglucose positron emission tomography; BTC, biliary tract cancer; CA, carbohydrate antigen; TNM, tumour-node metastasis; AJCC, American
Joint Committee on Cancer; UICC, Union for International Cancer Control; GBC, gallbladder cancer; iCCA, intrahepatic cholangiocarcinoma; pCCA,
perihilar cholangiocarcinoma; dCCA, distal cholangiocarcinoma; CCA, cholangiocarcinoma; PS, performance status; LFTs, liver function tests; LDH,
lactose dehydrogenase; CEA, carcinoembryonic antigen.
planned inclusions, after enrolment of 34 patients. There was an these studies has shown that patients benefit from the combin-
advantage for GemOx, both in terms of progression-free sur- ation independent of age (<65 versus ≥65 years), gender,
vival (median PFS, 11 versus 5.8 months) and overall survival primary tumour site (intra- versus extrahepatic versus gallblad-
(OS, 20 versus 13.5 months). der versus ampullary), stage of disease (locally advanced versus
Recently, intensity-modulated radiotherapy (IMRT) was metastatic) and prior therapy (surgery or stenting); however,
shown to allow safe dose escalation [33]. Whether a dose escal- patients with performance status 2 may derive the least benefit
ation could improve the local control and survival remains from the combination [39]. In patients where there is a concern
unknown. about renal function, oxaliplatin may be substituted for cisplatin
Experience is growing in the use of radioembolisation using [II, B]; in patients with performance status 2, gemcitabine
90
Y-microspheres for patients with iCCA. Prospective, rando- monotherapy may be considered [I, B].
mised data are lacking; a pooled analysis of 12 studies including There is no established second-line systemic therapy follow-
298 patients showed a median OS of 15.5 months and response ing progression after first-line treatment although fluoropyrimi-
rate of 28% [34]. Importantly, 7/73 (10%) patients in three dine-based therapy (either in monotherapy or in combination
selected studies were converted to resectable disease, highlighting with other cytotoxics) is sometimes used [III, C]. A systematic
the importance of reassessment of patients in the multidisciplin- review including 761 patients showed disappointing median
ary team in the event of a good response to any treatment [IV, B]. PFS (3.2 months; 95% CI: 2.7–3.7) and response rates (7.7%;
95% CI: 4.6–10.9); the mean OS was 7.2 months (95% CI: 6.2–
systemic chemotherapy 8.2) and no recommendation could be made about the most ap-
propriate second-line regimen [40]. Moreover, the magnitude of
Prospective randomised, controlled studies have shown that sys-
benefit to patients (if any) over best supportive care is not
temic chemotherapy extends survival in patients with advanced
known; results of study NCT01926236 (ABC-06) will inform
BTC compared with best supportive care [35, 36]. The phase III
this question.
UK ABC-02 study has established the cisplatin/gemcitabine
combination as a standard of care in this disease [I, A], achiev-
ing a median survival close to a year (11.7 months) for cisplatin/
gemcitabine, compared with 8.1 months for gemcitabine alone
personalised medicine
(95% CI: 0.53-0.79; P<0.001) [37] with a similar benefit in the The modest gains obtained with systemic chemotherapy high-
randomised phase II Japanese study [38]. A meta-analysis of light the need for improved therapies; the epithelial growth
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw324 | v

by guest
on 05 February 2018
Table 4. Levels of evidence and grades of recommendation clinical examination, laboratory investigation (including LFTs
(adapted from the Infectious Diseases Society of America–United and lactate dehydrogenase), tumour markers (carcinoembryonic
States Public Health Service Grading Systema) antigen, CA19-9) and CT scan of the thorax, abdomen and pelvis
[IV, A]. Regular visits can be extended to 6-monthly thereafter
Levels of evidence
and prolonged to yearly visits after 5 years of follow-up.
I Evidence from at least one large randomised, controlled trial of In patients receiving treatment of advanced, recurrent or
good methodological quality (low potential for bias) or meta- metastatic disease, best supportive care should include active
analyses of well-conducted randomised trials without identification and management of obstructive complications.
heterogeneity These may include biliary obstruction (requiring biliary drain-
II Small randomised trials or large randomised trials with a suspicion age and stents, as appropriate); gastric outlet obstruction (re-
of bias (lower methodological quality) or meta-analyses of such quiring duodenal stent or, occasionally, bypass surgery) and/or
trials or of trials with demonstrated heterogeneity
pancreatic duct obstruction (requiring pancreatic enzyme re-
placement therapy). Percutaneous transhepatic drainage is
recommended if endoscopic stenting is not possible; a metal
stent is preferred in patients with a life expectancy of >3
months. Some patients require repeat stenting on multiple occa-
A Strong evidence for efficacy with a substantial clinical benefit, sions; anticipation for such an event is required when planning
strongly recommended stent placement. In addition, patients should have full access to
B Strong or moderate evidence for efficacy but with a limited clinical palliative care and symptom management (including pain
benefit, generally recommended control) throughout their treatment.
risk or the disadvantages (adverse events, costs, ...), optional
D Moderate evidence against efficacy or for adverse outcome, methodology
These clinical practice guidelines were developed in accordance
E Strong evidence against efficacy or for adverse outcome, never
with the ESMO standard operating procedures for clinical
recommended
practice guidelines development, www.esmo.org/Guidelines/
a
By permission of the Infectious Diseases Society of America [45]. ESMO-Guidelines-Methodology. The relevant literature has
been selected by the expert authors. A summary of recommen-
dations is shown in Table 3. Levels of evidence and grades of
recommendation have been applied using the system shown in
factor receptor (EGFR) and vascular endothelial growth factor
(VEGF) axes have been studied the most.
standard clinical practice by the experts and the ESMO faculty.
Despite a promising response rate (63%) when adding the
This manuscript has been subjected to an anonymous peer-
EGFR-targeted monoclonal antibody, cetuximab, to gemcitabine
review process.
and oxaliplatin in a single-arm study (leading to salvage surgery
in n = 9 [30%] of patients) [41], there was no incremental benefit
observed from the addition of cetuximab to GemOx in the rando- conflict of interest
mised phase II BINGO study [42]; similar negative findings were
JWV has reported honoraria/consultancy for Lilly, AstraZeneca
observed with erlotinib or panitumumab with no clear correlation
and Sirtex; research support from Lilly, AstraZeneca and
with EGFR overexpression or k-ras mutation status.
NuCana. SAK has reported speakers’ fees for Bayer and BTG.
Prospective randomised phase II studies targeting VEGF have
FH has reported research support from Merck Serono and ad-
failed to show a benefit of adding sorafenib (an oral multi-tyro-
visory board participation for Merck Serono, MSD and Celgene.
sine kinase inhibitor) to single-agent gemcitabine [43] or cedira-
TG has reported speakers’ bureau member of and research
nib (an oral VEGFR-1, −2 and −3, PDGF and c-Kit tyrosine
support from Roche, Merck-Serono, Amgen, Sanofi-Aventis and
kinase inhibitor) to the cisplatin/gemcitabine combination [44].
Bayer. DA has reported honoraria/consultancy for Roche,
There is currently no evidence to support the use of targeted
Merck-Serono, Bayer, Lilly and Servier; research support from
therapies outside the context of a clinical trial.
Roche. IB has declared no potential conflicts of interest.
A better understanding of the molecular pathology of BTC
may help identify suitable targets for therapy. Molecular profiling
has identified clear differences between CCA and GBC; moreover, references
intra- and extrahepatic CCAs have different profiles (e.g. IDH-1 1. Shaib Y, El-Serag HB. The epidemiology of cholangiocarcinoma. Semin Liver Dis
and FGFR fusion rearrangements appear in iCCAs only). The sig- 2004; 24: 115–125.
nificance and clinical relevance (particularly with therapeutic 2. Nakeeb A, Pitt HA, Sohn TA et al. Cholangiocarcinoma. A spectrum of intrahepatic,
intent) of identifying these signatures are under evaluation. perihilar, and distal tumors. Ann Surg 1996; 224: 463–473; discussion 473–
475.
3. Bragazzi MC, Cardinale V, Carpino G et al. Cholangiocarcinoma: epidemiology and
follow-up and long-term implications risk factors. Transl Gastrointest Cancer 2012; 1: 21–32.
4. Shin HR, Oh JK, Masuyer E et al. Comparison of incidence of intrahepatic and
Follow-up after potentially curative treatment should consist of extrahepatic cholangiocarcinoma - focus on East and South-Eastern Asia. Asian
3-monthly visits during the first 2 years after therapy including Pac J Cancer Prev 2010; 11: 1159–1166.
v | Valle et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
5. Patel T. Worldwide trends in mortality from biliary tract malignancies. BMC Cancer 28. Ito H, Ito K, D’Angelica M et al. Accurate staging for gallbladder cancer:
2002; 2: 10. implications for surgical therapy and pathological assessment. Ann Surg 2011;
6. Khan SA, Emadossadaty S, Ladep NG et al. Rising trends in cholangiocarcinoma: 254: 320–325.
is the ICD classification system misleading us? J Hepatol 2012; 56: 848–854. 29. Horgan AM, Amir E, Walter T, Knox JJ. Adjuvant therapy in the treatment of biliary
7. McLean L, Patel T. Racial and ethnic variations in the epidemiology of intrahepatic tract cancer: a systematic review and meta-analysis. J Clin Oncol 2012; 30:
cholangiocarcinoma in the United States. Liver Int 2006; 26: 1047–1053. 1934–1940.
8. Palmer WC, Patel T. Are common factors involved in the pathogenesis of primary 30. Ben-Josef E, Guthrie KA, El-Khoueiry AB et al. SWOG S0809: a phase II
liver cancers? A meta-analysis of risk factors for intrahepatic cholangiocarcinoma. Intergroup trial of adjuvant capecitabine and gemcitabine followed by radiotherapy
J Hepatol 2012; 57: 69–76. and concurrent capecitabine in extrahepatic cholangiocarcinoma and gallbladder
9. Welzel TM, Graubard BI, El-Serag HB et al. Risk factors for intrahepatic and carcinoma. J Clin Oncol 2015; 33: 2617–2622.
extrahepatic cholangiocarcinoma in the United States: a population-based case- 31. Thet Cho M. Adjuvant gemcitabine plus docetaxel followed by 5FU chemoradiation
control study. Clin Gastroenterol Hepatol 2007; 5: 1221–1228. for patients with resected pancreaticobiliary cancers: a single-institution, phase II
10. Tyson GL, El-Serag HB. Risk factors for cholangiocarcinoma. Hepatology 2011; study. J Clin Oncol 2014; 32 (Suppl.): abstr e22243.
54: 173–184. 32. Phelip JM, Vendrely V, Rostain F et al. Gemcitabine plus cisplatin versus
11. Khan SA, Thomas HC, Davidson BR, Taylor-Robinson SD. Cholangiocarcinoma. chemoradiotherapy in locally advanced biliary tract cancer: Fédération
Lancet 2005; 366: 1303–1314. Francophone de Cancérologie Digestive 9902 phase II randomised study. Eur J
12. Kumagai S, Kurumatani N, Arimoto A, Ichihara G. Cholangiocarcinoma among Cancer 2014; 50: 2975–2982.
offset colour proof-printing workers exposed to 1,2-dichloropropane and/or 33. Fuller CD, Dang ND, Wang SJ et al. Image-guided intensity-modulated
dichloromethane. Occup Environ Med 2013; 70: 508–510. radiotherapy (IG-IMRT) for biliary adenocarcinomas: initial clinical results.
13. Nordenstedt H, Mattsson F, El-Serag H, Lagergren J. Gallstones and Radiother Oncol 2009; 92: 249–254.
cholecystectomy in relation to risk of intra- and extrahepatic cholangiocarcinoma. 34. Al-Adra DP, Gill RS, Axford SJ et al. Treatment of unresectable intrahepatic
Br J Cancer 2012; 106: 1011–1015. cholangiocarcinoma with yttrium-90 radioembolization: a systematic review and
14. Bertran E, Heise K, Andia ME, Ferreccio C. Gallbladder cancer: incidence and pooled analysis. Eur J Surg Oncol 2015; 41: 120–127.
survival in a high-risk area of Chile. Int J Cancer 2010; 127: 2446–2454. 35. Glimelius B, Hoffman K, Sjödén PO et al. Chemotherapy improves survival and
15. Park MJ, Kim YK, Lim S et al. Hilar cholangiocarcinoma: value of adding DW imaging quality of life in advanced pancreatic and biliary cancer. Ann Oncol 1996; 7:
to gadoxetic acid-enhanced MR imaging with MR cholangiopancreatography for 593–600.
preoperative evaluation. Radiology 2014; 270: 768–776. 36. Sharma A, Dwary AD, Mohanti BK et al. Best supportive care compared with
16. Pitman MB, Layfield LJ. Guidelines for pancreaticobiliary cytology from the chemotherapy for unresectable gall bladder cancer: a randomized controlled study.
Papanicolaou Society of Cytopathology: a review. Cancer Cytopathol 2014; 122: J Clin Oncol 2010; 28: 4581–4586.
399–411. 37. Valle J, Wasan H, Palmer DH et al. Cisplatin plus gemcitabine versus gemcitabine
17. Razumilava N, Gleeson FC, Gores GJ. Awareness of tract seeding with endoscopic for biliary tract cancer. N Engl J Med 2010; 362: 1273–1281.
ultrasound tissue acquisition in perihilar cholangiocarcinoma. Am J Gastroenterol 38. Okusaka T, Nakachi K, Fukutomi A et al. Gemcitabine alone or in combination with
2015; 110: 200. cisplatin in patients with biliary tract cancer: a comparative multicentre study in
18. Bosman FT; World Health Organization; International Agency for Research on Japan. Br J Cancer 2010; 103: 469–474.
Cancer. WHO Classification of Tumours of the Digestive System. Lyon, France: 39. Valle JW, Furuse J, Jitlal M et al. Cisplatin and gemcitabine for advanced biliary
IARC Press 2010. tract cancer: a meta-analysis of two randomised trials. Ann Oncol 2014; 25:
19. Levy C, Lymp J, Angulo P et al. The value of serum CA 19-9 in predicting 391–398.
cholangiocarcinomas in patients with primary sclerosing cholangitis. Dig Dis Sci 40. Lamarca A, Hubner RA, David Ryder W, Valle JW. Second-line chemotherapy in
2005; 50: 1734. advanced biliary cancer: a systematic review. Ann Oncol 2014; 25: 2328–2338.
20. Nakamura H, Arai Y, Totoki Y et al. Genomic spectra of biliary tract cancer. Nat 41. Gruenberger B, Schueller J, Heubrandtner U et al. Cetuximab, gemcitabine, and
Genet 2015; 47: 1003–1010. oxaliplatin in patients with unresectable advanced or metastatic biliary tract
21. In Edge SB, Byrd DR, Compton CC et al (eds). Cancer Staging Handbook from the cancer: a phase 2 study. Lancet Oncol 2010; 11: 1142–1148.
AJCC Cancer Staging Manual. New York, NY: Springer 2010; 247–276. 42. Malka D, Cervera P, Foulon S et al. Gemcitabine and oxaliplatin with or without
22. Bismuth H, Nakache R, Diamond T. Management strategies in resection for hilar cetuximab in advanced biliary-tract cancer (BINGO): a randomised, open-label,
cholangiocarcinoma. Ann Surg 1992; 215: 31–38. non-comparative phase 2 trial. Lancet Oncol 2014; 15: 819–828.
23. European Association for the Study of the Liver. EASL clinical practice guidelines: 43. Moehler M, Maderer A, Schimanski C et al. Gemcitabine plus sorafenib versus
management of cholestatic liver diseases. J Hepatol 2009; 51: 237–267. gemcitabine alone in advanced biliary tract cancer: a double-blind placebo-
24. Andrén-Sandberg A. Diagnosis and management of gallbladder polyps. N Am J controlled multicentre phase II AIO study with biomarker and serum programme.
Med Sci 2012; 4: 203–211. Eur J Cancer 2014; 50: 3125–3135.
25. de Jong MC, Nathan H, Sotiropoulos GC et al. Intrahepatic cholangiocarcinoma: an 44. Valle JW, Wasan H, Lopes A et al. Cediranib or placebo in combination
international multi-institutional analysis of prognostic factors and lymph node with cisplatin and gemcitabine chemotherapy for patients with advanced biliary
assessment. J Clin Oncol 2011; 29: 3140–3145. tract cancer (ABC-03): a randomised phase 2 trial.Lancet Oncol 2015; 16:
26. Matsumoto N, Ebata T, Yokoyama Y et al. Role of anatomical right hepatic 967–978.
trisectionectomy for perihilar cholangiocarcinoma. Br J Surg 2014; 101: 261–268. 45. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
27. Dickson PV, Behrman SW. Distal cholangiocarcinoma. Surg Clin North Am 2014; among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
94: 325–342. 139–144.
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw324 | v

by guest
on 05 February 2018
clinical practice guidelines Annals of Oncology 23 (Supplement 7): vii41–vii48, 2012
doi:10.1093/annonc/mds225
Hepatocellular carcinoma: ESMO–ESDO Clinical Practice

C. Verslype1,2, O. Rosmorduc3 & P. Rougier4, on behalf of the ESMO Guidelines Working Group*
ESDO, European Society of Digestive Oncology
Departments of 1Hepatology; 2Digestive Oncology, University Hospitals Leuven, Leuven, Belgium; 3Department of Gastroenterology and Hepatology, Saint-Antoine
Hospital, Paris, France; 4Department of Digestive Oncology, European Georges Pompidou Hospital, Paris, France
incidence and epidemiology diagnostic tests and the availability of effective treatment. Cost-
effectiveness studies suggest surveillance of HCC is warranted
The oncologic community is faced with a steady increase in in cirrhotic patients irrespective of its etiology [5]. Surveillance
the incidence of hepatocellular carcinoma (HCC) [1]. Liver of non-cirrhotic patients is also advocated, especially in HBV
cancer represents the sixth most common cancer in the world carriers with serum viral load >10 000 copies/ml [6] or HCV-
(749 000 new cases) and the third cause of cancer-related death infected patients with bridging fibrosis (F3) [III, A]. Patients
(692 000 cases). The incidence varies from 3 out of 100 000 in with HCV infection and advanced fibrosis remain at risk for
Western countries, to more than 15 out of 100 000 in certain HCC even after achieving sustained virological response
areas of the world, mapping the geographical distribution of following antiviral treatment [III, A].
viral hepatitis B (HBV) and hepatitis C (HCV), the most Japanese cohort studies have shown that surveillance by
clinical practice
important causes of chronic liver disease and HCC [2]. Most abdominal ultrasound resulted in an average size of the
guidelines
cases arise in those regions with limited resources. The detected tumors of 1.6 ± 0.6 cm, with <2% of the cases
incidence of HCC increases progressively with advancing age exceeding 3 cm [7]. In the Western world and in less
in all populations, with a strong male preponderance. experienced centers, sensitivity of finding early-stage HCC by
The association of chronic liver disease and HCC represents ultrasound is considerably less effective [8]. There are no data
the basis for preventive strategies, including universal to support the use of contrast-enhanced computed
vaccination at birth against hepatitis B , programs to stop tomography (CT) or magnetic resonance imaging (MRI) for
transmission and early antiviral eradication of viral hepatitis B surveillance. In many centers, ultrasound surveillance is
and C [III, A]. It is unclear whether HBV vaccination will complemented with the determination of serum alpha-
result in a decline in HCC as was seen in Taiwan, given the fetoprotein (AFP), which can lead to a 6%–8% gain in the
importance of other risk factors in Europe, such as alcoholic tumor detection rate but at the price of false-positive results.
and non-alcoholic fatty liver disease. The control of other risk A randomized, controlled trial (RCT) of Chinese patients
factors for chronic liver disease and cancer is more difficult to with chronic hepatitis B infection compared surveillance
implement, such as cutting down on the consumption of (ultrasound and serum AFP measurements every 6 months)
alcohol and programs aiming at a healthier lifestyle in the light versus no surveillance [9]. Despite low compliance with the
of the obesity pandemic [3, 4]. In Africa, reduction of exposure surveillance program (55%), HCC-related mortality was
to aflatoxin B1, especially in HBV-infected individuals, may reduced by 37% in the surveillance arm. Considering the most
lower the risk of HCC. appropriate surveillance interval, a randomized study
HCC may evolve from subclasses of adenomas, and in <10% comparing a 3- versus 6-month based schedule failed to detect
of cases HCC occurs in a normal liver. any differences [10]. Therefore, surveillance of patients at risk
Surveillance of HCC involves the repeated application of for HCC should be carried out by abdominal ultrasound every
screening tools in patients at risk for HCC and aims for the 6 months [I, A].
reduction in mortality of this patient population. The success
of surveillance is influenced by the incidence of HCC in the
target population, the availability and acceptance of efficient diagnosis and pathology
Although ultrasonography by an experienced physician is
widely accepted for tumor surveillance in patients at risk for
*Correspondence to: ESMO Guidelines Working Group, ESMO Head Office, Via
L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland; E-mail: clinicalguidelines@esmo. HCC, lesion characterization (the formal diagnosis of HCC) is
org based on either a tissue specimen or, in selected cases, on very
†
specific CT/MRI findings, often referred to as ‘non-invasive’
Approved by the ESMO Guidelines Working Group: November 2007, last update June
2012. This publication supersedes the previously published version—Ann Oncol 2010; criteria [III, A]. The diagnostic work-up of a patient suspected
21 (Suppl. 5): v59–v64. with HCC is given in Table 1.
Downloaded from https://academic.oup.com/annonc/article-abstract/23/suppl_7/vii41/144007

by guest
on 05 February 2018
Table 1. Diagnostic work-up in a patient with suspected hepatocellular scan or dynamic contrast-enhanced MRI). Diagnosis should be
carcinoma on ultrasound based on the identification of the typical vascular hallmark of
HCC (hypervascular in the arterial phase with washout in the
portal venous or delayed phases) [III, B]. A lesion that displays
arterial hypervascularization and venous washout on contrast-
• History and clinical examination enhanced ultrasound may also be a cholangiocarcinoma,
- Risk-factors for chronic liver disease: i.v. drug use, alcohol intake
making this technique less suitable for non-invasive diagnosis
- Symptoms and signs of chronic liver disease ( jaundice, ascites,
of HCC. Angiography and FDG-PET-scan are not
encephalopathy, splenomegaly)
recommended for early diagnosis.
- Performance status and nutritional state
In addition to the above-mentioned comments, every
• Laboratory analysis
decision on biopsy of a focal liver lesion should be discussed
- Etiology of liver disease: HBV, HCV, iron status, auto-immunity, etc.
- Liver function: prothrombin time, albumin, bilirubin
by the multidisciplinary team, including a hepatobiliary and
- Platelets transplant surgeon. A negative biopsy does not rule out
- Tumor marker: serum alpha-fetoprotein malignancy and there is a risk of tumor seeding that varies
• Imaging studies between 0% and 11%, with a median time interval between
- Dynamic (multiple phase) MRI or CT studies for diagnosis and biopsy and seeding of 17 months. There is no indication for
evaluation of tumor extent (number and size of nodules, vascular biopsy of a focal lesion in a cirrhotic liver: (i) when the patient
invasion, extra-hepatic spread) is not a candidate for any form of therapy because of serious
- Consider: chest CT and a bone scan in advanced disease comorbidity; (ii) in case of decompensated cirrhosis and the
• Assessment of portal hypertension patient is on the waiting list for liver transplantation and (iii)
- Upper endoscopy: varices and/or hypertensive gastropathy when the patient is a candidate for resection that can be
- Optional: transjugular measurement of hepatic-venous pressure carried out with an acceptable morbidity and mortality risk. In
gradient the future, it is anticipated that obtaining tissue for molecular
studies and targeted therapy will be increasingly important in
HCC. So far, different tumoral subclasses of HCC have already
CT, computed tomography; MRI, magnetic resonance imaging.
been characterized (Wnt, proliferation and inflammation
class), but non-responding to a specific targeted therapy [14].
Pathological diagnosis of HCC requires a biopsy of the In addition, the gene profile of adjacent non-tumoral tissue
tumor or a resection specimen, and preferably contains may also determine outcome.
surrounding non-tumoral parenchyma for comparison. HCC is
an adenocarcinoma and the composing tumor cells resemble
normal hepatocytes. An international expert panel of
staging
pathologists has set up diagnostic criteria for HCC and Staging of HCC is important to determine outcome, planning
especially for the challenging differential diagnosis with of optimal therapy and includes assessment of tumor extent,
premalignant dysplastic lesions [11]. Stromal invasion, or liver function, portal pressure and clinical performance status
tumor cell invasion into the portal tracts or fibrous septa, [III, A]. Relevant techniques to evaluate tumor extent (number
defines HCC and is not present in dysplastic lesions [III, A]. and size of nodules, vascular invasion, extrahepatic spread)
Other histological features of HCC, however, may also be seen include contrast-enhanced MRI or helical CT; chest CT and a
in dysplastic lesions: (i) increased cell density more than two bone scan should be considered in advanced disease. Liver
times that of the surrounding tissue, with an increased nuclear/ function is assessed by the Child–Pugh scoring system
cytoplasm ratio and irregular thin-trabecular pattern; (ii) (bilirubin, albumin, ascites, prothrombin time and hepatic
intratumoral portal tracts); (iii) pseudoglandular pattern; (iv) encephalopathy). The finding of esophageal varices and/or
diffuse fatty change (up to 40% in early well-differentiated splenomegaly with blood platelet counts of <100 000/µl suggest
tumors, uncommon in tumors >3 cm; and (v) varying clinically important portal hypertension, which can also be
numbers of unpaired arteries. Additional measured by the transjugular route (hepatic-venous pressure
immunohistochemical staining may be helpful: glypican-3 to gradient >10 mmHg). Several staging systems—incorporating
differentiate high-grade dysplastic nodules from early HC and some or all of the above-mentioned items—have been
cytokeratin 19 may point to progenitor cell features or biliary developed. Every system has advantages and drawbacks. The
features in mixed forms of HCC/cholangiocarcinoma, which pTNM system is based on the pathology report and may be
are not always detected on hematoxylin–eosin stain [IV, B]. relevant to stratify patients for studies on adjuvant treatments.
Through the progression from dysplastic nodules, early and The Barcelona Clinic Liver Cancer (BCLC) staging system links
more advanced HCC, there is a development of arteries which staging of HCC in cirrhosis with treatment modalities [15, 16].
become the dominant blood supply. This neovascularization The BCLC system is widely used and encompasses all HCC
can be assessed by a CD34 stain. patients. The system identifies those patients with early HCC
In some cases, a formal pathological proof is not necessary who may benefit from curative therapies (stage 0 and A), those
for diagnosis and the clinician can rely on non-invasive at intermediate (stage B) or advanced stage (stage C) who may
imaging criteria for lesion characterization [12, 13]. These benefit from palliative treatments and those with a very poor
criteria only apply to cirrhotic patients and require state-of- life expectancy (stage D). Median survival without therapy is
the-art imaging techniques (multiple-phase multidetector CT >36 months for stage 0 and A, 16 months for stage B, 4–8
vii | Verslype et al. Volume 23 | Supplement 7 | October 2012

by guest
on 05 February 2018
Table 2. Impact of therapy on prognosis, according to BCLC stages
Natural history (median survival) Prognosis with therapy

Very early stage (stage 0) >36 months 70%–90% 5-year survival (OLT, ablation, resection)
Early stage (stage A) 36 months 50%–70% 5-year survival (OLT, ablation, resection)
Intermediate stage (stage B) 16 months 20-month median survival (transarterial chemoembolization)
Advanced stage (stage C) 4–8 months 6–11-month median survival (sorafenib)
Terminal stage D (stage D) <4 months
BCLC, The Barcelona Clinic Liver Cancer staging system; OLT, orthotopic liver transplantation.
Figure 1 Strategy for staging and treatment assignment in patients diagnosed with HCC (adapted from Bruix et al. [16]). PS, performance status. *Poor
liver synthetic function due to tumor involvement of the liver. °Only Child–Pugh A.
months for stage C and <4 months for stage D. The expected without advanced fibrosis, as long as an R0 resection can be
survival improvement with specific therapies is given in carried out without causing postoperative liver failure due to a
Table 2. A staging algorithm, largely based upon the BCLC too small liver remnant [17] [III, B].
system, is given in Figure 1, with two modifications. Portal In the case of cirrhosis, resection is effective and safe
hypertension is taken out of the algorithm and gives more ( postoperative mortality <5%) in early BCLC stages (0 and A)
freedom regarding the clinical decision concerning resection, provided that one is dealing with a single lesion, a good
which still represents the main approach in Japan. In addition, performance status and no clinically important portal
patients with poor liver synthetic function (Child–Pugh C) and hypertension [18, 19] [III, B]. In individual cases and based
tumor extent within the Milan criteria (one nodule <5 cm or upon Japanese experience, more lesions can be safely resected,
three nodules <3 cm) should, in our opinion, not be denied the but at a higher risk of postoperative morbidity and mortality.
possibility of liver transplantation and are therefore not Tumor recurrence is between 50% and 70% at 5 years
classified as terminal stage. following surgery, which constitutes either intrahepatic
The etiology of co-existent liver disease has not been metastases (often within 2 years following surgery) or a new
identified as an independent prognostic factor. Nevertheless, HCC in the remaining cirrhotic liver (occurring beyond 2
finding a treatable underlying co-existent liver disease may be years). Microvascular invasion is a known predictor of
very relevant: e.g. antiviral treatment in case of hepatitis B or recurrence and survival, directly associated with histological
stopping alcohol may result in a marked improvement in liver differentiation, number and size of the nodule(s).
function and improving prognosis. Local ablation techniques (RFA and PEI) have been put
forward as alternatives for surgery. For small nodules <2 cm,
BCLC stage 0, both techniques achieve complete responses in
>90% of cases with good long-term outcome and can be
management of local disease: radical
considered alternatives to resection [III, B] [20]. RFA provides
therapies better local control than PEI, especially in HCCs >2 cm [II, A]
Radical treatments include surgical resection, liver [21]. The number and diameter of lesions treated by RFA
transplantation and local destruction methods [radiofrequency should not exceed five and 5 cm, respectively [III, B]. Bile
ablation (RFA) or percutaneous ethanol injection (PEI)]. There ducts may be damaged and tumor seeding is possible; the latter
are no randomized trials comparing the efficacy of these three has been solely reported for percutaneous procedures (not
approaches and all evidence is based on cure rates in patient surgical procedures) in 0–12.5% of cases (median 0.9%) [22].
series. Resection is the recommended treatment in patients The results of these local ablative techniques are also hampered
Volume 23 | Supplement 7 | October 2012 doi:10.1093/annonc/mds225 | vii

by guest
on 05 February 2018
by disease recurrence, varying between 4% and 60% depending transcatheter devices

on the size of the tumor and the approach used (with higher The preferential arterial vascularization of HCC resulted in the
risk in percutaneous than in laparoscopic series). application of intra-arterial administration of chemotherapy (e.
Neo-adjuvant or adjuvant therapies are not recommended to g. doxorubicin, cisplatin), embolizing material (e.g. coils,
improve outcome of patients treated with resection or local gelatin sponge particles) or radioactive particles.
ablation [II, B]. This recommendation is supported by a study Transarterial chemoembolization (TACE) involves the
of the Cochrane group that recently identified 12 RCTs with combination of selective injection through the hepatic artery of
843 patients, but concluded that there was no clear evidence antineoplastic agents and selective obstruction of tumoral
for the efficacy of any of the adjuvant and neo-adjuvant feeding vessels. TACE may induce partial responses in 15%–
protocols reviewed (including immunotherapy, retinoids, 55% of patients, which are associated with a benefit in overall
chemoembolization) [23]. The results of the STORM and progression-free survival [25]. The most common side
(Sorafenib as Adjuvant Treatment in the Prevention of effect is a post-embolization syndrome, characterized by fever,
Recurrence of Hepatocellular Carcinoma) international trial abdominal pain and risk of liver failure. Absolute
are pending. This international multicenter study randomly contraindications for TACE are decompensated cirrhosis
assigned 1100 patients to receive sorafenib 400 mg b.i.d. for up (Child–Pugh B ≥8, including jaundice, clinical encephalopathy,
to 4 years and placebo and included patients who have received refractory ascites), extensive tumor with massive replacement
surgical resection or local ablation. The primary end point of of both entire lobes, severely reduced portal vein flow ( portal
the study is recurrence-free survival. vein occlusion or hepatofugal blood flow), untreatable
Liver transplantation offers the possibility to cure both the arteriovenous fistula, bilio-enteric anastomosis or biliary stents
tumor and the underlying liver disease. Liver transplantation and a creatinine clearance <30 ml/min [26].
should be considered in patients with a solitary lesion of <5 cm TACE is recommended for patients with BCLC stage B, or
or three nodules <3 cm that are not suitable for resection [III, those with an excellent liver function and multinodular
A]. These Milan criteria guarantee a 5-year disease-free and asymptomatic tumors without macroscopic vascular invasion
overall survival of >65%. A recent systematic review including or extra hepatic spread [I, A]. However, patients with
90 studies comprising a total of 17 780 patients over 15 years intermediate stage cannot be cured by TACE as a single-
identified the Milan criteria as an independent prognostic modality treatment.
factor for outcome after liver transplantation [24]. Overall 5- The magnitude of the benefit is determined by the technique
year survival of patients within the Milan criteria (65%–78%) used and patient characteristics. Generally speaking, with
was similar compared with non-HCC indications according to increasing size and number of the lesions, results of TACE are
liver transplant registries European Liver Transplant Registry less favorable. The recently published and heavily contested
(ELTR) in Europe and Organ Procurement and Cochrane meta-analysis, which questioned TACE as a standard
Transplantation Network (OPTN) in the United States (65%– of care in patients with intermediate HCC, is an illustration of
87%). Criteria are subject to local (multidisciplinary) decisions the difficulty to extract the evidence from the literature [27].
within a (supra)national framework that defines priority rules Techniques have evolved during the last years. Studies with
(e.g. Eurotransplant) and are in evolution. Owing to organ doxorubicin-eluting beads (DEBDOX) have demonstrated less
shortage, liver transplant candidates are confronted with long systemic leakage of chemotherapy in the systemic circulation,
waiting times, which may be associated with tumor progression resulting in less side effects, with at least the same activity in
beyond the Milan criteria. In the case of a long anticipated randomized phase II trials with conventional TACE (gelfoam-
waiting time (>6 months), patients may be offered resection, lipiodol particles) as comparator [28, 29]. TACE with selective
local ablation or trans-arterial chemoembolization in order to administration with doxorubicin-eluting beads is an option to
minimize the risk of tumor progression and to offer a ‘bridge’ minimize systemic side effects of chemotherapy [II, A].
to transplant [III, B]. Unfortunately, no RCTs are available that The combination of TACE with systemic agents such as
support one of these options. There is no evidence in support sorafenib—either sequential or concomitant—cannot be
of sorafenib for patients with HCC on the waiting list for liver recommended today in clinical practice. An Asian phase III
transplantation. study evaluated the role of sorafenib versus placebo in patients
who achieved a response after TACE, but failed to show a gain
in time to progression (TTP). A subanalysis suggested that a
shorter treatment lag between TACE and sorafenib, a longer
management of locally advanced/ treatment duration and greater total daily dose may be
associated with clinical improvements [30]. The results of the
metastatic disease: palliative treatments ‘Sorafenib or Placebo in Combination with Transarterial
Palliative treatments include transcatheter devices, systemic Chemoembolization (DEBDOX) for Intermediate-Stage
therapy and external beam radiotherapy. These therapies are Hepatocellular Carcinoma’ (SPACE) study were recently
offered with the intention to improve survival or to maintain presented [31]. Although this global randomized phase II study
quality of life without the prospect of cure. Although primarily met its statistical end point of an improvement in TTP (HR of
intended for patients with intermediate or advanced-stage HCC, 0.797, P = 0.072), the clinical significance remains unclear.
these techniques may be used with success in patients with early- More patients in the non-Asian region, when compared with
stage HCC who have contraindications for radical therapies. the Asian countries, had a high number of early TACE
vii | Verslype et al. Volume 23 | Supplement 7 | October 2012

by guest
on 05 February 2018
discontinuations, together with a shorter treatment duration Table 3. Response assessment by modified RECISTa
than in Asian countries. The results of ongoing phase III trials
are awaited. Target lesions
Internal radiation with Iodine 131 or Yttrium 90 glass or Complete response (CR) Disappearance of any intratumoral arterial
resin particles has shown antitumoral effects with a safe profile enhancement in all target lesions
in phase II studies and registries [32]. Radioembolization may Partial response (PR) At least a 30% decrease in the sum of the
be competitive with sorafenib or TACE in subsets of patients, diameters of viable (enhancement in the
such as those with prior TACE failure, excellent liver function, arterial phase) target lesions, taking as
macrovascular invasion and the absence of extra-hepatic disease reference the baseline sum of the diameters
[III, C]. Current phase III studies are evaluating the place of of target lesions
radioembolization versus TACE in patients with intermediate Stable disease (SD) Any cases that do not qualify for either PR or
stage HCC, and as single modality or combined with sorafenib PD
in patients with advanced HCC compared with sorafenib. Progressive disease (PD) An increase of at least 20% in the sum of the
diameters of viable (enhancement in the
arterial phase) target lesions recorded since
systemic therapy treatment started
Systemic therapy with cytotoxic drugs (doxorubicin or Non-target lesions
cisplatin) yields low objective response rates (<10%). Although Complete response (CR) Disappearance of any intratumoral arterial
combinations like XELOX or GEMOX may offer a more enhancement in all non-target lesions
interesting disease control rate [33, 34], they are without a Stable disease (SD) or Persistence of intratumoral arterial
proven survival benefit. In addition, chemotherapy is poorly incomplete response (IR) enhancement in one or more non-target
tolerated, due to underlying cirrhosis, coexisting cytopenias lesions
and unpredictable pharmacokinetics (altered activity of drug Progressive disease (PD) Appearance of one or more new lesions and/
metabolizing enzymes, fluid retention). or unequivocal progression of existing
The results of a randomized, placebo-controlled, double- non-target lesions
blind phase III study with the multikinase inhibitor sorafenib Additional recommendations
New lesion A new lesion can be classified as HCC if its
represented a breakthrough in the field [35]. Sorafenib is an
longest diameter is at least 1 cm and the
oral drug which blocks PDGF, VEGF, c-Kit and raf signaling,
enhancement pattern is typical for HCC.
both on the tumor cell and on the surrounding endothelial
A lesion with atypical radiological pattern
cells. Six hundred and two patients with advanced HCC, no
can be diagnosed as HCC by evidence of at
prior systemic treatment and good liver synthetic capacity least 1-cm interval growth.
(Child–Pugh A) were randomized between sorafenib 400 mg b. Pleural effusion or ascites Cytopathological confirmation of the
i.d. or placebo. Sorafenib was well tolerated and yielded a neoplastic nature of any effusion that
substantially relative improvement (44%) in overall survival. appears or worsens during treatment is
Median survival increased from 7.9 to 10.7 months [hazard required to declare PD
ratio (HR) 0.69, 95% CI 0.55–0.87]. Side effects include hand– Lymph nodes in the porta Lymph nodes detected at the porta hepatis
foot skin reaction, diarrhea and fatigue, but sorafenib was not hepatis can be considered malignant if the lymph-
found hepatotoxic. A similar benefit of sorafenib was node short axis is at least 2 cm
demonstrated in a subsequent Asian-Pacific RCT [36]. Portal vein thrombosis Malignant portal vein thrombosis should be
Sorafenib is the standard systemic therapy for patients with considered as a non-measurable lesion and
advanced HCC and well-preserved liver function (BCLC stage thus included in the non-target lesion
C) and those with intermediate-stage HCC who progress group
following TACE [I, A]. There are no clinical or molecular
HCC, hepatocellular carcinoma.
biomarkers available to identify the best responders to a
mRECIST, modified Response Evaluation Criteria in Sold Tumors
sorafenib.
(adapted from Lencioni et al. [37]).
In case of progression or intolerance to sorafenib, best
supportive care is preferred or patients should be included in
clinical trials. Systemic chemotherapy, tamoxifen, surrounding non-tumoral liver parenchyma and represents a
immunotherapy, anti-androgen or somatostatin analogues are promising powerful technique which needs further validation
not recommended for the clinical management of HCC [III, C]. External beam radiotherapy can be used to control
patients [I–II, A–B]. pain in patients with bone metastases [II, B]. The intrahepatic
For patients with end-stage disease with heavily impaired application of radioactive (e.g. Yttrium 90) microspheres via
liver function or a poor performance status (both due to the the hepatic artery was discussed above.
tumor involvement of the liver), only symptomatic treatment is
advocated, as they will die within 6 months [III, B].
response evaluation and follow-up
external beam radiotherapy Many HCC treatments act by induction of tumor necrosis or
Three-dimensional conformal radiotherapy makes it possible reduction in vascularity, which is not necessarily accompanied
to direct high-dose radiation to HCC with sparing of the by tumor shrinkage. Viable tumor needs to be assessed using
Volume 23 | Supplement 7 | October 2012 doi:10.1093/annonc/mds225 | vii

by guest
on 05 February 2018
1. Incidence and epidemiology

• Preventive strategies for HCC should include universal vaccination at birth against hepatitis B, programs to stop transmission and early antiviral eradication of viral
hepatitis B and C [III, A].
• All patients at risk for HCC should enter surveillance programs: cirrhotic patients (irrespective of etiology), non-cirrhotic HBV carriers with high viral load or non-
cirrhotic patients with chronic hepatitis C and advanced fibrosis (at least Metavir F3)
• Patients with HCV infection and advanced fibrosis remain at risk for HCC even after achieving sustained virological response following antiviral treatment [III, A].
• Surveillance of patients at risk for HCC should be carried out by abdominal ultrasound every 6 months [I, A].
2. Diagnosis
• The diagnosis of HCC is based on either a tissue specimen or in selected cases on very specific CT/MRI findings, often referred to as ‘non-invasive’ criteria [III, A].
• Pathological diagnosis of HCC requires a biopsy of the tumor or a resection specimen, and preferably contains surrounding non-tumoral parenchyma for comparison.
Stromal invasion or tumor cell invasion into the portal tracts or fibrous septa, defines HCC and is not present in dysplastic lesions [III, A].
• Non-invasive diagnosis of HCC is only possible in cirrhotic patients and requires state-of-the-art imaging techniques (multiple-phase multidetector CT scan and/or
dynamic contrast-enhanced MRI), with identification of the typical vascular hallmark of HCC (hypervascular in the arterial phase with washout in the portal venous
or delayed phases) [III, A].
• There is no indication for biopsy of a focal lesion in a cirrhotic liver when the patient is: (i) not a candidate for any form of therapy because of serious co-morbidity;
(ii) in case of decompensated cirrhosis and the patient is on the waiting list for liver transplantation and (iii) when the patient is a candidate for resection that can be
carried out with an acceptable morbidity and mortality risk.
3. Staging
• Staging of HCC is important to determine outcome, planning of optimal therapy and includes assessment of tumor extent, liver function, portal pressure and clinical
performance status [III, A].
• The Barcelona Clinic Liver Cancer (BCLC) staging system should be used in patients with HCC and underlying cirrhosis [III, A]. The system identifies those patients
with early HCC who may benefit from curative therapies (stage 0 and A), those at intermediate (stage B) or advanced stage (stage C) who may benefit from palliative
treatments and those with a very poor life expectancy (stage D).
4. Management of local disease: radical therapies
• Resection is the recommended treatment in patients without advanced fibrosis, as long as an R0-resection can be carried out without causing postoperative liver
failure due to a too small liver remnant [III, B].
• In the case of cirrhosis, resection is effective and safe (postoperative mortality <5%) in early BCLC stages (0 and A) provided that one is dealing with a single lesion, a
good performance status and no clinical significant portal hypertension [III, B].
• Local ablation with radiofrequency ablation (RFA) or percutaneous ethanol injection (PEI) is an alternative for resection for a single nodule <2 cm, BCLC stage 0 or
those early stages that are not candidates for resection [III, B].
• RFA provides better local control than PEI, especially in HCCs >2 cm [II, A].
• The number and diameter of lesions treated by RFA should not exceed five and 5 cm, respectively [III, B].
• Neo-adjuvant or adjuvant therapies are not recommended to improve outcome of patients treated with resection or local ablation [II, B].
• Liver transplantation should be considered in patients with a solitary lesion of <5 cm or three nodules <3 cm that are not suitable for resection [III, A].
• In the case of a long anticipated waiting time (>6 months) for liver transplantation, patients may be offered resection, local ablation or TACE in order to minimize the
risk of tumor progression [III, B].
5. Management of locally advanced/metastatic disease: palliative treatments
• TACE is recommended for patients with HCC BCLC stage B, or those with an excellent liver function and multinodular asymptomatic tumors without macroscopic
vascular invasion or extra-hepatic spread [I, A].
• TACE with selective administration with doxorubicin-eluting beads is recommended to minimize systemic side effects of chemotherapy [II, A].
• The combination of TACE with sorafenib—either sequential or concomitant—cannot be recommended outside clinical trials.
• Sorafenib is the standard systemic therapy for patients with advanced HCC and well-preserved liver function (BCLC stage C) and those with intermediate stage HCC
who progress following TACE [I, A].
• In case of progression or intolerance to sorafenib, best supportive care is preferred or patients should be included in clinical trials.
• Systemic chemotherapy, tamoxifen, immunotherapy, anti-androgen or somatostatin analogues are not recommended for the clinical management of HCC patients [I–
II, A–B].
• The role of radioembolization with glass or resin Y-90 spheres may be competitive with sorafenib or TACE in subsets of patients, such as those with prior TACE
failure, excellent liver function, macrovascular invasion and the absence of extra-hepatic disease [III, C].
• External beam radiotherapy can be used to control pain in patients with bone metastases [II, B].
• For patients with end-stage disease with heavily impaired liver function or a poor performance status (both due to the tumor involvement of the liver) only
symptomatic treatment is advocated [III, B].
6. Response evaluation and follow-up
• Response assessment should be based on dynamic CT or MRI studies [III, A] and the modified RECIST criteria [III, B].
• Serum tumor markers (such as alpha-fetoprotein levels) may be helpful, particularly in the case of not easily measurable disease, but should not be used as the only
determinant for treatment decision [IV, B].
• Follow-up of patients who underwent radical treatments should consist of the clinical evaluation of liver decompensation and the early detection of recurrence by
dynamic CT or MRI studies every 3 months the first 2 years and surveillance every 6 months later on [III, A].
• Patients with more advanced stages of HCC who are treated with TACE or systemic agents (e.g. sorafenib) are evaluated clinically for signs of liver decompensation
and by dynamic CT or MRI for tumor progression every 2 months to guide therapy decisions [III, A].
HCC, hepatocellular carcinoma; CT/MRI, computed tomography/magnetic resonance imaging; TACE, transarterial chemoembolization.
vii | Verslype et al. Volume 23 | Supplement 7 | October 2012

by guest
on 05 February 2018
dynamic CT or MRI studies and viable tumor should be 4. Marrero J, Fontana R, Fu S et al. Alcohol, tobacco and obesity are synergistic
defined as uptake of contrast agent in the arterial phase [III, risk factors for hepatocellular carcinoma. J Hepatol 2005; 42: 218–224.
A]. The current RECIST criteria were merely designed for the 5. Sarasin FP, Giostra E, Hadengue A. Cost-effectiveness of screening for detection
of small hepatocellular carcinoma in Western patients with Child-Pugh class A
evaluation of cytotxic agents. A modification of the RECIST
cirrhosis. Am J Med 1996; 101: 422–434.
(mRECIST) criteria is available and is based on the
6. Chen CJ, Yang HI, Su J et al.; REVEAL-HBV Study Group. Risk of hepatocellular
measurement of the diameter of the viable tumor component carcinoma across a biological gradient of serum hepatitis B virus DNA level.
of target lesions [37] (Table 3). The mRECIST criteria also JAMA 2006; 295: 65–73.
include guidelines regarding evaluation of vascular invasion, 7. Sato T, Tateishi R, Yoshida H et al. Ultrasound surveillance for early detection of
lymph nodes, effusions and new lesions. These mRECIST hepatocellular carcinoma among patients with chronic hepatitis C. Hepatol Int
criteria have been partially validated in retrospective cohort 2009; 3: 544–550.
studies and are now used in ongoing prospective phase II and 8. Singal A, Volk ML, Waljee A et al. Meta-analysis: surveillance with ultrasound for
III studies with new drugs or locoregional treatments [38]. early-stage hepatocellular carcinoma in patients with cirrhosis. Ailment
Pharmacol Ther 2009; 30: 37–47.
Although these mRECIST criteria need further prospective
9. Chen JG, Parkin DM, Chen QG et al. Screening for liver cancer: results of a
validation, it is recommended in daily clinical practice to randomised controlled trial in Qidong, China. J Med Screen 2003; 10:
consider not only tumor diameters but also lesion viability in 204–209.
therapy decision making [III, B]. Serum tumor markers (such 10. Trinchet J, Beaugrand M for GRETH. A Randomized Trial Comparing 3-Month vs
as AFP levels) may be helpful particularly in the case of not 6-Month Screening for HCC by Ultrasonography in Cirrhosis [Abstract]. Book of
easily measurable disease, but should not be used as the only Abstracts. International Liver Cancer Association (ILCA) 2007; Abstract 023.
determinant for treatment decision [IV, B]. www.ilca-online.org.
Follow-up of patients who underwent radical treatments 11. International Consensus Group for Hepatocellular Neoplasia. The International
Consensus Group for Hepatocellular Neoplasia. Pathologic diagnosis of early
(resection or RFA) should consist of the clinical evaluation of
hepatocellular carcinoma: a report of the international consensus group for
liver decompensation and the early detection of recurrence by hepatocellular neoplasia. Hepatology 2009; 49: 658–664.
dynamic CT or MRI studies every 3 months the first 2 years 12. Lencioni R, Cioni D, Della Pina C et al. Imaging diagnosis. Semin Liver Dis 2005;
and surveillance every 6 months later on [III, A]. Patients with 25: 162–170.
recurrence following radical therapies may still be candidates 13. Forner A, Vilana R, Ayuso C et al. Diagnosis of hepatic nodules 20 mm or
for curative therapies. smaller in cirrhosis: prospective validation of the noninvasive diagnostic criteria
Patients with more advanced stages of HCC who are treated for hepatocellular carcinoma. Hepatology 2008; 47: 97–104.
with TACE or systemic agents (e.g. sorafenib) are evaluated 14. Villanueva A, Newell P, Chiang DY et al. Genomics and signaling pathways in
clinically for signs of liver decompensation and for tumor hepatocellular carcinoma. Semin Liver Dis 2007; 27: 55–76.
progression by dynamic CT or MRI every 2 months to guide 15. Llovet JM, Brú C, Bruix J. Prognosis of hepatocellular carcinoma: the BCLC
staging classification. Semin Liver Dis 1999; 19: 329–338.
therapy decisions [III, A].
16. Bruix J, Sherman M; Practice Guidelines Committee, American Association for
the Study of Liver Diseases. Management of hepatocellular carcinoma.
note Hepatology 2005; 42: 1208–1236.
17. Lang H, Sotiropoulos GC, Dömland M et al. Liver resection for hepatocellular
Summary of recommendations is provided in Table 4. Levels of carcinoma in non-cirrhotic liver without underlying viral hepatitis. Br J Surg
evidence [I–V] and grades of recommendation [A–D] are 2005; 92: 198–202.
given in square brackets, according to the adapted Infectious 18. Belghiti J, Hiramatsu K, Benoist S et al. Seven hundred forty-seven
Diseases Society of American-United States Public Health hepatectomies in the 1990s: an update to evaluate the actual risk of liver
resection. J Am Coll Surg 2000; 191: 38–46.
Service Grading System.
19. Llovet JM, Schwartz M, Mazzaferro V. Resection and liver transplantation for
Statements without grading were considered justified
hepatocellular carcinoma. Semin Liver Dis 2005; 25: 181–200.
20. Livraghi T, Meloni F, Di Stasi M et al. Sustained complete response and
complications rates after radiofrequency ablation of very early hepatocellular
carcinoma in cirrhosis: is resection still the treatment of choice? Hepatology
conflict of interest 2008; 47: 82–89.
Prof. Verslype has reported research funding from Bayer, Sirtex 21. Lencioni R. Loco-regional treatment of hepatocellular carcinoma. Hepatology
and Novartis. Prof. Rougier has reported: honoraria from 2010; 52: 762–773.
Sanofi Aventis, Amgen, Keocyte, Merck Serono, Pfizer, Roche 22. Stigliano R, Marelli L, Yu D et al. Seeding following percutaneous diagnostic and
and Lilly; advisory board for Sanofi Aventis and Keocyte. therapeutic approaches for hepatocellular carcinoma. What is the risk and the
outcome? Seeding risk for percutaneous approach of HCC. Cancer Treat Rev
Dr. Rosmorduc has not reported any potential conflicts of 2007; 33: 437–447.
interest. 23. Samuel M, Chow PK, Chan Shih-Yen E et al. Neoadjuvant and adjuvant therapy
for surgical resection of hepatocellular carcinoma. Cochrane Database Syst Rev
2009; 21: CD001199.
references 24. Mazzaferro V, Bhoori S, Sposito C et al. Milan Criteria in Liver Transplantation for
1. El-Serag HB, Mason AC. Rising incidence of hepatocellular carcinoma in the HCC: an evidence-based analysis on 15 years of experience. Liver Transplant
United States. N Engl J Med 1999; 340: 745–750. 2011; 17(Suppl. 2): S44–S57.
2. Parkin DM, Bray F, Ferlay J et al. Global cancer statistics, 2002. CA Cancer J 25. Llovet JM, Bruix J. Systematic review of randomized trials for unresectable
Clin 2005; 55: 74–108. hepatocellular carcinoma: chemoembolization improves survival. Hepatology
3. El-Serag HB, Richardson PA, Everhart JE. The role of diabetes in hepatocellular 2003; 37: 429–442.
carcinoma: a case-control study among United States Veterans. Am J 26. Raoul JL, Sangro B, Forner A et al. Evolving strategies for the management of
Gastroenterol 2001; 96: 2462–2467. intermediate-stage hepatocellular carcinoma: available evidence and expert
Volume 23 | Supplement 7 | October 2012 doi:10.1093/annonc/mds225 | vii

by guest
on 05 February 2018
opinion on the use of transarterial chemoembolization. Cancer Treat Rev 2011; clinic liver cancer stages: a European evaluation. Hepatology 2011; 54:
37: 212–220. 868–878.
27. Oliveri RS, Wetterslev J, Gluud C. Transarterial (chemo)embolisation for unresectable 33. Boige V, Raoul JL, Pignon JP et al; Fédération Francophone de Cancérologie
hepatocellular carcinoma. Cochrane Database Syst Rev 2011; 3: CD004787. Digestive. Multicentre phase II trial of capecitabine plus oxaliplatin (XELOX) in
28. Lammer J, Malagari K, Vogl T et al. Prospective randomised study of patients with advanced hepatocellular carcinoma: FFCD 03-03 trial. Br J Cancer
doxorubicin-eluting-bead embolization in the treatment of hepatocellular 2007; 97: 862–867.
carcinoma: results of the PRECISION V study. Cardiovasc Intervent Radiol 2010; 34. Louafi S, Boige V, Ducreux M et al. Gemcitabine plus oxaliplatin (GEMOX) in
33: 41–52. patients with advanced hepatocellular carcinoma (HCC): results of a phase II
29. Van Malenstein H, Maleux G, Vandecaveye V et al. A randomized phase II study study. Cancer 2007; 109: 1384–1390.
of drug-eluting beads versus transarterial chemoembolization for unresectable 35. Llovet JM, Ricci S, Mazzaferro V et al.; SHARP Investigators Study Group.
hepatocellular carcinoma. Onkologie 2011; 34: 368–376. Sorafenib in advanced hepatocellular carcinoma. N Engl J Med 2008; 359:
30. Kudo M, Imanaka K, Chida N et al. Phase III study of sorafenib after transarterial 378–390.
chemoembolisation in Japanese and Korean patients with unresectable 36. Cheng AL, Kang YK, Chen Z et al. Efficacy and safety of sorafenib in patients in
hepatocellular carcinoma. Eur J Cancer 2011; 47: 2117–2127. the Asia-Pacific region with advanced hepatocellular carcinoma: a phase III
31. Lencioni R, Llovet JM, Han G et al. Sorafenib or placebo in combination with randomised, double-blind, placebo-controlled trial. Lancet Oncol 2009; 10:
transarterial chemoembolization (TACE) with doxorubicin-eluting beads 25–34.
(DEBDOX) for intermediate-stage hepatocellular carcinoma (HCC): phase II, 37. Lencioni R, Llovet JM. Modified RECIST (mRECIST) assessment for hepatocellular
randomized, double-blind SPACE trial. J Clin Oncol 2012; 30(Suppl. 4): Abstract carcinoma. Semin Liver Dis 2010; 30: 52–60.
LBA154. 38. Edeline J, Boucher E, Rolland Y et al. Comparison of tumor response by
32. Sangro B, Carpanese L, Cianni R et al. European Network on Radioembolization Response Evaluation Criteria in Solid Tumors (RECIST) and modified RECIST in
with Yttrium-90 Resin Microspheres (ENRY). Survival after yttrium-90 resin patients treated with sorafenib for hepatocellular carcinoma. Cancer 2012; 118:
microsphere radioembolization of hepatocellular carcinoma across Barcelona 147–156.
vii | Verslype et al. Volume 23 | Supplement 7 | October 2012

by guest
on 05 February 2018
clinical practice guidelines Annals of Oncology 24 (Supplement 6): vi64–vi72, 2013
doi:10.1093/annonc/mdt354
Early colon cancer: ESMO Clinical Practice Guidelines

R. Labianca1, B. Nordlinger2, G. D. Beretta3, S. Mosconi1, M. Mandalà1, A. Cervantes4 & D. Arnold5
on behalf of the ESMO Guidelines Working Group*
1
Ospedale Papa Giovanni XXIII, Bergamo, Italy; 2Hospital Ambroise Parè, Paris, France; 3Humanitas Gavazzeni Clinic, Bergamo, Italy; 4Department of Hematology
and Medical Oncology, INCLIVA, University of Valencia, Valencia, Spain; 5Department of Medical Oncology, Tumor Biology Center, Freiburg, Germany
These Clinical Practice Guidelines are endorsed by the Japanese Society of Medical Oncology (JSMO)
general information declined by 6% per quinquennium in men and 8% per

quinquennium in women. The analysis updated to 2007
incidence and epidemiology showed a greater reduction of the mortality rate in the young
Colorectal cancer (CRC) is the third most common tumour in population (aged 30–49 years) with ∼10% per quinquennium
men and the second in women, accounting for 10% of all [4]. In Europe, the 5-year survival for colon cancer in different
tumour types worldwide. Incidence is higher in males (ratio: geographical settings ranged from 28.5% to 57% in men and
1.4) and for both genders there is a 10-fold difference in from 30.9% to 60% in women: the pooled estimation for 51
incidence between several regions. With 608 000 deaths registries of 23 countries is 46.8% in men and 48.4% in
clinical practice
estimated each year (∼8% of all cancer deaths), CRC is the women [5].
guidelines
fourth most common cancer-related cause of death in the The risk of developing colon cancer depends on factors
world [1]. which can be classified into lifestyle or behavioural factors
As a general observation, there has been an increasing (such as smoking, high red meat consumption, obesity,
incidence in countries where the overall risk of large bowel physical inactivity) and genetically determinant factors.
cancer was low, while in historically high-risk countries either a According to international guidelines [6, 7], screening tests
stabilisation (Western Europe and Australia) or a decrease are stratified according to the personal risk of disease. Age is
(USA, Canada and New Zealand) in incidence was reported [2]. considered the major unchangeable risk factor for sporadic
A gradient of incidence and mortality between North Western colon cancer: nearly 70% of patients with colon cancer are
and South Eastern Europe has been observed: new CRC cases over 65 years of age, and this disease is rare before 40 years
increased in historically low-risk areas such as Spain and even if data from SEER and Western registries show an
Eastern Europe [3]. This growing incidence reflects increased incidence in the 40–44 years group and a decrease
modifications in lifestyle behaviours and their consequences in the oldest groups [8].
related with ‘westernisation’ such as obesity, physical inactivity, Individuals with:
heavy alcohol consumption, high red meat consumption and
(i) a personal history of adenoma, colon cancer, inflammatory
smoking.
bowel disease (Crohn’s disease and ulcerative colitis),
Mortality has declined progressively in many Western
(ii) significant family history of CRC or polyps,
countries: this can be attributed to cancer screening
(iii) an inherited syndrome (5–10% of all colon cancers) such
programmes, removal of adenomas, early detection of
as familial adenomatous polyposis coli and its variants
cancerous lesions and availability of more effective therapies,
(1%), Lynch-associated syndromes [hereditary non-
chiefly for early stage disease. Mortality rates for CRC in the
polyposis colon cancer (3–5%)], Turcot-, Peutz-Jeghers-
European Union (EU) vary between 15 and 20 of 100 000
and MUTYH-associated polyposis syndromes,
males and between 9 and 14 of 100 000 females and have
decreased in both Western and Northern Countries, are considered at high risk of colon cancer and must be actively
particularly in females. In 10 years (1997–2007), EU mortality screened and, in cases of inherited syndromes, also referred for
genetic counselling [7, 9].
*Correspondence to: ESMO Guidelines Working Group, ESMO Head Office, Via, L. Taddei
4, CH-6962 Viganello-Lugano, Switzerland;
E-mail: clinicalguidelines@esmo.org screening principles
†
The aim of screening is to detect a pre-cancer condition in a
Approved by the ESMO Guidelines Working Group: April 2002, last update July 2013.
This publication supersedes the previously published version—Ann Oncol 2010; 21 healthy population, as well as very early-stage malignancies
(Suppl. 5): v70–v77. which can be treated with a clearly curative intention.
Downloaded from https://academic.oup.com/annonc/article-abstract/24/suppl_6/vi64/161392

by guest
on 05 February 2018
For average-risk populations, the European Guidelines for survival [11]. A systematic review and meta-analysis of the
quality assurance in CRC screening and diagnosis [10] provide published literature were carried out to assess the diagnostic
‘guiding principles and evidence-based recommendations on accuracy (sensitivity, specificity, and positive and negative
quality assurance which should be followed when implementing ratios) of alarm features in predicting large bowel cancer,
CRC screening using the various modalities currently adopted in resulting in a pooled prevalence of CRC of 6% (95% CI: 5% to
publically mandated programmes in EU member States’. 8%) in >19 000 cases, and only dark red rectal bleeding and
The recommendations are: abdominal mass had a specificity of >95%, suggesting that the
presence of either characteristic strongly indicates a diagnosis of
Only the faecal occult blood test (FOBT) for men and CRC [12]. Colon cancer can occur as multiple or synchronous
women aged 50–74 (or 70) years has been recommended (2.5%) with identical or different histological patterns and stages
to date. In average-risk populations, the guaiac (g) FOBT of development.
reduced mortality from CRC by ∼15% [I] in different age Patients with synchronous primary tumours have the same
groups [I, V]. The benefit from annual screening appears to prognosis as patients with single site colon cancers.
be greater than for biennial screening and the test interval Metachronous primary tumours arise in up to 3% during 5
should not exceed 2 years [II, B]. years after surgery, and the incidence increases up to 9% after
Faecal immunochemical testing appears to be superior to several decades in long-term survivors.
gFOBT with respect to the detection rate and positive
predictive value for adenomas and cancer [III]; the test
interval should not exceed 3 years [V]. diagnostic procedures
Flexible sigmoidoscopy (FS) reduces CRC incidence and
Endoscopy is the main procedure for diagnosis and can be
mortality when carried out in an organised screening
carried out by either sigmoidoscopy (as >35% of tumours are
programme [II]; the optimal interval should not be <10
located in the rectosigmoid) or ( preferably) a total colonoscopy.
years and may even be extended to 20 years [IV, C]. The
The advantages of endoscopy are many, e.g. determination of
preferred age range is likely to be between 55 and 64 years
the exact localisation and biopsy of the lesion, detection of
[III, C]. After age 74, average risk FS screening should be
(further) synchronous precancerous or cancerous lesions and
discontinued, given the increasing co-morbidity in this
removal of polyps. Before surgery, if a complete colonoscopy
population [V, D].
cannot be carried out for whatever reason, the rest of the colon
Colonoscopy: limited evidence exists on the efficacy in
should be visualised by combining limited left-sided
reducing CRC incidence and mortality [III]. A note of
colonoscopy with barium enema in order to study the proximal
caution is the observation that colonoscopy screening may
colon. Virtual colonoscopy or CT colonography are not yet
not be as effective in the right colon as in other segments of
standard investigations, but are valuable instruments to identify
the large bowel [IV]. The age range is 50–74 years [V, D]
with precision the location of the tumour or to detect
with the optimal age for a single colonoscopy being around
synchronous lesions or polyps, and they are potentially helpful
55 years [IV, C]. The optimal interval should not be <10
for patients eligible for laparoscopic resection. In any case, if not
years and may even be extended up to 20 years [III, C].
carried out before, a complete colonoscopy should be carried
Combination of FOBT and sigmoidoscopy: there is no
out within 3–6 months after surgery [V, B].
current evidence for extra benefit from adding a once-only
sigmoidoscopy to FOBT screening [II].
New screening technologies are still under evaluation:
pathology
computed tomography (CT) colonography, stool DNA
testing and capsule endoscopy should therefore not be used The standard assessment should include the morphological
for screening in the average-risk population [V, D]. description of the specimen, surgical procedure carried out,
definition of tumour site and size, presence or absence of
macroscopic tumour perforation, histological type and
diagnosis grade, extension of tumour into the bowel wall and
adjacent organs (T stage), distance of cancer from resected
symptoms margins ( proximal, distal and radial), presence or absence
Colon cancer arises from the mucosa of the bowel, generally of tumour deposits, lymphovascular and/or perineural
growing towards the lumen and/or spreading to adjacent invasion, presence of tumour budding, site and number of
organs. Symptoms are associated with relatively large tumours removed regional lymph nodes and their possible
and/or advanced disease stages, and are generally not specific infiltration by cancer cells (N stage), and finally the
for colon cancer. Change in bowel habits, general or localised possible involvement of other organs (e.g. liver) if
abdominal pain, weight loss without other specific causes, submitted for removal or biopsy (M stage) [13].
weakness, iron deficiency and anaemia are the most common The pathological stage must be reported according to the
symptoms, and depends on the location and stage of the American Joint Cancer Committee (AJCC)/ Union for
primary tumour; they are associated with worse prognosis and International Cancer Control (UICC) TNM classification, 7th
their number (but not their duration) is inversely related to edition (Table 1).
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt354 | vi

by guest
on 05 February 2018
staging and risk assessment Table 1. Continued
staging procedures Stage T N M Dukes MACe

Once a colon cancer is diagnosed, clinical examination, Anatomic stage/prognostic groups
laboratory tests and instrumental screening should be carried 0 Tis N0 M0 – –
out in order to detect or to exclude metastatic disease. Clinical I T1 N0 M0 A A
examination may show visceromegaly (hepatomegaly or T2 N0 M0 A B1
lymphadenopathy), ascites and/or synchronous tumours IIA T3 N0 M0 B B2
(chiefly in women: ovarian, endometrial and breast cancers). IIB T4a N0 M0 B B2
Liver enzymes are generally obtained preoperatively, even if they IIC T4b N0 M0 B B3
can be normal in the presence of metastases. Ultrasonography IIIA T1-T2 N1/N1c M0 C C1
of the liver and the whole abdomen may be useful, but a CT T1 N2a M0 C C1
scan is usually more appropriate, in order to detect a metastatic IIIB T3-T4a N1/N1c M0 C C2
T2-T3 N2a M0 C C1/C2
spread to the liver or complications related to the tumour
T1-T2 N2b M0 C C1
( perforation, fistula, obstruction…) [V, B]. However, sensitivity
IIIC T4a N2a M0 C C2
of the CT scan in detecting peritoneal implants is relatively poor
T3-T4a N2b M0 C C2
(and influenced by lesion size). Magnetic resonance imaging
T4b N1-N2 M0 C C3
might be useful for locally advanced tumours and could also be IVA Any T Any N M1a – –
the preferred first-line investigation for evaluating liver IVB Any T Any N M1b – –
metastases in patients who have not previously undergone
a
therapy [14]. The clinical benefit of routine chest CT scan is This includes cancer cells confined within the glandular basement
controversial and its use is not generally recommended [III, D]. membrane (intraepithelial) or mucosal lamina propria (intramucosal) with
Similarly, the routine use of positron emission tomography no extension through the muscularis mucosae into the submucosa.
b
(PET) with the glucose analogue 18-fluoro-2-deoxy-D-glucose Direct invasion in T4 includes invasion of other organs or other segments
(FDG-PET) is not recommended at the time of initial diagnosis, of the colorectum as a result of direct extension through the serosa, as
as it does not modify the treatment approach in the vast confirmed on microscopic examination (for example, invasion of the
majority of patients [15]. sigmoid colon by a carcinoma of the caecum) or, for cancers in a
The preoperative evaluation of the serum marker retroperitoneal or subperitoneal location, direct invasion of other organs or
carcinoembryonic antigen (CEA) is useful for postoperative structures by virtue of extension beyond the muscularis propria (that is, a
tumour on the posterior wall of the descending colon invading the left
kidney or lateral abdominal wall; or a mid or distal rectal cancer with
Table 1. The TNM staging system, AJCC/UICC 7th edition
invasion of prostate, seminal vesicles, cervix or vagina).
c
Tumour that is adherent to other organs or structures, grossly, is classified
Primary tumour (T) cT4b. However, if no tumour is present in the adhesion, microscopically, the
TX Primary tumour cannot be assessed classification should be pT1-4a depending on the anatomical depth of wall
T0 No evidence of primary tumour invasion. The V and L classifications should be used to identify the presence
Tis Carcinoma in situ: intraepithelial or invasion of lamina propriaa or absence of vascular or lymphatic invasion, whereas the PN site-specific
T1 Tumour invades submucosa factor should be used for perineural invasion.
T2 Tumour invades muscularis propria d
A satellite peritumoural nodule in the pericolorectal adipose tissue of a
T3 Tumour invades through the muscularis propria into the primary carcinoma without histologic evidence of residual lymph node in
pericolorectal tissues the nodule may represent discontinuous spread, venous invasion with
T4a Tumour penetrates into the surface of the visceral peritoneumb extravascular spread (V1/2) or a totally replaced lymph node (N1/2).
T4b Tumour directly invades or is adherent to other organs or structuresb,c Replaced nodes should be counted separately as positive nodes in the N
Regional lymph nodes (N)d category, whereas discontinuous spread or venous invasion should be
NX Regional lymph nodes cannot be assessed classified and counted in the site-specific factor category Tumour Deposits.
N0 No regional lymph node metastasis e
MAC is the modified Astler–Coller classification.
N1 Metastasis in one to three regional lymph nodes Edge et al. [17]. Used with the permission of the American Joint Committee
N1a Metastasis in one regional lymph node on Cancer (AJCC), Chicago, IL. The original source for this material is the
N1b Metastasis in two to three regional lymph nodes AJCC Cancer Staging Handbook, 7th edition (2010) published by Springer
N1c Tumour satellite deposits in subsierose or in non peritonealised Science and Business Media LLC, www.springer.com.
tissues
N2 Metastases in ≥4 regional lymph nodes (a: 4–6, b: ≥7)
follow-up of CRC patients (or for use in the treatment of
Distant metastases (M)
metastatic disease), while it has a low predictive value for
M0 No distant metastases
diagnosis in asymptomatic patients due to its relatively low
M1 Distant metastases
sensitivity and specificity [16]. The CEA level may have a
M1a Metastases confined to one organ or site (for example liver, lung,
ovary, nonregional node)
prognostic value in the preoperative setting (>5 ng/dl suggests a
M1b Metastases in more than one organ/site or the peritoneum
worse prognosis). An increased preoperative value not
normalised after 1 month following surgical resection may
continued indicate persistent disease.
vi | Labianca et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
Surgical staging includes an assessment of liver metastases, and therefore the use of any predictive marker information for
nodal spread of disease and extension of the tumour through decision making is not indicated [IV, C].
the bowel wall and onto adjacent structures. For adequate pN- Generally, adjuvant treatment is recommended for stage III
staging, at least 12 nodes have to be examined [17]. This is and ‘high-risk’ stage II patients [A]. The first issue is therefore
particularly important for determination of stage II status, as it how to define the risk. The 5-year survival after surgical
has been shown that patient prognosis is much better if at least resection alone is:
14 tumour-free nodes have been presented. It is not entirely
clear, however, if this is a surgical (resecting more nodes) or a (i) stage I: 85%–95%,
pathological (finding more nodes and preventing inaccurate (ii) stage II: 60%–80%,
classification of stage II) issue. Intra-operative ultrasound is a (iii) stage III: 30%–60%.
more accurate assessment for liver metastases: occult liver The wide ranges reflect major differences in prognosis
metastases can be found in 15% of patients; in 5% these are depending upon the stage subset, tumour grading and the other
solitary and could easily be resected. biological characteristics discussed below.
Several newer predictors have been recently examined,
including microsatellite instability (MSI)/mismatch repair
risk assessment (MMR), 18q deletion, k-ras mutations, TP53, TGFBR2, DCC
Although local failure rates are very low in colon cancer, and thymidilate synthase gene expression.
systemic recurrence of the disease following surgery is frequent The most promising risk factors at the present time are
and is very often the ultimate cause of death. The prognosis of represented by allelic loss of chromosome 18q (negative for
colon cancer is clearly related to the staging features of the TNM prognosis) and MSI/MMR ( positive for prognosis). In
classification, including the degree of penetration of the tumour particular, MSI/MMR may be useful to identify a small (10%–
through the bowel wall and the presence, or absence, of nodal 15%) subset of stage II patients who are at a very low risk of
involvement. However, many additional parameters such as recurrence and in whom the benefits of chemotherapy are very
grading, lymphatic or venous or perineural invasion, lymphoid unlikely. Beyond this prognostic information, the MSI/MMR
inflammatory response and involvement of resection margins, status is not useful for guidance on treatment decisions,
which are reflected by the Dukes’ and TNM classifications, have reflecting the heterogeneity of data for the potential predictive
been shown to have strong prognostic impact. Furthermore, value [19–22]. In stage III, the role of MSI/MMR status is not
factors such as p53, k-ras and bcl-2 expression, TGF-alpha, clear: conflicting data exist on the potential benefit of treatment
EGFR, proliferation index and aneuploidy are under evaluation with 5-FU alone in the older studies and in the more recent
for their single or combined value under high-risk conditions. analyses [23], whereas no conclusive data are available for
Bowel obstruction and perforation are clinical indicators of a oxaliplatin. Therefore, MSI/MMR does not need to be
poor prognosis. determined if an oxaliplatin combination is planned [IV, D].
Elevated pre-treatment serum levels of CEA and/or The general consensus suggests that patients with stage II are
carbohydrate antigen 19-9 (CA 19-9) have a negative prognostic considered at high risk if they present at least one of the
significance. An age of >70 years at presentation is not a following clinical characteristics: lymph nodes sampling <12;
contraindication to standard therapies; acceptable morbidity poorly differentiated tumour; vascular or lymphatic or
and mortality, as well as long-term survival, are achieved in this perineural invasion; tumour presentation with obstruction or
patient population. Some retrospective studies suggested that tumour perforation and pT4 stage [II].
perioperative blood transfusions could impair the prognosis, but During risk assessment, one must integrate all known
these findings were not confirmed by a large, multi-institutional, tumour-related prognostic factors starting from the stage and
prospective randomised trial which demonstrated no benefit for grade and deriving a rough estimate of the chances of relapse.
autologous blood transfusions when compared with allogeneic For example, a patient with a stage II G3 adenocarcinoma with
transfusions [18]. blood vessel invasion, presence of tumour budding and high
Risk assessment is particularly important in order to decide thymidine labelling index, is likely to have >70% chance of
when to propose an adjuvant treatment to an individual patient. relapse, much higher in comparison to another patient with a
As it is well known, adjuvant therapy is a systemic treatment stage IIIA G1 lesion but with opposite pathological and
administered after primary tumour resection with the aim of biological parameters.
reducing the risk of relapse and death. It is well established that Another important problem is tailoring the decision to each
in colon cancer, adjuvant therapy decreases the risk of death by individual patient’s clinical characteristics. In this context, the
absolute 3%–5% in stage II with single-agent 5-FU and by most debated issue is the impact of age on decision making.
10%–15% in stage III with fluoropyrimidines alone plus a The median age of patients presenting with CRC is 72 years,
further 4%–5% with oxaliplatin-containing combinations [I, A]. whereas the median age of patients in clinical trials is 63 years
Each treatment option, including observation alone, should be and <10% of patients >70 years are accrued in the studies.
thoroughly discussed with the patient, taking into consideration When facing an elderly patient (>age 70) with a resected high-
prognostic aspects of the tumour disease, non-disease-related risk CRC, one must keep in mind that:
characteristics (such as performance status, age, comorbidities, (i) the life expectancy of a 70-year old otherwise healthy
etc.) and the individual’s preferences. individual is ∼8 years for men and 14 years for women; (ii)
Notably, there is no evidence for a predictive marker toxicity of chemotherapy is similar below and above age 70 [II];
regarding the benefit of adjuvant chemotherapy for early CRC, (iii) the efficacy of adjuvant treatments is similar in elderly
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt354 | vi

by guest
on 05 February 2018
people compared with that in the general population [II]; (iv) supply and distribution of regional lymph nodes. The resection
data from pooled analyses indicate that patients >70 years may should include a segment of colon of at least 5 cm on either side
not benefit significantly from oxaliplation-based combinations of the tumour, although wider margins are often included
in the adjuvant setting. However, they may have a similar because of obligatory ligation of the arterial blood supply [IV, B].
benefit to younger patients from 5-FU-based chemotherapy To clearly define stage II versus III and to identify and
[24]. A subset analysis of the MOSAIC trial also confirms that eradicate potential lymph node metastases, at least 12 lymph
patients over 70 years may not further benefit from the addition nodes must be resected [IV, B].
of oxaliplatin [25]. Laparoscopic approach has now received wide acceptance for
Recently, nomograms have been developed and are also several types of surgical procedures of major abdominal surgery.
available for CRC. These statistics-based tools attempt to Laparoscopic colectomy can be safely carried out for colon
provide all proven prognostic factors and to quantify the risk of cancer, particularly for left-sided cancer [I]. For right-sided
5- and 10-year death as precisely as possible [26]. colonic cancers, the benefit is less obvious since anastomosis
must be hand sewn, which requires a laparotomy [IV]. The
long-term oncological results of laparoscopic colectomy are
management of local/locoregional similar to those of the conventional approach [27] [I].
disease Advantages of laparoscopy over the conventional approach are
treatment of malignant polyps reduced pain, reduced length of hospital stay and reduced
duration of ileus [28] [II]. It is recognised that a laparoscopic
Complete endoscopic polypectomy should be carried out approach should only be carried out if the following criteria are
whenever the morphological structure of the polyp permits. The met:
presence of invasive carcinoma in a polyp requires a thorough
review with the pathologist for histological features that are (i) technically experienced surgeons,
associated with an adverse outcome. Making the decision to (ii) lack of serious abdominal adhesion due to prior major
undergo surgical resection for a neoplastic polyp that contains abdominal surgery,
invasive carcinoma involves the uncertainties of predicting and (iii) no locally advanced disease and/or acute bowel obstruction
balancing adverse disease outcome against operative risk. or perforation.
Unfavourable histological findings include lymphatic or venous
invasion, grade 3 differentiation, level 4 invasion (invades the Obstructive colorectal cancers can be treated in one or two
submucosa of the bowel wall below the polyp) or involved stages. Two-stage procedures can include colostomy followed by
margins of excision. Although level 4 invasion and involved colonic resection, or Hartmann’s procedure followed by
margins of excision are two of the most important prognostic colostomy closure and anastomosis. An alternative is a one-
factors, their absence does not necessarily preclude an adverse stage procedure with either subtotal colectomy and ileorectal
outcome. Several staging systems to stratify the aggressiveness of anastomosis or, in selected cases, segmental resection after
polyps have been proposed such as involvement of submucosae intraoperative colonic lavage [III]. Endoscopic stenting can be
(sm1, sm2, sm3: involves the superficial, middle and deep thirds used to relieve obstruction from rectosigmoid cancer and
of the submucosa, respectively), invasion into the stalk, absolute allow subsequent one-step resection. Obstructive right-sided
thickness of the invasive tumour beyond the muscolaris cancers can be treated by colonic resection and immediate
mucosae. When unfavourable histological features are present in anastomosis [IV].
a polyp from a patient with an average operative risk, resection
is recommended [IV, B]. The pedunculated polyp with invasive treatment by stage
carcinoma confined to the head, with no other unfavourable Stage 0 (Tis N0 M0)
factors, has a minimal risk for an adverse outcome. The Treatment options are:
consensus is that endoscopic polypectomy is adequate treatment
with proper follow-up examination [IV, B]. Invasion of the stalk (i) Local excision or simple polypectomy.
but with clear margins of excision and favourable histological (ii) Segmentary en-bloc resection for larger lesions not amenable
features may be treated with endoscopic polypectomy with a to local excision.
similar risk as level 2 invasion (invades the muscularis mucosa Stage I (T1-2 N0 M0)
but is limited to the head and neck of the stalk). Pedunculated
polypoid carcinomas can be treated using the same criteria as (old staging: Dukes’ A or modified Astler–Coller A and B1).
other pedunculated polyps with invasive carcinoma. Invasive Wide surgical resection and anastomosis. No adjuvant
carcinoma in a sessile polyp should usually be interpreted as chemotherapy.
having level 4 invasion. Consequently, standard surgical Stage II A, B, C (T3 N0 M0, T4 a-b N0 M0)
resection is recommended in patients with average operative Standard treatment options:
risk [IV, B]. (i) Wide surgical resection and anastomosis.
(ii) Following surgery, adjuvant therapy should not be routinely
localised disease recommended for unselected patients. In high-risk patients
The goal of surgery is a wide resection of the involved segment who present at least one of the previously mentioned
of bowel together with the removal of its lymphatic drainage. clinical high-risk features (see above), adjuvant therapy
The extent of the colonic resection is determined by the blood could be considered in clinical practice [II, B].
vi | Labianca et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
Stage III (any T, N1-N2, M0) event-free survival for IFL, when compared with FL. The
PETACC-3 trial [35] compared LV5FU2 or AIO regimen plus
(i) Wide surgical resection and anastomosis. irinotecan with LV5FU2 or AIO regimen alone. Results did not
(ii) Following surgery, the standard treatment is a doublet show any significant advantage for the regimen with irinotecan
schedule with oxaliplatin and a fluoropyrimidine. Although in terms of DFS.
all three combination regimens are superior to 5-FU/FA There is currently no role for targeted agents associated with
alone [I, A], FOLFOX4 or XELOX should be preferred to chemotherapy in the adjuvant setting for colon cancer. All trials
FLOX. When oxaliplatin is contraindicated, monotherapy evaluating bevacizumab, NSABP C-08 [36], AVANT [37] or
with infusional or oral fluoropyrimidines should be cetuximab, NCCTG NO147 [38] and PETACC-8 [39] are
preferred to bolus 5-FU FU/LV. negative, probably due to different biological characteristics in
early when compared with advanced disease.
In the adjuvant setting many questions are still unanswered:
treatment options
The benefit of combinations with oxaliplatin has been (i) The optimal duration of adjuvant treatment: 3 or 6
demonstrated in three landmark trials. In the MOSAIC study months? In Italy, the TOSCA trial investigates whether 3
[29], the addition of oxaliplatin to 5-FU/LV (FOLFOX schema), months of FOLFOX4 or XELOX treatment are not inferior
demonstrated a significantly increased disease-free survival to 6 months with the same schedule in terms of
(DFS) at 3 years, with a reduction in the risk of recurrence of recurrence-free survival (RFS) in stage II and III colon
23% compared with the control arm (LV5FU2).The update at cancer patients. Together with other studies (SCOT,
the 6-year follow-up confirmed the benefit in DFS of adjuvant France, US, Greece, Japan…), this trial forms the backbone
treatment with FOLFOX4, and an advantage was also observed of a large international collaboration (‘IDEA’) which will
in overall survival (OS), but for stage III patients only [30]. give a definitive answer regarding the duration of adjuvant
The NSABP C-07 trial compared the efficacy of bolus FU/ therapy in stage III patients.
LV + oxaliplatin (FLOX) versus FU/LV alone (Roswell Park (ii) The validation of prognostic/predictive factors: interesting
schedule); 3-year DFS was 76.5% versus 71.6% for FLOX and and potentially positive data are expected from large subset
FULV, respectively [31], and the magnitude of reduction in the analyses from large trials, such as PETACC-3, AVANT and
risk of recurrence was similar to that of the MOSAIC trial. The PETACC-8.
spectrum of toxicity between MOSAIC and NSABP-C07 was (iii) The possible role of aspirin in this setting: from a large
different: grade 3–4 diarrhoea occurred more often with FLOX study [40], it appears that the regular use of this drug after
than with FOLFOX, while grade 3 sensory neuropathy was diagnosis of CRC leads to an increase in cancer-specific
observed in 12% with FOLFOX and 8% with FLOX. FLOX survival and OS but only in patients with mutated PIK3CA
should probably not be used in clinical practice, due to its cancer. Further studies on this topic are needed to confirm
toxicity and also due to a lack of OS benefit. The XELOXA these exciting findings.
international phase III study [32] assessed the safety and efficacy
of adjuvant capecitabine plus oxaliplatin (XELOX) versus bolus personalised medicine
FU/LV (Mayo Clinic or Roswell Park regimen) in stage III In this disease setting, more research is needed to identify
patients: the arm including the oral compound was well molecular markers which could lead to advances in
tolerated and superior to the i.v fluoropyrimidine. As personalised medicine.
capecitabine does not require a central venous access, it may be
preferred in many patients [IV, B]. follow-up and long-term implications
In case a clinically relevant neurotoxicity occurs, oxaliplatin
should be stopped and fluoropyrimidine continued, as it follow-up
contributes to about two-third to the therapeutic effect of Despite optimal primary treatment, with adequate surgery with
adjuvant FOLFOX/XELOX. or without adjuvant chemotherapy, 30%–50% of patients with
As stated before, in special situations monotherapy with colon cancer will relapse, and most of those patients will die
capecitabine or 5-FU/LV in infusion can be an alternative from their disease.
approach. The X-ACT trial showed that capecitabine is an Detecting relapse in advance is the main goal of surveillance
active agent with a favourable toxicity profile and may reduce after primary treatment, but this is clinically meaningful only if
overall costs compared with i.v. treatments [I]: after 4.3 years it improves survival. Furthermore, follow-up can be expensive
of follow-up, the data still confirm the equivalence in terms of and resource-consuming in terms of both money and
DFS between capecitabine and 5-FU/LV [33] in stage procedures for a national health system, so intensive
III patients. surveillance needs to be justified with a good level of evidence.
Negative trials are related to irinotecan in combination with In the past, there was no strong evidence that regular follow-up
5-FU (bolus or infusional). The CALGB-89803 trial [34] could improve the outcome for patients radically resected for
compared 5-FU/LV + irinotecan (IFL) with the Roswell Park colon cancer. Several trials failed to demonstrate a benefit, and
scheme in more than 1200 patients. The trial was prematurely results of old meta-analyses were unfortunately based on
closed because of an elevated rate of mortality in the IFL group nonrandomised data or mix of randomised and cohort studies.
with respect to the FL regimen (2.2% versus 0.8%). Efficacy In the last 10 years, four further systematic reviews [41–44]
results indicated no improvement in terms of either OS or have been published. All of these studies demonstrated improved
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt354 | vi

by guest
on 05 February 2018
survival in patients undergoing more intensive surveillance when detection of intraluminal recurrent cancer, and therefore there
compared with those with minimal or no follow-up. The is no indication of intensive endoscopic follow-up. If a colon
estimated OS gain was between 7% and 13%. On the basis of without tumour or polyps is observed 1 year after resection,
these data, intensive follow-up should now be considered colonoscopy should be carried out after 3–5 years. In this field,
standard practice for colon cancer patients [I, A] [45]. specific recommendations are based mainly on level II and III
The improvement in OS has been attributed to earlier evidence, particularly concerning timing intervals and duration
detection of recurrent disease and in particular, to a higher rate of follow-up [48].
of detection of isolated locoregional relapses. The same rate of A recently reported analysis of individual patient data from
recurrence for intensive and minimal follow-up was reported large adjuvant colon cancer randomised trials including >20 000
[46], but with an anticipation of 8.5 months in the intensive patients indicated that 82% of stage III and 74% of stage II colon
group. Detection of isolated local recurrences was increased in cancer recurrences are diagnosed within the first 3 years after
the intensive group (15% compared with 9%, with RR 1.61 and primary cancer resection [49].
P = 0.011), and also a small non-significant increase in the Suggested recommendations are as follows [50]:
detection of hepatic metastases was reported. Absolute
reduction in mortality was 9%–13%, comparable with the (i) Intensive follow-up must be carried out in colon cancer
benefit observed with adjuvant chemotherapy in stage III. In patients [I, A].
addition, trials included in this analysis were conducted before (ii) History and physical examination and CEA determination
the modern multidisciplinary approach to metastatic disease, are advised every 3–6 months for 3 years and every 6–12
and therefore, the real benefit in clinical practice at the present months at years 4 and 5 after surgery [II, B].
time could be even more evident. (iii) Colonoscopy must be carried out at year 1 and every 3–5
Despite these results, recent data still showed low adherence years thereafter, looking for metachronous adenomas and
to follow-up recommendations. This could be due to difficulties cancers [III, B].
in determining which is the best surveillance test. (iv) CT scan of chest and abdomen every 6–12 months for the
Indeed, although pooled data suggest a survival advantage first 3 years can be considered in patients who are at higher
related to intensive follow-up, the heterogeneity of the studies risk of recurrence [II, B].
included in these meta-analyses does not allow assessment of (v) CEUS could substitute for abdominal CT scan [III, C].
which kind of surveillance must be applied in clinical practice. It (vi) Other laboratory and radiological examinations are of
seems clear that more investigations are better than fewer, which unproven benefit and must be restricted to patients with
in turn are better than no follow-up at all, but it is nearly suspicious symptoms.
impossible to recommend an optimal strategy with an adequate
level of evidence. As a matter of fact, ‘intensive’ procedures in
one trial can be similar to ‘minimal’ procedures in another trial, long-term implications: survivorship care plans
and surveillance intervals and duration of follow-up cannot be Follow-up is therefore standard practice after completing
extrapolated by meta-analyses data. Only trials including CEA cancer treatment. It consists of periodic visits and investigations,
testing and/or liver imaging achieve significant improvements in which usually take place in a specialist cancer setting sometimes
survival, but all studies considering liver imaging also included for prolonged periods after the treatment ends. However, there
blood CEA monitoring; CEA testing alone does not show has been growing awareness in recent years that optimal
benefit in individual studies and has demonstrated a reduction cancer survivorship care involves more than surveillance tests.
in mortality only in meta-analyses [47]. Despite this, CEA rise is In this setting, the primary practitioner should have a significant
often the first signal of recurrence: a positive value could be role [51].
detected 1.5–6 months before clinical/instrumental detection Survivors of CRC now represent the third largest group of
with other test(s). There are false-positive rates of CEA long-term cancer survivors in Western countries, ∼11% of the
elevation of 7%–16% and false-negative rates of up to 40%. CEA population.
test monitoring is also effective in patients without elevation in Survivorship care plans are an increasing priority within the
the preoperative setting: in these patients, a subsequent cancer system, and supporting colon cancer patients and their
elevation was observed in 44% of recurrent patients. There is no providers in staying on track with recommended follow-up is
evidence that other laboratory tests can be useful. only part of helping patients to regain their health and stay well
As far as liver imaging is concerned, a CT scan has been after treatment.
shown to be more sensitive than ultrasonography (0.67 Data from the Tjandra surveillance meta-analysis [44]
compared with 0.43), but a modern contrast enhancement reported benefit in the intensive arm, but this benefit was less
ultrasound scan (CEUS) can substantially increase the significant. In fact, in this systematic revision it was observed
sensitivity of ultrasonography. Chest recurrence could be that, despite benefit in OS and in re-resection rate, the cancer-
detected by the CT scan: in colon cancer, lung is the first site of related mortality was not improved and the survival benefit was
relapse in 20% of patients and pulmonary resection could not due to early detection and treatment of recurrent disease.
determine a 30% 5-year survival. In contrast, there are no data Other factors contributing to a survival advantage of
in favour of regular use of chest X-rays. surveillance in these patients might include the management of
Metachronous primary cancer could be detected with an co-morbidities, promotion of beneficial dietary and lifestyle
incidence of 0.7% within the first 2 years after curative surgery, factors and increased psychosocial support.
but there is no evidence favouring a survival benefit through the Major elements in survivorship care are as follows [52]:
vi | Labianca et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
(i) Prevention of recurrent and new cancer (classic end-point issues are equal in importance in the care of index cancer. This
of follow-up), simple consideration is often not applied in clinical practice. A
(ii) Intervention for cancer sequelae and their treatment recent analysis [54] showed that survivors were overall less likely
(rehabilitation), to receive recommended follow-up for chronic conditions, such
(iii) Assessment of medical and psychological late effects as angina, congestive heart failure and chronic lung disease, and
(modern end-point of follow-up), to receive less of some types of preventive care compared with
(iv) Health promotion (lifestyle promotion, co-morbidity matched controls. For example, diabetic cancer survivors were
prevention, etc.). less likely to have preventive eye examinations, and the data
showed a trend toward less intensive monitoring of HbA1c.
Most long-term survivors of CRC report very good quality of Even though at the present time, this issue can be considered
life following their treatment, but several problems are still still experimental, the development of Survivorship Care Plans
observed. Some patients can have bowel dysfunction: diarrhoea, will certainly be one of the major topics in the near future for
constipation, bowel obstruction, pain. Hemicolectomy can lead gastro intestinal medical oncologists.
to loose stools but this usually improves over time and surgery
can also lead to adhesions. It is important to carry out dietary
counselling and suggest use of over-the-counter medications note
(e.g., fibre laxative, stool softeners, antidiarrhoeals). Survivors Levels of evidence and grades of recommendation have been
who received oxaliplatin can experience numbness or painful applied using the system shown in Table 2. Statements without
sensations [53]. grading were considered justified standard clinical practice by
More than 76% of colon cancer survivorships are >65 years of the experts and the ESMO faculty.
age, so long-term employment is not a problem for a majority,
but in young people employment and financial concerns should
be an important issue to consider. conflict of interest
Despite colon cancer survivors often staying well, higher rates Prof. Labianca has reported research grants from Roche and
of psychological depression have been reported. Assessment of Sanofi. Prof. Arnold has reported research grants from Roche
distress should be considered but evidence on the effectiveness and Sanofi. The other authors have reported no potential
of psychosocial interventions among survivors of CRC is conflicts of interest.
limited.
The majority of colon cancer survivors die of other causes.
Consequently, care for general medical and preventive health references
Table 2. Levels of evidence and grades of recommendation (adapted from 1. GLOBOCAN 2008 website: globocan.iarc.fr. (12 July 2013, date last accessed).
the Infectious Diseases Society of America-United States Public Health 2. Jemal A, Center M, deSantis C, Ward EM. Global patterns of cancer incidence and
mortality rates and trends. Cancer Epidemiol Biomarkers Prev 2010; 19:
1893–1907.
3. Center MM, Jemal A, Ward E. International trends in colorectal cancer incidence
Levels of evidence rates. Cancer Epidemiol Biomarkers Prev 2009; 18: 1688–1694.
I Evidence from at least one large randomised, controlled trial of good 4. Bossetti C, Levi F, Rosato V et al. Recent trends in colorectal cancer mortality in
methodological quality (low potential for bias) or meta-analyses of Europe. Int J Cancer 2011; 129: 180–191.
well-conducted randomised trials without heterogeneity 5. Coleman MP, Quaresma M, Berrino F et al. Cancer survival in five continents: a
II Small randomised trials or large randomised trials with a suspicion of worldwide population-based study (CONCORD). Lancet Oncol 2008; 9: 730–756.
bias (lower methodological quality) or meta-analyses of such trials 6. American Cancer Society Guidelines. Colorectal cancer early detection. http://
or of trials with demonstrated heterogeneity www.cancer.org.
III Prospective cohort studies 7. Schmoll HJ, Van Cutsem E, Stein A et al. ESMO consensus guidelines for
IV Retrospective cohort studies or case–control studies management of patients with colon and rectal cancer: a personalized approach to
V Studies without control group, case reports, experts opinions clinical decision making. Ann Oncol 2012; 23: 2479–2516.
Grades of recommendation 8. Davis DM, Marcet JE, Frattini JC et al. Is it time to lower the recommended
screening age for colorectal cancer? J Am Coll Surg 2011; 213: 352–361.
A Strong evidence for efficacy with a substantial clinical benefit, strongly
9. Balmaña J, Balaguer F, Cervantes A, Arnold D. Familial risk-colorectal cancer:
recommended
ESMO Clinical Practice Guidelines. Ann Oncol 2013; 24(Suppl. 6): vi72–79.
B Strong or moderate evidence for efficacy but with a limited clinical
10. European guidelines for quality assurance in colorectal cancer screening and
benefit, generally recommended
diagnosis. Segnan N, Patnick J, von Karsa L (eds). 2010.
C Insufficient evidence for efficacy or benefit does not outweigh the risk
11. McDermott FT, Hughes ES, Pihl E et al. Prognosis in relation to symptom duration
or the disadvantages (adverse events, costs, …), optional in colon cancer. Br J Surg 1981; 68: 846–849.
D Moderate evidence against efficacy or for adverse outcome, generally 12. Ford AC, Veldhuyzen van Zanten SJO, Rodgers CC et al. Diagnostic utility of alarm
not recommended features for colorectal cancer: systematic review and meta-analysis. Gut 2008; 57:
E Strong evidence against efficacy or for adverse outcome never 1545–1553.
recommended 13. Washington MK, Berlin J, Branton P et al. Protocol for the examination of
a
specimens from patients with primary carcinoma of the colon and rectum. Arch
Dykewicz CA. Summary of the guidelines for preventing opportunistic Pathol Lab Med 2009; 133: 1539–1551.
infections among hematopoietic stem cell transplant recipients. Clin Infect 14. Koh JL, Yan TD, Glenn D, Morris DL. Evaluation of preoperative computed
Dis 2001; 33: 139–144. By permission of the Infectious Diseases Society of tomography in estimating peritoneal cancer index in colorectal peritoneal
America. carcinomatosis. Ann Surg Oncol 2009; 16: 327–333.
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt354 | vi

by guest
on 05 February 2018
15. Niekel MC, Bipat S, Stoker J. Diagnostic imaging of colorectal liver metastases 35. Van Cutsem E, Labianca R, Bodoky G et al. Randomized phase III trial comparing
with CT, MR imaging, FDG PET, and/or FDG PET/CT: a meta-analysis of biweekly infusional fluorouracil/leucovorin alone or with irinotecan in the adjuvant
prospective studies including patients who have not previously undergone treatment of stage III colon cancer: PETACC-3. J Clin Oncol 2009; 27:
treatment. Radiology 2010; 257: 674–684. 3117–3125.
16. Thirunavukarasu P, Sukumar S, Sathaiah M et al. C-stage in colon cancer: 36. Allegra CJ, Yothers G, O’Connell MJ et al. Phase III trial assessing bevacizumab in
implications of carcinoembryonic antigen biomarker in staging, prognosis, and stages II and III carcinoma of the colon: results of NSABP protocol C-08. J Clin
management. J Natl Cancer Inst 2011; 103: 689–697. Oncol 2011; 29: 11–16.
17. Edge SB, Byrd DR, Compton CC et al. (eds). AJCC Cancer Staging Manual. 7th 37. de Gramont A, Van Cutsem E, Schmoll HJ et al. Bevacizumab plus oxaliplatin-
edition. New York, NY: Springer, 2010, pp. 143–164. based chemotherapy as adjuvant treatment for colon cancer (AVANT): a phase 3
18. Busch OR, Hop WC, Hoynck van Pependrecht MA et al. Blood transfusions and randomised controlled trial. Lancet Oncol 2012; 13: 1225–1233.
prognosis in colorectal cancer. N Engl J Med 1993; 328: 1372–1376. 38. Alberts SR, Sargent DJ, Nair S et al. Effect of oxaliplatin, fluorouracil and
19. Ribic CM, Sargent DJ, Moore MJ et al. Tumor microsatellite instability status as a leucovorin with or without cetuximab on survival among patients with
predictor of benefit from fluorouracil-based adjuvant chemotherapy for colon resected stage III colon cancer: a randomized trial. JAMA 2012; 307:
cancer. N Engl J Med 2003; 349: 247–257. 1383–1393.
20. Sargent DJ, Marsoni S, Monges G et al. Defective mismatch repair as a predictive 39. Taieb J, Tabernero J, Mini E et al. Adjuvant FOLFOX-4 with or without cetuximab
marker for the lack of efficacy of fluorouracil-based adjuvant therapy in colon (CTX) in patients (PTS) with resected stage III colon cancer: DFS and OS results
cancer. J Clin Oncol 2010; 28: 3219–3226. and subgroup analyses of the PETACC-8 Intergroup phase III trial. Ann Oncol
21. Sinicrope FA, Foster NR, Thibodeau SN et al. DNA mismatch repair status and 2012; 23(suppl 9): abstr # LBA4.
colon cancer recurrence and survival in clinical trials of 5-fluorouracil-based 40. Liao X, Lochhead P, Nishihara R et al. Aspirin use, tumor PIK3CA mutation and
adjuvant therapy. J Natl Cancer Inst 2011; 103: 863–875. colorectal cancer survival. N Engl J Med 2012; 367: 1596–1606.
22. Tejpar S, Saridaki Z, Delorenzi M et al. Microsatellite instability, prognosis and drug 41. Bruinvels DJ, Stiggelbout AM, Kievit J et al. Follow-up of patients with colorectal
sensitivity of stage II and III colorectal cancer: more complexity to the puzzle. J Natl cancer. A meta-analysis. Ann Surg 1994; 219: 174–182.
Cancer Inst 2011; 103: 841–844. 42. Rosen M, Chan L, Beart RW, Jr et al. Follow-up of colorectal cancer: a meta-
23. Roth AD, Delorenzi M, Tejpar S et al. Integrated analysis of molecular and clinical analysis. Dis Colon Rectum 1998; 41: 1116–1126.
prognostic factors in stage II/III colon cancer. J Natl Cancer Inst 2012; 104: 43. Renehan AG, Egger M, Saunders MP et al. Impact on survival of intensive follow
1635–1646. up after curative resection for colorectal cancer: systematic review and meta-
24. McCleary NJ, Meyerhardt JA, Green E et al. Impact of age on the efficacy of newer analysis of randomised trials. BMJ 2002; 324: 813–819.
adjuvant therapies in patients with stage II/III colon cancer: findings from the 44. Tjandra JJ, Chan MK. Follow-up after curative resection of colorectal cancer: a
ACCENT database. J Clin Oncol 2013; 31: 2600–2606. meta-analysis. Dis Colon Rectum 2007; 50: 1783–1799.
25. Tournigand C, André T, Bonnetain F et al. Adjuvant therapy with fluorouracil and 45. Jeffery GM, Hickey BE, Hider P. Follow-up strategies for patients treated for non-
oxaliplatin in stage II and elderly patients (between ages 70 and 75 years) with metastatic colorectal cancer. Cochrane Database Syst Rev 2007; CD002200.
colon cancer: subgroup analyses of the Multicenter International Study of 46. Renehan AG, Egger M, Saunders MP et al. Mechanisms of improved survival from
Oxaliplatin, Fluorouracil and Leucovorin in the Adjuvant Treatment of Colon Cancer intensive follow-up in colorectal cancer: a hypothesis. Br J Cancer 2005; 92:
trial. J Clin Oncol 2012; 30: 3353–3360. 430–433.
26. Weiser MR, Landmann RG, Kattan MW et al. Individualized prediction of colon 47. Chau I, Allen MJ, Cunningham D et al. The value of routine serum carcino-
cancer recurrence using a nomogram. J Clin Oncol 2008; 26: 380–385. embryonic antigen measurement and computed tomography in the surveillance of
27. Clinical Outcomes of Surgical Therapy Study Group. A comparison of patients after adjuvant chemotherapy for colorectal cancer. J Clin Oncol 2004; 22:
laparoscopically assisted and open colectomy for colon cancer. N Engl J Med 1420–1429.
2004; 350: 2050–2059. 48. Rex DK, Kahi CJ, Levin B et al. Guidelines for colonoscopy surveillance after
28. Hewett PJ, Allardyce RA, Bagshaw PF et al. Short-term outcomes of the cancer resection: a consensus update by the American Cancer Society and US
Australasian randomized clinical study comparing laparoscopic and conventional Multi-Society Task Force on Colorectal Cancer. Gastroenterology 2006; 130:
open surgical treatments for colon cancer: the ALCCaS trial. Ann Surg 2008; 248: 1865–1871.
728–738. 49. Sargent DJ, Patiyil S, Yothers G et al. Endpoints for colon cancer adjuvant trials:
29. Andrè T, Boni C, Mounedji-Boudiaf L et al. Oxaliplatin, fluorouracil and leucovorin observations and recommendations based on individual patient data from 20,898
as adjuvant treatment for colon cancer. N Engl J Med 2004; 350: 2343–2351. patients enrolled onto 18 randomized trials from the ACCENT group. J Clin Oncol
30. Andrè T, Boni C, Navarro M et al. Improved overall survival with oxaliplatin, 2007; 29: 4569–4574.
fluorouracil and leucovorin as adjuvant treatment in stage II or III colon cancer in 50. Tsikitis VL, Malireddy K, Gree EA et al. Postoperative surveillance
the MOSAIC trial. J Clin Oncol 2009; 27: 3109–3116. recommendations for early stage colon cancer based on results from the clinical
31. Kuebler JP, Wieand HS, O’Connell MJ et al. Oxaliplatin combined with weekly outcomes of surgical therapy trial. J Clin Oncol 2009; 27: 3671–3676.
bolus fluorouracil and leucovorin as surgical adjuvant chemotherapy for stage II 51. Grunfeld E, Earle CC. The interface between primary and oncology specialty care:
and III colon cancer: results from NSABP C-07. J Clin Oncol 2007; 25: treatment through survivorship. J Natl Cancer Inst Monogr 2010; 25–30.
2198–2204. 52. Sisler JJ, Taylor-Brown J, Nugent Z et al. Continuity of care of colorectal cancer
32. Haller DG, Tabernero J, Maroun J et al. Capecitabine plus oxaliplatin compared survivors at the end of treatment: the oncology–primary care interface. J Cancer
with fluorouracil and folinic acid as adjuvant therapy for stage III colon cancer. J Surviv 2012; 6: 468–475.
Clin Oncol 2011; 29: 1465–1471. 53. Howell D, Hack TF, Oliver TK et al. Models of care for post-treatment follow-up of
33. Twelves C, Wong A, Nowacki MP et al. Capecitabine as adjuvant treatment for adult cancer survivors: a systematic review and quality appraisal of the evidence.
stage III colon cancer. N Engl J Med 2005; 352: 2696–2704. J Cancer Surviv 2012; 6: 359–371.
34. Saltz LB, Niedzwiecki D, Hollis D et al. Irinotecan fluorouracil plus leucovorin is not 54. Grunfeld E, Earle CC, Stovall E. A framework for cancer survivorship research
superior to fluorouracil plus leucovorin alone as adjuvant treatment for stage III and translation to policy. Cancer Epidemiol Biomarkers Prev 2011; 20:
colon cancer: results of CALGB 89803. J Clin Oncol 2007; 25: 3456–3461. 2099–2104.
vi | Labianca et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
Annals of Oncology 28 (Supplement 4): iv22–iv40, 2017
doi:10.1093/annonc/mdx224
CLINICAL PRACTICE GUIDELINES
Rectal cancer: ESMO Clinical Practice Guidelines

R. Glynne-Jones1, L. Wyrwicz2, E. Tiret3,4, G. Brown5, C. Rödel6, A. Cervantes7 & D. Arnold8, on behalf of

the ESMO Guidelines Committee*
1
Department of Radiotherapy, Mount Vernon Centre for Cancer Treatment, Northwood, London, UK; 2Department of Gastrointestinal Cancer, Maria Sklodowska-
Curie Memorial Cancer Center and Institute of Oncology, Warsaw, Poland; 3Department of Surgery, Sorbonne Universités, UPMC Univ Paris 06, Paris; 4APHP, Hôpital
Saint-Antoine, Paris, France; 5Department of Radiology, The Imperial College and Royal Marsden Hospital, Sutton, Surrey, UK; 6Department of Radiotherapy and
Oncology, University of Frankfurt, Frankfurt, Germany; 7CIBERONC, Medical Oncology Department, INCLIVA University of Valencia, Valencia, Spain; 8Instituto CUF de
Oncologia (I.C.O.), Lisbon, Portugal
*Correspondence to: ESMO Guidelines Committee, ESMO Head Office, Via L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland. E-mail: clinicalguidelines@esmo.org
†
Approved by the ESMO Guidelines Committee: August 2002, last update May 2017. This publication supersedes the previously published version—Ann Oncol 2013;
24 (Suppl. 6): vi81–vi88.
Incidence and epidemiology is a critical component of management, driving surveillance and

potential interventions for the patient and affected family mem-
The incidence of rectal cancer in the European Union is bers [6].
125 000 per year, i.e. 35% of the total colorectal cancer inci-
dence, reflecting 15–25 cases/100 000 population per year and
is predicted to increase further in both genders. The mortality is Diagnosis and pathology/molecular biology
4–10/100 000 population per year. Median age at diagnosis Diagnosis is based on a digital rectal examination (DRE) and en-
is 70 years, but predictions suggest that this figure will rise in doscopy with biopsy for histopathological confirmation (Figure 1).
the future. Tumours with distal extension to 15 cm from the anal margin (as
Evidence is accumulating that rectal cancer is distinct from measured by rigid sigmoidoscopy) are classified as rectal and more
colon cancer with different aetiologies and risk factors [1–2], pos- proximal tumours as colonic. Cancers are categorised as low (up to
sibly reflecting different environmental exposures. High body 5 cm), middle (from > 5 to 10 cm) or high (from > 10 up to
mass index, body or abdominal fatness and diabetes type II are 15 cm).
seen as risk factors. Longstanding ulcerative colitis and Crohn’s The Cancer Genome Atlas Network analysis showed common
disease affecting the rectum, excessive consumption of red or genomic profiles for non-hypermutated colon and rectal cancers
processed meat and tobacco as well as moderate/heavy alcohol [7]. Unique subtypes are characterised by accumulation of
use increase the risk. distinct genetic and epigenetic alterations (DNA methylation),
A healthy lifestyle and exercise can reduce the risk of develop- differing slightly from colon cancer [8, 9]. A transcriptional
ing rectal cancer [3, 4]. Consumption of garlic, milk, calcium and subtype with high Wnt signalling, stem cell and mesenchymal
high dietary fibre are regarded as protective [5]. Although regular signatures occurs in rectal cancer and has a poor prognosis. Such
use of non-steroidal anti-inflammatory drugs (NSAIDs) is asso- patients may also gain less benefit from adjuvant chemotherapy
ciated with reduced incidence, and there may be a protective ef- (ChT) [10].
fect of vitamin D via antitumour immunity, no formal guidelines
for pharmacological primary prevention should be advised.
The majority of rectal cancers develop via the chromosomal in-
stability (CIN) pathway. About 13% are caused by deficient mis- Staging and risk assessment
match repair (dMMR). There is a recognised hereditary A specialised and dedicated multidisciplinary team (MDT) of
component, although this is more pronounced for colon than named radiologists, surgeons, radiation oncologists, medical on-
rectal cancer. The most common disorders are Lynch syndrome cologists and pathologists should attend regular meetings and
and familial adenomatous polyposis. Hence, genetic counselling discuss all (relevant) patients [III, A]. Core members should be
VC The Author 2017. Published by Oxford University Press on behalf of the European Society for Medical Oncology.
All rights reserved. For Permissions, please email: journals.permissions@oup.com.
Downloaded from https://academic.oup.com/annonc/article-abstract/28/suppl_4/iv22/3958158

by guest
on 05 February 2018
Annals of Oncology Clinical Practice Guidelines
DRE
Biopsy for
Endoscopy histopathological
confirmation
Distal extension Distal extension

to ≤ 15 cm from to > 15 cm from
anal margin anal margin
Rectal cancer Colon cancer
Up to 5 cm from > 5 to 10 cm from > 10 to 15 cm from

anal margin anal margin anal margin
Low rectal cancer Middle rectal cancer High rectal cancer
Figure 1. Rectal cancer diagnosis.

DRE, digital rectal examination.
present for the discussion of all cases where their input is needed Table 1. Diagnostic work-up in primary rectal cancer
[11]. There should be a MDT coordinator, and clinical guidelines Parameter Method of choice
should be taken into account in decision-making. The MDT
should also audit whether their decisions are implemented [12] Location (distance DRE/palpation
and review patient outcomes with standardised quality from anal verge) Rigid sigmoidoscopy
assurance. (flexible endoscopy)
A history and physical examination including DRE, full blood
Morphological verification Biopsy
count, liver and renal function tests, serum carcinoembryonic
antigen (CEA) and computed tomography (CT) scan of thorax cT stage
and abdomen should be carried out to define functional status Early ERUS
and presence of metastases [III, A] (Figure 2). Positron emission MRI
tomography (PET) may provide additional information in terms Intermediate/advanced MRI (ERUS)
of disease outside the pelvis. However, current evidence is not Sphincter infiltration MRI (ERUS, palpation, EUA)
considered strong enough to recommend the use of PET in all pa-
cN stage MRI (CT, ERUS)
tients [V, C] (Table 1) [13].
Increasing age, comorbidity and decreasing functional reserves M stage CT, MRI (or US) of the liver/abdomen
are associated with higher early postoperative mortality and CT of the thorax
worse toxicity from radiotherapy (RT) and ChT in older patients. PET-CT if extensive EMVI for other sites
Hence, for patients over 70 years, formal geriatric assessment or Evaluation for all patients MDT discussion
at least screening tools for frailty are recommended before any
treatment [III, C] [14]. Methods within brackets are less optimal.
Rigid rectoscopy and preoperative colonoscopy to the caecal CT, computed tomography; DRE, digital rectal examination; EMVI, extra-
pole are required, or, in the case of obstruction, virtual colonos- mural vascular invasion; ERUS, endorectal ultrasound; EUA, examination
copy to exclude synchronous colonic tumours. If no preoperative under anaesthesia; MDT, multidisciplinary team; MRI, magnetic reson-
(virtual) colonoscopy was carried out, completion colonoscopy ance imaging; PET, positron emission tomography; US, ultrasound.
is recommended within 6 months of surgery [III, A].
Volume 28 | Supplement 4 | August 2017 doi:10.1093/annonc/mdx224 | iv23

by guest
on 05 February 2018
Clinical Practice Guidelines Annals of Oncology
Table 2. UICC TNM staging (8th edition) classification for colon and rectal cancer [16]
TNM Clinical Classification
T—Primary tumour
TX Primary tumour cannot be assessed
T0 No evidence of primary tumour
Tis Carcinoma in situ: invasion of lamina propriaa
T1 Tumour invades submucosa
T2 Tumour invades muscularis propria
T3 Tumour invades subserosa or into non-peritonealised pericolic or perirectal tissues
T4 Tumour directly invades other organs or structuresb,c,d and/or perforates visceral peritoneum
T4a Tumour perforates visceral peritoneum
T4b Tumour directly invades other organs or structures
N—Regional lymph nodes
NX Regional lymph nodes cannot be assessed
N0 No regional lymph node metastasis
N1 Metastasis in 1–3 regional lymph nodes
N1a Metastasis in 1 regional lymph node
N1b Metastasis in 2–3 regional lymph nodes
N1c Tumour deposit(s), i.e. satellites,e in the subserosa, or in non-peritonealised pericolic or perirectal
soft tissue without regional lymph node metastasis
N2 Metastasis in 4 or more regional lymph nodes
N2a Metastasis in 4–6 regional lymph nodes
N2b Metastasis in 7 or more regional lymph nodes
M—Distant metastasis
M0 No distant metastasis
M1a Metastasis confined to one organ (liver, lung, ovary, non-regional lymph node(s)) without peritoneal metastases
M1b Metastasis in more than one organ
M1c Metastasis to the peritoneum with or without other organ involvement
a
Tis includes cancer cells confined within the mucosal lamina propria (intramucosal) with no extension through the muscularis mucosae into the
submucosa.
b
Invades through to visceral peritoneum to involve the surface.
c
Direct invasion in T4b includes invasion of other organs or segments of the colorectum by way of the serosa, as confirmed on microscopic examin-
ation, or for tumours in a retroperitoneal or subperitoneal location, direct invasion of other organs or structures by virtue of extension beyond the mus-
cularis propria.
d
Tumour that is adherent to other organs or structures, macroscopically, is classified cT4b. However, if no tumour is present in the adhesion, microscop-
ically, the classification should be pT1–3, depending on the anatomical depth of wall invasion.
e
Tumour deposits (satellites) are discrete macroscopic or microscopic nodules of cancer in the pericolorectal adipose tissue’s lymph drainage area of a
primary carcinoma that are discontinuous from the primary and without histological evidence of residual lymph node or identifiable vascular or neural
structures. If a vessel wall is identifiable on H&E, elastic or other stains, it should be classified as venous invasion (V1/2) or lymphatic invasion (L1).
Similarly, if neural structures are identifiable, the lesion should be classified as perineural invasion (Pn1). The presence of tumour deposits does not
change the primary tumour T category, but changes the node status (N) to pN1c if all regional lymph nodes are negative on pathological examination.
H&E, hematoxilin and eosin; UICC, the Union for International Cancer Control; TNM, tumour, node, metastasis.
Reprinted from [16], with permission from John Wiley & Sons, Inc.
Endoscopic rectal ultrasound (ERUS) may define treatment for to select patients for the respective preoperative management and
the earliest tumours. T1 tumours appropriate for transanal endo- to define the extent of surgery [III, A]. A standard proforma for
scopic microsurgery (TEM) can be selected by determining MRI and pathology ensures a comprehensive report. The version
whether a lesion is limited to the mucosa or submucosa (sm) [15]. of TNM staging used by the histopathologist and the MDT
ERUS offers less value in locally advanced rectal cancer (LARC). should be documented, acknowledged by all members of the
Pelvic magnetic resonance imaging (MRI) is the most accurate MDT and regularly updated. The Union for International Cancer
test to define locoregional clinical staging. By detecting extra- Control (UICC) TNM staging classification (8th edition) is
mural vascular invasion (EMVI), and determining the T substage shown in Table 2 [16].
and distance to the circumferential resection margin (CRM), High-quality MRI allows further subclassification of cT3,
MRI can also predict the risks of local recurrence and synchron- which is recommended as described in Table 3 [17, 18]. Stage
ous/metachronous distant metastases, and should be carried out grouping is shown in Table 4 [16].
iv24 | Glynne-Jones et al. Volume 28 | Supplement 4 | August 2017

by guest
on 05 February 2018
Table 3. Subclassification of T3 rectal cancer [18] Table 4. Stage grouping of colon and rectal cancer [16]
Depth of invasion beyond TNM Pathological Classification
the muscularis propria (in mm)
The pT and pN categories correspond to the T and N categories.
T3aa <1 pN0 Histological examination of a regional lymphadenectomy
T3b 1–5 specimen will ordinarily include 12 or more lymph nodes.
T3c 6–15 If the lymph nodes are negative, but the number ordinarily
T3d > 15 examined is not met, classify as pN0.
Stage
a
This sub-classification based upon an evaluation using MRI before Stage 0 Tis N0 M0
treatment decision is clinically valuable, and is used in these recom- Stage I T1, T2 N0 M0
mendations. It can be used also in the histopathological classification Stage II T3, T4 N0 M0
but is not validated and not incorporated in any of the TNM versions Stage IIA T3 N0 M0
(5–7). Stage IIB T4a N0 M0
MRI, magnetic resonance imaging; TNM, tumour, node, metastasis. Stage IIC T4b N0 M0
Reprinted from [18] with permission from Springer. Stage III Any T N1, N2 M0
Stage IIIA T1, T2 N1 M0
T1 N2a M0
Stage IIIB T1, T2 N2b M0
Meta-analyses and population data show clinical nodal staging T2, T3 N2a M0
is unreliable even using ERUS, CT and MRI combined. The use T3, T4a N1 M0
of node size > 10 mm as a criterion for node-positive disease has Stage IIIC T3, T4a N2b M0
been shown to be inaccurate. Irregular border and heterogeneous T4a N2a M0
signal provide more relevant additional information [19]. T4b N1, N2 M0
Nomograms have been suggested as a predictor of lymph node Stage IV Any T Any N M1
involvement but have yet to be validated [20]. The assessment of Stage IVA Any T Any N M1a
the relationship between tumour and mesorectal fascia (MRF) is Stage IVB Any T Any N M1b
more crucial to decision-making than lymph node status. Stage IVC Any T Any N M1c
PET-CT should not be used routinely for initial staging, but
can, in conjunction with liver MRI and contrast enhanced CT of TNM, tumour, node, metastasis.
the thorax, abdomen and pelvis be used to assess features at pres- Reprinted from [16], with permission from John Wiley & Sons, Inc.
entation associated with a high risk of metastases, e.g. extensive
EMVI on MRI (see above) or high levels of CEA. Its value for as-
sessment of primary tumour and nodal status is unproven, al-
though it may assist RT target delineation. Bone scan and brain
imaging should only be carried out if symptoms warrant. of vascular or lymphatic invasion, and an invasion depth of
> 1000 micrometres [II, A] [25].
For mesorectal resections, histopathological examination
should include a photographic record of the surgical specimen
Histopathology and assessment of total mesorectal excision (TME) quality [III,
T1 tumours can be classified according to Haggitt’s subclassifica- B] [26–28], which is a strong quality control measure (Figure 3).
tion if the cancer is pedunculated and according to the Kudo/ The classification has three grades based on the completeness of
Kikuchi sm-system if in a sessile adenoma [21, 22]. The two sys- the removal of the mesorectum and/or plane of surgical excision
tems overlap. If sessile, the level of infiltration into the sm and the (Table 5) and impacts on both local recurrence and survival.
width of invasion compared with the width of the cancer should Along with the involved CRM rate (i.e. 1 mm), TME quality
be assessed [23]. If pedunculated, the grade, lymphovascular in- represents a surrogate parameter for good oncological outcomes
vasion (LVI) and presence of budding predict the risk of lymph [29]. More advanced T-stage, tumour distance from the anal
node metastases, enable a risk/benefit assessment of the require- verge < 8 cm, more advanced age and low surgical case volume
ment for further surgery and define the method of excision [III, have been independently associated with moderate or poor TME
B] [24]. These specimens should be pinned-out on cork before quality [30].
pathology assessment to facilitate this subclassification. At least 12 regional lymph nodes should be examined.
Endoscopic resection for small tumours/polyps can be useful Proximal, distal and circumferential margins should be docu-
for both diagnosis and treatment, but en bloc resection is recom- mented in millimetres (separately for tumour and involved
mended for accurate assessment of invasion in the resection mar- lymph nodes). A proforma report such as the one by the Royal
gin and the deepest area [II, B]. Piecemeal resection makes the College of Pathologists is recommended [IV, B] [31].
specimen impossible to assess for the above and should be Uncertainties in the interpretation of CRM and the residual (R)
avoided. tumour classification (and the distinction pT4/R1) according to
However, radical surgery and removal of lymph nodes is rec- the TNM version can cause confusion. An expanded classification
ommended for high-risk pathological features according has been suggested [32]. Extranodal extension (ENE) of nodal
to Japanese guidelines, i.e. poorly differentiated with evidence metastases, EMVI, perineural invasion (PNI) and tumour

by guest
on 05 February 2018
Figure 2. Rectal cancer staging.

CEA, carcinoembryonic antigen; CT, computed tomography; DRE, digital rectal examination; EMVI, extramural vascular invasion; ERUS, endor-
ectal ultrasound; MRI, magnetic resonance imaging; PET-CT, positron emission tomography computed tomography; TEM, transanal endo-
scopic microsurgery.
A B C
Figure 3. The definitions for defining quality of mesorectal excision [28].

(A) A complete mesorectal excision—shows good bulk of mesorectum with a smooth surface and no defects. (B) A nearly complete meso-
rectal excision shows good bulk of mesorectum, but some defects or irregularities in the surface (arrowed) are present. (C) An incomplete
mesorectal excision demonstrating a deep defect on the mesorectum below the peritoneal reflection, which allows visualisation of the mus-
cularis propria (arrowed).

by guest
on 05 February 2018
Table 5. Grading of quality and completeness of the mesorectum in a total mesorectal excision specimen according to the plane of surgical excision [27]
Mesorectal plane (good plane of Intact mesorectum with only minor irregularities of a smooth mesorectal surface; no defect deeper than
surgery achieved) 5 mm; no coning; and smooth circumferential resection margin on slicing
Intramesorectal plane (moderate Moderate bulk to mesorectum, with irregularities of the mesorectal surface; moderate distal coning; muscularis
plane of surgery achieved) propria not visible with the exception of levator insertion; and moderate irregularities of circumferential
resection margin
Muscularis propria plane (poor Little bulk to mesorectum with defects down onto muscularis propria; very irregular circumferential resection
plane of surgery achieved) margin; or both
The specimen is examined as a whole (fresh) and as cross-sectional slices (fixed) to make an adequate interpretation. A TME specimen ideally should
have a smooth surface, without incisions, defects or cracks, as an indication of successful surgical excision of all mesorectal tissue. ‘Coning’ represents the
tendency for the surgeon to cut inwards towards the central tube of the rectum during distal dissection, rather than staying outside the visceral meso-
rectal fascia. The specimen then shows a tapered, conical appearance representing suboptimal surgical quality.
TME, total mesorectal excision.
Reprinted from [27] with permission from Elsevier.
budding should also be evaluated [III, A]. Analysis of the rectal In rare situations, local excision can be an option in patients
‘doughnut’ is not required [IV, C] [33]. with a cT1 tumour or in elderly or fragile patients. TEM is then
the procedure of choice.
In selecting laparoscopic or open surgery, the surgeon should
take into account his/her experience with the technique, the stage
Management of local/locoregional disease and location of the cancer and patient factors such as obesity and
previous open abdominal surgery. In the case of low rectal tu-
Risk-adapted treatment mours, transanal TME (TaTME) may facilitate pelvic and distal
mesorectal dissection, but standardisation and assessment of the
Very early cT1N0, with low grade (G1/G2). Local excisional pro- technique are necessary [38].
cedures such as TEM are appropriate as a single modality for early Robotic-assisted rectal cancer surgery provides some technical
cancers (cT1N0 without adverse features like G3, V1, L1) [III, A] advantages for surgeons compared with conventional laparos-
[34, 35]. Only patients with cT1N0 should be considered for such copy, but is still under evaluation [39]. If an abdominoperineal
treatment [36], although TEM for more advanced T-stage may be excision is planned and the tumour extends into the levators, a
appropriate for patients at high surgical risk after discussion with cylindrical specimen should be achieved, avoiding a ‘waist’ effect
the patient. and minimising the risk of a positive CRM and/or an R1/2 resec-
TEM permits more accurate en bloc, full-thickness local exci- tion [40]. Selection of patients suitable for extralevator abdomi-
sion of rectal tumours than local excision, and can provide simi- noperineal excision is recommended using MRI [41].
lar oncological results in pT1sm1 (clinical cN0) rectal cancers In Japan, lateral node dissection (LND) is practised if the tu-
compared with results achieved by TME, without compromising mour is sited below the peritoneal reflection to reduce the risk of
anorectal function. pelvic recurrence and improve overall survival (OS). Lateral pel-
Local recurrence after local excision often occurs in the tumour vic nodes are often invaded if multiple mesorectal nodes are
bed. If there is an unfavourable pTNM assessment following local involved [42]. LND is rarely practised in Europe, unless involve-
excision, the value of adjuvant CRT in preventing local recurrence ment is suspected on imaging with enlarged lateral nodes persist-
is unproven, and TME should remain the standard salvage option. ing following chemoradiotherapy (CRT).
Local RT (brachytherapy or contact therapy—Papillon tech- For cT2 tumours < 4 cm, local excision after preoperative RT/
nique) may also be used as an alternative to local surgery [37], CRT has been considered as alternative management to abdom-
alone or combined with CRT [III, C]. inal surgery [43–45], with minimal adverse impact on anorectal
function 1 year after surgery [45]. More mature data from other
studies suggests some compromise to function [46]. This strategy
Early rectal cancer not suitable for local excision [cT1–cT2; cT3a/ is not routinely recommended outside clinical trials, except for
b if middle or high, N0 (or also cN1 if high), MRF clear, no elderly, fragile patients at high surgical risk [47, 48].
EMVI]. More advanced tumours up to and including cT2c/T3a/b These early, favourable cases, which are not suitable for local
should be treated by radical TME surgery because of higher risks excision, i.e. cT1-2 but with adverse pathological features (e.g.
of recurrence and the higher risk of mesorectal lymph node in- G3, V1, L1), and some cT3a/b without clear involvement of MRF
volvement [36]. The standard of care for surgery is TME, imply- (MRF-) according to MRI, when located above the levators, may
ing that all of the mesorectal fat, including all lymph nodes, be appropriate for surgery alone with TME [II, A], as the risk of
should be meticulously excised [III, A]. A partial mesorectal exci- local failure is very low. Although not prospectively assessed,
sion with a distal margin of at least 5 cm of mesorectum can be EMVI on MRI, even in the case of cT3a/b tumours, confers a
considered in high rectal cancer. higher risk of local and distant recurrence [49].

by guest
on 05 February 2018
For complete responders, in patients with cT2-3a tumours, a margin at risk, involved MRF or CRMþ) are imprecise, but MRI
‘watch-and-wait’ approach or local excision after neoadjuvant CRT can predict rectal cancers that are unlikely to be amenable to a
is feasible, but because of limited data about long-term outcomes curative resection without multivisceral resection, either because
should be implemented only in prospective protocols [IV, C] [50]. the tumour abuts or breaches the MRF or there is macroscopic
tumour outside the MRF with local extension to pelvic side wall
Intermediate/more locally advanced rectal cancers [cT3a/b (very and sacrum or in terms of tumour spread involvement into the
low, levators clear, MRF clear or (cT3a/b in mid- or high rectum, lateral compartment. In these circumstances, preoperative treat-
cN1-2 (not extranodal), no EMVI]. The routine delivery of ment is necessary to shrink the cancer back away from the threat-
preoperative RT, either CRT or short-course preoperative radio- ened margin i.e. the MRF/CRM. Without preoperative treatment
therapy (SCPRT), to all patients with imaging predicted or in the case of no response, surgery is likely to lead to either an
cNþ remains controversial in view of the poor accuracy if categor- R1 or an R2 resection. For such patients, CRT has been shown to
ised by nodal size alone [19], and the lack of prognostic relevance significantly increase the chance of performing an R0 resection
of the preoperative MRI assessment of involved lymph nodes compared with RT alone [60].
on the risk of local recurrence. This is particularly valid because Treatment recommendations for rectal cancer are summarised
data suggest a low risk of local recurrence if the surgeon routinely in Figure 4. Recommended treatment options for primary rectal
carries out good-quality TME and removes the mesorectal cancer without distant metastases are summarised in Table 6.
nodes en bloc [27, 51]. However, it is the responsibility of the sur-
geon to demonstrate that consistent, good-quality TME is being
achieved. Risk of recurrence according to postoperative
histology
Locally advanced rectal cancers (>cT3b, and EMVIþ). For pa- Historical studies prior to TME suggest that the postoperative
tients with LARC, treatment decisions regarding neoadjuvant histopathological features, which have an impact on the risk of local
therapy should be based on preoperative, MRI-predicted CRM recurrence, include: pathological TNM stage, T substage (Table 3),
( 1 mm), EMVI and more advanced T3 substages (T3c/T3d), CRM status, the number/proportion of involved lymph nodes,
which define the risk of both local recurrence and/or synchron- extracapsular extension, extranodal deposits, tumour differenti-
ous and subsequent metastatic disease [52, 53]. MRI also allows ation, LVI, EMVI and PNI. Hence, it is recommended that patholo-
risk stratification in terms of the predicted required extent of sur- gists review MRI scan reports when assessing EMVI status [61].
gery [37], and the achievement of a clear CRM (>1 mm). Histologically involved nodes have in the past been associated
For resectable cancers, where there is no indication on MRI with a high risk of local recurrence. However, the risk of local re-
that surgery is likely to be associated with either an R2 or an R1 currence is reduced if the quality of the mesorectal excision is
resection, standard TME should achieve a curative resection, good (i.e. with a complete, smooth mesorectum with no defects
and downstaging/downsizing is not necessary to achieve this. and no coning), ensuring removal of all mesorectal lymph nodes.
The use of CRT or SCPRT aims to reduce local recurrence. No
differences in oncological outcomes between CRT and SCPRT
were reported in two prospective studies offering preoperative Selection between short-course preoperative
therapy in unselected clinically determined T3/T4 or node- radiotherapy and long-course chemoradiotherapy
positive rectal cancer patients [54, 55]. The latter trial showed Two different schedules of preoperative therapy are standards of
that CRT had significantly higher adverse events compared with care:
SCPRT, with no statistically significant differences in postoper-
• SCPRT with a 25 Gy total dose at 5 Gy/fraction during 1
ative complications [56]. Two phase III trials showed that
week, followed by immediate surgery (< 10 days from the
SCPRT with delayed surgery is a useful alternative to conven-
first radiation fraction) [I, A]; SCPRT with delayed surgery is
tional short-course RT with immediate surgery [57, 58], which
also a useful alternative to conventional short-course RT,
is associated with significantly lower postoperative complica-
with immediate surgery offering similar oncological out-
tions [57].
comes and lower postoperative complications [57].
Previous recommendations aimed to reduce the overall risk
• CRT with a recommended dose of 45–50 Gy in 25–28 frac-
of an involved CRM to < 3% and local recurrence to (preferably)
tions; a boost with a further 5.4 Gy in 3 fractions can be con-
< 5% in the population in whom curative treatment is intended
sidered for preoperative RT if the CRM is threatened, and for
[59]. Evidence from the UK CR07 trial suggests that, without RT,
postoperative RT routinely with 5.4–9.0 Gy in 3–5 fractions
a local recurrence rate of 5% (27/543) can be achieved if a com-
according to CRM [I, A].
plete mesorectal excision is carried out with a negative CRM [27].
MDTs and surgeons are, therefore, required to audit their local It is not possible to give a rigid definition of which T and N
recurrence rates. There are recognised long-term adverse conse- sub-stages require SCPRT or CRT. The selection of preoperative
quences of surgery and RT. Symptoms such as chronic pain, fae- approach in LARC is based more regarding the risk of a CRMþ at
cal incontinence and sexual difficulties are reported in both sexes. TME surgery. If CRM and/or R0 resection status are predicted at
Good communication between surgeons, clinicians and patients risk, CRT is advised [60]. Otherwise, either SCPRT or CRT can
will optimise joint decision-making. be administered [I, A] [54, 55].
However, more recent evidence suggests that even if the pre-
Tumours with threatened resection margin. The terms dicted resection margin is at risk (CRM 1 mm, cT4 or fixed cT3
‘unresectable/borderline cancers’ (i.e. cT4, with the resection tumours), similar R0 resection rates and disease-free survival

by guest
on 05 February 2018
by guest
on 05 February 2018
Annals of Oncology
Volume 28 | Supplement 4 | August 2017

Figure 4. Rectal cancer treatment.
cCR, clinical complete response; CRT, chemoradiotherapy; EMVI, extramural vascular invasion; FOLFOX, leucovorin/fluorouracil/oxaliplatin; MRF, mesorectal fascia; MRI, magnetic resonance
imaging; RT, radiotherapy; SCPRT, short-course preoperative radiotherapy; TEM, transanal endoscopic microsurgery; TME, total mesorectal excision.
doi:10.1093/annonc/mdx224 | iv29
Clinical Practice Guidelines
Table 6. Recommended choice of treatment options within TNM risk category of primary rectal cancer without distant metastases
Risk group TN substage Possible therapeutic options Further considerations
Very early cT1 sm1 N0 (on ERUS and MRI) Local excision (TEM) Alternatively, in the case of adverse
If pT1 and no adverse features, TEM is sufficient features on pathology, TEM plus sal-
If adverse histopathology (sm 2, G3, V1, L1), vage (or adjuvant) CRT in periopera-
requires radical resection (TME) as standard tive high-risk patients (but
unproven benefit—with high risk
of local recurrence for pT2)
Early (Good) cT1-cT2; cT3a/b if middle or high, Surgery (TME) alone is standard. If unexpected For fragile, high-risk patients or those
N0 (or also cN1 if high), MRF poor prognostic signs on histopathology rejecting radical surgery (CRT with
clear, no EMVI (CRMþ, extranodal/N2), consider postopera- evaluation, local excision or if
tive CRT/CT (see postoperative recommen- achieving cCR, ‘watch-and-wait’,
dations in Table 7) organ preservation)
Intermediate cT3a/b very low, levators clear, MRF Surgery (TME) alone is a standard only if good- If CRT is given and cCR is achieved,
clear or cT3a/b in mid- or high quality mesorectal resection assured (and ‘watch-and-wait’ in high-risk pa-
rectum, cN1-2 (not extranodal), local recurrence 0.5% or, if not, preopera- tients for surgery may be
no EMVI tive SCPRT (55 Gy) or CRT followed by TME considered
Bad cT3c/d or very low localisation le- Preoperative SCPRT (55cGy) or CRT followed If CRT and cCR achieved, ‘watch-and-
vators threatened, MRF clear by TME, depending on need for regression wait’ in high-risk patients may be
cT3c/d mid-rectum, cN1–N2 considered
(extranodal), EMVIþ, limited
cT4aN0
Advanced (Ugly) cT3 with any MRF involved, any Preoperative CRT followed by surgery (TME Alternatively, 55 Gy alone with a
cT4a/b, lateral nodeþ and more extended surgery if needed due delay to surgery in fragile/elderly or
to tumour overgrowth), or preoperative in patients with severe comorbidity
SCPRT (55 Gy) plus FOLFOX and delay to who cannot tolerate CRT
surgery
Other factors besides T and N stages are relevant, such as EMVI, MRF involvement, distance from the anus and sphincters, size of mesorectum and patient
characteristics. Patient preferences are also important.
cCR, clinical complete response; CRM, circumferential resection margin; CRT, chemoradiotherapy; CT, computed tomography; EMVI, extramural vascular
invasion; ERUS, endoscopic rectal ultrasound; FOLFOX, leucovorin/fluorouracil/oxaliplatin; MRF, mesorectal fascia; MRI, magnetic resonance imaging;
SCPRT, short-course preoperative radiotherapy; TEM, transanal endoscopic microsurgery; TME, total mesorectal excision; TNM, tumour, node, metastasis.
(DFS) are achieved for CRT or SCPRT followed by ChT with oxa- Radiotherapy field sizes
liplatin/leucovorin/fluorouracil prior to surgery [58].
It is beyond the scope of these guidelines to present a detailed rec-
Biological molecularly targeted agents have not been successfully
ommendation of field sizes for each T- and N-stage depending on
integrated into CRT. Several meta-analyses indicated that oxalipla-
the location within the rectum. Most current guidelines are based
tin added to CRT may slightly increase pathological complete re-
on a consensus of experts rather than being evidence-based. Widely
sponse (pCR) rates and DFS in selected patients, but also enhances
encompassing nodal regions will be more appropriate for patients
acute toxicity [62]. Given the contradictory results and lack of a
with advanced tumours for whom radical surgery is not intended,
clear long-term oncological benefit in the seven randomised trials
and smaller volumes for early cancers with the same plan.
[63–72] testing this combination so far, oxaliplatin as a radiosensi-
tiser is not currently recommended to be routinely added to
fluoropyrimidine-based CRT [I, D]. Continuous intravenous infu-
sions of 5-fluorouracil (5-FU) or oral capecitabine during CRT are Preoperative (neoadjuvant) chemotherapy
recommended rather than bolus 5-FU [I, A] [63, 73]. Strategies using induction ChT before/following CRT or SCPRT
In summary, preoperative RT or CRT reduces the rate of local and surgery are being investigated in multiple trials.
recurrence without improvement of OS for mid/low stage II/III NACT alone using a fluoropyrimidine and oxaliplatin or com-
rectal cancers [I, A], but is associated with significantly worse in- bined with targeted agents has been proposed instead of preopera-
testinal and sexual functions after surgery [I, A]. tive CRT in cT3 tumours not threatening the CRM and cT4
Upper rectal cancers (>12 cm from the anal verge) above the tumours in the mid- and upper- rectum, with the aim of promptly
peritoneal reflection do not benefit from preoperative SCPRT or treating potential micrometastases and individualising treatment
CRT and should be treated as colon cancer [I, A]. Patients with options [72]. After NACT, pCR is achieved in some 25% of early
cT4 tumours falling back into the pelvis might benefit from neoad- stage cases. However, limited long-term oncological outcome data
juvant CRT or neoadjuvant ChT (NACT) alone [IV, D]. are available for more advanced stages, particularly with CRM

by guest
on 05 February 2018
involvement; hence, NACT alone is not recommended for the Table 7. Potential indications for postoperative chemoradiotherapy if
treatment of localised, non-metastatic disease outside clinical preoperative chemoradiotherapy not given
trials.
Sufficient and necessary Insufficient and
unnecessary
Reassessment/response assessment after CRM 1 mm pT1/pT2
preoperative (chemo)radiotherapy pT4b pT3
pN2 extracapsular spread close to MRF CRM > 2 mm
Assessment of the primary tumour response. The standard meth-
Extranodal deposits (N1c) pT4a above peritoneal
ods of clinically re-assessing patients following preoperative ther- reflection
apy rely on clinical examination using DRE, proctoscopy, and re- pN2 if poor mesorectal quality/defects pN1
imaging by MRI. These findings direct appropriate surgical strat- If good quality smooth
egy, the type of operation intended and the possibility of choos- intact mesorectum
ing a ‘watch-and-wait’ strategy.
Sufficient
pN2 low tumours within 4 cm of anal
Clinical complete response and a ‘watch-and-wait’ approach.
verge (risk of involved LPLN)
Following CRT or SCPRT, a clinical complete response (cCR)
Extensive extramural vascular invasion/
can be obtained in 10%–40% of patients when assessed after an perineural invasion close to MRF
interval of 12 weeks from the start of treatment. The likelihood of
achieving a cCR will depend partly on initial stage and currently Borderline sufficient
unknown molecular factors. cCR has only partial concordance pN2 in mid/upper rectum if good
with pCR [74]. Although not universally agreed, a cCR is defined mesorectal quality
as the absence of any palpable tumour or irregularity at DRE, no CRM 1–2 mm
Circumferential obstructing tumours
visible lesion at rectoscopy except a flat scar, telangiectasia or
whitening of the mucosa. These minimal criteria can be comple- CRM, circumferential resection margin; LPLN, lateral pelvic lymph node;
mented by absence of any residual tumour in the primary site and MRF, mesorectal fascia.
draining lymph nodes on imaging with MRI or ERUS, and nega-
tive biopsies from the scar. An initially raised CEA level which re-
turns to normal (< 5 ng/ml) after CRT is associated with an
increased likelihood of cCR and pCR, and hence supports the
opinion that a cCR has been achieved [IV, C]. inter-reader agreement has not been widely tested. mriTRG does
Dedicated centres have reported encouraging oncological and not correlate well with histopathological TRG, and there is dis-
functional outcome results for selected patients treated with cordance with RECIST (response evaluation criteria in solid
standardised CRT and a non-operative strategy. However, such tumours) tumour measurements [77].
patients have been subjected to rigorous and meticulous follow- The additional value of diffusion-weighted imaging, gadofosveset-
up, where MRI surveillance is available [75], and more frequent enhanced MRI or Apparent Diffusion Coefficient (ADC) measure-
than routine surveillance (see below) to ensure that surgical sal- ments have not been validated. The value of CT in assessment of local
vage is feasible and timely. response is relatively low. PET should not be routinely used as re-
Substantially more follow-up and larger numbers of patients sponse tool, although reduction in uptake can be quantified. The
treated within properly controlled prospective studies are needed relevance of these changes is not understood, and the extent of sur-
to validate the ‘watch-and-wait’ approach. Ongoing experiences gery should not be modified based on these findings [IV, D].
from large databases, such as the European Registry of Cancer In the case of persistent potential CRM involvement on imag-
Care (EURECCA) ‘International Watch & Wait Database’ www. ing following CRT, the consensus is that such patients should not
iwwd.org [50], will provide more information on its safety and undergo trial dissection but rather should be formally referred to
efficacy, and help to select appropriate patients. Patients should a MDT with experience in multivisceral resection, so the treated
be informed that the strategy remains unproven and that a small tumour can be removed en bloc [78]. Further ChT may be useful
increased oncological risk of uncontrolled pelvic and metastatic for some but is unproven [79]. The ultimate decision should be
disease exists, although the prognosis of patients with cCR is ex- made by the MDT. Difficulty in distinguishing between tumour
cellent even without surgery. A standardised protocol for inten- and fibrosis on restaging MRI may lead to potential discordance
sive surveillance is, therefore, recommended. between imaging and clinical findings at surgery.
Patients planned for surgery. In LARC, it is recommended to re-

evaluate the primary tumour/CRM with MRI after CRT prior to re- Distant metastases. Routine restaging of chest and abdomen after
section to achieve clear margins [41, 76], although re-imaging after neoadjuvant CRT is not recommended, but patients with more
CRT may both underestimate (poor discrimination between re- advanced cT4 cancers, threatened CRM and the presence of
sidual tumour and radiation-induced fibrosis), and overestimate EMVI should be re-staged within 3 months of original staging to
pathological response and T downstaging (tumour fragmentation). exclude metastatic disease prior to surgery. If metastatic disease is
Comparison of sequential MRIs provides MRI tumour regres- diagnosed, the patient should be re-evaluated by the MDT to de-
sion grading (mriTRG), which can discriminate/determine good termine appropriate management. Earlier stage tumours do not
and poor responders and predict survival outcomes, even though merit this practice unless clinical progression, including new

by guest
on 05 February 2018
Local recurrence
Curative approaches Palliative approaches
Preoperative CRT Re-irradiation in SCPRT

previously irradiated
patients to facilitate Re-irradiation in previously irradiated
a curative resection patients to palliate symptoms
Fluoropyrimidine and
oxaliplatin-based
ChT
Systemic palliative ChT to
downstage tumour
Surgical salvage by specialist team

Palliative surgical diversion procedures
in patients with reasonable life expectancy
Brachytherapy
Figure 5. Rectal cancer treatment of local recurrence.

ChT, chemotherapy; CRT, chemoradiotherapy; RT, radiotherapy; SCPRT, short-course preoperative radiotherapy.
symptoms which may be related to metastasis or dramatically of the RT to be fully expressed (but before tumour repopulation)
increased serum CEA, is observed. and for the acute reaction to settle so that surgery can be carried
out safely.
Pathological assessment of response. A pCR after CRT is associ- In the case of SCPRT in resectable cancers, where downstaging
ated with low rates of local and distant recurrence. A standardised is not required, ‘immediate’ surgery is recommended to take
definition of pCR is recommended [80]. Several tumour regres- place within 7 days from the end of neoadjuvant treatment, and
sion grades (TRGs) are in use, but interobserver agreement is lim- ideally within 0–3 days if the patient is 75 years (<10 days from
ited. pCR is classified by Mandard as TRG1 but by Dworak as the first radiation fraction) [I, A] [83, 86].
TRG5 [81, 82]. The optimal system (e.g. reproducibility and Longer intervals after SCPRT or CRT may enhance pCR rates
prognostic information) remains unclear; as a minimum, tu- (with unknown prognostic implications), but risks repopulation
mours should be graded as having either complete response, par- delays the use of postoperative systemic adjuvant ChT and risks
tial response or no response [IV, B]. Other dynamic subsequent metastases. In practice, there is a wide variation in the
histopathological features, i.e. amount of necrosis, regression of timing of surgery (4–12 weeks) due to patient/surgeon choice, re-
EMVI and downstaging of T and N stage, may also define covery from treatment and/or waiting list issues. Prospective tri-
outcomes. als have been carried out, one randomising between 6 and
12 weeks and the other between 7 and 11 weeks after CRT. The
Interval to surgery. The optimal timing of surgical resection of latter shows that the longer interval does not increase cCR and is
LARC after preoperative CRT or SCPRT remains controversial associated with higher surgical morbidity [84]. In contrast, pre-
and is addressed in trials [83–89]. The ideal interval requires a liminary results from the former suggest a significant increase in
balance between allowing sufficient time for the maximal effects pCR [85].

by guest
on 05 February 2018
Metastatic disease
Refer to ESMO consensus guidelines on

metastatic colorectal cancer
RAS and BRAF mutational status testing

should be carried out at the time of
diagnosis of metastatic disease, before
treatment with EGFR-targeted
monoclonal antibodies
Curative approaches Palliative approaches
SCPRT plus triplet ChT SCPRT

Systemic ChT (capecitabine, oxaliplatin and
bevacizumab)
Systemic chemotherapy
+/- targeted agents
Metastasectomy
Figure 6. Rectal cancer treatment of metastatic disease.

ChT, chemotherapy; EGFR, epidermal growth factor receptor; ESMO, European Society for Medical Oncology; SCPRT, short-course preopera-
tive radiotherapy.
Postoperative therapy subsequent local and distant recurrence, but there is no auto-
matic benefit from the use of adjuvant ChT.
Postoperative chemoradiotherapy. Preoperative CRT (i.e. 45– After surgery alone for rectal cancer, individual trials and
54 Gy, 1.8–2.0 Gy/fraction) or SCPRT has better outcomes meta-analyses indicate that there is a benefit for adjuvant 5-FU-
than postoperative CRT [87, 88]. Traditionally, postoperative based ChT in terms of DFS and OS [89, 90], but the magnitude of
CRT was administered for all patients with pT3-4 or benefit is smaller than for colon cancer. However, only few stud-
pNþ tumours, and combined with additional 4 months of ad- ies included in the meta-analysis-mandated TME surgery and/or
juvant bolus 5-FU ChT, but the routine use of CRT to reduce preoperative RT/CRT. In contrast, following SCPRT or CRT, in-
local recurrence can be questioned if a good-quality TME can dividual randomised trials [91–93] and meta-analyses [94] have
be assured [53]. not shown any benefit for 5-FU alone.
Postoperative CRT could be selectively used in patients with The addition of oxaliplatin to 5-FU may improve DFS [70, 95],
unexpected adverse histopathological features after primary sur- but results are not consistent [96] and there is no effect on OS. A
gery—e.g. positive CRM, perforation in the tumour area, incom- single randomised, phase II study suggests that adding oxaliplatin
plete mesorectal resection, extranodal deposits or nodal deposits to 5-FU/leucovorin in a modified leucovorin/fluorouracil/oxali-
with extracapsular spread close to the MRF, or in other cases with platin regimen (mFOLFOX6) improves relapse-free survival and
high risk of local recurrence if preoperative RT has not been given OS in high-risk rectal cancers without downstaging after pre-
[I, A] (see Table 7). operative 5-FU-based CRT [97], but this data should not be used
to recommend that all patients with ypNþ disease should receive
Postoperative chemotherapy. In colon cancer, adjuvant ChT has oxaliplatin-based postoperative ChT.
an established role for patients with ‘high-risk’ stage II and stage It also remains unclear whether the initial clinical (yc) or patho-
III disease. Patients with rectal cancer were specifically excluded logical (yp) stage should be used to determine the risk/benefit of
from most phase III adjuvant studies because of the potential tox- adjuvant treatment. In general, downgrading in T or N stage has
icity and confounding impact of RT or CRT. Postoperative been recognised more as a prognostic factor of favourable outcome
pathological staging (ypTNM) can predict a high risk of rather than predictive biomarker for adjuvant treatment.

by guest
on 05 February 2018
Surveillance and follow-up
Clinical assessment every 6 months for 2 years Long-term side effects of treatment should be monitored,
including assessment of lower genitourinary toxicities
Completion colonoscopy within the first year if not done at the Late effects/survivorship clinics for patients who
time of diagnostic work-up (e.g. if obstruction was present) have received pelvic RT
Regular serum CEA tests

(at least every 6 months in the first 3 years)
Minimum of two CT scans of the chest, abdomen and pelvis

in the first 3 years for distant metastases
History of colonoscopy with resection of colonic polyps every

5 years up to the age of 75 years
High-risk patients (CRM+) may merit more proactive

surveillance for local recurrence
Figure 7. Rectal cancer surveillance and follow-up.

CEA, carcinoembryonic antigen; CRM, circumferential resection margin; CT, computed tomography; RT, radiotherapy.
Summarising, it is reasonable to consider adjuvant ChT in rectal [III, A] prior to an attempt at resection [60]. Alternatively,
cancer patients after preoperative CRT/RT with yp stage III (and SCPRT followed by a fluoropyrimidine and oxaliplatin-based
‘high-risk’ yp stage II). The level of scientific evidence for sufficient ChT as used in the Polish-2 study can be also applied [58].
benefit is much lower than in colon cancer and is probably limited In patients previously irradiated, re-irradiation to lower doses
to DFS rather than to OS [II, C]. Hence, the decision on postopera- (with concomitant ChT is safe and can be used in selected pa-
tive ChT (fluoropyrimidine alone or combined with oxaliplatin) tients to facilitate a curative resection or per se to palliate symp-
should be risk-balanced, taking into account both the predicted toms [IV, C] [98]. If salvage surgery is not currently an option,
toxicity for a particular patient and the risk of relapse, and should systemic palliative ChT may be used to downstage the tumour, al-
be made jointly by the individual and the clinician. though reports of efficacy are rare [V, C] [99]. Palliative surgical
diversion procedures in patients with reasonable life expectancy
are also recommended. Brachytherapy can be an effective pallia-
Management of local recurrence tive option [100]. See Figure 5.
Local recurrence is less frequent with good-quality TME and pre-

operative RT/CRT. Recurrent pelvic tumour can cause severe pain,
often requiring opiate and non-opiate pain relief with an offensive
mucinous discharge and incontinence. Surgical salvage is compli- Management of advanced/metastatic disease
cated by the loss of the normal anatomical planes. Hence surgical Metastatic rectal cancer stages are covered already in the ESMO
salvage is recommended to be carried out by specialist teams. consensus guidelines on metastatic colorectal cancer [13], but, in
If RT has not already been given, patients should be considered principle, should reflect the goals of treatment: tumour- and
for standard-dose, preoperative CRT (45–50 Gy in 5–6 weeks) disease-related characteristics, patient-related factors (comorbidity,

by guest
on 05 February 2018
• A history and physical examination including DRE, full blood count, liver and renal function tests, serum CEA and CT scan of thorax and abdomen should
be carried out to define functional status and presence of metastases [III, A].
• Rigid rectoscopy and preoperative colonoscopy to the caecal pole are required, or, in the case of obstruction, virtual colonoscopy to exclude synchronous
colonic tumours. If no preoperative (virtual) colonoscopy was carried out, completion colonoscopy is recommended within 6 months of surgery [III, A].
• Pelvic MRI is the most accurate test to define locoregional clinical staging. By detecting EMVI, and determining the T substage and distance to the CRM, it
can predict the risks of synchronous/metachronous distant metastases, and should be carried out to select patients for the respective preoperative
management and to define the extent of surgery [III, A].
• At least 12 regional lymph nodes should be examined. Proximal, distal and circumferential margins should be documented in millimetres (separately for tu-
mour and involved lymph nodes). A proforma report such as the one by the Royal College of Pathologists is recommended [IV, B]. For mesorectal resec-
tions, histopathological examination should include a photographic record of the surgical specimen and assessment of TME quality [III, B], which is a
strong quality control measure.
• Local excisional procedures such as TEM are appropriate as a single modality for early cancers (cT1N0 without adverse features like G3, V1, L1) [III, A]. Local
RT (brachytherapy or contact therapy—Papillon technique) may also be used as an alternative to local surgery, alone or combined with CRT [III, C].
• More advanced tumours up to and including cT2c/T3a/b should be treated by radical TME surgery because of higher risks of recurrence and the higher
risk of mesorectal lymph node involvement. The standard of care for surgery is TME, implying that all of the mesorectal fat, including all lymph nodes,
should be meticulously excised [III, A].
• For patients with LARC, treatment decisions regarding neoadjuvant therapy should be based on preoperative, MRI-predicted CRM (1 mm), EMVI and
more advanced T3 substages (T3c/T3d), which define the risk of both local recurrence and/or synchronous and subsequent metastatic disease. For resect-
able cancers, where there is no indication on MRI that surgery is likely to be associated with either an R2 or an R1 resection, standard TME should achieve a
curative resection. The use of CRT or SCPRT aims to reduce local recurrence.
• The selection of preoperative approach in LARC is based more on the MDT decision regarding the risk of a CRMþ at TME surgery. If CRM and/or R0 resec-
tion status are predicted at risk, CRT is advised. Otherwise, either SCPRT or CRT can be administered [I, A]. Continuous intravenous infusions of 5-FU or oral
capecitabine during CRT are recommended rather than bolus 5-FU [I, A].
• Preoperative RT or CRT reduces the rate of local recurrence without improvement of OS for mid/low stage II/III rectal cancers [I, A], but is associated with
significantly worse intestinal and sexual functions after surgery [I, A].
• Upper rectal cancers (>12 cm from the anal verge) above the peritoneal reflection do not benefit from preoperative SCPRT or CRT and should be treated
as colon cancer [I, A].
• In the case of SCPRT in resectable cancers, where downstaging is not required, ‘immediate’ surgery is recommended to take place within 7 days from the
end of neoadjuvant treatment, and ideally within 0–3 days if the patient is 75 years (<10 days from the first radiation fraction) [I, A].
• Postoperative CRT could be selectively used in patients with unexpected adverse histopathological features after primary surgery—e.g. positive CRM, per-
foration in the tumour area, incomplete mesorectal resection, extranodal deposits or nodal deposits with extracapsular spread close to the MRF, or in other
cases with high risk of local recurrence if preoperative RT has not been given [I, A].
• During follow-up, clinical examination, completion colonoscopy and pelvic imaging using MRI and/or CT and for distant metastases CT of the chest, ab-
domen and pelvis are recommended [V, B].
A minimum provisional recommendation for average-risk patients is as follows:
– Clinical assessment: every 6 months for 2 years [V, D].
– A completion colonoscopy within the first year if not done at the time of diagnostic work-up (e.g. if obstruction was present) [I, A].
– History and colonoscopy with resection of colonic polyps every 5 years up to the age of 75 years [I, B].
– It is reasonable to offer a minimum of two CTs of the chest, abdomen and pelvis in the first 3 years and regular serum CEA tests (at least every 6 months
in the first 3 years).
• High-risk patients (CRMþ) may merit more proactive surveillance for local recurrence.
CEA, carcinoembryonic antigen; CRM, circumferential resection margin; CRT, chemoradiotherapy; CT, computed tomography; DRE, digital rectal examin-
ation; EMVI, extramural vascular invasion; 5-FU, 5-fluorouracil; LARC, locally advanced rectal cancer; MDT, multidisciplinary team; MRF, mesorectal fascia;
MRI, magnetic resonance imaging; OS, overall survival; RT, radiotherapy; SCPRT, short-course preoperative radiotherapy; TEM, transanal endoscopic micro-
surgery; TME, total mesorectal excision.
socioeconomic factors and expectations of the patient), and This latter strategy palliates symptoms in 80% of patients and
treatment-related factors such as toxicity. See Figure 6. avoids a salvage stoma for selected patients [101]. If the patient
Whether the primary tumour remains in situ and untreated has a chance for cure (oligometastatic disease), the treatment
may impact on the treatment strategy. ChT alone may be insuffi- should aim for rapid local control with effective systemic ChT
cient in those cases, and local palliation of rectal symptoms with and appropriate sequence/timing of metastasectomy. Single-
RT may be required. SCPRT (if feasible) is preferred to CRT since institution series suggest that SCPRT can be safely combined
systemic ChT can start within 2 weeks from the start of treatment. with triplet ChT (capecitabine, oxaliplatin and bevacizumab) to

by guest
on 05 February 2018
Table 9. Levels of evidence and grades of recommendation (adapted from the Infectious Diseases Society of America-United States Public Health Service
Grading Systema)
Levels of evidence
I Evidence from at least one large randomised, controlled trial of good methodological quality (low potential for bias) or meta-analyses of well-con-
ducted randomised trials without heterogeneity
A Strong evidence for efficacy with a substantial clinical benefit, strongly recommended
B Strong or moderate evidence for efficacy but with a limited clinical benefit, generally recommended
C Insufficient evidence for efficacy or benefit does not outweigh the risk or the disadvantages (adverse events, costs, . . .), optional
D Moderate evidence against efficacy or for adverse outcome, generally not recommended
E Strong evidence against efficacy or for adverse outcome, never recommended
a
facilitate the resection of borderline resectable liver metastasis 105]. Isolated CEA monitoring is insufficiently sensitive [106].
and the primary tumour [102]. There are no randomised studies, Routine monitoring of CEA and CT imaging is only recom-
so the MDT should be responsible for critical decisions in pa- mended up to 5 years following surgery.
tients with potentially curable metastatic disease. Both rectal cancer surgery and the additional pre- or postopera-
tive (C)RT may result in late sequelae, which impact daily func-
tion. Long-term side effects of treatment should be monitored.
Personalised medicine These include assessment of lower genitourinary toxicities (e.g.
erectile dysfunction, dyspareunia and urinary incontinence).
There are no molecular markers in rectal cancers available that An increased risk of developing a second primary cancer follow-
can evaluate specific situations or treatments (e.g. whether a pa- ing RT for rectal cancer within or outside of the irradiated volume
tient needs preoperative treatment for a localised or locally may have been overestimated [107]. However, with better treat-
advanced rectal cancer, indicating that surgery will not be radi- ments, increasing numbers of patients are living with the long-term
cal). Similarly, there are no known markers that can predict re- consequences of surgery, ChT and RT—such as stomas, poor
sponse to RT or CRT. Rectal cancers with distant metastases mobility, and attendant co-morbidity (osteopaenia, malabsorption,
should be studied for RAS and BRAF mutational status and the endocrinology problems and cardiovascular disease). Surveillance
other requirements addressed in the ESMO consensus guidelines should address the social, financial and emotional aspects as well as
on metastatic colorectal cancer [13]. practical and functional consequences to maximise survivors’ long-
term well-being. Important components include guidelines for the
proactive detection of likely future effects and an educational pro-
Follow-up, long-term implications and gram (before and after treatment) to promote engagement with the
healthcare system and an appropriate and healthy lifestyle.
survivorship Evidence supports late effects/survivorship clinics for patients
Follow-up/surveillance with clinical examination, imaging and who have received pelvic RT.
colonoscopy aims to improve prognosis by early detection and A minimum provisional recommendation for average-risk pa-
salvage of local recurrence and metastases, and to prevent/detect tients is as follows:
second colorectal cancers. See Figure 7.
• Clinical assessment: every 6 months for 2 years [V, D].
Clinical examination and pelvic imaging using MRI and/or CT
• A completion colonoscopy within the first year if not done at
and for distant metastases CT of the chest, abdomen and pelvis are
the time of diagnostic work-up (e.g. if obstruction was present)
recommended [V, B]. Patients with rectal tumours (particularly
[I, A].
more advanced stages) have a higher risk of recurrence and benefit
• History and colonoscopy with resection of colonic polyps
more from follow-up [103], although <10% may have salvageable
every 5 years up to the age of 75 years [I, B].
recurrence. Routine use of PET-CT as surveillance is not recom-
• A minimum of two CTs of the chest, abdomen and pelvis in
mended, although when recurrence is diagnosed, PET-CT may be
the first 3 years and regular serum CEA tests (at least every
helpful for defining other unrecognised sites of disease.
6 months in the first 3 years).
CEA screening and CT monitoring increase the rate of surgical
resection of recurrence with curative intent, although the opti- High-risk patients (CRMþ) may merit more proactive surveil-
mum modality, intensity and frequency remain undefined [104, lance for local recurrence.

by guest
on 05 February 2018
Methodology 6. Balma~ na J, Balaguer F, Cervantes A, Arnold D. Familial risk-colorectal cancer:
ESMO Clinical Practice Guidelines. Ann Oncol 2013; 24 (Suppl 6): vi73–vi80.
These Clinical Practice Guidelines were developed in accordance 7. Cancer Genome Atlas Network. Comprehensive molecular character-
with the ESMO standard operating procedures for Clinical ization of human colon and rectal cancer. Nature 2012; 487: 330–337.
Practice Guidelines development http://www.esmo.org/ 8. Sadanandam A, Wang X, de Sousa E Melo F et al. Reconciliation of clas-
Guidelines/ESMO-Guidelines-Methodology. The relevant litera- sification systems defining molecular subtypes of colorectal cancer. Cell
Cycle 2014; 13: 353–357.
ture has been selected by the expert authors. A summary of key
9. Guinney J, Dienstmann R, Wang X et al. The consensus molecular sub-
recommendations is given in Table 8. Levels of evidence and types of colorectal cancer. Nat Med 2015; 21: 1350–1356.
grades of recommendation have been applied using the system 10. Song N, Pogue-Geile KL, Gavin PG et al. Clinical outcome from oxali-
shown in Table 9. Statements without grading were considered platin treatment in stage II/III colon cancer according to intrinsic sub-
justified standard clinical practice by the experts and the ESMO types: secondary analysis of NSABP C-07/NRG oncology randomized
Faculty. This manuscript has been subjected to an anonymous clinical trial. JAMA Oncol 2016; 2: 1162–1169.
11. National Cancer Action Team. The characteristics of an effective multi-
peer review process.
disciplinary team (MDT), 2010. www.ncin.org.uk/view?rid¼136 (28
May 2017, date last accessed).
12. Munro A, Brown M, Niblock P et al. Do multidisciplinary team (MDT)
Acknowledgements processes influence survival in patients with colorectal cancer? A
RG-J would like to acknowledge the advice of Dr Bruce Mcfarlane, population-based experience. BMC Cancer 2015; 15: 686.
13. Van Cutsem E, Cervantes A, Adam R et al. ESMO consensus guidelines
Dr Marcia Hall, CNS Angela Wheeler, and Dr Teresa Leverton. for the management of patients with metastatic colorectal cancer. Ann
LW would like to express his gratitude to Dr Jacek Krynski and Pr Oncol 2016; 27: 1386–1422.
Krzysztof Bujko for discussions on multimodality treatment of rec- 14. Papamichael D, Audisio RA, Glimelius B et al. Treatment of colorectal can-
tal cancer. AC is a member of CIBERONC, a research network of cer in older patients: International Society of Geriatric Oncology (SIOG)
the Spanish National Institute of Health Charles III. consensus recommendations 2013. Ann Oncol 2015; 26: 463–476.
15. Burdan F, Sudol-Szopinska I, Staroslawska E et al. Magnetic resonance
imaging and endorectal ultrasound for diagnosis of rectal lesions. Eur J
Med Res 2015; 20: 4.
Disclosure 16. Brierley JD, Gospodarowicz MK and Wittekind C. (eds). TNM
Classification of Malignant Tumours, 8th edition. Oxford: John Wiley
RG-J has received honoraria for lectures from Roche, Sanofi, & Sons, Inc. 2016.
Merck Serono and Servier and for attending advisory boards 17. Zinicola R, Pedrazzi G, Haboubi N, Nicholls RJ. The degree of extra-
from Eli Lilly, Roche, Servier, Eisai Amgen, Bristol-Myers mural spread of T3 rectal cancer: an appeal to the American Joint
Squibb and Mundipharma. He has received research funding for Committee on Cancer. Colorectal Dis 2017; 19: 8–15.
investigator-led phase II trials from Merck and Roche. LW has 18. Edge SB, Byrd DR, Compton CC (eds). AJCC Cancer Staging
Handbook, 7th edition. New York: Springer 2010.
received honoraria for lectures from Amgen, Merck Serono,
19. Brown G, Richards CJ, Bourne MW et al. Morphologic predictors of
Roche and Servier and for advisory boards from Halozyme and lymph node status in rectal cancer with use of high-spatial-resolution MR
Eisai. AC has received honoraria for lectures from Amgen, imaging with histopathologic comparison. Radiology 2003; 227: 371–377.
Roche, Merck Serono, Takeda and Servier and for advisory 20. Liu Y, Wang R, Ding Y et al. A predictive nomogram improved diag-
boards from Eli Lilly, Bayer, Roche, Merck Serono, Servier and nostic accuracy and interobserver agreement of perirectal lymph nodes
Amgen; he has received research funding from Merck Serono, Eli metastases in rectal cancer. Oncotarget 2016; 7: 14755–14764.
21. Haggitt RC, Glotzbach RE, Soffer EE, Wruble LD. Prognostic factors in
Lilly, Bayer, Genentech and Roche. DA has reported research
colorectal carcinomas arising in adenomas: implications for lesions
grants from Roche and Sanofi and has received honoraria for lec- removed by endoscopic polypectomy. Gastroenterology 1985; 89: 328–336.
tures from Roche, Merck Serono, Bayer and Servier and for ad- 22. Kikuchi R, Takano M, Takagi K et al. Management of early invasive
visory boards from Eli Lilly, Bayer, Roche, Merck Serono and colorectal cancer. Risk of recurrence and clinical guidelines. Dis Colon
Servier. ET, GB and CR have reported no potential conflicts of Rectum 1995; 38: 1286–1295.
interest. 23. Toh EW, Brown P, Morris E et al. Area of submucosal invasion and
width of invasion predicts lymph node metastasis in pT1 colorectal can-
cers. Dis Colon Rectum 2015; 58: 393–400.
24. Sigourakis G, Lanitis S, Gockel I et al. Transanal endoscopic microsur-
References gery for T1 and T2 rectal cancers: a meta-analysis and meta-regression
1. World Cancer Research Fund/American Institute for Cancer Research. analysis of outcomes. Am Surg 2011; 77: 761–772.
Continuous Update Project Report Summary. Food, Nutrition, 25. Ikematsu H, Yoda Y, Matsuda T et al. Long-term outcomes after resec-
Physical Activity, and the Prevention of Colorectal Cancer 2011. tion for submucosal invasive colorectal cancers. Gastroenterology 2013;
2. Wei EK, Giovannucci E, Wu K et al. Comparison of risk factors for 144: 551–559.
colon and rectal cancer. Int J Cancer 2004; 108: 433–442. 26. Hermanek P, Hermanek P, Hohenberger W et al. The pathological as-
3. Kirkegaard H, Johnsen NF, Christensen J et al. Association of adherence sessment of mesorectal excision: implications for further treatment and
to lifestyle recommendations and risk of colorectal cancer: a prospective quality management. Int J Colorectal Dis 2003; 18: 335–341.
Danish cohort study. BMJ 2010; 341: c5504. 27. Quirke P, Steele R, Monson J et al. Effect of the plane of surgery
4. Aleksandrova K, Pischon T, Jenab M et al. Combined impact of healthy achieved on local recurrence in patients with operable rectal cancer: a
lifestyle factors on colorectal cancer: a large European cohort study. prospective study using data from the MRC CR07 and NCIC-CTG
BMC Med 2014; 12: 168. CO16 randomised clinical trial. Lancet 2009; 373: 821–828.
5. Murphy N, Norat T, Ferrari P et al. Dietary fibre intake and risks of can- 28. Garcıa-Granero E, Faiz O, Mu~ noz E et al. Macroscopic assessment of
cers of the colon and rectum in the European prospective investigation mesorectal excision in rectal cancer: a useful tool for improving quality
into cancer and nutrition (EPIC). PLoS One 2012; 7: e39361. control in a multidisciplinary team. Cancer 2009; 115: 3400–3411.

by guest
on 05 February 2018
29. Kreis ME, Ruppert R, Ptok H et al. Preoperative magnetic resonance 50. Van de Valk M. The International Watch and Wait database
imaging to select patients with rectal cancer for neoadjuvant (IWWD) for rectal cancer: an update. J Clin Oncol 2017; 35 (Suppl 4S):
chemoradiation–interim analysis of the German OCUM Trial abstr 521.
(NCT01325649). J Gastrointest Surg 2016; 20: 25–32. 51. Taylor FG, Quirke P, Heald RJ et al. Preoperative high-resolution mag-
30. Garlipp B, Ptok H, Schmidt U et al. Factors influencing the quality of netic resonance imaging can identify good prognosis stage I, II, and III
total mesorectal excision. Br J Surg 2012; 99: 714–720. rectal cancer best managed by surgery alone: a prospective, multicenter,
31. Quirke P, Williams GT. Minimum Dataset for Colorectal Cancer European study that recruited consecutive patients with rectal cancer.
Histopathology Reports. London: Royal College of Pathologists 1998. Ann Surg 2011; 253: 711–719.
32. Wittekind C, Compton C, Quirke P et al. A uniform residual tumor (R) 52. Hunter CJ, Garant A, Vuong T et al. Adverse features on rectal
classification: integration of the R classification and the circumferential MRI identify a high-risk group that may benefit from more intensive
margin status. Cancer 2009; 115: 3483–3488. preoperative staging and treatment. Ann Surg Oncol 2012; 19:
33. Ng CW, Lieske B, Tan KK. Routine histological sampling of doughnuts 1199–1205.
post oncologic anterior resection is not necessary. Int J Colorectal Dis 53. Taylor FG, Quirke P, Heald RJ et al. Preoperative magnetic resonance
2014; 29: 843–845. imaging assessment of circumferential resection margin predicts
34. Bach SP, Hill J, Monson JR et al. A predictive model for local recurrence disease-free survival and local recurrence: 5-year follow-up results of
after transanal endoscopic microsurgery for rectal cancer. Br J Surg the MERCURY study. J Clin Oncol 2014; 32: 34–43.
2009; 96: 280–290. 54. Bujko K, Nowacki MP, Nasierowska-Guttmejer A et al. Long-term re-
35. Junginger T, Goenner U, Hitzler M et al. Long-term oncologic outcome sults of a randomized trial comparing preoperative short-course radio-
after transanal endoscopic microsurgery for rectal carcinoma. Dis therapy with preoperative conventionally fractionated chemoradiation
Colon Rectum 2016; 59: 8–15. for rectal cancer. Br J Surg 2006; 93: 1215–1223.
36. Stornes T, Wibe A, Nesbakken A et al. National early rectal cancer treat- 55. Ngan SY, Burmeister B, Fisher RJ et al. Randomized trial of short-
ment revisited. Dis Colon Rectum 2016; 59: 623–629. course radiotherapy versus long-course chemoradiation comparing
37. Gérard JP, Ortholan C, Benezery K et al. Contact X-ray therapy for rec- rates of local recurrence in patients with T3 rectal cancer: Trans-
tal cancer: experience in Centre Antoine-Lacassagne, Nice, 2002–2006. Tasman Radiation Oncology Group Trial 01.04. J Clin Oncol 2012; 31:
Int J Radiat Oncol Biol Phys 2008; 72: 665–670. 3827–3833.
38. Penna M, Hompes R, Arnold S et al. Transanal total mesorectal exci- 56. Ansari N, Solomon MJ, Fisher RJ et al. Acute adverse events and post-
sion: international registry results of the first 720 cases. Ann Surg 2017; operative complications in a randomized trial of preoperative short-
266: 111–117. course radiotherapy versus long-course chemoradiotherapy for T3
39. Sun Y, Xu H, Li Z et al. Robotic versus laparoscopic low anterior resec- adenocarcinoma of the rectum: Trans-Tasman Radiation Oncology
tion for rectal cancer: a meta-analysis. World J Surg Oncol 2016; 14: 61. Group Trial (TROG 01.04). Ann Surg 2017; 265: 882–888.
40. Holm T, Ljung A, H€aggmark T et al. Extended abdominoperineal resec- 57. Erlandsson J, Holm T, Pettersson D et al. Optimal fractionation of pre-
tion with gluteus maximus flap reconstruction of the pelvic floor for operative radiotherapy and timing to surgery for rectal cancer
rectal cancer. Br J Surg 2007; 94: 232–238. (Stockholm III): a multicentre, randomised, non-blinded, phase 3, non-
41. Battersby NJ, How P, Moran B et al. Prospective validation of a low rec- inferiority trial. Lancet Oncol 2017; 18: 336–346.
tal cancer magnetic resonance imaging staging system and development 58. Bujko K, Wyrwicz L, Rutkowski A et al. Long-course oxaliplatin-based
of a local recurrence risk stratification model: the MERCURY II study. preoperative chemoradiation versus 5 5 Gy and consolidation
Ann Surg 2016; 263: 751–760. chemotherapy for cT4 or fixed cT3 rectal cancer: results of a random-
42. Ishihara S, Kanemitsu Y, Murono K et al. Oncological benefit of lateral ized phase III study. Ann Oncol 2016; 27: 834–842.
pelvic lymph node dissection for rectal cancer treated without preopera- 59. Glimelius B, Tiret E, Cervantes A et al. Rectal cancer: ESMO Clinical
tive chemoradiotherapy: a multicenter retrospective study using pro- Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol
pensity score analysis. Int J Colorectal Dis 2016; 31: 1315–1321. 2013; 24 (Suppl 6): vi81–vi88.
43. Lezoche E, Baldarelli M, Lezoche G et al. Randomized clinical trial of 60. Braendengen M, Tveit KM, Berglund A et al. Randomized phase III
endoluminal locoregional resection versus laparoscopic total mesorectal study comparing preoperative radiotherapy with chemoradiotherapy in
excision for T2 rectal cancer after neoadjuvant therapy. Br J Surg 2012; nonresectable rectal cancer. J Clin Oncol 2008; 26: 3687–3694.
99: 1211–1218. 61. Chand M, Siddiqui MR, Swift I, Brown G. Systematic review of prog-
44. Bujko K, Richter P, Smith FM et al. Preoperative radiotherapy and local nostic importance of extramural venous invasion in rectal cancer.
excision of rectal cancer with immediate radical re-operation for poor World J Gastroenterol 2016; 22: 1721–1726.
responders: a prospective multicentre study. Radiother Oncol 2013; 62. Yang YJ, Cao L, Li ZW et al. Fluorouracil-based neoadjuvant chemora-
106: 198–205. diotherapy with or without oxaliplatin for treatment of locally advanced
45. Garcia-Aguilar J, Renfro LA, Chow OS et al. Organ preservation for rectal cancer: an updated systematic review and meta-analysis.
clinical T2N0 distal rectal cancer using neoadjuvant chemoradiotherapy Oncotarget 2016; 7: 45513–45524.
and local excision (ACOSOG Z6041): results of an open-label, single- 63. O’Connell MJ, Colangelo LH, Beart RW et al. Capecitabine and oxali-
arm, multi-institutional, phase 2 trial. Lancet Oncol 2015; 16: platin in the preoperative multimodality treatment of rectal cancer: sur-
1537–1546. gical end points from National Surgical Adjuvant Breast and Bowel
46. Gornicki A, Richter P, Polkowski W et al. Anorectal and sexual func- Project trial R-04. J Clin Oncol 2014; 32: 1927–1934.
tions after preoperative radiotherapy and full-thickness local excision of 64. Allegra CJ, Yothers G, O’Connell MJ et al. Neoadjuvant 5-FU or capeci-
rectal cancer. Eur J Surg Oncol 2014; 40: 723–730. tabine plus radiation with or without oxaliplatin in rectal cancer pa-
47. Morino M, Risio M, Bach S et al. Early rectal cancer: the European tients: a phase III randomized clinical trial. J Natl Cancer Inst 2015; 107:
Association for Endoscopic Surgery (EAES) clinical consensus confer- pii: djv248.
ence. Surg Endosc 2015; 29: 755–773. 65. Aschele C, Cionini L, Lonardi S et al. Primary tumor response to pre-
48. Smith FM, Rao C, Oliva Perez R et al. Avoiding radical surgery im- operative chemoradiation with or without oxaliplatin in locally
proves early survival in elderly patients with rectal cancer, demonstrat- advanced rectal cancer: pathologic results of the STAR-01 randomized
ing complete clinical response after neoadjuvant therapy: results of a phase III trial. J Clin Oncol 2011; 29: 2773–2780.
decision-analytic model. Dis Colon Rectum 2015; 58: 159–171. 66. Aschele C, Lonardi S, Cionini L et al. Final results of STAR-01: a
49. Bugg WG, Andreou AK, Biswas D et al. The prognostic significance of randomized phase III trial comparing preoperative chemoradiation
MRI-detected extramural venous invasion in rectal carcinoma. Clin with or without oxaliplatin in locally advanced rectal cancer. J Clin
Radiol 2014; 69: 619–623. Oncol 2016; 34 (Suppl): abstr 3521.

by guest
on 05 February 2018
67. Gérard JP, Azria D, Gourgou-Bourgade S et al. Clinical outcome of the 85. Lefevre JH, Mineur L, Kotti S et al. Effect of interval (7 or 11 weeks) be-
ACCORD12/0405 PRODIGE 2 randomized trial in rectal cancer. J Clin tween neoadjuvant radiochemotherapy and surgery on complete patho-
Oncol 2012; 30: 4558–4565. logic response in rectal cancer: a multicenter, randomized, controlled
68. Francois E, Gourgou-Bourgade S, Azria D et al. ACCORD12/0405- trial (GRECCAR-6). J Clin Oncol 2016; 34: 3773–3780.
Prodige 2 phase III trial neoadjuvant treatment in rectal cancer: re- 86. van den Broek CB, Vermeer TA, Bastiaannet E et al. Impact of the inter-
sults after 5 years of follow-up. J Clin Oncol 2016; 34 (Suppl 4S): abstr val between short-course radiotherapy and surgery on outcomes of rec-
490. tal cancer patients. Eur J Cancer 2013; 49: 3131–3139.
69. Schmoll HJ, Haustermans K, Price TJ et al. Preoperative chemoradiother- 87. Sauer R, Becker H, Hohenberger W et al. Preoperative versus postoper-
apy and postoperative chemotherapy with capecitabine and oxaliplatin ative chemoradiotherapy for rectal cancer. N Engl J Med 2004; 351:
versus capecitabine alone in locally advanced rectal cancer: disease-free 1731–1740.
survival at interim analysis. J Clin Oncol 2014; 32 (Suppl): abstr 3501. 88. Sebag-Montefiore D, Stephens RJ, Steele R et al. Preoperative radiother-
70. Rödel C, Graeven U, Fietkau R et al. Oxaliplatin added to fluorouracil- apy versus selective postoperative chemoradiotherapy in patients with
based preoperative chemoradiotherapy and postoperative chemotherapy rectal cancer (MRC CR07 and NCIC-CTG C016): a multicentre, rando-
of locally advanced rectal cancer (the German CAO/ARO/AIO-04 study): mised trial. Lancet 2009; 373: 811–820.
final results of the multicentre, open-label, randomised, phase 3 trial. 89. Quasar Collaborative Group, Gray R, Barnwell J et al. Adjuvant chemo-
Lancet Oncol 2015; 16: 979–989. therapy versus observation in patients with colorectal cancer: a rando-
71. Jiao D, Zhang R, Gong Z et al. Fluorouracil-based preoperative chemo- mised study. Lancet 2007; 370: 2020–2029.
radiotherapy with or without oxaliplatin for stage II/III rectal cancer: a 90. Petersen SH, Harling H, Kirkeby LT et al. Postoperative adjuvant
3-year follow-up study. Chin J Cancer Res 2015; 27: 588–596. chemotherapy in rectal cancer operated for cure. Cochrane Database
72. Deng Y, Chi P, Lan P et al. Modified FOLFOX6 with or without radi- Syst Rev 2012; 3: CD004078.
ation versus fluorouracil and leucovorin with radiation in neoadjuvant 91. Bosset JF, Calais G, Mineur L et al. Fluorouracil-based adjuvant chemo-
treatment of locally advanced rectal cancer: initial results of the Chinese therapy after preoperative chemoradiotherapy in rectal cancer: long-
FOWARC multicenter, open-label, randomized three-arm phase III term results of the EORTC 22921 randomised study. Lancet Oncol
trial. J Clin Oncol 2016; 34: 3300–3307. 2014; 15: 184–190.
73. Hofheinz RD, Wenz F, Post S et al. Chemoradiotherapy with capecita- 92. Sainato A, Cernusco Luna Nunzia V, Valentini V et al. No benefit of ad-
bine versus fluorouracil for locally advanced rectal cancer: a rando- juvant fluorouracil leucovorin chemotherapy after neoadjuvant chemo-
mised, multicentre, non-inferiority, phase 3 trial. Lancet Oncol 2012; radiotherapy in locally advanced cancer of the rectum (LARC): long
13: 579–588. term results of a randomized trial (I-CNR-RT). Radiother Oncol 2014;
74. Glynne-Jones R, Hughes R. Critical appraisal of the ‘wait and see’ ap- 113: 223–229.
proach in rectal cancer for clinical complete responders after chemora- 93. Breugom AJ, van Gijn W, Muller EW et al. Adjuvant chemotherapy for
diation. Br J Surg 2012; 99: 897–909. rectal cancer patients treated with preoperative (chemo)radiotherapy
75. Martens MH, Maas M, Heijnen LA et al. Long-term outcome of an and total mesorectal excision: a Dutch Colorectal Cancer Group
organ preservation program after neoadjuvant treatment for rectal can- (DCCG) randomised phase III trial. Ann Oncol 2015; 26: 696–701.
cer. J Natl Cancer Inst 2016; 108: pii: djw171. 94. Breugom AJ, Swets M, Bosset JF et al. Adjuvant chemotherapy after pre-
76. van der Paardt MP, Zagers MB, Beets-Tan RG et al. Patients who operative (chemo)radiotherapy and surgery for patients with rectal can-
undergo preoperative chemoradiotherapy for locally advanced rectal cer: a systematic review and meta-analysis of individual patient data.
cancer restaged by using diagnostic MR imaging: a systematic review Lancet Oncol 2015; 16: 200–207.
and meta-analysis. Radiology 2013; 269: 101–112. 95. Zhao L, Liu R, Zhang Z et al. Oxaliplatin/fluorouracil-based adjuvant
77. Siddiqui MR, Bhoday J, Battersby NJ et al. Defining response to radiochemotherapy for locally advanced rectal cancer after neoadjuvant che-
therapy in rectal cancer using magnetic resonance imaging and histo- moradiotherapy and surgery: a systematic review and meta-analysis of
pathological scales. World J Gastroenterol 2016; 22: 8414–8434. randomized controlled trials. Colorectal Dis 2016; 18: 763–772.
78. Beyond TME Collaborative. Consensus statement on the multidiscip- 96. Glynne-Jones R, Counsell N, Quirke P et al. Chronicle: results of a
linary management of patients with recurrent and primary rectal cancer randomised phase III trial in locally advanced rectal cancer after neoad-
beyond total mesorectal excision planes. Br J Surg 2013; 100: E1–E33. juvant chemoradiation randomising postoperative adjuvant capecita-
79. Sclafani F, Brown G, Cunningham D et al. Systemic chemotherapy bine plus oxaliplatin (XELOX) versus control. Ann Oncol 2014; 25:
(CT) as salvage treatment for locally advanced rectal cancer (LARC) pa- 1356–1362.
tients (pts) who fail to respond to neoadjuvant chemoradiotherapy 97. Hong YS, Nam BH, Kim KP et al. Oxaliplatin, fluorouracil, and leuco-
(CRT). J Clin Oncol 2017; 35 (Suppl 4S): abstr 709. vorin versus fluorouracil and leucovorin as adjuvant chemotherapy for
80. Loughrey MB, Quirke P, Shepherd NA, Royal College of Pathologists. locally advanced rectal cancer after preoperative chemoradiotherapy
Standards and Datasets for Reporting Cancers. Dataset for Colorectal (ADORE): an open-label, multicentre, phase 2, randomised controlled
Cancer Histopathology Reports July 2014, 3rd edition. https://www. trial. Lancet Oncol 2014; 15: 1245–1253.
rcpath.org/asset/E94CE4A2-D722-44A7-84B9D6829413CFC/ (28 May 98. Guren MG, Undseth C, Rekstad BL et al. Reirradiation of locally recur-
2017, date last accessed). rent rectal cancer: a systematic review. Radiother Oncol 2014; 113:
81. Mandard AN, Dalibar DF, Mandard JC. Pathological assessment of tu- 151–157.
mour regression after preoperative chemoradiotherapy of oesophageal 99. Alberda WJ, Haberkorn BC, Morshuis WG et al. Response to chemo-
carcinoma. Clinicopathologic correlations. Cancer 1994; 73: 2680–2686. therapy in patients with recurrent rectal cancer in previously irradiated
82. Dworak O, Kelholz L. Hoffman A. Pathological features of rectal cancer area. Int J Colorectal Dis 2015; 30: 1075–1080.
after preoperative radiochemotherapy. Int J Colorectal Dis 1997; 12: 19–23. 100. Bishop AJ, Gupta S, Cunningham MG et al. Interstitial brachytherapy
83. Pettersson D, Holm T, Iversen H et al. Preoperative short-course radio- for the treatment of locally recurrent anorectal cancer. Ann Surg Oncol
therapy with delayed surgery in primary rectal cancer. Br J Surg 2012; 2015; 22 (Suppl 3): S596–S602.
99: 577–583. 101. Tyc-Szczepaniak D, Wyrwicz L, Kepka L et al. Palliative radiotherapy
84. Evans J, Bhoday J, Sizer B et al. Results of a prospective randomised and chemotherapy instead of surgery in symptomatic rectal cancer with
control 6 vs 12 trial: is greater tumour downstaging observed on post synchronous unresectable metastases: a phase II study. Ann Oncol
treatment MRI if surgery is delayed to 12-weeks versus 6-weeks after 2013; 24: 2829–2834.
completion of neoadjuvant chemoradiotherapy?. Ann Oncol 2016; 27 102. van Dijk TH, Tamas K, Beukema JC et al. Evaluation of short-course
(Suppl 6): vi149 (abstract 4520). radiotherapy followed by neoadjuvant bevacizumab, capecitabine, and

by guest
on 05 February 2018
oxaliplatin and subsequent radical surgical treatment in primary stage colorectal cancer: the FACS randomized clinical trial. JAMA 2014; 311:
IV rectal cancer. Ann Oncol 2013; 24: 1762–1769. 263–270.
103. Pugh SA, Shinkins B, Fuller A et al. Site and stage of colorectal cancer 106. Nicholson BD, Shinkins B, Pathiraja I et al. Blood CEA levels for detect-
influence the likelihood and distribution of disease recurrence and ing recurrent colorectal cancer. Cochrane Database Syst Rev 2015; 12:
post-recurrence survival: data from the FACS randomized controlled CD011134.
trial. Ann Surg 2016; 263: 1143–1147. 107. Martling A, Smedby KE, Birgisson H et al. Risk of second primary can-
104. Baca B, Beart RW, Jr, Etzioni DA. Surveillance after colorectal cancer cer in patients treated with radiotherapy for rectal cancer. Br J Surg
resection: a systematic review. Dis Colon Rectum 2011; 54: 2017; 104: 278–287.
1036–1048. 108. Dykewicz CA. Summary of the guidelines for preventing opportunistic
105. Primrose JN, Perera R, Gray A et al. FACS Trial Investigators. Effect of 3 infections among hematopoietic stem cell transplant recipients. Clin
to 5 years of scheduled CEA and CT follow-up to detect recurrence of Infect Dis 2001; 33: 139–144.

by guest
on 05 February 2018
special articles Annals of Oncology
Annals of Oncology 27: 1386–1422, 2016

ESMO consensus guidelines for the management

of patients with metastatic colorectal cancer
E. Van Cutsem1*, A. Cervantes2, R. Adam3, A. Sobrero4, J. H. Van Krieken5, D. Aderka6, E. Aranda
Aguilar7, A. Bardelli8, A. Benson9, G. Bodoky10, F. Ciardiello11, A. D’Hoore12, E. Diaz-Rubio13,
J.-Y. Douillard14, M. Ducreux15, A. Falcone16,17, A. Grothey18, T. Gruenberger19, K. Haustermans20,
V. Heinemann21, P. Hoff22, C.-H. Köhne23, R. Labianca24, P. Laurent-Puig25, B. Ma26, T. Maughan27,
K. Muro28, N. Normanno29, P. Österlund30,31, W. J. G. Oyen32, D. Papamichael33,
G. Pentheroudakis34, P. Pfeiffer35, T. J. Price36, C. Punt37, J. Ricke38, A. Roth39, R. Salazar40,
W. Scheithauer41, H. J. Schmoll42, J. Tabernero43, J. Taïeb25, S. Tejpar1, H. Wasan44,
T. Yoshino45, A. Zaanan25 & D. Arnold46
1
Digestive Oncology, University Hospitals Gasthuisberg Leuven and KU Leuven, Leuven, Belgium; 2Medical Oncology Department, INCLIVA University of Valencia,
Valencia, Spain; 3Hepato-Biliary Centre, Paul Brousse Hospital, Villejuif, France; 4Medical Oncology, IRCCS San Martino Hospital, Genova, Italy; 5Research Institute for
Oncology, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands; 6Division of Oncology, Sheba Medical Center, Tel Aviv University, Tel Aviv, Israel;
7
Medical Oncology Department, University Hospital Reina Sofia, Cordoba, Spain; 8School of Medicine, University of Turin, Turin, Italy; 9Division of Hematology/Oncology,
Northwestern Medical Group, Chicago, USA; 10Department of Oncology, St László Hospital, Budapest, Hungary; 11Division of Medical Oncology, Seconda Università di
Napoli, Naples, Italy; 12Abdominal Surgery, University Hospitals Gasthuisberg Leuven and KU Leuven, Leuven, Belgium; 13Medical Oncology Department, Hospital Clínico
San Carlos, Madrid, Spain; 14Medical Oncology, Institut de Cancérologie de l’Ouest (ICO), St Herblain; 15Department of Medical Oncology, Gustave Roussy, Université
Paris-Saclay, Villejuif, France; 16Department of Medical Oncology, University of Pisa, Pisa, Italy; 17Division of Medical Oncology, Department of Oncology, University Hospital
‘S. Chiara’, Istituto Toscano Tumori, Pisa, Italy; 18Division of Medical Oncology, Mayo Clinic, Rochester, USA; 19Department of Surgery I, Rudolfstiftung Hospital, Vienna,
Austria; 20Department of Radiation Oncology, University Hospitals Gasthuisberg and KU Leuven, Leuven, Belgium; 21Comprehensive Cancer Center, University Clinic
Munich, Munich, Germany; 22Instituto do Câncer do Estado de São Paulo, University of São Paulo, São Paulo, Brazil; 23Northwest German Cancer Center, University
Campus Klinikum Oldenburg, Oldenburg, Germany; 24Cancer Center, Ospedale Giovanni XXIII, Bergamo, Italy; 25Digestive Oncology Department, European Hospital
Georges Pompidou, Paris, France; 26Department of Clinical Oncology, Prince of Wales Hospital, State Key Laboratory in Oncology in South China, Chinese University of
Hong Kong, Shatin, Hong Kong; 27CRUK/MRC Oxford Institute for Radiation Oncology, Gray Laboratories, University of Oxford, Oxford, UK; 28Department of Clinical
Oncology and Outpatient Treatment Center, Aichi Cancer Center Hospital, Nagoya, Japan; 29Cell Biology and Biotherapy Unit, I.N.T. Fondazione G. Pascale, Napoli, Italy;
30
Helsinki University Central Hospital, Comprehensive Cancer Center, Helsinki, Finland; 31Department of Oncology, University of Helsinki, Helsinki, Finland; 32The Institute of
Cancer Research and The Royal Marsden Hospital, London, UK; 33Department of Medical Oncology, Bank of Cyprus Oncology Centre, Nicosia, Cyprus; 34Department of
Medical Oncology, University of Ioannina, Ioannina, Greece; 35Department of Oncology, Odense University Hospital, Odense, Denmark; 36Haematology and Medical
Oncology Unit, Queen Elizabeth Hospital, Woodville, Australia; 37Department of Medical Oncology, Academic Medical Centre, University of Amsterdam, Amsterdam, The
Netherlands; 38Department of Radiology and Nuclear Medicine, University Clinic Magdeburg, Magdeburg, Germany; 39Digestive Tumors Unit, Geneva University Hospitals
(HUG), Geneva, Switzerland; 40Catalan Institute of Oncology (ICO), Barcelona, Spain; 41Department of Internal Medicine I and Comprehensive Cancer Center, Medical
University of Vienna, Vienna, Austria; 42Department of Internal Medicine IV, University Clinic Halle, Martin-Luther-University Halle-Wittenberg, Halle, Germany; 43Medical
Oncology Department, Vall d’ Hebron University Hospital, Vall d’Hebron Institute of Oncology (V.H.I.O.), Barcelona, Spain; 44Department of Cancer Medicine,
Hammersmith Hospital, Imperial College Healthcare NHS Trust, London, UK; 45Department of Gastroenterology and Gastrointestinal Oncology, National Cancer Center
Hospital East, Chiba, Japan; 46Instituto CUF de Oncologia (ICO), Lisbon, Portugal
Received 31 March 2016; revised 27 May 2016; accepted 31 May 2016
Colorectal cancer (CRC) is one of the most common malignancies in Western countries. Over the last 20 years, and the
last decade in particular, the clinical outcome for patients with metastatic CRC (mCRC) has improved greatly due not only
to an increase in the number of patients being referred for and undergoing surgical resection of their localised metastatic
disease but also to a more strategic approach to the delivery of systemic therapy and an expansion in the use of ablative
techniques. This reflects the increase in the number of patients that are being managed within a multidisciplinary team
environment and specialist cancer centres, and the emergence over the same time period not only of improved imaging
techniques but also prognostic and predictive molecular markers. Treatment decisions for patients with mCRC must be
evidence-based. Thus, these ESMO consensus guidelines have been developed based on the current available evidence
*Correspondence to: Prof. Eric Van Cutsem, ESMO Guidelines Committee, ESMO
Head Office, Via L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland.
Downloaded from https://academic.oup.com/annonc/article-abstract/27/8/1386/1741549

by guest
on 05 February 2018
Annals of Oncology special articles
to provide a series of evidence-based recommendations to assist in the treatment and management of patients with
mCRC in this rapidly evolving treatment setting.
Key words: colorectal cancer, ESMO, consensus, clinical practice guidelines
introduction (ii) local and ablative treatment (LAT) [including surgery and
the management of patients with oligometastatic disease
Colorectal cancer (CRC) is the second most commonly diag- (OMD)];
nosed cancer in Europe and a leading cause of death both in (iii) the treatment of metastatic disease.
Europe and worldwide [1, 2]. In 2012, there were 447 000 new
cases of CRC in Europe with 215 000 deaths and worldwide, Each panel member was assigned to one of the above working
there were 1.4 million new cases with 694 000 deaths. Over the groups. Three consensus conference chairs (EVC, AC and DA)
last decade in particular, the clinical outcome for patients with were also appointed. Before the consensus conference, clinically
metastatic CRC (mCRC) has improved. Today, the median relevant questions were identified for each working group. Each
overall survival (OS) for patients with mCRC being treated both working group was responsible for reviewing relevant literature
in phase III trials and in large observational series or registries is in order to draft preliminary recommendations relating to each
∼30 months and more than double that of 20 years ago. of their assigned questions. No systematic literature search was
However, it is unclear which improvements and strategic undertaken. The experts in each group were invited to submit their
changes in the treatment and management of patients with recommendations in advance to structure the on-site discussions.
mCRC in recent years have been responsible for the improved During the conference, in parallel sessions, the three working
treatment outcomes for these patients. Factors which may have groups discussed and reached agreement on recommendations
contributed are: relating to each of their assigned questions. Recommendations
from each group were then presented to the entire panel of experts,
(i) changes in the clinical presentation of patients, before the where they were discussed and modified as required until consen-
commencement of treatment, due to closer follow-up after sus was reached.
resection of the primary tumour and earlier detection of An adapted version of the ‘Infectious Diseases Society of
metastatic disease; America-United States Public Health Service Grading System’
(ii) improvements in the efficacy of systemic therapies in terms was used (Table 1, [4]) to define the level of evidence and
of regimens used, sequence of administration, number of strength of each recommendation proposed by the group, as for all
lines of therapy administered and biomarker-based patient of the ESMO Consensus and ESMO Clinical Practice Guidelines,
selection; and are given in the text in square brackets. Statements made based
(iii) an increase in the number of patients being treated with a on expert opinion were also considered to be justified standard
view to facilitating resection of their metastases, offering an clinical practice by the experts and the ESMO faculty. These
increased number of patients the chance of cure and/or ESMO Consensus Guidelines follow on from those published in
durable relapse-free survival and, more recently, the utilisa- 2012 [3] and should be used to support the 2014 ESMO Clinical
tion of other ablative therapy techniques with the aim of Practice Guidelines [5].
achieving the same outcome;
(iv) implementation of ‘continuum of care’ treatment strategies
coupled with the early integration of optimal supportive molecular pathology and biomarkers
care measures. A clinical or biological suspicion that a patient may have
These ESMO Consensus Guidelines therefore aim to reflect the mCRC should always be confirmed by adequate radiological
diagnostic, therapeutic and strategic improvements which have imaging, and the histology of the primary tumour or metas-
contributed to the current ‘state-of-the-art’ treatment approaches tases, as appropriate, conducted before the commencement of
and to provide guidance for the comprehensive management of systemic therapy, as described previously [5]. Tissue samples
patients with mCRC going forward. will typically range from large tumour samples to smaller
biopsy/endoscopy samples. Whenever possible, any diagnostic
biopsy or tissue sampling procedure should aim to maximise
methodology
the number of samples collected (ideally n = 10 biopsies). In
In 2014, the ESMO Guidelines Committee decided to update addition to samples taken for embedding, additional frozen
the clinical recommendations for mCRC using a consensus con- material should be collected to provide the opportunity for
ference approach. An international panel of experts in the man- future ‘new’ tests to be conducted on frozen tissue if required. It
agement of patients with CRC, from a range of diagnostic and is also essential that all tissue and biopsy samples are handled
therapeutic disciplines, was convened in Zurich in December appropriately in order to facilitate meaningful and accurate
2014 to update the existing ESMO Consensus Guidelines for the molecular testing.
management of patients with colon and rectal cancer [3]. A set
of pre-formulated topics was prepared and three working tissue handling
groups convened in the areas of:
Standardisation of tissue processing for patients with mCRC
(i) molecular pathology and biomarkers; still remains a challenge. The time from tissue sampling to
Volume 27 | No. 8 | August 2016 doi:10.1093/annonc/mdw235 | 

by guest
on 05 February 2018
Table 1. Levels of evidence and grades of recommendation (adapted from the Infectious Diseases Society of America—United States Public Health
Coding Systema [4])
Levels of evidence
II Small randomised trials or large randomised trials with a suspicion of bias (low methodological quality) or meta-analyses of such trials or of trials
IV Retrospective cohort studies of case–control studies
C Insufficient evidence for efficacy or benefit does not outweigh the risk of the disadvantages (adverse events, costs, …) optional
a
By permission of the Infectious Diseases Society of America.
fixation should be minimised to only a few minutes if possible, sufficient quantity of neoplastic cells for the analysis [13]. This
to prevent any degradation of proteins and nucleic acids that is particularly crucial for DNA- or RNA-based biomarker
might occur during cold ischaemia [6, 7]. Fixation in 10% testing, such as RAS mutation analysis, because a low fraction of
neutral buffered formalin (4% formaldehyde solution), which is neoplastic cells can lead to dilution of mutant alleles and false-
widely available, is generally compatible with any procedure for negative results [14, 15]. To evaluate the tumour content of the
protein, RNA and/or DNA biomarker analysis. The fixation sample, it is recommended that the pathologist assesses a haema-
time should be between 6 and 48 h [8]. Longer or shorter fixa- toxylin and eosin-stained section of the paraffin block designated
tion times may adversely affect biomarker testing, while under- for DNA extraction and mutation analysis before DNA extrac-
fixation is also associated with poor tissue morphology [9]. tion. The minimum fraction of tumour versus non-tumour cells
Acidic fixatives (e.g. Bouin) are not recommended since they required will depend on the genotyping method. It has been
lead to the rapid degradation of nucleic acids [10]. Similarly, demonstrated that a tumour cell content of 30% or less might
accelerated fixation with heated formalin is discouraged as it lead to false-negative results when a technique with low sensitivity
degrades tissue morphology and affects the results of molecular such as Sanger sequencing is used for testing [16, 17]. A neoplas-
studies [11]. Biomarker analyses should be carried out within tic cell content of at least 50% is therefore recommended when
4–6 weeks of the sections being cut, as ageing of formalin-fixed, using a technique with low sensitivity. Sections of tissue with high
paraffin-embedded tissue sections causes the degradation of tumour content may be used directly. In samples with a low
both epitopes and DNA [12]. tumour cell content, and where feasible, suitable areas identified
by the pathologist may be scraped (manual macro-dissection)
recommendation 1: tissue handling. from the tissue slide(s) in order to enrich the tumour cell content.
Laser capture micro-dissection can also be used, but this technol-
• Fixation with 10% neutral buffered formalin (4% formalde-
ogy is not widely available, and requires the skills of a pathologist,
hyde) is recommended [V, A].
additional work and, therefore, high costs.
• Fixation time should be no less than 6 h, and no greater than
48 h in duration. In the case of microwave-enhanced fixation,
the quality of both nucleic acids and proteins must be verified recommendation 2: selection of specimens for biomarker testing.
[IV, A].
• The primary pathologist should review all available tumour
• Sections for biomarker testing should ideally be cut immedi-
specimens to select those that are most suitable for biomarker
ately before analysis [IV, A].
analyses [IV, A].
• Enrichment of samples by macro-dissection to maximise
selection of specimens for biomarker testing tumour cell content (>50%) before DNA extraction is recom-
The pathologist plays a central role in biomarker testing and can mended [III, A].
either perform the biomarker tests at his/her laboratory if it has
been accredited for biomarker testing, or send the tissue block
to an accredited reference laboratory for external testing. In both tissue selection for biomarker testing
instances, the primary pathologist should review the available Most patients undergo surgery of their primary tumour, although
material for each patient and choose the most appropriate block in some cases, only an endoscopic biopsy of the primary is
to be used for testing. The pathologist should also ensure that carried out. Thus, archival samples of primary tumour tissue are
the tissue block selected for biomarker analysis contains a usually available for biomarker testing for the majority of patients
 | Van Cutsem et al. Volume 27 | No. 8 | August 2016

by guest
on 05 February 2018
with advanced or mCRC. However, for ∼20% of patients who growth factor receptor (EGFR) monoclonal antibodies, cetuximab
present with metastatic disease, archival material from their and panitumumab, have shown that patients with mCRC, whose
primary tumour will not always be available. For these patients, tumours contain activating mutations in KRAS exon 2 (codons
biomarker testing is usually carried out using specimens obtained 12/13), do not derive a benefit from EGFR monoclonal antibody
from primary tumour biopsies or the metastatic tumour, for therapy [41–47]. Furthermore, recent evidence from the PRIME
example, from resected liver metastases or positive lymph nodes. study with panitumumab [48], from the CRYSTAL study with
For some patients, both the primary tumour and metastatic tissue cetuximab [49] and from other studies of EGFR monoclonal
specimens may be available for mutation testing. Indeed, a antibody therapies has shown that mutations other than those in
number of studies have addressed the concordance in KRAS KRAS exon 2 [i.e. exons 3 and 4 of KRAS and exons 2, 3 and 4 of
mutation status between primary colorectal tumours and their NRAS (expanded RAS analysis)] also predict a lack of response to
metastases with conflicting results. While some studies have EGFR-targeting monoclonal antibodies and that these therapies
failed to find any difference in KRAS mutation status between the may in fact have a detrimental effect in patients with RAS-mutant
primary tumours and their metastases [18–22], others have disease, specifically when combined with an oxaliplatin-based
reported discordant results in 4%–32% of the patients [23–35]. cytotoxic backbone [48–54].
However, many of these studies involved the analysis of small In the PRIME study, in which patients were randomised to
numbers of samples, involving heterogeneous metastatic sites and receive panitumumab plus FOLFOX4 [infusional 5-fluorouracil
the use of techniques with low sensitivity that might have led to (5-FU), leucovorin, oxaliplatin] versus FOLFOX4 alone first-
false-negative results if adequate enrichment of the tumour cells line, additional RAS mutations were detected in the tumours of
was not carried out. In a large study of 305 matched primary col- 17% of patients with mCRC originally classified as KRAS exon 2
orectal tumours and liver metastases, the discordance rate was wild-type. These patients also failed to benefit from panitumu-
3.6% [36]. When these data are pooled with results from different mab therapy, and had inferior progression-free survival (PFS)
previous small studies, the overall rate of discordance is ∼5% for and OS times compared with those treated with FOLFOX4
liver metastases. In contrast, a discordance rate of 25% has been alone (not statistically significant). In fact, this study was the
described for lymph node metastases. Although these data are first to hint at a detrimental effect of panitumumab in patients
limited to KRAS exon 2 mutations, they can be extrapolated to whose tumours carried RAS mutations at sites other than KRAS
situations where expanded RAS analysis has been conducted (see exon 2 [48].
below), for which no information is available. Based on this evi- Conversely, those patients whose tumours did not have RAS
dence, tissue from either a patient’s primary tumour or a liver mutations at the tested sites had significantly better outcomes
metastasis may be used for RAS mutation testing. Lymph node from the addition of panitumumab to FOLFOX4 than those
metastases do not seem to be suitable for the determination of patients whose tumours contained RAS mutations. The phase II
the RAS mutation status of colorectal tumours. In patients for PEAK study that evaluated FOLFOX6 plus panitumumab
whom both primary tumour and metastases are available, testing versus FOLFOX6 plus bevacizumab in untreated patients with
of a sample from either site is sufficient. KRAS exon 2 wild-type mCRC supported these findings.
Patients with KRAS and NRAS exon 2, 3 and 4 wild-type mCRC
recommendation 3: tissue selection. treated with FOLFOX6 plus panitumumab achieved a better
PFS than those treated with FOLFOX6 plus bevacizumab and a
• Tissue from either the primary tumour or a liver metastasis trend towards improved OS was also observed [53]. Using next-
may be used for RAS mutation testing [III, A]. generation sequencing (NGS) techniques, investigators analysed
• Other metastatic sites such as lymph node or lung metastases tumour samples previously assessed for KRAS exon 2 codon 12
may be used only if primary tumour or liver metastases and 13 mutations from patients enrolled in the phase III
samples are not available [II, B]. 20020408 trial of panitumumab in patients with chemorefrac-
tory mCRC [52] for additional RAS-activating mutations.
Patients with RAS wild-type tumours achieved a response rate
definition and validation of biomarkers
(RR) with panitumumab of 15% compared with 1% for those
Biomarkers can be diagnostic, predictive or prognostic. Ideally, patients with RAS-mutant tumours.
a biomarker should only serve one of these purposes, but there These findings with panitumumab have been upheld by
are good and clinically relevant examples of prognostic biomar- trials evaluating cetuximab. Using sensitive BEAMing (Beads,
kers that predict a response to a specific therapy, for example, Emulsions, Amplification, and Magnetics) technology, KRAS
human epidermal growth factor receptor 2 (HER2) in breast exon 2 wild-type tumours from the pivotal CRYSTAL and
cancer and BRAF (strongly prognostic and, to a lesser extent, OPUS studies were retrospectively evaluated for mutations in
predictive) in CRC [37–39]. It is also essential to follow strict KRAS exons 3 and 4 and NRAS exons 2, 3 and 4 [49, 50]. In the
rules for the development and validation of biomarkers that are phase III CRYSTAL study, which randomised patients to receive
specific to the purpose and sometimes also specific to the nature first-line FOLFIRI (infusional 5-FU, leucovorin, irinotecan)
of each biomarker. Establishing clinical utility in the appropriate with or without cetuximab, other RAS mutations were detected
clinical setting is essential [40]. in nearly 15% of evaluable patients previously assessed to be
KRAS exon 2 wild-type. Similarly, in the phase II OPUS study,
RAS testing which randomised patients to receive first-line FOLFOX4 with
evidence that tumour RAS mutational status is predictive. or without cetuximab, mutations at other RAS loci were detected
Retrospective analyses of pivotal clinical trials for the epidermal in 31% of evaluable tumours previously assessed to be KRAS
Volume 27 | No. 8 | August 2016 doi:10.1093/annonc/mdw235 | 

by guest
on 05 February 2018
exon 2 wild-type. In patients with RAS wild-type tumours recommendation 4: RAS testing.
(according to the expanded RAS analysis), the addition of cetuxi-
mab to FOLFIRI or FOLFOX4 was associated with improved • RAS mutational status is a negative predictive biomarker for
treatment outcomes across all efficacy end points. Conversely, in therapeutic choices involving EGFR antibody therapies in the
patients with RAS-mutant tumours, no benefit from the addition metastatic disease setting [I, A].
of cetuximab to FOLFIRI versus FOLFIRI alone was observed ° RAS testing should be carried out on all patients at the time
[49]. In the OPUS study, the addition of cetuximab to FOLFOX4 of diagnosis of mCRC [I, A].
was associated with a non-significant improvement in PFS and • RAS testing is mandatory before treatment with the EGFR-
OS in patients with RAS wild-type tumours; it seemed to be detri- targeted monoclonal antibodies cetuximab and panitumumab
mental in patients whose tumours carried RAS mutations. [I, A].
Data from the phase III FIRE-3 trial also underscore the • A network of arrangements should be established to ensure
importance of expanded RAS mutational analysis in the selection the rapid and robust transit of tissue samples from referral
of patients for treatment with cetuximab. Previously untreated centres to testing laboratories, to minimise the turnaround
patients, with KRAS exon 2 wild-type mCRC, were randomised time and avoid delays in having this information available for
to receive FOLFIRI with either cetuximab or bevacizumab. all patients with mCRC.
Additional RAS mutations were identified in the tumours of 16% • Primary or metastatic colorectal tumour tissue can be used for
of assessable patients, with an improvement in OS (median 33.1 RAS testing (see also Recommendation 3).
versus 28.7 months) observed for patients with RAS wild-type • RAS analysis should include at least KRAS exons 2, 3 and 4
tumours treated with cetuximab compared with those with KRAS (codons 12, 13, 59, 61, 117 and 146) and NRAS exons 2, 3 and
exon 2 wild-type tumours treated with cetuximab [55]. 4 (codons 12, 13, 59, 61 and 117).
Confirmation of these observations was provided by a sys- • Turnaround time for RAS testing (expanded RAS analysis)
tematic review and meta-analysis of randomised, controlled should be ≤7 working days from the time of receipt of the
trials evaluating EGFR antibody therapy [56]. The analysis specimen by the testing laboratory to the time of issuing of
showed that across nine trials involving 5948 patients, patients the final report, for >90% of specimens.
with tumours without any RAS mutations were found to have a • Validation (or verification, where more applicable) of RAS
significantly better treatment outcome with EGFR monoclonal testing assays should be carried out and recorded before
antibody therapy than those whose tumours harboured RAS implementation in clinical use. Laboratory audit mechanisms
mutations [56]. should be in place.
In summary, the cumulative data clearly show that patients • Laboratories providing RAS testing of colorectal tumours
whose tumours harbour any RAS mutation are unlikely to should demonstrate their successful participation in a relevant
benefit from EGFR antibody therapy, confirming the presence external quality assessment scheme, and be appropriately
of a RAS mutation (according to expanded RAS analysis) as a accredited.
negative predictive marker of treatment outcome in patients with
mCRC who might be under consideration for EGFR monoclo-
nal antibody therapy. Thus, cetuximab and panitumumab BRAF testing
should only be considered for the treatment of patients with BRAF mutations (nearly always V600E) are found in the tumours
RAS wild-type mCRC. Expanded RAS analyses should be con- of between 8% and 12% of patients with mCRC included in clini-
ducted on all patients eligible/being considered for EGFR anti- cal trials and are almost exclusively non-overlapping with RAS
body therapy. mutations [38, 60, 61]. A retrospective analysis of patients with
mCRC demonstrated that two-thirds of BRAF-mutant patients’
timing of testing. Wong et al. [57] discuss whether RAS testing primary lesions were located on the right side of the colon and
of CRC is better practised as a ‘reflex’ or an ‘on-demand’ associated with an increased incidence of peritoneal and distant
process. However, the general consensus of the expert panel was lymph node metastases, but fewer pulmonary metastases [60].
that patients should be assessed for their tumour RAS mutation Just under one-third of BRAF-mutant tumours also had microsa-
status at the time of diagnosis of their metastatic disease, to tellite instability (MSI), and the same proportion of tumours with
facilitate strategic treatment decisions within a multidisciplinary MSI contained BRAF mutations.
team (MDT) environment, local reimbursement regulations BRAF mutations are a significant negative prognostic marker
permitting. However, it should also be noted that an external for patients with mCRC. Tran et al. [60] reported a median sur-
quality assessment has uncovered differences in the quality of vival for patients with BRAF-mutant mCRC of 10.4 months
RAS testing for EGFR antibody therapy [58] and that, to date, compared with 34.7 months for patients with BRAF wild-type
the exact cut-off for clinically relevant RAS-mutant allele tumours. In a multivariate analysis, the hazard ratio (HR) for
frequencies has not been determined. survival was 10.662 (P < 0.001) [60]. This particularly poor
Investigation of cost estimates and the economic implications prognosis for patients with BRAF-mutant tumours is supported
of expanded RAS testing in patients with mCRC showed the by a number of randomised studies with specific chemotherapy
increased societal cost of expanded RAS testing versus KRAS regimens [38, 44, 48, 61–63]. Although the evidence of BRAF
exon 2 testing to be inconsequential when compared with the mutations as a negative predictive biomarker for EGFR antibody
amount of money saved by not treating the additional up to therapy in later lines is accumulating [64, 65], its role in
18% of patients who harbour additional RAS mutations (beyond earlier lines in combination studies with chemotherapy has
those in KRAS exon 2) with EGFR antibody therapies [59]. not been ascertained [44]. Indeed, two meta-analyses [66, 67]
 | Van Cutsem et al. Volume 27 | No. 8 | August 2016

by guest
on 05 February 2018
showed the efficacy benefit of EGFR antibody therapies to be tumourigenesis in patients with CRC. Tumours with MSI repre-
greater in patients with RAS wild-type/BRAF wild-type tumours sent only 4%–8% of tumours in patients with mCRC. Data are
than in those with RAS wild-type/BRAF-mutant tumours. In currently scarce on the prognostic and predictive values of an
the meta-analysis that included two second-line trials and two MSI phenotype in the metastatic disease setting [78–80]. A recent
trials involving chemorefractory patients [66], the lack of the retrospective analysis demonstrated that the median age was a bit
conferral of a significant efficacy benefit by EGFR-antibody younger (67 years), poor differentiation was more frequent
therapies over standard chemotherapy alone in patients with (58%), and that 45% of patients whose tumours had an MSI phe-
BRAF-mutant tumours was considered to support the assessment notype had stage IV disease at presentation. BRAF V600E muta-
of tumour BRAF mutation status before the initiation of EGFR- tions were present in 30% of patients with MSI [79]. In mCRC,
antibody therapy. Conversely, authors of the second meta-analysis some data have suggested that MSI tumours tend to have lower
[67] concluded that there was insufficient evidence to exclude disease control rates when treated with oxaliplatin-based first-line
EGFR antibody therapy from patients with RAS wild-type/BRAF- therapy [81], although most studies show MSI status to be not
mutant disease. However, in a small subgroup analysis (n = 28) of relevant as a single predictive marker for response to irinotecan-
the TRIBE study, the cohort of patients with BRAF-mutant or oxaliplatin-based chemotherapy regimens and not predictive
tumours treated with the chemotherapy triplet FOLFOXIRI plus for the effect of chemotherapy more generally in these patients
bevacizumab showed a non-statistically significant increase in OS [78, 82, 83].
compared with those treated with FOLFIRI plus bevacizumab In a pooled analysis of four phase III studies in the first-line
[68]. treatment of mCRC (CAIRO, CAIRO2, COIN and FOCUS),
Also, BRAF V600E-mutated melanomas are sensitive to the BRAF mutations have been shown to be more frequent in
BRAF-mutant inhibitor vemurafenib [69], but BRAF-mutated patients whose tumours exhibit MSI than in those whose
CRCs are not as sensitive [70, 71]. Feedback reactivation of tumours exhibit MSS [62]. The same pooled analysis showed
EGFR in CRC could explain why CRCs generally have a lower PFS and OS to be significantly worse for patients with tumours
response to BRAF inhibitors [37, 71]. Clinical trials are ongoing with MSI when compared with those with tumours showing
to test targeted therapies in patients with metastatic BRAF MSS [HR, 1.33; 95% confidence interval (CI) 1.12–1.57 and HR,
(V600E) mutant CRC, using combinations of BRAF-mutant 1.35; 95% CI 1.13–1.61, respectively], and for patients with
inhibitors (dabrafenib, vemurafenib or encorafenib) in combina- BRAF-mutant tumours when compared with those with BRAF
tion with MEK and EGFR inhibition, and in some cases conven- wild-type tumours (HR, 1.34; 95% CI 1.17–1.54 and HR, 1.91;
tional cytotoxic therapy. Early results are promising [72, 73]. 95% CI 1.66–2.19, respectively) [62]. Emerging data have shown
Furthermore, somatic BRAF V600E mutations have been MMR status to predict the clinical benefit of immune check-
associated with sporadic cases of DNA mismatch repair defi- point blockade with pembrolizumab in patients with mCRC.
ciency (dMMR) showing an MSI phenotype [74]. However, The immune-related objective RR and immune-related
BRAF V600E mutation is not associated with the MSI pheno- 6-month PFS rate were 40% (4 out of 10 patients) and 78%
type due to a germline MMR mutation (Lynch syndrome) [75, (7 out of 9 patients), respectively, for patients with dMMR CRC
76]. BRAF V600E mutation testing has therefore been proposed and 0% and 11% for those with MMR-proficient CRC, with
as a means to exclude Lynch syndrome. Recently, however, excellent median PFS and survival (not reached) in the cohort
patients with BRAF-mutant tumours with mutations in codons with dMMR CRCs versus 2.2 and 5.0 months, respectively, in
594 and 596 were shown to exhibit microsatellite stability (MSS) the cohort with MMR-proficient tumours [84].
and markedly longer OS when compared with patients with Thus, the prevalence of MSI and BRAF mutations in the
BRAF V600E-mutant disease [77]. tumours of patients with mCRC is low. Both biomarkers confer
Tumour BRAF mutation status should be determined for an inferior prognosis, which in the case of patients with
every case of CRC, ideally at the time of diagnosis, as this repre- tumours with MSI may be driven by the presence of BRAF
sents a different biological subtype, and in combination with mutations. These conclusions are supported by the data from
testing for dMMR, can assist in the identification of a germline other studies which show the presence of a BRAF V600E muta-
versus somatic cause of dMMR. In patients with mCRC, BRAF tion to be as poor a prognostic factor in patients with tumours
mutation status should be assessed at the same time as RAS with MSI as it is in other patients with mCRC [60].
mutational status for prognostic assessment (and/or potential
selection for clinical trials). recommendation 6: MSI testing.
recommendation 5: BRAF testing. • MSI testing in the metastatic disease setting can assist clini-
cians in genetic counselling [II, B].
• Tumour BRAF mutation status should be assessed alongside the • MSI testing has strong predictive value for the use of immune
assessment of tumour RAS mutational status for prognostic check-point inhibitors in the treatment of patients with
assessment (and/or potential selection for clinical trials) [I, B]. mCRC [II, B].
MSI testing biomarkers of chemotherapy sensitivity or toxicity

Tumours with MSI retain their chromosomal numbers intact dihydropyrimidine dehydrogenase. Dihydropyrimidine
but contain microsatellite repeats, which vary in length due to dehydrogenase (DPD) is a key enzyme in the metabolic
dMMR, and are thought to contribute to the early steps of catabolism of 5-FU and capecitabine. About 85% of 5-FU is
Volume 27 | No. 8 | August 2016 doi:10.1093/annonc/mdw235 | 

by guest
on 05 February 2018
eliminated through a catabolic process involving DPD. excision repair cross-complementation group 1. The function of
Numerous genetic mutations have been identified in the DPD the excision repair cross-complementation group 1 (ERCC1)
gene locus (DPYD), with a few key variants having functional protein is predominantly in the nucleotide excision repair of
consequences for enzymatic activity. Deficiencies in DPD damaged DNA. Nucleotide excision repair is the primary DNA
activity have been shown to cause 5-FU-treated cancer patients repair mechanism involved in the removal of therapeutic
to experience severe drug-related toxicities [85], and DPD platinum-DNA adducts from tumour DNA. A variety of
activity is a predictive biomarker of potential toxicity when methods can be used to measure the level of ERCC1 activity,
using 5-FU and capecitabine [86]. Polymorphism has been namely immunohistochemistry (IHC) for protein expression,
documented mainly on the DPYD*2A gene at a frequency of reverse transcription–polymerase chain reaction (RT–PCR) for
2%–3% with geographical variation. mRNA expression and DNA single-nucleotide polymorphism
Several methods are available to detect DPD deficiency such (SNP) for genotyping. High ERCC1 levels have been shown to
as the functional dihydrouracil/uracil ratio in plasma, the uracil be a negative predictive marker for platinum-based therapy in
breath test or DPYD*2 mutations. Patients with known partial patients with lung cancer [90, 91]. In CRC, depending on the
DPD deficiency benefit from dose adaptation of their 5-FU/ techniques used, high ERRC1 expression levels have been
capecitabine therapy to avoid severe toxicity. In patients with shown to be associated with poor prognosis and to be predictive
complete DPD deficiency, fluoropyrimidines should not be used of a poor outcome in patients receiving oxaliplatin-based
and an alternative treatment offered. therapy (RT–PCR mRNA evaluation). A meta-analysis showed
DPD deficiency is generally not assessed in routine practice ERCC1-C118T polymorphisms to predict clinical outcome in
before 5-FU administration. There is no recommended standar- patients with CRC receiving oxaliplatin-based therapy [92].
dised assessment technique, although several methods are avail- More specifically, PFS and OS were significantly shorter in
able (see above). None of the current strategies are adequate to patients with T/T or T/C genotypes of ERCC1-C118T when
mandate routine DPD testing before starting fluoropyrimidine- compared with those with the C/C genotype. Thus, high ERCC1
based therapy [II, C]. gene expression seems to confer oxaliplatin resistance, while
Testing for DPD deficiency, however, remains an option. In ERCC1-C118T polymorphisms are predictive of treatment
the case of patients who experience severe 5-FU toxicity, DPD outcome in patients receiving oxaliplatin-based therapy [92].
levels should be tested before 5-FU is re-introduced. Recently it has been proposed that ERRC1 induction after
exposure to oxaliplatin may be dependent on KRAS mutational
status [93].
UDP glucuronosyltransferase 1 family, polypeptide A1. UDP At the present time, the use of ERCC1 protein levels cannot
glucuronosyltransferase 1 family, polypeptide A1 (UGT1A1) is be recommended for treatment decisions involving the use of
an enzyme of the glucuronidation pathway that transforms small oxaliplatin in routine practice. Clinical trials have not been able
lipophilic molecules, such as steroids, bilirubin, hormones and to demonstrate a predictive role for ERCC1 status for treatment
drugs, into water-soluble, excretable metabolites. The gene is part of with oxaliplatin.
a complex locus that encodes several UDP-glucuronosyltransferases.
Polymorphism may be associated with increased toxicity to
irinotecan. UGT1A1 is responsible for bilirubin glucuronidation as thymidylate synthase. Thymidylate synthase (TS) is the primary
well as glucuronidation of SN-38, the active metabolite of irinotecan. target for 5-FU. 5-FU is an inhibitor of TS. Experimentally, it has
Genetic variations within the UGT1A1 gene have also been been shown that low levels of TS expression lead to a better
associated with the development of certain drug toxicities. The response to 5-FU and improved survival of colon cancer patients
UGT1A1*28 variant, the allele behind many cases of Gilbert [94]. The TS gene (TYMS) is under the control of a promoter
syndrome, has been associated with an increased risk for neutro- acting as an enhancer (TSER). Earlier studies have shown that
paenia in patients receiving irinotecan [87, 88], and the United higher numbers of TSER repeats (TSER2*, TSER3* or higher)
States Food and Drug Administration recommends on the irino- lead to higher TS expression and activity. TS activity and CRC
tecan drug label that patients with the *28/*28 genotype should sensitivity to 5-FU seem to correlate with TSER polymorphisms.
receive a lower starting dose of irinotecan [89]. The *28 allele These correlations, however, need to be confirmed in a larger
has also been shown to be associated with an increased risk of randomised study.
developing diarrhoea in patients receiving irinotecan [87, 88].
The UGT1A1*6 variant, more common in Asian populations
recommendation 7: biomarkers of chemotherapy sensitivity and
than the *28 variant, has also been associated with the develop- toxicity:
ment of irinotecan-related toxicities. Patients who are heterozy-
gous or homozygous for the *6 allele may have a higher risk of • DPD testing before 5-FU administration remains an option
developing neutropaenia and diarrhoea than those with the but is not routinely recommended [II, D].
UGT1A1*1/*1 genotype. • UGT1A1 phenotyping remains an option and should be
Thus, UGT1A1 gene polymorphisms are predictive of irinote- carried out in patients with a suspicion of UGT1A1 deficiency
can-related side-effects, including diarrhoea, neutropaenia and as reflected by low conjugated bilirubin and in patients where
vomiting. However, in everyday practice, UGT1A1/UGT1A1 an irinotecan dose of >180 mg/m2 per administration is
status is rarely used as a predictive biomarker of irinotecan toxi- planned [95] [III, C].
city. Attention should be paid to bilirubin levels, especially in • ERCC1 expression cannot be recommended for use as a bio-
patients where conjugated bilirubin is <20% of total bilirubin. marker for treatment decisions involving the use of oxaliplatin
 | Van Cutsem et al. Volume 27 | No. 8 | August 2016

by guest
on 05 February 2018
in routine clinical practice, but could be included prospec- prognosis. A systematic review and meta-analysis of the prog-
tively in clinical trials [III, D]. nostic value of the CpG island methylator phenotype (CIMP) in
• TS activity and TSER genotyping are not recommended for patients with CRC showed the CIMP to be independently asso-
use in clinical practice [II, D]. ciated with a significantly worse prognosis [115]. However, epi-
genetic DNA hypermethylation inactivation of the SRBC gene,
the product of which interacts with the product of the BRCA1
emerging biomarkers gene, predicted a shorter PFS, particularly in oxaliplatin-treated
A list of biomarkers beyond RAS mutational status is emer- patients with mCRC for whom metastasectomy was not indi-
ging which may impact on the response to all classes of cated (HR, 1.96; 95% CI 1.13–3.40; log-rank P = 0.01). SRBC
targeted agents, and specifically the current perspective of hypermethylation was also associated with a shorter PFS (HR,
EGFR-antibody therapies. These include HER2, MET and 1.90; log-rank P = 0.045), in a validation cohort of unresectable
KRAS gene amplification, ligands such as transforming colorectal tumours treated with oxaliplatin [116].
growth factor-α (TGF-α), amphiregulin and epiregulin,
EGFR mutations and alterations/mutations in HER3, PI3KCA recommendation 8: emerging biomarkers not recommended for
and PTEN. routine patient management outside of a clinical trial setting:
Mutations in KRAS, NRAS and BRAF and amplification of
HER2 and MET drive primary (de novo) resistance to anti-EGFR • Detection of mutations in PIK3CA, exon 20 [II, D].
treatment. Recently, the emergence of alterations in these genes • Evaluation of PTEN loss by IHC [V, D].
was detected in patients who responded to EGFR blockade and • Evaluation of the levels of the EGFR ligands amphiregulin,
then relapsed. Molecular heterogeneity impairs the efficacy of epiregulin and TGF-α [II, D].
EGFR-antibody therapy in patients with mCRC by fuelling • Evaluation of levels of EGFR protein expression [II, E].
de novo and acquired resistance [96]. With the exception of EGFR • Evaluation of EGFR amplification and copy number and
mutations, which are described only in the acquired setting, all of EGFR ectodomain mutations [IV, D].
the genetic alterations defined as a mechanism of de novo resis- • Evaluation of HER2 amplification or HER2 activating mutations.
tance are also responsible for acquired resistance. Differences can • Evaluation of HER3, and MET receptor overexpression [IV, D].
be found in the frequency of individual genetic alterations, such as
KRAS and NRAS exon 3 mutations, which occur more frequently
in the acquired rather than in the de novo setting. Acquired resis- emerging technologies
tance to EGFR-antibody therapy is driven by the selection of cell A number of novel tools for the assessment of diagnostic, prog-
clones that carry RAS or RAF mutations but account for only nostic and/or predictive biomarkers in patients with mCRC have
0.4%–17% of tumour cells. Mutations in KRAS exons 3 and 4 and been proposed, with an increasing interest in liquid biopsies
NRAS exons 2, 3, and 4 as well as amplification of KRAS, HER2 involving the analysis of either circulating tumour cells (CTCs) or
and MET [96–99] account for around 20% of mCRC patients who circulating tumour DNA (ctDNA). Although the levels of CTCs
do not benefit from anti-EGFR treatment, although initially as assessed (mostly using the CellSearch system) have been shown
selected for anti-EGFR treatment based on KRAS exon 2 wild- to correlate with prognosis in patients with mCRC [117], the clin-
type status [48, 52–54, 97, 100–104]. The prognostic role of ical utility of CTC assessments in patients with mCRC has hardly
PIK3CA mutations is uncertain [105], but a PIK3CA exon 20 been explored.
mutation may predict resistance to EGFR-antibody therapy [106– Conversely, analysis of ctDNA is emerging as a new tool for
110], although the correlation is not strong enough to be applied molecular profiling that has more possibilities for translation into
as a negative predictive marker [111]. PIK3CA and PTEN altera- the clinic than CTCs. The seminal work of Bardelli and colleagues
tions often co-occur with KRAS or BRAF mutations [107, 112], has shown very promising results from ctDNA liquid biopsies
but there is insufficient evidence for their use as biomarkers of [118, 119]. In addition to the seminal papers from Bardelli and
resistance to EGFR-antibody therapy. There is no clear evidence colleagues and Montagut et al. [120], a number of tumour–blood
for HER3 overexpression and HER3 mutations, mesenchymal– concordance studies are currently being conducted that will
epidermal transition (MET)/MET alterations (overexpression or undoubtedly validate the clinical utility of these technologies for
gene amplification) or KRAS amplification, EGFR mutations [tyr- identifying tumour mutations in the blood of patients. Currently,
osine kinase (TK) or ligand-binding domains] or amplification in its use as a monitoring tool for secondary resistance to EGFR
the resistance to EGFR antibody therapies. Emerging data indicate antibody therapies is under investigation. It can be anticipated
that HER2 activating mutations or HER2 amplification may that liquid biopsies will be used therapeutically in the near future
mediate in some instances resistance to EGFR antibodies [100, as more and better drugs are developed against mutant clones (or
113]. A phase II clinical trial also showed that HER2 amplification those with other molecular alterations, e.g. amplifications, etc.)
is predictive of response to HER2 dual inhibition with trastuzu- that emerge upon exposure to EGFR-targeted therapies [40, 118,
mab and lapatinib in a cohort of CRC patients failing EGFR anti- 120–135].
body therapy [114]. Similarly, increasing evidence suggests that micro-RNA
Thus, although CRC is primarily considered to be a genetic (miRNA) is involved in the pathogenesis and progression of mCRC
disease, characterised by the sequential accumulation of genetic [136]. However, the prognostic and predictive role of miRNA needs
alterations, there is growing evidence that epigenetic alterations to be demonstrated in a randomised clinical trial setting. Finally,
add an additional layer of complexity to its pathogenesis and NGS can provide important information on tumour heterogeneity
characterise a subgroup of CRCs with a distinct aetiology and and clonal evolution. NGS has already been published as a reliable
Volume 27 | No. 8 | August 2016 doi:10.1093/annonc/mdw235 | 

by guest
on 05 February 2018
technology for use in patients with mCRC and has the potential to provide the opportunity to register patients for the local and/or
screen for larger cancer gene panels in clinical trials [137]. national registries with extreme/unusual patients’ details just
noted, to provide information on the diversity of patients seen.
recommendation 9: emerging technologies. Several (observational) studies have shown improved clinical out-
comes, including improved OS, when patients with CRC are
• Although CTC number correlates with prognosis in patients managed by MDTs [141, 142].
with mCRC, the clinical utility of CTC assessments is not yet
The role of the MDT is to define the initial diagnostic workup
clear and therefore cannot be recommended [IV, D].
and then the treatment focus, based on the best diagnostic and
• The utility of liquid ctDNA biopsies to guide treatment deci- therapeutic decision-making available [3]. Furthermore, an
sions is currently under investigation in clinical trials, but
MDT-managed treatment strategy has to be maintained for the
cannot yet be recommended in routine practice [V, D].
duration of a patient’s treatment, to allow the refinement of
• Whole genome, whole exome and whole transcriptome analy- treatment strategies according to on-treatment information (e.g.
sis should be carried out only in a research setting [V, D].
response to a selected treatment) and evaluation of the potential
need for the integration of ablative treatments (such as second-
view on how molecular classification should be ary surgery and LAT strategies, see below).
developed going forward The first step in the process is for the MDT members to criti-
cally define whether or not a patient has initially clearly resect-
CRC is a heterogeneous disease with heterogeneous outcomes
able or initially unresectable metastatic disease and to define the
and drug responses. To date, pathological staging and gene
status of the resection of the primary tumour when considering
expression signatures have failed to accurately predict disease
the management of both synchronous and/or oligometastatic
recurrence and prognosis. In an attempt to identify biologically
CRC, and the first-line treatment of patients with metastatic
homogeneous subtypes of CRC, many independent groups have
disease. Conversely, for patients whose disease is deemed ‘never
reported the results of gene expression-based subtyping, with
to be resectable’, the discussion may be left to the treating
Marisa et al. [138], the first to present a robust transcriptome-
medical oncologist (after discussion with the MDT) and patient
based classification of colon cancer. Subsequently, an inter-
as to the pros and cons of various approaches and sequences
national consortium dedicated to large-scale data sharing and
based on the perceived aims [e.g. duration of disease control
analytics has recently provided a robust and unified classifica-
versus quality of life (QoL), and toxicity profiles, etc.].
tion, defining four different subtypes: CMS1 (MSI Immune),
hypermutated, microsatellite unstable, with strong immune activa-
tion; CMS2 (Canonical), epithelial, chromosomally unstable, with oligometastatic disease
marked WNT and MYC signalling activation; CMS3 (Metabolic),
OMD is characterised by the localisation of the disease to a few
epithelial, with evident metabolic dysregulation; and CMS4
sites and lesions and is associated with the option to use LAT
(Mesenchymal), prominent TGF-β activation, stromal inva-
approaches in patient treatment strategies with a view to improving
sion and angiogenesis [139]. This effort provides the most
disease control and therefore clinical outcome in these patients.
robust and reproducible classification system currently avail-
Generally, OMD may be characterised by the existence of
able for CRC and may form the basis for future clinical trials.
metastases at up to 2 or occasionally 3 sites and 5 or sometimes
more lesions, predominantly visceral and occasionally lymphono-
local ablative treatment (LAT), including dal. Typically, these are the primary, and other involved sites such
surgery, and management of patients as the liver, lung, peritoneum, nodes and ovary. Patients with
with OMD disease at other sites, such as multiple lesions in the bones and the
brain, may also be treated using a local ablative approach, but as
the role of MDTs and tumour boards these patients are associated with an unfavourable prognosis, local
The optimal treatment strategies for patients with mCRC are evol- ablative treatment strategies are only used to prevent immediate
ving rapidly with improved clinical outcomes being achieved complications. This latter group of patients should be excluded
when the treatment approaches for individual patients are dis- from a classification of OMD. On the other hand, a patient with
cussed within an MDT of experts who meet regularly as a tumour one or two resectable liver metastases, and a single bone lesion,
board to review mCRC cases [140, 141]. An ideal MDT should should be classified as having OMD, because for a patient with
include access to both a colorectal surgeon (preferably with exper- this disease profile, locally ablative treatment strategies could be
tise in peritoneal approaches) and a specialist hepatobiliary and/ used and meaningfully contribute to their prognosis.
or, lung surgeon as necessary, with the obligatory inclusion of a Thus, treatment strategies for patients with OMD should be
pathologist and a diagnostic radiologist, as well as radiation and based on the possibility of achieving complete ablation of all
medical oncologists. An interventional radiologist/nuclear physi- tumour masses, using surgical R0 resection (complete resection
cian may also be included as appropriate, as the role of ablative with clear resection margins and no evidence of microscopic
treatments gains increasing importance (see below). Ideally, residual tumour) and/or LAT, either initially or possibly after
patients should be treated either in specialist cancer centres or, induction treatment with systemic therapy, for both the primary
alternatively, where this is not possible, as part of a network of tumour and metastases.
individuals dedicated to the management of CRC with an estab- For patients with OMD confined to a single organ (most fre-
lished referral route between their site or centre and a specialist quently the liver), or a few organs (pre-dominantly visceral metas-
cancer centre (virtual MDTs). Wherever possible, MDTs should tases, e.g. lung), a potentially curative approach exists. Numerous
 | Van Cutsem et al. Volume 27 | No. 8 | August 2016

by guest
on 05 February 2018
case series have shown that in this setting, long-term survival or involvement is associated with better outcomes, it may be appro-
even cure can be attained in 20%–50% of patients who undergo priate to ‘watch and wait’ or at least employ a sequential approach
complete R0 resection of their metastases [143]. Even in the [144, 145]. The data showing different outcomes depending on the
absence of randomised trials comparing surgical with non-surgical site(s) of OMD are likely to reflect molecular differences. For
disease management, surgery has become the standard treatment example, patients whose mCRC is associated with RAS and BRAF
approach for patients with resectable OMD. mutations have worse clinical outcomes, with RAS mutations
For patients with more extensive OMD involving more sites shown to be associated with an increased incidence of lung, bone
or lesions, e.g. primary, liver, lung, peritoneum, nodes, bones, and brain metastases [146]. Moreover there are data to suggest that
brain, ovary and >4 organs, the value of a surgical approach is tumour TS expression levels and RAS mutation status are altered
controversial. In these patients, surgery may contribute to long- by site of metastasis compared with the primary [23–36, 147].
term survival but is rarely curative [143]. For this group of
patients, the consideration of localised interventions (LAT) recommendation 10: OMD.
becomes relevant, in combination with systemic therapy (as part
of a multimodal therapy approach), following a careful MDT • For patients with OMD, systemic therapy is the standard of
discussion and assessment. The goal for this group of patients is care and should be considered as the initial part of every treat-
to achieve long-term disease control, potentially contributing to ment strategy (exception: patients with single/few liver or lung
OS (and, although unlikely, potentially cure), with well- lesions, see below).
controlled sites of metastases, but without continued systemic • The best local treatment should be selected from a ‘toolbox’ of
therapy. Liver-directed therapy is probably the best established procedures according to disease localisation, treatment goal
of the LAT interventions; however, the increasing use of the (‘the more curative the more surgery’/higher importance of
appropriate ablative treatment strategy from a ‘toolbox’ of local/complete control), treatment-related morbidity and patient-
options, including, for example, stereotactic ablative body radio- related factors such as comorbidity/ies and age [IV, B].
therapy (SBRT) and radiofrequency ablation (RFA) for visceral or
nodal involvement, peritonectomy with or without hyperthermic
intraperitoneal chemotherapy (HIPEC) for peritoneal disease, liver metastases and surgical resection
and nodal dissection, sees the management of this subgroup of For patients with colorectal liver metastases (CLM), the treat-
patients becoming increasingly complex (Figure 1). Furthermore, ment strategy should be directed towards complete resection
the potential still exists for isolated bone, pancreatic and brain whenever possible, with both ‘oncological’ ( prognostic) and
metastases, but these are rare and likely to not have a defined ‘technical’ (surgical) criteria being considered when evaluating
treatment pathway. patients for surgery [148]. However, prospective evaluations do
Subcharacterisation of OMD according to site also impacts on not exist either for ‘oncological’ or for ‘technical’ criteria, and
the treatment options and the timing of treatment. Patients with for many of these, there is no (international) consensus.
liver and lung metastases have a much better prognosis than those The ‘technical’ definitions of resectable CLM have evolved
with other metastatic disease locations. In fact, because lung over time, with the current consensus proposing that disease
Toolbox of ablative treatments
Local treatments Locoregional treatments
Local
Thermal devices Non-thermal devices Embolic devices
chemotherapy
Radiofrequency Brachytherapy Radioembolisation

ablation or electroporation SIRT
cryoablation
External Body Chemoembolisation

Microwave ablation Radiotherapy with TACE/Beads
high-precision RT
Figure 1. Toolbox of ablative treatments. SIRT, selective internal radiation therapy; RT, radiation therapy; TACE, transarterial chemoembolisation.
Volume 27 | No. 8 | August 2016 doi:10.1093/annonc/mdw235 | 

by guest
on 05 February 2018
should be considered technically resectable as long as complete contrast enhancers (such as gadoxetate) which is associated with
macroscopic resection is feasible, while maintaining at least a a higher accuracy of lesion detection [157].
30% future liver remnant (FLR) or a remnant liver to body For the detection of extrahepatic metastases and local recur-
weight ratio >0.5 (e.g. >350 g of liver per 70 kg patient) [149– rence at the site of the initial colorectal surgery, CT and positron
151]. However, the concern remains that not all patients with emission tomography (PET)/CT scans are used [158]. A pro-
technically resectable liver-limited metastases benefit from spective randomised trial evaluating high-quality CT and PET
surgery, with approximately half developing widespread sys- imaging involving 263 patients showed only a 7.6% change in
temic disease within 3 years of resection [152]. management following PET [159], while a retrospective analysis
The ‘oncological’ criteria provide prognostic information that reported a change in intended curative therapy to palliative
predict a longer disease-free survival (DFS) or a higher likeli- therapy or vice versa in one-third of patients [160]. Also, a
hood of cure. These include, as strong parameters, the number recently published meta-analysis of studies evaluating PET and
of lesions, the presence (or suspicion) of extrahepatic disease PET/CT in patients with liver metastases reported PET findings
and the criteria used in numerous retrospective evaluations and to result in changes in the management of a mean of 24% of
in the FONG score [153]. Thus, for some patients, neoadjuvant patients, with a mean incidence of PET-based extrahepatic
chemotherapy may be a better option than upfront surgery.
In practice, patients can be categorised into groups based on
technological and oncological criteria as outlined in Figure 2 Table 2. Contraindications to hepatic resection in patients with
and according to the new system for deciding whether or not a CRC liver metastases (adapted from Adam et al. [148] with
patient is eligible for resection proposed by Adam et al. [148], permission from AlphaMed Press)
and described in Table 2. Category Contraindication
Technical (A)
imaging in the identification of resectable/
1. Absolute Impossibility of R0 resection with ≥30% liver remnant
unresectable disease Presence of unresectable extrahepatic disease
Computed tomography (CT) scans are routinely used for
2. Relative R0 resection possible only with complex procedure
primary staging and disease surveillance in patients with CRC.
(portal vein embolisation, two-stage hepatectomy,
Although practice varies between treatment centres, the evi-
hepatectomy combined with ablationa)
dence suggests that the best methods for detection of liver R1 resection
metastases from CRC are CT and magnetic resonance imaging Oncological (B)
(MRI) [154]. However, many teams alternate liver ultrasonogra- 1. Concomitant extrahepatic disease (unresectable)
phy (US) and CT for detection of disease to decrease the expo- 2. Number of lesions ≥5
sure of patients to the radiation resulting from repeated CT 3. Tumour progression
scans. For the characterisation of focal liver lesions, CT, con-
trast-enhanced US (CEUS) and MRI can be used [155]. For Patients should be categorised as A1 or A2/B1, B2 or B3.
a
lesions <10 mm in diameter, MRI is a more sensitive modality All methods, including radiofrequency ablation.
than CT [156] and specifically hepatobiliary MRI with specific
Oncological
criteria
(prognostic)
Bad Preoperative FOLFOX
Good Conversion with

Perioperative FOLFOX ‘best systemic therapy’
No preoperative
Excellent therapy
(adjuvant?)
Surgical
criteria
Easy Difficult (technical)
Figure 2. Categorisation of patients according to technical and oncological criteria. FOLFOX, infusional 5-fluorouracil, leucovorin, oxaliplatin.
 | Van Cutsem et al. Volume 27 | No. 8 | August 2016

by guest
on 05 February 2018
disease of 32% [161]. However, although PET may provide addi- doublet plus monoclonal antibody therapy or FOLFOXIRI
tional information, mainly in patients with a high risk of extra- either alone or in combination with bevacizumab should be
hepatic disease, there is currently no consensus as to the patient considered preoperatively [V, consensus >75%].
population with the most to gain. The current evidence is not In the case of patients with good oncological and technical
considered strong enough to recommend the use of PET in all (surgical) criteria, who have not received perioperative che-
patients. motherapy, there is no strong evidence to support the use of
adjuvant chemotherapy [166]. However, the experience of
recommendation 11: imaging in the identification and management Kemeny et al. [167] indicates that patients with unfavourable
of disease. prognostic criteria (e.g. by FONG score) may benefit from adju-
vant treatment. However, the expert opinion is that if patients
• Imaging should comprise first an abdominal/pelvic and thor- have not received any previous chemotherapy for metastatic
acic CT scan and, in the case of doubt, a second method such disease, then chemotherapy is recommended (low level of evi-
as US (CEUS), MRI or PET/CT scan depending on the locali- dence—expert opinion), with the recommendations being
sation of the metastases. US may be helpful to characterise FOLFOX or CAPOX, unless patients were previously recently
liver metastases, MRI liver, peritoneal or pelvic metastases (<6–12 months) exposed to oxaliplatin-based adjuvant che-
and PET/CT extrahepatic disease [IV, B]. motherapy for stage II or III CRC.
• A stepwise imaging approach is the recommended policy, in
relation to the therapeutic possibilities, rather than the use of
recommendation 12: perioperative treatment.
all imaging modalities in all patients [V, B].
• Both technical criteria for resection and prognostic considera-
tions define the need for systemic perioperative therapy [IV, B].
liver metastases that are technically resectable • In patients with clearly resectable disease and favourable prog-
up front nostic criteria, perioperative treatment may not be necessary
The primary goal for patients who present with technically and upfront resection is justified [I, C; consensus >75%].
resectable liver metastases is clearly cure, with the primary goal • In patients with technically resectable disease where the prog-
R0 resection, although it should be noted that a 10-year follow nosis is unclear or probably unfavourable, perioperative com-
up is required for confirmation of this [162]. In the management bination chemotherapy (FOLFOX or CAPOX) should be
of these patients, imaging is used to determine the nature and administered [I, B; consensus >75%].
true extent of their disease. • Targeted agents should not be used in resectable patients
In patients with ‘favourable oncological’ criteria (i.e. >50% like- where the indication for perioperative treatment is prognostic
lihood of cure based on various factors including long-term meta- in nature [II, E].
chronous disease), and ‘favourable surgical’ criteria (no massive • In situations where the criteria for prognosis and resectability
disease infiltration), both upfront surgery [R0 resection/no evi- are not sharply defined, perioperative therapy should be con-
dence of disease (NED)] and perioperative chemotherapy are sidered (as part of a continuum of treatment option) [IV, B]
options. The panel expressed no clear preference for one option (Figure 2). Patients with synchronous onset of metastases
over the other, since the 5-year OS rate reported for the EPOC should be allocated to this group and therapeutic pathway.
study with perioperative chemotherapy, 51% (95% CI 45–58) in • In patients with favourable oncological and technical (surgi-
the perioperative chemotherapy group versus 48% (95% CI 40– cal) criteria, who have not received perioperative chemother-
55) in the surgery-only group, is not convincing, despite the fact apy, there is no strong evidence to support the use of adjuvant
that the DFS in eligible patients was significantly improved [163]. chemotherapy [II, C], whereas patients with unfavourable cri-
However, in patients with disease that is technically easy to teria may benefit from adjuvant treatment [III, B].
resect but where the prognostic situation is unclear or likely not • In patients who have not received any previous chemotherapy,
to be ‘excellent’, perioperative chemotherapy should be the treat- adjuvant treatment with FOLFOX or CAPOX is recom-
ment approach of choice (Figure 2). Perioperative chemotherapy mended (unless patients were previously recently exposed to
in this group should comprise 3 months chemotherapy before oxaliplatin-based adjuvant chemotherapy) [IV, B].
surgery and 3 months chemotherapy post-surgery, only. The • Decision-making should include patients’ characteristics and
preferred treatment in this setting should be FOLFOX [or alter- preferences [IV, B].
natively capecitabine with oxaliplatin (CAPOX)] as reported for
the EPOC trial [163, 164]. EGFR-targeting monoclonal antibo-
dies (cetuximab and panitumumab) are not to be used in this unresectable CLM with ‘conversion’ as a strategic
setting, based on the data from the New EPOC trial [165]. No treatment goal
data with bevacizumab are available for this specific patient Any patient with limited liver and/or lung metastases should be
group; therefore, bevacizumab should not be used [V, consensus considered a candidate for potential secondary resection as cur-
>75%]. rently, there are no criteria that allow us to distinguish between
In patients with disease that is technically easy to resect but those patients for whom purely palliative treatment and those
with one or more unfavourable prognostic features, resulting in for whom potentially curative treatment is appropriate.
a relatively low chance of ‘cure’, there is uncertainty regarding Systemic therapy given with a view to rendering technically
the best treatment strategy. Either FOLFOX alone, as used in the unresectable colorectal metastases resectable is called conversion
EPOC study, or a highly active regimen such as a chemotherapy therapy, and offers the best means of ‘converting’ patients with
Volume 27 | No. 8 | August 2016 doi:10.1093/annonc/mdw235 | 

by guest
on 05 February 2018
unresectable metastatic disease to resectability [168]. Also, conversion treatment

although survival times are slightly shorter for those patients The observation that patients with initially unresectable CLM
with mCRC who undergo conversion therapy followed by whose metastases are rendered resectable after responding to
surgery than for those patients with initially resectable meta- chemotherapy have a better long-term outcome than patients
static disease, they are far better than if resection was not carried treated with chemotherapy alone has led to the introduction of
out at all [168, 169]. conversion chemotherapy into clinical practice [148, 171–174].
In patients receiving conversion therapy, response to systemic Resection rates have been shown to be correlated with
therapy is a strong prognostic indicator but is also unpredict- response to systemic therapy [175]. However, these correlations
able. With the increasing efficacy of systemic therapy regimens, may be biased by other factors such as the year of the trial (in
it is recommended that resectability is first evaluated after (only) later years with more active systemic regimens, resection is more
2 months of optimal treatment and again after 4 months, when frequently integrated into the therapeutic algorithms), as well as
the maximal tumour shrinkage is deemed to have occurred in patient selection and criteria for resection. Furthermore, only a
most patients, so that the opportunity for resection is not few of the trials specifically designed to investigate conversion
missed in patients who a priori have a low chance of further chemotherapy as a treatment strategy in patients with initially
resection [148]. However, due to the limitations of RECIST (1.1; unresectable CLM (Table 3) were randomised, controlled trials,
potentially 2.0) [170], radiologists should be advised to pay making it difficult to reach any decision regarding the ‘best’
special attention to the treatment effects if the vascular endothe- regimen to use in this clinical setting.
lial growth factor (VEGF)-targeting antibody bevacizumab is a In the CELIM trial, one of the first of these trials to be con-
component of the therapy regimen. ducted, patients with technically unresectable and/or ≥5 liver
As reported previously [5], up to 75% of these patients will metastases treated with either FOLFOX plus cetuximab or
suffer a relapse following resection of their hepatic metastases, FOLFIRI plus cetuximab were evaluated for resectability every
with the majority occurring in the liver. There is no role for 2 months [176]. A tumour RR of 62% was achieved for all
partial palliative resection of metastases, but other ablative tech- patients and 70% in patients with KRAS exon 2 wild-type disease.
niques, such as RFA or SBRT, may be used as an adjunct to An encouraging 33% of patients across the two treatment arms
surgery to achieve a situation where there is NED. They may underwent R0 resection of their liver metastases. However, as this
also provide an alternative to resection in the case of patients trial involved randomisation between two different chemotherapy
with poor anatomical localisation of their metastases for resec- regimens, both in combination with cetuximab, no conclusion
tion, and in order to retain sufficient FLR. Resection of resect- can be drawn regarding either the benefit of different treatment
able lung metastases offers 25%–35% 5-year survival rates in intensities or the benefit of any specific drug used.
carefully selected patients. Although resection of lung metas- More importantly, two randomised phase II trials in patients
tases is less well studied, R0 resection of lung metastases can with unresectable disease have shown treatment intensification to
also be recommended [5]. lead to increased RRs with a consequential increase in the rates of
In addition ∼20%–30% of newly diagnosed patients with R0 resection and therefore improved prognosis [177, 178]. The
mCRC present with synchronous metastases. There is no stan- first of these was a prospective, randomised, Chinese trial in 138
dard of care for treating patients with synchronous CRC liver patients with KRAS exon 2 wild-type liver-limited disease where
metastases, although in this potentially curative setting treat- an increased RR in the cetuximab-containing combination che-
ment typically involves a two-stage resection. However, some- motherapy (FOLFIRI/mFOLFOX6) arm was associated with an
times surgery is not the first step for these patients who may increase in R0 resection rate [177]. Twenty patients (29%) in the
also require systemic therapy. The majority opinion was that cetuximab-containing arm and nine (13%) in the chemotherapy
patients presenting with synchronous metastatic disease should alone arm became eligible for resection. Overall, 18 patients
be treated more aggressively, with the recommendation that pre- (26%) in the cetuximab arm and five (7%) in the chemotherapy
operative chemotherapy should be used. alone arm underwent an R0 resection. Significantly, patients in
Table 3. Conversion chemotherapy approach in patients with liver-limited disease

Study CTx Controlled study n RR, % Liver resection rate, %
Vie-LM-Bev [302] CAPOX + bevacizumab No 56 73 93

CELIM [176] FOLFOX6/FOLFIRI + cetuximab No 106 70 33
GONO [179] FOLFOXIRI + bevacizumab No 30 80 40
POCHER [303] Chrono-IFLO + cetuximab No 43 79 60
BOXER [304] CAPOX + bevacizumab No 45 78 40
OLIVIA [178] FOLFOXIRI + bevacizumab versus FOLFOX + bevacizumab Yes 80 81 versus 62 49 versus 23 (R0)
Ye et al. [177] FOLFIRI/FOLFOX ± cetuximab Yes 116 57 versus 29 26 versus 7 (R0)
CAPOX (XELOX), capecitabine and oxaliplatin; Chrono-IFLO; chronomodulated irinotecan, 5-fluorouracil, leucovorin and oxaliplatin; CTx,
chemotherapy; FOLFIRI, infusional 5-fluorouracil, leucovorin and irinotecan; FOLFOX, infusional 5-fluorouracil, leucovorin and oxaliplatin; FOLFOXIRI,
infusional 5-fluorouracil, leucovorin, irinotecan and oxaliplatin; RR, response rate.
 | Van Cutsem et al. Volume 27 | No. 8 | August 2016

by guest
on 05 February 2018
either treatment arm undergoing hepatic resection had a longer The pathological response after preoperative chemotherapy
median survival time than those who did not [46.4 versus 25.7 also provides strong prognostic information and could serve in
months for the cetuximab arm (P = 0.007) and 36.0 versus 19.6 the future as a stratification parameter for further treatment
months for the chemotherapy alone arm (P = 0.004)]. decisions. To date, no prospective pathological response data
The second trial was the European, multinational, open-label, from randomised trials are available, and therefore, pathological
phase II OLIVIA trial in which patients with unresectable liver response should not yet be used as a decision-making factor.
metastases were randomised to receive bevacizumab plus either The time to maximum response is typically ∼12–16 weeks
FOLFOXIRI (n = 41) or mFOLFOX6 (n = 39) [178]. The overall (FIRE-3 trial DpR analysis) [193] in patients with disease that is
resection rates were 61% and 49%, respectively, and the R0 re- borderline in terms of resectability and who are receiving perio-
section rates were 49% and 23%, respectively. The correspond- perative therapy (Figure 2). According to the expert consensus
ing tumour RRs were 81% and 62%, respectively. In this trial, discussion, the total therapy duration pre- and post-surgery
FOLFOXIRI plus bevacizumab was associated with higher re- should not exceed 6 months.
sponse and resection rates than mFOLFOX6 plus bevacizumab The role of continued systemic treatment, post-conversion treat-
in patients with initially unresectable CLM. However, as bevaci- ment and surgery is unclear. It is also unclear whether the mono-
zumab was included in both arms and the intensification was clonal antibody therapies should be continued post-resection.
set by the addition of a third chemotherapy compound, the rela- Intra-arterial chemotherapy and chemoembolisation have been
tive value of bevacizumab in this setting remains unclear, as shown to achieve high RRs and R0 resection rates in small series
FOLFOXIRI alone is known to achieve a high RR [179]. Also, [195–197] and may be used to shrink a larger tumour so that it
consideration of the other available data from these studies [176, can be removed by surgery, but the data on chemoembolisation
177, 180] clearly shows both FOLFOX and FOLFIRI to be active for liver metastases from CRC are exploratory.
in combination with EGFR inhibitors in patients with RAS
wild-type disease in this treatment setting, while FOLFOXIRI
plus (or minus) bevacizumab has been shown to be superior to recommendation 13: conversion therapy.
the corresponding FOLFOX or FOLFIRI regimens and its activ-
• In potentially resectable patients (if conversion is the goal), a
ity to be independent of tumour RAS and BRAF mutation status
regimen leading to high RRs and/or a large tumour size
[68, 178, 181, 182].
reduction (shrinkage) is recommended [II, A].
Studies involving the retrospective analysis of RR (specifically
• There is uncertainty surrounding the best combination to use
in patients with liver-limited disease) and the corresponding R0
as only few trials have addressed this specifically:
resection rates provide additional information [44, 46, 183, 184],
but need to be regarded with caution. However, it seems clear ° In patients with RAS wild-type disease, a cytotoxic doublet
plus an anti-EGFR antibody seems to have the best benefit
that regimens that achieve high RRs are beneficial and are asso-
risk/ratio, although the combination of FOLFOXIRI plus
ciated with higher R0 resection rates. Thus, the standard che-
bevacizumab may also be considered and, to a lesser extent,
motherapy regimens used in the CRYSTAL, PRIME and OPUS
a cytotoxic doublet plus bevacizumab [II, A].
trials with EGFR-targeting monoclonal antibodies versus che-
motherapy alone in patients with RAS wild-type disease, and ° In patients with RAS-mutant disease: a cytotoxic doublet
plus bevacizumab or FOLFOXIRI plus bevacizumab [II, A].
FOLFOXIRI plus bevacizumab versus the doublet mFOLFOX6
• Patients must be re-evaluated regularly in order to prevent the
plus bevacizumab should be regarded as standard treatment
overtreatment of resectable patients as the maximal response
options. Moreover, data from the FIRE-3 [55] and CALGB
is expected to be achieved after 12–16 weeks of therapy in
[185] studies show that a cytotoxic doublet plus cetuximab in
most patients.
RAS wild-type patients is associated with higher RRs compared
with bevacizumab, although this did not translate into higher
resection rates in either of these studies.
metastases at unfavourable/uncommon sites and
role of ablative treatment with or without surgery
role of other efficacy (response) parameters Patients with a limited number of lesions and involved sites and
who therefore do not belong to the group of patients with limited
The new metric response parameters early tumour shrinkage CLM should be regarded as having OMD and be treated accord-
(ETS) and depth of response (DpR) are emerging to characterise ing to the standard treatment algorithm presented in Figure 3.
the response, but may also be useful as predictors of long-term In these patients, the use of local ablation therapies such as RFA
outcome in patients with mCRC [186], and in particular those or cryoablation has been shown to be feasible, as well as precision
receiving EGFR-antibody therapy [187–189], although recent radiotherapy (SBRT) and, to a lesser extent, chemoembolisation.
data have also shown this for a triplet compared with a doublet of The selection of the best instruments from the ‘toolbox’ of abla-
cytotoxics in combination with bevacizumab [68, 190]. In trials tive therapies (Figure 3) for use in this setting differs according to:
investigating treatment intensification [44, 46, 182, 191, 192], the
more intensive therapy arms had higher DpR and ETS rates, and • the size and localisation of the metastases—and therefore
higher RRs. In the FIRE-3 and PEAK trials, the DpR and ETS access with regard to the use of the best treatment method;
rates were higher for the EGFR-inhibitor containing combina- • the rates of local control achieved (with the local control
tions (cetuximab and panitumumab, respectively) than for the greater for surgery than for the other options);
bevacizumab-containing regimen [193, 194]. • the invasiveness of the technique;
Volume 27 | No. 8 | August 2016 doi:10.1093/annonc/mdw235 | 

by guest
on 05 February 2018
recommendation 14: ablative techniques.

Best systemic treatment
in terms of induction of response • Despite the lack of more available prospective data, this strate-
gic treatment approach should be evaluated and pursued
further in suitable patients [II, B].
Evaluation at 6–8 weeks

toolbox of LATs
At time of "best response" also evaluate
use of best treatment strategies available The most important discriminator for the usage of different
(patient-/expertise-dependent) toolbox instruments is, after tumour location, the type of energy
administered. Current technologies comprise invasive thermal
ablation with distinct size limitations (e.g. RFA and others),
conformal radiation techniques which are directed against iso-
"Toolbox" instruments for local ablative lated lesions, and chemoembolisation or radioembolisation with
treatment (surgery, invasive local ablation
[RFA, microwave], precision radiotherapy
yttrium-labelled microspheres, both of which are limited to the
[SBRT], embolisation techniques liver for use in the management of CLM that are rather diffuse.
[any particles/beads, SIRT])
Consider (recommended) re-uptake

thermal ablation
of systemic treatment, but limit treatment In patients with advanced CLM, thermal ablation such as RFA
duration to a total of 6 months
often cannot be used due to the inherent size limitation of
∼3 cm [199]. However, in the phase II CLOCC trial (chemother-
Figure 3. Standard treatment algorithm for patients with oligometastatic apy plus or minus RFA) [200], RFA combined with surgical
disease. RFA, radiofrequency ablation; SBRT, stereotactic body radiation resection for the treatment of patients with CLM suggested an
therapy; SIRT, selective internal radiation therapy. improvement in both PFS and OS [198]. A considerable amount
of data are available on the use of thermal ablation in combina-
tion with liver resection for the treatment of patients with CLM
either as part of a two-stage approach or intraoperatively using
• the non-tumour-related prognostic considerations and patient ultrasound guidance [201].
factors as well as patient preferences; Thermal ablation techniques also have proven efficacy in the
• the local expertise with regard to the use of a particular abla- ablation of lung metastases from CRC. Local control rates of
tive treatment method; 88%–92% at 1 year and 77% at 3 years have been reported for
• consideration of patient frailty and life expectancy. RFA of lung metastases [202, 203]. Mortality and major compli-
Selection of the best ‘situation-adapted’ treatment strategy cation rates may be as low as 1%, whereas grade 1 and 2 events
should consider all of these factors as part of an MDT treatment occur in up to 33% of treatments [204, 205]. However, a recent
decision before the start of systemic treatment and at the time of meta-analysis of four RFA patient series and 23 surgical patient
best response. Adoption of the treatment approach outlined in series demonstrated that the data currently available for lung
Figure 3 requires repeated MDT discussions for the duration of metastases from CRC do not allow a firm conclusion to be drawn
an individual patient’s treatment pathway. with regard to the use of surgery or RFA, although most evidence
supports surgery as the most effective treatment option [205].
use of local and ablative therapy in patients stereotactic ablative body radiotherapy
with OMD (with non-curative intent) High conformal hypofractionated irradiation [e.g. SBRT, high-
A treatment goal of ablation is a relatively new concept for dose rate (HDR)-brachytherapy] of CLM has been reported to
patients with mCRC and involves an attempt to eradicate all achieve high local control rates. The risk of recurrence correlates
visible metastatic lesions using the best instrument from the with increasing tumour size as well as the applied dose regimen
toolbox of LATs, in combination with systemic therapy. The [206, 207].
overall goal of this strategy is not necessarily to cure the patient, SBRT and HDR-brachytherapy achieve similar results to
as the prognosis for these patients is generally poor due to the RFA, with local tumour control >80% at 12 months depending
unfavourable localisation of their metastases and the number of on size [208–212]. Also, although grade 2 toxicity may be as
involved organs coupled with the limitations of local ablative high as 70%, grade ≥3 events have not been recorded across
treatments, compared with surgical resection. several series. Support for the use of SBRT in the liver is growing
However, full ablation of all visible sites may allow disconti- with data reported for five retrospective studies [213–215] and
nuation of the standard of care, systemic therapy, with the possi- eight prospective studies [216–222] of SBRT in the treatment of
bility of a (relevant) relapse-/disease-free interval. The CLOCC liver metastases from various primaries. SBRT has also been
trial, a prematurely terminated randomised phase II trial, has used successfully in patients with unresectable visceral pulmon-
shown that the combined approach with surgery and RFA of ary or hepatic metastases [223]. Prospective trials will validate
unresectable metastases plus systemic therapy may be associated which patients benefit most from SBRT with its short treatment
with a significant improvement in OS [198]. time course, lack of a need for recovery and favourable overall
 | Van Cutsem et al. Volume 27 | No. 8 | August 2016

by guest
on 05 February 2018
toxicity profile. The use of SBRT together with systemic therapy of the study) and the OS data are not yet available (SIRFLOX
should also be investigated prospectively. study) [226]. However, a (potentially relevant) improved time to
liver progression has been shown for patients treated with che-
recommendation 15: local ablation techniques. motherapy plus radioembolisation. In this trial, around 45% of
patients had the primary tumour in place and around 40% had
• In patients with unresectable liver metastases only, or OMD, extrahepatic disease, suggesting that radioembolisation may be
local ablation techniques such as thermal ablation or high most beneficial in patients with liver-limited disease.
conformal radiation techniques (e.g. SBRT, HDR-brachyther- Yttrium-90-labelled particles may also currently be a good
apy) can be considered. The decision should be taken by an alternative in patients who are potential candidates for resec-
MDT based on local experience, tumour characteristics and tion, but display a small FLR volume. A matched-pair analysis
patient preference [IV, B]. comparing yttrium-90-labelled particles with portal vein embo-
• In patients with lung only or OMD of the lung, ablative high lisation showed a lesser, but still pronounced benefit of yttrium-
conformal radiation or thermal ablation may be considered if 90-labelled particles with regard to contralateral liver hypertrophy,
resection is limited by comorbidity, the extent of lung par- following simultaneous treatment of the ipsilateral tumour load
enchyma resection, or other factors [IV, B]. with yttrium-90-labelled particles [227].
• SBRT is a safe and feasible alternative treatment for oligome-
tastatic colorectal liver and lung metastases in patients not
amenable to surgery or other ablative treatments [IV, B]. recommendation 16: embolisation.
• RFA can be used in addition to surgery with the goal of eradi-
cating all visible metastatic sites [II, B]. • For patients with liver-limited disease failing the available
chemotherapeutic options
° Radioembolisation with yttrium-90 microspheres should
chemoembolisation be considered [II, B].
To date, the data on chemoembolisation for liver metastases from ° Chemoembolisation may be also considered as a treatment
CRC are mostly observational series in various treatment situa- option [IV, B].
tions [195–197]. Comparative data are limited to irinotecan-based • Radioembolisation (and chemoembolisation) of CLM in
drug-eluting beads in a small phase II cohort in previously treated earlier treatment lines may be interesting as ‘consolidation
patients showing a benefit versus systemic chemotherapy [224], treatment’ but should be limited to clinical trials.
and the role of intra-arterial irinotecan in patients pre-exposed to
intravenous irinotecan is unclear. Numerous trials with che-
motherapy-loaded particles (beads) are ongoing, also in combina- cytoreductive surgery and HIPEC for patients with
tion with systemic treatment and in the neoadjuvant setting. peritoneal metastases
In selected patients with peritoneal metastasis, complete cytoreduc-
radioembolisation tive surgery and HIPEC may provide prolonged survival when
Radioembolisation [selective internal radiation therapy (SIRT)] carried out in experienced high-volume centres (in view of the rela-
typically involves a single delivery of yttrium-90 connected to tively high morbidity associated with the procedure) [228–230].
either resin or glass particles into the hepatic artery with the The efficacy of this multimodality treatment depends on the extent
therapeutic effect essentially limited to irradiation. of peritoneal dissemination and is scored using the peritoneal
For patients with liver-limited metastases failing the available cancer index (PCI), which is the main prognostic factor [231].
chemotherapeutic options, radioembolisation with yttrium-90 Involvement of the lower ileum is a negative prognostic factor.
resin microspheres has been shown to prolong the time to Cytoreductive surgery is particularly effective in patients with
tumour progression in the liver, based on a small randomised low-volume peritoneal disease (a PCI <12 is often suggested) and
phase III study [225]. no evidence of systemic disease. With recommendations on stan-
Recently, a randomised phase III study of SIRT with resin micro- dardising the delivery of HIPEC in patients with CRC [232] and
spheres as an adjunct to chemotherapy in the first-line treatment evaluation of oxaliplatin versus mitomycin C for HIPEC [233],
setting failed to show an overall PFS benefit (as primary end point cytoreductive surgery and HIPEC is on the verge of becoming the
Table 4. Drivers for first-line treatment

Tumour characteristics Patient characteristics Treatment characteristics
Clinical presentation: Age Toxicity profile

Tumour burden
Tumour localisation
Tumour biology Performance status Flexibility of treatment administration
RAS mutation status Organ function Socioeconomic factors
BRAF mutation status Comorbidities, patient attitude, expectation and preference Quality of life
Volume 27 | No. 8 | August 2016 doi:10.1093/annonc/mdw235 | 

by guest
on 05 February 2018
Table 5. Historical ESMO groups for treatment stratification of fit patients with metastatic CRC [3]
Group 0 Group 1 Group 2 Group 3
Resectable Potentially resectable Not resectable Not resectable
Clinical presentation Clearly resectable R0 liver Unresectable liver/lung- Multiple metastases/sites Asymptomatic
and/or lung disease limited disease which Tumour-related Multiple metastases
might become resectable symptoms Never able to undergo resection
after response to Able to withstand Unsuitable for intensive therapy
conversion therapy intensive therapy Frail with co-morbidities
Treatment goal Cure (NED) Maximum tumour Clinically relevant tumour Halt/slow tumour progression
shrinkage shrinkage Tumour shrinkage less relevant
Disease control Tolerability most relevant
Treatment intensity Surgery Intensive treatment approach Less intensive treatment approach
Immediate surgery with Upfront most active Upfront active Treatment selected according to
no prior chemotherapy combination regimen combination (at least a patient preference
or chemotherapy doublet) Sequential approach (start with
moderate (FOLFOX) single agent or doublet with low
perioperative toxicity)
chemotherapy FOLFOX an exception
CRC, colorectal cancer; FOLFOX, infusional 5-fluorouracil, leucovorin, oxaliplatin; NED, no evidence of disease.
accepted standard treatment approach for patients with peritoneal • 5–7 months of second-line therapy;
metastases from a colorectal primary. • a treatment break before initiation of a further line;
• approximately 3 months of third-line therapy;
recommendation 17: cytoreductive surgery and HIPEC. • potentially a fourth line (in patients with RAS wild-type disease);
• a few months of re-challenge of initial induction or first-line
• Complete cytoreductive surgery and HIPEC can be consid-
therapy;
ered for patients with limited peritoneal metastases in centres
• a few months best supportive care only.
which are very experienced in the use of HIPEC [III, B].
treatment of metastatic disease determination of the therapeutic strategy

The optimal therapeutic strategy for each patient is determined
The definition of a (potential) treatment aim and strategy is
following a clinical examination, blood counts, determination of
important for both the upfront integration of a multimodal treat-
liver and renal function parameters, measurement of tumour
ment approach and for the choice of a systemic treatment strategy
marker [the most relevant being carcinoembryonic antigen
(first-line and later-line) as part of a ‘continuum of care’.
(CEA)] levels, an abdominal and thoracic CT/MRI scan and an
Relevant factors for determination of the treatment goal are: assessment of the patient’s general clinical condition (health),
independent of their malignant disease.
• tumour- and disease-related characteristics, such as clinical pre- The general condition and performance status of a patient are
sentation and patterns of tumour biology (e.g. metastases strong prognostic and predictive factors for chemotherapy.
limited to the liver and/or lung, the dynamics of progression, Whether a patient is classified as ‘fit’ or ‘unfit’ is now used to
symptoms and prognostic molecular or biochemical markers); determine whether or not they will be assigned to a more inten-
• patient-related factors (co-morbidity, socioeconomic factors sive (combination of 2 or 3 cytotoxics with a biological) or less
and expectations of the patient); intensive treatment approach with the classical drivers of treat-
• treatment-related factors such as toxicity (Table 4). ment choice being tumour, patient and treatment characteristics
A patient with classical mCRC may typically achieve an OS of as outlined in Table 4. Historically, ‘fit’ patients with mCRC
∼30 months as the result of an MDT-managed ‘continuum of were categorised according to the previous ESMO consensus
care’. An example of a typical ‘continuum of care’ treatment guidelines into four groups (0, 1, 2, and 3) to determine the stra-
sequence is outlined below: tegic treatment approaches (Table 5) [3, 5].
The decision as to whether a patient has initially resectable or
• approximately 4–6 months of first-line ‘induction’ therapy; initially unresectable metastatic disease should be made at the
• 4–6 (–8) months of ‘maintenance’ therapy—or no treatment first meeting of the MDT. Patients with initially resectable meta-
after resection and/or ablation following first-line treatment; static disease should be referred for immediate resection or peri-
• about 3 months re-introduction (or treatment beyond pro- operative chemotherapy with the goal being to achieve complete
gression); R0 resection and/or a situation where the patient can be treated
 | Van Cutsem et al. Volume 27 | No. 8 | August 2016

by guest
on 05 February 2018
with another ablative treatment (LAT). In the case of patients than a fluoropyrimidine (5-FU/leucovorin) alone [I, B] [234,
with OMD, the goal would be the creation of a situation where 235]. Infusional regimens of 5-FU/leucovorin [234, 235] are
the patient has NED as described previously. generally less toxic than bolus regimens [236, 237] and should
However, in the case of fit patients with mCRC, whose metas- be used in preference. The oral fluoropyrimidine capecitabine
tases are not initially resectable, it is becoming increasingly can be used as an alternative to 5-FU/leucovorin alone [238]
obvious that the original ESMO groups 1 and 2 are becoming and in combination with oxaliplatin [239]. Capecitabine is less
less clearly delineated and the treatment strategies less strict (see frequently used in combination with irinotecan due to early
shading Table 5). concerns that it was more toxic than FOLFIRI [240, 241].
Indeed, two clinically relevant categories are evolving for the However, the results are controversial [242, 243]. The monoclo-
systemic treatment of ‘fit’ patients with CRC whose metastatic nal antibodies bevacizumab (anti-VEGF) and cetuximab and
disease is not resectable at presentation: panitumumab (anti-EGFR) have been shown to improve the
clinical outcome of patients with mCRC when combined with
A. 1) Those for whom intensive treatment is appropriate with combination chemotherapy regimens in the first-line setting [I,
the goal of cytoreduction (tumour shrinkage) and con- B] [43–46, 48–50, 101, 244, 245].
version to resectable disease The triplet combination chemotherapy regimen FOLFOXIRI
or has been demonstrated to be superior to FOLFIRI in an Italian
2) Those who need intensive treatment, although they will study [181]. FOLFOXIRI plus bevacizumab has also been shown
never make it to resection or LAT, since they need a rapid to be superior to both FOLFIRI plus bevacizumab and
reduction in tumour burden because of impending clinical FOLFOX6 plus bevacizumab [68, 178, 182]. However, the super-
threat, impending organ dysfunction or severe symptoms. iority of the cytotoxic triplet over a cytotoxic doublet has not
been demonstrated in all studies [246]. The contribution of bev-
B. Those for whom intensive treatment is not necessary and
acizumab in the triplet combination is also uncertain.
where the goal is disease control.
Thus, the chemotherapy options for the treatment of patients
The application of LAT within the context of OMD and the with mCRC in the first-line setting are typically (for most patients)
sequence of induction chemotherapy followed by LAT (without a cytotoxic doublet such as FOLFOX, CAPOX or FOLFIRI or, pos-
further systemic treatment) may also need to be considered as a sibly, in very selected patients the cytotoxic triplet FOLFOXIRI or
pre-defined treatment sequence. Such patients should be consid- fluoropyrimidine monotherapy in selected patients with asympto-
ered as belonging to group A1 above. matic primarily unresectable metastases that are likely to be eligible
For patients in both categories, knowledge of the RAS and for multiple lines of treatment and who are not candidates for a
BRAF mutational status of their disease is used to further refine combination chemotherapy.
treatment strategies (Table 6).
anti-VEGF therapy. The monoclonal antibody bevacizumab,
which binds circulating VEGF-A, has been shown to increase
the systemic therapy options in the first-line the activity (either RR, PFS and/or OS) in combination with
treatment setting bolus 5-FU/leucovorin/irinotecan and in combination with 5-
The typical first-line chemotherapy backbone comprises a fluor- FU/leucovorin or capecitabine alone in the first-line treatment
opyrimidine (intravenous 5-FU or oral capecitabine) used in setting [I, B] [244, 247–249]. Bevacizumab in combination with
various combinations and schedules with irinotecan or oxalipla- a fluoropyrimidine plus oxaliplatin has been shown to increase
tin [5]. Combination chemotherapy with a fluoropyrimidine PFS but not RR or OS in the first-line setting in a large phase III
plus oxaliplatin or irinotecan (FOLFOX or FOLFIRI) provides study [I, B] [245]. However, in smaller randomised trials
higher RRs and better progression-free and ( partly) OS times evaluating the addition of bevacizumab to FOLFOX or FOLFIRI
Table 6. Revised ESMO groups for treatment stratification of patients according to whether patients are ‘fit’ or ‘unfit’
Patient’s ‘Fit’ patients ‘Unfit’ patients
classification Group 1 Group 2
Clinical presentation A) Conversion and achievement of NED Asymptomatic patients Best supportive
B) Impending clinical threat, impending organ dysfunction No impending clinical threat care
and severe (disease-related) symptoms Resection not an option
Treatment biomarker driven: RAS wt, RAS mt, BRAF mt Treatment biomarker driven: RAS wt, RAS mt,
patient subgroups BRAF mt patient subgroups
Treatment goal A) Cytoreduction, followed by R0 resection; NED achieved Disease control and hence prolonged survival Palliative
by LAT
B) Improvement of symptoms and hence avoidance of rapid
evolution and prolonged survival
LAT, local and ablative therapy; mt, mutant; NED, no evidence of disease; wt, wild-type.
Volume 27 | No. 8 | August 2016 doi:10.1093/annonc/mdw235 | 

by guest
on 05 February 2018
failed to demonstrate an improved outcome [250, 251] which is recommendation 18: first-line systemic therapy combinations
somewhat at odds with the randomised trial comparisons of according to targeted agent used:
both chemotherapy backbones plus bevacizumab versus each
• Biologicals (targeted agents) are indicated in the first-line
other [242, 252], data from the CALGB 80405 trial where
treatment of most patients unless contraindicated [I, A].
investigator-based selection did not lead to a difference between
chemotherapy backbones [185], and the data from large
• The VEGF antibody bevacizumab should be used in combina-
tion with:
observational trials with nearly 5000 patients where no difference
was detected [253, 254]. ° the cytotoxic doublets FOLFOX/CAPOX/FOLFIRI,
FOLFOXIRI in combination with bevacizumab has also been ° the cytotoxic triplet FOLFOXIRI in selected fit and moti-
vated patients where cytoreduction (tumour shrinkage) is
shown to enhance RR and PFS compared with FOLFIRI plus
the goal—and potentially also in fit patients with tumour
bevacizumab [68] and to produce one of the longest median OS
BRAF mutations [II, B],
recorded in this clinical setting, but is limited to very selected
patients. The contribution of bevacizumab to the cytotoxic ° fluoropyrimidine monotherapy in patients unable to toler-
ate aggressive treatment [I, B].
regimens in both arms of this study is uncertain as it was not
investigated. • EGFR antibodies should be used in combination with:
Bevacizumab is usually continued in combination with any ° FOLFOX/FOLFIRI [I, A],
cytotoxic agent or any combination of cytotoxic agents until ° capecitabine-based and bolus 5-FU based regimens should
disease progression or unacceptable toxicity [5]. Currently, not be combined with EGFR antibodies [I, E].
there is no validated predictive marker for bevacizumab.
Consideration also needs to be given to the clear evidence that
anti-EGFR therapy. The EGFR antibodies cetuximab and patients should receive all three available cytotoxic agents
panitumumab are active in various combinations with their (fluoropyrimidine, oxaliplatin and irinotecan) and all tar-
activity, either alone or in combination with cytotoxics, limited geted treatments (anti-VEGF and, if RAS wild-type, anti-
to those patients whose tumours do not harbour a RAS mutation. EGFR) during the course of their treatment whenever possible
It has been shown that expanding RAS mutational analysis of [257, 258], although the optimal sequence remains to be elu-
tumours to include detection of mutations in exons 3 and 4 of cidated.
KRAS and exons 2, 3 and 4 of NRAS is superior to KRAS exon 2 To date, there is no unequivocal evidence for the superiority
analysis in predicting which patients are unlikely to respond of one class of biological over another (bevacizumab versus the
(negative predictive factor) or in whom EGFR antibody therapy EGFR antibody therapies) in the first-line treatment of patients
may be detrimental. Thus, a tumour RAS mutation is a negative with RAS wild-type mCRC, although the combination with an
predictive marker for treatment outcome with the EGFR EGFR antibody led to an improved RR in both phase III trials
monoclonal antibody therapies [II, B], and as stated previously and to an improved OS in the FIRE 3 study, but not in the
(Recommendation 4), knowledge of the expanded RAS mutational CALGB study. The PFS was identical for bevacizumab- and
status of a patient’s tumour is therefore a prerequisite for the use cetuximab-containing combinations in both phase III studies
of cetuximab or panitumumab as mandated by the European [55, 193, 259, 260].
Medicines Agency (EMA) [255, 256]. Also, although the treatment goal is a moving target, depend-
Expanded RAS analysis should be carried out at diagnosis in ing on the course of the disease, the aim should be for 70%–80%
order to determine whether EGFR antibody therapies are likely of ‘fit’ patients to receive second-line therapy and 50%–60% of
to be of clinical benefit. Moreover, the evidence is increasing ‘fit’ patients, third-line therapy.
that a BRAF mutation is predictive for a lack of benefit from
EGFR-targeting monoclonal antibodies administered as mono-
therapy in later lines [64, 65]. However, its role in combination discontinuation of treatment and the concept
with cytotoxic agents has not been ascertained [44]. of maintenance therapy
Cetuximab has been shown to improve the RR and median Historically, continuing patients on chemotherapy until
PFS and OS rates in first line in combination with FOLFIRI disease progression or unacceptable toxicity has been routine
when compared with FOLFIRI alone in mCRC patients with in clinical trials. However, clinical trials using this approach as
RAS wild-type tumours [43, 44, 49] [I, B]. Both cetuximab and well as clinical observations made during routine practice have
panitumumab also increase the activity of the cytotoxic doublet indicated the dangers of continuing cytotoxic therapy, specifi-
FOLFOX in mCRC patients with RAS wild-type tumours cally oxaliplatin-containing therapy, as cumulative toxicity
[38, 45, 46, 48, 50, 183]. However, in contrast, the addition of often occurs before clinical progression. As a result, disconti-
EGFR antibodies to oxaliplatin-based regimens where non-infu- nuation and/or intermittent combination chemotherapy/main-
sional fluoropyrimidines were used (e.g. bolus administration, tenance strategies have been investigated in a number of
FLOX; capecitabine, CAPOX) has not resulted in any benefit clinical trials with the result that these approaches provide an
[38, 61]. attractive treatment option for patients with a response or
Biologicals are generally indicated for the first-line treatment stable disease.
of patients with mCRC unless contraindicated due to, for The early UK MRC CR06 trial randomised patients with
example, reduced organ function, poor performance status or either an objective response or stable disease following 3 months
cardiovascular insufficiency. Capecitabine-based therapy should of single-agent fluoropyrimidine therapy to continue on che-
not be used in combination with EGFR antibody therapies [38]. motherapy or take a treatment break with further chemotherapy
 | Van Cutsem et al. Volume 27 | No. 8 | August 2016

by guest
on 05 February 2018
reserved for disease progression [261]. No clear difference in OS as tumour shrinkage continues or disease stabilisation is main-
was observed between the two treatment arms with an HR of tained and the treatment remains tolerable.
0.87 favouring intermittent therapy. Individualisation of the treatment approach after discussion
Since then the administration of intermittent combination with the patient is an essential component of this process and
chemotherapy has been investigated in a number of clinical should include discussions of projected survival time, time free
trials. The GISCAD study showed that intermittent treatment from cancer symptoms, time free from the side-effects and con-
with FOLFIRI compared with continued treatment did not straints of treatment, and impact on career and family life (social
diminish the efficacy of treatment [262]. The OPTIMOX-1 trial and financial).
randomised patients to receive FOLFOX4 until progression (or It is important to acknowledge that the discussion on mainte-
unacceptable toxicity) or FOLFOX7 (using a higher dose of oxa- nance treatment raises the question of a pre-planned abbrevia-
liplatin) for six cycles after which patients whose disease tion and shortening of the time on first-line therapy, followed
responded continued on maintenance 5-FU with the reintro- by a well-defined treatment algorithm.
duction of oxaliplatin only at disease progression [263]. No dif- However, ‘treatment holidays’ per se, in the meaning of a pro-
ference in PFS or OS time was noted and was taken as an longed pausing of any treatment for a limited time, can be discussed
indication that oxaliplatin-free intervals did not shorten survival for any patient with indolent and asymptomatic presentation of
time. The randomised OPTIMOX-2 and UK MRC COIN trials their disease.
subsequently investigated treatment breaks without mainte- In any patient, re-introduction of an initially successful induc-
nance 5-FU [264, 265]. In both studies, a detrimental effect with tion regimen should be considered as a treatment option either
inferior outcomes due to treatment-free intervals could not be following the ‘maintenance’ strategy or at a later stage in the
excluded, but was small probably due to the short ‘induction’ therapeutic pathway.
phase of chemotherapy.
More recently, the concept of treatment discontinuation has recommendation 19: maintenance therapy.
been refined further. Randomised trials involving more than
1000 patients have investigated the concept of ‘maintenance’
• Patients receiving FOLFOX or CAPOX plus bevacizumab-
based therapy as induction therapy should be considered for
treatment as a separate phase in the treatment strategy and con-
maintenance therapy after 6 cycles of CAPOX and 8 cycles of
tinuum of care [266–270]. The data from these randomised
FOLFOX. The optimal maintenance treatment is a combina-
phase II/III trials comparing maintenance therapy with biologi-
tion of a fluoropyrimidine plus bevacizumab. Bevacizumab as
cals plus or minus chemotherapy with a chemotherapy-free
monotherapy is not recommended [I, B].
interval [265, 269–271] show any fluoropyrimidine plus bevaci-
zumab to have the best activity in terms of interval PFS and a
• Patients receiving FOLFIRI can continue on induction
therapy—at a minimum—for as long as tumour shrinkage
trend towards an improved survival. Also, although a study
continues and the treatment is tolerable [V, B].
from the Nordic group did not show a benefit from the combi-
nation of bevacizumab and erlotinib [272], the DREAM study
• For patients receiving initial therapy with FOLFOXIRI plus or
minus bevacizumab, a fluoropyrimidine plus bevacizumab
showed a significant OS advantage for a maintenance strategy
may be considered as maintenance therapy (as was done in
with bevacizumab plus erlotinib [266]. However, this combina-
the pivotal trials examining FOLFOXIRI).
tion is not considered as a standard treatment because of the rela-
tively small size of the DREAM study and the lack of activity of
• For patients receiving initial therapy with a single-agent fluor-
opyrimidine ( plus bevacizumab), induction therapy should be
erlotinib in mCRC [266]. Currently, the integration of approaches
maintained [V, A].
other than de-escalation as maintenance strategies [266, 272],
should be reserved for clinical trials.
• Individualisation and discussion with the patient is essential
[V, A].
Thus, after ‘induction’ treatment, an active maintenance treat-
ment is seen as a possible option, especially in patients treated
• Initial induction therapy or a second-line therapy have to be
reintroduced at radiological or first signs of symptomatic pro-
with oxaliplatin-based chemotherapy, as it allows an early and
gression. If a second-line therapy is chosen, re-introduction of
upfront pre-planned discontinuation of the initially chosen sys-
the initial induction treatment should be a part of the entire
temic therapy combination. The optimal maintenance treatment
treatment strategy as long as no relevant residual toxicity is
following induction with fluoropyrimidine/oxaliplatin and bev-
present [III, B].
acizumab is a combination of a fluoropyrimidine (capecitabine)
plus bevacizumab as demonstrated in the CAIRO3 and AIO
0207 trials [268, 270] [I, B], and may also be considered for second line
patients following initial FOLFOXIRI plus or minus bevacizu- Second-line therapy describes the therapy received from the time
mab therapy [182, 273]. However, the future challenge is to that the first-line chemotherapy backbone has to be changed,
determine in which patients’ treatment should be deescalated mostly after failure of a first-line strategy, and should be offered to
and in which patients it can be stopped completely. Very limited as many patients as possible. Second-line therapy is normally
and preliminary data exist on maintenance strategies involving proposed for patients with good performance status and adequate
EGFR-antibody therapies, which do not allow any conclusion to organ function, and is dependent on the first-line therapy choice.
be drawn [61, 266, 267, 274]. Second-line therapy with oxaliplatin and irinotecan is known to
For patients receiving FOLFIRI first-line, the optimum dura- be superior to best supportive care and 5-FU [275–277].
tion of induction therapy is unclear and continuation of In patients in whom the initial chemotherapy backbone has
FOLFIRI induction therapy is recommended for at least as long failed, the chemotherapy backbone should be changed [5].
Volume 27 | No. 8 | August 2016 doi:10.1093/annonc/mdw235 | 

by guest
on 05 February 2018
Following failure on 5-FU/leucovorin, patients who can tolerate it randomised phase II trial and also in view of the previously men-
should be switched to an irinotecan or oxaliplatin-containing tioned data on the similar relative benefit of EGFR antibodies in
combination chemotherapy regimen such as FOLFIRI, FOLFOX later lines compared with second line.
or possibly irinotecan/oxaliplatin [278–280]. Patients who receive
FOLFIRI up front should receive FOLFOX and those patients who recommendation 20: second-line combinations with targeted agents.
receive FOLFOX up front should receive an irinotecan-containing
regimen, preferably FOLFIRI, with early evidence of the efficacy of • Patients who are bevacizumab naïve should be considered for
this strategy provided by the trial of Tournigand et al. [281]. treatment with an antiangiogenic (bevacizumab or aflibercept)
Also, as stated previously, treatment with all three cytotoxics second line [I, A]. The use of aflibercept should be restricted
(fluoropyrimidine, irinotecan and oxaliplatin) during the course to combination with FOLFIRI for patients progressing on an
of a patient’s treatment is associated with longer survival [257, oxaliplatin-containing regimen [I, A].
258]. However, when considering current treatment strategies, • Patients who received bevacizumab first line should be consid-
biologicals and predictive markers (e.g. tumour RAS mutation ered for treatment with:
status for EGFR antibody therapy) need to be added to the mix ° Bevacizumab post-continuation strategy [I, A].
which makes the decision-making more complex. ° Aflibercept or ramucirumab (in combination with FOLFIRI)
If bevacizumab was not used as the biological first line, it should when treated in first line with oxaliplatin [I, A].
be considered in second line, as FOLFOX plus bevacizumab was ° EGFR antibodies in combination with FOLFIRI/irinotecan
shown to improve OS compared with FOLFOX alone in a phase for patients with RAS wild-type (BRAF wild-type) disease
III trial [282] and confirmed in subsequent studies [283–285]. • Relative benefit of EGFR antibodies is similar in later
Data from the randomised phase III TML study [283], and from lines compared with second line [II, A].
the BEBYP study [286], showed continuation of bevacizumab
treatment with second-line chemotherapy to benefit patients pre- • Patients who are fast progressors on first-line bevacizumab-
viously treated with bevacizumab, suggesting that patients treated containing regimens should be considered for treatment with
first line with bevacizumab can benefit from subsequent therapies aflibercept or ramucirumab (only in combination with
that target VEGF. The anti-angiogenic fusion protein aflibercept FOLFIRI) [II, B], and—in the case of patients with RAS wild-
has been shown to confer a survival advantage when added to type disease and no pre-treatment with anti-EGFR therapy—
FOLFIRI in patients previously progressing on a prior oxaliplatin- EGFR antibody therapy, preferably in combination with
containing regimen compared with FOLFIRI plus placebo [287]. chemotherapy [II, B].
A benefit has also been reported for patients treated with afliber-
cept who had received prior bevacizumab therapy [288]. Recently,
a similar OS benefit has been reported for the anti-VEGFR2 anti- third line
body ramucirumab, also in combination with FOLFIRI, as Both cetuximab and panitumumab have shown efficacy in the
second-line treatment following first-line treatment with a fluoro- third-line/salvage-therapy setting in patients with RAS wild-
pyrimidine, oxaliplatin and bevacizumab [289]. In total, four trials type tumours [293–295], and are equally active as single agents
have reported a gain in OS by the addition of an antiangiogenic [296]. The combination of cetuximab with irinotecan is more
compound, irrespective of the various first-line regimens [282, active than cetuximab alone, in irinotecan refractory patients
283, 287, 289]. [293]. Any activity in patients with BRAF-mutant tumours, if
Both EGFR antibodies, cetuximab and panitumumab, have active at all, seems to be limited to patients with chemorefractory
been shown to increase RR and PFS, but not OS when combined mCRC [64, 65]. There is no unequivocal evidence to support
with irinotecan-containing therapy in the second-line treatment administration of the alternative EGFR antibody, if a patient is
setting [47, 65, 290] and can be considered if not used previously refractory to the other.
in the treatment of patients with RAS wild-type disease. However, The multi-targeted kinase inhibitor regorafenib has reported
generally, there is a similar relative benefit when cetuximab (and activity versus placebo plus best supportive care in two phase III
panitumumab) is used in later lines compared with second line, trials [297, 298]. Regorafenib has demonstrated a significant
which was confirmed in a recent randomised trial [291]. improvement in OS (and maintenance of QoL over time) in
No randomised phase III studies have been carried out which patients pre-treated with all available cytotoxics and bevacizu-
compare the different biologicals available, specifically in patients mab and EGFR antibodies [297], and can be proposed as a stan-
who are fast progressors (PFS <3–4 months) on a first-line bevaci- dard treatment in this setting [I, B]. However, some concerns
zumab-containing regimen. In view of the inclusion criteria of over safety have raised some doubt as to whether the labelled
the bevacizumab, aflibercept and ramucirumab trials [283, 287, dose (160 mg/day day 1–21 q4 weeks) is the optimal dose. In
289], aflibercept and ramucirumab may be considered for the reality, it seems that in some regions many physicians start with
treatment of patients with RAS mutant or unclassified tumours, a lower dose and then increase the dose to the approved dose if
and EGFR inhibitors for patients with RAS wild-type disease, no toxicity is observed. Frequent and close monitoring for
especially when a higher RR is desired. The toxicity profiles of regorafenib toxicity is recommended.
bevacizumab, aflibercept, ramucirumab and cetuximab/panitu- Recently, an oral agent that combines trifluridine and tipiracil
mumab also need to be considered. A randomised phase II trial, hydrochloride, has been shown to be effective in the treatment of
however, suggested no difference in OS or in PFS between bevaci- patients with refractory mCRC, leading to a significant survival
zumab and panitumumab when combined with FOLFIRI [292]. benefit that is similar to that of regorafenib, but with limited toxi-
This trial does not influence the recommendations, because it is a city and is therefore a potential new option [299, 300].
 | Van Cutsem et al. Volume 27 | No. 8 | August 2016

by guest
on 05 February 2018
recommendation 21: third-line therapy. • Trifluridine/tipiracil is recommended for patients pre-
treated with fluoropyrimidines, oxaliplatin, irinotecan, bev-
• In RAS wild-type and BRAF wild-type patients not previously
acizumab and in RAS wild-type patients with EGFR antibo-
treated with EGFR antibodies, cetuximab or panitumumab
dies [I, B].
therapy should be considered
° Cetuximab and panitumumab are equally active as single consensus recommendations on the
agents [I, A]. use of cytotoxics and biologicals in the
° The combination of cetuximab with irinotecan is more active first- and subsequent-line treatment
than cetuximab alone, in irinotecan refractory patients [II, B].
° There is no unequivocal evidence to administer the alterna- of patients with mCRC
tive EGFR antibody, if a patient is refractory to one of the Within the expert panel, the consensus was that the initial cate-
EGFR antibodies [I, C]. gorisation of patients with mCRC for treatment should be made
• Regorafenib is recommended in patients pre-treated with according to whether they were clinically ‘fit’ or ‘unfit’ and sub-
fluoropyrimidines, oxaliplatin, irinotecan, bevacizumab and sequently according to treatment goal. There was also the recog-
in RAS wild-type patients with EGFR antibodies [I, B]. nition that there may be an intermediate category of patients
° Regorafenib is superior to placebo in terms of OS, although who are ‘unfit’ but may benefit from treatment. It was also
there are toxicity concerns in frail patients. recognised that all treatment decisions involving patients
Assessment of clinical condition of the patient
Fita Unfita (but may be suitable) Unfita
FP+bevacizumab: reduced BSC

GOAL
dose doublet; anti-EGFR
Patients with clearly OMD Cytoreduction (Shrinkage)** Disease control (control of progression)
resectable metastases See
figure 2
MOLECULAR PROFILE MOLECULAR PROFILE
Surgery alone
surgery with perioperative RAS wt RAS mt BRAF mt RAS wt RAS mt BRAF mt
postoperative CT
CT doublet + Combination CT CT triplet + CT doublet + CT doublet + CT triplet +/–

anti-EGFR + bevacizumab bevacizumab biological agent bevacizumab bevacizumab
Re-evaluation/assessment of response every 2 months* Re-evaluation/assessment of response every 2–3 months*
GOAL
Progressive
Progressive disease Second-line Second-line
disease
Surgery Cytoreduction (Shrinkage)** Disease control
Continue;
maintenance;
or pause
Continue Continue;
maintenance;
or pause
Figure 4. Zurich treatment algorithm. BSC, best supportive care; CT, chemotherapy; EGFR, epidermal growth factor receptor; FP, fluoropyrmidine; mt,
mutant, NED, no evidence of disease; OMD, oligometastatic disease; wt, wild-type. aPatients assessed as fit or unfit according to medical condition not due to
malignant disease. *After two re-evaluations, consider maintenance. **(A) Includes two subgroups: (1) those for whom intensive treatment is appropriate with
the goal of cytoreduction (tumour shrinkage) and conversion to resectable disease; (2) those who need an intensive treatment, although they will never make it
to resection or LAT, since they need a rapid reduction of tumour burden because of impending clinical threat, impending organ dysfunction, severe symptoms.
Volume 27 | No. 8 | August 2016 doi:10.1093/annonc/mdw235 | 

by guest
on 05 February 2018
special articles
 | Van Cutsem et al.
Table 7. Systemic therapy choices according to the Zurich treatment algorithm for patients with unresectable metastatic disease (excluding those with oligometastatic disease)a
Category Fit patientsb Unfitb
May be unfit Unfit
Treatment goal Cytoreduction (tumour shrinkage) Disease control (control of progression) Palliation
Molecular profile RAS wt RAS mt BRAF mt RAS wt RAS mt BRAF mt Any Any
First-line
Preferred choice (s) CT doublet + EGFR antibodyc,d CT doublet + bevacizumab FOLFOXIRI + bevacizumab CT doublet + bevacizumab CT doublet FOLFOXIRI ± bevacizumab FP + bevacizumab BSC
or + bevacizumab
CT doublet + EGFR
antibodyc
Second choice FOLFOXIRI ± bevacizumab FOLFOXIRI + bevacizumab CT doublet + bevacizumab FP + bevacizumab CT doublet + bevacizumab Reduced-dose CT doublet —
Third choice CT doublet + bevacizumab FOLFOXIRI FOLFOXIRI If RAS wt may consider —

EGFR antibody therapy
Maintenance
Preferred choice FP + bevacizumabe FP + bevacizumab FP + bevacizumab FP + bevacizumabe FP + bevacizumab FP + bevacizumab FP + bevacizumab —
Second choice Pause Pause Pause Pause Pause Pause FP —

Second line
Preferred choice(s) CT doublet + bevacizumab CT doublet + bevacizumab CT doublet + bevacizumab CT doublet + bevacizumab CT doublet CT doublet + bevacizumab —
or + bevacizumab
CT doublet + EGFR antibody
Second choice CT doublet + EGFR antibodyc,f FOLFIRI + aflibercept/ FOLFIRI + aflibercept/ FOLFIRI + aflibercept/ FOLFIRI + aflibercept/ FOLFIRI + aflibercept/ -
or ramucirumab ramucirumab ramucirumab ramucirumab ramucirumab
FOLFIRI + aflibercept/
ramucirumab
Third line
Preferred choice (s) CT doublet + EGFR antibodyc,f Regorafenib or trifluridine/ Regorafenib or trifluridine/ CT doublet + EGFR antibodyc Regorafenib or Regorafenib or —
or tipiracil tipiracil or irinotecan + cetuximab trifluridine/tipiracil trifluridine/tipiracil
irinotecan + cetuximabf
Second choice EGFR antibody monotherapyf EGFR antibody monotherapyf —
Third choice Regorafenib or trifluridine/ Regorafenib or trifluridine/ —

tipiracil tipiracil
BSC, best supportive care; CT, chemotherapy; EGFR, epidermal growth factor receptor; FP, fluoropyrmidine; FOLFOXIRI, infusional 5-fluorouracil, leucovorin, irinotecan and oxaliplatin; mt, mutant; wt,
wild-type.
Volume 27 | No. 8 | August 2016
a
Cross references to Figure 4.
b
Patients assessed as fit or unfit according to medical condition not due to malignant disease.
c
EGFR antibodies: cetuximab and panitumumab.
d
In patients in need of a rapid reduction in tumour burden because of impending clinical threat, impending organ dysfunction and severe disease-related symptoms, a similar strategy can be proposed, although
Annals of Oncology
the consensus on the preferred treatment of choice was less strong. For those patients who have RAS wild-type disease, a cytotoxic doublet plus an EGFR antibody is a preferred option, although a cytotoxic
doublet plus bevacizumab is an equally valid alternative. A cytotoxic triplet plus or minus bevacizumab may be an alternative for selected, very fit and motivated patients.
e
In patients where a bevacizumab-containing regimen was started. In patients where a cetuximab-containing combination was started: pause or less intensive regimen.
f
If not yet pretreated with an EGFR antibody.

by guest
on 05 February 2018
categorised as being clinically fit must be made at a tumour ablative methods after response to systemic therapy are the
board meeting by an MDT, informed by the appropriate mole- goals, intensive treatment (cytotoxics in combination with
cular analyses. The appropriate molecular analyses are to be biologicals) should be the treatment of choice for first-line
carried out at the time of initial diagnosis of mCRC and should induction therapy.
comprise a full analysis of tumour RAS mutational status However, uncertainty remains as to which is the best combi-
(KRAS: exon 2, 3 and 4 and NRAS: exon 2, 3 and 4) with a nation to use for patients stratified according to the molecular
simultaneous analysis of tumour BRAF mutational status, con- profile of their disease, with the treatment recommendations of
ducted in a validated laboratory/testing centre, to facilitate the the panel presented in Figure 4 and below, with the alternative
best diagnostic and prognostic decision making possible. All options presented in Table 7.
patients considered for systemic therapy should be stratified
according to whether their tumours were RAS wild-type, RAS
mutant or BRAF mutant. Consensus recommendation for patients where cytoreduction
The expert panel consensus outlined below led to the genera- with ‘conversion’ and/or the integration of local ablative treat-
tion of the Zurich treatment algorithm (Figure 4). ment is the goal
• A1a. For those patients who have RAS wild-type disease, a
allocation to first-line treatment cytotoxic doublet plus an EGFR antibody should be the treat-
1) Fit patients with resectable metastatic disease ment of choice.
• A1b. For those patients with RAS-mutant disease, a cytotoxic
The assignment of ‘fit’ patients to surgery is dependent on doublet plus bevacizumab or cytotoxic triplet plus bevacizu-
surgical evaluation within an MDT according to technical mab are the preferred options.
and prognostic criteria, as described above, and consideration • A1c. Patients should be revaluated for their disease status
of any contraindications for resection; as outlined in Table 2. every 2 months in order to ensure that resectable patients are
In the case of ‘fit’ patients with initially resectable metastatic not over-treated.
disease, the recommendation is (Recommendation 12 this docu- • A1d. If, after the first re-evaluation at 2 months, there is evi-
ment) that they can either be referred immediately for potentially dence of tumour shrinkage, patients should be recommended
curative surgery or for perioperative chemotherapy (FOLFOX) for either potentially curative surgery or the most suitable
[163, 164], dependent on the available prognostic information LAT strategy—with a view to eliminating all evidence of
and surgical considerations. disease (i.e. R0 resection, NED).
• A1e. If no response is evident at first evaluation, it is suggested
2) Fit patients with unresectable metastatic disease that the cytotoxic doublet is changed in order to maximise the
For fit patients with unresectable metastatic disease, the treat- chance of resection [5].
ment goals are either cytoreduction (A) or disease control and • A1f. Where there is evidence for cytoreduction, but the
hence prolongation of survival (Figure 4 and Table 6). patients are not suitable for surgery, they should continue on
combination chemotherapy plus the appropriate biological
A. cytoreduction. dependent on RAS and BRAF mutation status as indicated in
Figure 4.
1) For those patients with potentially resectable mCRC for • A1g. Where there is evidence of disease progression, patients
whom cytoreduction and conversion to resectable disease should continue to second-line therapy (Figure 5).
and/or for patients with OMD, the integration of local or • A1h. Toxicity might also require a change to an alternative
regimen.
Reasons for first-line discontinuation 2) A specific group of patients who need an intensive treatment,
although neither resection nor LAT of metastases are a treat-
ment goal: are patients in need of a rapid reduction in tumour
burden because of impending clinical threat, impending
Tumour progression Excessive toxicity Patient/doctor decision
(clinical resistance) (intolerance) (stop CT) +/– maintenance organ dysfunction and severe disease-related symptoms. In
these patients, a similar strategy can be proposed, although
the consensus on the preferred treatment of choice was less
strong.
Consider second-line CT Early progession Late progression
(PS, organ function) (during 1–2 months (occurring after at least
of discontinuation) 2 months of treatment)
Consensus recommendation for patients where cytoreduction is
needed because of aggressive biology and/or risk of developing or
existing severe symptoms
Progression Reintroduction of
first-line CT • A2a. For those patients who have RAS wild-type disease, a
cytotoxic doublet plus an EGFR antibody is a preferred
Figure 5. Maintenance and second-line treatment options. CT, chemother- option, although a cytotoxic doublet plus bevacizumab is an
apy; PS, performance status. equally valid alternative. A cytotoxic triplet plus or minus
Volume 27 | No. 8 | August 2016 doi:10.1093/annonc/mdw235 | 

by guest
on 05 February 2018

LAT (including surgery and the management of patients with OMD)
Recommendation 1: Tissue handling

• Fixation with 10% neutral buffered formalin (4% formaldehyde) is recommended [V, A]
• Fixation time should be no less than 6 h, and no greater than 48 h in duration. In the case of microwave-enhanced fixation, the quality of both nucleic
acids and proteins must be verified [IV, A]
• Sections for biomarker testing should ideally be cut immediately before analysis [IV, A]
Recommendation 2: Selection of specimens for biomarker testing
• The primary pathologist should review all available tumour specimens to select those that are most suitable for biomarker analyses [IV, A]
• Enrichment of samples by macro-dissection to maximise tumour cell content (>50%) before DNA extraction is recommended [III, A]
Recommendation 3: Tissue selection
• Tissue from either the primary tumour or a liver metastasis may be used for RAS mutation testing [III, A]
• Other metastatic sites such as lymph node or lung metastases may be used only if primary tumour or liver metastases samples are not available [II, B]
Recommendation 4: RAS testing
• RAS mutational status is a negative predictive biomarker for therapeutic choices involving EGFR antibody therapies in the metastatic disease setting [I, A]
° RAS testing should be carried out on all patients at the time of diagnosis of mCRC [I, A]
• RAS testing is mandatory before treatment with the EGFR-targeted monoclonal antibodies cetuximab and panitumumab [I, A]
• A network of arrangements should be established to ensure the rapid and robust transit of tissue samples from referral centres to testing laboratories, to
minimise the turnaround time and avoid delays in having this information available for all patients with mCRC
• Primary or metastatic colorectal tumour tissue can be used for RAS testing (see also Recommendation 3)
• RAS analysis should include at least KRAS exons 2, 3 and 4 (codons 12, 13, 59, 61, 117 and 146) and NRAS exons 2, 3 and 4 (codons 12, 13, 59, 61 and 117)
• Turnaround time for RAS testing (expanded RAS analysis) should be ≤7 working days from the time of receipt of the specimen by the testing laboratory
to the time of issuing of the final report, for >90% of specimens.
• Validation (or verification, where more applicable) of RAS testing assays should be carried out and recorded before implementation in clinical use.
Laboratory audit mechanisms should be in place
• Laboratories providing RAS testing of colorectal tumours should demonstrate their successful participation in a relevant external quality assessment
scheme, and be appropriately accredited.
Recommendation 5: BRAF testing
• Tumour BRAF mutation status should be assessed alongside the assessment of tumour RAS mutational status for prognostic assessment (and/or
potential selection for clinical trials) [I, B]
Recommendation 6: Microsatellite instability testing
• MSI testing in the metastatic disease setting can assist clinicians in genetic counselling [II, B]
• MSI testing has strong predictive value for the use of immune check-point inhibitors in the treatment of patients with mCRC [II, B]
Recommendation 7: Biomarkers of chemotherapy sensitivity and toxicity
• DPD testing before 5-FU administration remains an option but is not routinely recommended [II, D]
• UGT1A1 phenotyping remains an option and should be carried out in patients with a suspicion of UGT1A1 deficiency as reflected by low conjugated
bilirubin and in patients where an irinotecan dose of >180 mg/m2 per administration is planned [95] [III, C]
• ERCC1 expression cannot be recommended for use as a biomarker for treatment decisions involving the use of oxaliplatin in routine clinical practice, but
could be included prospectively in clinical trials [III, D]
• TS activity and TSER genotyping are not recommended for use in clinical practice [II, D]
Recommendation 8: Emerging biomarkers not recommended for routine patient management outside of a clinical trial setting:
• Detection of mutations in PIK3CA exon 20 [II, D]
• Evaluation of PTEN loss by IHC [V, D]
• Evaluation of the levels of the EGFR ligands amphiregulin, epiregulin and transforming growth factor-α [II, D]
• Evaluation of levels of EGFR protein expression [II, E]
• Evaluation of EGFR amplification and copy number and EGFR ectodomain mutations [IV, D]
• Evaluation of HER2 amplification or HER2 activating mutations
• Evaluation of HER3, and MET receptor overexpression [IV, D]
Recommendation 9: Emerging technologies
• Although CTC number correlates with prognosis in patients with mCRC, the clinical utility of CTC assessments is not yet clear and therefore cannot be
recommended [IV, D]
• The utility of liquid ctDNA biopsies to guide treatment decisions is currently under investigation in clinical trials, but cannot yet be recommended in
routine practice [V, D]
• Whole genome, whole exome and whole transcriptome analysis should be carried out only in a research setting [V, D]
Continued
 | Van Cutsem et al. Volume 27 | No. 8 | August 2016

by guest
on 05 February 2018
Table 8. Continued
Recommendation 10: OMD

• For patients with OMD, systemic therapy is the standard of care and should be considered as the initial part of every treatment strategy (exception:
patients with single/few liver or lung lesions, see below).
• The best local treatment should be selected from a ‘toolbox’ of procedures according to disease localisation, treatment goal (‘the more curative the more
surgery’/higher importance of local/complete control), treatment-related morbidity and patient-related factors such as comorbidity/ies and age [IV, B]
Recommendation 11: Imaging in the identification and management of disease
• Imaging should comprise first an abdominal/pelvic and thoracic CT scan and, in the case of doubt, a second method such as US (CEUS), MRI or PET/
CT scan depending on the localisation of the metastases. US may be helpful to characterise liver metastases, MRI liver, peritoneal or pelvic metastases
and PET/CT extrahepatic disease [IV, B]
• A stepwise imaging approach is the recommended policy, in relation to the therapeutic possibilities, rather than the use of all imaging modalities in all
patients [V, B]
Recommendation 12: Perioperative treatment
• Both, technical criteria for resection and prognostic considerations define the need for systemic perioperative therapy [IV, B]
• In patients with clearly resectable disease and favourable prognostic criteria, perioperative treatment may not be necessary and upfront resection is
justified [I, C; consensus >75%]
• In patients with technically resectable disease where the prognosis is unclear or probably unfavourable, perioperative combination chemotherapy
(FOLFOX or CAPOX) should be administered [I, B; consensus >75%]
• Targeted agents should not be used in resectable patients where the indication for perioperative treatment is prognostic in nature [II, E]
• In situations where the criteria for prognosis and resectability are not sharply defined, perioperative therapy should be considered (as part of a
continuum of treatment option) [IV, B] (Figure 2). Patients with synchronous onset of metastases should be allocated to this group and therapeutic
pathway
• In patients with favourable oncological and technical (surgical) criteria, who have not received perioperative chemotherapy, there is no strong evidence
to support the use of adjuvant chemotherapy [II, C], whereas patients with unfavourable criteria may benefit from adjuvant treatment [III, B]
• In patients who have not received any previous chemotherapy, adjuvant treatment with FOLFOX or CAPOX is recommended (unless patients were
previously recently exposed to oxaliplatin-based adjuvant chemotherapy) [IV, B]
• Decision-making should include patients’ characteristics and preferences [IV, B]
Recommendation 13: Conversion therapy
• In potentially resectable patients (if conversion is the goal), a regimen leading to high response rates and/or a large tumour size reduction (shrinkage) is
recommended [II, A]
• There is uncertainty surrounding the best combination to use as only few trials have addressed this specifically:
° In patients with RAS wild-type disease, a cytotoxic doublet plus an EGFR antibody seems to have the best benefit risk/ratio, although the combination
of FOLFOXIRI plus bevacizumab may also be considered and, to a lesser extent, a cytotoxic doublet plus bevacizumab [II, A]
° In patients with RAS mutant disease: a cytotoxic doublet plus bevacizumab or FOLFOXIRI plus bevacizumab [II, A]
• Patients must be re-evaluated regularly in order to prevent the overtreatment of resectable patients as the maximal response is expected to be achieved
after 12–16 weeks of therapy in most patients.
Recommendation 14: Ablative techniques
• Despite the lack of more available prospective data, this strategic treatment approach should be evaluated and pursued further in suitable patients [II, B]
Recommendation 15: Local ablation techniques
• In patients with unresectable liver metastases only, or OMD, local ablation techniques such as thermal ablation or high conformal radiation techniques
(e.g. SBRT, HDR-brachytherapy) can be considered. The decision should be taken by an MDT based on local experience, tumour characteristics and
patient preference [IV, B]
• In patients with lung only or OMD of the lung, ablative high conformal radiation or thermal ablation may be considered if resection is limited by
comorbidity, the extent of lung parenchyma resection, or other factors [IV, B]
• SBRT is a safe and feasible alternative treatment for oligometastatic colorectal liver and lung metastases in patients not amenable to surgery or other
ablative treatments [IV, B]
• RFA can be used in addition to surgery with the goal of eradicating all visible metastatic sites [II, B]
Recommendation 16: Embolisation
• For patients with liver-limited disease failing the available chemotherapeutic options
° Radioembolisation with yttrium-90 microspheres should be considered [II, B]
° Chemoembolisation may be also considered as a treatment option [IV, B]
• Radioembolisation (and chemoembolisation) of CLM in earlier treatment lines may be interesting as ‘consolidation treatment’ but should be limited to
clinical trials
Continued
Volume 27 | No. 8 | August 2016 doi:10.1093/annonc/mdw235 | 

by guest
on 05 February 2018
Table 8. Continued
Recommendation 17: Cytoreductive surgery and HIPEC

• Complete cytoreductive surgery and HIPEC can be considered for patients with limited peritoneal metastases in centres which are very experienced in
the use of HIPEC [III, B]
Treatment of metastatic disease
Recommendation 18: First-line systemic therapy combinations according to targeted agent used
• Biologicals (targeted agents) are indicated in the first-line treatment of most patients unless contraindicated [I, A]
• The VEGF antibody bevacizumab should be used in combination with:
° The cytotoxic doublets FOLFOX/CAPOX/FOLFIRI
° The cytotoxic triplet FOLFOXIRI in selected fit and motivated patients where cytoreduction (tumour shrinkage) is the goal—and potentially also in fit
patients with tumour BRAF mutations [II, B]
° Fluoropyrimidine monotherapy in patients unable to tolerate aggressive treatment [I, B]
• EGFR antibodies should be used in combination with:
° FOLFOX/FOLFIRI [I, A]
° Capecitabine-based and bolus 5-FU based regimens should not be combined with EGFR antibodies [I, E]
Recommendation 19: Maintenance therapy
• Patients receiving FOLFOX or CAPOX plus bevacizumab-based therapy as induction therapy, should be considered for maintenance therapy after
6 cycles of CAPOX or 8 cycles of FOLFOX. The optimal maintenance treatment is a combination of a fluoropyrimidine (plus bevacizumab).
Bevacizumab as monotherapy is not recommended [I, B]
• Patients receiving FOLFIRI can continue on induction therapy—at a minimum—for as long as tumour shrinkage continues and the treatment is
tolerable [V, B]
• For patients receiving initial therapy with FOLFOXIRI plus or minus bevacizumab, a fluoropyrimidine plus bevacizumab may be considered as
maintenance therapy (as was done in the pivotal trials examining FOLFOXIRI)
• For patients receiving initial therapy with single-agent fluoropyrimidine (plus bevacizumab), induction therapy should be maintained [V, A].
Individualisation and discussion with the patient is essential [V, A]
• Initial induction therapy or a second-line therapy have to be reintroduced at radiological or first signs of symptomatic progression. If a second-line therapy is
chosen, re-introduction of the initial induction treatment should be a part of the entire treatment strategy as long as no residual toxicity is present [III, B]
Recommendation 20: Second-line combinations with targeted agents
• Patients who are bevacizumab naïve should be considered for treatment with an antiangiogenic (bevacizumab or aflibercept) second-line [I, A]. The use
of aflibercept should be restricted to combination with FOLFIRI for patients progressing on an oxaliplatin-containing regimen [I, A]
• Patients who received bevacizumab first-line should be considered for treatment with:
° Bevacizumab post-continuation strategy [I, A]
° Aflibercept or ramucirumab (in combination with FOLFIRI) when treated in first line with oxaliplatin [I, A]
° EGFR antibodies in combination with FOLFIRI/irinotecan for patients with RAS wild-type (BRAF wild-type) disease
• Relative benefit of EGFR antibodies is similar in later lines compared with second-line [II, A]
• Patients who are fast progressors on first-line bevacizumab-containing regimens should be considered for treatment with aflibercept or ramucirumab
(only in combination with FOLFIRI) [II, B], and—in the case of patients with RAS wild-type disease and no pre-treatment with anti-EGFR therapy—
EGFR antibody therapy, preferably in combination with chemotherapy [II, B]
Recommendation 21: Third-line therapy
• In RAS wild-type and BRAF wild-type patients not previously treated with EGFR antibodies, cetuximab or panitumumab therapy should be considered
° Cetuximab and panitumumab are equally active as single agents [I, A]
° The combination of cetuximab with irinotecan is more active than cetuximab alone, in irinotecan refractory patients [II, B]
° There is no unequivocal evidence to administer the alternative EGFR antibody, if a patient is refractory to one of the EGFR antibodies [I, C].
• Regorafenib is recommended in patients pre-treated with fluoropyrimidines, oxaliplatin, irinotecan, bevacizumab and in RAS wild-type patients with
EGFR antibodies [I, B]
° Regorafenib is superior to placebo in terms of OS, although there are toxicity concerns in frail patients.
• Trifluridine/tipiracil is a new option for patients pre-treated with fluoropyrimidines, oxaliplatin, irinotecan, bevacizumab and in RAS wild-type patients
with EGFR antibodies [I, B]
Consensus recommendations on the use of cytotoxics and biologicals in the first- and subsequent-line treatment of patients with mCRC
Consensus recommendation for patients where cytoreduction with ‘conversion’ and/or the integration of local ablative treatment is the goal
• A1a. For those patients who have RAS wild-type disease, a cytotoxic doublet plus an EGFR antibody should be the treatment of choice
• A1b. For those patients with RAS mutant disease, a cytotoxic doublet plus bevacizumab or cytotoxic triplet plus bevacizumab are the preferred options
• A1c. Patients should be revaluated for their disease status every 2 months in order to ensure that resectable patients are not over-treated
• A1d. If, after the first re-evaluation at 2 months, there is evidence of tumour shrinkage, patients should be recommended for either potentially curative
surgery or the most suitable LAT strategy—with a view to eliminating all evidence of disease (i.e. R0 resection, NED)
Continued
 | Van Cutsem et al. Volume 27 | No. 8 | August 2016

by guest
on 05 February 2018
Table 8. Continued
• A1e. If no response is evident at first evaluation, it is suggested that the cytotoxic doublet is changed in order to maximise the chance of resection [5]
• A1f. Where there is evidence for cytoreduction but the patients are not suitable for surgery, they should continue on combination chemotherapy plus the
appropriate biological dependent on RAS and BRAF mutation status as indicated in Figure 4.
• A1g. Where there is evidence of disease progression, patients should continue to second-line therapy (Figure 5).
• A1h. Toxicity might also require a change to an alternative regimen.
Consensus recommendation for patients where cytoreduction is needed because of aggressive biology and/or risk of developing or existing severe symptoms
• A2a. For those patients who have RAS wild-type disease, a cytotoxic doublet plus an EGFR antibody is a preferred option, although a cytotoxic doublet
plus bevacizumab is an equally valid alternative. A cytotoxic triplet plus or minus bevacizumab may be an alternative for selected, very fit and motivated
patients
• A2b. For those patients with RAS mutant disease, a cytotoxic doublet plus bevacizumab is the preferred option. A cytotoxic triplet plus or minus
bevacizumab may be an alternative for selected, very fit and motivated patients
• A2c. Patients should be revaluated for their disease status every 2 months
• A2d. Treatment should not be changed in patients without tumour progression and not suffering from major toxicity
Consensus recommendation for patients where disease control is the goal
• B1a. For these patients, a cytotoxic doublet in combination with bevacizumab or in patients with RAS wild-type tumours, a cytotoxic doublet plus an
EGFR antibody are recommended
• B1b. Patients should be revaluated for their disease status every 2–3 months
• B1c. In patients with a good response or at least disease control, active maintenance therapy should be considered. A fluoropyrimidine plus bevacizumab
is the preferred option if they started their treatment with a cytotoxic doublet plus bevacizumab
• B1d. Where there is evidence of disease progression, patients should continue to second-line therapy (Figure 5)
• B1e. Toxicity might also require a change to second-line therapy.
bevacizumab may be an alternative for selected, very fit and tumours a cytotoxic doublet plus an EGFR antibody are
motivated patients. recommended.
• A2b. For those patients with RAS-mutant disease, a cytotoxic • B1b. Patients should be revaluated for their disease status
doublet plus bevacizumab is the preferred option. A cytotoxic every 2–3 months.
triplet plus or minus bevacizumab may be an alternative for • B1c. In patients with a good response or at least disease
selected, very fit and motivated patients. control, active maintenance therapy should be considered. A
• A2c. Patients should be revaluated for their disease status fluoropyrimidine plus bevacizumab is the preferred option if
every 2 months. they started with a cytotoxic doublet plus bevacizumab.
• A2d. Treatment should not be changed in patients without • B1d. Where there is evidence of disease progression, patients
tumour progression and not suffering from major toxicity. should continue to second-line therapy (Figure 5).
• B1e. Toxicity might also require a change to second-line
B. disease control. The recommendation for fit patients, for therapy.
whom surgery or induction therapy plus LAT are not options and,
where the treatment goal is long-term disease control without
3) Unfit patients
symptomatic toxicity, is that they should receive chemotherapy
(usually a doublet) plus bevacizumab or a cytotoxic doublet plus Patients with mCRC who are assessed as being unfit for any
EGFR antibody therapy as an alternative option for patients with treatment should receive best supportive care. For the other
RAS wild-type disease. patients in this group who are unfit, but may be suitable for
Patients should be re-evaluated every 2–3 months. Where treatment, physician experience should guide treatment choice
there is evidence of good disease control, patients should con- with potential treatment options being capecitabine plus bevaci-
tinue on therapy and if after two re-evaluations, a patient has zumab or a reduced-dose doublet of cytotoxics.
achieved a good response/disease control, active maintenance In the case of unfit patients with RAS wild-type disease where
therapy with chemotherapy might be considered (see there is the fear that they may be receiving their last line of treat-
Recommendation 19). Patients with progressive disease or ment, anti-EGFR therapy can be considered (Figure 4).
excessive toxicity should progress to second-line therapy (see
Figure 5 and Table 7). 4) Treatment of elderly patients with mCRC
Consensus recommendation for patients where disease control is Fit older patients should be treated with systemic combination
the goal chemotherapy plus targeted agents as they derive the same
benefit as younger patients [301]. For older patients unfit for
• B1a. For these patients, a cytotoxic doublet in combination standard combination chemotherapy (with or without targeted
with bevacizumab or in patients with RAS wild-type agents), less intensive therapies including capecitabine plus
Volume 27 | No. 8 | August 2016 doi:10.1093/annonc/mdw235 | 

by guest
on 05 February 2018
bevacizumab or reduced dose fluoropyrimidine plus oxaliplatin has reported member of speaker’s bureau for Merck-Serono,
or irinotecan are appropriate first-line options [301]. Taiho, Chugai and Takeda. PO has reported honoraria for lec-
tures, advisory boards or travel grants from Amgen, Bayer,
note Celgene, Merck-Serono, Nordic Drugs, Roche, Sanofi Oncology,
Prime Oncology and Finnish medical associations. DP has
A summary of recommendations is provided in Table 8.
reported advisory boards and lectures for Roche, Merck-Serono,
Amgen. GP has reported research grants from Roche, Sanofi,
funding Amgen. PP has reported advisory boards, symposia, lectures for
All costs relating to the consensus conference were covered from Amgen, Bayer, Celgene, Merck-Serono, Merck Sharp & Dohme,
the European Society for Medical Oncology central funds. Nordic Drugs, Roche, Sanofi, Taiho. TJP has reported advisory
There was no external funding of the event or manuscript pro- boards for Amgen, Roche, Merck. CP has reported advisory role
duction. for Merck-Serono, Roche, Bayer, Nordic Pharma, Amgen. JR
has reported research funding, speaker fees, advisory board
member for Bayer, Simtex, Merck-Serono. RS has reported
disclosure member of advisory board and speaker’s bureau for Merck-
Serono and Amgen; research funding from Roche, Amgen and
RA has reported honoraria from Amgen, Merck-Serono, Sanofi,
Merck-Serono. WS has reported research funding and honoraria
Roche. EAA has reported advisory role for Amgen, Bayer,
for consultancy activities from Roche, Merck, Bayer, Amgen. HJS
Celgene, Merck, Roche, Sanofi. DAr has reported honoraria/
has reported support for clinical trials from Roche; advisory boards
consultancy for Roche, Merck-Serono, Bayer, Amgen; research
for Amgen, Roche and Bayer. AS has reported advisory boards and
funding from Roche. ABa has reported advisory boards for
satellite symposia for Roche, Merck-Serono, Amgen, Takeda, Bayer,
Horizon Discovery, Trovagene and Biocartis; stock holder for
Sanofi, Lilly, Celgene. JTab has reported consultancy/advisory role
Horizon Discovery. ABe has reported research support directly
for Amgen, Bayer, Boehringer Ingelheim, Celgene, Chugai, Lilly,
to institution from Amgen, Genentech, Astellas/Aveo, Gilead,
MSD, Merck-Serono, Novartis, Roche, Sanofi, Symphogen, Taiho
Bayer; scientific consultancy for Genentech, Lilly/ImClone,
and Takeda. JTai has reported advisory, research grants, speaker for
Sanofi, Bayer, Bristol-Myers Squibb; no stock and no speakers’
Roche, Merck, Amgen, Sanofi, Bayer. EVC has reported research
bureau. GB has reported research grants from Bayer, Celgene,
grants Bayer, Boehringer, Amgen, Celgene, Ipsen, Lilly, Merck,
Lilly, Roche, Abbvie; speaker for Roche, Bayer, Lilly, Amgen,
Novartis, Roche, Sanofi. JHVK has reported research funding from
Janssen, Taiho, Pfizer, Novartis. AC has reported member of
Amgen and Merck-Serono. HW has reported advisory boards and
speaker’s bureau for Roche and Merck-Serono; advisory boards
speakers’ bureau for Sanofi, Bayer, Roche, Merck, Sirtex Medical,
for Merck-Serono, Roche, Amgen, Bayer and Lilly. FC has
Lilly. TY has reported research funding from Pfizer, Eli Lilly,
reported consultancy/advisory roles for Bayer, Roche, Merck-
Sumitomo Dainippon, Taiho, Yakult and Daiichi Sankyo. DAd,
Serono, Sanofi, Lilly, Astra Zeneca. ED-R has reported advisory
ADH, KH, PH, NN, WJGO, AR, ST and AZ have reported no
boards for Roche, Merck-Serono, Sanofi, Bayer, Amgen;
potential conflicts of interest.
research grants from Roche, Merck-Serono, Amgen. J-YD has
reported advisory boards, symposia, lectures for Amgen, Roche,
Merck-Serono, Sanofi, Bayer. MD has reported advisory boards
for Celgene, Merck-Serono, Roche, Amgen, Novartis, Sanofi; references
symposia lectures for Merck-Serono, Roche, Novartis, Celgene; 1. Ferlay J, Steliarova-Foucher E, Lortet-Tieulent J et al. Cancer incidence and
research funding from Roche, Pfizer, Chugai. AF has reported mortality patterns in Europe: estimates for 40 countries in 2012. Eur J Cancer
honoraria for advisory role, speaker at symposia and research 2013; 49: 1374–1403.
funding from Amgen, Bayer, Merck-Serono, Roche, Sanofi and 2. Ferlay J, Soerjomataram I, Dikshit R et al. Cancer incidence and mortality
Lilly. AG has reported research fundings and honoraria for con- worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J
sultancy activities from Genentech, Roche, Eisai, Bayer, Amgen. Cancer 2015; 136: E359–E386.
TG has reported speakers’ bureau member and research funding 3. Schmoll HJ, Van Cutsem E, Stein A et al. ESMO consensus guidelines for
management of patients with colon and rectal cancer. A personalized approach
from Roche, Merck-Serono, Amgen, Sanofi-Aventis, Bayer. VH
to clinical decision making. Ann Oncol 2012; 23: 2479–2516.
has reported research funding from Merck-Serono, Roche,
4. Dykewicz CA, Centers for Disease Control and Prevention, Infectious Diseases
Sanofi, Amgen, Lilly; speaker’s honoraria and advisory boards Society of America, American Society of Blood and Marrow Transplantation.
for Merck-Serono, Roche, Sanofi, Amgen and Lilly. C-HK has Summary of the guidelines for preventing opportunistic infections among
reported speakers’ honoraria and advisory boards for Amgen, hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
Merck, Pfizer, Sanofi; research grants from Novartis, Roche, 139–144.
Celgene. RL has reported advisory boards for Merck, Roche, 5. Van Cutsem E, Cervantes A, Nordlinger B et al. Metastatic colorectal cancer:
Sanofi, Amgen. PL-P has reported honoraria for conferences ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann
Oncol 2014; 25(Suppl 3): iii1–iii9.
and consulting from Amgen, Sanofi, Merck-Serono, Boehringer
6. Neumeister VM, Anagnostou V, Siddiqui S et al. Quantitative assessment of effect
Ingenheim, Astra Zeneca, Roche, Lilly, Integragen. BM has
of preanalytic cold ischemic time on protein expression in breast cancer tissues.
reported research grants from Merck Serono, Hoffmann La J Natl Cancer Inst 2012; 104: 1815–1824.
Roche; consultancy/advisory board for Bayer, Boehringer- 7. Portier BP, Wang Z, Downs-Kelly E et al. Delay to formalin fixation ‘cold ischemia
Ingelheim, Novartis. TM has reported research funding from time’: effect on ERBB2 detection by in-situ hybridization and immunohistochemistry.
Astra Zeneca, GlaxoSmithKline and Almac Diagnostics. KM Mod Pathol 2013; 26: 1–9.
 | Van Cutsem et al. Volume 27 | No. 8 | August 2016

by guest
on 05 February 2018
8. Engel KB, Moore HM. Effects of preanalytical variables on the detection of 30. Garm Spindler KL, Pallisgaard N, Rasmussen AA et al. The importance of KRAS
proteins by immunohistochemistry in formalin-fixed, paraffin-embedded tissue. mutations and EGF61A>G polymorphism to the effect of cetuximab and
Arch Pathol Lab Med 2011; 135: 537–543. irinotecan in metastatic colorectal cancer. Ann Oncol 2009; 20: 879–884.
9. Arber DA. Effect of prolonged formalin fixation on the immunohistochemical reactivity 31. Loupakis F, Pollina L, Stasi I et al. PTEN expression and KRAS mutations on
of breast markers. Appl Immunohistochem Mol Morphol 2002; 10: 183–186. primary tumors and metastases in the prediction of benefit from cetuximab plus
10. Babic A, Loftin IR, Stanislaw S et al. The impact of pre-analytical processing on irinotecan for patients with metastatic colorectal cancer. J Clin Oncol 2009; 27:
staining quality for H&E, dual hapten, dual color in situ hybridization and 2622–2629.
fluorescent in situ hybridization assays. Methods 2010; 52: 287–300. 32. Molinari F, Martin V, Saletti P et al. Differing deregulation of EGFR and
11. Chafin D, Theiss A, Roberts E et al. Rapid two-temperature formalin fixation. downstream proteins in primary colorectal cancer and related metastatic sites
PLoS One 2013; 8: e54138. may be clinically relevant. Br J Cancer 2009; 100: 1087–1094.
12. Kerr KM, Bubendorf L, Edelman MJ et al. Second ESMO consensus conference 33. Perrone F, Lampis A, Orsenigo M et al. PI3KCA/PTEN deregulation contributes to
on lung cancer: pathology and molecular biomarkers for non-small-cell lung impaired responses to cetuximab in metastatic colorectal cancer patients. Ann
cancer. Ann Oncol 2014; 25: 1681–1690. Oncol 2009; 20: 84–90.
13. van Krieken JH, Jung A, Kirchner T et al. KRAS mutation testing for predicting 34. Baldus SE, Schaefer KL, Engers R et al. Prevalence and heterogeneity of KRAS,
response to anti-EGFR therapy for colorectal carcinoma: proposal for an BRAF, and PIK3CA mutations in primary colorectal adenocarcinomas and their
European quality assurance program. Virchows Arch 2008; 453: 417–431. corresponding metastases. Clin Cancer Res 2010; 16: 790–799.
14. Dijkstra JR, Heideman DA, Meijer GA et al. KRAS mutation analysis on low 35. Italiano A, Hostein I, Soubeyran I et al. KRAS and BRAF mutational status in
percentage of colon cancer cells: the importance of quality assurance. Virchows primary colorectal tumors and related metastatic sites: biological and clinical
Arch 2013; 462: 39–46. implications. Ann Surg Oncol 2010; 17: 1429–1434.
15. Tsiatis AC, Norris-Kirby A, Rich RG et al. Comparison of Sanger sequencing, 36. Knijn N, Mekenkamp LJ, Klomp M et al. KRAS mutation analysis: a comparison
pyrosequencing, and melting curve analysis for the detection of KRAS between primary tumours and matched liver metastases in 305 colorectal cancer
mutations: diagnostic and clinical implications. J Mol Diagn 2010; 12: patients. Br J Cancer 2011; 104: 1020–1026.
425–432. 37. Corcoran RB, Ebi H, Turke AB et al. EGFR-mediated re-activation of MAPK
16. Carotenuto P, Roma C, Rachiglio AM et al. Detection of KRAS mutations in signaling contributes to insensitivity of BRAF mutant colorectal cancers to RAF
colorectal carcinoma patients with an integrated PCR/sequencing and real-time inhibition with vemurafenib. Cancer Discov 2012; 2: 227–235.
PCR approach. Pharmacogenomics 2010; 11: 1169–1179. 38. Maughan TS, Adams RA, Smith CG et al. Addition of cetuximab to oxaliplatin-
17. Tol J, Dijkstra JR, Vink-Borger ME et al. High sensitivity of both sequencing and based first-line combination chemotherapy for treatment of advanced colorectal
real-time PCR analysis of KRAS mutations in colorectal cancer tissue. J Cell Mol cancer: results of the randomised phase 3 MRC COIN trial. Lancet 2011; 377:
Med 2010; 14: 2122–2131. 2103–2114.
18. Etienne-Grimaldi MC, Formento JL, Francoual M et al. K-Ras mutations and 39. Swain SM, Baselga J, Kim SB et al. Pertuzumab, trastuzumab, and docetaxel in
treatment outcome in colorectal cancer patients receiving exclusive HER2-positive metastatic breast cancer. N Engl J Med 2015; 372: 724–734.
fluoropyrimidine therapy. Clin Cancer Res 2008; 14: 4830–4835. 40. Dancey JE, Dobbin KK, Groshen S et al. Guidelines for the development and
19. Losi L, Benhattar J, Costa J. Stability of K-ras mutations throughout the natural incorporation of biomarker studies in early clinical trials of novel agents. Clin
history of human colorectal cancer. Eur J Cancer 1992; 28A: 1115–1120. Cancer Res 2010; 16: 1745–1755.
20. Suchy B, Zietz C, Rabes HM. K-ras point mutations in human colorectal 41. Amado RG, Wolf M, Peeters M et al. Wild-type KRAS is required for
carcinomas: relation to aneuploidy and metastasis. Int J Cancer 1992; 52: 30–33. panitumumab efficacy in patients with metastatic colorectal cancer. J Clin Oncol
21. Weber JC, Meyer N, Pencreach E et al. Allelotyping analyses of synchronous 2008; 26: 1626–1634.
primary and metastasis CIN colon cancers identified different subtypes. Int J 42. Karapetis CS, Khambata-Ford S, Jonker DJ et al. K-ras mutations and benefit
Cancer 2007; 120: 524–532. from cetuximab in advanced colorectal cancer. N Engl J Med 2008; 359:
22. Zauber P, Sabbath-Solitare M, Marotta SP, Bishop DT. Molecular changes in the 1757–1765.
Ki-ras and APC genes in primary colorectal carcinoma and synchronous 43. Van Cutsem E, Kohne CH, Hitre E et al. Cetuximab and chemotherapy as initial
metastases compared with the findings in accompanying adenomas. Mol Pathol treatment for metastatic colorectal cancer. N Engl J Med 2009; 360:
2003; 56: 137–140. 1408–1417.
23. Oudejans JJ, Slebos RJ, Zoetmulder FA et al. Differential activation of ras genes 44. Van Cutsem E, Kohne CH, Lang I et al. Cetuximab plus irinotecan, fluorouracil,
by point mutation in human colon cancer with metastases to either lung or liver. and leucovorin as first-line treatment for metastatic colorectal cancer: updated
Int J Cancer 1991; 49: 875–879. analysis of overall survival according to tumor KRAS and BRAF mutation status. J
24. Al-Mulla F, Going JJ, Sowden ET et al. Heterogeneity of mutant versus wild-type Clin Oncol 2011; 29: 2011–2019.
Ki-ras in primary and metastatic colorectal carcinomas, and association of 45. Bokemeyer C, Bondarenko I, Makhson A et al. Fluorouracil, leucovorin, and
codon-12 valine with early mortality. J Pathol 1998; 185: 130–138. oxaliplatin with and without cetuximab in the first-line treatment of metastatic
25. Albanese I, Scibetta AG, Migliavacca M et al. Heterogeneity within and between colorectal cancer. J Clin Oncol 2009; 27: 663–671.
primary colorectal carcinomas and matched metastases as revealed by analysis 46. Bokemeyer C, Bondarenko I, Hartmann JT et al. Efficacy according to biomarker
of Ki-ras and p53 mutations. Biochem Biophys Res Commun 2004; 325: status of cetuximab plus FOLFOX-4 as first-line treatment for metastatic
784–791. colorectal cancer: the OPUS study. Ann Oncol 2011; 22: 1535–1546.
26. Oliveira C, Velho S, Moutinho C et al. KRAS and BRAF oncogenic mutations in 47. Peeters M, Price TJ, Cervantes A et al. Randomized phase III study of
MSS colorectal carcinoma progression. Oncogene 2007; 26: 158–163. panitumumab with fluorouracil, leucovorin, and irinotecan (FOLFIRI) compared
27. Artale S, Sartore-Bianchi A, Veronese SM et al. Mutations of KRAS and BRAF in with FOLFIRI alone as second-line treatment in patients with metastatic colorectal
primary and matched metastatic sites of colorectal cancer. J Clin Oncol 2008; cancer. J Clin Oncol 2010; 28: 4706–4713.
26: 4217–4219. 48. Douillard JY, Oliner KS, Siena S et al. Panitumumab-FOLFOX4 treatment and
28. Santini D, Loupakis F, Vincenzi B et al. High concordance of KRAS status RAS mutations in colorectal cancer. N Engl J Med 2013; 369: 1023–1034.
between primary colorectal tumors and related metastatic sites: implications for 49. Van Cutsem E, Lenz HJ, Kohne CH et al. Fluorouracil, leucovorin, and irinotecan
clinical practice. Oncologist 2008; 13: 1270–1275. plus cetuximab treatment and RAS mutations in colorectal cancer. J Clin Oncol
29. Cejas P, Lopez-Gomez M, Aguayo C et al. KRAS mutations in primary colorectal 2015; 33: 692–700.
cancer tumors and related metastases: a potential role in prediction of lung 50. Bokemeyer C, Kohne CH, Ciardiello F et al. FOLFOX4 plus cetuximab treatment
metastasis. PLoS One 2009; 4: e8199. and RAS mutations in colorectal cancer. Eur J Cancer 2015; 51: 1243–1252.
Volume 27 | No. 8 | August 2016 doi:10.1093/annonc/mdw235 | 

by guest
on 05 February 2018
51. Peeters M, Oliner KS, Price TJ et al. Analysis of KRAS/NRAS mutations in a 69. Sosman JA, Kim KB, Schuchter L et al. Survival in BRAF V600-mutant
Phase III Study of panitumumab with FOLFIRI compared with FOLFIRI alone as advanced melanoma treated with vemurafenib. N Engl J Med 2012; 366:
second-line treatment for metastatic colorectal cancer. Clin Cancer Res 2015; 707–714.
21: 5469–5479. 70. Kopetz S, Desai J, Chan E et al. Phase II pilot study of vemurafenib in patients
52. Poulin-Costello M, Azoulay L, Van Cutsem E et al. An analysis of the treatment with metastatic BRAF-mutated colorectal cancer. J Clin Oncol 2015; 33:
effect of panitumumab on overall survival from a phase 3, randomized, 4032–4038.
controlled, multicenter trial (20020408) in patients with chemotherapy refractory 71. Prahallad A, Sun C, Huang S et al. Unresponsiveness of colon cancer to BRAF
metastatic colorectal cancer. Target Oncol 2013; 8: 127–136. (V600E) inhibition through feedback activation of EGFR. Nature 2012; 483:
53. Schwartzberg LS, Rivera F, Karthaus M et al. PEAK: a randomized, multicenter 100–103.
phase II study of panitumumab plus modified fluorouracil, leucovorin, and 72. Corcoran RB, Atreya CE, Falchook GS et al. Combined BRAF and MEK inhibition
oxaliplatin (mFOLFOX6) or bevacizumab plus mFOLFOX6 in patients with with dabrafenib and trametinib in BRAF V600-mutant colorectal cancer. J Clin
previously untreated, unresectable, wild-type KRAS exon 2 metastatic colorectal Oncol 2015; 33: 4023–4031.
cancer. J Clin Oncol 2014; 32: 2240–2247. 73. Hyman DM, Puzanov I, Subbiah V et al. Vemurafenib in multiple
54. Stintzing S, Jung A, Rossius L et al. Mutations within the EGFR signaling nonmelanoma cancers with BRAF V600 mutations. N Engl J Med 2015; 373:
pathway: influence on efficacy in FIRE-3—a randomized phase III study of 726–736.
FOLFIRI plus cetuximab or bevacizumab as first-line treatment for wild-type (WT) 74. Bettstetter M, Dechant S, Ruemmele P et al. Distinction of hereditary
KRAS (exon 2) metastatic colorectal cancer (mCRC) patients. J Clin Oncol 2014; nonpolyposis colorectal cancer and sporadic microsatellite-unstable colorectal
32(3 Suppl): abstr 445. cancer through quantification of MLH1 methylation by real-time PCR. Clin Cancer
55. Heinemann V, von Weikersthal LF, Decker T et al. FOLFIRI plus cetuximab versus Res 2007; 13: 3221–3228.
FOLFIRI plus bevacizumab as first-line treatment for patients with metastatic 75. Domingo E, Niessen RC, Oliveira C et al. BRAF-V600E is not involved in the
colorectal cancer (FIRE-3): a randomised, open-label, phase 3 trial. Lancet Oncol colorectal tumorigenesis of HNPCC in patients with functional MLH1 and MSH2
2014; 15: 1065–1075. genes. Oncogene 2005; 24: 3995–3998.
56. Sorich MJ, Wiese MD, Rowland A et al. Extended RAS mutations and anti-EGFR 76. Loughrey MB, Waring PM, Tan A et al. Incorporation of somatic BRAF mutation
monoclonal antibody survival benefit in metastatic colorectal cancer: a meta- testing into an algorithm for the investigation of hereditary non-polyposis
analysis of randomized, controlled trials. Ann Oncol 2015; 26: 13–21. colorectal cancer. Fam Cancer 2007; 6: 301–310.
57. Wong NA, Gonzalez D, Salto-Tellez M et al. RAS testing of colorectal carcinoma 77. Cremolini C, Di Bartolomeo M, Amatu A et al. BRAF codons 594 and 596
—a guidance document from the Association of Clinical Pathologists Molecular mutations identify a new molecular subtype of metastatic colorectal cancer at
Pathology and Diagnostics Group. J Clin Pathol 2014; 67: 751–757. favorable prognosis. Ann Oncol 2015; 26: 2092–2097.
58. Tack V, Ligtenberg MJ, Tembuyser L et al. External quality assessment unravels 78. Des Guetz G, Uzzan B, Nicolas P et al. Microsatellite instability: a predictive
interlaboratory differences in quality of RAS testing for anti-EGFR therapy in marker in metastatic colorectal cancer? Target Oncol 2009; 4: 57–62.
colorectal cancer. Oncologist 2015; 20: 257–262. 79. Goldstein J, Tran B, Ensor J et al. Multicenter retrospective analysis of metastatic
59. Kircher SM, Mohindra N, Nimeiri H. Cost estimates and economic implications of colorectal cancer (CRC) with high-level microsatellite instability (MSI-H). Ann
expanded RAS testing in metastatic colorectal cancer. Oncologist 2015; 20: 14–18. Oncol 2014; 25: 1032–1038.
60. Tran B, Kopetz S, Tie J et al. Impact of BRAF mutation and microsatellite 80. Muller CI, Schulmann K, Reinacher-Schick A et al. Predictive and prognostic
instability on the pattern of metastatic spread and prognosis in metastatic value of microsatellite instability in patients with advanced colorectal cancer
colorectal cancer. Cancer 2011; 117: 4623–4632. treated with a fluoropyrimidine and oxaliplatin containing first-line chemotherapy.
61. Tveit KM, Guren T, Glimelius B et al. Phase III trial of cetuximab with continuous A report of the AIO Colorectal Study Group. Int J Colorectal Dis 2008; 23:
or intermittent fluorouracil, leucovorin, and oxaliplatin (Nordic FLOX) versus FLOX 1033–1039.
alone in first-line treatment of metastatic colorectal cancer: the NORDIC-VII 81. Nopel-Dunnebacke S, Schulmann K, Reinacher-Schick A et al. Prognostic value
study. J Clin Oncol 2012; 30: 1755–1762. of microsatellite instability and p53 expression in metastatic colorectal cancer
62. Venderbosch S, Nagtegaal ID, Maughan TS et al. Mismatch repair status and treated with oxaliplatin and fluoropyrimidine-based chemotherapy. Z
BRAF mutation status in metastatic colorectal cancer patients: a pooled analysis Gastroenterol 2014; 52: 1394–1401.
of the CAIRO, CAIRO2, COIN, and FOCUS studies. Clin Cancer Res 2014; 20: 82. des Guetz G, Mariani P, Cucherousset J et al. Microsatellite instability and
5322–5330. sensitivitiy to FOLFOX treatment in metastatic colorectal cancer. Anticancer Res
63. Tol J, Nagtegaal ID, Punt CJ. BRAF mutation in metastatic colorectal cancer. N 2007; 27: 2715–2719.
Engl J Med 2009; 361: 98–99. 83. Kim JE, Hong YS, Ryu MH et al. Association between deficient mismatch repair
64. Peeters M, Oliner KS, Parker A et al. Massively parallel tumor multigene system and efficacy to irinotecan-containing chemotherapy in metastatic colon
sequencing to evaluate response to panitumumab in a randomized phase III cancer. Cancer Sci 2011; 102: 1706–1711.
study of metastatic colorectal cancer. Clin Cancer Res 2013; 19: 1902–1912. 84. Le DT, Uram JN, Wang H et al. PD-1 blockade in tumors with mismatch-repair
65. Seymour MT, Brown SR, Middleton G et al. Panitumumab and irinotecan versus deficiency. N Engl J Med 2015; 372: 2509–2520.
irinotecan alone for patients with KRAS wild-type, fluorouracil-resistant advanced 85. Sanoff HK, McLeod HL. Predictive factors for response and toxicity in
colorectal cancer (PICCOLO): a prospectively stratified randomised trial. Lancet chemotherapy: pharmacogenomics. Semin Colon Rectal Surg 2008; 19:
Oncol 2013; 14: 749–759. 226–230.
66. Pietrantonio F, Petrelli F, Coinu A et al. Predictive role of BRAF mutations in 86. Meulendijks D, Henricks LM, Sonke GS et al. Clinical relevance of DPYD variants
patients with advanced colorectal cancer receiving cetuximab and panitumumab: c.1679T>G, c.1236G>A/HapB3, and c.1601G>A as predictors of severe
a meta-analysis. Eur J Cancer 2015; 51: 587–594. fluoropyrimidine-associated toxicity: a systematic review and meta-analysis of
67. Rowland A, Dias MM, Wiese MD et al. Meta-analysis of BRAF mutation as a individual patient data. Lancet Oncol 2015; 16: 1639–1650.
predictive biomarker of benefit from anti-EGFR monoclonal antibody therapy 87. Marsh S, Hoskins JM. Irinotecan pharmacogenomics. Pharmacogenomics 2010;
for RAS wild-type metastatic colorectal cancer. Br J Cancer 2015; 112: 11: 1003–1010.
1888–1894. 88. Barbarino JM, Haidar CE, Klein TE, Altman RB. PharmGKB summary: very
68. Cremolini C, Loupakis F, Antoniotti C et al. FOLFOXIRI plus bevacizumab versus important pharmacogene information for UGT1A1. Pharmacogenet Genomics
FOLFIRI plus bevacizumab as first-line treatment of patients with metastatic 2014; 24: 177–183.
colorectal cancer: updated overall survival and molecular subgroup analyses of 89. Camptosar prescribing information. http://labeling.pfizer.com/ShowLabeling.
the open-label, phase 3 TRIBE study. Lancet Oncol 2015; 16: 1306–1315. aspx?id=533. 2014.
 | Van Cutsem et al. Volume 27 | No. 8 | August 2016

by guest
on 05 February 2018
90. Olaussen KA, Mountzios G, Soria JC. ERCC1 as a risk stratifier in platinum-based 111. Misale S, Yaeger R, Hobor S et al. Emergence of KRAS mutations and acquired
chemotherapy for nonsmall-cell lung cancer. Curr Opin Pulm Med 2007; 13: resistance to anti-EGFR therapy in colorectal cancer. Nature 2012; 486:
284–289. 532–536.
91. Soria JC. ERCC1-tailored chemotherapy in lung cancer: the first prospective 112. Laurent-Puig P, Cayre A, Manceau G et al. Analysis of PTEN, BRAF, and EGFR
randomized trial. J Clin Oncol 2007; 25: 2648–2649. status in determining benefit from cetuximab therapy in wild-type KRAS
92. Qian YY, Liu XY, Wu Q et al. The ERCC1 C118T polymorphism predicts clinical metastatic colon cancer. J Clin Oncol 2009; 27: 5924–5930.
outcomes of colorectal cancer patients receiving oxaliplatin-based chemotherapy: 113. Kavuri SM, Jain N, Galimi F et al. HER2 activating mutations are targets for
a meta-analysis based on 22 studies. Asian Pac J Cancer Prev 2014; 15: colorectal cancer treatment. Cancer Discov 2015; 5: 832–841.
8383–8390. 114. Siena S, Sartore-Bianchi A, Lonardi S et al. Trastuzumab and lapatinib in HER2-
93. Orlandi A, Di Salvatore M, Bagala C et al. ERCC1 induction after oxaliplatin amplified metastatic colorectal cancer patients (mCRC): the HERACLES trial. J
exposure may depend on KRAS mutational status in colorectal cancer cell line: in Clin Oncol 2015; 33(15 Suppl): abstr 3508.
vitro veritas. J Cancer 2015; 6: 70–81. 115. Juo YY, Johnston FM, Zhang DY et al. Prognostic value of CpG island methylator
94. Peters GJ, Backus HH, Freemantle S et al. Induction of thymidylate synthase as a phenotype among colorectal cancer patients: a systematic review and meta-
5-fluorouracil resistance mechanism. Biochim Biophys Acta 2002; 1587: analysis. Ann Oncol 2014; 25: 2314–2327.
194–205. 116. Moutinho C, Martinez-Cardus A, Santos C et al. Epigenetic inactivation of the
95. Hoskins JM, Goldberg RM, Qu P et al. UGT1A1*28 genotype and irinotecan- BRCA1 interactor SRBC and resistance to oxaliplatin in colorectal cancer. J Natl
induced neutropenia: dose matters. J Natl Cancer Inst 2007; 99: 1290–1295. Cancer Inst 2014; 106: djt322.
96. Misale S, Di Nicolantonio F, Sartore-Bianchi A et al. Resistance to anti-EGFR 117. Cohen SJ, Punt CJ, Iannotti N et al. Relationship of circulating tumor cells to
therapy in colorectal cancer: from heterogeneity to convergent evolution. Cancer tumor response, progression-free survival, and overall survival in patients with
Discov 2014; 4: 1269–1280. metastatic colorectal cancer. J Clin Oncol 2008; 26: 3213–3221.
97. Bardelli A, Siena S. Molecular mechanisms of resistance to cetuximab and 118. Diaz LA, Jr., Bardelli A. Liquid biopsies: genotyping circulating tumor DNA. J Clin
panitumumab in colorectal cancer. J Clin Oncol 2010; 28: 1254–1261. Oncol 2014; 32: 579–586.
98. Mohan S, Heitzer E, Ulz P et al. Changes in colorectal carcinoma genomes under 119. Siravegna G, Bardelli A. Genotyping cell-free tumor DNA in the blood to detect
anti-EGFR therapy identified by whole-genome plasma DNA sequencing. PLoS residual disease and drug resistance. Genome Biol 2014; 15: 449.
Genet 2014; 10: e1004271. 120. Montagut C, Dalmases A, Bellosillo B et al. Identification of a mutation
99. Yonesaka K, Zejnullahu K, Okamoto I et al. Activation of ERBB2 signaling causes in the extracellular domain of the Epidermal Growth Factor Receptor
resistance to the EGFR-directed therapeutic antibody cetuximab. Sci Transl Med conferring cetuximab resistance in colorectal cancer. Nat Med 2012; 18:
2011; 3: 99ra86. 221–223.
100. Bertotti A, Migliardi G, Galimi F et al. A molecularly annotated platform of patient- 121. Aung KL, Board RE, Ellison G et al. Current status and future potential of somatic
derived xenografts (‘xenopatients’) identifies HER2 as an effective therapeutic mutation testing from circulating free DNA in patients with solid tumours. Hugo J
target in cetuximab-resistant colorectal cancer. Cancer Discov 2011; 1: 2010; 4: 11–21.
508–523. 122. Bettegowda C, Sausen M, Leary RJ et al. Detection of circulating tumor DNA
101. Douillard JY, Siena S, Cassidy J et al. Randomized, phase III trial of panitumumab in early- and late-stage human malignancies. Sci Transl Med 2014; 6:
with infusional fluorouracil, leucovorin, and oxaliplatin (FOLFOX4) versus 224ra224.
FOLFOX4 alone as first-line treatment in patients with previously untreated 123. Dawson SJ, Tsui DW, Murtaza M et al. Analysis of circulating tumor DNA to
metastatic colorectal cancer: the PRIME study. J Clin Oncol 2010; 28: monitor metastatic breast cancer. N Engl J Med 2013; 368: 1199–1209.
4697–4705. 124. De Mattos-Arruda L, Weigelt B, Cortes J et al. Capturing intra-tumor genetic
102. Stintzing S, Jung AS, Rossius L et al. Analysis of KRAS/NRAS and BRAF heterogeneity by de novo mutation profiling of circulating cell-free tumor DNA: a
mutations in FIRE-3. Eur J Cancer 2013; 49(Suppl 3): abstr LBA17. proof-of-principle. Ann Oncol 2014; 25: 1729–1735.
103. Valtorta E, Misale S, Sartore-Bianchi A et al. KRAS gene amplification in 125. Diaz LA, Jr., Williams RT, Wu J et al. The molecular evolution of acquired
colorectal cancer and impact on response to EGFR-targeted therapy. Int J Cancer resistance to targeted EGFR blockade in colorectal cancers. Nature 2012; 486:
2013; 133: 1259–1265. 537–540.
104. Vaughn CP, Zobell SD, Furtado LV et al. Frequency of KRAS, BRAF, and NRAS 126. Forshew T, Murtaza M, Parkinson C et al. Noninvasive identification and
mutations in colorectal cancer. Genes Chromosomes Cancer 2011; 50: 307–312. monitoring of cancer mutations by targeted deep sequencing of plasma DNA. Sci
105. Ogino S, Lochhead P, Giovannucci E et al. Discovery of colorectal cancer PIK3CA Transl Med 2012; 4: 136ra168.
mutation as potential predictive biomarker: power and promise of molecular 127. Gormally E, Caboux E, Vineis P, Hainaut P. Circulating free DNA in plasma or
pathological epidemiology. Oncogene 2014; 33: 2949–2955. serum as biomarker of carcinogenesis: practical aspects and biological
106. Jhawer M, Goel S, Wilson AJ et al. PIK3CA mutation/PTEN expression status significance. Mutat Res 2007; 635: 105–117.
predicts response of colon cancer cells to the epidermal growth factor receptor 128. Higgins MJ, Jelovac D, Barnathan E et al. Detection of tumor PIK3CA status in
inhibitor cetuximab. Cancer Res 2008; 68: 1953–1961. metastatic breast cancer using peripheral blood. Clin Cancer Res 2012; 18:
107. Karapetis CS, Jonker D, Daneshmand M et al. PIK3CA, BRAF, and PTEN status 3462–3469.
and benefit from cetuximab in the treatment of advanced colorectal cancer— 129. Murtaza M, Dawson SJ, Tsui DW et al. Non-invasive analysis of acquired
results from NCIC CTG/AGITG CO.17. Clin Cancer Res 2014; 20: 744–753. resistance to cancer therapy by sequencing of plasma DNA. Nature 2013; 497:
108. Prenen H, De Schutter J, Jacobs B et al. PIK3CA mutations are not a major 108–112.
determinant of resistance to the epidermal growth factor receptor inhibitor 130. Newman AM, Bratman SV, To J et al. An ultrasensitive method for quantitating
cetuximab in metastatic colorectal cancer. Clin Cancer Res 2009; 15: circulating tumor DNA with broad patient coverage. Nat Med 2014; 20:
3184–3188. 548–554.
109. Sartore-Bianchi A, Martini M, Molinari F et al. PIK3CA mutations in colorectal 131. Punnoose EA, Atwal S, Liu W et al. Evaluation of circulating tumor cells and
cancer are associated with clinical resistance to EGFR-targeted monoclonal circulating tumor DNA in non-small cell lung cancer: association with clinical
antibodies. Cancer Res 2009; 69: 1851–1857. endpoints in a phase II clinical trial of pertuzumab and erlotinib. Clin Cancer Res
110. Tian S, Simon I, Moreno V et al. A combined oncogenic pathway signature of 2012; 18: 2391–2401.
BRAF, KRAS and PI3KCA mutation improves colorectal cancer classification and 132. Schwarzenbach H, Hoon DS, Pantel K. Cell-free nucleic acids as biomarkers in
cetuximab treatment prediction. Gut 2013; 62: 540–549. cancer patients. Nat Rev Cancer 2011; 11: 426–437.
Volume 27 | No. 8 | August 2016 doi:10.1093/annonc/mdw235 | 

by guest
on 05 February 2018
133. Taly V, Pekin D, Benhaim L et al. Multiplex picodroplet digital PCR to detect KRAS 155. Bartolotta TV, Taibbi A, Midiri M et al. Characterisation of focal liver lesions
mutations in circulating DNA from the plasma of colorectal cancer patients. Clin undetermined at grey-scale US: contrast-enhanced US versus 64-row MDCT and
Chem 2013; 59: 1722–1731. MRI with liver-specific contrast agent. Radiol Med 2010; 115: 714–731.
134. Taniguchi K, Uchida J, Nishino K et al. Quantitative detection of EGFR mutations 156. Niekel MC, Bipat S, Stoker J. Diagnostic imaging of colorectal liver metastases
in circulating tumor DNA derived from lung adenocarcinomas. Clin Cancer Res with CT, MR imaging, FDG PET, and/or FDG PET/CT: a meta-analysis of
2011; 17: 7808–7815. prospective studies including patients who have not previously undergone
135. Thierry AR, Mouliere F, El Messaoudi S et al. Clinical validation of the detection of treatment. Radiology 2010; 257: 674–684.
KRAS and BRAF mutations from circulating tumor DNA. Nat Med 2014; 20: 157. Zech CJ, Korpraphong P, Huppertz A et al. Randomized multicentre trial of
430–435. gadoxetic acid-enhanced MRI versus conventional MRI or CT in the staging of
136. Li J, Mansmann UR. A microRNA molecular modeling extension for prediction of colorectal cancer liver metastases. Br J Surg 2014; 101: 613–621.
colorectal cancer treatment. BMC Cancer 2015; 15: 472. 158. Hendlisz A, Golfinopoulos V, Garcia C et al. Serial FDG-PET/CT for early outcome
137. Ciardiello F, Normanno N, Maiello E et al. Clinical activity of FOLFIRI plus prediction in patients with metastatic colorectal cancer undergoing
cetuximab according to extended gene mutation status by next-generation chemotherapy. Ann Oncol 2012; 23: 1687–1693.
sequencing: findings from the CAPRI-GOIM trial. Ann Oncol 2014; 25: 159. Moulton C, Levine MN, Law C et al. An Ontario Clinical Oncology Group (OCOG)
1756–1761. randomized controlled trial (RCT) assessing FDG PET/CT in resectable liver
138. Marisa L, de Reynies A, Duval A et al. Gene expression classification of colon colorectal adenocarcinoma metastases (CAM). J Clin Oncol 2011; 29(Suppl):
cancer into molecular subtypes: characterization, validation, and prognostic abstr 3520.
value. PLoS Med 2013; 10: e1001453. 160. Petersen RK, Hess S, Alavi A, Hoilund-Carlsen PF. Clinical impact of FDG-PET/CT
139. Guinney J, Dienstmann R, Wang B et al. The consensus molecular subtypes of on colorectal cancer staging and treatment strategy. Am J Nucl Med Mol Imaging
colorectal cancer. Nat Med 2015; 21: 1350–1356. 2014; 4: 471–482.
140. Prades J, Borras JM. Shifting sands: adapting the multidisciplinary team model 161. Maffione AM, Lopci E, Bluemel C et al. Diagnostic accuracy and impact on
to technological and organizational innovations in cancer care. Future Oncol management of (18)F-FDG PET and PET/CT in colorectal liver metastasis: a
2014; 10: 1995–1998. meta-analysis and systematic review. Eur J Nucl Med Mol Imaging 2015; 42:
141. Shah S, Arora S, Atkin G et al. Decision-making in Colorectal Cancer Tumor 152–163.
Board meetings: results of a prospective observational assessment. Surg Endosc 162. Tomlinson JS, Jarnagin WR, DeMatteo RP et al. Actual 10-year survival after
2014; 28: 2783–2788. resection of colorectal liver metastases defines cure. J Clin Oncol 2007; 25:
142. Prades J, Remue E, van Hoof E, Borras JM. Is it worth reorganising cancer 4575–4580.
services on the basis of multidisciplinary teams (MDTs)? A systematic review of 163. Nordlinger B, Sorbye H, Glimelius B et al. Perioperative FOLFOX4 chemotherapy
the objectives and organisation of MDTs and their impact on patient outcomes. and surgery versus surgery alone for resectable liver metastases from colorectal
Health Policy 2015; 119: 464–474. cancer (EORTC 40983): long-term results of a randomised, controlled, phase 3
143. Weiser MR, Jarnagin WR, Saltz LB. Colorectal cancer patients with trial. Lancet Oncol 2013; 14: 1208–1215.
oligometastatic liver disease: what is the optimal approach? Oncology (Williston 164. Nordlinger B, Sorbye H, Glimelius B et al. Perioperative chemotherapy with
Park) 2013; 27: 1074–1078. FOLFOX4 and surgery versus surgery alone for resectable liver metastases from
144. Khattak MA, Martin HL, Beeke C et al. Survival differences in patients with colorectal cancer (EORTC Intergroup trial 40983): a randomised controlled trial.
metastatic colorectal cancer and with single site metastatic disease at initial Lancet 2008; 371: 1007–1016.
presentation: results from South Australian clinical registry for advanced 165. Primrose J, Falk S, Finch-Jones M et al. Systemic chemotherapy with or without
colorectal cancer. Clin Colorectal Cancer 2012; 11: 247–254. cetuximab in patients with resectable colorectal liver metastasis: the New EPOC
145. Price TJ, Townsend AR, Beeke C et al. ‘Watchful waiting’ for metastatic randomised controlled trial. Lancet Oncol 2014; 15: 601–611.
colorectal cancer, antediluvian or an option to be considered again? Asia Pac J 166. Mitry E, Fields AL, Bleiberg H et al. Adjuvant chemotherapy after potentially
Clin Oncol 2012; 8: 10–13. curative resection of metastases from colorectal cancer: a pooled analysis of two
146. Yaeger R, Cercek A, Chou JF et al. BRAF mutation predicts for poor outcomes randomized trials. J Clin Oncol 2008; 26: 4906–4911.
after metastasectomy in patients with metastatic colorectal cancer. Cancer 167. Kemeny N, Capanu M, D’Angelica M et al. Phase I trial of adjuvant hepatic
2014; 120: 2316–2324. arterial infusion (HAI) with floxuridine (FUDR) and dexamethasone plus systemic
147. Ohrling K, Edler D, Hallstrom M, Ragnhammar P. Expression of thymidylate oxaliplatin, 5-fluorouracil and leucovorin in patients with resected liver
synthase in liver and lung metastases of colorectal cancer and their matched metastases from colorectal cancer. Ann Oncol 2009; 20: 1236–1241.
primary tumours. Anticancer Res 2008; 28: 1741–1747. 168. Adam R, Delvart V, Pascal G et al. Rescue surgery for unresectable colorectal
148. Adam R, De Gramont A, Figueras J et al. The oncosurgery approach to managing liver metastases downstaged by chemotherapy: a model to predict long-term
liver metastases from colorectal cancer: a multidisciplinary international survival. Ann Surg 2004; 240: 644–657; discussion 657–658.
consensus. Oncologist 2012; 17: 1225–1239. 169. Adam R, Barroso E, Laurent C et al. Impact of the type and modalities of
149. Clavien PA, Petrowsky H, DeOliveira ML, Graf R. Strategies for safer liver surgery preoperative chemotherapy on the outcome of liver resection for colorectal
and partial liver transplantation. N Engl J Med 2007; 356: 1545–1559. metastases. J Clin Oncol 2011; 29(15 Suppl): abstr 3519.
150. Zorzi D, Laurent A, Pawlik TM et al. Chemotherapy-associated hepatotoxicity and 170. Chun YS, Vauthey JN, Boonsirikamchai P et al. Association of computed
surgery for colorectal liver metastases. Br J Surg 2007; 94: 274–286. tomography morphologic criteria with pathologic response and survival in patients
151. Jones RP, Stattner S, Sutton P et al. Controversies in the oncosurgical treated with bevacizumab for colorectal liver metastases. JAMA 2009; 302:
management of liver limited stage IV colorectal cancer. Surg Oncol 2014; 23: 2338–2344.
53–60. 171. Adam R, Bhangui P, Poston G et al. Is perioperative chemotherapy useful for
152. Kanas GP, Taylor A, Primrose JN et al. Survival after liver resection in metastatic solitary, metachronous, colorectal liver metastases? Ann Surg 2010; 252: 774–787.
colorectal cancer: review and meta-analysis of prognostic factors. Clin Epidemiol 172. Nordlinger B, Van Cutsem E, Gruenberger T et al. Combination of surgery and
2012; 4: 283–301. chemotherapy and the role of targeted agents in the treatment of patients with
153. Fong Y, Fortner J, Sun RL et al. Clinical score for predicting recurrence after colorectal liver metastases: recommendations from an expert panel. Ann Oncol
hepatic resection for metastatic colorectal cancer: analysis of 1001 consecutive 2009; 20: 985–992.
cases. Ann Surg 1999; 230: 309–318; discussion 318–321. 173. Nordlinger B, Van Cutsem E, Rougier P et al. Does chemotherapy prior to liver
154. Floriani I, Torri V, Rulli E et al. Performance of imaging modalities in diagnosis of resection increase the potential for cure in patients with metastatic colorectal
liver metastases from colorectal cancer: a systematic review and meta-analysis. J cancer? A report from the European Colorectal Metastases Treatment Group. Eur
Magn Reson Imaging 2010; 31: 19–31. J Cancer 2007; 43: 2037–2045.
 | Van Cutsem et al. Volume 27 | No. 8 | August 2016

by guest
on 05 February 2018
174. Van Cutsem E, Nordlinger B, Adam R et al. Towards a pan-European consensus 192. Cremolini C, Loupakis F, Masi G et al. FOLFOXIRI/bevacizumab versus FOLFIRI/
on the treatment of patients with colorectal liver metastases. Eur J Cancer 2006; bevacizumab as first-line treatment in unresectable metastatic colorectal cancer:
42: 2212–2221. results of phase III TRIBE trial by GONO Group. Ann Oncol 2014; 24(Suppl 4):
175. Folprecht G, Grothey A, Alberts S et al. Neoadjuvant treatment of unresectable abstr O-0026.
colorectal liver metastases: correlation between tumour response and resection 193. Stintzing S, Modest DP, Fischer von Weikersthal L et al. Independent radiological
rates. Ann Oncol 2005; 16: 1311–1319. evaluation of objective response, early tumor shrinkage, and depth of response in
176. Folprecht G, Gruenberger T, Bechstein WO et al. Tumour response and FIRE-3 (AIO KRK-0306) in the final RAS evaluable population. Ann Oncol 2014;
secondary resectability of colorectal liver metastases following neoadjuvant 25(Suppl 5): abstr LBA11.
chemotherapy with cetuximab: the CELIM randomised phase 2 trial. Lancet 194. Rivera F, Karthaus M, Hecht JR et al. First-line treatment with modified FOLFOX6
Oncol 2010; 11: 38–47. (mFOLFOX6) + panitumumab or bevacizumab in patients with RAS/BRAF wild-
177. Ye LC, Liu TS, Ren L et al. Randomized controlled trial of cetuximab plus type (WT) metastatic colorectal carcinoma (mCRC). Ann Oncol 2015; 26(Suppl
chemotherapy for patients with KRAS wild-type unresectable colorectal liver- 4): abstr PD-014.
limited metastases. J Clin Oncol 2013; 31: 1931–1938. 195. Bhutiani N, Akinwande O, Martin RC, 2nd. Efficacy and toxicity of hepatic intra-
178. Gruenberger T, Bridgewater J, Chau I et al. Bevacizumab plus mFOLFOX-6 or arterial drug-eluting (irinotecan) bead (DEBIRI) therapy in irinotecan-refractory
FOLFOXIRI in patients with initially unresectable liver metastases from colorectal unresectable colorectal liver metastases. World J Surg 2015.
cancer: the OLIVIA multinational randomised phase II trial. Ann Oncol 2014; 26: 196. Liu DM, Thakor AS, Baerlocher M et al. A review of conventional and drug-eluting
702–708. chemoembolization in the treatment of colorectal liver metastases: principles and
179. Masi G, Loupakis F, Salvatore L et al. Bevacizumab with FOLFOXIRI (irinotecan, proof. Future Oncol 2015; 11: 1421–1428.
oxaliplatin, fluorouracil, and folinate) as first-line treatment for metastatic 197. Martin RC, 2nd, Scoggins CR, Schreeder M et al. Randomized controlled trial of
colorectal cancer: a phase 2 trial. Lancet Oncol 2010; 11: 845–852. irinotecan drug-eluting beads with simultaneous FOLFOX and bevacizumab for
180. Folprecht G, Gruenberger T, Bechstein W et al. Survival of patients with initially patients with unresectable colorectal liver-limited metastasis. Cancer 2015; 121:
unresectable colorectal liver metastases treated with FOLFOX/cetuximab or 3649–3658.
FOLFIRI/cetuximab in a multidisciplinary concept (CELIM study). Ann Oncol 198. Ruers T, Punt C, Van Coevorden F et al. Radiofrequency ablation (RFA)
2014; 25: 1018–1025. combined with chemotherapy for unresectable colorectal liver metastases
181. Falcone A, Ricci S, Brunetti I et al. Phase III trial of infusional fluorouracil, (CRC LM): long-term survival results of a randomized phase II study of the
leucovorin, oxaliplatin, and irinotecan (FOLFOXIRI) compared with infusional EORTC-NCRI CCSG-ALM Intergroup 40004 (CLOCC). J Clin Oncol 2015; 33
fluorouracil, leucovorin, and irinotecan (FOLFIRI) as first-line treatment for (15 Suppl): abstr 3501.
metastatic colorectal cancer: the Gruppo Oncologico Nord Ovest. J Clin Oncol 199. Tanis E, Nordlinger B, Mauer M et al. Local recurrence rates after radiofrequency
2007; 25: 1670–1676. ablation or resection of colorectal liver metastases. Analysis of the European
182. Loupakis F, Cremolini C, Masi G et al. Initial therapy with FOLFOXIRI and Organisation for Research and Treatment of Cancer #40004 and #40983. Eur J
bevacizumab for metastatic colorectal cancer. N Engl J Med 2014; 371: Cancer 2014; 50: 912–919.
1609–1618. 200. Ruers T, Punt C, Van Coevorden F et al. Radiofrequency ablation combined with
183. Douillard JY, Siena S, Cassidy J et al. Final results from PRIME: randomized systemic treatment versus systemic treatment alone in patients with non-
phase III study of panitumumab with FOLFOX4 for first-line treatment of resectable colorectal liver metastases: a randomized EORTC Intergroup phase II
metastatic colorectal cancer. Ann Oncol 2014; 25: 1346–1355. study (EORTC 40004). Ann Oncol 2012; 23: 2619–2626.
184. Okines A, Puerto OD, Cunningham D et al. Surgery with curative-intent in 201. Evrard S, Poston G, Kissmeyer-Nielsen P et al. Combined ablation and resection
patients treated with first-line chemotherapy plus bevacizumab for metastatic (CARe) as an effective parenchymal sparing treatment for extensive colorectal
colorectal cancer First BEAT and the randomised phase-III NO16966 trial. Br J liver metastases. PLoS One 2014; 9: e114404.
Cancer 2009; 101: 1033–1038. 202. Lencioni R, Crocetti L, Cioni R et al. Response to radiofrequency ablation of
185. Venook A, Niedzwiecki D, Lenz HJ et al. CALGB/SWOG 80405: Phase III trial of pulmonary tumours: a prospective, intention-to-treat, multicentre clinical trial (the
irinotecan/5-FU/leucovorin (FOLFIRI) or oxaliplatin/5-FU/leucovorin (mFOLFOX6) RAPTURE study). Lancet Oncol 2008; 9: 621–628.
with bevacizumab (BV) or cetuximab (CET) for patients ( pts) with KRAS wild-type 203. Petre EN, Jia X, Thornton RH et al. Treatment of pulmonary colorectal metastases
(wt) untreated metastatic adenocarcinoma of the colon or rectum (MCRC). J Clin by radiofrequency ablation. Clin Colorectal Cancer 2013; 12: 37–44.
Oncol 2014; 32(15 Suppl): abstr LBA 3. 204. Kashima M, Yamakado K, Takaki H et al. Complications after 1000 lung
186. Piessevaux H, Buyse M, Schlichting M et al. Use of early tumor shrinkage to radiofrequency ablation sessions in 420 patients: a single center’s experiences.
predict long-term outcome in metastatic colorectal cancer treated with Am J Roentgenol 2011; 197: W576–W580.
cetuximab. J Clin Oncol 2013; 31: 3764–3775. 205. Schlijper RC, Grutters JP, Houben R et al. What to choose as radical local
187. Mansmann UR, Laubender RP, Giessen CA et al. Validating the prognostic treatment for lung metastases from colo-rectal cancer: surgery or radiofrequency
relevance of initial change in tumor size using a series of therapeutic regimens ablation? Cancer Treat Rev 2014; 40: 60–67.
for patients with metastatic colorectal cancer (mCRC). J Clin Oncol 2012; 30 206. Ricke J, Mohnike K, Pech M et al. Local response and impact on survival after
(Suppl 4): abstr 580. local ablation of liver metastases from colorectal carcinoma by computed
188. Mansmann UR, Laubender RP, Sartorius U et al. Improved early prediction of tomography-guided high-dose-rate brachytherapy. Int J Radiat Oncol Biol Phys
individual prognosis for patients with mCRC: joint modeling of tumor shrinkage 2010; 78: 479–485.
with volume data for PFS and OS. J Clin Oncol 2012; 30(Suppl 15): abstr 3603. 207. Sterzing F, Brunner TB, Ernst I et al. Stereotactic body radiotherapy for liver
189. Piessevaux H, Buyse M, De Roock W et al. Radiological tumor size decrease at tumors: principles and practical guidelines of the DEGRO Working Group on
week 6 is a potent predictor of outcome in chemorefractory metastatic Stereotactic Radiotherapy. Strahlenther Onkol 2014; 190: 872–881.
colorectal cancer treated with cetuximab (BOND trial). Ann Oncol 2009; 20: 208. Collettini F, Schnapauff D, Poellinger A et al. Percutaneous CT-guided high-dose
1375–1382. brachytherapy (CT-HDRBT) ablation of primary and metastatic lung tumors in
190. Cremolini C, Loupakis F, Antoniotti C et al. Early tumor shrinkage and depth of nonsurgical candidates. Rofo 2012; 184: 316–323.
response predict long-term outcome in metastatic colorectal cancer patients 209. Comito T, Cozzi L, Clerici E et al. Stereotactic Ablative Radiotherapy (SABR) in
treated with first-line chemotherapy plus bevacizumab: results from phase III inoperable oligometastatic disease from colorectal cancer: a safe and effective
TRIBE trial by the Gruppo Oncologico del Nord Ovest. Ann Oncol 2015; 26: approach. BMC Cancer 2014; 14: 619.
1188–1194. 210. Filippi AR, Badellino S, Guarneri A et al. Outcomes of single fraction stereotactic
191. Cremolini C, Loupakis F, Falcone A. FOLFOXIRI and bevacizumab for metastatic ablative radiotherapy for lung metastases. Technol Cancer Res Treat 2014; 13:
colorectal cancer. N Engl J Med 2015; 372: 291–292. 37–45.
Volume 27 | No. 8 | August 2016 doi:10.1093/annonc/mdw235 | 

by guest
on 05 February 2018
211. Tselis N, Ferentinos K, Kolotas C et al. Computed tomography-guided interstitial bowel involvement for peritoneal carcinomatosis of colorectal origin. Eur J Surg
high-dose-rate brachytherapy in the local treatment of primary and secondary Oncol 2014; 40: 1467–1473.
intrathoracic malignancies. J Thorac Oncol 2011; 6: 545–552. 232. Turaga K, Levine E, Barone R et al. Consensus guidelines from The American
212. Peters N, Wieners G, Pech M et al. CT-guided interstitial brachytherapy of Society of Peritoneal Surface Malignancies on standardizing the delivery of
primary and secondary lung malignancies: results of a prospective phase II trial. hyperthermic intraperitoneal chemotherapy (HIPEC) in colorectal cancer patients
Strahlenther Onkol 2008; 184: 296–301. in the United States. Ann Surg Oncol 2014; 21: 1501–1505.
213. Blomgren H, Lax I, Naslund I, Svanstrom R. Stereotactic high dose fraction 233. Prada-Villaverde A, Esquivel J, Lowy AM et al. The American Society of Peritoneal
radiation therapy of extracranial tumors using an accelerator. Clinical experience Surface Malignancies evaluation of HIPEC with mitomycin C versus oxaliplatin in
of the first thirty-one patients. Acta Oncol 1995; 34: 861–870. 539 patients with colon cancer undergoing a complete cytoreductive surgery. J
214. Wada H, Takai Y, Nemoto K, Yamada S. Univariate analysis of factors correlated Surg Oncol 2014; 110: 779–785.
with tumor control probability of three-dimensional conformal hypofractionated 234. de Gramont A, Figer A, Seymour M et al. Leucovorin and fluorouracil with or
high-dose radiotherapy for small pulmonary or hepatic tumors. Int J Radiat Oncol without oxaliplatin as first-line treatment in advanced colorectal cancer. J Clin
Biol Phys 2004; 58: 1114–1120. Oncol 2000; 18: 2938–2947.
215. Wulf J, Guckenberger M, Haedinger U et al. Stereotactic radiotherapy of primary 235. Douillard JY, Group VS. Irinotecan and high-dose fluorouracil/leucovorin for
liver cancer and hepatic metastases. Acta Oncol 2006; 45: 838–847. metastatic colorectal cancer. Oncology (Williston Park) 2000; 14: 51–55.
216. Ambrosino G, Polistina F, Costantin G et al. Image-guided robotic stereotactic 236. de Gramont A, Bosset JF, Milan C et al. Randomized trial comparing monthly
radiosurgery for unresectable liver metastases: preliminary results. Anticancer low-dose leucovorin and fluorouracil bolus with bimonthly high-dose leucovorin
Res 2009; 29: 3381–3384. and fluorouracil bolus plus continuous infusion for advanced colorectal cancer: a
217. Goodman KA, Wiegner EA, Maturen KE et al. Dose-escalation study of single- French intergroup study. J Clin Oncol 1997; 15: 808–815.
fraction stereotactic body radiotherapy for liver malignancies. Int J Radiat Oncol 237. Meta-Analysis Group in CancerLevy E, Piedbois P et al. Toxicity of fluorouracil in
Biol Phys 2010; 78: 486–493. patients with advanced colorectal cancer: effect of administration schedule and
218. Hoyer M, Roed H, Traberg Hansen A et al. Phase II study on stereotactic body prognostic factors. J Clin Oncol 1998; 16: 3537–3541.
radiotherapy of colorectal metastases. Acta Oncol 2006; 45: 823–830. 238. Van Cutsem E, Hoff PM, Harper P et al. Oral capecitabine vs intravenous 5-
219. Lee MT, Kim JJ, Dinniwell R et al. Phase I study of individualized fluorouracil and leucovorin: integrated efficacy data and novel analyses from two
stereotactic body radiotherapy of liver metastases. J Clin Oncol 2009; 27: large, randomised, phase III trials. Br J Cancer 2004; 90: 1190–1197.
1585–1591. 239. Cassidy J, Clarke S, Diaz-Rubio E et al. Randomized phase III study of
220. Mendez Romero A, Wunderink W, Hussain SM et al. Stereotactic body radiation capecitabine plus oxaliplatin compared with fluorouracil/folinic acid plus
therapy for primary and metastatic liver tumors: a single institution phase i-ii oxaliplatin as first-line therapy for metastatic colorectal cancer. J Clin Oncol
study. Acta Oncol 2006; 45: 831–837. 2008; 26: 2006–2012.
221. Rusthoven KE, Kavanagh BD, Cardenes H et al. Multi-institutional phase I/II trial 240. Fuchs CS, Marshall J, Mitchell E et al. Randomized, controlled trial of irinotecan
of stereotactic body radiation therapy for liver metastases. J Clin Oncol 2009; 27: plus infusional, bolus, or oral fluoropyrimidines in first-line treatment of
1572–1578. metastatic colorectal cancer: results from the BICC-C Study. J Clin Oncol 2007;
222. Scorsetti M, Arcangeli S, Tozzi A et al. Is stereotactic body radiation therapy an 25: 4779–4786.
attractive option for unresectable liver metastases? A preliminary report from a 241. Kohne CH, De Greve J, Hartmann JT et al. Irinotecan combined with infusional 5-
phase 2 trial. Int J Radiat Oncol Biol Phys 2013; 86: 336–342. fluorouracil/folinic acid or capecitabine plus celecoxib or placebo in the first-line
223. Fumagalli I, Bibault JE, Dewas S et al. A single-institution study of stereotactic treatment of patients with metastatic colorectal cancer. EORTC study 40015.
body radiotherapy for patients with unresectable visceral pulmonary or hepatic Ann Oncol 2008; 19: 920–926.
oligometastases. Radiat Oncol 2012; 7: 164. 242. Schmiegel W, Reinacher-Schick A, Arnold D et al. Capecitabine/irinotecan or
224. Fiorentini G, Aliberti C, Tilli M et al. Intra-arterial infusion of irinotecan-loaded capecitabine/oxaliplatin in combination with bevacizumab is effective and safe as
drug-eluting beads (DEBIRI) versus intravenous therapy (FOLFIRI) for hepatic first-line therapy for metastatic colorectal cancer: a randomized phase II study of
metastases from colorectal cancer: final results of a phase III study. Anticancer the AIO colorectal study group. Ann Oncol 2013; 24: 1580–1587.
Res 2012; 32: 1387–1395. 243. Moosmann N, von Weikersthal LF, Vehling-Kaiser U et al. Cetuximab plus
225. Hendlisz A, Van den Eynde M, Peeters M et al. Phase III trial comparing capecitabine and irinotecan compared with cetuximab plus capecitabine and
protracted intravenous fluorouracil infusion alone or with yttrium-90 resin oxaliplatin as first-line treatment for patients with metastatic colorectal cancer:
microspheres radioembolization for liver-limited metastatic colorectal cancer AIO KRK-0104—a randomized trial of the German AIO CRC study group. J Clin
refractory to standard chemotherapy. J Clin Oncol 2010; 28: 3687–3694. Oncol 2011; 29: 1050–1058.
226. van Hazel GA, Heinemann V, Sharma NK et al. SIRFLOX: randomized phase iii 244. Hurwitz H, Fehrenbacher L, Novotny W et al. Bevacizumab plus irinotecan,
trial comparing first-line mFOLFOX6 ( plus or minus bevacizumab) versus fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004;
mFOLFOX6 ( plus or minus bevacizumab) plus selective internal radiation therapy 350: 2335–2342.
in patients with metastatic colorectal cancer. J Clin Oncol 2016; 34: 245. Saltz LB, Clarke S, Diaz-Rubio E et al. Bevacizumab in combination with
1723–1731. oxaliplatin-based chemotherapy as first-line therapy in metastatic colorectal
227. Garlipp B, de Baere T, Damm R et al. Left-liver hypertrophy after therapeutic cancer: a randomized phase III study. J Clin Oncol 2008; 26: 2013–2019.
right-liver radioembolization is substantial but less than after portal vein 246. Souglakos J, Androulakis N, Syrigos K et al. FOLFOXIRI (folinic acid, 5-
embolization. Hepatology 2014; 59: 1864–1873. fluorouracil, oxaliplatin and irinotecan) vs FOLFIRI (folinic acid, 5-fluorouracil and
228. Cao C, Yan TD, Black D, Morris DL. A systematic review and meta-analysis of irinotecan) as first-line treatment in metastatic colorectal cancer (MCC): a
cytoreductive surgery with perioperative intraperitoneal chemotherapy for multicentre randomised phase III trial from the Hellenic Oncology Research Group
peritoneal carcinomatosis of colorectal origin. Ann Surg Oncol 2009; 16: (HORG). Br J Cancer 2006; 94: 798–805.
2152–2165. 247. Cunningham D, Lang I, Marcuello E et al. Bevacizumab plus capecitabine versus
229. Cotte E, Passot G, Mohamed F et al. Management of peritoneal carcinomatosis capecitabine alone in elderly patients with previously untreated metastatic
from colorectal cancer: current state of practice. Cancer J 2009; 15: 243–248. colorectal cancer (AVEX): an open-label, randomised phase 3 trial. Lancet Oncol
230. Elias D, Lefevre JH, Chevalier J et al. Complete cytoreductive surgery plus 2013; 14: 1077–1085.
intraperitoneal chemohyperthermia with oxaliplatin for peritoneal carcinomatosis 248. Kabbinavar F, Irl C, Zurlo A, Hurwitz H. Bevacizumab improves the overall and
of colorectal origin. J Clin Oncol 2009; 27: 681–685. progression-free survival of patients with metastatic colorectal cancer treated
231. Elias D, Mariani A, Cloutier AS et al. Modified selection criteria for complete with 5-fluorouracil-based regimens irrespective of baseline risk. Oncology 2008;
cytoreductive surgery plus HIPEC based on peritoneal cancer index and small 75: 215–223.
 | Van Cutsem et al. Volume 27 | No. 8 | August 2016

by guest
on 05 February 2018
249. Tebbutt NC, Wilson K, Gebski VJ et al. Capecitabine, bevacizumab, and 267. Diaz-Rubio E, Gomez-Espana A, Massuti B et al. First-line XELOX plus
mitomycin in first-line treatment of metastatic colorectal cancer: results of the bevacizumab followed by XELOX plus bevacizumab or single-agent bevacizumab
Australasian Gastrointestinal Trials Group Randomized Phase III MAX Study. J as maintenance therapy in patients with metastatic colorectal cancer: the phase
Clin Oncol 2010; 28: 3191–3198. III MACRO TTD study. Oncologist 2012; 17: 15–25.
250. Passardi A, Nanni O, Tassinari D et al. Effectiveness of bevacizumab added to 268. Hegewisch-Becker S, Graeven U, Lerchenmuller CA et al. Maintenance strategies
standard chemotherapy in metastatic colorectal cancer: final results for first-line after first-line oxaliplatin plus fluoropyrimidine plus bevacizumab for patients with
treatment from the ITACa randomized clinical trial. Ann Oncol 2015; 26: metastatic colorectal cancer (AIO 0207): a randomised, non-inferiority, open-
1201–1207. label, phase 3 trial. Lancet Oncol 2015; 16: 1355–1369.
251. Souglakos J, Ziras N, Kakolyris S et al. Randomised phase-II trial of CAPIRI 269. Koeberle D, Betticher DC, von Moos R et al. Bevacizumab continuation versus no
(capecitabine, irinotecan) plus bevacizumab vs FOLFIRI (folinic acid, 5- continuation after first-line chemotherapy plus bevacizumab in patients with
fluorouracil, irinotecan) plus bevacizumab as first-line treatment of patients with metastatic colorectal cancer: a randomized phase III non-inferiority trial (SAKK
unresectable/metastatic colorectal cancer (mCRC). Br J Cancer 2012; 106: 41/06). Ann Oncol 2015; 26: 709–714.
453–459. 270. Simkens LH, van Tinteren H, May A et al. Maintenance treatment with
252. Yamazaki K, Nagase M, Tamagawa H et al. A randomized phase III trial of capecitabine and bevacizumab in metastatic colorectal cancer (CAIRO3): a phase
mFOLFOX6 plus bevacizumab versus FOLFIRI plus bevacizumab as first-line 3 randomised controlled trial of the Dutch Colorectal Cancer Group. Lancet
treatment for metastatic colorectal cancer: West Japan Oncology Group study 2015; 385: 1843–1852.
4407G (WJOG4407G). J Clin Oncol 2014; 32(15 Suppl): abstr 3534. 271. Wasan H, Meade AM, Adams R et al. Intermittent chemotherapy plus either
253. Kozloff M, Yood MU, Berlin J et al. Clinical outcomes associated with intermittent or continuous cetuximab for first-line treatment of patients with KRAS
bevacizumab-containing treatment of metastatic colorectal cancer: the BRiTE wild-type advanced colorectal cancer (COIN-B): a randomised phase 2 trial.
observational cohort study. Oncologist 2009; 14: 862–870. Lancet Oncol 2014; 15: 631–639.
254. Van Cutsem E, Rivera F, Berry S et al. Safety and efficacy of first-line 272. Johnsson A, Hagman H, Frodin JE et al. A randomized phase III trial on
bevacizumab with FOLFOX, XELOX, FOLFIRI and fluoropyrimidines in metastatic maintenance treatment with bevacizumab alone or in combination with erlotinib
colorectal cancer: the BEAT study. Ann Oncol 2009; 20: 1842–1847. after chemotherapy and bevacizumab in metastatic colorectal cancer: the Nordic
255. European Medicines Agency. CHMP summary of opinion—erbitux. http://www. ACT Trial. Ann Oncol 2013; 24: 2335–2341.
ema.europa.eu/docs/en_GB/document_library/Summary_of_opinion/human/ 273. Stein A, Atanackovic D, Hildebrandt B et al. Upfront FOLFOXIRI+bevacizumab
000558/WC500155463.pdf (6 January 2015, date last accessed). followed by fluoropyrimidin and bevacizumab maintenance in patients with
256. European Medicines Agency. CHMP summary of opinion—vectibix. http://www. molecularly unselected metastatic colorectal cancer. Br J Cancer 2015; 113:
ema.europa.eu/docs/en_GB/document_library/Summary_of_opinion/human/ 872–877.
000741/WC500144827.pdf (6 January 2015, date last accessed). 274. Garcia Alfonso P, Benavides M, Sanchez Ruiz A et al. Phase II study of first-line
257. Grothey A, Sargent D, Goldberg RM, Schmoll HJ. Survival of patients with mFOLFOX plus cetuximab (C) for 8 cycles followed by mFOLFOX plus C or
advanced colorectal cancer improves with the availability of fluorouracil- single agent (S/A) C as maintenance therapy in patients (P) with metastatic
leucovorin, irinotecan, and oxaliplatin in the course of treatment. J Clin Oncol colorectal cancer (mCRC): The MACRO-2 trial (Spanish Cooperative Group for
2004; 22: 1209–1214. the Treatment of Digestive Tumors [TTD]). Ann Oncol 2014; 25(Suppl 4): abstr
258. Grothey A, Sargent D. Overall survival of patients with advanced colorectal cancer 499O.
correlates with availability of fluorouracil, irinotecan, and oxaliplatin regardless of 275. Cunningham D, Pyrhonen S, James RD et al. Randomised trial of irinotecan plus
whether doublet or single-agent therapy is used first line. J Clin Oncol 2005; 23: supportive care versus supportive care alone after fluorouracil failure for patients
9441–9442. with metastatic colorectal cancer. Lancet 1998; 352: 1413–1418.
259. Heinemann V, Modest DP, Fischer von Weikersthal L et al. Independent radiological 276. Rougier P, Lepille D, Bennouna J et al. Antitumour activity of three second-line
evaluation of objective response, early tumor shrinkage, and depth of response in treatment combinations in patients with metastatic colorectal cancer after optimal
FIRE-3 (AIO KRK-0306). Ann Oncol 2014; 25(Suppl 2): abstr O-0030. 5-FU regimen failure: a randomised, multicentre phase II study. Ann Oncol 2002;
260. Lenz HJ, Niedzwiecki D, Innocenti F et al. CALGB/SWOG 80405: phase III trial of 13: 1558–1567.
irinotecan/5-FU/leucovorin (FOLFIRI) or oxaliplatin/5-FU/leucovorin (mFOLFOX6) 277. Rougier P, Van Cutsem E, Bajetta E et al. Randomised trial of irinotecan versus
with bevacizumab (BV) or cetuximab (CET) for patients ( pts) with expanded RAS fluorouracil by continuous infusion after fluorouracil failure in patients with
analyses untreated metastatic adenocarcinoma of the colon or rectum (mCRC). metastatic colorectal cancer. Lancet 1998; 352: 1407–1412.
Ann Oncol 2014; 25(Suppl 5): abstr 501O. 278. Haller DG, Rothenberg ML, Wong AO et al. Oxaliplatin plus irinotecan compared
261. Maughan TS, James RD, Kerr DJ et al. Comparison of intermittent and with irinotecan alone as second-line treatment after single-agent fluoropyrimidine
continuous palliative chemotherapy for advanced colorectal cancer: a multicentre therapy for metastatic colorectal carcinoma. J Clin Oncol 2008; 26: 4544–4550.
randomised trial. Lancet 2003; 361: 457–464. 279. Koopman M, Antonini NF, Douma J et al. Sequential versus combination
262. Labianca R, Sobrero A, Isa L et al. Intermittent versus continuous chemotherapy chemotherapy with capecitabine, irinotecan, and oxaliplatin in advanced
in advanced colorectal cancer: a randomised ‘GISCAD’ trial. Ann Oncol 2011; colorectal cancer (CAIRO): a phase III randomised controlled trial. Lancet 2007;
22: 1236–1242. 370: 135–142.
263. Tournigand C, Cervantes A, Figer A et al. OPTIMOX1: a randomized study of 280. Seymour MT, Maughan TS, Ledermann JA et al. Different strategies of sequential
FOLFOX4 or FOLFOX7 with oxaliplatin in a stop-and-go fashion in advanced and combination chemotherapy for patients with poor prognosis advanced
colorectal cancer—a GERCOR study. J Clin Oncol 2006; 24: 394–400. colorectal cancer (MRC FOCUS): a randomised controlled trial. Lancet 2007;
264. Adams RA, Meade AM, Seymour MT et al. Intermittent versus continuous 370: 143–152.
oxaliplatin and fluoropyrimidine combination chemotherapy for first-line treatment 281. Tournigand C, Andre T, Achille E et al. FOLFIRI followed by FOLFOX6 or the
of advanced colorectal cancer: results of the randomised phase 3 MRC COIN reverse sequence in advanced colorectal cancer: a randomized GERCOR study. J
trial. Lancet Oncol 2011; 12: 642–653. Clin Oncol 2004; 22: 229–237.
265. Chibaudel B, Maindrault-Goebel F, Lledo G et al. Can chemotherapy be 282. Giantonio BJ, Catalano PJ, Meropol NJ et al. Bevacizumab in combination with
discontinued in unresectable metastatic colorectal cancer? The GERCOR oxaliplatin, fluorouracil, and leucovorin (FOLFOX4) for previously treated
OPTIMOX2 Study. J Clin Oncol 2009; 27: 5727–5733. metastatic colorectal cancer: results from the Eastern Cooperative Oncology
266. Tournigand C, Chibaudel B, Samson B et al. Bevacizumab with or without Group Study E3200. J Clin Oncol 2007; 25: 1539–1544.
erlotinib as maintenance therapy in patients with metastatic colorectal cancer 283. Bennouna J, Sastre J, Arnold D et al. Continuation of bevacizumab after first
(GERCOR DREAM; OPTIMOX3): a randomised, open-label, phase 3 trial. Lancet progression in metastatic colorectal cancer (ML18147): a randomised phase 3
Oncol 2015; 16: 1493–1505. trial. Lancet Oncol 2013; 14: 29–37.
Volume 27 | No. 8 | August 2016 doi:10.1093/annonc/mdw235 | 

by guest
on 05 February 2018
284. Beretta GD, Petrelli F, Stinco S et al. FOLFIRI + bevacizumab as second-line 294. Geva R, Vecchione L, Tejpar S et al. Bevacizumab plus chemotherapy as salvage
therapy for metastatic colorectal cancer pretreated with oxaliplatin: a pooled treatment in chemorefractory patients with metastatic colorectal cancer. Onco
analysis of published trials. Med Oncol 2013; 30: 486. Targets Ther 2013; 6: 53–58.
285. Lievre A, Samalin E, Mitry E et al. Bevacizumab plus FOLFIRI or FOLFOX in 295. Hecht JR, Patnaik A, Berlin J et al. Panitumumab monotherapy in patients
chemotherapy-refractory patients with metastatic colorectal cancer: a with previously treated metastatic colorectal cancer. Cancer 2007; 110:
retrospective study. BMC Cancer 2009; 9: 347. 980–988.
286. Masi G, Salvatore L, Boni L et al. Continuation or reintroduction of bevacizumab 296. Price TJ, Peeters M, Kim TW et al. Panitumumab versus cetuximab in patients
beyond progression to first-line therapy in metastatic colorectal cancer: final with chemotherapy-refractory wild-type KRAS exon 2 metastatic colorectal
results of the randomized BEBYP trial. Ann Oncol 2015; 26: 724–730. cancer (ASPECCT): a randomised, multicentre, open-label, non-inferiority phase
287. Van Cutsem E, Tabernero J, Lakomy R et al. Addition of aflibercept to fluorouracil, 3 study. Lancet Oncol 2014; 15: 569–579.
leucovorin, and irinotecan improves survival in a phase III randomized trial in 297. Grothey A, Van Cutsem E, Sobrero A et al. Regorafenib monotherapy for
patients with metastatic colorectal cancer previously treated with an oxaliplatin- previously treated metastatic colorectal cancer (CORRECT): an international,
based regimen. J Clin Oncol 2012; 30: 3499–3506. multicentre, randomised, placebo-controlled, phase 3 trial. Lancet 2013; 381:
288. Tabernero J, Van Cutsem E, Lakomy R et al. Aflibercept versus placebo in 303–312.
combination with fluorouracil, leucovorin and irinotecan in the treatment of 298. Li J, Qin S, Xu R et al. Regorafenib plus best supportive care versus placebo plus
previously treated metastatic colorectal cancer: prespecified subgroup analyses best supportive care in Asian patients with previously treated metastatic
from the VELOUR trial. Eur J Cancer 2014; 50: 320–331. colorectal cancer (CONCUR): a randomised, double-blind, placebo-controlled,
289. Tabernero J, Yoshino T, Cohn AL et al. Ramucirumab versus placebo in phase 3 trial. Lancet Oncol 2015; 16: 619–629.
combination with second-line FOLFIRI in patients with metastatic colorectal 299. Mayer RJ, Van Cutsem E, Falcone A et al. Randomized trial of TAS-102 for
carcinoma that progressed during or after first-line therapy with bevacizumab, refractory metastatic colorectal cancer. N Engl J Med 2015; 372:
oxaliplatin, and a fluoropyrimidine (RAISE): a randomised, double-blind, 1909–1919.
multicentre, phase 3 study. Lancet Oncol 2015; 16: 499–508. 300. Yoshino T, Mizunuma N, Yamazaki K et al. TAS-102 monotherapy for pretreated
290. Sobrero AF, Maurel J, Fehrenbacher L et al. EPIC: phase III trial of cetuximab plus metastatic colorectal cancer: a double-blind, randomised, placebo-controlled
irinotecan after fluoropyrimidine and oxaliplatin failure in patients with metastatic phase 2 trial. Lancet Oncol 2012; 13: 993–1001.
colorectal cancer. J Clin Oncol 2008; 26: 2311–2319. 301. Papamichael D, Audisio RA, Glimelius B et al. Treatment of colorectal cancer in
291. Cascinu S, Lonardi S, Rosati G et al. A phase III multicenter trial comparing two older patients: International Society of Geriatric Oncology (SIOG) consensus
different sequences of second/third line therapy (cetuximab/irinotecan followed recommendations 2013. Ann Oncol 2015; 26: 463–476.
by FOLFOX versus FOLFOX followed by cetuximab/irinotecan) in metastatic K- 302. Gruenberger B, Tamandl D, Schueller J et al. Bevacizumab, capecitabine, and
RAS wt colorectal cancer (mCC) patients, refractory to FOLFIRI/Bevacizumab. Eur oxaliplatin as neoadjuvant therapy for patients with potentially curable metastatic
J Cancer 2015; 51(Suppl S3): abstr 2006. colorectal cancer. J Clin Oncol 2008; 26: 1830–1835.
292. Hecht JR, Cohn A, Dakhil S et al. SPIRITT: a randomized, multicenter, phase II 303. Garufi C, Torsello A, Tumolo S et al. Cetuximab plus chronomodulated irinotecan,
study of panitumumab with FOLFIRI and bevacizumab with FOLFIRI as second- 5-fluorouracil, leucovorin and oxaliplatin as neoadjuvant chemotherapy in
line treatment in patients with unresectable wild type KRAS metastatic colorectal colorectal liver metastases: POCHER trial. Br J Cancer 2010; 103: 1542–1547.
cancer. Clin Colorectal Cancer 2015; 14: 72–80. 304. Wong R, Cunningham D, Barbachano Y et al. A multicentre study of
293. Cunningham D, Humblet Y, Siena S et al. Cetuximab monotherapy and cetuximab capecitabine, oxaliplatin plus bevacizumab as perioperative treatment of patients
plus irinotecan in irinotecan-refractory metastatic colorectal cancer. N Engl J with poor-risk colorectal liver-only metastases not selected for upfront resection.
Med 2004; 351: 337–345. Ann Oncol 2011; 22: 2042–2048.
 | Van Cutsem et al. Volume 27 | No. 8 | August 2016

by guest
on 05 February 2018
Renal cell carcinoma: ESMO Clinical Practice Guidelines

B. Escudier1, C. Porta2, M. Schmidinger3, N. Rioux-Leclercq4, A. Bex5, V. Khoo6,7, V. Gruenvald8 &
A. Horwich9 on behalf of the ESMO Guidelines Committee*
1
Department of Medical Oncology, Gustave Roussy, Villejuif, France; 2IRCCS San Matteo University Hospital Foundation, Pavia, Italy; 3Department of Medicine I, Clinical
Division of Oncology and Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria; 4Department of Pathology, Rennes Hospital and University, Rennes,
France; 5Division of Surgical Oncology—Urology, The Netherlands Cancer Institute, Amsterdam, The Netherlands; 6Institute of Cancer Research, Royal Marsden Hospital,
London, UK; 7University of Melbourne and Monash University, Victoria, Australia; 8Clinic for Hematology, Hemostasis, Oncology and Stem Cell Transplantation, Medical
School Hanover, Hanover, Germany; 9The Institute of Cancer Research, London, UK
incidence and epidemiology diagnosis

Kidney cancer accounts for 5% and 3% of all adult malignancies As stated above, >50% of RCCs are currently detected inciden-
in men and women, respectively, thus representing the 7th most tally, making the classical triad of flank pain, gross haematuria
common cancer in men and the 10th most common cancer in and palpable abdominal mass less frequent than in the past.
women [1]. However, available statistics include not only renal Despite this, RCC remains the ‘Internist’s cancer’ with paraneo-
clinical practice
parenchymal tumours, but also urothelial cancer of the renal plastic syndromes such as hypercalcaemia, unexplained fever,
guidelines
pelvis; renal cell carcinoma (RCC) accounts for ∼80% of all erythrocytosis and Stauffer’s syndrome (signs of cholestasis un-
kidney cancers. related to tumour infiltration of the liver or intrinsic liver
After over two decades of increasing rates, RCC incidence disease, which typically resolve after kidney tumour resection),
trends worldwide have shown signs of plateauing or decreasing still being relatively frequent.
in recent years. Furthermore, kidney cancer mortality rates Suspicion of RCC should prompt laboratory examinations of
overall have levelled. These patterns are consistent with reports serum creatinine, haemoglobin, leukocyte and platelet counts,
of incidental diagnosis and downward shift of tumour stage and lymphocyte to neutrophil ratio, lactate dehydrogenase, C-react-
size; indeed, the widespread use of non-invasive radiological ive protein (CRP) and serum-corrected calcium, in addition to
techniques [e.g. ultrasonography (US), computed tomography other symptom-derived tests [IV, B]. Some of these tests are
(CT)] allows the frequent detection of early and small RCCs, prognosticators for survival and are used for risk assessment
which are potentially curable. within different prognostic score systems (see later).
Beyond well-known risk factors for RCC, such as cigarette Most cases of RCC are strongly suspected by imaging.
smoking, obesity and hypertension, evidence is accumulating to Diagnosis is usually suggested by US and further investigated by
suggest an aetiological or, on the contrary, a protective role, for CT scan, which allows for assessment of local invasiveness,
additional factors [2], such as trichloroethylene. Furthermore, lymph node involvement, or distant metastases. Magnetic res-
RCC also appears to be more common in patients with end- onance imaging (MRI) may provide additional information in
stage renal failure or acquired renal cystic disease, and in investigating local advancement and venous involvement by
patients on dialysis, those who have had kidney transplantation, tumour thrombus.
or those with tuberous sclerosis syndrome. For accurate staging of RCC, contrast-enhanced chest, ab-
Approximately 2%–3% of all RCCs are hereditary and several dominal, and pelvic CT is mandatory [III, A]; unless indicated
autosomal dominant syndromes are described, each with a dis- by clinical or laboratory signs or symptoms, the use of bone
tinct genetic basis and phenotype, the most common one being scan or CT (or MRI) of the brain is not recommended for
Von Hippel Lindau (VHL) disease. routine clinical practice [III, A]. In case of an allergy to CT con-
trast medium, adequate staging should include a high-resolution
CT scan of the chest without contrast medium, together with an
abdominal MRI. 18Fluorodeoxyglucose-positron emission tom-
6962 Viganello-Lugano, Switzerland. ography (18FDG-PET) is not a standard investigation in the
E-mail: clinicalguidelines@esmo.org diagnosis and staging of clear cell RCC (ccRCC) and should not
†
be used. The role of new tracers is under investigation only.
Approved by the ESMO Guidelines Committee: September 2008, last update August
2016. This publication supersedes the previously published version—Ann Oncol 2014; A renal tumour core biopsy provides histopathological con-
25 (Suppl. 3): iii49–iii56. firmation of malignancy with high sensitivity and specificity; it

by guest
on 05 February 2018
is especially recommended before treatment with ablative ther- - In the 2004 WHO classification of RCC, the maximum size
apies [III, B] as well as in patients with metastatic disease before of papillary adenoma was 5 mm. From now, papillary
starting systemic treatment [III, B]. Nowadays, complications adenoma is defined as a papillary renal tumour ≤15 mm in
(e.g. bleeding or tumour seeding) are rare or even exceptional its largest dimension.
(as in the case of tumour seeding) [3], while diagnostic accuracy - The main prognostic factors in chromophobe RCC are
remains high [4]. The final histopathological diagnosis, classifi- tumour stage, the presence of necrosis, a sarcomatoid and/or
cation, grading and evaluation of prognostic factors are based rhabdoid component and small vessel invasion.
on the nephrectomy specimen when available. - Hybrid tumours present overlapping features of oncocyto-
mas and chromophobe RCC. They have indolent behaviour
and are usually observed in Birt–Hogg–Dubé syndrome.
pathology assessment - The diagnosis of the highly aggressive collecting duct carcin-
The new edition of the World Health Organization (WHO) oma is based on six histological features: medullary location,
histological classification of renal tumours has been recently infiltrative growth pattern, tubular architecture, desmoplastic
reported (Table 1) and was based on tumour histology, chromo- stromal reaction, high-grade atypia and that the tumour is
somal alterations and molecular pathways [5]. Changes in the neither an RCC nor a transitional cell carcinoma.
2016 WHO classification are as follows: - Renal medullary carcinoma occurs in young patients with
sickle traits and is histologically similar to collecting duct car-
- The molecular genetic profile of ccRCC is characterised in cinoma.
>80% of sporadic cases, by biallelic VHL gene alterations: - MiTF translocation RCCs harbour gene fusions involving
VHL gene mutations, hypermethylation of VHL gene pro- mainly TFE3 and TFEB genes, and occur in young patients.
moter, and loss of heterozygosity. Mutations in chromatin re- The median age at diagnosis is 31 years. The diagnosis is
modelling genes have also been reported (PBRM1: 41%, based on a strong nuclear TFE3/TFEB immunoreactivity and
BAP1: 8%–10%, and SETD2: 11.5%) and are associated with the presence of a translocation involving TFE3 or TFEB
an increased risk of ccRCC-related death. genes: t(X;1)( p11.2;q21) and t(6,11)( p21;q12), respectively.
- Multilocular cystic ccRCC has been renamed as multilocular - Among angiomyolipomas (AMLs), epithelioid AML is now
cystic renal neoplasia of low malignant potential due to its in- recognised as a separate entity with a risk of progression or
dolent behaviour. metastasis. The prognostic factors for the risk of progression
- Even though papillary RCC are histologically and cytogen- are association with tuberous sclerosis, multiple AML, the
etically defined by two main subtypes, type 1 and type 2, they presence of necrosis, tumour size >7 cm, extrarenal extension
represent a heterogeneous disease including tumours with in- and/or renal vein invasion and the presence of a carcinoma-
dolent outcome and tumours with aggressive and lethal like growth pattern.
phenotype. RCCs associated with the hereditary leiomyoma- - Two new entities are now recognised:
tosis are usually type 2 papillary RCC and have a poor prog- ○ acquired cystic disease-associated RCCs that have indolent
nosis with a high risk of dissemination. outcome and occur in patients with end-stage renal disease
- The oncocytic variant of papillary RCC should be reclassified and acquired cystic disease and
as type 1 (mainly) or type 2 papillary RCC. ○ succinate dehydrogenase (SDH)-deficient RCCs that occur
in patients with germline mutations in an SDH gene,

Table 1. WHO 2016 classification of renal cell tumours leading to a dysfunction of mitochondrial complex II.
Clear cell renal cell carcinoma

Multilocular cystic renal neoplasm of low malignant potential The prognostic factors validated by the International Society
Papillary renal cell carcinoma of Urological Pathology (ISUP) consensus, and the new WHO
Hereditary leiomyomatosis and renal cell carcinoma-associated renal cell 2016 classification of RCC, and to be reported in routine prac-
carcinoma tice are as follows [6]:
Chromophobe renal cell carcinoma
Collecting duct carcinoma - The tumour histological subtype
Renal medullary carcinoma - The ISUP nucleolar grading system that should be applied
MiT family translocation renal cell carcinoma only to ccRCC and papillary RCC
Succinate dehydrogenase-deficient renal cell carcinoma - A sarcomatoid and/or rhabdoid differentiation that defines a
Mucinous tubular and spindle cell carcinoma grade 4 tumour
Tubulocystic renal cell carcinoma - The presence of necrosis
Acquired cystic disease-associated renal cell carcinoma - The presence of microscopic vascular invasion
Clear cell papillary renal cell carcinoma - The pTNM staging
Unclassified renal cell carcinoma
Papillary adenoma
Oncocytoma
biology
Reprinted with permission from Moch et al. [5]. Beyond the classical one gene–one histology paradigm, a more
WHO, World Health Organization. complex biological classification of RCC (and especially of its
clear cell histotype) is currently emerging [7].
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw328 | v

by guest
on 05 February 2018
First, RCC proved to be an extremely heterogeneous disease recurrence-free survival was reported [16], indicating that a
[8]; beyond the seminal genetic alteration (mutation, deletion or plateau has been reached for prognostication with available
hypermethylation) of the VHL tumour suppressor gene, which models. Hence, no clear preference for a specific prognostic
is present in the vast majority of sporadic RCCs, other genetic model may be given.
alterations may occur, especially over time [9], contributing to
worsen the prognosis of patients harbouring these tumours. advanced disease. The Memorial Sloane Kettering Cancer
Notably enough, three of these other genes (PBRM1, BAP1, and Centre (MSKCC) was the gold standard for the risk assessment
SETD2) are located on the same short arm of chromosome 3 during cytokine treatment in metastatic (m)RCC [17]. Its
where the VHL gene is also located. Genetic abnormalities to applicability to targeted agents was shown more recently [18].
these genes seem to increase tumour aggressiveness [10], defin- Further refinement was introduced with the International
ing these cancers as ‘diseases of chromosome 3p’. Metastatic RCC Database Consortium (IMDC) score, which
On the contrary, some RCCs are characterised by mutations extended the previous factors to a total number of 6 to increase
in the mTOR pathway and especially in the highly conserved concordance [19, 20]:
FAT (FRAP–ATM–TTRAP) and kinase domains of the MTOR
gene; these cancers have been defined as metabolic RCCs [11]. • Karnofsky performance status (PS) <80%
When metastatic, they are thought to be more sensitive to • Haemoglobin <lower limit of normal
mTOR inhibitors [12]. • Time from diagnosis to treatment of <1 year
Finally, according to another comprehensive molecular char- • Corrected calcium above the upper limit of normal
acterisation of papillary RCCs, type 1 and type 2 papillary RCCs • Platelets greater than the upper limit of normal
were shown to be clinically and biologically distinct. Alterations • Neutrophils greater than the upper limit of normal
in the MET pathway were indeed associated with type 1, and ac-
A recent evaluation of this model in second-line treatment
tivation of the NRF2-ARE pathway was associated with type 2,
underscored its predictive value in previously treated mRCC
while CDKN2A loss, and a CpG island methylator phenotype in
[21] (Table 5).
type 2 contributed to convey a poor prognosis. Based on this
genomic profile, type 2 papillary RCC consisted of at least three
subtypes based on molecular and phenotypic features [13]. molecular prognostication. Gene signatures were known to
Finally, some of the escape mechanisms, namely cMET detect different risk groups in RCC [22]. More recently, a 16-
(cabozantinib) and FGF (fibroblast growth factor, lenvatinib) gene assay was shown to improve prediction of recurrence-free
activation, have been used to develop new strategies in vascular survival in localised RCC compared with the SSIGN score
endothelial growth factor (VEGF) refractory patients. according to Leibovich (concordance: 0.81versus 0.74) [23].
These data indicate that molecular analysis may exert add-
itional benefit to already established clinical and histo-anatom-
staging and risk assessment ical parameters, which may lead to an individual risk
assessment in the future. Other putative markers such as circu-
staging
lating DNA, microRNA or DNA methylation status were shown
The American Joint Committee on Cancer (AJCC)/Union for to have prognostic relevance in RCC and warrant future investi-
International Cancer Control (UICC) tumour–node–metastasis gation. As of today, no specific molecular marker can be recom-
(TNM) staging system should be used (Table 2). mended for clinical use.
risk assessment
The natural clinical course varies in RCC, which has led to the management of local/locoregional
development of different prognostic models for the assessment disease
of the patient’s individual risk. Extent of disease, histology, A summary of the recommendations for treatment of localised
grading and clinical factors have been recognised as having and locally advanced disease is shown in Table 6.
prognostic value in RCC and may be used in localised or meta-
static disease [5].
T1 tumours (<7 cm)
localised disease. Different pre- or postoperative scores have • Partial nephrectomy (PN) is recommended as the preferred
been applied to assess prognosis in RCC, which are used for option in organ-confined tumours measuring up to 7 cm
risk-adapted follow-up strategies. Integrated prognostic scores (elective indication). This is based on a systematic review in-
offer some predictive advantages over single tumour cluding multiple retrospective studies and a prospective ran-
characteristics and are used preferentially. These models are domised, controlled trial (RCT) which compared radical
composed of histological and clinical factors. The most recent nephrectomy (RN) with PN in solitary T1a-b N0M0 renal
modifications of the stage, size, grade and necrosis (SSIGN) tumours <5 cm with normal contralateral kidney function [I,
score [14] (Table 3) and the University of California Los A] [24].
Angeles Integrated Staging System (UISS) (Table 4) [15] score • PN can be carried out via open, laparoscopic or laparoscopic
are frequently used. robot-assisted approaches.
However, among different prognostic scores, a concordance • Laparoscopic RN is recommended if PN is not technically
of 0.68–0.89 for cancer-specific survival (CSS) and 0.74–0.82 for feasible.
v | Escudier et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Table 2. Staging of RCC (AJCC/UICC TNM classification of malignant tumours, 7th edition)
Primary tumour (T)

T1 Tumour ≤7 cm in greatest dimension, limited to the kidney
T1a Tumour ≤4.0 cm
T1b Tumour >4.0 cm but ≤7.0 cm
T2 Tumour >7.0 cm in greatest dimension, limited to the kidney
T2a Tumour >7 cm but ≤10 cm
T2b Tumour >10 cm, limited to the kidney
T3 Tumour extends into major veins or perinephric tissues but not into the ipsilateral adrenal gland and not beyond Gerota’s fascia
T3a Tumour grossly extends into the renal vein or its segmental (muscle
containing) branches, or tumour invades perirenal and/or renal sinus fat
(peripelvic) but not beyond Gerota’s fascia
T3b Tumour grossly extends into the vena cava below the diaphragm
T3c Tumour grossly extends into the vena cava above the diaphragm or
invades the wall of the vena cava
T4 Tumour invades beyond Gerota’s fascia (including contiguous extension into the ipsilateral adrenal gland)

N1 Metastasis in regional lymph node(s)
N2 More than one lymph node involved
Distant metastases (M)
cM0 Clinically no distant metastasis

cM1 Clinically distant metastasis
pM1 Pathologically proven distant metastasis, e.g. needle biopsy
Stage I T1 N0 M0
Stage II T2 N0 M0
Stage III T1–2 N1 M0
T3 Any M0
Stage IV T4 Any M0
Any Any M1
Used with the permission of the American Joint Committee on Cancer (AJCC), Chicago, IL, USA. The original source for this material is the AJCC Cancer
Staging Handbook, 7th edition (2010) published by Springer Science and Business Media LLC, www.springerlink.com.
RCC, renal cell carcinoma; AJCC, American Joint Committee on Cancer; UICC, Union for International Cancer Control; TNM, tumour–node–metastases.
• In patients with compromised renal function, solitary kidney • Systematic reviews of RFA and PN suggest that RFA has a
or bilateral tumours, PN is also the standard of care, with no long-term CSS equal to PN with a low metastasis rate but
tumour size limitation (imperative indication). slightly higher local recurrence rate compared with PN and
• Systematic reviews comparing surgical management of localised CA [25]. The quality of the available evidence prevents defini-
RCC (T1-2N0M0) were unable to identify prospective comparative conclusions regarding morbidity and oncological out-
tive studies reporting on oncological outcomes for minimally in- comes for RFA and CA [III].
vasive ablative procedures compared with RN [24]. • Active surveillance is an option in elderly patients with sig-
• Radio frequency ablation (RFA) or cryoablation (CA) treat- nificant co-morbidities or those with a short-life expect-
ments are options in patients with small cortical tumours ancy and solid renal tumours measuring <40 mm. The
(≤3 cm), especially for patients who are frail, present a high growth of renal tumours (mean 3 mm/year) is low in most
surgical risk and those with a solitary kidney, compromised cases, and progression to metastatic disease is reported in
renal function, hereditary RCC or multiple bilateral tumours. 1%–2% [26]. Renal biopsy is recommended to select
Renal biopsy is recommended to confirm malignancy and patients with small masses for active surveillance [III] with
subtype in this setting. high accuracy [3, 4].
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw328 | v

by guest
on 05 February 2018
Table 3. SSIGN score for localised RCC [14] Table 4. UISS (UCLA Integrated Staging System)
Features Score Patient group Prognostic group
T stage Fuhrman’s ECOG 5-year disease-
Pathological T category of primary tumour (TNM 2002)
grade status specific survival
pT1a 0
pT1b 2 Localized disease (N0, M0)
pT2 3 Low risk 1 1–2 0 91.1%
pT3a-4 4
Intermediate 1 1–2 1 or more 80.4%
Regional lymph node status (TNM 2002) risk 1 3–4 Any
pNx or pN0 0 2 Any Any
pN1 or pN2 2 3 1 Any
3 2–4 Any
Tumour size
<10 cm 0 High risk 3 2–4 1 or more 54.7%
10 cm or more 1 4 Any Any
Metastatic disease
Nuclear grade
Low risk N1M0 Any Any 32%
1 or 2 0
N2M0/M1 1–2 0
3 1
4 3 Intermediate N2M0/M1 1–2 1 or more 19.5%
risk 3 0, 1, or
Histological tumour necrosis
more
No 0
4 0
Yes 1
High risk N2M0/M1 4 1 or more 0%
Scores Group 5-year metastasis-free survival
0–2 Low risk 97.1% Risk groups and 5-year disease-specific survival.
3–5 Intermediate risk 73.8% UCLA, University of California Los Angeles; ECOG, Eastern
6 or more High risk 31.2% Cooperative Oncology Group.
SSIGN, size, stage, grade, and necrosis; RCC, renal cell carcinoma; TNM,
tumour–node–metastases. Table 5. Median overall survival estimates in first- and second-line
according to IMDC risk groups
Number of Risk category First-line [8] Second-line [9]
risk factors median OS median OS
T2 tumours (>7 cm) (months) (months)
Laparoscopic RN is the preferred option.
0 Favourable 43.2 35.3
1–2 Intermediate 22.5 16.6
locally advanced RCC (T3 and T4) 3–6 Unfavourable 7.8 5.4
• Open RN remains the standard of care even though a laparo-
scopic approach can be considered. IMDC, International Metastatic RCC Database Consortium; OS, overall
• Systematic adrenalectomy or extensive lymph node dissection survival; RCC, renal cell carcinoma.
is not recommended when abdominal CT shows no evidence
of adrenal or lymph node invasion [27].
• The evidence regarding management of venous tumour placebo [28]. However, recently a press release announced
thrombus is based on retrospective studies with significant that the S-TRAC trial, comparing sunitinib to placebo in
risks of bias and confounding. Resection of venous thrombi is high-risk localised RCC, met its primary end point. Full data
challenging and associated with a high risk of complications. should be presented at the ESMO 2016 meeting. Depending
Surgical intervention should be considered, but the most ef- on these data, the role of sunitinib in the adjuvant setting will
fective approach remains unknown and outcome depends on have to be discussed. Neoadjuvant approaches are experimen-
tumour thrombus level [III]. tal and should not be proposed outside clinical trials.
• Currently, there is no evidence from randomised phase III • Attempting to downsize venous tumour thrombi with system-
trials that adjuvant therapy is of survival benefit or prolongs ic targeted therapy cannot be recommended.
disease-free survival (DFS). Several RCTs of adjuvant suniti-
nib, sorafenib, pazopanib, axitinib and everolimus are
ongoing. Data from a large adjuvant trial of sunitinib versus management of metastatic disease
sorafenib versus placebo were reported in 2015 (ASSURE)
after an interim analysis carried out with 62% information. role of surgery and local therapy
Results demonstrated no significant differences in DFS or ○ In the era of immunotherapy, cytoreductive nephrectomy was
overall survival (OS) between the experimental arms and recommended in patients with good PS [I, A] [29]. Whether
v | Escudier et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Table 6. Recommendations for the treatment of localised and locally advanced RCC
Level of evidence and grade of
recommendation
Partial nephrectomy is recommended for the treatment of all T1 tumours if negative margins are obtained and III, C
risk of morbidity is acceptable.
Laparoscopic radical nephrectomy is the preferred option for the treatment of organ-confined RCC (stages II, B
T1T2N0NxM0) when partial nephrectomy is not feasible.
Routine adrenalectomy and lymph node dissection are not required for all radical nephrectomies. III, D
Open radical nephrectomy with the goal of obtaining negative margins is still the standard of care for locally III, C
advanced RCC.
Ablative treatments are options in patients with small cortical tumours (≤3 cm) and age >70 years, high surgical III, C
risk, solitary kidney, compromised renal function, hereditary RCC or multiple bilateral tumours.
RCC, renal cell carcinoma.
this recommendation will remain with current targeted ther- option [II, C]. The safety of observation has also been suggested
apies is currently being investigated in two prospective trials. by retrospective and prospective studies.
In routine practice, cytoreductive nephrectomy is recom-
mended in patients with good PS and large primary tumours first-line treatment of patients with good or intermediate
with limited volumes of metastatic disease and for patients prognosis.
with a symptomatic primary lesion. Cytoreductive nephrec-
tomy is not recommended in patients with poor PS [III, B]. • Three treatments have demonstrated efficacy in pivotal phase
○ Metastasectomy and other local treatment strategies including III trials: bevacizumab (combined with interferon), sunitinib
whole brain radiotherapy (WBRT), conventional radiotherapy, and pazopanib [30–32]. All three drugs have been registered
stereotactic radiosurgery (SRS), stereotactic body radiotherapy based on improvement of progression-free survival (PFS) over
(SBRT), cyberknife radiotherapy and hypofractionated radio- either interferon or placebo. More recently, pazopanib has
therapy can be considered and carried out for selected patients been shown not to be inferior to sunitinib in a large phase III
after multidisciplinary review. A recent systematic review of 16 trial [33]. Efficacy of both sunitinib and pazopanib has been
studies including 2350 patients sought to identify the evidence confirmed by real-world evidence studies. These two tyrosine
base for local treatment strategies of metastases from RCC kinase inhibitors (TKIs) are currently the most commonly
[25]. The results consistently point towards a benefit of com- used treatments. Considering all of the published trials, the
plete metastasectomy for OS and CSS. No systemic treatment level of recommendation is considered to be [I, A] for all three
is recommended after metastasectomy. regimens.
○ No general guidelines can be given as to whether a patient • Sorafenib [II, B], high-dose interleukin-2 [III, C], and low-
should be referred for local treatment of metastases. Patient dose interferon combined with bevacizumab [III, A] are alter-
selection should be discussed in a multidisciplinary team. native options.
Good PS, solitary or oligo metastases, metachronous disease • Single-agent interferon-alpha, as the inferior arm of three
with disease-free interval >2 years, the absence of progression RCTs, should no longer be regarded as a standard option [I,
on systemic therapy, low or intermediate Fuhrmann grade and D]. There is currently no evidence that new checkpoint inhi-
complete resection have been associated with favourable bitors should be used in first line, although numerous
outcome after local treatment of metastases from RCC. ongoing trials are exploring their role, either as monotherapy
or in combination (with either VEGF inhibitors or other
checkpoint inhibitors).
systemic treatment • Interestingly, very recently, cabozantinib has been reported to
An algorithm for systemic treatment in mRCC is presented in be superior to sunitinib in a randomised phase 2 trial. If these
Figure 1. results are confirmed, the role of cabozantinib in the first-line
Recommendations mainly relate to clear cell histology, since setting will have to be assessed.
most of the pivotal trials have been done in this common histo-
logical subtype. In addition, recommendations will differ
first-line treatment of patients with poor prognosis.
according to risk stratification (see above).
The time to start systemic therapy is not well defined. Because • Temsirolimus is currently the only drug tested in a phase III
some RCCs have a very indolent course, a period of observation study, demonstrating evidence of activity in this patient popu-
before starting treatment should be considered, especially in lation [II, A] [34]. The pivotal trial demonstrated improve-
patients with limited tumour burden and few symptoms. ment of OS compared with interferon or the combination of
Indeed, the outcome of patients who crossed over to an active temsirolimus and interferon.
agent after a brief period of treatment with placebo, within • Based on subgroup analysis from the pivotal trial, as well as
placebo-controlled phase III trials, indirectly supports this expanded access programmes, sunitinib is another reasonable
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw328 | v

by guest
on 05 February 2018
Clear cell histology Non clear cell histology
Standard:
Sunitinib [II, B]
Good or intermediate risk Poor risk
Option:
Temsirolimus [III, B]
Sorafenib [III, B]
Pazopanib [III, B]
Standard: Standard: Everolimus [III, B]
Sunitinib [I, A] Temsirolimus [II, A]
First line
Bevacizumab + IFN [I, A]
treatment:
Pazopanib [I, A] Option:
Sunitinib [II, B]
Option: Sorafenib [III, B]
High dose IL2 [III, C] Pazopanib [III, B]
Sorafenib [II, B]
Bevacizumab + low dose IFN
[III, B]
Post cytokines Post TKIs
Second line Standard: Standard:

treatment: Axitinib [II, A] Nivolumab [I, A; MCBS 5]
Sorafenib [I, A] Cabozantinib [I, A]
Pazopanib [II, A]
Option:
Option: Axitinib [II, B]
Sunitinib [III, A] Everolimus [II, B]
Sorafenib [III, B]
Post 2 TKIs Post TKI and mTOR Post TKI / nivolumab Post TKI / Cabozantinib
Third line Sorafenib [I, B] Standard: Standard:

Standard:
treatment: Nivolumab [V, A] Cabozantinib [V, A] Nivolumab [V, A]
Nivolumab [II, A]
Cabozantinib [V, A]
Cabozantinib [II, A]
Option: Option:
Option: Axitinib [IV, C] Everolimus [V, B]
Option:
Other TKI [IV, B] Everolimus [IV, C] Axitinib [V, B]
Everolimus [II, B]
Rechallenge [IV, B]
Figure 1. Algorithm for systemic treatment in mRCC. mRCC, metastatic renal cell carcinoma; IFN, interferon; IL2, interleukin 2; TKI, tyrosine kinase inhibi-
tor; mTOR, mammalian target of rapamycin; MCBS, ESMO Magnitude of Clinical Benefit Scale v1.0.
option in this setting [II, B]. Sorafenib as well as pazopanib, ○ Based on recent phase III trials, sorafenib can also be used
based on expanded access programmes or real-world evidence as an option [III, B].
studies, are other possible alternatives [III, B].
• There is no clear recommendation on whether temsirolimus • However, second-line treatment has recently been dramatical-
or TKIs should be used in poor risk patients, although TKIs ly modified by the report of two large trials showing improve-
are more commonly used in patients with good PS (expert ment in OS with nivolumab [an anti-programmed death 1
opinion). The advantage of using TKIs in this setting will be (PD-1) inhibitor] and cabozantinib [38–40] over everolimus.
to use second-line recommendations below, as some patients Both trials showed very significant improvement in OS and
in the second-line trials were in the poor prognostic group. response rate, while PFS was improved only in the cabozanti-
• It is clear that, for some poor prognosis patients, best support- nib trial. In both trials, patients could be treated after either
ive care remains the only suitable treatment option. one or two TKIs.
• Obviously, availability of these two drugs is still very limited,
and several situations should be differentiated:
second-line treatment. ○ Only nivolumab is available: It should be recommended [I,
A; ESMO-Magnitude of Clinical Benefit Scale (MCBS)

• Evidence that TKIs are active after cytokines has been seen v1.0 score: 5].
with sorafenib [I, A], pazopanib [II, A] and, recently, axitinib ○ Nivolumab and cabozantinib are both available: either drug
[II, A] [32, 35, 36]. Sunitinib also has activity is this setting is recommended [I, A; ESMO-MCBS v1.0 score: 5 (nivolu-
[III, A]. However, since VEGF-targeted therapy is now the mab)]
first-line standard of care, the number of patients treated with ○ Neither of these drugs is available: either everolimus [II, B]
cytokines is decreasing. or axitinib [II, B] can be used.

• After first-line treatment with VEGF-targeted therapy,
○ Both axitinib [II, B] and everolimus [II, B] are active [36, • Of note, the combination of lenvatinib and everolimus has re-
37]. Both drugs have shown significantly improved PFS cently been approved by the FDA (Food and Drug
over sorafenib (axitinib) or placebo (everolimus). Administration) based on a randomised study of 150 patients,
v | Escudier et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
showing PFS and OS benefit over everolimus [41]. Recently in chromosome 7 was shown to lead to a loss of the folliculin gene
Europe, the CHMP (Committee for Medicinal Products for with up-regulation of mTOR. Finally, collecting duct tumours
Human Use) gave a positive opinion for this combination. (and also medullary carcinomas) were reported to behave more
However, based on the size of this study, this combination like aggressive urothelial tumours rather than RCCs and may
cannot, at this stage, be added to current European guidelines. therefore be considered for chemotherapy. None of these
• Finally, the optimal duration of treatment, especially for nivo- ‘genetic’ recommendations can be graded, as data are limited
lumab, remains questionable, as well as the benefit of treat- and no clear treatment recommendation can be made for these
ment beyond progression. subgroups with distinct biology.
third-line treatment. Beyond second-line treatment, enrolment role of radiotherapy and bisphosphonates
into clinical trials is recommended where possible. However, The spectrum of radiosensitivity in RCC is wide, but it is not a
based on recent trials with nivolumab and cabozantinib, radioresistant disease. Radiotherapy has been shown to provide
different situations should be defined: good symptom palliation and local control in RCC depending
on the dose that can be delivered [46]. There is a developing ra-
• In patients already treated with two TKIs, either nivolumab or
tionale with emerging data suggesting that the apparent radiore-
cabozantinib is recommended [II, A]. If neither of these drugs
sistance of RCC can be overcome through the ceramide pathway
is available, everolimus remains the standard option [II, B]
with the use of higher dose per fraction treatments usually deliv-
• In patients previously treated with one TKI and nivolumab,
ered by new high-precision radiotherapy methods such as SBRT
cabozantinib is recommended, if available [V, A]. In the
[IV, B] [47]. This can be exploited and used in many different
absence of cabozantinib, either everolimus or axitinib can be
clinical situations particularly for unresectable local recurrences
used [IV, C].
or oligometastatic disease.
• In patients previously treated with one TKI and cabozantinib,
nivolumab is recommended [V, A], and either everolimus or • There is no current evidence for the use of radiotherapy in the
axitinib remains an acceptable option [V, B]. neoadjuvant or adjuvant setting. This is on the basis of four
• In patients previously treated with VEGF-targeted therapy negative ‘old’ trials with two pre-operative and two adjuvant
and an mTOR inhibitor, sorafenib [II, B] has shown activity studies. Despite being randomised trials, there are several
[42]. In addition, nivolumab or cabozantinib can be recom- major limitations in trial design and methodology that
mended in this setting [V, A]. Finally, another TKI or rechal- included inappropriate case selection, sub-therapeutic radio-
lenge with the same TKI is considered as an option [IV, B]. therapy regimes and inadequate patient numbers.
Furthermore, treatment morbidity was substantially high and
medical treatment of metastatic disease of non-clear cell the radiotherapy techniques used then have now been super-
histology. For these patients, enrolment into specifically seded by improved modern irradiation methods such as in-
designed clinical trials is strongly recommended. However, tensity-modulated radiotherapy or SBRT [II, D].
small prospective trials as well as subgroup analyses from larger • Radiotherapy can be used to treat unresectable local or recur-
trials have recently been reported [43–45]. In these trials, rent disease with the aim of improving local control. For
sunitinib and everolimus have been compared, and in every patients in whom surgery cannot be carried out due to poor
trial, there is a trend in favour of sunitinib. In addition, some PS or unsuitable clinical condition, radiotherapy can be an al-
recommendations can be provided according to the results of ternative if other local therapies such as radioablation are not
the expanded access programmes of sunitinib and sorafenib, of appropriate. Modern image-guided radiotherapy techniques
small retrospective studies, and of the subgroup analysis of the are needed to enable a high biological dose to be delivered,
temsirolimus registration trial. Overall, sunitinib has the most such as volumetric modulated arc therapy (VMAT) or SBRT
reproducible efficacy [II, B]. These studies also suggest that [IV, B]. As discussed earlier, there is an emerging role for its
patients with non-clear cell histology may benefit from use in the synchronous or metachronous development of oli-
treatment with everolimus [III, B], sorafenib, pazopanib or gometastatic mRCC disease, oligoprogression or in mixed re-
temsirolimus [III, B]. However, in most of these studies, only sponse scenarios with immuno- or targeted therapies [IV, B].
patients with papillary and chromophobe tumours were • Radiotherapy is an effective treatment for palliation of local
enrolled. and symptomatic mRCC disease or to prevent the progression
In the absence of prospective data, genetic considerations of metastatic disease in critical sites: bones, brain [I, A]. For
may influence treatment decisions: in papillary type 1 tumours, symptomatic bone metastasis, local radiotherapy (either as a
activation of the c-MET pathway has commonly been reported. single fraction or as fractionated course) can provide good
Novel agents inhibiting the cMET receptor are currently under symptom relief in up to two-thirds of cases with complete
investigation. However, as the c-MET receptor and VEGF-re- symptomatic responses in up to 20%–25% [1, A].
ceptor were shown to cooperate, VEGF-inhibiting agents may • For the management of spinal cord compression, an ambula-
be a reasonable choice. Similarly, there is no evidence for the tory status at diagnosis and limited metastatic disease are fa-
optimal treatment of papillary type 2, which is characterised by vourable prognostic factors in those patients able to undergo
inactivation of the fumarate-hydratase gene, fumarate accumu- surgery. The use of initial surgery and postoperative radio-
lation and HIF upregulation. Again, VEGF inhibitors may be therapy was reported in a randomised trial to improve sur-
considered in this context. Patients with chromophobe RCC vival and maintenance of ambulation compared with
may benefit from mTOR inhibitors since mutation on irradiation alone [48–50] [1, A].
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw328 | v

by guest
on 05 February 2018
• In the management of mRCC patient with brain metastases, its location and potential consequences (see sections above on
the use of corticosteroids can provide effective temporary radiotherapy palliation and spinal cord compression). In wide-
relief of cerebral symptoms. WBRT between 20 and 30 Gy in spread mRCC bone metastasis, bisphosphonate therapy with
4–10 fractions, respectively, is effective for symptom control zoledronic acid has been shown to significantly reduce skeletal-
[II, B]. Most trials in brain metastasis include only a small related events (SREs) in patients and increase time to first SRE
proportion of RCC cases [48–50]. With the use of SRS deliver- [51]. Denosumab is a synthetic RANK ligand inhibitor that may
ing larger doses per fraction, the mRCC response outcomes have a greater bone effect as it is capable of reaching all sites
are not thought to differ from other solid tumours. For the within bone (being a circulating antibody), compared with
subset of good prognosis patients with a single unresectable bisphosphonates, which have a greater affinity for sites of active
brain metastasis, SRS with or without WBRT should be con- bone turnover. Denosumab has been shown in a randomised
sidered [II, A]. There is less reported late cognitive dysfunc- trial to extend the time to first SRE by 4.3 months and was non-
tion using SRS alone compared with the combination therapy inferior to zoledronic acid [52]. In addition, denosumab has the
[II, A]. Adequate control of brain metastases before initiation convenience of subcutaneous administration with no require-
of anti-VEGF therapy is recommended (expert opinion). ment for renal monitoring or dose adjustment [I, A]. Bone-
targeted therapy with either zoledronic acid or denosumab
Multidisciplinary management is needed to optimise care for should be considered in mRCC patients with reasonable life ex-
mRCC patients suffering from bone metastasis. The approach pectancy and widespread bony metastasis weighting the poten-
will need to be individualised to the extent of bone metastasis, tial benefits of the treatment (supposed benefit in terms of OS)
Table 7. Magnitude of Clinical Benefit Scale (MCBS) table for new therapies/indications in renal cell carcinomaa
Therapy Disease Trial Control Absolute Hazard QoL/toxicity MCBS
setting survival ratio (95% scoreb
gain CI)
Nivolumab, a Advanced Study of nivolumab versus Everolimus, in patients OS gain: OS: HR for Improved 5 (Form
PD-1 everolimus in pre-treated with renal cell 5.4 death toxicity 2a)
checkpoint advanced or metastatic clear carcinoma who had months 0.73 profile and
inhibitor cell renal cell carcinoma received previous TKI (0.57– QoL
(CheckMate 025) [40] treatment. Median OS 0.93)
Phase III 19.6 months
NCT01668784
CI, confidence interval; QoL, quality of life; PD-1, programmed death 1; TKI, tyrosine kinase inhibitor; OS, overall survival; HR, hazard ratio.
a
EMA approvals in 2016 to end of August 2016.
b
ESMO-MCBS version 1.0 [53].
Levels of evidence
C Insufficient evidence for efficacy or benefit does not outweigh the risk or the disadvantages (adverse events, costs, etc.), optional
a
v | Escudier et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
with the potential harms (risk of osteonecrosis of the mandible) to synchronise the surgical recommendations with the surgical
[II, A]. Further trials are ongoing to explore its other applications. recommendations of the EAU RCC guidelines.
personalised medicine conflict of interest

In this disease setting, more research is needed to identify mo-
BE has reported honoraria for lectures and advisory boards
lecular markers which could lead to advances in personalised
received from Novartis, Pfizer, Bristol-Myers Squibb, Exelixis,
medicine.
Roche, Ipsen, and Eisai. CP acted as a consultant and/or speaker
for Novartis, Pfizer, Bristol-Myers Squibb, Roche/Genentech,
response evaluation, follow-up, long- Exelixis, Ipsen, Eisai and Peloton. MS has reported honoraria
for lectures and advisory boards from Pfizer, Roche, Novartis,
term implications and survivorship Exelixis, Eisai, Bristol-Myers Squibb, Aveo and Astellas. AB took
So far, there is no evidence that early treatment of metastasis part in advisory boards for Pfizer, Novartis, Nektar, and Eisai.
results in better outcome compared with delayed treatment. VG has reported honoraria from Bristol-Myers Squibb,
Overall, there is no evidence that any particular follow-up proto- Novartis, Pfizer, and Roche, and advisory role in Bristol-Myers
col influences the outcome in early RCC. No standard recom- Squibb, Exelixis, Ipsen, Novartis, and Pfizer. All remaining
mendation can be given for the follow-up in advanced RCC authors have declared no conflicts of interest.
either.
The follow-up scheme for localised RCC following surgery
should depend on the therapeutic possibilities upon recurrence. references
CT scans of thorax and abdomen are routinely carried out, with
1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA Cancer J Clin 2016;
time intervals depending on risk factors. It is recommended to 66: 7–30.
perform CT scans every 3–6 months in high-risk patients for 2. Chow WH, Dong LM, Devesa SS. Epidemiology and risk factors for kidney cancer.
the first 2 years, while a yearly CT scan is probably sufficient in Nat Rev Urol 2010; 7: 245–257.
low-risk patients (expert opinion). 3. Volpe A, Kachura JR, Geddie WR et al. Techniques, safety and accuracy of
Long-term follow-up is proposed in some institutions, due to sampling of renal tumors by fine needle aspiration and core biopsy. J Urol 2007;
the possibility of late relapse, but its benefit has never been 178: 379–386.
demonstrated. 4. Marconi L, Dabestani S, Lam TB et al. Systematic review and meta-analysis of
During systemic therapy in mRCC patients, 2- to 4-month diagnostic accuracy of percutaneous renal tumour biopsy. Eur Urol 2016; 69:
660–673.
follow-up schemes with CT scan should be advised to determine
5. Moch H, Cubilla AL, Humphrey PA et al. The 2016 WHO classification of tumours
response and resistance. Although not perfect, RECIST
of the urinary system and male genital organs-part A: renal, penile, and testicular
(Response Evaluation Criteria in Solid Tumours) criteria tumours. Eur Urol 2016; 70: 93–105.
remain the most frequently used method to assess drug efficacy. 6. Delahunt B, Cheville JC, Martignoni G et al. The International Society of Urological
However, in the case of RECIST-defined disease progression, Pathology (ISUP) grading system for renal cell carcinoma and other prognostic
there is no clinical evidence that this quantity of progression is a parameters. Am J Surg Pathol 2013; 37: 1490–1504.
clinically valid end point that should require treatment interrup- 7. Brugarolas J. Molecular genetics of clear-cell renal cell carcinoma. J Clin Oncol
tion or modification. 2014; 32: 1968–1976.
8. Gerlinger M, Rowan AJ, Horswell S et al. Intratumor heterogeneity and branched
evolution revealed by multiregion sequencing. N Engl J Med 2012; 366:
methodology 883–892.
9. Gerlinger M, Horswell S, Larkin J et al. Genomic architecture and evolution of clear
These clinical practice guidelines were developed in accordance cell renal cell carcinomas defined by multiregion sequencing. Nat Genet 2014; 46:
with the ESMO standard operating procedures for clinical prac- 225–233.
tice guidelines development, www.esmo.org/Guidelines/ESMO- 10. Hakimi AA, Ostrovnaya I, Reva B et al. Adverse outcomes in clear cell renal cell
Guidelines-Methodology. The relevant literature has been carcinoma with mutations of 3p21 epigenetic regulators BAP1 and SETD2: a
selected by the expert authors. An MCBS table with ESMO- report by MSKCC and the KIRC TCGA research network. Clin Cancer Res 2013;
19: 3259–3267.
MCBS scores is included in Table 7. ESMO-MCBS v1.0 [53] was
11. Hakimi AA, Pham CG, Hsieh JJ. A clear picture of renal cell carcinoma. Nat Genet
used to calculate scores for new therapies/indications approved
2013; 45: 849–850.
by the EMA since 1st January 2016. Levels of evidence and 12. Voss MH, Hakimi AA, Pham CG et al. Tumor genetic analyses of patients with
grades of recommendation have been applied using the system metastatic renal cell carcinoma and extended benefit from mTOR inhibitor therapy.
shown in Table 8. Statements without grading were considered Clin Cancer Res 2014; 20: 1955–1964.
justified standard clinical practice by the experts and the ESMO 13. Cancer Genome Atlas Research NetworkLinehan WM, Spellman PT et al.
faculty. This manuscript has been subjected to an anonymous Comprehensive molecular characterization of papillary renal-cell carcinoma. N Engl
peer-review process. J Med 2016; 374: 135–145.
14. Leibovich BC, Blute ML, Cheville JC et al. Prediction of progression after radical
nephrectomy for patients with clear cell renal cell carcinoma: a stratification tool
acknowledgements for prospective clinical trials. Cancer 2003; 97: 1663–1671.
15. Zisman A, Pantuck AJ, Dorey F et al. Improved prognostication of renal cell
AB took part as practicing urologist and not as vice chair of the carcinoma using an integrated staging system. J Clin Oncol 2001; 19:
European Association of Urology (EAU) RCC guideline panel 1649–1657.
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw328 | v

by guest
on 05 February 2018
16. Sun M, Shariat SF, Cheng C et al. Prognostic factors and predictive models in renal 37. Motzer RJ, Escudier B, Oudard S et al. Efficacy of everolimus in advanced renal
cell carcinoma: a contemporary review. Eur Urol 2011; 60: 644–661. cell carcinoma: a double-blind, randomised, placebo-controlled phase III trial.
17. Motzer RJ, Mazumdar M, Bacik J et al. Survival and prognostic stratification of Lancet 2008; 372: 449–456.
670 patients with advanced renal cell carcinoma. J Clin Oncol 1999; 17: 38. Choueiri TK, Escudier B, Powles T et al. Cabozantinib versus everolimus in
2530–2540. advanced renal-cell carcinoma. N Engl J Med 2015; 373: 1814–1823.
18. Motzer RJ, Escudier B, Bukowski R et al. Prognostic factors for survival in 1059 39. Choueiri TK, Escudier B, Powles T et al. Cabozantinib versus everolimus in
patients treated with sunitinib for metastatic renal cell carcinoma. Br J Cancer advanced renal cell carcinoma (METEOR): final results from a randomised, open-
2013; 108: 2470–2477. label, phase 3 trial. Lancet Oncol 2016; 17: 917–927.
19. Heng DY, Xie W, Regan MM et al. Prognostic factors for overall survival in patients 40. Motzer RJ, Escudier B, McDermott DF et al. Nivolumab versus everolimus in
with metastatic renal cell carcinoma treated with vascular endothelial growth advanced renal-cell carcinoma. N Engl J Med 2015; 373: 1803–1813.
factor-targeted agents: results from a large, multicenter study. J Clin Oncol 2009; 41. Motzer RJ, Hutson TE, Glen H et al. Lenvatinib, everolimus, and the combination in
27: 5794–5799. patients with metastatic renal cell carcinoma: a randomised, phase 2, open-label,
20. Heng DY, Xie W, Regan MM et al. External validation and comparison with other multicentre trial. Lancet Oncol 2015; 16: 1473–1482.
models of the International Metastatic Renal-Cell Carcinoma Database Consortium 42. Motzer RJ, Porta C, Vogelzang NJ et al. Dovitinib versus sorafenib for third-line
prognostic model: a population-based study. Lancet Oncol 2013; 14: 141–148. targeted treatment of patients with metastatic renal cell carcinoma: an open-label,
21. Ko JJ, Xie W, Kroeger N et al. The International Metastatic Renal Cell Carcinoma randomised phase 3 trial. Lancet Oncol 2014; 15: 286–296.
Database Consortium model as a prognostic tool in patients with metastatic renal 43. Motzer RJ, Barrios CH, Kim TM et al. Phase II randomized trial comparing
cell carcinoma previously treated with first-line targeted therapy: a population- sequential first-line everolimus and second-line sunitinib versus first-line sunitinib
based study. Lancet Oncol 2015; 16: 293–300. and second-line everolimus in patients with metastatic renal cell carcinoma. J Clin
22. Jones J, Otu H, Spentzos D et al. Gene signatures of progression and metastasis Oncol 2014; 32: 2765–2772.
in renal cell cancer. Clin Cancer Res 2005; 11: 5730–5739. 44. Armstrong AJ, Halabi S, Eisen T et al. Everolimus versus sunitinib for patients with
23. Rini B, Goddard A, Knezevic D et al. A 16-gene assay to predict recurrence after metastatic non-clear cell renal cell carcinoma (ASPEN): a multicentre, open-label,
surgery in localised renal cell carcinoma: development and validation studies. randomised phase 2 trial. Lancet Oncol 2016; 17: 378–388.
Lancet Oncol 2015; 16: 676–685. 45. Tannir NM, Jonasch E, Albiges L et al. Everolimus versus sunitinib prospective
24. MacLennan S, Imamura M, Lapitan MC et al. Systematic review of oncological evaluation in metastatic non-clear cell renal cell carcinoma (ESPN): a randomized
outcomes following surgical management of localised renal cancer. Eur Urol 2012; multicenter phase 2 trial. Eur Urol 2016; 69: 866–874.
61: 972–993. 46. Khoo VS, Pyle L. Radiotherapy and supportive care. In Eisen T, Christmas T (eds),
25. Dabestani S, Marconi L, Hofmann F et al. Local treatments for metastases of renal Clinical Progress in Renal Cancer, Vol 15. Oxford: Informa UK Ltd 2007;
cell carcinoma: a systematic review. Lancet Oncol 2014; 15: e549–e561. 191–201. ISBN 10-1-84184-604-X.
26. Jewett MA, Mattar K, Basiuk J et al. Active surveillance of small renal masses: 47. De Meerleer G, Khoo V, Escudier B et al. Radiotherapy for renal-cell carcinoma.
progression patterns of early stage kidney cancer. Eur Urol 2011; 60: 39–44. Lancet Oncol 2014; 15: e170–e177.
27. Bekema HJ, MacLennan S, Imamura M et al. Systematic review of adrenalectomy 48. Patchell RA, Tibbs PA, Regine WF et al. Direct decompressive surgical resection in
and lymph node dissection in locally advanced renal cell carcinoma. Eur Urol the treatment of spinal cord compression caused by metastatic cancer: a
2013; 64: 799–810. randomised trial. Lancet 2005; 366: 643–648.
28. Haas NB, Manola J, Uzzo RG et al. Adjuvant sunitinib or sorafenib for high-risk, 49. Andrews DW, Scott CB, Sperduto PW et al. Whole brain radiation therapy with or
non-metastatic renal-cell carcinoma (ECOG-ACRIN E2805): a double-blind, without stereotactic radiosurgery boost for patients with one to three brain
placebo-controlled, randomised, phase 3 trial. Lancet 2016; 387: 2008–2016. metastases: phase III results of the RTOG 9508 randomised trial. Lancet 2004;
29. Flanigan RC, Mickisch G, Sylvester R et al. Cytoreductive nephrectomy in patients 363: 1665–1672.
with metastatic renal cancer: a combined analysis. J Urol 2004; 171: 50. Kocher M, Soffietti R, Abacioglu U et al. Adjuvant whole-brain radiotherapy versus
1071–1076. observation after radiosurgery or surgical resection of one to three cerebral
30. Escudier B, Pluzanska A, Koralewski P et al. Bevacizumab plus interferon alfa-2a metastases: results of the EORTC 22952-26001 study. J Clin Oncol 2011; 29:
for treatment of metastatic renal cell carcinoma: a randomised, double-blind phase 134–141.
III trial. Lancet 2007; 370: 2103–2111. 51. Henry DH, Costa L, Goldwasser F et al. Randomized, double-blind study of
31. Motzer RJ, Hutson TE, Tomczak P et al. Sunitinib versus interferon alfa in denosumab versus zoledronic acid in the treatment of bone metastases in patients
metastatic renal-cell carcinoma. N Engl J Med 2007; 356: 115–124. with advanced cancer (excluding breast and prostate cancer) or multiple myeloma.
32. Sternberg CN, Davis ID, Mardiak J et al. Pazopanib in locally advanced or J Clin Oncol 2011; 29: 1125–1132.
metastatic renal cell carcinoma: results of a randomized phase III trial. J Clin Oncol 52. Rosen LS, Gordon D, Tchekmedyian NS et al. Long-term efficacy and safety of
2010; 28: 1061–1068. zoledronic acid in the treatment of skeletal metastases in patients with nonsmall
33. Motzer RJ, Hutson TE, Cella D et al. Pazopanib versus sunitinib in metastatic renal- cell lung carcinoma and other solid tumors: a randomized, Phase III, double-blind,
cell carcinoma. N Engl J Med 2013; 369: 722–731. placebo-controlled trial. Cancer 2004; 100: 2613–2621.
34. Hudes G, Carducci M, Tomczak P et al. Temsirolimus, interferon alfa, or both for 53. Cherny NI, Sullivan R, Dafni U et al. A standardised, generic, validated approach to
advanced renal-cell carcinoma. N Engl J Med 2007; 356: 2271–2281. stratify the magnitude of clinical benefit that can be anticipated from anti-cancer
35. Escudier B, Eisen T, Stadler WM et al. Sorafenib in advanced clear-cell renal-cell therapies: the European Society for Medical Oncology Magnitude of Clinical
carcinoma. N Engl J Med 2007; 356: 125–134. Benefit Scale (ESMO-MCBS). Ann Oncol 2015; 26: 1547–1573.
36. Rini BI, Escudier B, Tomczak P et al. Comparative effectiveness of axitinib versus 54. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
sorafenib in advanced renal cell carcinoma (AXIS): a randomised phase 3 trial. among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
Lancet 2011; 378: 1931–1939. 139–144.
v | Escudier et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Published online 5 August 2014
Bladder cancer: ESMO Practice Guidelines

J. Bellmunt1,2, A. Orsola3, J. J. Leow1,2, T. Wiegel4, M. De Santis5 & A. Horwich6 on behalf of the
ESMO Guidelines Working Group*
1
Department of Medical Oncology, University Hospital del Mar-IMIM, Barcelona, Spain; 2Bladder Cancer Center, Dana-Farber Cancer Institute/Brigham and Women’s
Cancer Center, Boston, USA; 3Department of Urology, Vall d’Hebron University Hospital, Barcelona, Spain; 4Department of Radio Oncology, University Hospital Ulm, Ulm,
Germany; 5Ludwig Boltzmann Institute for Applied Cancer Research, Kaiser Franz Josef- Spital, Vienna, Austria; 6Institute of Cancer Research and Royal Marsden Hospital,
Sutton, UK
incidence and epidemiology and adenocarcinomas [4]. This guideline relates to transitional
cell carcinoma.
In Europe, an estimated 151 297 new cases of bladder cancer
were diagnosed in 2012, with an age-standardised incidence rate
( per 100 000 persons) of 17.7 for males and 3.5 for females.
Overall, the annual crude incidence rate is 20.4/100 000. In A complete history and physical examination should be under-
2012, there were 52 395 deaths from bladder cancer with an taken, together with laboratory tests evaluating full blood counts
clinical practice
annual crude mortality rate of 7.1/100 000 [1]. Approximately and renal function. Bladder ultrasonography most frequently
guidelines
70% of patients with bladder cancer are >65 years of age. gives an initial suspicious image, but final diagnosis of bladder
The most common presenting symptom is painless haema- cancer is based on cystoscopy and evaluation of the resected
turia, seen in >80% of patients. Others may also present with tissue. Cystoscopic examination and TURBT under anaesthesia
irritative symptoms such as dysuria, frequency or urgency. should be carried out following a standardised protocol
Symptoms of metastases such as bone or flank pain are rare. (Figure 1). Complete resection of all tumour tissue should be
Most diagnosed cases of muscle-invasive bladder cancer (MIBC; achieved when possible. At the time of TURBT, the number of
80%–90%) present as primary invasive bladder cancer. However, tumours, their size(s) and the presence of extra-vesical extension
up to 15% of patients have a history of non-muscle-invasive or invasion of adjacent organs by bimanual examination should
bladder cancer (NMIBC), mainly high-risk cases. be documented. Ideally, both the base of the tumour and the
tumour edges should be sent separately to the pathologist to
pathological diagnosis ensure the presence of lamina propria and bladder muscle in the
specimen, essential for accurate staging.
Pathological diagnosis should be made according to the World Because associated carcinoma in situ (CIS) has been shown to
Health Organisation (WHO) classification (Table 1) from a be an adverse prognostic factor, bladder biopsies should be taken
biopsy obtained during transurethral resection of the bladder from reddish, suspicious areas when present or random biopsies
tumour (TURBT). Tumours should be graded as high and low from normal looking urothelium if there is a positive cytology or
grade according to the latest WHO criteria and can concomi- a previous diagnosis of associated CIS. Similarly, biopsies from
tantly be graded according to the 1973 classifications of high, the prostatic urethra should be taken if the tumour is located at
low and intermediate grade carcinoma [3]. Ninety percent of the trigone or bladder neck area, or when there is no bladder
bladder carcinomas are transitional cell carcinomas. The other tumour and the procedure is carried out to study a positive cy-
types of urothelial cancer are relatively uncommon, including tology, since the tumour could be located in the urothelium
lymphoepithelioma-like or sarcomatoid carcinomas, micropapil- lining the prostatic urethra or the ducts [III, C] [6]. Management
lary or nested variants and primary squamous cell carcinomas of bladder cancer is based on the pathological findings of the
biopsy, with attention to histology, grade and depth of invasion.
MIBC should be staged according to the tumour–node–metasta-
*Correspondence to: ESMO Guidelines Working Group, ESMO Head Office, Via L. Taddei
sis (TNM) system and grouped into categories (Table 2).
4, CH-6962 Viganello-Lugano, Switzerland. Once histology confirms muscle invasion, local staging can be
E-mail: clinicalguidelines@esmo.org carried out with further imaging studies such as computed tom-
† ography (CT) or magnetic resonance imaging. Either test can
Approved by the ESMO Guidelines Working Group: February 2011, last update June
2014. This publication supersedes the previously published version—Ann Oncol 2011; be used to assess extra-vesical invasion but these tests are often
22 (Suppl 6): vi45–vi49. unable to reliably differentiate T2 from T3a, T3b or even T4a.

by guest
on 05 February 2018
Importantly, because of interference by post-TURBT peri- management of local/locoregional
vesical reactions, imaging is recommended before TURBT, if disease
possible, when an invasive tumour is suspected (by ultrasound
or cystoscopy). Similarly, both tests are useful to detect enlarged treatment of non-muscle-invasive bladder cancer
nodes—over 8 mm in the pelvic area and over 1 cm for abdom- Complete TURBT is the treatment of choice for any initial
inal nodes—and distant metastasis. Hydronephrosis should also bladder tumour [9], followed by instillations according to risk
be taken into account as it has been shown to be an independent stratification in NMIBC [I, A]. A second TURBT is a reasonable
predictor of advanced stage bladder cancer and poor clinical option in high-risk NMIBC tumours, either before intravesical
outcome, and it predicts extra-vesical disease and node-positive therapy [II, B] or thereafter [III, B]. Presentations with very
disease [8]. A chest CT should be carried out at the same time as high-risk features, e.g. multiple grade 3 T1 tumours with TIS or
the abdomino-pelvis CT. Because a synchronous upper tract increased depth of invasion, may be considered for cystectomy.
urothelial tumour may exist in 2.5% of patients, upper urinary In case of TIS or high-grade T1 failing Bacillus Calmette–
tract imaging with either CT urograms, or i.v. or retrograde pye- Guérin (BCG), cystectomy should be considered due to the high
lograms should be undertaken to exclude this. In patients with risk of progression [III, B] (Figure 3).
high risk of metastases, additional tests may be undertaken, for
example, bone scans and chest imaging.
treatment of muscle-invasive bladder cancer
Radical cystectomy (RC) with extended lymphadenectomy is
usually considered to be the standard treatment of MIBC [10].
Table 1. Pathologic Diagnosis of Urothelial Carcinoma of the Extended lymphadenectomy has potentially been shown to
Bladder (WHO/ISUP 1998 Consensus; WHO, 2004) be beneficial [III, A], and may be curative in patients with me-
tastasis or micro-metastasis to a few nodes. Progression-free
Papilloma survival (PFS) and overall survival (OS) have been correlated
Papillary urothelial neoplasm of low malignant potential with number of lymph nodes removed during surgery.
Urothelial carcinoma low grade
Reconstruction may be carried out either by ileal conduit or
Urothelial carcinoma high grade
bladder replacement, depending on tumour characteristics and
patient choice. Age is no longer a limiting factor for surgery,
Reprinted with permission from [2].
WHO, World Health Organization; ISUP, International Society of
even though postoperative morbidity increases with age [11].
Urological Pathology.
External beam radiotherapy may be considered as a curative
therapeutic option as part of a multimodality bladder-preserv-
ing approach [III]. When the patient is unfit for cystectomy,
Prognostic factors in first line advanced disease

1.0
n = 199
Risk factors:
0.8 0 = KPS > 80, no visceral mets
1 = KPS < 80, or visceral mets
2 = KPS < 80, and visceral mets
Proportion surviving
0.6
0.4 No risk factors

33.0 m.
13.4/13.6 m.
0.2
9.3 m. One risk factor
0.0 Two risk factors
0 12 24 36 48 60 72 84 96 108 120 132 144 156 168

Time in months
Bajorin, D.F. et al. J Clin Oncol 1999; 17: 3173–3181
Copyright © American Society of Clinical Oncology
Figure 1. Prognostic factors in first-line advanced disease. Reprinted from [5] with permission of © 1999 American Society of Clinical Oncology. All rights reserved .
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu223 | iii

by guest
on 05 February 2018
Table 2. TNM staging system for urothelial carcinoma of the

carbogen nicotinamide) to radiotherapy significantly reduced the
bladder risk of relapse and death [16]. A third and recently published mul-
ticentre randomised trial (the BC2001 trial) has demonstrated
Stage I T1 N0 M0 improved results for chemoradiotherapy using the combination of
Stage II T2a-T2b N0 M0 5-fluorouracil and mitomycin C in terms of locoregional control
Stage III T3a-T3b, T4a N0 M0 [17]. A cystoscopy with bladder biopsy is mandatory for response
Stage IV T4b N0 M0 evaluation, either midway through treatment or 2–3 months there-
Any T N1-N3 M0 after. If persistent or recurrent disease is observed at response
Any T Any N M1 evaluation or during follow-up (cystoscopy and urinary cytology
every 3 months during the first 2 years, and every 6 months there-
Reprinted from [7]. Used with the permission of the American Joint after), prompt salvage cystectomy is recommended when possible
Committee on Cancer (AJCC), Chicago, Illinois. The original source [II, A].
for this material is the AJCC Cancer Staging Handbook, Seventh Over the past 20 years, organ preservation by trimodality
Edition (2010) published by Springer Science and Business Media
treatment has been investigated in prospective series from single
LLC, www.springerlink.com.
centres and cooperative groups, with more than 1000 patients
included [18]. Generally, ∼20% of patients will present with re-
sidual tumour at re-staging, and ≤70% of the patients are
tumour free after the first cystoscopy control. An additional
radiotherapy can also be offered for palliation (bleeding, pain). 20%–30% of patients with initial complete response will develop
Curative external beam radiotherapy should be delivered with de novo or recurrent disease in the preserved bladder, requiring
3D conformal radiation therapy or intensity-modulated radio- additional treatment. Patients require the same regular follow-
therapy techniques, and ideally with image guidance. up as with radiotherapy (see previous paragraph). However,
during follow-up, one-quarter of these individuals developed a
neoadjuvant and adjuvant therapy new lesion requiring additional treatment. Five-year OS rates in
The use of cisplatin-based neoadjuvant chemotherapy for the range of 50%–60% have been reported, and about three-
bladder cancer is supported by a meta-analysis of 11 rando- quarters of the surviving patients retained their bladder [19, 20].
mised trials including 3005 patients. There was a 5% absolute Clinical criteria helpful in determining whether patients are
increase in 5-year OS and a 9% absolute increase in 5-year ideal for bladder preservation include early tumour stage (in-
disease-free survival (DFS) compared with RC alone [12]. This cluding high-risk T1 disease [21], T2 <5 cm), a visibly complete
demonstrated survival benefit encourages the use of platinum- TURBT, absence of associated CIS and ureteral obstruction and
based combination chemotherapy before RC or definitive radio- adequate bladder capacity and function [22]. Close coordination
therapy [I, A]. Alternatively, for adjuvant chemotherapy, an among all disciplines and the willingness of the patients to
updated meta-analysis of nine randomised trials including 945 undergo lifelong surveillance are required to achieve optimal
patients found an OS benefit [hazard ratio (HR) 0.77, 95% con- results.
fidence interval (CI) 0.59–0.99, P = 0.049] and DFS benefit (HR
0.66, 95% CI 0.45–0.91, P = 0.014) among those who received
cisplatin-based adjuvant chemotherapy. The DFS benefit was management of advanced and
more apparent among those with positive lymph node involve- metastatic disease
ment [13]. While there is still insufficient evidence for the
routine use of adjuvant chemotherapy in clinical practice [I, A] Cisplatin-containing combination chemotherapy with GC
[14], it is likely that high-risk patients, such as those with extra- (gemcitabine/cisplatin), or MVAC (methotrexate, vinblastine,
vesical and/or node-positive disease that have not received adriamycin and cisplatin) is standard in advanced surgically
neoadjuvant chemotherapy, will benefit most from adjuvant unresectable and metastatic patients fit enough to tolerate cisplatin
chemotherapy. [I, A]. Median survival in these patients is about 14 months; long-
term DFS has been reported in about 15% of patients; in 20.9%
with lymph-node-only disease compared with only 6.8% with vis-
organ preservation therapy ceral metastases [23–25]. So far, no improvement in survival has
The approach of organ preservation therapy for MIBC is a rea- been achieved with newer triplets, novel four-drug regimens or
sonable option for patients seeking an alternative to cystectomy dose-dense chemotherapy [26–28]. GC is less toxic than MVAC
and a palliative option for those who are medically unfit for [I, A] [25]. MVAC is better tolerated with the use of granulocyte
surgery [III, B]. Contemporary protocols utilise aggressive endo- colony-stimulating factor (G-CSF) [29, 30] [III, B]. High-dose
scopic TURBT alone, TURBT plus radiotherapy, TURBT plus intensity MVAC with G-CSF, delivered in half the time of trad-
chemotherapy or—as the preferred treatment—a tri-modality itional MVAC, is an option for fit patients with limited advanced
combination of TURBT plus radiotherapy and chemotherapy. disease, given its lower toxicity profile and superior response
The initial prospective, randomised comparison of radiotherapy rate compared with standard MVAC [31]. The addition of a
alone versus concomitant chemoradiotherapy in bladder cancer third agent ( paclitaxel) to GC has been shown to be of some
demonstrated an improved local control rate when cisplatin was benefit in a subset of patients having the bladder as the primary
given in conjunction with radiotherapy [II, A] [15]. A second trial origin of the disease [I, B], and should be considered as an
showed that the addition of carbogen and nicotinamide (bladder option in highly selected patients [28]. Performance status (PS)
iii | Bellmunt et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
(Karnofsky PS of 80% or less) and the presence of visceral me- independent, adverse prognostic factors for survival (PS >0,
tastases are independent poor prognostic factors for survival [5] haemoglobin level <10 g/dl, and the presence of liver metastasis)
(Figure 1). for patients failing platinum-based chemotherapy have been
About 50% of patients are unfit for cisplatin-containing defined and validated (Figure 2). These factors should therefore
chemotherapy due to a poor PS, impaired renal function or co- be considered for stratification in future trials and for assessing
morbidity. Patients unfit for cisplatin-based chemotherapy may phase II data [34].
be palliated with a carboplatin-based regimen or single-agent The only valid randomised phase III trial in patients progres-
taxane or gemcitabine. Methotrexate/carboplatin/vinblastine (M- sing after first-line treatment with platinum-containing combin-
CAVI) and carboplatin/gemcitabine (CarboGem) are active in ation chemotherapy for metastatic disease tested vinflunine, a
patients unfit for cisplatin, but without a statistically significant novel third-generation vinca alkaloid, plus best supportive care
difference in OS and PFS [I, A]. Severe acute toxicity was slightly (BSC) versus BSC alone [35]. The results showed modest activity
higher on M-CAVI, which makes CarboGem the preferred and (overall response rate 8.6%), a clinical benefit with a favourable
reference treatment in unfit patients [I, A] [32]. Patients with PS safety profile and a survival benefit in favour of vinflunine,
2 and impaired renal function and unfit patients in Bajorin prog- which was statistically significant in the eligible patient population.
nostic group 2 have limited benefit from combination chemo- This trial reached the highest level of evidence ever reported for
therapy, and new strategies are needed [II, A] [32]. second-line treatment. In Europe, vinflunine is the only approved
Selected patients with locally advanced disease (T4b N1) may drug in this setting [I, B]: however, it is unknown whether other
be candidates for cystectomy and lymph node dissection or de- agents used in this setting would have a similar benefit.
finitive radiotherapy following systemic therapy [33]. The role of
anti-angiogenic therapy is investigational in first- and second-line
therapy. personalised medicine
Palliative radiotherapy may be used to reduce symptoms such Overall, personalised cancer therapies hold the promise to
as pain or bleeding. The role of consolidative radiation therapy improve clinical outcomes, using readily obtainable biomarkers
after chemotherapy in patients with locoregional relapses is of response to predict their clinical benefits.
under evaluation [III, B]. Advanced technologies such as high-throughput transcript
profiling, microarrays, metabolomics and proteomics have pro-
treatment of relapse vided us with tools to enhance our understanding of the mo-
Second-line phase II data are highly variable with results de- lecular pathways underlying bladder cancer. Intense research
pending on patient selection. Response rates for treatment of efforts in this area have led to the discovery of numerous mo-
relapse with mono-chemotherapy are lower than with combina- lecular markers that may be useful for screening, early diagnosis
tions, but PFS has been short with both options. Recently, and surveillance as well as staging and prognosis [36]. Current
Prognostic factors in second line

• 4 subgroups formed, based on the presence of 0, 1, 2 or 3 prognostic factors
Strata:
Risk = 0 Censored risk = 0 Risk = 1 Censored risk = 1
Risk = 2 Censored risk = 2 Risk = 3
Median (months) 95% CI:
1.00 Risk 0 = 11.5 (9.3 to 17.9; n = 65)
Risk 1 = 7.3 (5.6 to 8.3; n = 64)
Overall survival (proportion)
Variables Ref. Risk 2 = 3.8 (2.8 to 5.4; n = 19)

at baseline category Risk 3 = 2.4 (1.6 to 3.0; n = 3)
0.75
Haemoglobin < 10 g/dL Log-rank test: P = < .0001
Liver No
involvement involvement 0.50
ECOG-PS ≥1
0.25
0 5.0 7.5 10.0 15.0 12.5 15.0 17.5 20.0

Fig2. Kaplan-Meier estimates for each risk group (zero, Time (months)
one, two, or three factors) in the CA 183001 phase II
study, including (A) all patients (N = 151)
Kaplan-Meier estimates of each risk group

Bellmunt J, et al. J Clin Oncol. 2010;28:1850–1855
Figure 2. Prognostic factors in second line. Reprinted from [34] with permission of © 2010 American Society of Clinical Oncology. All rights reserved.
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu223 | iii

by guest
on 05 February 2018
Denotes recommended therapy

Denotes optional therapy
Presentation
1. Painless haematuria (80% of patients)
2. Irritative symptoms (e.g. dysuria, frequency, urgency) (invasive or high grade tumours)
3. Bone pain (advanced metastatic cases) or flank pain (from retroperitoneal metastases or ureteral obstruction)
Workup
1. History and physical examination
2. Cystoscopic evaluation including biopsy by transurethral resection (TUR) with bimanual examination
3. Urine cytology
4. Blood work (Haematology and biochemistry)
5. Upper urinary tract imaging (mainly CT urogram, alternatively intravenous or retrograde pyelogram) (to exclude 2.5% of patients
who have synchronous upper tract urothelial cancer)
6. Metastatic workup in patients with high risk of metastases [CT scan of chest, abdomen and pelvis, liver function tests, bone scan
(especially in those with bone pain, elevated calcium or alkaline phosphatase)]
Staging and Grading (refer to Tables 1 and 2)
Management of Local Disease Organ Preservation Therapy Management of Metastatic Disease
Management of Local Disease
Depending on the findings at TUR
Non-muscle Invasive (NMIBC): Muscle Invasive

ideally TUR should have been
complete and followed by
IPOMC* Neoadjuvant
Chemotherapy
Radical Cystectomy
with
Lymphadenectomy
Contemplate 2nd
TURBT if:
incomplete initial
TUR, no muscle
present in
specimen or Further Adjuvant Adjuvant
high-risk NMIBC Chemotherapy Chemotherapy
(limited data) (if no neoadjuvant)
Every three months follow up (see text)

Intravesical instillations (Mitomycin C or BCG)
according to risk group
Cystoscopic Surveillance according to risk
group (for high risk, 10 year follow-up)
* IPOMC: immediate postoperative Mitomycin C
Figure 3. Overview of clinical management of patient with suspected bladder cancer (local disease, organ preservation therapy, and metastatic disease).
iii | Bellmunt et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
Organ Preservation Therapy
(cystectomy ineligible, patient preference)
TURBT (should aim for complete)
Combined Radiotherapy
Combined
chemoradiotherapy alone
chemoradiotherapy
Total dose 55-64 Gy
40 Gy
Imaging + TURBT evaluation
Salvage
cystectomy if
persisting Complete
tumour Radiotherapy if
complete
response (CR)
Surveillance
Cystoscopic evaluation (3 months)
with TUR bladder biopsy
(every 6 months)
Management of Metastatic Disease

Patients with poor
comorbid status or impaired
renal function “unfit”
Carboplatin-based
regimens or single-
agents: taxane, gemcitabine
PS £ 2 + Poor renal Cisplatin-based

function combination
chemotherapy (e.g.
MVAC, GC,
HDMVAC, PCG)
Clinical trial
Best Supportive Care
Progression < 12 months Progression > 12 months

Second line chemotherapy
1.-Platinum based
1.-Vinflunine rechallenge
2.-Taxane based
3.-Clinical trial
Fig. 3 Continued
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu223 | iii

by guest
on 05 February 2018

G1-S cell cycle progression such as TP53 and RB1 have been the
(adapted from the Infectious Diseases Society of America-United most consistently reported [40].
States Public Health Service Grading Systema) In addition, a large proportion of urothelial tumours also
harbour mutation and/or gene amplification that are potentially
Levels of evidence therapeutic targets, and these include FGFR3 mutations, PTEN
I Evidence from at least one large randomised, controlled trial deletions and FGFR1, CCND1 and MDM2 amplifications [41].
of good methodological quality (low potential for bias) or Moreover, aberrations of the chromatin remodelling genes
meta-analyses of well-conducted randomised trials without (UTX, MLL-MLL3, CREBBP-EP300, NCOR1, ARID1A and CHD6)
heterogeneity and, more recently, STAG2 mutations have also been documen-
II Small randomised trials or large randomised trials with a ted in more than half of urothelial carcinomas, including low-
suspicion of bias (lower methodological quality) or and high-grade tumours [40, 42–44]. However, the functional
meta-analyses of such trials or of trials with demonstrated effect of mutations in these genes encoding epigenomic regula-
heterogeneity tory proteins remains relatively unknown. The identification of
III Prospective cohort studies these driving genomic alterations, even if occurring in only a
IV Retrospective cohort studies or case–control studies small subset of bladder cancer patients, may lead to the develop-
V Studies without control group, case reports, experts opinions ment of patient-specific therapies. This has been the case of the
Grades of recommendation recently described mutations in TSC1 predicting response to
mTOR inhibitors like everolimus [40, 45], or in the PIK3CA
A Strong evidence for efficacy with a substantial clinical benefit, gene, mutated in up to 26% of cases [46] that may predict
strongly recommended sensitivity to PIK3CA/mTOR inhibitors.
B Strong or moderate evidence for efficacy but with a limited
clinical benefit, generally recommended
C Insufficient evidence for efficacy or benefit does not outweigh follow-up and long-term implications
the risk or the disadvantages (adverse events, costs, …),
There is no generally accepted follow-up protocol; therefore,
optional
the possible options could be as follows: in NMIBC, regular
cystoscopy and cytology is mandatory every 3–6 months based
on the high or low risk during the first 2 years, and every 6–12
recommended months thereafter to assess tumour response, progression or
recurrence.
a
By permission of the Infectious Diseases Society of America [47]. After definitive treatment of MIBC with RC, urine cytology,
liver function and renal function tests should be carried out
every 3–6 months for 2 years, and subsequently as clinically
indicated. Imaging of the chest, upper tract, abdomen and pelvis
evidence suggests that screening for bladder cancer on a popula- every 3–6 months for 2 years should also be undertaken based
tion level is not helpful for improving survival [V, C]. on the risk of recurrence, and subsequently as clinically indi-
For NMIBC, cystoscopy alone remains the most cost-effective cated. Additionally, urethral wash cytology may be carried out
method to detect bladder cancer recurrence [II or III, B], despite every 6–12 months if urethrectomy has not been carried out or
the fact that it is an invasive and relatively expensive procedure. if there is prior history of CIS.
As for voided urine cytology, it is highly specific but not sensi- For MIBC patients in whom a bladder preservation strategy
tive enough, especially for low-grade tumours. Improvement in has been adopted, there is a need to evaluate response to treat-
scoring systems, such as the European Organisation for ment after induction chemoradiation. After completion, the
Research and Treatment of Cancer (EORTC) scoring system, or same follow-up regimen as for RC is recommended; however,
the Spanish Urological Club for Oncological Treatment cystoscopy and urine cytology plus random biopsies every 3–6
(CUETO) scoring system, for the broad spectrum of NMIBC, as months for 2 years are necessary. During follow-up, monitoring
well as deeper knowledge on the impact of depth of invasion of long-term treatment toxicities and potential recurrences of
and associated CIS in high-grade T1 (HGT1) bladder cancer, secondary tumours should be carried out.
should aid in a better risk stratification of NMIBC [37–39]. For those who undergo systemic chemotherapy, response
Presently, there are five bladder tumour marker tests, namely evaluation every two to three cycles using the initial radiograph-
BTA-Stat, BTA-TRAK, NMP-22, uCyt+ and UroVysion that ic tests carried out during the work-up is also necessary.
may be used for diagnosis and/or follow-up [36], but none has Providing optimal care for patients will also involve addressing
been shown to be superior to urine cytology and cystoscopy. psychosocial implications of all above-mentioned treatment
Molecular analyses have identified genetic and epigenetic al- strategies.
teration in high-grade urothelial carcinomas, including ≤60% of
genomic alterations that could be treated by drugs that are
already available or are in clinical testing [40]. Some potential
note
new targets for treatment intervention have been described for Levels of evidence and grades of recommendation have been
urothelial tumours. Mutations in the receptor tyrosine kinases applied using the system shown in Table 3. Statements without
(RTK)-RAS-RAF, phosphoinositide 3-kinase (PI3K)/AKT/mam- grading were considered justified standard clinical practice by
malian target of rapamycin (mTOR) pathways and regulators of the experts and the ESMO faculty.
iii | Bellmunt et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
conflict of interest 19. Rödel C, Grabenbauer GG, Kühn R et al. Combined-modality treatment and
selective organ preservation in invasive bladder cancer: long-term results. J Clin
JB has reported Advisory board for and lecture fees from Pierre Oncol 2002; 20: 3061–3071.
Fabre. MdS has reported study grants from Pierre Fabre 20. Shipley WU, Zietman AL, Kaufman DS et al. Selective bladder preservation by
Oncology; she also reported Honoraria and Consultancy fees trimodality therapy for patients with muscularis propria-invasive bladder cancer and
from Amgen, Astellas, Bayer, Celgene, Dendreon, Ferring, who are cystectomy candidates—the Massachusetts General Hospital and Radiation
Therapy Oncology Group experiences. Semin Radiat Oncol 2005; 15: 36–41.
GlaxoSmithKline, Janssen Cilag, Novartis, Pfizer, Pierre Fabre,
21. Weiss C, Wolze C, Engehausen DG et al. Radiochemotherapy after transurethral
Roche, Sanofi, Shionogi, Takeda and Teva/Oncogenex. AO, JL, resection for high-risk T1 bladder cancer: an alternative to intravesical therapy or
TW and AH have reported no potential conflicts of interest. early cystectomy? J Clin Oncol 2006; 24: 2318–2324.
22. Milosevic M, Gospodarowicz M, Zietman A et al. Radiotherapy for bladder cancer.
Urology 2007; 69: 80–92.
references 23. von der Maase H, Sengelov L, Roberts JT et al. Long-term survival results of a
randomized trial comparing gemcitabine plus cisplatin, with methotrexate,
1. GLOBOCAN 2012: Estimated Cancer Incidence, Mortality and Prevalence vinblastine, doxorubicin, plus cisplatin in patients with bladder cancer. J Clin Oncol
Worldwide in 2012. http://globocan.iarc.fr (18 February 2014, date last accessed). 2005; 23: 4602–4608.
2. Epstein JI, Amin MB, Reuter VR, Mostofi FK. The World Health Organization/ 24. Loehrer PJ, Sr, Einhorn LH, Elson PJ et al. A randomized comparison of cisplatin
International Society of Urological Pathology consensus classification of urothelial alone or in combination with methotrexate, vinblastine, and doxorubicin in patients
(transitional cell) neoplasms of the urinary bladder. Bladder Consensus Conference with metastatic urothelial carcinoma: a cooperative group study. J Clin Oncol
Committee. Am J Surg Pathol 1998; 22: 1435–1448. 1992; 10: 1066–1073.
3. Eble JN, Sauter G, Epstein JI, Sesterhenn IA. World Health Organization 25. von der Maase H, Hansen SW, Roberts JT et al. Gemcitabine and cisplatin versus
Classification of Tumours. Pathology and Genetics of Tumours of the Urinary methotrexate, vinblastine, doxorubicin, and cisplatin in advanced or metastatic
System and Male Genital Organs. Lyon: IARC Press 2004. bladder cancer: results of a large, randomized, multinational, multicenter, phase III
4. Amin MB, McKenney JK, Paner GP et al. ICUD-EAU International Consultation on study. J Clin Oncol 2000; 18: 3068–3077.
Bladder Cancer 2012: pathology. Eur Urol 2013; 63: 16–35. 26. Milowsky MI, Nanus DM, Maluf FC et al. Final results of sequential doxorubicin
5. Bajorin DF, Dodd PM, Mazumdar M et al. Long-term survival in metastatic plus gemcitabine and ifosfamide, paclitaxel, and cisplatin chemotherapy in
transitional-cell carcinoma and prognostic factors predicting outcome of therapy. J patients with metastatic or locally advanced transitional cell carcinoma of the
Clin Oncol 1999; 17: 3173–3181. urothelium. J Clin Oncol 2009; 27: 4062–4067.
6. Witjes JA, Compérat E, Cowan NC et al. EAU guidelines on muscle-invasive and 27. Galsky MD, Iasonos A, Mironov S et al. Phase II trial of dose-dense doxorubicin
metastatic bladder cancer: summary of the 2013 Guidelines. Eur Urol 2014; 65: plus gemcitabine followed by paclitaxel plus carboplatin in patients with advanced
778–792. urothelial carcinoma and impaired renal function. Cancer 2007; 109: 549–555.
7. Edge SB, Byrd DR, Compton CC (eds), AJCC Cancer Staging Handbook, 7th 28. Bellmunt J, von der Maase H, Mead GM et al. Randomized phase III study
edition. New York, NY: Springer 2010. comparing paclitaxel/cisplatin/gemcitabine and gemcitabine/cisplatin in patients
8. Stimson CJ, Cookson MS, Barocas DA et al. Preoperative hydronephrosis predicts with locally advanced or metastatic urothelial cancer without prior systemic
extravesical and node positive disease in patients undergoing cystectomy for therapy: EORTC Intergroup Study 30987. J Clin Oncol 2012; 30: 1107–1113.
bladder cancer. J Urol 2010; 183: 1732–1737. 29. Bamias A, Aravantinos G, Deliveliotis C et al. Docetaxel and cisplatin with
9. Babjuk M, Burger M, Zigeuner R et al. EAU guidelines on non-muscle- granulocyte colony-stimulating factor (G-CSF) versus MVAC with G-CSF in
invasive urothelial carcinoma of the bladder: update 2013. Eur Urol 2013; 64: advanced urothelial carcinoma: a multicenter, randomized, phase III study from the
639–653. Hellenic Cooperative Oncology Group. J Clin Oncol 2004; 22: 220–228.
10. Gakis G, Efstathiou J, Lerner SP et al. ICUD-EAU International Consultation on 30. Gabrilove JL, Jakubowski A, Scher H et al. Effect of granulocyte colony-stimulating
Bladder Cancer 2012: radical cystectomy and bladder preservation for muscle- factor on neutropenia and associated morbidity due to chemotherapy for
invasive urothelial carcinoma of the bladder. Eur Urol 2013; 63: 45–57. transitional-cell carcinoma of the urothelium. N Engl J Med 1988; 318:
11. Stimson CJ, Chang SS, Barocas DA et al. Early and late perioperative outcomes 1414–1422.
following radical cystectomy: 90-day readmissions, morbidity and mortality in a 31. Sternberg CN, de Mulder P, Schornagel JH et al. Seven year update of an EORTC
contemporary series. J Urol 2010; 184: 1296–1300. phase III trial of high-dose intensity M-VAC chemotherapy and G-CSF versus classic
12. Advanced Bladder Cancer (ABC) Meta-analysis Collaboration. Adjuvant M-VAC in advanced urothelial tract tumours. Eur J Cancer 2006; 42: 50–54.
chemotherapy in invasive bladder cancer: a systematic review and meta-analysis 32. De Santis M, Bellmunt J, Mead G et al. Randomized phase II/III trial assessing
of individual patient data Advanced Bladder Cancer (ABC) meta-analysis gemcitabine/carboplatin and methotrexate/carboplatin/vinblastine in patients with
collaboration. Eur Urol 2005; 48: 189–199; discussion 199–201. advanced urothelial cancer who are unfit for cisplatin-based chemotherapy:
13. Leow JJ, Martin-Doyle W, Rajagopal PS et al. Adjuvant chemotherapy for invasive EORTC study 30986. J Clin Oncol 2012; 30: 191–199.
bladder cancer: a 2013 updated systematic review and meta-analysis of 33. Bellmunt J, Maroto P, Mellado B et al. Phase II study of sunitinib as first line
randomized trials. Eur Urol 2014; 66: 42–54. treatment in patients with advanced urothelial cancer ineligible for cisplatin-based
14. Advanced Bladder Cancer (ABC) Meta-analysis Collaboration. Adjuvant chemotherapy. ASCO Genitourinary Cancers Symposium 2008; abstract 291.
chemotherapy for invasive bladder cancer (individual patient data). Cochrane 34. Bellmunt J, Choueiri TK, Fougeray R et al. Prognostic factors in patients with
Database Syst Rev 2006; 2: CD006018. advanced transitional cell carcinoma of the urothelial tract experiencing
15. Coppin CM, Gospodarowicz MK, James K et al. Improved local control of invasive treatment failure with platinum-containing regimens. J Clin Oncol 2010; 28:
bladder cancer by concurrent cisplatin and preoperative or definitive radiation. The 1850–1855.
National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 1996; 14: 35. Bellmunt J, Théodore C, Demkov T et al. Phase III trial of vinflunine plus best
2901–2907. supportive care compared with best supportive care alone after a platinum-
16. Hoskin PJ, Rojas AM, Bentzen SM, Saunders MI. Radiotherapy with concurrent containing regimen in patients with advanced transitional cell carcinoma of the
carbogen and nicotinamide in bladder carcinoma. J Clin Oncol 2010; 28: urothelial tract. J Clin Oncol 2009; 27: 4454–4461.
4912–4918. 36. Kamat AM, Hegarty PK, Gee JR et al. ICUD-EAU International Consultation on
17. James ND, Hussain SA, Hall E et al. Radiotherapy with or without chemotherapy in Bladder Cancer 2012: screening, diagnosis, and molecular markers. Eur Urol
muscle-invasive bladder cancer. N Engl J Med 2012; 366: 1477–1488. 2013; 63: 4–15.
18. Rödel C, Weiss C, Sauer R. Trimodality treatment and selective organ preservation 37. Fernandez-Gomez J, Madero R, Solsona E et al. The EORTC tables overestimate
for bladder cancer. J Clin Oncol 2006; 24: 5536–5544. the risk of recurrence and progression in patients with non-muscle-invasive
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu223 | iii

by guest
on 05 February 2018
bladder cancer treated with bacillus Calmette-Guérin: external validation of the 42. Taylor BS, Barretina J, Socci ND et al. Functional copy-number alterations in
EORTC risk tables. Eur Urol 2011; 60: 423–430. cancer. PLoS One 2008; 3: e3179.
38. Xylinas E, Kent M, Kluth L et al. Accuracy of the EORTC risk tables and of the 43. Gui Y, Guo G, Huang Y et al. Frequent mutations of chromatin remodeling genes in
CUETO scoring model to predict outcomes in non-muscle-invasive urothelial transitional cell carcinoma of the bladder. Nat Genet 2011; 43: 875–878.
carcinoma of the bladder. Br J Cancer 2013; 109: 1460–1466. 44. Solomon DA, Kim JS, Bondaruk J et al. Frequent truncating mutations of STAG2 in
39. Fernandez-Gomez J, Madero R, Solsona E et al. Predicting nonmuscle invasive bladder cancer. Nat Genet 2013; 45: 1428–1430.
bladder cancer recurrence and progression in patients treated with bacillus 45. Iyer G, Hanrahan AJ, Milowsky MI et al. Genome sequencing identifies a basis for
Calmette-Guérin: the CUETO scoring model. J Urol 2009; 182: 2195–2203. everolimus sensitivity. Science 2012; 338: 221.
40. Iyer G, Al-Ahmadie H, Schultz N et al. Prevalence and co-occurrence of actionable 46. Ross JS, Wang K, Al-Rohil RN et al. Advanced urothelial carcinoma: next-
genomic alterations in high-grade bladder cancer. J Clin Oncol 2013; 31: generation sequencing reveals diverse genomic alterations and targets of therapy.
3133–3140. Mod Pathol 2014; 27: 271–280.
41. Al-Ahmadie HA, Iyer G, Janakiraman M et al. Somatic mutation of fibroblast growth 47. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
factor receptor-3 (FGFR3) defines a distinct morphological subtype of high-grade among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
urothelial carcinoma. J Pathol 2011; 224: 270–279. 139–144.
iii | Bellmunt et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
Testicular seminoma and non-seminoma: ESMO

Clinical Practice Guidelines for diagnosis, treatment
and follow-up†
J. Oldenburg1, S. D. Fosså1, J. Nuver2, A. Heidenreich3, H-J Schmoll4, C. Bokemeyer5, A. Horwich6,
J. Beyer7 & V. Kataja8, on behalf of the ESMO Guidelines Working Group*
1
Department of Oncology, Oslo University Hospital, Oslo, Norway; 2Department of Medical Oncology, University Medical Center Groningen, Groningen, The Netherlands;
3
Department of Urology, RWTH University Hospital, Aachen; 4Department of Oncology/Haematology/Haemostaseology, University Hospital Halle, Halle; 5Department of
Oncology, Haematology and Bone Marrow Transplantation with Section Pneumology, University Hospital, Hamburg, Germany; 6Department of Academic Radiotherapy,
Institute of Cancer Research, Royal Marsden Hospital, Sutton Hospital, UK; 7Department of Haematology and Oncology, Vivantes Klinikum Am Urban, Berlin, Germany;
8
Cancer Centre, Kuopio University Hospital, Kuopio, Finland;
epidemiology tissue), pT category (according to the American Joint

Committee on Cancer, AJCC, Union for International Cancer
Germ cell tumours (GCT) represent a rare malignancy affecting Control, UICC), all histological components with
mostly Caucasian males aged between 15 and 40 years. Among corresponding percentages, and presence or absence of vascular
these young men, testicular GCT (TGCT) is the most common invasion and testicular intraepithelial neoplasia (TIN). In
clinical practice
cancer with noted geographic differences [1]. Cure rates seminomas, the presence of syncytiotrophoblasts should be
guidelines
approximate 100% in stage I disease and exceed 80% in reported. Increased copy numbers of iso-chromosome 12p are
metastatic cases. found in both TGCT and EGGCT and provide a
Approximately 50% of the TGCTs are pure seminomas and pathognomonic test, which might be useful in challenging
50% are non-seminomas. The vast majority of GCT arise in the histologic diagnoses, e.g. somatically transformed teratoma.
testicles with ∼5% occurring outside of the gonads, i.e.
extragonadal germ cell tumour (EGGCT). EGGCTs are usually management of the primary tumour
found in the body’s mid-line, e.g. retroperitoneum, mediastinum
or cerebrum, sometimes posing diagnostic difficulties. ‘Radical orchiectomy’ provides the histological diagnosis and
should be carried out before any further treatment, unless the
clinical situation requires immediate chemotherapy in patients
diagnosis with a clear germ cell malignancy based on elevated tumour
In patients with a testicular mass, testicular sonography markers. Any testicular mass of uncertain ranking must be
(7.5 MHz transducer) should be carried out, also noting the size explored by the inguinal approach to verify or exclude
and any structural alterations of the contralateral testis. malignancy. As benign testicular lesions are recognised with
Diagnosis of a testicular germ cell cancer (TGCC) is based on increasing frequency, frozen section analysis should be
histology of the testicular mass. Elevation of ‘tumour markers’, considered intra-operatively, which differentiates malignant
i.e. serum levels of α-fetoprotein (AFP) and/or human chorionic from benign testicular lesions [IV, B] [2]. Tumour marker
gonadotropin (HCG) support the diagnosis. Biopsy of mid-line analysis should be carried out before and after surgery until
extragonadal tumours is mandatory, unless the patient is very sick normalisation, progression or plateau development, since this
and has high tumour markers. The biopsy should be preceded by information is used for final staging.
testicular sonography to exclude a TGCT. Radical orchiectomy is carried out through an inguinal
Histology of GCT should be reported according to the World incision [III, A]. Any scrotal violation for biopsy or open
Health Organisation (WHO) classification, specifying tumour surgery should be avoided. The tumour-bearing testis is resected
size, multiplicity, extension of tumour (e.g. in rete testis or other with the spermatic cord at the level of the internal inguinal ring.
In experienced centres, ‘organ-preserving surgery’ may be
feasible in case of a small tumour, particularly in patients with
*Correspondence to: ESMO Guidelines Working Group, ESMO Head Office, Via L.
Taddei 4, CH-6962 Viganello-Lugano, Switzerland; synchronous bilateral testicular tumours, tumour in a solitary
E-mail: clinicalguidelines@esmo.org testis or contralateral atrophic testis. However, mandatory
†
postresection testicular radiotherapy renders the residual
Approved by the ESMO Guidelines Working Group: April 2002, last update July 2013.
This publication supersedes the previously published versions-Ann Oncol 2010; testicular tissue azoospermic but retains some testosterone
21 (Suppl 5): v140-v146 and Ann Oncol 2010; 21 (Suppl 5): v147–v154. production [IV, B] [3].

by guest
on 05 February 2018
biopsy for diagnosis of TIN in the contralateral testis For stage I disease different risk factors have been identified for
and subsequent management seminoma and non-seminoma based on histological features in the
In 2%–5% of TGCT patients, a contralateral TGCT is diagnosed primary tumour. For metastatic cases the IGCCCG has identified
either metachronously or synchronously. Accordingly, between three prognostic groups (see Table 1). If treatment is carried out
3% and 5% of testicular cancer patients have TIN in the correctly, the 5-year survival rate of patients with TGCT
contralateral testis with the highest risk (∼30%) in men with approximates 99% in stage I, and 91%, 79% and 48% in metastatic
testicular atrophy (volume <12 ml) and age <40 years, and in disease with good, intermediate and poor prognosis, respectively.
patients with EGGCT. The IGCCCG provided prognostic information for
The majority of European Germ Cell Cancer Consensus chemotherapy-treated metastatic disease. For patients with non-
group (EGCCCG) experts did not consider a routine biopsy of seminoma, a good, intermediate or poor risk group is identified.
the contralateral testis as indicated [V, C] [4]. If a biopsy is Patients with seminoma are categorised as either good or
carried out and TIN is diagnosed, however, the condition may intermediate risk (there is no poor-risk group). However, not all
be managed by surveillance, irradiation with 20 Gy in 2 Gy patients with metastases receive chemotherapy, e.g.
fractions (with potential damage to the contralateral, non- radiotherapy for seminoma IIA or retroperitoneal lymph node
affected testis by scattered radiation) or orchiectomy, depending dissection (RPLND) for non-seminoma IIA (Figures 1 and 2).
on fertility issues. Imaging: computed tomography (CT) scan of the abdomen and
In patients with metastatic disease treated with three or more pelvis [III, B] is mandatory. Thoracic CT should be carried out in
cycles of cisplatin-based chemotherapy, TIN in the contralateral case of non-seminoma, but can be omitted in seminoma patients
non-resected testicle may be eradicated or progression may be without infradiaphragmatic metastases. Magnetic resonance
slowed down, although the risk of developing an invasive imaging (MRI) of the central nervous system is indicated in
tumour is still substantial. advanced stages, particularly in case of choriocarcinoma/high HCG,
or in those with cerebral symptoms. Positron emission tomography
(PET) scanning does not contribute to initial staging [II, D].
post-orchiectomy staging and risk Blood tests: tumour markers (AFP, HCG, LDH) should be
assessment determined before orchiectomy and followed until
normalisation or lack of further decrease. The half-life for HCG
Post-orchiectomy management should be the responsibility of is up to 3 days and 5–7 days for AFP. Serum levels of total
clinicians with experience in the classification and treatment testosterone, luteinizing hormone (LH) and follicle-stimulating
of TGCT [V, A]. hormone (FSH) should be determined. Semen analysis and
‘Staging and risk group categorisation’ are carried out sperm banking should be discussed with all patients.
according to the AJCC/UICC and the International Germ Cell
Cancer Collaborative Group (IGCCCG), reflecting the extent of
the disease based on clinical and radiological examinations and
post-orchiectomy treatment
the results of serum tumour markers after orchiectomy, Before any treatment (Figures 1 and 2), the patient should be
including serum lactate dehydrogenase (LDH) [5]. informed about the potential treatment modalities, their acute
Table 1. Post-orchiectomy staging of metastatic seminoma and non-seminoma according to AJCC/UICC and IGCCCG classification
Clinical stage TNM (AJCC/UICC) Serum tumour markers (S) to be determined after IGCCCG prognostic group
orchiectomy
Ta N M S LDH HCG AFP (ng/ml)
IS Tany N0 M0 S1 <1.5xN and <5000 and <1000 Good
S2 1.5–10xN or 5000–50 000 or 1000–10 000 Intermediate
S3 >10xN or >50 000 or >10 000 Poor
IIA Tany N1 M0 S0 Normal Normal Normal Good
(≤2 cm) S1 <1.5xN and <5000 and <1000
IIB Tany N2 M0 S0 Normal Normal Normal Good
(>2–5 cm) S1 <1.5xN and <5000 and <1000
IIC Tany N3 M0 S0 Normal Normal Normal Good
(>5 cm) S1 <1.5xN and <5000 and <1000
IIIA Tany Nany M1a S0 Normal Normal Normal Good
S1 <1.5xN and <5000 and <1000
IIIB Tany N1-3 M0 S2 1.5–10xN or 5000–50 000 or 1000–10 000 Intermediate
Nany M1a
IIIC Tany N1-3 M0 S3 >10xN or >50 000 or >10 000 Poor
Nany M1a S3 >10xN or >50 000 or >10 000 Poor
M1b Sany Any level Any level Any level Poor
Primary mediast EGGCT Nany Many Sany Any level Any level Any level Poor
a
Primary retroperitoneal EGGCT is staged like TGCT (Tany).
vi | Oldenburg et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
Figure 1. Standard treatment strategies for seminoma.
and late toxicity, and the overall outcome. Based on multiple in similar relapse rates, but less protracted treatment-related
clinical studies, three or four cycles with bleomycin, etoposide, lethargy, sick leave and probably treatment-induced
cisplatin (BEP) (Table 2) represent the standard treatment of malignancies [I, A] [6, 7], although the true long-term adverse
metastatic patients. effects after >10 years are still unknown.
In order to maintain treatment intensity, chemotherapy If a relapse occurs, it is usually located in the retroperitoneal
cycles should be repeated every 3 weeks, independent of or iliac lymph nodes. Rarely, late occurring relapses may contain
leukocyte count. However, infection at day 22 warrants delay of non-seminoma components [IV, B] [8].
chemotherapy until recovery.
Tumour markers are to be determined immediately before stage IIA (lymph nodes 1–2 cm)
the start of each new chemotherapy cycle.
The treatment options consist of either cisplatin-based
chemotherapy or radiotherapy to para-aortic and ipsilateral iliac
lymph nodes with 30 Gy in 2 Gy fractions (Figure 1). A recent
seminoma study reported three relapses among 29 irradiated stage IIA
stage I patients (10.9%), compared with no relapses after cisplatin-
based chemotherapy among six stage IIA and 79 stage IIB
Approximately 80% of the patients with seminoma present with
patients [II, B] [9]. Neoadjuvant carboplatin before
stage I disease, with a survival of ∼99%, independent of the
radiotherapy may further reduce relapse rates, according to a
chosen strategy, if accepted by the patient. In light of this very
recent single centre pilot study in 51 seminoma patients, but
high cure rate, minimising toxicity is the priority. Surveillance is
this strategy needs further validation [III, B] [10].
considered the preferred strategy. The predictive value of ‘risk
factors’, such as rete testis infiltration and tumour size ≥4 cm, is
controversial, but these factors are sometimes used to apply one stage IIB/IIC
course of carboplatin (AUC 7) or radiotherapy (20 Gy/10 Three cycles of BEP represent the standard therapy. If there are
fractions to para-aortic target volume) as adjuvant treatment. arguments against bleomycin, e.g. reduction in lung capacity,
Compared with radiotherapy, one course of carboplatin results emphysaema, heavy smoking (including former smokers) or
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt304 | vi

by guest
on 05 February 2018
Figure 2. Standard treatment strategies for non-seminoma.
poor renal function, four cycles of etoposide, cisplatin (EP) are biopsy [IV, B] [12]. If PET is unavailable, lesions >3 cm can
used (Table 2). either be biopsied, resected or followed until resolution or
Patients unsuitable for chemotherapy should receive para- progression.
aortic and ipsilateral iliac field radiotherapy to 36 Gy in 2 Gy A negative PET scan warrants follow-up only. In the case of
fractions. a positive PET scan, the possibility of residual seminoma is
high, though a false-positive result cannot be excluded [IV, B]
stage III [12]. A biopsy might be carried out before treatment by
Chemotherapy with BEP is standard treatment: three cycles for irradiation or resection. However, perioperative complications
good prognosis patients according to IGCCCG (alternatively are more common than in non-seminoma due to desmoplastic
four cycles of EP) and four cycles for intermediate prognosis reactions of the chemotherapy-exposed seminoma metastases.
patients according to IGCCCG (alternatively four cycles of
etoposide, ifosfamide and cisplatin (VIP), if there are arguments non-seminoma
against bleomycin) [5, 11].
stage I
post-chemotherapy management Stage I disease implies excellent survival rates of 98%–100% and
Patients with complete response do not require further is categorised by absence or presence of vascular invasion into
treatment and are followed-up. In case of residual tumour, a ‘low risk’ (20% relapse rate) or ‘high risk’ (40%–50% relapse
2-fluor-2-deoxy-D-glucose PET (FDG-PET) scan a minimum of rate), respectively.
6 weeks after ending chemotherapy may be carried out:
low-risk non-seminoma stage I
- in lesions >3 cm, FDG-PET is the recommended approach
- in lesions <3 cm, FDG-PET may be considered, but its Surveillance is the standard for low-risk disease. If surveillance
positive predictive value is lower and surveillance is preferred. is not feasible, e.g. due to difficulties with repeated imaging, low
compliance or patient’s preference, adjuvant chemotherapy with
Based on the negative predictive value >90%, a negative PET one or two cycles of BEP is given. Efficacy appears to be similar
scan of a non-growing or regressing lesion may substitute a between one and two cycles of BEP [III, C] [13]. In patients not
vi | Oldenburg et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
Table 2. Chemotherapy regimens in metastatic seminoma recommended above. Some experts consider nerve-sparing
and non-seminoma RPLND the preferred treatment of patients with teratoma and
somatic transformation in the primary tumour.
BEP a (Repeat cycles every 3 weeks)
Cisplatin 20 mg/m2 Day 1–5 stage IIA (IIB), marker-negative
Etoposide 100 mg/m2 Day 1–5
Bleomycin 30 mg Day 1, 8, 15
Metastatic non-seminoma not purely consisting of teratoma
EP b (Repeat cycles every 3 weeks) should be treated according to the IGCCCG’s recommendations
Cisplatin 20 mg/m2 Day 1–5 (Figure 2).
Etoposide 100 mg/m2 Day 1–5 Small lymph nodes might not represent metastases thus
VIP/PEI c (Repeat cycles every 3 weeks) implying the risk of over-treatment, which may be avoided by
Cisplatin 20 mg/m2 Day 1–5 the following two strategies:
Etoposide 75 mg/m2 Day 1–5
• Close follow-up with abdominal imaging every 6 weeks until
Ifosfamide 1.2 g Day 1–5
TIP d (Repeat cycles every 3 weeks)
regression or progression, resulting in observation only or
Paclitaxel 250 mg/m2 Day 1
treatment, respectively. Treatment may consist of primary
Cisplatin 25 mg/m2 Day 2–5 nerve-sparing RPLND in case of a single progressing lymph
Ifosfamide 1.5 g Day 2–5 node, and the presence of normal markers suggestive of
VeIP e (Repeat cycles every 3 weeks) teratoma or chemotherapy. In case of multiple progressive
Vinblastine 0.11 mg/kg Day 1 + 2 lymph nodes and/or rising tumour markers suggestive of
Ifosfamide 1.2 g/m2 Day 1–5 non-teratomatous TGCT, chemotherapy (3 cycles of BEP) is
Cisplatin 20 mg/m2 Day 1–5 indicated.
TI-CE f (TI cycles 1–2 every 2 weeks) • Lymph node biopsy or primary nerve-sparing RPLND. The
Paclitaxel 200 mg/m2 Day 1 latter approach comprising both diagnostic and therapeutic
Ifosfamide 2.0 g Day 2–4 potential. Adjuvant chemotherapy post-RPLND in form of
(CE cycles 3–5 every 3 weeks) two cycles BEP may be considered, in case of vital GCT in the
Carboplatin AUC = 7 Day 1–3 specimen. Completely resected teratoma warrants follow-up
Etoposide 400 mg/m2 Day 1–3 only.
CE g (Two cycles, may be preceded by VeIP)
Carboplatin 700 mg/m2 Day 1
Etoposide 750 mg/m2 Day 1–3 Stage IS/II/III
a
Three cycles BEP in IGCCCG good prognosis, four cycles BEP in IGCCCG Patients with good prognosis should receive three cycles of BEP
intermediate or poor prognosis. or four cycles of EP if contra-indications against bleomycin
b
Four cycles EP only in IGCCCG good prognosis in case of contra- exist. BEP can be substituted by VIP (Table 2). Four cycles of
indications to bleomycin. BEP still represent standard treatment of patients with
c
Four cycles VIP only in IGCCCG intermediate or poor prognosis in case of intermediate or poor prognosis [I, A] [5, 14]. In case of contra-
contra-indications to bleomycin as first-line chemotherapy; or as salvage indication against bleomycin, four cycles of VIP are used. First-
chemotherapy. PEI is synonymous to VIP. line high-dose chemotherapy has not been proven superior to
d
Four cycles TIP, typically as conventional dose salvage chemotherapy. standard dose chemotherapy in three randomised trials.
e
Four cycles VeIP, typically as conventional dose salvage chemotherapy. A prospective randomised trial has indicated that poor
f
Two cycles TI before stem cell harvesting, thereafter three cycles CE as high- prognosis patients with an insufficient tumour marker decline
dose treatment. after the first cycle of BEP might benefit from dose
g
Two cycles CE as high-dose treatment, may be preceded by cyto-reductive intensification of first-line therapy, rather than continuation of
VeIP. standard BEP treatment [15], though the evidence of an
optimal dose-dense regimen is still needed.
suitable for surveillance or adjuvant chemotherapy, open nerve- post-chemotherapy management

sparing RPLND in highly experienced centres is an option. Four (to eight) weeks after the last cycle, determination of
Some experts consider nerve-sparing RPLND the preferred tumour markers as well as imaging (chest X-ray, CT scan or
treatment of patients with teratoma and somatic transformation MRI of the initial sites) should be carried out.
in the primary tumour. In case of complete response (normal tumour markers,
no residual tumour lesions, no retroperitoneal lymph nodes
≥10 mm), no further treatment is necessary. Residual lymph
high-risk non-seminoma stage I nodes, exceeding 10 mm in diameter, should be removed by
There are two standard treatment options: surveillance with open nerve-sparing RPLND [III] [16, 17].
40%–50% relapse rate or adjuvant chemotherapy (one or two Principally, any residual tumour with normal markers should
cycles of BEP, relapse rate of 3%–4%). Survival is the same be resected if technically feasible. This applies in particular to
whichever option is used. Nerve-sparing RPLND may be carried liver and lung metastases after chemotherapy. Patients with
out in case of contra-indications against the strategies multiple visceral metastases should be discussed with experts
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt304 | vi

by guest
on 05 February 2018
and treated at specialised centres. Patients in whom all residual cisplatin (TIP) or high-dose chemotherapy with three cycles of
lesions are deemed resectable should be operated by surgeons paclitaxel, ifosfamide, carboplatin and etoposide (TI-CE) is
with appropriate experience and without delay, also in case of under preparation.
plateauing tumour markers. In 2005, Pico et al. reported on 280 relapsing TC patients
Also, patients with elevated tumour markers should receive randomised to either four cycles of cisplatin, ifosfamide and
treatment based on individualised recommendations by experts. etoposide/vinblastine or three such cycles followed by high-dose
Elevated tumour markers should be assessed at least once carboplatin, etoposide and cyclophosphamide (CarboPEC) with
weekly: rising tumour markers indicate progressive GCT, haematopoietic stem cell support without significant differences
usually requiring highly specialised multi-disciplinary therapy of OS or progression-free survival (PFS) [II, D] [21] Alternative
(see salvage treatment). A laparoscopic RPLND should only be conventional dose cisplatin-based regimens with similar efficacy
carried out within clinical studies since not all potentially comprise TIP, VeIP (vinblastine, ifosfamide, cisplatin) or
affected lymph nodes can be assessed by this approach. VIP/PEI (etoposide, ifosfamide, cisplatin) [14]. Carboplatin-
Patients with complete response or with complete resection of based high-dose chemotherapy has been reported to achieve
differentiated teratoma or fibrotic tissue only, require no further complete remissions in relapsing patients as third line or later
treatment. Good prognosis patients with completely resected and is the preferred option of some authorities, despite absence
viable malignant tumour, comprising <10% of the specimen, do of randomised trials in this area. [III, B] [22].
not benefit from adjuvant chemotherapy [IV, C] [18]. In refractory patients, i.e. those not reaching a marker-
However, in patients with IGCCCG intermediate or poor negative complete response after first-line treatment or those
prognosis, >10% viable tumour in the specimen, and/or without favourable response to salvage treatment, further
incomplete resection, consolidation chemotherapy, e.g. two cycles treatment must be individualised by GCT experts [V, B] [23].
of VIP, may be considered, although a surveillance strategy is also These patients should be included in clinical trials, if available.
justified according to ∼30% of EGCCCG experts [4]. Surgery should be part of the strategy whenever possible,
particularly in those patients with localised or late relapse, and
with poor response to chemotherapy.
salvage treatment of seminoma
and non-seminoma
Conclusive recommendations as to an optimal salvage approach
late relapse
in patients relapsing after cisplatin-based first-line treatment A late relapse occurs in 2%–3% of survivors and is defined as
cannot be made at present. The prognosis of relapsing GCC new tumour growth >2 years after at least three cycles of
patients is variable as shown by the ‘International Prognostic preceding chemotherapy. These relapses do not respond so well
Factor Study Group’ who categorised 1594 relapsing GCC to new chemotherapy (often yolk sac tumour, usually AFP-
patients into five prognostic groups, with 2-year survival rates positive, or slow-growing teratoma) [IV, C] [24].
ranging from 75% (very low risk) to 6% (very high risk), Table 3 In particular, in marker-negative relapses histological
[IV, C] [19]. The same group demonstrated superior survival assessment of the relapsing lesions should be carried out by
rates for patients treated with high-dose chemotherapy [n = 812, radical surgical resection of all lesions, if technically feasible.
51.2% 5-year overall survival (OS)] compared with conventional Further chemotherapy must be individualised based on the
dose chemotherapy (n = 773, 5-year OS 42.8) [IV, C] [20]. The histology of the late relapse and tumour marker development. If
retrospective nature of this study limits its conclusive power, salvage chemotherapy is the first treatment option of a late
such that an international prospective study randomising relapse, radical post-chemotherapy surgery should be conducted
relapsing patients to either four cycles of paclitaxel, ifosfamide, whenever possible.
Table 3. Prognostic score for patients with relapsing non-seminoma or seminoma. From Lorch et al. [19]. Reprinted with permission. @2010 American
Society of Clinical Oncology. All rights reserved.
Parameter Score points

0 1 2 3 Score
Primary site Gonadal Extragonadal – Mediastinal non-seminoma
Prior response CR/PRm- PRm+/SD PD –
PFI, months >3 ≤3 – –
AFP salvage Normal ≤1000 >1000 –
HCG salvage ≤1000 >1000 – –
Score sum (values from 0 to 10)
Regroup score sum into categories: (0) = 0; (1 or 2) = 1; (3 or 4) = 2; (5 or more) = 3
Add histology score points: pure seminoma = −1; non-seminoma or mixed tumours = 0
Final prognostic score (−1 = very low risk; 0 = low risk; 1 = intermediate risk; 2 = high risk; 3 = very high risk)
CR, complete remission; PRm-, partial remission, negative markers; PRm+, partial remission, positive markers; SD, stable disease; PD, progressive disease; PFI,
progression-free interval; LBB, liver, bone, brain metastases; AFP, α-fetoprotein; HCG, human chorionic gonadotrophin.
vi | Oldenburg et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
Grading Systema
Levels of evidence
I Evidence from at least one large, randomised, controlled trial of good methodological quality (low potential for bias) or meta-analyses of well-conducted
randomised trials without heterogeneity
II Small randomised trials or large randomised trials with a suspicion of bias (lower methodological quality) or meta-analyses of such trials or of trials with
demonstrated heterogeneity
C Insufficient evidence for efficacy or benefit does not outweigh the risk or the disadvantages (adverse events, costs..,), optional
a
Dykewicz CA. Summary of the guidelines for preventing opportunistic infections among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001;
33: 139–144. By permission of the Infectious Diseases Society of America.
late toxicity association with both impaired physical and mental quality of
life. Furthermore, anxiety levels are higher in GCC survivors
Besides early detection of relapse, follow-up should be directed than in the general male population.
towards prevention, detection and treatment of late toxicity for Perhaps most importantly, TGCT survivors and their family
the increasing number of GCC survivors. doctors should be adequately informed (verbally and using
Semen cryopreservation should be considered in each patient. written information) about potential late toxicity and their
Compared with the general population the 10-year post- prevention, both during and at the end of treatment and in the
treatment paternity rate is significantly reduced, in part due to course of specialised follow-up.
pre-existing fertility problems. Nevertheless, the 15-year
fatherhood rate among testicular cancer survivors wishing to
father a child is ∼70%, with a strong association with treatment personalised medicine
intensity [IV, B] [25]. Hypogonadism is present in 11%–35% of
TGCT survivors, depending on cut-off levels of testosterone In this disease setting, more research is needed to identify
used, age, cumulative cisplatin dose and follow-up duration. molecular markers which could lead to advances in
Therefore, determination of testosterone levels is recommended personalised medicine.
during follow-up, although it is not always clear when and at
what testosterone level replacement should be offered.
Compared with the general population, there is about a twofold follow-up
increased risk of late post-chemotherapy cardiovascular disease Early detection and treatment of relapse represents the primary
(coronary heart disease, myocardial infarction, congestive heart objective of follow-up visits during the first 5–10 years.
failure and stroke) among TGCT survivors. Early-onset Recommendations for the follow-up schedule need to be
(starting 3–5 years after treatment) metabolic syndrome occurs adapted according to national and institutional requirements.
in about 20%–30% of long-term survivors [IV, C] [26]. Many follow-up recommendations that have been published
Therefore, survivors need to be counselled on a healthy lifestyle most likely expose TGCT survivors to unnecessary radiation,
(no smoking, regular physical exercise) and screened for other increasing the risk of a radiation-induced second cancer.
known risk factors such as hypertension, dyslipidaemia and Replacing CT by MRI scan would reduce this risk, but is not
excessive weight gain. Pulmonary and renal toxicity, oto- and considered feasible for the majority of European countries.
neurotoxicity are further dose-related sequelae. However, effort should be made to reduce the frequency of CT
The relative risk (RR) of a second solid non-germ cell scans and limit their overall number. PET-CT scanning has no
tumour, particularly in the gastro-intestinal and urinary tract, is role in the routine follow-up of TGCT patients.
approximately doubled after radiotherapy (latency ≥10 years)
and is probably also increased after chemotherapy. The
estimated cumulative risk of leukaemia depends on the
cumulative etoposide dose and occurs earlier in the course of
note
follow-up, i.e. usually <10 years. Levels of evidence and grades of recommendation have been
Health-related quality of life in long-term TGCT survivors applied using the system shown in Table 4. Statements without
appears to be similar to the normal male population, but grading were considered justified standard clinical practice by
persisting long-term treatment-related side-effects show a strong the experts and the ESMO faculty.
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt304 | vi

by guest
on 05 February 2018
conflict of interest lesions: a retrospective validation of the SEMPET trial. Ann Oncol 2012; 23:
59–64.
Prof. Kataja has reported institutional clinical research support 13. Tandstad T, Dahl O, Cohn-Cedermark G et al. Risk-adapted treatment in clinical
from Sanofi, Bayer Health Care, Orion Pharma and Merck; the stage I nonseminomatous germ cell testicular cancer: the SWENOTECA
arrangements do not involve personal financial support from management program. J Clin Oncol 2009; 27: 2122–2128.
the companies mentioned. The other authors have declared no 14. Feldman DR, Bosl GJ, Sheinfeld J, Motzer RJ. Medical treatment of advanced
potential conflicts of interest. testicular cancer. JAMA 2008; 299: 672–684.
15. Fizazi K, Pagliaro LC, Flechon A et al. A phase III trial of personalized
chemotherapy based on serum tumor marker decline in poor-prognosis
germ-cell tumors: results of GETUG 13. J Clin Oncol 2013; 31 (Suppl. 18):
references LBA4500.
1. Bray F, Richiardi L, Ekbom A et al. Trends in testicular cancer incidence and 16. Ehrlich Y, Brames MJ, Beck SD et al. Long-term follow-up of cisplatin
mortality in 22 European countries: continuing increases in incidence and declines combination chemotherapy in patients with disseminated nonseminomatous germ
in mortality. Int J Cancer 2006; 118: 3099–3111. cell tumors: is a postchemotherapy retroperitoneal lymph node dissection needed
2. Elert A, Olbert P, Hegele A et al. Accuracy of frozen section examination of after complete remission? J Clin Oncol 2010; 28: 531–536.
testicular tumors of uncertain origin. Eur Urol 2002; 41: 290–293. 17. Heidenreich A, Pfister D. Retroperitoneal lymphadenectomy and resection
3. Heidenreich A, Weissbach L, Holtl W et al. Organ sparing surgery for malignant for testicular cancer: an update on best practice. Ther Adv Urol 2012; 4:
germ cell tumor of the testis. J Urol 2001; 166: 2161–2165. 187–205.
4. Beyer J, Albers P, Altena R et al. Maintaining success, reducing treatment burden, 18. Fizazi K, Oldenburg J, Dunant A et al. Assessing prognosis and optimizing
focusing on survivorship: highlights from the third European consensus conference treatment in patients with postchemotherapy viable nonseminomatous germ-cell
on diagnosis and treatment of germ-cell cancer. Ann Oncol 2013; 24: 878–888. tumors (NSGCT): results of the sCR2 international study. Ann Oncol 2008; 19:
5. International Germ Cell Cancer Collaborative Group. International germ cell 259–264.
consensus classification: a prognostic factor-based staging system for metastatic 19. Lorch A, Beyer J, Bascoul-Mollevi C et al. Prognostic factors in patients with
germ cell cancers. J Clin Oncol 1997; 15: 594–603. metastatic germ cell tumors who experienced treatment failure with cisplatin-
6. Oliver RT, Mason MD, Mead GM et al. Radiotherapy versus single-dose carboplatin based first-line chemotherapy. J Clin Oncol 2010; 28: 4906–4911.
in adjuvant treatment of stage I seminoma: a randomised trial. Lancet 2005; 366: 20. Lorch A, Bascoul-Mollevi C, Kramar A et al. Conventional-dose versus high-dose
293–300. chemotherapy as first salvage treatment in male patients with metastatic germ cell
7. Oliver RT, Mead GM, Rustin GJ et al. Randomized trial of carboplatin versus tumors: evidence from a large international database. J Clin Oncol 2011; 29:
radiotherapy for stage I seminoma: mature results on relapse and contralateral 2178–2184.
testis cancer rates in MRC TE19/EORTC 30982 study (ISRCTN27163214). J Clin 21. Pico JL, Rosti G, Kramar A et al. A randomised trial of high-dose chemotherapy in
Oncol 2011; 29: 957–962. the salvage treatment of patients failing first-line platinum chemotherapy for
8. Oldenburg J, Alfsen GC, Waehre H, Fossa SD. Late recurrences of germ cell advanced germ cell tumours. Ann Oncol 2005; 16: 1152–1159.
malignancies: a population-based experience over three decades. Br J Cancer 22. Lorch A, Neubauer A, Hackenthal M et al. High-dose chemotherapy (HDCT) as
2006; 94: 820–827. second-salvage treatment in patients with multiple relapsed or refractory germ-cell
9. Tandstad T, Smaaland R, Solberg A et al. Management of seminomatous testicular tumors. Ann Oncol 2010; 21: 820–825.
cancer: a binational prospective population-based study from the Swedish 23. Oechsle K, Kollmannsberger C, Honecker F et al. Long-term survival after
Norwegian testicular cancer study group. J Clin Oncol 2011; 29: 719–725. treatment with gemcitabine and oxaliplatin with and without paclitaxel plus
10. Horwich A, Dearnaley DP, Sohaib A et al. Neoadjuvant carboplatin before secondary surgery in patients with cisplatin-refractory and/or multiply relapsed
radiotherapy in stage IIA and IIB seminoma. Ann Oncol 2013; 24: 2104–2107. germ cell tumors. Eur Urol 2011; 60: 850–855.
11. de Wit R, Stoter G, Sleijfer DT et al. Four cycles of BEP versus four cycles of VIP in 24. Oldenburg J, Martin JM, Fossa SD. Late relapses of germ cell malignancies:
patients with intermediate-prognosis metastatic testicular non-seminoma: a incidence, management, and prognosis. J Clin Oncol 2006; 24: 5503–5511.
randomized study of the EORTC Genitourinary Tract Cancer Cooperative Group. 25. Brydoy M, Fossa SD, Klepp O et al. Paternity following treatment for testicular
European Organization for Research and Treatment of Cancer. Br J Cancer 1998; cancer. J Natl Cancer Inst 2005; 97: 1580–1588.
78: 828–832. 26. de Haas EC, Altena R, Boezen HM et al. Early development of the metabolic
12. Bachner M, Loriot Y, Gross-Goupil M et al. 2-18fluoro-deoxy-D-glucose Positron syndrome after chemotherapy for testicular cancer. Ann Oncol 2013; 24:
emission tomograpshy (FDG-PET) for postchemotherapy seminoma residual 749–755.
vi | Oldenburg et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
Penile cancer: ESMO Clinical Practice Guidelines

H. Van Poppel1, N. A. Watkin2, S. Osanto3, L. Moonen4, A. Horwich5 & V. Kataja6, on behalf of the
1
Department of Urology, University Hospital, KU Leuven, Leuven, Belgium; 2Department of Urology, St. George’s Hospital, London, UK; 3Department of Oncology, Leiden
University Medical Centre, Leiden; 4Department of Radiation Oncology, The Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital, Amsterdam,
The Netherlands; 5Institute of Cancer Research and Royal Marsden Hospital, Sutton, UK; 6Cancer Centre, Kuopio University Hospital, Kuopio, Finland
incidence lesion(s); and penis length. Physical examination alone can

assess infiltration of the tumour into the corpora cavernosa.
Penile cancer is an uncommon malignancy in Western Where there is doubt as to the presence of corpora cavernosa
countries with an incidence of <1 per 100 000 men in Europe invasion and to determine whether limited surgery is possible,
and the United States but is more frequent in Africa, Asia and magnetic resonance imaging (MRI) combined with an intra-
South America and accounts for 10% of all cancers in men in cavernosal injection of prostaglandin E1 that causes an artificial
certain areas. Incidence varies according to racial group, erection may be helpful [1].
ethnicity, and geographical location. Important risk factors
clinical practice
include social and cultural habits, and hygienic and religious
guidelines
regional lymph nodes
practices. Penile cancer is rare in circumcised men, particularly
if they are circumcised as newborns. Penile cancer in men and Evaluation of the LNs is also critical as characteristics such as
cervical cancer in women have a strong correlation with human the involvement of the inguinal LNs, the number and site of
papilloma virus (HPV) infection [1, 2]. positive nodes and extracapsular nodal involvement provide the
strongest prognostic factors of survival.
diagnosis non-palpable nodes. If nodes are non-palpable (cN0) at

Accurate histological diagnosis and staging of both the primary physical examination, dynamic sentinel node biopsy (DSNB) is
tumour and regional nodes are of utmost importance for indicated in intermediate (T1G2) or high-risk (T1G3 or worse)
selecting appropriate therapy. Penile cancer drains primarily to disease. Early detection of LN metastases by DSNB and
the inguinal nodes. The diagnosis of the primary tumour, the subsequent resection in clinically node negative T2-3 penile
inguinal lymph nodes (LNs) and regional and distant cancer improves survival compared with a policy of surveillance
metastases are discussed in Table 1. An incisional or excisional [III, C] [3]. Moreover, DSNB is a less morbid approach
biopsy is advised. Punch biopsy or scrapings may not be compared with prophylactic inguinal lymph reproducibility of
sufficiently representative. the technique, for which the sentinel node identification rate
was 97%, the false-negative rate was 7% and the complication
rate was 4.7% [III] [4]. A recent meta-analysis of 17 studies
primary tumour
demonstrated that DSNB is a method with a high detection rate
With regard to the primary tumour, the initial assessment of sentinel nodes ( pooled; 88.3%) and sensitivity ( pooled;
should be made by physical examination. The physical 88.0%). The highest detection rate and sensitivity was seen in
examination of suspected penile cancer must record: diameter studies using radiotracer and blue dye for sentinel LN mapping
of penile lesion(s) or suspicious areas; location of lesion(s) on and including only cN0 cases [II] [5]. If DSNB is not available,
the penis; number of lesions; morphology of lesion(s): papillary, ultrasound-guided fine-needle aspiration cytology (FNAC)
nodular, ulcerous or flat; relationship of lesion(s) to other biopsy of visualised nodes can be used [1].
structures, e.g. submucosa, tunica albuginea, urethra, corpus
spongiosum and corpus cavernosum; colour and boundaries of palpable nodes. If nodes are palpable, LN metastases can be
diagnosed using a percutaneous FNAC biopsy and/or histology.
*Correspondence to: ESMO Guidelines Working Group, ESMO Head Office, Via In the case of a negative biopsy and clinically suspicious nodes,
L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland. a repeat biopsy or node excision is advised [C] [6]. At the time
of diagnosis, almost half of palpable inguinal nodes are enlarged
†
Approved by the ESMO Guidelines Working Group: July 2013. due to inflammatory changes; however, those that become

by guest
on 05 February 2018
Table 1. Guidelines on the diagnosis of penile cancer [C]
Primary tumour
Physical examination recording morphological and physical characteristics of the lesion
Cytological and/or histological diagnosis
Regional lymph node disease
Physical examination of both groins, recording morphological and physical characteristics of the nodes
a) Non-palpable nodes → DSNB (if not available: ultrasound-guided FNAC biopsy/risk factors)
b) Palpable nodes → FNAC biopsy
Regional metastases (inguinal and pelvic nodes)
Pelvic CT scan/PET-CT scan in patients with metastatic inguinal nodes
More distant metastases
PET-CT scan (if not available CT scan and chest X-ray)
Bone scan in symptomatic patients
Molecular markers
Investigational, currently not useful in clinical practice
Reprinted from Pizzocaro G, et al. [1], with permission from Elsevier.

CT, computed tomography; DSNB, dynamic sentinel node biopsy; FNAC, fine-needle aspiration cytology; PET, positron emission tomography.
palpable during follow-up are malignant in nearly 100% of cases staging and risk assessment
[7]. MRI and computed tomography (CT) scanning can detect
enlarged inguinal and pelvic nodes. CT scan is used primarily, tumour-node-metastasis (TNM) classification
despite low sensitivity (36%). The use of 18F-fluorodeoxyglucose Penile cancer should preferably be staged according to the
positron emission tomography-computed tomography American Joint Committee on Cancer/Union for International
(18F-FDG PET/CT) remains uncertain. Cancer Control (AJCC/UICC) seventh edition TNM
classification (see Table 2) [9].
distant metastases
Scanning with 18F-FDG PET/CT appears encouraging for risk assessment
detection of pelvic LN metastases with great accuracy and also Patients can be prognostically stratified based on stage and/or
identifies more distant metastases in patients with inguinal grade into three risk groups according to the likelihood of
node-positive penile cancer [III/IV, C] [8]. harbouring occult node-positive disease: low-risk group; Tis,
TaG1-2 or T1G1, the intermediate-risk group; T1G2 and the
pathological categories high-risk group; T2 or any G3 [10]. Patients with T1G1 penile
Squamous cell carcinoma (SCC) accounts for more than 95% of SCC do not need further nodal assessment after local treatment
cases of penile cancer. Bowenoid papulosis (BP), Bowen’s [1]. In patients with intermediate T1G2 tumours, 13% up to 29%
disease (BD) and erythroplasia of Queyrat (EQ) are three develop LN metastases during follow-up. The 2009 European
recognised clinical manifestations of penile intra-epithelial Association of Urology (EAU) guidelines recommend either
neoplasia (carcinoma in situ) which are histologically DSNB or modified inguinal lymph node dissection (ILND) in
indistinguishable. BP is typically a raised papule on the penile clinical N0 patients with T1G2 nodular growth or vascular
shaft skin in a young male with a history of HPV exposure. BD invasion, T1G3 tumours and in all tumours T2 or higher [C] [10].
is a red scaly patch on the penile shaft, and EQ is a shiny At present, the risk for LN metastasis may be predicted by
erythematous plaque on the mucosal surface of the inner several tumour characteristics other than T and G categories [1].
prepuce and/or glans penis. EQ has the highest risk of Specifically, these risk factors include pathological subtypes,
developing SCC and BP the lowest. Balanitis xerotica obliterans invasion of perineural spaces, lymphovascular invasion, tumour
(lichen sclerosus et atrophicus) is a common lesion that is depth or thickness, anatomical site, size of the primary tumour,
associated with SCC but has no proven direct causal link. growth pattern, irregular front of invasion, positive margins of
SCC of the penis is classified as classic/usual type, basaloid, resection and urethral invasion. High histological grade,
verrucous, sarcomatoid or adenosquamous. Growth patterns perineural invasion and lymphovascular invasion appear to be
include superficial spreading, nodular or vertical-phase growth the strongest predictors of metastasis of penile cancer [III] [1].
and verrucous growth [1]. Verrucous carcinoma almost never invades LNs but gives rise to
distinct inflammatory nodal enlargement. The presence of
molecular biology central node necrosis and/or an irregular nodal border of the
Although several molecular prognostic markers have been regional LNs are very useful to identify high-risk pathological
evaluated, currently these markers are not useful in clinical node-positive penile cancer. Graafland et al. [11] demonstrated
practice. SCC antigen is not a sensitive marker of tumour an association between these unfavourable factors and poor
burden and has little prognostic significance for survival in prognosis, with a sensitivity of 95%, a specificity of 82% and a
patients with penile cancer treated with surgery [1]. diagnostic accuracy of 87% [IV]. Nomograms have been
vi | Van Poppel et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
Table 2. American Joint Committee on Cancer/Union for International Cancer Control (AJCC/UICC) seventh edition TNM clinical and pathological
classification of penile cancer
Clinical classification T: primary tumour

Tis Carcinoma in situ
Ta Non-invasive verrucous carcinoma, not associated with destructive invasion
T1 Tumour invades sub-epithelial connective tissue
T1a Without lymphovascular invasion and well or moderately differentiated (T1G1-2)
T1b With lymphovascular invasion or poorly differentiated/undifferentiated (T1G3-4)
T2 Tumour invades corpus spongiosum/corpora cavernosa
T3 Tumour invades urethra
T4 Tumour invades other adjacent structures
N: Regional lymph nodes
N0 No palpable or visibly enlarged inguinal lymph node
N1 Palpable mobile unilateral inguinal lymph node
N2 Palpable mobile multiple or bilateral inguinal lymph nodes
N3 Fixed inguinal nodal mass or pelvic lymphadenopathy, unilateral or bilateral
M: Distant metastasis
Pathological classification
The pT categories correspond to the T categories.
The pN categories are based upon biopsy or surgical excision.
pN: Regional lymph nodes
pNX Regional lymph nodes cannot be assessed
pN0 No regional lymph node metastasis
pN1 Intra-nodal metastasis in a single inguinal lymph node
pN2 Metastasis in multiple or bilateral inguinal lymph nodes
pN3 Metastasis in pelvic lymph node(s), unilateral or bilateral or
extranodal extension of regional lymph node metastasis
pM: Distant metastasis
pM0 No distant metastasis
pM1 Distant metastasis
G: Histological grading
GX Grade of differentiation cannot be assessed
G1 Well differentiated
G2 Moderately differentiated
G3-4 Poorly differentiated/undifferentiated
From [12]. Used with the permission of the American Joint Committee on Cancer (AJCC), Chicago, Illinois. The original source for this material is the AJCC
Cancer Staging Handbook, Seventh Edition (2010) published by Springer Science and Business Media LLC, www.springer.com.
developed based on clinical and pathologic parameters to excision and circumcision or more radical amputative
predict and identify patients at the risk of nodal metastasis. procedures loosely divided into partial penectomy and radical
Recently, Thuret et al. [13] reviewed several nomograms and penectomy.
found that the staging system developed by the AJCC combined The pathological argument for extent of resection was based
with tumour grade was the most simple and most accurate on the principle that at least a 2cm clear margin from
method (80.9%) to predict cancer specific mortality after macroscopic disease was required for local control. In many
primary tumour excision for penile cancer. Even the most cases, a partial penectomy carried out would leave the patient
sensitive nomogram is inaccurate in predicting patients with with a short penile stump, perhaps, but not always, suitable for
positive LNs with as many as 75% of patients predicted to be at erect micturition, and with limited sexual function. The
high risk of LN metastasis being pathologically negative [14]. psychological morbidity of these procedures was and still
remains significant. Radical radiotherapy offered patients a
treatment choice of penile preservation albeit with its own treatment
complications (see section below). Certainly in the UK, many
primary surgery patients were treated routinely with either external beam or
Until 15 years ago, surgical options to treat invasive penile brachytherapy for distal stage T1 and T2 tumours with surgery
cancer were limited to minor procedures such as wide local for salvage recurrences. The evidence that local recurrence did
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt286 | vi

by guest
on 05 February 2018
not adversely influence survival was an important factor in Compared with external beam treatment, the volume of the area
supporting this approach. treated to a high dose is smaller, but the dose inhomogeneity
Around this time, two important factors encouraged a change within this volume is more pronounced. Penile brachytherapy
in practice. First, surgeons began to question the premise that can be performed under general anesthesia or penile block with
surgical margins had to be so generous. A margin of <5 mm is systemic sedation. Low dose rate brachytherapy consists of
adequate for most tumours [III] [15]. The recurrence rate of either manually afterloaded 192 Ir or pulse dose rate
patients with resection margins of 5 mm or less could still be brachytherapy with automated afterloading with a high-
<5%, and this led to newer penile-preserving techniques being intensity 192 Ir source to deliver hourly pulses. A typical
developed. Secondly, and particularly in the UK and the brachytherapy schedule consists of 55–60 Gy given in 4–6 days.
Netherlands, specialist centres treating high volumes of penile Results of brachytherapy have been reported in about 20
cancer patients allowed these techniques to be evaluated. studies. All except two of these studies reported on fewer than
Previously, the main block to progress was the rarity of the 80 patients. The largest study by Rozan et al. [16] reported on
disease and small numbers of patients reported in the literature. 259 patients, of whom 184 had been treated by brachytherapy
In general, we can now group surgical techniques into three only and 75 had a combination of external beam treatment and
broad categories. First, for small volume and superficial penile brachytherapy. In the vast majority of the studies, the patients
lesions, circumcision, wide local excision and epithelial ablative were treated over a period exceeding one or even two or three
techniques are still mainstay treatments. One has to remember decades. Treatment parameters such as tumour dose, dose rate,
that the glanular and preputial mucosa is probably affected as a fractionation schedule etc. varied considerably among the
field change. Local recurrences over time may occur and patients reported within the individual studies. Also, patient
retreatment is a distinct possibility. Secondly, for glanular and selection criteria were not uniformly applied in most of the
distal penile tumours, it is now possible to preserve much more reports. Despite this wide variety in treatment parameters and
length, and cosmetic and functional results are far superior to patient characteristics, the outcome of the studies is remarkably
conventional partial penectomy. This balances the argument concordant [17–19]. Long-term (5–10 years) local control rates
between the choice of surgery and radiotherapy for such vary between 60% and 90% and seem more related to tumour
tumours (which in developed countries are the majority of characteristics than treatment parameters. According to the
presentations). Thirdly, surgeons are now being more aggressive 2013 ABS-GEC-ESTRO consensus statement, the good tumor
in extending the preservation techniques, with ongoing studies control rates, acceptable morbidity, and functional organ
looking at both less resecting for superficial tumours and more preservation warrant recommendation of brachytherapy as the
penile reconstruction ( phalloplasty) for more advanced initial treatment for invasive T1, T2, and selected T3 penile
tumours suitable only for total penectomy. cancers [20]. Adequate surgical salvage possibilities with a
The role of salvage surgery after radio/chemotherapy remains success rate between 70% and 100% are observed and reported
an area of controversy. For patients with extensive regional penis conservation rate is between 52% and 86%. The most
disease in which primary surgery is deemed unlikely to result in important predictors for successful brachytherapy seem to be
a clear margin, attempts at down-staging are appropriate. If the tumour size (less or more than 4 cm) and tumour location
patient responds to treatment, surgical resection can be limited to the glans or the prepuce without corpus cavernosum
considered. There is no evidence that surgery in this situation is involvement. For patients meeting these criteria, different
inherently more complicated than for patients with bulky nodal studies report local recurrence rates of about 20% after 5–10
disease. years with a secondary control of about 85% of the recurrences
There are no direct comparisons between radiotherapy and by salvage surgery.
the newer penile-preserving techniques, and studies with External beam treatment as single treatment modality has
chemotherapy are very limited. been used in only a small number of studies, most of these
reporting on limited numbers of patients [21]. One exception is
a study by Gotsadze et al. [22] analysing results in 155 patients.
radiotherapy Reported local control rates for stages I and II range from 65%
radiotherapy as treatment of the primary tumour. In order to to 90%. Sarin et al. [23] analysed the impact of various radiation
deliver the radiation dose to the tumour, there are two options: parameters such as total dose, dose per fraction, total treatment
external beam treatment or brachytherapy. By using external time and ‘biological equivalent dose’ with or without time factor
megavoltage radiation beams, a relatively homogeneous dose is on local failure in 44 patients with T1 tumours. A higher
delivered in the target region. Tissue-equivalent bolus is often incidence of local failure was observed with total dose <60 Gy,
required to provide sufficient dose build-up to the surface of the dose per fraction <2 Gy and treatment time exceeding 45 days.
lesion. Using fractionated treatment, schedules can spare Radiotherapy can induce adverse effects and complications
normal tissues. A typical radical external beam course consists such as teleangiectasia, atrophy and depigmentation of the skin,
of one daily fraction of about 2 Gy, five fractions per week fibrosis, urethral stenosis and ( painful) ulcerations and necrosis.
during 6–7 weeks to a total dose of 66–70 Gy. Brachytherapy Most serious complications are urethral stenosis and persisting
(brachy is from the Greek for short distance) consists of placing ulceration or necrosis. The reported incidence ranges from 8%
sealed radioactive sources very close to or in contact with the to 45% and 0% to 23%, respectively. The reported incidence of
tumour. Because the absorbed dose falls off very rapidly with penectomy for radiation complications varies from 0% to 5%
increasing distance from the sources, high doses can be [19, 23]. Only very few data are available concerning functional
delivered safely to a localised target region over a short time. and psychosexual outcome of organ preservation by
vi | Van Poppel et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
radiotherapy. Opjordsmoen et al. [24] reported on post-therapy referral bias, and pointing to inherent changes in clinical staging
sexuality in 30 men. Patients underwent a semi-structured and patient management over the period of patient inclusion.
interview and completed three self-administered questionnaires There are only phase II and no randomised clinical studies
focusing on psychosocial adjustment to severe illness, mental reported; thus, the level of evidence on chemotherapy is no
symptoms and quality of life. A global score of overall sexual more than grade III and the recommendation C.
function was constructed, consisting of sexual interest, sexual Cisplatin has been the cornerstone of combination regimens
ability, sexual satisfaction, sexual identity, partner relationship used. The above mentioned limitations of the various studies,
and frequency of coitus. In 10 of 12 patients treated with often containing heterogeneous patient populations without
irradiation, the global sexual score was not reduced or was only stratification for prognostic factors, preclude the drawing of any
slightly reduced. Eleven out of 14 patients treated with local firm conclusions concerning an optimal first-line or second-line
surgery, laser beam treatment or partial penectomy had chemotherapy regimen.
moderate to severe reduction of global sexual score. All four The largest prospective study was reported in 1999 and
patients who had undergone total penectomy recorded a included only 45 patients, who were treated with the
severely reduced global sexual score. It was remarkable that, in combination regimen bleomycin/methotrexate/cisplatin (BMP)
the patients treated by irradiation, physicians evaluated the [31]. It was advised to omit bleomycin from future studies based
patients’ post-treatment sexuality, especially with regard to the on the high incidence of fatalities and severe lung toxic effect.
ability to perform coitus, to be more impaired than was Although the question remains unanswered, it seems likely that
evaluated by the patients themselves. a bleomycin-containing cisplatin-based regimen can be safely
applied in patients after exclusion of older age, heavy smokers
radiotherapy for the management of regional lymph node and those with compromised lung function.
metastases Efficacy of anticancer treatment regimens is usually based on
elective radiation: Elective radiation of clinically uninvolved results obtained in metastatic cancer patients. Activity/efficacy
LN regions has proven to be an effective policy in many observed in the neoadjuvant setting before definitive local
tumours, especially in SCC. In penile cancer, elective treatment (often surgery) may fuel enthusiasm and serve as an
radiotherapy has never been widely used. Arguments against its incentive to apply the same drug or combination of drugs in the
use have been related to the unproven efficacy, the possibility of metastatic setting. Various regimens have been reported to be
complications and the concern that radiation-induced fibrotic effective and to induce partial responses in about 20%–33% of
changes might hinder reliable follow-up. advanced penile cancer patients (prospective studies). In none of
the studies were modern Response Evaluation Criteria in Solid
adjuvant postoperative radiation: The role of adjuvant Tumors (RECIST) criteria applied to measure clinical responses.
postoperative radiation is controversial. The incidence of
inguinal failure in patients with inguinal LN metastasis treated perioperative chemotherapy. Cytoreductive neoadjuvant
with lymphadenectomy varies between 25% and 77% [25–27]. chemotherapy has been applied to allow surgery or radiotherapy
Adjuvant radiation to the inguinal lymphatic area has been to obtain local control, but there are no prospective trials of
advocated by some, if histopathological examination of the adjuvant chemotherapy, only small retrospective series.
inguinal specimen revealed more than one metastatic LN and/or Cisplatin combination chemotherapy regimens are the most
extranodal extension. Chen et al. reported that postoperative widely used and seem to be the most effective [III, C]. Four
groin radiation reduced the inguinal recurrence rate from 60% cycles of neoadjuvant paclitaxel in combination with cisplatin
(three recurrences in five patients) to 11% (one recurrence in and ifosfamide chemotherapy has been shown to be well
nine patients) [28]. However, larger series confirming this tolerated and effective in patients with bulky regional disease
benefit are lacking. (any T, N2 or N3 according to the seventh edition of the AJCC/
UICC TNM staging classification system) but who had no
chemoradiotherapy: Although studies in other SCCs of the evidence of distant metastatic disease [III, C] [32]. An overall
perineal area, e.g. vulvar and anal cancer, have demonstrated the ( partial and complete) response rate of 50% was achieved with
efficacy of chemoradiation regimes, prospective studies of such 30% of patients experiencing stable disease and 20% progressive
strategies are unavailable in penile cancer. A few retrospective disease. The high response rate allowed consolidation surgery in
studies suggest some benefit of radiotherapy with concurrent the majority of patients with a pathologic complete response in
cisplatin-based chemotherapy in locally advanced unresectable 13.6% of patients who completed chemotherapy and underwent
disease [29, 30]. surgery. Neoadjuvant chemotherapy followed by consolidation
The precise role of chemoradiation, eventually in surgery resulted in a meaningful remission in a considerable
combination with targeted therapies, remains an important number of patients, with 30% of patients who remained disease-
research question for the near future. free at a median follow-up of 36 months (14–59 months) [32].
Also paclitaxel, cisplatin and 5-fluorouracil (5-FU) may provide
chemotherapy an attractive regimen and preliminary data of the combination
The literature on the role of chemotherapy in penile cancer is in neoadjuvant setting were reported in 2009 by Pizzocaro et al.
limited due to the rarity of the disease in Western Europe and [33]. Adjuvant chemotherapy is recommended in pN2-3
the United States with invariably small, prospective and patients [C] [34]. Neoadjuvant chemotherapy followed by
retrospective studies. Results are often obtained over a radical surgery is advisable in unresectable or recurrent LN
prolonged period of time reflecting potential selection and metastases [C] [32, 35, 36].
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt286 | vi

by guest
on 05 February 2018
vi | Van Poppel et al.
Table 3. Recommendations for the treatment of penile cancer
Disease stage Recommendation Grade

Tis or Ta Penile-preserving techniques, including topical therapy (5% 5-fluorouracil and 5% imiquimod cream) [IV] [41], circumcision and wide local excision [IV] [42], laser C
therapy using CO2 or Nd:YAG laser [III] [43, 44], partial/total glans resurfacing [III] [45, 46]
T1G1-2 Penile-preserving techniques, including wide local excision plus reconstructive surgery with split-thickness skin graft or full-thickness skin graft [III] [47] laser therapy C
[IV] [48], radiotherapy delivered as EBRT or brachytherapy with interstitial implant [IV] [18, 49, 50]
T1G3–4, T ≥ 2 If tumour <50% of the glans and no invasion of the corpora cavernosa: wide local excision or glansectomy [III] [47] B
Tumours with invasion into corpora cavernosa: partial or total penectomy [III] [51] B
T1-2, N0 and tumour <4 cm: circumcision followed by brachytherapy alone or EBRT with or without chemotherapy [III] [18, 20, 49, 50] C
T1-2, N0 and tumour >4 cm: circumcision followed by either EBRT with chemotherapy or brachytherapy in select cases and with post-treatment surveillance [52] D
For T2 tumours only, radiotherapy with or without concurrent chemotherapy [53] D
T3-4 or N+: circumcision followed by EBRT with chemotherapy [53] D
Non-palpable lymph nodes Low- risk (Tis, Ta, T1G1) and intermediate-risk (T1G2) patients without lymphovascular invasion are followed with surveillance [54] B
DSNB is recommended in patients with non-palpable inguinal lymph nodes T1G2 or greater [1, 53] B
If positive nodes are found on DSNB, ILND is recommended [III] [55] B
If DSNB is not available: ILND based on risk factors or nomograms [IV] [10] C
Unilateral or bilateral palpable Fine-needle aspiration (FNA) of the LN is standard for these patients (omitting the procedure for high-risk tumours to avoid delay of ILND) [III] [56] B
inguinal nodes A negative FNA result should be confirmed with an excisional biopsy or followed with careful surveillance [1, 53] C
Positive findings from either procedure warrant an immediate ILND [III] [53, 54] C
a) If 0-1 nodes are positive → no further treatment C
b) If ≥2 nodes are positive or when extranodal extension is found (consider adjuvant chemotherapy) → pelvic LND (PLND) (consider postoperative radiotherapy)
When pelvic LNs (PLNs) are enlarged → systemic chemotherapy or radiotherapy with concurrent chemotherapy + percutaneous biopsy or PET/CT [53] C
Fixed or ulcerated inguinal nodes Patients with non-fixed nodes can be considered for inguinal node dissection with the option to use a skin flap to cover the defect C
Patients with fixed nodes should be considered for neoadjuvant chemoradiotherapy [III] [1, 36] C
Responders receive consolidation surgery (bilateral and deep ILND and ipsilateral PLND if possible) [III] [32] C
Volume 24 | Supplement 6 | October 2013
Patients with disease progression or unresectable LN may consider additional systemic chemotherapy, local-field radiotherapy or participation in a clinical trial C
Recurrent disease For recurrences without invasion of the corpora cavernosa salvage penile-sparing options can be considered [IV] [43–45, 47] C
Invasion of the corpora cavernosa warrants partial or total penectomy [IV] [22] B
For local recurrences in the inguinal region, consider systemic chemotherapy, EBRT, surgery or a combination [57] C
Metastatic penile cancer Treatment options include systemic chemotherapy or radiotherapy or radiotherapy with concurrent chemotherapy [58]
a) Responders receive consolidation ILND B
b) For those with no response/ disease progression, consider salvage systemic chemotherapy or radiotherapy for local control and/or best supportive care/ clinical trial C
Annals of Oncology
Nd:YAG, neodynium yttrium-aluminium-garnet; EBRT, external beam radiation therapy; ILND, inguinal lymph node dissection; DSNB, dynamic sentinel node biopsy; FNA, fine-needle aspiration; PLND, pelvic
lymph node dissection; PET/CT, positron emission tomography/computed tomography.

by guest
on 05 February 2018
Figure 1. Guidelines on treatment strategies for the primary tumour.

* grade of recommendation is D.
Figure 2. Guidelines on treatment strategies for the regional LNs.
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt286 | vi

by guest
on 05 February 2018
Grading Systema)
Levels of evidence
I Evidence from at least one large randomised, controlled trial of good methodological quality (low potential for bias) or meta-analyses of well-conducted
C Insufficient evidence for efficacy or benefit does not outweigh the risk or the disadvantages (adverse events, costs, …), optional
a
chemotherapy for advanced disease. Cisplatin has been used in the adverse effects of the treatment and the treatment centre’s
combination with agents such as 5-FU or irinotecan. Patients experience. For superficial and glans confined tumours, a
were treated in the neoadjuvant setting for T3 or N1-N2 disease penile-preserving strategy is recommended [B] [1]. Guidelines
with a maximum of four cycles before surgery, or up to eight on treatment strategies for the primary tumour are presented in
cycles for T4 or N3 or M1 disease. There were eight clinical Figure 1.
responses (30.8%) in 26 eligible patients (two complete and six
partial responses) [37]. A sustained palliative response has been regional lymph nodes. Lymphadenectomy is the standard
observed with combination chemotherapy using cisplatin and treatment of patients with inguinal LN metastases [B] [59].
gemcitabine for the management of metastatic penile cancer Guidelines on treatment strategies for the regional LNs are
[38]. The chemotherapy combination ifosfamide, paclitaxel and presented in Figure 2.
cisplatin may also be a rational regimen in metastatic disease,
based on the activity in the neoadjuvant setting [32]. In a
retrospective study, cisplatin and 5-FU were used in 25 patients recurrence. Local recurrence rate after conservative surgery
with metastatic penile cancer. Partial responses were observed in does not seem to have a negative impact on long-term survival.
8 patients (32%). The median progression-free survival (PFS)
and overall survival (OS) were 20 weeks and 8 months, metastatic disease. The OS of patients with metastatic disease
respectively. The combination was well tolerated. Severe (beyond the pelvic nodes) is 0% at 5 years and <10% at 2 years.
neutropenia was the most important grade 3-4 adverse effect Patients who present with metastatic disease have a very poor
observed, occurring in 20% of patients [39]. prognosis and early consideration of palliative care is
Paclitaxel in combination with carboplatin may provide an recommended.
alternative regimen in patients who are ineligible for cisplatin
treatment. The taxane paclitaxel demonstrated efficacy as a
single-agent therapy in 25 metastatic patients who were response evaluation
previously treated in the neoadjuvant or adjuvant setting with
Penectomy is disfiguring and can have an intense effect on the
cisplatin combination chemotherapy regimens, with a partial
patient’s quality of life, sexual function, self-esteem and general
response rate of 20%. The median PFS was only 11 weeks and
mental health. Therefore, there is an increased trend for penile-
the median OS was 23 weeks [III, C] [40].
preserving strategies, despite the fact that recurrence rates may
The phase II or retrospective nature of the studies reported
be higher than those of radical surgical procedures. In the
and their small sample size, plus the lack of any randomised
choice of treatment, the patient’s preference must be considered.
clinical trial comparing different regimens, preclude assessment
Adverse effects must be weighed against one another to allow
of a superior drug regimen in the setting of patients with distant
correct treatment selection. Also important is the patient’s
metastatic disease or bulky regional/locally advanced disease.
compliance in attending follow-up visits after treatment.
Recommendations for treatment strategies for the primary
tumour in penile cancer per stage and grade for nodal
metastases, recurrent disease and metastatic penile cancer are
presented in Table 3.
In this disease setting, more research is needed to identify
primary tumour. The choice of treatment is influenced by the molecular markers which could lead to advances in
size and position of the tumour on the glans or in the corpora, personalised medicine.
vi | Van Poppel et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
follow-up 3. Lont AP, Horenblas S, Tanis PJ et al. Management of clinically node negative
penile carcinoma: improved survival after the introduction of dynamic sentinel
The aim of follow-up is to detect early recurrences when they node biopsy. J Urol 2003; 170: 783–786.
can still be cured. New imaging modalities such as 18F-FDG 4. Leijte JA, Hughes B, Graafland NM et al. Two-center evaluation of dynamic
PET/CT scanning can improve the detection of early regional sentinel node biopsy for squamous cell carcinoma of the penis. J Clin Oncol 2009;
and distant metastases. A large retrospective study based on a 27: 3325–3329.
two-centre analysis of 700 patients with penile cancer showed 5. Sadeghi R, Gholami H, Zakavi SR et al. Accuracy of sentinel lymph node biopsy for
inguinal lymph node staging of penile squamous cell carcinoma: systematic review
that 74.1% and 92.2% of all recurrences occurred within the first and meta-analysis of the literature. J Urol 2012; 187: 25–31.
2 and 5 years, respectively [60]. 6. Graafland NM, Leijte JA, Olmos RA et al. Repeat dynamic sentinel node biopsy in
locally recurrent penile carcinoma. BJU Int 2010; 105: 1121–1124.
primary tumour 7. Ornellas AA, Seixas AL, Marota A et al. Surgical treatment of invasive squamous
cell carcinoma of the penis: retrospective analysis of 350 cases. J Urol 1994;
Local recurrence has been reported in 28% of patients during
151: 1244–1249.
the first 2 years of follow-up after penis-preserving surgery and 8. Schlenker B, Scher B, Tiling R et al. Detection of inguinal lymph node involvement
is significantly higher than the local recurrence rate after in penile squamous cell carcinoma by 18f-fluorodeoxyglucose PET/CT: a
penectomy (6%). Following penile-preserving treatment, a prospective single-center study. Urol Oncol 2012; 30: 55–59.
follow-up visit every 3 months is advised in the first 2 years and 9. Sobin L, Gospodariwics M, Wittekind C. TNM Classification of Malignant Tumors.
every 6 months in the following 3 years. Patients must continue UICC International Union Against Cancer, 7th edition. West Sussex, UK: Willy-
with regular self-examination and must report any changes even Blackwell 2009; 239–242.
after a 5-year follow-up. After penectomy, the intervals of 10. Solsona E, Algaba F, Horenblas S et al. EAU Guidelines on penile cancer. Eur Urol
2004; 46: 1–8.
follow-up visits are 6 months in the first 2 years and 1 year in
11. Graafland NM, Teertstra HJ, Besnard AP et al. Identification of high risk
the following 3 years [C] [60].
pathological node positive penile carcinoma: value of preoperative computerized
tomography imaging. J Urol 2011; 185: 881–887.
regional recurrences 12. Edge SB, Byrd DR, Compton CC, eds. AJCC Cancer Staging Handbook, 7th ed.
Most regional recurrences occur within 2 years after ILND or New York, NY: Springer, 2010.
13. Thuret R, Sun M, Abdollah F et al. Tumor grade improves the prognostic ability of
DSNB. For patients subjected to a ‘wait-and-see’ policy (groins
American Joint Committee on Cancer stage in patients with penile carcinoma. J
not surgically staged) and patients with positive nodes, Urol 2011; 185: 501–507.
meticulous follow-up with ultrasound-guided FNAC biopsy of 14. Graafland NM, Lam W, Leijte JA et al. Prognostic factors for occult inguinal lymph
the groin is advised every 3 months in the first 2 years and every node involvement in penile carcinoma and assessment of the high-risk EAU
6 months in the following 3 years [C] [60]. subgroup: a two-institution analysis of 342 clinically node-negative patients. Eur
For patients surgically staged for negative nodes, the regional Urol 2010; 58: 742–747.
relapse rate is lower and the intervals for follow-up visits 15. Minhas S, Kayes O, Hegarty P et al. What surgical resection margins are required
(ultrasound-guided FNAC biopsy) are 3 months in the first 2 to achieve oncological control in men with primary penile cancer? BJU Int 2005;
96: 1040–1043.
years and 6 months for years 3–5. The risk of recurrence rates
16. Rozan R, Albuisson E, Giraud B et al. Interstitial brachytherapy for penile
for patients that underwent a ‘wait-and-see’ policy, for patients
carcinoma: a multicentric survey (259 patients). Radiother Oncol 1995; 36:
staged as pN0 and for patients staged as pN+ were 9%, 2.3% and 83–93.
19%, respectively [60]. 17. Kiltie AE, Elwell C, Close HJ et al. Iridium-192 implantation for node-negative
carcinoma of the penis: the Cookridge Hospital experience. Clin Oncol (R Coll
Radiol) 2000; 12: 25–31.
note 18. de Crevoisier R, Slimane K, Sanfilippo N et al. Long-term results of brachytherapy
Levels of evidence and grades of recommendation have been for carcinoma of the penis confined to the glans (N- or NX). Int J Radiat Oncol Biol
Phys 2009; 74: 1150–1156.
applied using the system shown in Table 4. Statements without
19. Crook J, Grimard L, Tsihlias J et al. Interstitial brachytherapy for penile cancer: an
grading were considered justified standard clinical practice by
alternative to amputation. J Urol 2002; 167: 506–511.
the experts and the ESMO faculty. 20. Crook JM, Haie-Meder C, Demanes DJ et al. American Brachytherapy Society-
Groupe Europeén de Curiethérapie-European Society of Therapeutic Radiation
Oncology (ABS-GEC-ESTRO) consensus statement for penile brachytherapy.
conflict of interest Brachytherapy 2013; 12: 191–198.
Prof. Kataja has reported institutional clinical research support 21. Zouhair A, Coucke PA, Jeanneret W et al. Radiation therapy alone or combined
from Sanofi, Bayer Health Care, Orion Pharma and Merck; the surgery and radiation therapy in squamous-cell carcinoma of the penis? Eur J
Cancer 2001; 37: 198–203.
arrangements do not involve personal financial support from
22. Gotsadze D, Matveev B, Zak B, Mamaladze V. Is conservative organ-sparing
the companies mentioned. The other authors have declared no treatment of penile carcinoma justified? Eur Urol 2000; 38: 306–312.
potential conflicts of interest. 23. Sarin R, Norman AR, Steel GG, Horwich A. Treatment results and prognostic
factors in 101 men treated for squamous carcinoma of the penis. Int J Radiat
Oncol Biol Phys 1997; 38: 713–722.
references 24. Opjordsmoen S, Waehre H, Aass N, Fossa SD. Sexuality in patients treated for
1. Pizzocaro G, Algaba F, Horenblas S et al. EAU Penile cancer guidelines 2009. Eur penile cancer: patients’ experience and doctors’ judgement. Br J Urol 1994; 73:
Urol 2010; 57: 1002–1012. 554–560.
2. Ferlay J, Shin HR, Bray F et al. Estimates of worldwide burden of cancer in 2008: 25. Horenblas S, van Tinteren H, Delemarre JF et al. Squamous cell carcinoma of the
Globocan 2008. Int J Cancer 2010; 127: 2893–2917. penis. III. treatment of regional lymph nodes. J Urol 1993; 149: 492–497.
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt286 | vi

by guest
on 05 February 2018
26. Culkin DJ, Beer TM. Advanced penile carcinoma. J Urol 2003; 170: 359–365. 44. Windahl T, Andersson SO. Combined laser treatment for penile carcinoma: results
27. Ozsahin M, Jichlinski P, Weber DC et al. Treatment of penile carcinoma: to cut or after long-term followup. J Urol 2003; 169: 2118–2121.
not to cut? Int J Radiat Oncol Biol Phys 2006; 66: 674–679. 45. Hadway P, Corbishley CM, Watkin NA. Total glans resurfacing for premalignant
28. Chen MF, Chen WC, Wu CT et al. Contemporary management of penile cancer lesions of the penis: initial outcome data. BJU Int 2006; 98: 532–536.
including surgery and adjuvant radiotherapy: an experience in Taiwan. World J Urol 46. Shabbir M, Muneer A, Kalsi J et al. Glans resurfacing for the treatment of
2004; 22: 60–66. carcinoma in situ of the penis: surgical technique and outcomes. Eur Urol 2011;
29. Pedrick TJ, Wheeler W, Riemenschneider H. Combined modality therapy for locally 59: 142–147.
advanced penile squamous cell carcinoma. Am J Clin Oncol 1993; 16: 501–505. 47. Smith Y, Hadway P, Biedrzycki O et al. Reconstructive surgery for invasive
30. Eliason M, Bowen G, Bowen A et al. Primary treatment of verrucous carcinoma of squamous carcinoma of the glans penis. Eur Urol 2007; 52: 1179–1185.
the penis with fluorouracil, cis-diamino-dichloro-platinum, and radiation therapy. 48. Schlenker B, Tilki D, Seitz M et al. Organ-preserving neodymium-yttrium-
Arch Dermatol 2009; 145: 950–952. aluminium-garnet laser therapy for penile carcinoma: a long-term follow-up. BJU
31. Haas GP, Blumenstein BA, Gagliano RG et al. Cisplatin, methotrexate and Int 2010; 106: 786–790.
bleomycin for the treatment of carcinoma of the penis: a Southwest Oncology 49. Azrif M, Logue JP, Swindell R et al. External-beam radiotherapy in T1–2 N0 penile
Group study. J Urol 1999; 161: 1823–1825. carcinoma. Clin Oncol (R Coll Radiol) 2006; 18: 320–325.
32. Pagliaro LC, Williams DL, Daliani D et al. Neoadjuvant paclitaxel, ifosfamide, and 50. Crook JM, Girmard L, Esche B, Pond G. Penile brachytherapy: results for 60
cisplatin chemotherapy for metastatic penile cancer: a phase II study. J Clin Oncol patients. Brachytherapy 2007; 6: 82.
2010; 28: 3851–3857. 51. Lont AP, Gallee MP, Meinhardt W et al. Penis conserving treatment for T1 and T2
33. Pizzocaro G, Nicolai N, Milani A. Taxanes in combination with cisplatin and penile carcinoma: clinical implications of a local recurrence. J Urol 2006; 176:
fluorouracil for advanced penile cancer: preliminary results. Eur Urol 2009; 55: 575–580; discussion 580.
546–551. 52. Crook J, Jezioranski J, Cygler JE. Penile brachytherapy: technical aspects and
34. Pizzocaro G, Piva L, Bandieramonte G, Tana S. Up-to-date management of postimplant issues. Brachytherapy 2010; 9: 151–158.
carcinoma of the penis. Eur Urol 1997; 32: 5–15. 53. National Comprehensive Cancer Network (NCCN). NCCN Clinical Practice
35. Leijte JA, Kerst JM, Bais E et al. Neoadjuvant chemotherapy in advanced penile Guidelines in Oncology: Penile Cancer, version 1, 2013. http://www.nccn.org/
carcinoma. Eur Urol 2007; 52: 488–494. professionals/physician_gls/pdf/penile.pdf. (15 July 2013, date last accessed).
36. Bermejo C, Busby JE, Spiess PE et al. Neoadjuvant chemotherapy followed by 54. Ornellas AA, Kinchin EW, Nóbrega BL et al. Surgical treatment of invasive
aggressive surgical consolidation for metastatic penile squamous cell carcinoma. J squamous cell carcinoma of the penis: Brazilian National Cancer Institute long-
Urol 2007; 177: 1335–1338. term experience. J Surg Oncol 2008; 97: 487–495.
37. Theodore C, Skoneczna I, Bodrogi I et al. A phase II multicentre study of irinotecan 55. Leijte JA, Kroon BK, Valdés Olmos RA et al. Reliability and safety of current
(CPT 11) in combination with cisplatin (CDDP) in metastatic or locally advanced dynamic sentinel node biopsy for penile carcinoma. Eur Urol 2007; 52:
penile carcinoma (EORTC PROTOCOL 30992). Ann Oncol 2008; 19: 1304–1307. 170–177.
38. Power DG, Galvin DJ, Cuffe S et al. Cisplatin and gemcitabine in the management 56. Saisorn I, Lawrentschuk N, Leewansangtong S, Bolton DM. Fine-needle aspiration
of metastatic penile cancer. Urol Oncol 2009; 27: 187–190. cytology predicts inguinal lymph node metastasis without antibiotic pretreatment in
39. Di Lorenzo G, Buonerba C, Federico P et al. Cisplatin and 5-fluorouracil in penile carcinoma. BJU Int 2006; 97: 1225–1228.
inoperable, stage IV squamous cell carcinoma of the penis. BJU Int 2012; 110: 57. Graafland NM, Moonen LM, van Boven HH et al. Inguinal recurrence following
E661–E666. therapeutic lymphadenectomy for node positive penile carcinoma: outcome and
40. Di Lorenzo G, Federico P, Buonerba C et al. Paclitaxel in pretreated metastatic implications for management. J Urol 2011; 185: 888–893.
penile cancer; final results of a phase 2 study. Eur Urol 2011; 60: 1280–1284. 58. Pettaway CA, Pagliaro L, Theodore C, Haas G. Treatment of visceral, unresectable,
41. Alnajjar HM, Lam W, Bolgeri M et al. Treatment of carcinoma in situ of the glans or bulky/unresectable regional metastases of penile cancer. Urology 2010; 76
penis with topical chemotherapy agents. Eur Urol 2012; 62: 923–928. (Suppl. 1): S58–S65.
42. Feldman AS, McDougal WS. Long-term outcome of excisional organ sparing 59. Horenblas S. Lymphadenectomy in penile cancer. Urol Clin North Am 2011; 38:
surgery for carcinoma of the penis. J Urol 2011; 186: 1303–1307. 459–469, vi–vii.
43. Bandieramonte G, Colecchia M, Mariani L et al. Peniscopically controlled CO2 60. Leijte JA, Kirrander P, Antonini N et al. Recurrence patterns of squamous cell
laser excision for conservative treatment of in situ and T1 penile carcinoma: report carcinoma of the penis: recommendations for follow-up based on a two- centre
on 224 patients. Eur Urol 2008; 54: 875–882. analysis of 700 patients. Eur Urol 2008; 54: 161–168.
vi | Van Poppel et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
Cancer of the prostate: ESMO Clinical Practice

C. Parker1, S. Gillessen2, A. Heidenreich3 & A. Horwich4, on behalf of the ESMO
1
Royal Marsden Hospital, Sutton, UK; 2Department of Oncology/Hematology, Kantonsspital St Gallen, St Gallen, Switzerland; 3Department of Urology,
Uniklinik RWTH Aachen, Aachen, Germany; 4Institute of Cancer Research, Sutton, UK
screening and early detection recommendations

• A single elevated PSA level should not prompt a prostate
Subclinical prostate cancer is common in men >50 years of age.
biopsy, and should be verified by a second value [IV, B].
Population-based screening of men aged between 55 and 69
years, using prostate-specific antigen (PSA) testing, has been
• The decision whether or not to have a prostate biopsy should
be made in the light of DRE findings, ethnicity, age, co-mor-
evaluated in randomised trials [1, 2]. After a median follow-up
bidities, PSA values, free/total (f/t) PSA, history of previous
of 13 years, the European screening trial demonstrated a relative
biopsy and patient values [II, B].
reduction in the risk of prostate cancer mortality of 21% (29% if
adjusted for non-compliance). However, 781 men needed to be
• Transrectal ultrasound-guided prostate biopsy should be
clinical practice
carried out under antibiotic cover and local anaesthesia, and a
invited for screening and 27 patients needed to be treated to
guidelines
minimum of 10–12 cores obtained [II, B].
prevent one death from prostate cancer. Risk-adapted early de-
tection of prostate cancer using a baseline PSA level has also
• Before repeat biopsy, multi-parametric MRI is recommended
with a view to MRI-guided or MRI-transrectal ultrasound
been evaluated in retrospective cohort studies. The baseline PSA
(TRUS) fusion biopsy [5] [III, B].
at or before the age of 50 years is associated with the risk of
prostate cancer mortality over the subsequent 25 years [3]. The most dominant Gleason pattern and the pattern with the
highest Gleason grade determine the biopsy Gleason score [6].
recommendations Biopsy pathology should be reported using the International
• Population-based PSA screening for prostate cancer reduces Society of Urologic Pathology recommendations.
prostate cancer mortality at the expense of over diagnosis and
overtreatment and is not recommended [I, C]. recommendation
• Testing for prostate cancer in asymptomatic men should not • The extent of involvement of each biopsy core, and the com-
be done in men over the age of 70 years [I, B]. monest and the worst Gleason grades should be reported [II, A].

diagnosis and pathology
General health and co-morbidities should be assessed. Patients
The risk of clinically significant prostate cancer is related to age, who are not suitable for treatment with curative intent, by virtue
ethnicity, family history, PSA level, free/total PSA ratio and find- of poor general health, do not normally require staging investi-
ings on digital rectal examination (DRE) [4]. High-grade prostate gations.
cancer can occur in men with a ‘normal’ PSA level. After a previ-
ous negative biopsy, indications for repeated biopsies include a recommendation
rising PSA, suspicious DRE, abnormal multi-parametric magnetic • Localised disease should be classified as low-, intermediate- or
resonance imaging (MRI), atypical small acinar proliferation or high-risk (see Table 1) as a guide to prognosis and therapy
multifocal high-grade prostatic intraepithelial neoplasia. [III, A].
CH-6962 Viganello-Lugano, Switzerland. Clinical T stage (Table 2) should be evaluated by DRE. MRI pro-
E-mail: clinicalguidelines@esmo.org vides more accurate T staging and can inform surgical technique,
†
both with respect to nerve sparing and wide excision of areas of
Approved by the ESMO Guidelines Committee: February 2002, last update May 2015.
This publication supersedes the previously published version—Ann Oncol 2013; 24 potential extra-prostatic extension [9]. Within the low-risk cat-
(suppl. 6): vi106–vi114. egory, higher % positive cores, length of core involvement, PSA

by guest
on 05 February 2018
Table 1. Risk groups for localised prostate cancer [7] positron emission tomography/CT (PET/CT) or pelvic nodal
dissection [III, B].
Low risk T1–T2a and GS ≤6 and PSA ≤10
• Patients with intermediate- or high-risk disease should be
Intermediate risk T2b and/or GS7 and/or PSA10-20
staged for metastases using technetium bone scan and
High risk ≥T2c or GS8-10 or PSA >20
thoraco-abdominal CT scan or whole-body MRI or choline
PET/CT [III, B].
GS, Gleason score; PSA, prostate-specific antigen.
management of local/locoregional
Table 2. Prostate cancer staging summary (7th edition of the AJCC/ disease
UICC Cancer Staging Handbook) [8] (Used with the permission of There is no consensus regarding optimum management of
the American Joint Committee on Cancer (AJCC), Chicago, Illinois. localised disease (Table 3). Patients should be informed of the
The original source for this material is the AJCC Cancer Staging potential benefits and harms of the different options. Given the
Handbook, Seventh Edition (2010) published by Springer Science range of treatment options and their side-effects, men should be
and Business Media LLC, www.springer.com)
offered the opportunity to consult with both an urologist and a
Primary tumor (T) radiation oncologist. Men should be counselled that treatment
of prostate cancer may cause sexual dysfunction, infertility,
TX Primary tumor cannot be assessed
bowel and urinary problems.
T0 No evidence of primary tumor
Watchful waiting with delayed hormone therapy for symptom-
T1 Clinically inapparent tumor neither palpable nor visible by
imaging
atic progression is an option for men who are not suitable for, or
T1a Tumor incidental histologic finding in 5% or less of tissue
unwilling to have, treatment with curative intent. Curative treat-
resected ment options include radical prostatectomy (RP), external beam
T1b Tumor incidental histologic finding in more than 5% of radiotherapy and brachytherapy. Active surveillance is a strategy
tissue resected of close monitoring, typically using serum PSA, repeat prostate
T1c Tumor identified by needle biopsy (e.g., because of elevated biopsies and/or MRI, keeping curative treatment in reserve for
PSA) those with early evidence of disease progression [10].
Two randomised, controlled trials have compared RP and
T2 Tumor confined within prostate
watchful waiting [11, 12]. The Scandinavian Prostate Cancer
T2a Tumor involves one-half of one lobe or less
Group Study 4 accrued 695 men in Scandinavia during the early
T2b Tumor involves more than one-half of one lobe but not both
lobes
1990s, at a time when PSA testing was not routinely carried out,
T2c Tumor involves both lobes
and the results may not be applicable to screen-detected cancers.
With up to 23 years follow-up, 200 men in the surgery group and
T3 Tumor extends through the prostate capsule 247 in the watchful waiting group had died. The actuarial risk of
T3a Extracapsular extension (unilateral or bilateral) death from prostate cancer at 18 years was 18% for surgery com-
T3b Tumor invade seminal vesicle(s) pared with 29% for watchful waiting (P = 0.001). RP increased the
T4 Tumor is fixed or invades adjacent structures other than rate of erectile dysfunction (80% versus 45%), and urinary leakage
seminal vesicles, such as external sphincter, rectum, (49% versus 21%) [13], but these side-effect rates may not be gen-
bladder, levator muscles, and/or pelvic wall eralisable to high-volume surgical centres. The PIVOT trial
Regional lymph nodes (N) recruited 731 North American men between 1994 and 2002 [12].
They were more representative of men with PSA-detected cancer,
NX Regional lymph nodes were not assessed
but these men had a higher than expected rate of co-morbidity.
No significant difference was seen in overall survival [hazard ratio
N1 Metastasis in regional node(s)
(HR) 0.88, 95% confidence interval (CI) 0.71–1.08]. In the low-
Distant Metastasis (M) risk subgroup of 296 men, the risk of death from prostate cancer
M0 No distant metastasis was <3% at 12 years, with no significant benefit for surgery.
M1 Distant metastasis Indeed, the trend both in terms of prostate cancer-specific mortal-
M1a Non-regional lymph node(s) ity (HR 1.48; 95% CI 0.42–5.24) and overall mortality (HR 1.15;
M1b Bone(s) 95% CI 0.80–1.66) favoured watchful waiting rather than surgery.
M1c Other site(s) with or without bone disease The case for adding radical local treatment for men with
high-risk localised and locally advanced disease is based on two
PSA, prostate-specific antigen. randomised, controlled trials. The Scandinavian Prostate
Cancer Group Study 7 (SPCG-7) trial included 875 men who
received 3 months of combined androgen blockade followed by
density and a lower f/t PSA ratio are associated with the risk of flutamide monotherapy. They were randomised by whether or
under-staging. not they were to receive radical radiotherapy (RT) to the prostate
[14]. It showed a beneficial impact of radical RT in terms of
recommendations cause-specific (11.9% versus 23.9%, P < 0.001) and overall mor-
• Patients with intermediate- or high-risk disease should have tality (29.6% versus 39.4%, P = 0.004). The NCIC/MRC trial
nodal staging using computed tomography (CT), MRI, choline randomised high-risk patients to either lifelong androgen
v | Parker et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Table 3. Stage-matched therapeutic strategies neoadjuvant and adjuvant hormone
Localised disease treatment
Low risk Active surveillance The value of neoadjuvant ADT, at least in men with high-risk
Brachytherapy localised and locally advanced disease, has been established by
Radical prostatectomy multiple randomised trials. For example, in the Trans-Tasman
Radical radiotherapy Radiation Oncology Group (TROG) 96-01 trial, 818 men with
Intermediate risk Active surveillance
locally advanced prostate cancer were randomly assigned to RT
Brachytherapy alone, RT plus 3 months neoadjuvant and concurrent combined
Radical prostatectomy androgen blockade (CAB) or RT plus 6 months CAB [16].
Radical radiotherapy ± neoadjuvant ADT Compared with RT alone, the use of 6 months hormone therapy
significantly improved overall mortality [HR 0.63 (0.48–0.83)].
High risk Neoadjuvant ADT + radical
Similarly, the Radiation Therapy Oncology Group (RTOG) trial
radiotherapy + adjuvant ADT
8610, in 456 men with T2–4 disease, found an improvement in
Radical prostatectomy + pelvic
10-year prostate cancer-specific mortality (23% versus 36%;
lymphadenectomy
P = 0.01) with the addition of 4 months neoadjuvant and con-
Locally advanced disease current ADT to RT [17].
Neoadjuvant ADT + radical Adjuvant ADT, after RT, has also been studied in multiple
radiotherapy + adjuvant ADT phase III trials. The RTOG 92-02 trial randomised 1554 patients
Radical prostatectomy + pelvic between 4 and 28 months of ADT in addition to RT [18]. In an
lymphadenectomy unplanned subgroup analysis, the addition of adjuvant ADT
improved overall survival in those with a Gleason score of 8–10
Metastatic disease
(81.0% versus 70.7%, P = 0.044). The European Organisation for
Hormone-naïve ADT Research and Treatment of Cancer (EORTC) 22961 trial rando-
mised 970 men with locally advanced disease between 6 and
Castration-resistant Abiraterone
36 months of ADT in addition to radical RT [19]. The 5-year
(first line) Docetaxel
overall mortality for short-term and long-term suppression was
Enzalutamide
Radium-223
19.0% and 15.2%, respectively (HR 1.42; 95% CI 1.09–1.85).
Sipuleucel-T
Second line (post- Abiraterone recommendations
docetaxel) Cabazitaxel • Neoadjuvant and concurrent ADT for 4–6 months are recom-
Enzalutamide mended for men receiving radical RT for high-risk disease,
Radium-223 and should be considered for men with intermediate-risk
disease [I, A].
Options listed in alphabetical order. • Adjuvant ADT, for 2–3 years, is recommended for men re-
ADT, androgen deprivation therapy.
ceiving neoadjuvant hormonal therapy and radical RT, who
are at high risk of prostate cancer mortality [I, A].
deprivation therapy (ADT) alone or to ADT plus RT. The add- post-operative radiotherapy
ition of RT improved the 7-year survival probability from 66% Three randomised trials, EORTC 22911, Southwest Oncology
to 74% (P = 0.003) [15]. Group (SWOG) 8794 and ARO 96-02, have compared post-op-
erative radiotherapy versus observation after RP in patients with
recommendations locally advanced disease [20–22]. Each trial has shown an ad-
• Watchful waiting with delayed hormone therapy is an option vantage to post-operative radiotherapy in terms of PSA failure,
for men with low-risk disease. but the impact on overall survival is less clear. SWOG 8794
• Watchful waiting with delayed hormone therapy is an option reported after 198 deaths that overall survival was improved
for men with localised or locally advanced disease who are not with adjuvant RT (HR 0.72; 95% CI 0.5–0.96; P = 0.023).
suitable for, or unwilling to have, radical treatment [I, A]. However, EORTC 22911, based on 245 events found no overall
• Active surveillance is an option for men with low-risk disease survival benefit (10-year overall survival 76.9% for adjuvant
[II, A]. radiation versus 80.7% for observation).
• RP or radiotherapy (external beam or brachytherapy) are options RT to the prostate bed has a risk of adverse effects on urinary,
for men with low- or intermediate-risk disease [I, B]. bowel and sexual function. For example, the SWOG 8794 trial
• Primary ADT alone is not recommended as standard initial [21] reported urethral strictures in 17.8% of men randomised
treatment of non-metastatic disease [III, B]. to adjuvant RT versus 9.5% in those randomised to observation
• Options for patients with high-risk or locally advanced pros- [response rate (RR) 1.9; 95% CI 1.1–3.1; P = 0.02]. Total urinary in-
tate cancer include external beam RT plus hormone treatment continence was seen in 6.5% versus 2.8% (RR 2.3; 95% CI 0.9–5.9;
[I, B] or RP plus pelvic lymphadenectomy [III, B]. P = 0.11), and rectal complications in 3.3% versus 0% (P = 0.02).
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv222 | v

by guest
on 05 February 2018
recommendation high-volume disease, and 0.63 (0.34–1.17) for those with low-
• Immediate post-operative radiotherapy after RP is not rou- volume disease. GETUG-15, based on 176 events, found a
tinely recommended. Patients with positive surgical margins similar progression-free survival but no survival difference (HR
or extra-capsular extension after RP, with undetectable serum 1.01; 95% CI 0.76–1.25) [29]. STAMPEDE was a larger trial
PSA, should be informed about the pros and cons of adjuvant with over 2000 patients and confirmed both progression-free
RT [I, A]. and overall survival benefit for adding docetaxel to ADT [30].
treatment of relapse after radical therapy recommendations

There are no randomised trials comparing salvage RT versus • Continuous ADT is recommended as first-line treatment of
observation in men with PSA failure after RP. A retrospective metastatic, hormone-naïve disease [I, A].
analysis of men with PSA failure after surgery compared the • Men starting ADT should be informed that regular exercise
long-term outcome of those managed by observation (n = 397), reduces fatigue and improves quality of life [31] [I, A].
with those managed by salvage RT (n = 160) [23]. Salvage radio- • ADT plus docetaxel is recommended as first-line treatment of
therapy was associated with a significant reduction in prostate metastatic, hormone-naïve disease in men fit enough for
cancer mortality (HR 0.32; 95% CI 0.19–0.54; P < 0.001). chemotherapy [1, A].
recommendation treatment of castrate-resistant

• Following RP, patients should have their serum PSA level prostate cancer
monitored. Salvage RT to the prostate bed is recommended in Corticosteroids decrease adrenal production of androgens and lead
the event of PSA failure. Salvage RT should start early (e.g. to favourable biochemical and clinical responses. Dexamethasone
PSA <0.5 ng/ml) [III, B]. appears to be more active than prednisolone [32]. The value of
corticosteroids and other hormonal manipulations, which do
ADT for relapse following RP or RT has been evaluated in retro-
not have a proven overall survival benefit, has not been estab-
spective series. The use of early ADT was associated with a delay
lished by randomised trials. The arguments for their use are
in time to progression but no impact on overall survival [24].
their favourable cost profile and, for some of these agents, their
Intermittent versus continuous ADT was studied in a rando-
low toxicity.
mised trial of 1386 patients with a PSA at relapse of >3.0 ng/ml
The COU-302 trial tested the use of abiraterone acetate plus
more than one year after RT. This study showed that intermit-
prednisone versus placebo plus prednisone, in over 1000 men with
tent therapy had a more favourable toxicity profile but no differ-
chemotherapy-naïve, asymptomatic or mildly symptomatic meta-
ence in overall survival (HR 1.02; 95% CI 0.86–1.21) [25].
static castrate-resistant prostate cancer (CRPC) [33]. Abiraterone
improved overall survival (HR 0.79; 95% CI 0.66–0.95). The main
recommendations specific side-effects were hypokalaemia, hypertension, oedema and
• Early ADT is not routinely recommended for men with bio- cardiac events.
chemical relapse unless they have symptomatic local disease, or Enzalutamide was tested against placebo in the same setting
proven metastases, or a PSA doubling time <3 months [IV, B]. in the PREVAIL trial [34]. Enzalutamide improved overall sur-
• Intermittent ADT is recommended for men with biochemical vival (HR 0.71; 95% CI 0.60–0.84). The main side-effects of
relapse after radical RT starting ADT [I, B]. enzalutamide were fatigue/asthenia and hypertension.
Sipuleucel-T, an immunotherapy using activated autologous
management of advanced/metastatic dendritic cells, was tested against ‘placebo’ [leucopheresis was
done three times, as in the active arm, but with reinfusion of
disease one-third of the (unactivated) antigen presenting cells] in a trial
SWOG 9346 randomised over 1500 patients, with metastatic of 512 patients [35]. Overall survival favoured Sipuleucel-T (HR
disease who achieved a PSA value <4 ng/ml on ADT, between 0.78; 95% CI 0.61–0.98), which was well tolerated. The lack of
intermittent and continuous ADT. The overall survival results any impact on disease response or progression, taken together
failed to demonstrate that intermittent treatment was non-inferior with logistic considerations and cost, has limited its use.
to continuous ADT (HR 1.10; 90% CI 0.99–1.23) [26]. Radium-223, a bone-targeted alpha emitter, was tested
Multiple phase III trials have studied the addition of an an- against placebo in over 900 men with bone-predominant, symp-
drogen receptor (AR) antagonist to ADT alone for initial treatment tomatic CRPC [36]. Radium-223 improved overall survival (HR
of metastatic disease. Meta-analysis of trials testing a non-steroidal 0.70; 95% CI 0.58–0.83) and time to first symptomatic skeletal
AR antagonist found an overall survival advantage (27.6% versus event (HR 0.66; 95% CI 0.52–0.83). Side-effects of radium-223
24.7%, P = 0.005) [27]. This modest benefit is typically consid- include myelosuppression, particularly thrombocytopenia, and
ered insufficient to justify combination treatment. diarrhoea.
Three phase III trials have compared ADT alone versus ADT Docetaxel has been shown in two phase III trials to improve
plus docetaxel in men with metastatic, hormone-naïve disease. overall survival in men with CRPC. TAX-327 compared doce-
Based on 237 events, the CHAARTED trial found that docetaxel taxel against mitoxantrone in over 1000 men [37]. Overall
improved overall survival (HR 0.61, 95% CI 0.47–0.80) [28]. survival favoured docetaxel (HR 0.76; 95% CI 0.62–0.94). The
The effect size was consistent across all subgroups. For example, side-effects of docetaxel included myelosuppression, fatigue, alo-
the HR for overall survival was 0.63 (0.45–0.81) for men with pecia, diarrhoea, neuropathy and peripheral oedema.
v | Parker et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
recommendations recommendation
• Abiraterone or enzalutamide are recommended for asymp- • In patients with bone metastases from CRPC, at high risk for
tomatic/mildly symptomatic men with chemotherapy-naïve clinically significant SREs, denosumab or zoledronate can be
metastatic CRPC [I, A]. recommended [I, B].
• Radium-223 is recommended for men with bone-predominant,
symptomatic metastatic CRPC without visceral metastases [I, A]. Spinal cord compression is a devastating complication of meta-
• Docetaxel is recommended for men with metastatic CRPC [I, A]. static prostate cancer and early detection is critical for successful
• Sipuleucel-T is an option in asymptomatic/mildly symptomatic management. A systematic review found that spinal cord com-
patients with chemotherapy-naïve metastatic CRPC [II, B]. pression is a common finding, even in asymptomatic patients
with metastatic prostate cancer and spinal metastases [44].
The optimal sequence or combination of these agents (abirater-
one, enzalutamide, radium-223, docetaxel and Sipuleucel-T) is recommendations
unknown. In practice, sequencing decisions will be made in the • MRI of the spine to detect subclinical cord compression is recom-
light of the distribution, extent and pace of disease, co-morbid- mended in men with CRPC with vertebral metastases [III, B].
ities, patient preferences and drug availability. • Urgent MRI of the spine to detect cord compression is very
TROPIC tested cabazitaxel against mitoxantrone in 755 strongly recommended in men with CRPC with vertebral me-
patients in the post-docetaxel setting [38]. Cabazitaxel improved tastases and neurological symptoms [III, A].
overall survival (HR 0.70; 95% CI 0.59–0.83), but was associated
with increased myelosuppression, including febrile neutropae- Beta-emitting, bone seeking radionuclides such as Sr-89 and
nia and diarrhoea. Sm-153-HEDP have proven symptomatic benefits in the treat-
Abiraterone plus prednisone was tested against placebo plus ment of metastatic CRPC. However, their use is limited by mye-
prednisone in the post-docetaxel setting in the COU-301 trial lotoxicity and they have largely been superseded by radium-223,
[39]. Abiraterone improved overall survival (HR 0.74; 95% CI where this is available.
0.64–0.86).
Enzalutamide was tested against placebo in the post-docetaxel personalised medicine
setting in the AFFIRM trial, and also improved overall survival
(HR 0.63; 95% CI 0.53–0.75) [40]. Although there are known prognostic factors to guide man-
agement, there are no established predictive biomarkers to
choose one particular treatment over another. Advanced
recommendation disease progressing without a significant rise in PSA should
• In patients with metastatic CRPC in the post-docetaxel setting, be investigated for neuro-endocrine markers, using biopsy or
abiraterone, enzalutamide, cabazitaxel and radium-223 (in blood analyses for neuron-specific enolase and/or chromogra-
those without visceral disease) are recommended options [I, A]. nin A [45].
palliative care recommendation

• Patients with evidence of neuro-endocrine change in their
Fractionated versus single-fraction RT for bone pain has been prostate cancer should receive chemotherapy in addition to
compared in multiple randomised trials. Single-fraction treat- ADT [IV, B].
ment provides similar pain relief [41].
recommendation follow-up and long-term implications

• A single fraction of external beam RT is recommended for DRE has been shown to be unnecessary in men whose disease
palliation of painful bone metastasis [I, A]. remains biochemically controlled after radical treatment [46].
Zoledronic acid, a bisphosphonate, was shown to prolong time

to first skeletal-related event (SRE), namely fracture, spinal cord recommendations
compression, surgery or RT for bone pain or a change in antic- • Routine DRE after local therapy is not required for asymp-
ancer treatment [42]. However, there were no differences in tomatic patients while the PSA remains controlled [II, B].
disease progression, overall survival or quality of life. Adverse • Biopsy of the prostate after RT should only be carried out in
effects included anaemia, fever, myalgia and osteonecrosis of the men with prostate cancer who are being considered for
jaw (ONJ). Denosumab, a RANK ligand inhibitor has been salvage local therapy [V, C].
compared with zoledronate [43]. Denosumab was superior with
Men developing bowel symptoms after prostate radiotherapy
respect to time to first SRE (HR 0.82; 95% CI 0.71–0.95,
may have inflammatory bowel disease, a primary colorectal
P = 0.0002), but was associated with an increased risk of hypo-
malignancy or a treatable radiation enteropathy [47].
calcaemia (13% versus 6%) and a trend towards higher inci-
dence of ONJ (2.3% versus 1.3%). There was no difference in
overall survival. Abiraterone, enzalutamide and radium-223 all recommendation
reduce the risk of SREs. The added value of zoledronate or • Chronic bowel symptoms after RT should be investigated by a
denosumab for SRE prevention is unclear. gastroenterologist [V, B].
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv222 | v

by guest
on 05 February 2018
Screening and early detection
• Population-based PSA screening for prostate cancer reduces prostate cancer mortality at the expense of over diagnosis and overtreatment and is not
recommended [I, C].
• Testing for prostate cancer in asymptomatic men should not be done in men over the age of 70 years [I, B].
Diagnosis and pathology
• A single elevated PSA level should not prompt a prostate biopsy, and should be verified by a second value [IV, B].
• The decision whether or not to have a prostate biopsy should be made in the light of DRE findings, ethnicity, age, co-morbidities, PSA values,
free/total (f/t) PSA, history of previous biopsy and patient values [II, B].
• Transrectal ultrasound-guided prostate biopsy should be carried out under antibiotic cover and local anaesthesia, and a minimum of 10–12 cores
obtained [II, B].
• Before repeat biopsy, multi-parametric MRI is recommended with a view to MRI-guided or MRI-transrectal ultrasound (TRUS) fusion biopsy [5] [III, B].
• The extent of involvement of each biopsy core, and the commonest and the worst Gleason grades should be reported [II, A].
•Localised disease should be classified as low, intermediate or high risk (see Table 1) as a guide to prognosis and therapy [III, A].
•Patients with intermediate- or high-risk disease should have nodal staging using computed tomography (CT), MRI, choline positron emission
tomography/CT (PET/CT) or pelvic nodal dissection [III, B].
• Patients with intermediate- or high-risk disease should be staged for metastases using technetium bone scan and thoraco-abdominal CT scan or
whole-body MRI or choline PET/CT [III, B].
• Watchful waiting with delayed hormone therapy is an option for men with low-risk disease.
• Watchful waiting with delayed hormone therapy is an option for men with localised or locally advanced disease who are not suitable for, or unwilling
to have, radical treatment [I, A].
• Active surveillance is an option for men with low-risk disease [II, A].
• Radical prostatectomy or radiotherapy (external beam or brachytherapy) are options for men with low- or intermediate-risk disease [I, B].
• Primary ADT alone is not recommended as standard initial treatment of non-metastatic disease [III, B].
• Options for patients with high-risk or locally advanced prostate cancer include external beam RT plus hormone treatment [I, B] or radical
prostatectomy plus pelvic lymphadenectomy [III, B].
Neoadjuvant and adjuvant hormone treatment
• Neoadjuvant and concurrent ADT for 4–6 months are recommended for men receiving radical RT for high-risk disease, and should be considered for
men with intermediate-risk disease [I, A].
• Adjuvant ADT, for 2–3 years, is recommended for men receiving neoadjuvant hormonal therapy and radical RT, who are at high risk of prostate
cancer mortality [I, A].
Post-operative radiotherapy
• Immediate post-operative radiotherapy after RP is not routinely recommended. Patients with positive surgical margins or extra-capsular extension
after RP, with undetectable serum PSA, should be informed about the pros and cons of adjuvant RT [I, A].
Treatment of relapse after radical therapy
•Following RP, patients should have their serum PSA level monitored. Salvage RT to the prostate bed is recommended in the event of PSA failure.
Salvage RT should start early (e.g. PSA <0.5 ng/ml) [III, B].
• Early ADT is not routinely recommended for men with biochemical relapse unless they have symptomatic local disease, or proven metastases, or a
PSA doubling time <3 months [IV, B].
• Intermittent ADT is recommended for men with biochemical relapse after radical RT starting ADT [I, B].
• Continuous ADT is recommended as first-line treatment of metastatic, hormone-naïve disease [I, A].
• Men starting ADT should be informed that regular exercise reduces fatigue and improves quality of life [30] [I, A].
• ADT plus docetaxel is recommended as first-line treatment of metastatic, hormone-naïve disease in men fit enough for chemotherapy [1, A].
Treatment of castrate-resistant prostate cancer (CRPC)
• Abiraterone or enzalutamide are recommended for asymptomatic/mildly symptomatic men with chemotherapy-naïve metastatic CRPC [I, A].
• Radium-223 is recommended for men with bone-predominant, symptomatic metastatic CRPC without visceral metastases [I, A].
• Docetaxel is recommended for men with metastatic CRPC [I, A].
• Sipuleucel-T is an option in asymptomatic/mildly symptomatic patients with chemotherapy-naïve metastatic CRPC [II, B].
• In patients with metastatic CRPC in the post-docetaxel setting, abiraterone, enzalutamide, cabazitaxel and radium-223 (in those without visceral
disease) are recommended options [I, A].
Continued
v | Parker et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Table 4. Continued
Palliative care
• A single fraction of external beam RT is recommended for palliation of painful bone metastasis [I, A].
• In patients with bone metastases from CRPC at high risk for clinically significant SREs, denosumab or zoledronate can be recommended [I, B].
• MRI of the spine to detect subclinical cord compression is recommended in men with CRPC with vertebral metastases [III, B].
• Urgent MRI of the spine to detect cord compression is very strongly recommended in men with CRPC with vertebral metastases and neurological
symptoms [III, A].
• Patients with evidence of neuro-endocrine change in their prostate cancer should receive chemotherapy in addition to ADT [IV, B].
Follow-up and long-term implications
• Routine DRE after local therapy is not required for asymptomatic patients while the PSA remains controlled [II, B].
• Biopsy of the prostate after RT should only be carried out in men with prostate cancer who are being considered for salvage local therapy [V, C].
• Chronic bowel symptoms after RT should be investigated by a gastroenterologist [V, B].
• Men on long-term ADT should be monitored for side-effects including osteoporosis (using bone densitometry) and metabolic syndrome [IV, B].
• In patients with CRPC on systemic treatment, regular imaging studies should be done to monitor disease response/progression [V, B].
PSA, prostate specific antigen; DRE, digital rectal examination; MRI, magnetic resonance imaging; ADT, androgen deprivation therapy; RT, radiotherapy;
RP, radical prostatectomy; SRE, skeletal related events.
Table 5. Levels of evidence and grades of recommendation (adapted from the Infectious Diseases Society of America-United States Public Health
Levels of evidence
I Evidence from at least one large randomised, controlled trial of good methodological quality (low potential for bias) or meta-analyses of
well-conducted randomised trials without heterogeneity
II Small randomised trials or large randomised trials with a suspicion of bias (lower methodological quality) or meta-analyses of such trials or of
trials with demonstrated heterogeneity
C Insufficient evidence for efficacy or benefit does not outweigh the risk or the disadvantages (adverse events, costs, …), optional
E Strong evidence against efficacy or for the adverse outcome, never recommended
a
ADT may cause hot flushes, lethargy, mood changes, osteopor- recommendation
osis, insulin resistance and muscle weakness. • In patients with CRPC on systemic treatment, regular imaging
studies should be done to monitor disease response/progres-
recommendation sion [V, B].
• Men on long-term ADT should be monitored for side-effects
including osteoporosis (using bone densitometry) and meta-
bolic syndrome [IV, B]. methodology
It is not adequate to rely on PSA alone to monitor response in These clinical practice guidelines were developed in accordance
men with CRPC. Rather, appropriate imaging tests should be with the ESMO standard operating procedures for clinical prac-
used. However, conventional imaging techniques such as tice guidelines development. A summary of recommendations is
CT and bone scan do not provide assessment of response/ shown in Table 4. The relevant literature has been selected by the
progression in bony metastatic disease. expert authors. Levels of evidence and grades of recommendation
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv222 | v

by guest
on 05 February 2018
have been applied using the system shown in Table 5. Statements 14. Widmark A, Klepp O, Solberg A et al. Endocrine treatment, with or without
without grading were considered justified standard clinical prac- radiotherapy, in locally advanced prostate cancer (SPCG-7/SFUO-3): an open
randomised phase III trial. Lancet 2009; 373: 301–308.
tice by the experts and the ESMO faculty. This manuscript has
15. Warde P, Mason M, Ding K et al. Combined androgen deprivation therapy and
been subjected to an anonymous peer review process.
radiation therapy for locally advanced prostate cancer: a randomised, phase 3 trial.
Lancet 2011; 378: 2104–2111.
acknowledgements 16. Denham JW, Steigler A, Lamb DS et al. Short-term neoadjuvant androgen
deprivation and radiotherapy for locally advanced prostate cancer: 10-year data
This project was supported by the National Institute for Health from the TROG 96.01 randomised controlled trial. Lancet Oncol 2011; 12:
Research Royal Marsden and Institute for Cancer Research 451–459.
Biomedical Research Centre. 17. Roach M, III, Bae K, Speight J et al. Short-term neoadjuvant androgen deprivation
therapy and external-beam radiotherapy for locally advanced prostate cancer:
long-term results of RTOG 8610. JCO 2008; 26: 585–591.
conflict of interest 18. Hanks GE, Pajak TF, Porter A et al. Phase III trial of long-term adjuvant androgen
CP has reported honoraria from Bayer, Janssen and Takeda; deprivation after neoadjuvant hormonal cytoreduction and radiotherapy in locally
advanced carcinoma of the prostate: the Radiation Therapy Oncology Group
consultancy fees from BNIT; and research grants from Bayer. A
Protocol 92-02. J Clin Oncol 2003; 21: 3972–3978.
Heidenreich has reported advisory boards of Astellas, Bayer,
19. Bolla M, de Reijke TM, Van Tienhoven G et al. Duration of androgen suppression in
Dendreon, Ferring, IPSEN, Jansen and Sanofi; research grants the treatment of prostate cancer. N Engl J Med 2009; 360: 2516–2527.
from Astellas and Sanofi; honoraria from Amgen, Astellas, Bayer, 20. Bolla M, van Poppel H, Tombal B et al. Postoperative radiotherapy after radical
Dendreon, Ferring, Ipsen, Jansen, Pfizer, Sanofi and Takeda. SG prostatectomy for high-risk prostate cancer: long-term results of a randomised
has reported compensated advisory boards of Bayer, Curevac, controlled trial (EORTC trial 22911). Lancet 2012; 380: 2018–2027.
Janssen Cilag, Janssen Diagnostics, Millenium, Astellas, Novartis, 21. Thompson IM, Tangen CM, Paradelo J et al. Adjuvant radiotherapy for pathological
Sanofi, Pfizer, Dendreon and Orion; uncompensated speakers’ T3N0M0 prostate cancer significantly reduces risk of metastases and improves
bureau for Amgen, Bayer, Janssen Cilag and Sanofi-Aventis; un- survival: long-term follow up of a randomized clinical trial. J Urol 2009; 181:
956–962.
compensated scientific advisor for ProteoMediX. A Horwich has
22. Wiegel T, Bottke D, Steiner U et al. Phase III postoperative adjuvant radiotherapy
reported no potential conflicts of interest. after radical prostatectomy compared with radical prostatectomy alone in pT3
prostate cancer with postoperative undetectable prostate-specific antigen: ARO
references 96-02/AUO AP 09/95. J Clin Oncol 2009; 27: 2924–2930.
23. Trock BJ, Han M, Freedland SJ et al. Prostate cancer-specific survival following
1. Schröder FH, Hugosson J, Roobol MJ et al. Screening and prostate cancer salvage radiotherapy vs observation in men with biochemical recurrence after
mortality: results of the European Randomised Study of screening for Prostate radical prostatectomy. JAMA 2008; 299: 2760–2769.
Cancer (ERSPC) at 13 years of follow-up. Lancet 2014; 384: 2027–2035. 24. Moul JW, Wu H, Sun L et al. Early versus delayed hormonal therapy for prostate
2. Andriole GL, Crawford ED, Grubb RL, 3rd et al. Prostate cancer screening in the antigen only recurrence of prostate cancer after radical prostatectomy. J Urol
randomized Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial: mortality 2004; 171: 1141–1147.
results after 13 years of follow-up. J Natl Cancer Inst 2012; 104: 125–132. 25. Crook JM, O’Callaghan CJ, Duncan G et al. Intermittent androgen suppression for
3. Lilja H, Cronin AM, Dahlin A et al. Prediction of significant prostate cancer rising PSA level after radiotherapy. N Engl J Med 2012; 367: 895–903.
diagnosed 20 to 30 years later with a single measure of prostate-specific antigen 26. Hussain M, Tangen CM, Higano CS et al. Intermittent (IAD) versus continuous
at or before age 50. Cancer 2011; 117: 1210–1219. androgen deprivation (CAD) in hormone sensitive metastatic prostate cancer
4. Thompson IM, Ankerst DP, Chi C et al. Assessing prostate cancer risk: results (HSM1PC) patients ( pts): results of S9346 (INT-0162), an international phase III
from the Prostate Cancer Prevention Trial. J Natl Cancer Inst 2006; 98: 529–534. trial. N Engl J Med 2013; 368: 1314–1325.
5. Sonn GA, Chang E, Natarajan S et al. Value of targeted prostate biopsy using 27. Prostate Cancer Trialists’ Collaborative Group. Maximum androgen blockade in
magnetic resonance-ultrasound fusion in men with prior negative biopsy and advanced prostate cancer: an overview of the randomised trials. Lancet 2000;
elevated prostate-specific antigen. Eur Urol 2014; 65: 809–815. 355: 1491–1498.
6. Epstein JI, Allsbrook WC, Jr, Amin MB et al. The 2005 International Society of 28. Sweeney C, Chen Y-H, Carducci M et al. Impact on overall survival (OS) with
Urological Pathology (ISUP) Consensus Conference on Gleason Grading of Prostatic chemohormonal therapy versus hormonal therapy for hormone-sensitive newly
Carcinoma. Am J Surg Pathol 2005; 29: 1228–1242. metastatic prostate cancer (mPrCa): an ECOG-led phase III randomized trial. J Clin
7. D’Amico AV, Whittington R, Malkovicz SB et al. Biochemical outcome after radical Oncol (ASCO Meeting Abstracts) 2014; 32 (18 suppl): LBA2.
prostatectomy, external beam radiation therapy, or interstitial radiation therapy for 29. Gravis G, Fizazi K, Joly F et al. Androgen-deprivation therapy alone or with
clinically localized prostate cancer. JAMA 1998; 280: 969–974. docetaxel in non-castrate metastatic prostate cancer (GETUG-AFU 15): a
8. Edge SB, Byrd DR, Compton CC (eds). AJCC Cancer Staging Handbook, 7th randomised, open-label, phase 3 trial. Lancet Oncol 2013; 14: 149–158.
edition. New York, NY: Springer 2010. 30. James ND, Sydes MR, Mason MD et al. Docetaxel and/or zoledronic acid for
9. Heidenreich A, Bastian PJ, Bellmunt J et al. EAU guidelines on prostate cancer. hormone-naïve prostate cancer: first overall survival results from STAMPEDE
Part 1: screening, diagnosis, and local treatment with curative intent-update. Eur (NCT00268476). J Clin Oncol 2015; 33 (suppl; abstract 5001)
Urol 2014; 65: 124–137. 31. Galvao DA, Taaffe DR, Spry N et al. Combined resistance and aerobic exercise
10. Suardi N, Capitanio U, Chun FK et al. Currently used criteria for active surveillance program reverses muscle loss in men undergoing androgen suppression therapy
in men with low-risk prostate cancer: an analysis of pathologic features. Cancer for prostate cancer without bone metastases: a randomized controlled trial. J Clin
2008; 113: 2068–2072. Oncol 2010; 28: 340–347.
11. Bill-Axelson A, Holmberg L, Garmo H et al. Radical prostatectomy or watchful 32. Venkitaraman R, Lorente D, Murthy V et al. A randomised phase 2 trial of
waiting in early prostate cancer. N Engl J Med 2014; 370: 932–942. dexamethasone versus prednisolone in castration-resistant prostate cancer. Eur
12. Wilt TJ, Brawer MK, Jones KM et al. Radical prostatectomy versus observation for Urol 2015; 67: 673–679.
localized prostate cancer. N Engl J Med 2012; 367: 203–213. 33. Ryan CJ, Smith MR, Fizazi K et al. Abiraterone acetate plus prednisone versus
13. Steineck G, Helgesen F, Adolfsson J et al. Quality of life after radical prostatectomy placebo plus prednisone in chemotherapy-naive men with metastatic castration-
or watchful waiting. N Engl J Med 2002; 347: 790–796. resistant prostate cancer (COU-AA-302): final overall survival analysis of a
v | Parker et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
randomised, double-blind, placebo-controlled phase 3 study. Lancet Oncol 2015; 42. Saad F, Gleason DM, Murray R et al. Long-term efficacy of zoledronic acid for the
16: 152–160. prevention of skeletal complications in patients with metastatic hormone-refractory
34. Beer TM, Armstrong AJ, Rathkopf DE et al. Enzalutamide in metastatic prostate prostate cancer. J Natl Cancer Inst 2004; 96: 879–882.
cancer before chemotherapy. N Engl J Med 2014; 371: 424–433. 43. Fizazi K, Carducci M, Smith M et al. Denosumab versus zoledronic acid for
35. Kantoff PW, Higano CS, Shore ND et al. Sipuleucel-T immunotherapy for treatment of bone metastases in men with castration-resistant prostate cancer: a
castration-resistant prostate cancer. N Engl J Med 2010; 363: 411–422. randomised, double-blind study. Lancet 2011; 377: 813–822.
36. Parker C, Nilsson S, Heinrich D et al. Alpha emitter radium-223 and survival in 44. Loblaw DA, Perry J, Chambers A, Laperriere NJ. Systematic review of the
metastatic prostate cancer. N Engl J Med 2013; 369: 213–223. diagnosis and management of malignant extradural spinal cord compression: the
37. Tannock IF, de Witt R, Berry WR et al. Docetaxel plus prednisone or mitoxantrone Cancer Care Ontario Practice Guidelines Initiative’s Neuro-Oncology Disease Site
plus prednisone for advanced prostate cancer. N Engl J Med 2004; 351: Group. J Clin Oncol 2003; 23: 2028–2037.
1502–1512. 45. Beltran H, Tagawa ST, Park K et al. Challenges in recognizing treatment-related
38. de Bono JS, Oudard S, Ozguroglu M et al. Prednisone plus cabazitaxel or neuroendocrine prostate cancer. J Clin Oncol 2012; 30: e386–e389.
mitoxantrone for metastatic castration-resistant prostate cancer progressing after 46. Doneux A, Parker CC, Norman A et al. The utility of digital rectal examination after
docetaxel treatment: a randomised open-label trial. Lancet 2010; 376: 1147–1154. radical radiotherapy for prostate cancer. Clin Oncol (R Coll Radiol) 2005; 17:
39. de Bono JS, Logothetis CJ, Molina A et al. Abiraterone and increased survival in 172–173.
metastatic prostate cancer. N Engl J Med 2011; 364: 1995–2005. 47. Andreyev J. Gastrointestinal symptoms after pelvic radiotherapy: a new understanding
40. Scher HI, Fizazi K, Saad F et al. Increased survival with enzalutamide in prostate to improve management of symptomatic patients. Lancet Oncol 2007; 8: 1007–1017.
cancer after chemotherapy. N Engl J Med 2012; 367: 1187–1197. 48. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
41. Chow E, Harris K, Fan G et al. Palliative radiotherapy trials for bone metastases: a among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
systematic review. J Clin Oncol 2007; 25: 1423–1436. 139–144.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv222 | v

by guest
on 05 February 2018
Adrenal cancer: ESMO Clinical Practice Guidelines for

A. Berruti1, E. Baudin2, H. Gelderblom3, H. R. Haak4, F. Porpiglia5, M. Fassnacht6 &
G. Pentheroudakis7 on behalf of the ESMO Guidelines Working Group*
1
Medical Oncology, Department of Clinical and Biological Sciences, University of Torino, Azienda Ospedaliero, Universitaria San Luigi, Orbassano, Italy; 2Service de
Médecine Nucléaire et de Cancerologie Endocrinienne, Institut Gustave-Roussy, Université Paris XI, Villejuif, France; 3Department of Clinical Oncology, Leiden University
Medical Center, Leiden, The Netherlands; 4Department of Internal Medicine, Máxima Medisch Centrum, Eindhoven, The Netherlands; 5Urology, Department of Clinical
and Biological Sciences, University of Torino, Azienda Ospedaliero, Universitaria San Luigi, Orbassano, Italy; 6Department of Internal Medicine I, Endocrine Unit,
University Hospital, University of Würzburg, Germany; 7Department of Medical Oncology, Medical School, University of Ioannina, Greece
incidence and epidemiology among different genetic backgrounds, it is below 10% for most
sporadic pheochromocytomas, except in patients with
Two different primary malignancies can arise from the adrenal mutations in the succinate dehydrogenase B (SDHB) gene and/
gland: the adrenocortical carcinoma (ACC) from the adrenal or extra-adrenal locations, among whom more than 30%–50%
cortex and the malignant pheochromocytoma from the adrenal may develop a malignant tumor.
medulla. Both malignancies are extremely rare. ACC has an
estimated incidence of ∼0.5–2 new cases per million people
per year. It follows a bimodal age distribution, with peaks in diagnosis and pathology/molecular
childhood and in the fourth to fifth decades of life. ACC is biology
more frequent in women than in men (ratio 1.5∶1). The
practice
Somatic mutations of the tumor suppressor gene TP53 are
clinical
majority of ACCs are sporadic; however, sometimes these
observed in one-third of ACCs. In addition, allelic losses
malignancies form part of hereditary syndromes such as the
(LOH) at the TP53 locus (17p13) are observed in >85% of
Li-Fraumeni syndrome, Beckwith-Wiedeman syndrome,
ACCs. The insulin-like growth factor II (IGF-II) locus (11p15)
multiple endocrine neoplasia (MEN) 1, congenital adrenal
is imprinted and IGF-II is overexpressed in 90% of ACCs.
hyperplasia, familial polyposis coli, and B-catenin mutations
About one-third of ACCs harbor somatic activating mutations
[1]. Germline p53 mutations without Li-Fraumeni are frequent
of the B-catenin gene [3].
in southern Brazilian children.
The gene encoding subunit B of the SDHB complex is by far
The frequency of sporadic ACC cases is consistently greater
the most important contributor to a hereditary malignant
in the population of patients with adrenal incidentaloma (2%
pheochromocytoma/paraganglioma. In addition, many
in the most conservative series) and, in our experience, the
sporadic metastatic pheochromocytomas/paragangliomas have
proportion of incidentally discovered ACC is increasing.
similar molecular profiles to those of hereditary tumors.
Pheochromocytomas are catecholamine-producing
Inactivation of mutations in SDHB is thought to reduce
neuroendocrine tumors arising from chromaffin cells of the
function of the SDHB complex, causing a pseudohypoxic state
adrenal medulla or extra-adrenal paraganglia. In the latter
and increased expression of angiogenic, growth and mitogenic
setting, the term paraganglioma is preferred. Their incidence is
factors via stabilization of the hypoxia-inducible pathway. If no
∼2–8 per million adults per year [2]. The incidence of
SDHB mutations are identified, genetic testing for VHL,
pheochromocytoma increases to 0.5% in patients with
SDHD, or succinate dehydrogenase C (SDHC) is indicated and
hypertensive symptoms and can be as high as 4% in patients
may provide future insights into the best therapeutic options to
with adrenal incidentalomas. Up to 30% of
be developed for these patients [2]. Rearranged during
pheochromocytomas are associated with a variety of inherited
transfection (RET) should be tested in case of calcitonine
conditions, including MEN2, Von Hippel–Lindau (VHL)
secretion and may be tested when all other gene mutations are
disease, neurofibromatosis type 1, and heredity paraganglioma
negative.
syndromes. Only a few (10%–17%) pheochromocytomas are
A detailed preoperative endocrine assessment is essential to
malignant. Although the likelihood of malignancy varies
establish the origin of the tumor (cortex versus medulla
versusothers). In all cases of an adrenal mass a comprehensive
hormonal analysis is strongly recommended. In 2005, the ACC
L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland. E-mail: clinicalguidelines@esmo.
working group of the European Network for the Study of
org Adrenal Tumors (ENSAT; www.ensat.org) proposed standards
†
for this situation in patients with suspected or established ACC
Approved by the ESMO Guidelines Working Group: June 2012.

by guest
on 05 February 2018
Table 1. Diagnostic work-up for adrenal cancer pheochromocytoma due to the risk of a hypertensive crisis
after infusion of i.v. contrast medium for CT. Although these
Hormonal work-up methods cannot determine the exact entity of the mass, both
Glucocorticoid excess (minimum 3 of 4 tests) are able to correctly diagnose benign tumors in most cases—
Dexamethasone suppression test (1 mg, 23:00 h) when performed according to the state-of-the-art criteria (e.g.
Excretion of free urinary cortisol (24 h urine) Hounsfield units <10 in unenhanced CT, rapid washout in 15-
Basal cortisol (serum) min delayed contrast-enhanced CT or signal intensity loss
Basal ACTH (plasma) using opposed-phase MRI strongly suggests a benign tumor).
Sexual steroids and steroid precursors Most ACCs and malignant pheochromocytomas are
DHEA-S (serum) inhomogeneous with irregular margins and irregular
17-OH-progesterone (serum) enhancement of solid components after the infusion of i.v.
Androstenedione (serum) contrast medium and, in many cases, it is difficult to
Testosterone (serum) distinguish these tumor entities using conventional imaging.
17-beta-estradiol (serum, only in men and postmenopausal women) The detection of local invasion or tumor extension into the
24-h urine steroid metabolite examination
inferior vena cava, as well as lymph node or other
Mineralocorticoid excess
metastases (lung and liver), is important for planning
Potassium (serum)
surgery. F-18/flourodeoxyglucose-positron emission
Aldosterone/renin ratio (only in patients with arterial hypertension
tomography (FDG-PET) is a useful tool for distinguishing
and/or hypokalemia)
potentially malignant lesions from benign tumors in
Catecholamine excess
Normetanephrine, metanephrine, and methoxytyramine (plasma)
radiologically indeterminate adrenal lesions. In patients with
Alternatively: fractionated metanephrine excretion (24 h urine) established pheochromocytoma, FDG-PET appears to be
Imaging superior to metaiodobenzylguanidine (MIBG) and
CT or MRI of abdomen and CT thorax somatostatin- or F-dopamin-based methods, particularly in
Bone scintigraphy (when suspecting skeletal metastases) patients with SDHB mutation and malignant tumors, and
FDG-PET (optional) should be indicated as the preferred method [6] (IV).
MIBG scintigraphy, DOTA-TATE-PET, Dopa/Dopamine PET or Fine needle biopsy of a suspected ACC is almost never
FDG-PET if pheochromocytoma is proved justified because of anticipated tumor spill, and in suspected
pheochromocytoma, it is contraindicated.
Adapted according to the recommendation of the ACC working group of The pathological differential diagnosis of adrenal neoplasias
the European Network for the Study of Adrenal Tumors (www.ensat.org/ is still largely based on morphological features requiring an
acc.htm), May 2005.
experienced pathologist. Several markers have been introduced
In patients with a clearly established diagnosis of an ACC, one can skip
to establish the adrenocortical origin of adrenal masses, with
the workup on catecholamine excess (and conversely for established
steroidogenesis factor-1 immunohistochemistry being
pheochromocytoma, one can skip the steroid analysis). DHEA,
particularly useful [7], although not yet widely available. The
dehydroepiandrosterone.
differential diagnosis between carcinoma and adenoma is
challenging as no single marker indicates malignancy. The
most widely used diagnostic score has been introduced by
(Table 1). It is important to acknowledge that the evidence Weiss et al. [8] and includes the following parameters: mitosis,
level for this proposition is formally low, although the atypical mitosis, necrosis, venous invasion, sinusal invasion,
diagnostic accuracy is high (V, B). Furthermore, the capsular invasion, nuclear atypia, diffuse architecture, and clear
preoperative hormone pattern may serve as a fingerprint of the cell. A score of ≥3 suggests malignancy. Ki67 as a marker of
tumor during follow-up. Whereas the measurement of proliferative activity can also be useful in this respect.
metanephrine and normetanephrine is well established for any For pheochromocytomas, the situation is similarly
pheochromocytoma, plasma and urine methoxytyramine levels demanding. Several histologic features (local invasiveness,
that provide useful information to assess the likelihood of growth pattern, presence of necrosis, cellularity, spindled
malignancy [4]. However, this marker is not yet widely morphology, nuclear pleomorphism and hyperchromasia,
available. mitotic activity, and atypical mitosis) comprise the
Urine steroid metabolomics revealed a pattern of nine Pheochromocytoma Adrenal gland Scaled Score. However,
predominantly immature, early-stage steroidogenesis in ACC there is currently no consensus on the adoption of a formal
that performed well in differentiating adenomas from ACCs. scoring system for these tumors [2]. According to the current
The use of this technique in clinics is, however, premature [5]. World Health Organization (WHO) classification, malignancy
For best patient care, adequate visualization of the tumor is defined by the presence of metastases to a site where
and potential metastases is essential. For differential diagnosis pheochromocytoma/paraganglionic tissue is not normally
of an adrenal mass computed tomography (CT) and magnetic present, e.g. liver or bone, to avoid confusion with multiple
resonance imaging (MRI) are currently considered equally primary tumors. No single histological finding can predict
effective. CT is less expensive and should be recommended as metastatic disease. However, patients with large tumor size (>5
the first choice in a suspected ACC. Conversely, MRI offers a cm), extra-adrenal location, or SDHB mutation have a higher
preferential application for the suspicion of risk of malignancy.
vii | Berruti et al. Volume 23 | Supplement 7 | October 2012

by guest
on 05 February 2018
Table 2. Disease staging system for adrenocortical carcinoma term survival [16] (V, A). To obtain an R0 resection of a
locally advanced ACC, it is often mandatory to resect ( parts
ENSAT TNM Definition of ) adjacent organs such as the wall of the vena cava, liver,
stage spleen, colon, pancreas and/or stomach (V, B). Locoregional
I T1, N0, M0 Tumor ≤5 cm lymphadenectomy improves tumor staging and seems to lead
II T2, N0, M0 Tumor >5 cm to a favorable oncologic outcome [17] (IV, A). No data on the
III T1–T2, N1, M0 Lymph node involvement and/or prognostic role of radical surgery and lymphadenectomy are
T3–T4, N0–N1, M0 tumor infiltration into surrounding available up to now for malignant pheochromocytoma.
tissue and/or a tumor thrombus in For ACC tumors not invading the kidney, concomitant
the vena cava and/or renal vein nephrectomy seems not to improve disease-free and overall
IV T1–T4, N0–N1, M1 Metastatic disease survival and can be avoided, although additional data are
needed (V, B).
In the case of inoperable local infiltrating or metastatic ACC,
surgical excision of the primary tumor and/or metastasis
should be considered in case of an objective response after
The disease stage and margin-free resection are the most neoadjuvant chemotherapy, when a radical resection seems to
important prognostic factors in ACC. In the assessment of the be feasible [16].
disease stage, we recommend the tumor–node–metastasis Cytoreductive (R2) resection in malignant
(TNM) classification proposed by the ENSAT network pheochromocytoma may improve the quality of life and
(Table 2) [9] (IV, A), since this system seems to be superior to survival by reducing the tumor burden and controlling
the staging system published by the Union Internationale hormonal hypersecretion [2] (V, B). It is also indicated in
Contre le Cancer in 2004. Age, mitotic count, the proliferation selected ACC patients with advanced tumors and severe
marker Ki67, and the glucocorticoid excess are additional symptomatic hormone excess when medical therapy is not able
prognostic parameters to refine the prognostic stratification to control endocrine symptoms, and life expectancy is >6
[10]. New molecular markers for aggressiveness and survival months [18] (V, B).
have been recently proposed, but their use in clinics needs Surgery for the recurrence of ACC (local and/or with a low
validation [11]. There is no staging system for malignant tumor metastatic burden) after primary treatment might be
pheochromocytoma and, in contrast to other neuroendocrine effective in improving survival, if R0 resection is achievable and
tumors, no poorly differentiated category is recognized. The the time to first recurrence was >12 months (IV, B).
survival rate depends mainly on the tumor size, primary tumor The hormone hypersecretion can increase the risk of
location (extra-adrenal location is associated with poor perioperative complications. For ACC patients, it is very
prognosis) [12], and the SDHB mutation status (SDHB important to diagnose any adrenocorticotropic hormone
mutations are predictive of poor prognosis) [13]. The overall 5- (ACTH)-independent glucocorticoid excess prior to surgery to
year survival rate is between 34% and 60%; however, adopt measures to prevent a postoperative adrenal crisis or
heterogeneous survival is a critical characteristic of this tumor insufficiency. In all patients with glucocorticoid excess
[12]. A significant number of patients die of hormone-related hydrocortisone must be administered during surgery (e.g. 150
complications (increased blood pressure, constipation) [12]. mg/days) and postoperatively.
In patients with pheochromocytoma and secreting
paraganglioma, exposure to high levels of circulating
management of local regional disease catecholamines during surgery could cause hypertensive crises
Surgery is of utmost importance in the treatment of both ACC and arrhythmias. Therefore, all patients with
and pheochromocytoma. Adrenal surgery should be performed pheochromocytoma or paraganglioma should receive
only in selected centers with >10 adrenalectomies for adrenal preoperative preparation at least 10–14 days before surgery (V,
cancer per year [14] (IV, A). A) [19]. Blood pressure targets for the treatment are <130/80
Open surgery with transperitoneal access is the standard mmHg in the supine position, and systolic blood pressure
treatment of all patients with localized (stage I–II) and local preferably >90 mmHg in the upright position. Traditionally,
advanced stage (stage III) ACCs when complete resection can the non-competitive α-adrenoreceptor antagonist
be achieved. Laparoscopic adrenalectomy is a safe and effective phenoxybenzamine is frequently used. A standard dose is 10
procedure for pheochromocytoma and a selected group of mg twice daily with adjustments every 2–4 days. Alternatively
patients with small ACCs (<8 cm) without preoperative doxazosine, a competitive and selective α1-adrenoreceptor
evidence for invasiveness and adrenal masses (e.g. antagonist might be as effective with fewer side effects. If the
incidentalomas) that are judged as only potentially malignant. target blood pressure is not reached, Ca-antagonists (nifedipine
This technique must be performed only in centers with a slow release) or metyrosine may be used. Blockade of β-
consolidated experience in laparoscopic adrenal surgery, in adrenergic receptors is indicated in patients developing
which principles of oncologic surgical treatment are strictly tachyarrhythmias, but should never be started before the
respected [15] (IV, B). blockade of α-receptors.
The resection status (R0, R1, R2) is a major predictor of Hypertension during surgery may be treated with magnesium
prognosis for ACC. A margin-free complete resection (R0 sulfate, intravenous α-adrenoreceptor antagonist
resection), in fact, provides the only means to achieve long- (phentolamine), calcium antagonist, and/or nitroprusside or
Volume 23 | Supplement 7 | October 2012 doi:10.1093/annonc/mds231 | vii

by guest
on 05 February 2018
nitroglycerine. Tachycardia can be treated with intravenous β-

adrenergic receptor blocker (esmolol). Postoperative hypotension
should be prevented by saline infusion the day before surgery. If
this adverse event occurs, it should be treated aggressively.
Postoperative care should also focus on glucose levels
because hypoglycemia may occur after the reduction of
catecholamine levels.
Most ACC patients have resectable disease at presentation,
however, more than half of the patients who have undergone
complete removal of the tumor are destined to have a relapse,
often with metastases [1]. Similarly, radical resection is not a
guarantee of cure for pheochromocytomas either. In one series
of 171 patients followed up after surgical resection of a
chromaffin cell tumor, 29 patients (17%) had recurrent or new
tumors, which were malignant in 15 (9%) cases [20].
The aggressive behavior and the high recurrence rate of ACC
provide the rationale for the use of adjuvant therapy. Adjuvant
radiotherapy to the tumor bed should be considered in patients
with incomplete/R1 resection or Rx resection [21] (V, A).
Mitotane has been the reference drug for the management of
ACC for decades. In a case control study involving 177 patients,
the outcome of 47 patients followed in Italian reference centers
that systematically adopted adjuvant mitotane to all radically
operated ACC was significantly higher (in terms of both disease-
free survival and overall survival) than the outcome of 55 Italian
patients and 75 German patients followed in institutions not
administering adjuvant mitotane therapy [22].
Although these data cannot be considered conclusive, they
suggest that adjuvant mitotane can delay and possibly prevent
a recurrence of disease (V, B). On these bases, a panel of
international experts unanimously stated that patients with Figure 1 Algorithm on management according to stage, risk factors, and
potential residual disease (R1 or Rx resection) and/or Ki67 disease characteristics for adrenocortical carcinoma (ACC) (A) and
more than 10% should be offered adjuvant mitotane, whereas pheochromocytoma (B). *Low-risk ACC is defined stage I–II and Ki67
adjuvant therapy was not considered mandatory in patients expression in ≤10% of neoplastic cells, high-risk ACC: stage III or Ki67
fulfilling all of the following criteria: stage I or II disease, expression in >10% of neoplastic cells.
histologically proven R0 resection; and Ki67 expressed in
≤10% of neoplastic cells [23] (V, B) (Figure 1A). For these
patients participation in the first prospective multinational involved in the follow-up of patients during mitotane
randomized trial, testing the efficacy of adjuvant mitotane treatment.
therapy (www.adiuvo-trial.org) is recommended whenever There are no data regarding the optimal duration of
possible. Mitotane is a difficult drug to manage, with a long adjuvant mitotane; we recommend that adjuvant mitotane
half-life, dose-limiting toxicity, and a narrow therapeutic should be administered for at least 2 years since the greatest
window. Although the evidence is limited, the panel frequency of disease recurrence is expected within this
recommends that the mitotane dosing should be guided by timeframe. It is noteworthy that while mitotane is well
plasma measurements, aiming at a concentration between 14 tolerated by a fraction of patients, the majority of patients find
and 20 mg/ml since there is less antitumor activity <14 mg/ml mitotane a difficult therapy that markedly impacts the quality
and the risk of toxicity (gastrointestinal and neurologic) is of their lives. So the treatment duration should be assessed
higher above 20 mg/ml (V, A) [24]. A suggested mitotane individually, taking carefully into account the cost/benefit ratio.
regimen is depicted in Table 3. Several blood analytes should
be monitored during mitotane therapy. Due to the adrenolytic management of advanced/metastatic
activity of mitotane, all patients must receive concomitant
administration of glucocorticoids to cover adrenal insufficiency.
disease
Owing to an increased metabolic clearance rate of Prognosis in locally advanced inoperable and metastatic ACC
glucocorticoids by mitotane therapy, high-dose glucocorticoid patients is poor, the 5-year overall survival being <15%.
replacement is needed. Mineralocorticoid supplementation is However, recent studies have highlighted a greater
necessary only in a subset of patients (Table 3). Testosterone heterogeneity of advanced ACC than previously thought.
and thyroxine supplementation may be also required. Mitotane Extremely long survival has been reported in patients with
can also induce a wide spectrum of hormone and metabolic resectable oligometastatic disease with considerable intervals
derangement and so an experienced endocrinologist should be among recurrences [10] (V, B).
vii | Berruti et al. Volume 23 | Supplement 7 | October 2012

by guest
on 05 February 2018
Table 3. Mitotane dose regimen, glucocorticoid, and mineralocorticoid supplementation, blood level monitoring, and dose adjustment according to toxicity
and blood level monitoring
Mitotane dose regimena • Start with 1.5 g/d and increase dose within 4–6 days to 6 g/days
• After 3 weeks, adjust dosage according tolerability and blood level (see below)
• Maximum dose 12 g/days, but most patients do not tolerate >8 g/days
• Target mitotane blood level 14–20 mg/l. Using this regimen, ∼50% of patients achieve the target level
within 3 months
Glucocorticoid and mineralocorticoid • A total daily dose of 50 mg hydrocortisone (divided as 20–20–10 mg) or 75 mg cortisone acetate and
supplementation more may be needed. Glucocorticoid replacement is monitored best with careful clinical assessment
• Fludrocortisone may be added depending on the blood pressure, serum potassium levels, and plasma
renin activity
Recommended blood monitoring • Mitotane serum levels every 2–3 weeks in the first 3 months. After reaching a plateau, the interval can
during mitotane therapy be extended (i.e. every 6 weeks)
• Glutamate-Oxaloacetate Transaminase (GOT), Glutamate-Pyruvate Transaminase (GPT), bilirubin,
Gamma-Glutamyl-Transferase (GGT). Initially every 4 weeks, after 6 months every 8 weeks. GGT is
invariably elevated without clinical consequences. If other liver enzymes are rapidly increasing (>3-fold
of baseline), there is a risk of liver failure: stop mitotane
• TSH, fT3, fT4 every 3–4 months. Thyroid hormone replacement is recommended in patients with
clinical symptoms of hypothyroidism
• Testosterone, free testosterone, and sexual hormone binding globulin (SHBG) should be tested in male
patients with symptoms of hypogonadism
• Renin every 3 months. If renin increases in the presence of symptoms suggestive of mineralocorticoid
deficiency, fludrocortisones should be added
• Cholesterol (High-Density Lipoprotein (HDL), Low-Density Lipoprotein (LDL)), triglycerides every 3–
4 months (in an adjuvant setting). If LDL/HDL cholesterol consistently increases, consider treatment
with statins not metabolized by CYP3A4 (e.g. pravastatin, rosuvastatin)
• Blood count every 3–4 months
Plasma mitotane level CNS (grade 2)/GI side effects (grade 3/4) Grade 3/4 CNS side effects
Absent Present Present
<14 mg/l Increase daily dose by 1 gb Reduce daily dose by 1 g Stop mitotanec
14–20 mg/l Maintain dose Reduce daily dose by 1.5 g Stop mitotanec
>20 mg/l Reduce daily dose to 50%–75% of the most Stop mitotanec Stop mitotanec
recent dose
Recommended dose adjustment according to the central nervous system (CNS)/gastrointestinal (GI) side effects and plasma mitotane level. HDL, high-
density lipoprotein; LDL, low-density lipoprotein; GOT, glutamate-oxaloacetate transaminase; GPT, glutamate-pyruvate transaminase; GGT, gamma-
glutamyl-transferase.
a
An alternative low-dose regimen is also available with potentially similar efficacy.
b
up to the maximum tolerated dose.
c
Until symptom resolution (grade 0 or 1)
Mitotane is the only drug approved in locally advanced of patients with a low tumor burden and/or more indolent
inoperable and metastatic patients although randomized, disease (Figure 1A). The combination with locoregional
controlled prospective trials are lacking (IV, A). Response therapies such as radiofrequency ablation (RFA) is
rates vary between 13% and 35%, but many of these results recommended. In case of rapidly progressing or life-
were derived from retrospective series since the 1960s with threatening extensive metastatic disease and/or radiological
variability in the response criteria [1, 10]. Patients with progression under mitotane, cytotoxic chemotherapy is
long-term maintenance of mitotane levels in the therapeutic indicated. Over the last 15 years, only 11 prospective single-
range may obtain a survival benefit [25]. Patients with arm chemotherapy studies with a total of 239 patients with or
hormonal excess often experience clinical benefit of this without mitotane have been published. Response rates vary
strategy and continuation of mitotane treatment can be between 7% and 54%, again with variability in the response
indicated in these patients even after radiological progression criteria. The association of mitotane to chemotherapy seems to
when alternative strategies to inhibit the hormonal excess are be more active than chemotherapy alone [1], although no
lacking. randomized trials have formally demonstrated this superiority.
Owing to the latency of mitotane to attain the therapeutic In the First International Randomized trial in advanced or
range, mitotane monotherapy is indicated in the management Metastatic Adrenocortical Carcinoma Treatment (FIRM-ACT
Volume 23 | Supplement 7 | October 2012 doi:10.1093/annonc/mds231 | vii

by guest
on 05 February 2018
trial), the two most active treatment regimens namely radiofrequency ablation. Radionuclide therapy is an effective
Etoposide, Doxorubicin, Cisplatin, and Mitotane (EDP-M) and treatment and 131I-MIBG, in activities ranging from 5.5 to 38
Streptozotocin and Mitotane (Sz-M) were compared with 304 GBq (150–1000 mCi), is the most frequently used.
chemotherapy-naïve patients. Patients with disease progression Approximately 50% of patients are eligible for 131I-MIBG
received the alternate regimen. The results of this trial, recently therapy based on the uptake on diagnostic scans. Several
published [26], indicate to us that EDP-M is the superior studies have been published on the efficacy of 131I-MIBG
regimen. treatment [2]; most are retrospective, and only one is a
Although no statistically substantial increase in overall prospective phase II study [30]. Objective responses were
survival was documented in patients receiving EDP-M as the observed in 22–47% of cases. Long-term survival of responders
first-line therapy, significantly better response rates and of 4.7 years or 72 months was reported but progression at
progression-free survival rates were achieved with EDP-M in study entry was not a prerequisite of most studies. Objective
comparison with Sz-M. The rate of serious adverse events was responses were mainly observed in patients with soft tissue
comparable. Of note, the results of the second-line regimens metastases. Grade 3–4 toxicity was reported in 16%–83% of
replicated the rates observed with the first-line therapy. Since patients.
EDP-M was superior to Sz-M in terms of progression-free 131
I-MIBG therapy should be considered as a first-line
survival either as first-line or second-line therapy, the crossover approach in patients with a good uptake of 123I-MIBG and
design may have attenuated its advantage on overall survival. unresectable, progressive pheochromocytoma/paraganglioma
On these bases, we recommend EDP-M as the first-line or symptomatic patients (not amenable to locoregional
therapy for ACC requiring cytotoxic therapy, and as the control), or patients with a high tumor burden with a low
reference for new therapies (I, A; Figure 1A). number of bone metastases. More recently peptide-
In patients unfit for the EDP-M regimen P-M may radiolabeled radiotherapy has also been developed.
constitute a reasonable alternative [27]. However, the FIRM- Cyclophosphamide- and dacarbazine-based regimens
ACT trial with a median overall survival between 12 and 14.8 combined with vincristine (CVD) or doxorubicin (CVDD or
months clearly indicated that new systemic therapy options are CDD) are the best studied chemotherapy regimens [2]. In the
urgently needed, but thus far positive results are lacking largest published study to date (n = 52 patients), 40% of
despite an increasing number of prospective studies, also patients treated with CVD, CDD, or CVDD experienced
incorporating modern targeted agents. Results of studies with clinical benefits, including a reduction in tumor size in 25% of
oral CYP3A4-mediated drugs may have been hampered by cases [31]. Systemic chemotherapy is debated as a first-line
subtherapeutic systemic exposure due to CYP3A4 induction by therapy in patients with a low uptake of 123I-MIBG and
mitotane [28, 29]. A multicenter prospective randomized, unresectable, rapidly progressive pheochromocytoma/
placebo-controlled clinical trial aimed to test the efficacy of paraganglioma, or patients with high tumor burden or with a
OSI 906–301, an IGR inhibitor, as second-/third-line approach high number of bone metastases. As in well-differentiated
in ACC patients has recently completed patient accrual. pancreatic neuroendocrine tumors, there is a strong rationale
However, the results of this study will not be available before and some first evidence on the potential efficacy of anti-
2013. angiogenic drugs in malignant pheochromocytomas. Thus, the
In case of painful metastasis, palliative radiotherapy is an European ENSAT network is currently launching a
option, particularly in bone lesions (IV, B). Arterial randomized, placebo-controlled trial testing sunitinib in
chemoembolization and radiofrequency ablation may be patients with malignant pheochromocytoma and
beneficial in selected patients (V, C). paraganglioma (FIRST-MAPPP trial).
The therapeutic strategy of metastatic pheochromocytoma/
paraganglioma aims to control excessive catecholamine
secretion and tumor burden, but no curative treatment is
follow-up
achievable. Treatment choices include a wait and see policy, International recommendations on follow-up are lacking both
locoregional therapies, systemic chemotherapy, and for ACC and pheochromocytoma/paraganglioma, and this is a
radiopharmaceutical agents (Figure 1B) [2], and they should be significant problem. The recommendations in these guidelines
discussed case by case in a multidisciplinary specialized setting. were formulated based on data available on the natural history
In the absence of any randomized studies and demonstrated of these diseases, personal experience, and consensus among
impact on survival, the patients’ quality of life should always be panelists.
considered. Indeed, due to the indolent course of subgroups of For patients with ACC after complete resection, we
patients, a wait and see policy coupled with a watchful follow- recommend regular follow-up every 3 months including
up can be considered as an option in asymptomatic patients abdominal CT (or MRI), thoracic CT, and monitoring of
with a low tumor burden. In these patients, an antineoplastic initially elevated steroids (V, B). After 2 years, intervals may be
treatment should be recommended in case of rapid progression gradually increased. In case of long-term persistence of the
and/or symptom onset. In the absence of tumor progression, disease-free status, follow-up should be continued for at least
surgery of the primary tumor or metastases can reduce 10 years.
hormone secretion and may prevent complications related to a For locally advanced or metastatic disease, overall survival
critical anatomical location and improve the efficacy of and time to progression (TTP) are the most important
subsequent therapies [2]. Metastatic disease palliation may also endpoints with the response rate (RR) as a secondary end
benefit from local therapy with embolization and or point. The TTP and RR guide the clinical decision-making in
vii | Berruti et al. Volume 23 | Supplement 7 | October 2012

by guest
on 05 February 2018
Table 4. Summary of Recommendations malignant recurrence and require long-term clinical
(adrenergic symptoms and blood pressure levels) and
• ACC is defined by a Weiss score of 3 or more. Malignant biochemical ( plasma or urinary metanephrine,
pheochromocytomas/paragangliomas are defined by the presence of normetanephrine, chromogranin A, and methoxythyramine)
metastasis. follow-up [2]. The follow-up is especially important for
• Patients suspected to harbor primary adrenal tumors should undergo a patients with extra-adrenal primary disease, tumor size >5 cm,
standardized diagnostic work-up consisting of endocrine assessment for or SDHB mutations. Biochemical testing is repeated ∼14 days
excess hormone production and modern imaging (CT/MRI of abdomen, following surgery to check for remaining disease and thereafter
chest CT, and in selected cases supplemented by isotope functional every 3–4 months for 2–3 years. This should subsequently be
imaging mainly FDG-PET). The diagnostic work-up differs between repeated every 6 months. Patients with new events (high blood
ACC and pheochromocytoma. pressure, adrenergic symptoms, or pain) or pathological
• Guided biopsies of potentially resectable primary adrenal tumors are not endocrine tests and/or elevated circulating chromogranin A
informative in most cases, but are potentially harmful and should be should undergo imaging that includes thorax and abdomen CT
avoided. and best functioning imaging (PET FDG in most cases). In
• The ENSAT TNM staging system should be used for ACC staging.
case of proven malignant disease, SDHB mutation, extra-
• Histological diagnosis should be done by an experienced pathologist and
adrenal primary disease, and in rare cases of
should rely on morphological, mitotic, and immunohistochemical
pheochromocytoma/paraganglioma without relevant
parameters.
preoperative hormone secretion, imaging could be repeated at
• Complete surgical extirpation of localized and locally advanced ACC or
least every 6 months during the first year and yearly afterward,
pheochromocytoma (R0 resection) is the mainstay of potentially curative
approaches. Additionally, a locoregional lymphadenectomy is suggested
irrespective of the negative results of biochemical tests. In these
for ACC. patients, lifelong follow-up is recommended [2].
• In pheochromocytoma cytoreductive surgery might be considered. In
advanced ACC, this approach is only reasonable for patients with severe conflict of interest
hormone excess.
• Meticulous perioperative management of hormonal, glucose, electrolytes, Dr. Berruti has reported: advisory board honorarium from
cardiac and fluid/blood pressure abnormalities is a critical component of Astellas. Prof. Fassnacht has reported Principal Investigator of
patient care. a study funded by HRA Pharma, Research grants: Pfizer,
• Despite the limited literature evidence, adjuvant systemic mitotane is Astellas Pharma.
recommended for patients with ACC and incomplete resection (R1, Rx The other authors have reported no potential conflicts of
stage III) or in the presence of high-risk features (Ki67>10%). R1 and Rx interest.
ACC resections may be followed by additional adjuvant radiotherapy to
the tumor bed.
• Fit patients with inoperable ACC, high tumor volume and rapid disease references
progression should be treated with combination cytotoxic chemotherapy 1. Fassnacht M, Libé R, Kroiss M et al. Adrenocortical carcinoma: a clinician’s
plus mitotane (EDP-M). Less fit patients and/or patients with low tumor update. Nat Rev Endocrinol 2011; 7: 323–335.
burden and slow progression can (first) be managed with mitotane 2. Pacak K, Eisenhofer G, Ahlman H et al. International symposium on
monotherapy combined or not with locoregional options. pheochromocytoma. Pheochromocytoma: recommendations for clinical practice
• Disease and symptom control is the main treatment goal for patients from the first international symposium. October 2005. Nat Clin Pract Endocrinol
with inoperable pheochromocytoma and can be attempted by Me tab 2007; 3: 92–102.
radiopharmaceuticals (131I-MIBG), locoregional ablative procedures, 3. Bertherat J, Bertagna X. Pathogenesis of adrenocortical cancer. Best Pract Res
and/or combination chemotherapy (CVD) in selected cases. Clin Endocrinol Metab 2009; 23: 261–271.
• Wait and see policy is recommended in low tumor burden and 4. Esenhofer G, Lenders JW, Siegert G et al. Plasma methoxytyramine: a novel
asymptomatic malignant pheochromocytoma and paraganglioma. biomarker of metastatic pheochromocytoma and paraganglioma in relation to
established risk factors of tumour size, location and SDHB mutation status. Eur J
• Patients with resected ACC or pheochromocytoma should be followed at
Cancer 2012; 48: 1739–1749.
regular intervals with clinical, imaging and biochemical screens for at
5. Arlt W, Biehl M, Taylor AE et al. Urine steroid metabolomics as a biomarker tool
least 10 years. Lifelong surveillance with an increased interval of time is
for detecting malignancy in adrenal tumors. J Clin Endocrinol Metab 2011; 96:
favored in malignant pheochromocytoma/paraganglioma. 3775–3784.
• The follow-up of patients with inoperable disease should be performed 6. Timmers HJ, Kozupa A, Chen CC et al. Superiority of fluorodeoxyglucose positron
every 2–4 months for ACC and every 3–6 months for emission tomography to other functional imaging techniques in the evaluation of
pheochromocytoma/paraganglioma during the first year of follow-up and metastatic SDHB-associated pheochromocytoma and paraganglioma. J Clin
then adjusted. Oncol 2007; 25: 2262–2269.
7. Sbiera S, Schmull S, Assie G et al. High diagnostic and prognostic value of
steroidogenic factor-1 expression in adrenal tumors. J Clin Endocrinol Metab
2010; 95: E161–E171.
8. Weiss LM, Medeiros LJ, Vickery AL, Jr. Pathologic features of prognostic
an individual patient and should be evaluated at regular significance in adrenocortical carcinoma. Am J Surg Pathol 1989; 13: 202–206.
9. Fassnacht M, Johanssen S, Quinkler M et al. German adrenocortical carcinoma
intervals (every 12 weeks or less depending on the therapy)
registry group; European Network for the Study of Adrenal Tumors. Limited
using CT scans. The role of PET scans is not yet clear in ACC. prognostic value of the 2004 International Union Against Cancer staging
Patients who underwent successful surgery for non- classification for adrenocortical carcinoma: proposal for a Revised TNM
metastatic pheochromocytoma/paraganglioma are at risk of Classification. Cancer 2009; 115: 243–250.
Volume 23 | Supplement 7 | October 2012 doi:10.1093/annonc/mds231 | vii

by guest
on 05 February 2018
10. Baudin E, Leboulleux S, Al Ghuzlan A et al. Therapeutic management of 21. Polat B, Fassnacht M, Pfreundner L et al. Radiotherapy in adrenocortical
advanced adrenocortical carcinoma: what do we know in 2011? Horm Cancer carcinoma. Cancer 2009; 115: 2816–2823.
2011; 6: 363–371. 22. Terzolo M, Angeli A, Fassnacht M et al. Adjuvant mitotane treatment for
11. de Reyniès A, Assié G, Rickman DS et al. Gene expression profiling reveals a adrenocortical carcinoma. N Engl J Med 2007; 356: 2372–2380.
new classification of adrenocortical tumors and identifies molecular predictors of 23. Berruti A, Fassnacht M, Baudin E et al. Adjuvant therapy in patients with
malignancy and survival. J Clin Oncol 2009; 27: 1108–1115. adrenocortical carcinoma: a position of an international panel. J Clin Oncol
12. Ayala-Ramirez M, Feng L, Johnson MM et al. Clinical risk factors for malignancy 2010; 28: e401–e402.
and overall survival in patients with pheochromocytomas and sympathetic 24. Haak HR, Hermans J, van de Velde CJ et al. Optimal treatment of adrenocortical
paragangliomas: primary tumor size and primary tumor location as prognostic carcinoma with mitotane: results in a consecutive series of 96 patients. Br J
indicators. J Clin Endocrinol Metab 2011; 96: 717–725. Cancer 1994; 69: 947–951.
13. Amar L, Baudin E, Burnichon N et al. Succinate dehydrogenase B gene 25. Hermsen IG, Fassnacht M, Terzolo M et al. Plasma concentrations of o,p’DDD, o,
mutations predict survival in patients with malignant pheochromocytomas or p’DDA, and o,p’DDE as predictors of tumor response to mitotane in
paragangliomas. J Clin Endocrinol Metab 2007; 92: 3822–3828. adrenocortical carcinoma: results of a retrospective ENS@T multicenter study. J
14. Lombardi CP, Raffaelli M, Boniardi M et al. Adrenocortical carcinoma: effect of Clin Endocrinol Metab 2011; 96: 1844–1851.
hospital volume on patient outcome. Langenbecks Arch Surg 2012; 397: 201–207. 26. Fassnacht M, Terzolo M, Allolio B et al. Combination chemotherapy in advanced
15. Porpiglia F, Fiori C, Daffara F et al. Retrospective evaluation of the outcome of adrenocortical carcinoma. N Engl J Med 2012; 366: 2189–2197.
open versus laparoscopic adrenalectomy for stage I and II adrenocortical cancer. 27. Bukowski RM, Wolfe M, Levine HS et al. Phase II trial of mitotane and cisplatin in
Eur Urol 2010; 57: 873–878. patients with adrenal carcinoma: a Southwest Oncology Group study. J Clin
16. Icard P, Goudet P, Charpenay C et al. Adrenocortical carcinomas: surgical trends Oncol 1993; 11: 161–165.
and results of a 253-patient series from the French Association of Endocrine 28. van Erp NP, Guchelaar HJ, Ploeger BA et al. Mitotane has a strong and a
Surgeons study group. World J Surg 2001; 25: 891–897. durable inducing effect on CYP3A4 activity. Eur J Endocrinol 2011; 164:
17. Reibetanz J, Jurowich C, Erdogan I et al. Impact of lymphadenectomy on the 621–626.
oncologic outcome of patients with adrenocortical carcinoma. Ann Surg 2012; 29. Ayala-Ramirez M, Feng L, Habra MA et al. Clinical benefits of systemic
255: 363–369. chemotherapy for patients with metastatic pheochromocytomas or sympathetic
18. Schteingart DE, Doherty GM, Gauger PG et al. Management of patients with 675 extra-adrenal paragangliomas: insights from the largest single-institutional
adrenal cancer: recommendations of an international consensus conference. experience. Cancer 2012; 118: 2804–2812.
Endocr Relat Cancer 2005; 12: 667–680. 30. Kroiss M, Quinkler M, Lutz WK et al. Drug interactions with mitotane by induction
19. Pacak K. Preoperative management of the pheochromocytoma patient. J Clin of CYP3A4 metabolism in the clinical management of adrenocortical carcinoma.
Endocrinol Metab 2007; 92: 4069–4079. 665 Clin Endocrinol (Oxf ) 2011; 75(5): 585–591.
20. Amar L, Servais A, Gimenez-Roqueplo AP et al. Year of diagnosis, features 31. Gonias S, Goldsby R, Matthay KK et al. Phase II study of high-dose [131I]
at presentation, and risk of recurrence in patients with pheochromocytoma metaiodobenzylguanidine therapy for patients with metastatic pheochromocytoma
or secreting paraganglioma. J Clin Endocrinol Metab 2005; 90: 2110–2116. and paraganglioma. J Clin Oncol 2009; 27: 4162–4168.
vii | Berruti et al. Volume 23 | Supplement 7 | October 2012

by guest
on 05 February 2018
Early and locally advanced non-small-cell lung cancer

(NSCLC): ESMO Clinical Practice Guidelines for
P. E. Postmus1, K. M. Kerr2, M. Oudkerk3, S. Senan4, D. A. Waller5, J. Vansteenkiste6, C. Escriu1 & S. Peters7,

on behalf of the ESMO Guidelines Committee*
1
The Clatterbridge Cancer Centre and Liverpool Heart and Chest Hospital, Liverpool; 2University of Aberdeen, Aberdeen, UK; 3Center for Medical Imaging,
University of Groningen, Groningen; 4Department of Radiation Oncology, VU University Medical Center, Amsterdam, The Netherlands; 5Department of Thoracic
Surgery, University Hospitals of Leicester NHS Trust, Leicester, UK; 6University Hospitals KU Leuven, Leuven, Belgium; 7Oncology Department, Service d’Oncologie
Médicale, Centre Hospitalier Universitaire Vaudois (CHUV), Lausanne, Switzerland
†
Approved by the ESMO Guidelines Committee: March 2010, last update May 2017. This publication supersedes the previously published version—Ann Oncol 2013; 24
(Suppl. 6): vi89–vi98.
Incidence and epidemiology rate of overdiagnosis of indolent cancers, such as lepidic adeno-
carcinomas (previously named bronchioloalveolar carcinoma)
Lung cancer is the leading cause of cancer mortality worldwide [4, 5], although a pathology review according to the recent classi-
with 1.8 million newly diagnosed cases, or 13% of all cancers fication [6] made this unlikely, as it categorised 97% of the de-
diagnosed, in 2012 [1]. The worldwide numbers are still rising tected cancers as invasive [7].
despite an ongoing small decline in the Western world. Global How screening for lung cancer should become part of standard
statistics estimate that 15% of lung cancers in men and 53% in evidence-based practice therefore needs to be analysed further.
women are not attributable to smoking, overall accounting for Nevertheless, for part of the Western world this positive trial has
25% [2]. resulted in guidelines for screening within high-risk groups
[8, 9]. Implementation in other health care systems has not yet
happened as confirmation of the results in a comparable trial in a
Screening for lung cancer different geographical area is crucial. Mature data of the
Lung cancer symptoms occur late in the disease, so the majority NELSON study [10] are expected in 2017 and may result in con-
of patients with lung cancer present with advanced disease. firmation. The NELSON study developed a non-invasive proto-
Unfortunately for those patients, the disease will not be curable col based on volume measurement and growth rate resulting in a
with currently available therapies. Therefore, early detection 10-fold reduction of the false-positive rate compared to the
might be a valuable approach to detect the disease at an earlier, NLST, maintaining the same lung cancer detection rate [11].
asymptomatic and potentially curable stage. Screening evaluated An important question is how to translate the findings of both
in relatively small trials failed to show benefit if periodical chest NLST and NELSON into advice on ‘who to screen’ (high-risk
X-ray and/or sputum cytology were used; screening by these tech- group), ‘how often’ (intervals between rounds), and ‘for how
niques is therefore not recommended. long’ (until which age). It is difficult to come to conclusions on
The much larger National Lung Cancer Screening Trial how to perform screening for the detection of incidence cases as a
(NLST) comparing low-dose computed tomography (LDCT) to screening study initially mainly deals with prevalence cases and
chest X-ray in over 53 000 current or former heavy smokers ( 30 the trial runs during a limited period of time. Questions such as,
pack-years or 15 years since smoking cessation), aged between ‘what is the optimal time between screening rounds?’, and ‘for
55 and 74 years, showed a 20% reduction in lung cancer-related how long should this be continued?’, are difficult to answer, be-
death and an overall all-cause mortality reduction of 6.7% [3]. cause the characteristics of tumours detected during the preva-
LDCT screening thus reduces lung cancer-related mortality. lence screening might differ from tumours detected during
However, this positive outcome generates new questions on the incidence screening [12]. Furthermore, findings detected during
C The Author 2017. Published by Oxford University Press on behalf of the European Society for Medical Oncology.
V

by guest
on 05 February 2018
earlier screening rounds provide further possibilities for risk Table 1. Work-up for diagnosis and staging
stratifications and may lead to guidance on the length of screen-
Mandatory Optional
ing [13].
General Medical historya
Recommendations: Physical examinationa
• Screening with LDCT reduces lung cancer-related mortality Assessing comorbidity
[I, A]. It is not yet ready for large-scale implementation, PS
mainly because the lung cancer mortality reduction rate lacks
Imaging X-ray thorax
definite proof of a second study result, and partly because of
CT thoraxa
remaining questions regarding definition of the at-risk popu- PET-CT thoraxa Bone scintigraphy
lation, timing, interval and method of computed tomography MRI brainb Contrast enhanced-
(CT, especially 2D versus 3D evaluation), how to handle CT brain
(false-) positive findings and especially cost-effectiveness,
notably in relation to smoking cessation [I, A]. Laboratory Blood cell counts
• LDCT screening can be carried out outside a clinical trial pro- Renal function
Liver enzymes
vided it is offered within a dedicated programme with quality
Bone parameters
control, in a centre with experience in CT screening, a large vol-
ume of thoracic oncology activity and multidisciplinary manage- Cardio-pulmonary FVC, FEV1, DLCO
ment of suspicious findings [I, B]. Candidates are current or function ECG
former heavy smokers ( 30 pack-years or 15 years since If indicated: CPET Ejection fraction,
smoking cessation) aged 55–74 years, who are well informed CAG
about potential benefits and risks. Individuals offered LDCT Tissue procurement Bronchoscopyb,c
screening should be referred to a smoking cessation programme. EBUS/EUS mediastinal Mediastinoscopy
• LDCT screening should not be offered on an ad hoc individ-
nodesa
ual basis, but patients requesting screening should be referred CT-guided biopsy
to a dedicated programme, as recommended above [V, B].
a
• Other screening methods, such as chest X-ray, sputum analysis Tests needed for clinical staging.
b
or biomarkers are not recommended for clinical use [I, C]. See text.
c
Depending on site and size of tumour with biopsy/aspiration/brush/
washing.
CAG, coronary angiography; CPET, cardio pulmonary exercise testing;
Diagnosis and pathology/molecular biology CT, computed tomography; DLCO, diffusing capacity of the lungs for
carbon monoxide; EBUS, endoscopic bronchial ultrasound; ECG, elec-
Diagnosis trocardiogram; EUS, endoscopic ultrasound; FEV1, forced expiratory vol-
ume in 1 second; FVC, forced expiratory vital capacity; MRI, magnetic
The most common diagnostic test for lung cancer is fibreoptic resonance imaging; PET-CT, positron emission tomography computed
bronchoscopy, often extended with evaluation of regional lymph tomography; PS, performance status.
nodes by endobronchial ultrasound (EBUS) and/or endoscopic
ultrasound (EUS). In most cases this will be sufficient to diagnose
non-small-cell lung cancer (NSCLC), although quite often the
amount of obtained material is not sufficient to sub-classify the becomes even more relevant as different histological subtypes dif-
tumour in more detail. fer with regards to metastatic pattern, recurrence and survival.
For earlier stages of NSCLC, the need for a detailed The beneficial effects of adjuvant chemotherapy (ChT) post-
pretreatment pathological diagnosis is not yet clear. In contrast resection may differ depending upon this adenocarcinoma sub-
to stage IV [14], the consequences of the upfront diagnosis for se- classification [17–19]; prospective trials are needed to evaluate
lecting the most effective therapy of stages I–III NSCLC are whether these retrospective findings have clinical consequences.
assumed to be less relevant. The pathological classification at diagnosis may influence ini-
For molecular analysis, the sample obtained through EBUS- tial treatment decisions such as the initial surgical approach. In a
guided aspirations of lymph nodes is often sufficient [15]. large surgical series (n ¼ 2268) of resected adenocarcinoma of
Commonly used tests are summarised in Table 1. 3 cm in diameter, the categories adenocarcinoma in situ (AIS),
minimally invasive adenocarcinoma (MIA) and lepidic predom-
inant (Lep) were found to have no metastasis in N1 or N2 lymph
Pathology nodes (n ¼ 329), whereas the other categories with predominance
As pathologists will not necessarily be aware of the disease stage at of acinar, papillary, micropapillary or solid growth patterns had
the time of pathological diagnosis, a thorough comprehensive N1 or N2 involvement in 22.9% of patients (445 of 1939). Until
diagnosis is always recommended whenever possible. now, these features are only detectable in full extent in resected
The recent World Health Organization (WHO) classification, material; further refining of preoperative work-up might make
with its further sub-classification of (surgically resected) adeno- this applicable for prospective use [20]. Future work may deter-
carcinoma, shows differences in metastatic pattern, recurrence mine if the extent of surgery could be limited to a segmentectomy
and survival between different histological subtypes [16]. This in the AIS and MIA subtypes, and a lobectomy could be
iv2 | Postmus et al. Volume 28 | Supplement 4 | August 2017

by guest
on 05 February 2018
performed with lymph node dissection for the invasive types. tumours indicate lack of mediastinal metastases [III, A]. This
Such decisions could be based on intra-operative frozen section diagnosis may be made intra-operatively by video-assisted
examination, which has a high concordance rate with final path- thoracoscopic biopsy and frozen section analysis.
ology [21], but is far from being a validated standard practice due • In isolated cases a diagnostic anatomical sublobar resection
to several technical and logistical problems [22–24]. may be acceptable.
Preoperative diagnostic work-up may identify patients at
higher risk for presence of regional lymph node metastases. By The solitary pulmonary nodule
measuring primary tumour low maximum standardised uptake
Solitary pulmonary nodules are a common problem and are usually
values (SUVmax) of fluorodeoxyglucose-positron emission tom-
a diagnostic challenge. Depending on presence or lack of benign
ography (FDG-PET) (< 3.0), it was possible to detect those cases
characteristics, such as calcification or no changes during at least
with low probability of mediastinal lymph node metastases, and
2 years, a diagnostic algorithm can be used to qualify the lesion as
to select the suitable candidates for a sublobar resection [25];
more or less likely to be malignant. However, it is important to note
however, this needs to be confirmed in comparable studies before
that validated diagnostic algorithms are not available in many
it can be concluded that a low SUVmax value of a peripheral tu-
populations. Guidelines developed by the British Thoracic Society
mour is useful for selection of patients for a sublobar resection.
(BTS) and the Fleischner Society were published recently [27, 28],
If bronchoscopy or transthoracic needle biopsy results in large
but like many previous guidelines, have focussed on Western popu-
( 0.7 mm2) and multiple ( 2) biopsies, the concordance with
lations. For other areas, such as Asia, with a high prevalence of
the final tumour classification after resection is 70% overall.
granulomatous disease and other infectious causes of pulmonary
For the acinar type, concordance was low, whereas the others
nodules, the recent Asian consensus guidelines are likely more ap-
were more favourable, but still relatively low at 70% [26]. These
propriate. The latter recommend a lesser reliance on positron emis-
types of study require further validation. Based on these observa-
sion tomography (PET) scans in Asian populations, and greater use
tions, and for other reasons, the idea of ‘minimal amounts of tissue
of non-surgical biopsy over surgical diagnosis or surveillance [29].
to come to a diagnosis’ of cancer needs to be re-evaluated, and
In general, it is important for clinicians to be aware of the em-
probably changed to ‘as much tissue as possible’ to allow better
phasis they would place on a ‘non-malignant’ result from a per-
diagnosis and classification as early as possible in the trajectory to
cutaneous biopsy. If the clinical and radiological evidence would
therapeutic decisions. In general, the rate of NOS (not otherwise
favour a surgical biopsy in any case, then the merits of non-
specified) after the complete diagnostic work-up should be <10%.
invasive methods should be discussed with the patient.
Recommendations: Recent data from the NELSON study on incidental nodules
• In patients with clinical stages I–III lesions, a pretreatment might be applied to the solitary nodule and incorporated in
pathological diagnosis is recommended prior to any curative guidelines [12]. Diagnostic procedures, as described in the previ-
treatment. ous section, will be of help in case further evaluation is needed.
• Bronchoscopy is the recommended test to obtain a patho-
logical diagnosis of centrally located tumours in stages I–III Recommendations:
with biopsy of any visible lesion [III, A]. • The diagnostic approach to non-calcified pulmonary nodules
• The pathological classification NOS should be used only in should be based on existing standard guidelines [III, A], al-
cases where it is impossible to obtain enough tissue for fur- though new evidence on nodule management is emerging.
• Likelihood of malignancy based upon risk calculation meth-
ther classification, or when steps to further classify the tu-
mour are inconclusive [V, A]. ods used in CT screening studies should be used only to
• An exception to the requirement for a pretreatment diagnosis guide the clinical assessment of pulmonary nodules detected
can be made if an experienced multidisciplinary group de- in the wider population [V, C].
cides that the risks of obtaining pathology may be unaccept-
able in a patient in whom the likelihood of malignancy is
high based on clinical and imaging findings [III, B]. Staging and risk assessment
• A pretreatment pathological diagnosis is strongly recom-
During the 16th World Congress of Lung Cancer, the Union for
mended for all patients before stereotactic ablative radiother- International Cancer Control (UICC) presented the revised tumour,
apy (SABR), unless a multidisciplinary tumour board is of node and metastasis (TNM) classification of malignant tumours
the opinion that the risk-benefit ratio of the procedure is un- (UICC TNM 8), published in December 2016 [30] and effective
acceptable. In such a situation, the predicted likelihood of since January 2017 (Table 2).
malignancy should preferably be at least 85%, based upon ac-
cepted criteria [III, B] [25]. Recommendation:
• The descriptive element of the recent WHO classification of
• In non-metastatic NSCLC, detailed locoregional staging accord-
adenocarcinoma subtypes should be used to describe bron-
ing to the 8th TNM staging system and the cardiopulmonary fit-
choscopic and CT-guided biopsies whenever possible [III, A].
ness of the patient determine the choice of treatment [III, A].
• The revised adenocarcinoma classification may identify pa-
tient subtypes for whom an anatomical sublobar resection,
rather than lobectomy, would be sufficient [III, A]. Locoregional staging
• FDG-PET may contribute for the selection of patients for ana- For locoregional staging, algorithms shown in Figures 1 and 2 are
tomical sublobar resections as low SUVmax values of peripheral still applicable.

by guest
on 05 February 2018
Table 2. Staging and stage grouping UICC TNM 8 [30] measuring the solid component of tumours, expert opinion fa-
vours lung or intermediate window settings [32].
Occult carcinoma TX N0 M0
CT follow-up studies have shown that incidental non-calcified
Stage 0 Tis N0 M0 non-solid lung lesions do not need shorter repeat CT examin-
Stage IA1 T1a(mi) N0 M0 ations than 1–2 years and are definitely less aggressive than solid
T1a N0 M0 or part-solid lesions and often even indolent.
Stage IA2 T1b N0 M0 The use of the staging system for tumours with additional nod-
Stage IA3 T1c N0 M0 ules has been left unchanged, although the approach to score
Stage IB T2a N0 M0 same lobe nodules as T3, different ipsilateral lobe as T4 and
Stage IIA T2b N0 M0 contralateral as M1a should be restricted to the same histological
Stage IIB T1a–c N1 M0 (sub)type and, as such, be considered as intrapulmonary metasta-
T2a–b N1 M0 ses [33]. In other situations, with more than one pulmonary site
T3 N0 M0 of disease, such as second primary tumours, these should be
Stage IIIA T1a–c N2 M0 staged differently. To conclude if two foci are indeed two differ-
T2a–b N2 M0 ent primaries is difficult; criteria are presented but often it will be
T3 N1 M0 impossible to come to a definitive conclusion and the role of a
T4 N0 M0 multidisciplinary tumour board is important [34]. When the
T4 N1 M0 conclusion is the presence of two primaries, each tumour should
Stage IIIB T1a–c N3 M0 be given a separate T, N and M category [35].
T2a–b N3 M0
A specific problem is the tumour with a specific growth pattern,
T3 N2 M0
such as ground glass or lepidic, and the pneumonic type. The
T4 N2 M0
International Association for the Study of Lung Cancer (IASLC)
Stage IIIC T3 N3 M0
proposes to determine the T of multifocal ground glass/lepidic tu-
T4 N3 M0
mours by the highest T-lesion, with either the number of tumours
UICC, Union for International Cancer Control; TNM, tumour, node and or m in parentheses to denote the multifocal nature, and that a
metastasis. single N and M category be used for all these lesions collectively.
Reprinted from [30] with permission from John Wiley & Sons, Inc. In daily practice, simply using m is to be preferred over trying to
estimate the number of groundglass opacity (GGO) areas. For the
pneumonic type, it is suggested to use size (or T3) if in one lobe,
T4 if involving a different ipsilateral lobe, and M1a if contralateral;
The two most striking changes in UICC TNM 8 are the further in that situation, the T stage will be based on the highest category
subdividing and detailing of both T and M stage, although the in the most involved lung. For N and M, a single category should
consequences for therapeutic approach are not yet obvious in all be used for all pulmonary areas of involvement [36]. Especially in
situations. the case where more than one lesion is present, and/or differences
The T stage was divided further by splitting T1 into three sub- in growth pattern are observed [34–36], accurate staging is vital to
groups based on size (T1a 1 cm, T1b > 1 cm to 2 cm, avoid erroneous interpretations leading to a false stage, resulting
T1c > 2 cm to 3 cm), this is continued into T2 (T2a > 3 cm in undertreatment.
to 4 cm, T2b > 4 cm to 5 cm), T3 (> 5 cm to 7 cm) and T4 For patients with abnormal mediastinal and/or hilar lymph
(> 7 cm). The T2 category was further enriched by adding the nodes at CT and/or PET, endosonography is recommended over
previous T3 classifiers, atelectasis/pneumonitis and/or involve- surgical staging [I, A]. If malignant nodal involvement is not
ment of main bronchus, irrespective of distance to main carina. found by this modality, subsequent surgical staging is recom-
Invasion of the diaphragm was found to have a similar prognosis mended [I, B]. For peripheral tumours without mediastinal in-
as other T4 tumours and has therefore been added to this cat- volvement on CT or PET-CT, mediastinal staging is advised in
egory [31]. case of no uptake of FDG by the primary tumour and/or a tu-
In addition to a further refinement of T stage overall, a number mour size 3 cm [II, C] [37].
of questions that were—despite the major improvements—left The proposed new staging suggests leaving the N categories un-
unanswered in the UICC TNM 7 classification, have now been changed, but to record for future testing the sub-classification of
addressed, and should therefore be incorporated in a new single (N1a, N2a) or multiple (N1b, N2b) affected nodes. For the
guideline. situation of so-called skip metastases, the N2a group is further
How to code and measure T and what size should be used? The divided into N2a1 (no N1) and N2a2 (with N1) [38].
new pathology classification for adenocarcinoma [6, 16] proposed Incorporation of specific consequences related to the new path-
that AIS be classified as Tis (AIS) and that MIA be coded as T1mi. ology classification [6, 16] and, through that, recognising specific
For part-solid tumours the size of the invasive component should categories with a much higher incidence of mediastinal metasta-
be used to assign the T category for clinical staging; however, the ses, even if the tumour size is < 3 cm [20], remains to be
whole size of the tumour should also be recorded. Pathological confirmed.
staging might be challenging in the situation of lepidic growth, and A specific problem is whether it is necessary to evaluate the
therefore, interaction with radiology might be needed to score the possible existence of brain metastases by brain magnetic reson-
invasive (solid) component. The display is best with wide (lung) ance imaging (MRI). There is some controversy between existing
window settings, particularly in the case of subsolid lesions. For guidelines: The National Comprehensive Cancer Network

by guest
on 05 February 2018
Figure 1. Suggested algorithm for locoregional lymph node staging in patients with non-metastatic NSCLC.
CT, computed tomography; EBUS, endoscopic bronchial ultrasound; EUS, endoscopic ultrasound; FDG, fluorodeoxyglucose; LN, lymph node;
NPV, negative predictive value; NSCLC, non-small-cell lung cancer; PET, positron emission tomography; VAM, video-assisted mediastinoscopy.
Reprinted from [137] with permission.
(NCCN) advises this for all patients except for those with stage IA • If EBUS and/or EUS does not reveal nodal involvement in a situ-
[39], the BTS and the National Institute for Health and Care ation of high clinical suspicion, mediastinoscopy is indicated [I, A].
Excellence (NICE) for all patients considered for curative therapy • Mediastinoscopy is the test with the highest negative predict-
[40, 41], whereas the American College of Chest Physicians ive value to rule out mediastinal lymph node disease [I, A].
(ACCP) restricts it to stage III/IV and symptomatic patients [42]. • Screening for brain metastases by MRI might be useful in pa-
Whether this is cost-effective is unclear as the detection rate of tients considered for curative therapy [III, B].
brain metastases is very low [43].
Recommendations: Pretreatment risk assessment

• For part-solid tumours, the size of the invasive component should Any locoregional therapy must consider the pre-therapy situ-
be used to assign the T category for clinical staging [III, A]. ation of the patient but, even more importantly, their predicted
• Subsolid lesions need dedicated radiological expertise for post-treatment status. Most patients are older than 65 years
evaluating the lung lesion composition [V, A]. of age and may have age- and life-style-related comorbidity.
• If two lung lesions fulfil the criteria for two primaries these A therapeutic intervention for lung cancer will reduce the pul-
should be evaluated and treated accordingly [III, A]. monary and vascular reserve capacity, either acutely following re-
• For patients with abnormal mediastinal and/or hilar lymph section, or more gradually following radiotherapy (RT). This
nodes at CT and/or PET, endosonography is recommended functional loss needs to be estimated pre-therapy to determine
over surgical staging [I, A]. whether an individual patient is able to cope with it and to main-
• The preferred first technique for pathological confirmation of tain an acceptable quality of life.
suspect nodes is needle aspiration under EBUS and/or EUS For surgical candidates, algorithms for pretreatment evaluation
guidance [I, A]. have been developed and are used widely (Figure 3) [41, 44]. The

by guest
on 05 February 2018
by guest
on 05 February 2018
No enlarged N2 nodes
No enlarged LNs but central tumour or Extensive mediastinal
iv6 | Postmus et al.

and peripheral tumour hilar LNs N2 infiltration
Imaging
CT-Scan*
Enlarged discrete N2 LNs
Not required if negative Not required
Invasive
LNs on PET
LN Result
NO-N1 N2 N3
Surgery: Unresectable N2
of N2
unforeseen N2
Category
Potentially
resectable N2

Dedicated
multidisciplinary
assessment
**
Adjuvant chemotherapy Surgical multimodality Non-surgical

(radiotherapy) treatment multimodality treatment
Approach
Therapeutic
Figure 2. Treatment recommendations for patients with locoregional NSCLC, based on imaging, invasive lymph node staging tests and multidisciplinary assessment.
*Category description according to CT imaging as in ACCP staging document [42].
**See text for factors involved in the choice between non-surgical and surgical multimodality treatment.
ACCP, American College of Chest Physicians; CT, computed tomography; LN, lymph node; NSCLC, non-small-cell lung cancer; PET, positron-emission tomography.
Annals of Oncology

Cardiac assessment: FEV1 Both
low risk or DL,CO > 80 %
treated patient
Either one < 80%
< 35 % or Exercise testing > 75% or

<10 mL·kg–1·min–1 Peak VO2 > 20 mL·kg–1·min–1
35%-75% or
10-20 mL·kg–1·min–1
Split function
ppo-FEV1 Both > 30%
ppo-D L,CO
At least one < 30%
< 35% or ppo-peak VO2

< 10 mL·kg–1·min–1
> 35% or
> 10 mL·kg–1·min–1
Lobectomy or Resection up to Resection

pneumonectomy calculated extent up to
are usually
pneumonectomy
not recommended.
Consider other options
Figure 3. Preoperative respiratory evaluation.

DLCO, diffusing capacity of the lungs for carbon monoxide; FEV1, forced expiratory volume in 1 second; ppo, predicted postoperative;
VO2, oxygen consumption.
Reprinted from [50], with permission from the European Respiratory Society.
relative risk of postoperative morbidity and mortality can be pre- Evaluation of the cardiac risk assessment for lung resections
dicted from preoperative forced expiratory volume in the first se- by the recalibrated thoracic revised cardiac risk index (RCRI)
cond (FEV1) and diffusing capacity of the lung for carbon is recommended (Table 3) [47], as it has been validated in
monoxide (DLCO). Patients with lower values might benefit from this setting [48]. Schematic description of the steps to be
a more extensive assessment through pulmonary exercise testing. taken for evaluating these aspects is given in Figure 5 (the fig-
When maximal oxygen consumption (VO2max) is < 10 mL/kg/ ure is based on the original RCRI rather than the recalibrated
min, patients are potentially at high risk for serious postoperative RCRI).
complications [III, A]. Surgical resection is usually acceptable if Evaluating all these pros and cons should be done within a
the predicted postoperative FEV1 and DLCO values are > 40%. multidisciplinary team and in consultation with the patient.
This can be estimated from the number of bronchopulmonary seg- Concentration of expertise will certainly improve decision-
ments to be resected taking into account the regional distribution making and be of benefit for treatment outcomes [49].
of ventilation and perfusion. The problematic area is where no real Unfortunately, the predicted tolerance for high-dose RT is less
guidelines exist, or the standard is not directly applicable, as resec- well defined and it is, in general, impossible to accurately deter-
tion of poorly functioning parts of the lung might improve the mine the related acute and long-term risks [50]. Based on the
situation instead of making it worse (Figure 4) [45]. known adverse effects of RT on vasculature and cardiac function,
The risk of in-hospital death can be estimated by a scoring sys- the dose to the heart should be minimised during RT planning
tem such as Thoracoscore [46]; however, it was designed for a [51–53].
general population and its value for use in cancer patients is In general, it is necessary to evaluate and optimise any comor-
limited. bidities before planned surgery [41]; furthermore, trying to

by guest
on 05 February 2018
Clinical stage I
lung cancer
Reduced lung
function?
(FEV1 < 65%)
No Yes
Lobectomy with Evidence of

confidence emphysema?
(based on TLC, RV, FEV1/FVC, CT
w/ hyperexpansion, etc.)
No Yes
Consider restrictive Moderately impaired lung Severely impaired

pathology with higher function? lung function?
associated risks (ILD, IPF) (FEV1 30%-65%) (FEV1 < 30%)
Consider non-surgical
Consider anatomy of modalities
emphysema, location of (SBRT, RFA)
mass, LVRS candidacy
Non-ideal LVRS Ideal LVRS

candidate candidate
Calculate COPD index Resection of mass

Predict postop FEV1 with concomitant
Consider sublobar LVRS
resection
Figure 4. Algorithm for patients with clinical stage I lung cancer and limited pulmonary function due to emphysema.
CT, computed tomography; COPD, chronic obstructive pulmonary disease; FEV1, forced expiratory volume 1; FVC, forced vital capacity; ILD,
interstitial lung disease; IPF, idiopathic pulmonary fibrosis; LVRS, lung volume reduction surgery; RFA, radiofrequency ablation; RV, reserve vol-
ume; SBRT, stereotactic body radiotherapy; TLC, total lung capacity.
Reprinted from [45], with permission from Elsevier.
optimise a patient’s condition prior to surgery is beneficial, espe- • The risk of postoperative morbidity and mortality can be esti-
cially for those with a poor preoperative condition [54]. mated using risk-specific models, although none have been
validated in a cancer population [III, B].
Recommendations: • Before considering surgical resection, precise assessment of
• In non-metastatic NSCLC, the cardiopulmonary fitness of the cardiac and pulmonary function is necessary to estimate risk
patient will determine the choice of treatment [III, A]. of operative morbidity [III, A].

by guest
on 05 February 2018
Table 3. Recalibrated thoracic revised cardiac risk index (adapted from [47]) segmentectomy resulted in equivalent outcomes [57]. The same
outcome of segmentectomy and lobectomy was reported in patients
Points
with radiologically pure solid cT1a adenocarcinomas [58]. As dis-
Weighted factors cussed in the section on diagnosis and pathology, the different types
Ischaemic heart disease 1.5 of adenocarcinoma have evolved in differences in metastatic pat-
History of cerebrovascular disease 1.5 tern, recurrence and survival, and based on this, one might expect
Serum creatinine > 2 mg/dL 1 that a limited resection will be adequate in the least invasive sub-
Pneumonectomy planned 1.5 types [20, 21]. Currently two phase III studies (CALGB 140503, and
JCOG0802/WJOG4607L) are recruiting [59] and waiting to mature
Class groupings
after accrual had been reached [60], respectively.
A 0
Whether surgery should be done through standard open thora-
B 1–1.5
cotomy, or a video-assisted thoracoscopic surgery (VATS) proced-
C 2–2.5
D > 2.5
ure, is probably less important from oncological perspective [61],
since comparative margin clearance and nodal dissection can be
Ischaemic heart disease: history of myocardial infarction, history of posi- achieved. A point of concern might be the extent of lymph node
tive exercise test, current complaint of chest pain (myocardial ischae- staging [62]. For patients, the major benefit is the reduced postop-
mia), nitrate therapy, ECG with pathological Q waves. Cerebrovascular erative morbidity and mortality, resulting in improved quality of
disease: transient ischaemic attack, stroke. life and making VATS the more attractive approach [63].
ECG, electrocardiogram. The management of lymph nodes during surgery is mainly dic-
tated by the staging requirements for guaranteed ‘R0 resection’
status. This implies surgical evaluation of a minimum of six
nodes/stations, three of which should be mediastinal, including
• For cardiac assessment, use of recalibrated RCRI is recom- the sub-carinal station, with no metastases found in most cranial
mended [III, A]. resected nodes [64]. While in stage I cases, overall survival (OS),
• Formal lung function testing should be undertaken to estimate local recurrence rate and distant metastasis do not appear to be
postoperative lung function. For patients with FEV1 and influenced by the method of lymph node assessment, systematic
DLCO values > 80% of their predicted pulmonary function nodal dissection is recommended in stages II and IIIA [65].
tests and no other major comorbidities, no further investiga- Intraoperative nodal management may be influenced by the ex-
tions are advised before surgical resection [III, A]. For others, tent of preoperative lymph node mapping, particularly prior
exercise testing and split lung function are recommended. In negative mediastinoscopy.
these patients, VO2max can be used to measure exercise cap- Patients presenting with multiple primaries should be assessed
acity and predict postoperative complications [III, A]. with curative intent. Complete resection is recommended, but
• Comorbidities should be evaluated and optimised before sur- combinations of resection and SABR have been found to be ef-
gery [III, A]. fective as well [66, 67].
• In patients with limited pulmonary function due to emphysema,
a lung volume reduction effect may be observed by resection of Recommendations:
the lung cancer within emphysematous lung tissue [III, B]. • Surgery should be offered to all patients with stage I and II
NSCLC as the preferred treatment to all who are willing to
accept procedure-related risks [III, A].
Treatment of early stages (stages I and II) • For patients with a non-centrally located resectable tumour
and absence of nodal metastasis on both CT and PET images,
Surgery surgical resection is recommended [I, A].
• Anatomical resection is preferred over wedge resection [I, A].
The cornerstone of treatment of potentially resectable lung cancer • Anatomical segmentectomy is generally considered accept-
is surgical removal of the tumour [55]. For those who are not will- able for pure GGO lesions or adenocarcinomas in situ or
ing to accept the risks, or are at very high risk, curative RT should with minimal invasion [III, B].
be offered, either SABR or hypofractionated high-dose RT. • Lobectomy is still considered the standard surgical treatment of tu-
Based on the Lung Cancer Study Group (LCSG) 821 trial, lobec- mours 2 cm in size that have a solid appearance on CT [II, B].
tomy is the current treatment of choice for T1 tumours as the local • Lymph node dissection should conform to IASLC specifica-
recurrence rate after a more limited resection (segmentectomy or tions for staging [III, A].
wedge resection) was found to be higher [56]. This study should be • Either open thoracotomy or VATS access can be carried out
viewed within the context that staging and surgical methods have as appropriate to the expertise of the surgeon [III, A].
progressed significantly since its publication more than two decades • VATS should be the approach of choice in stage I tumours
ago. Whether the conclusions are still applicable for smaller lesions [V, C].
(T1a) seems uncertain, research based on large databases suggest a • For patients with multifocal lung cancer, complete resection
(limited) practice change [57]. In squamous cell carcinoma, lobec- is recommended whenever possible. All patients with multi-
tomy is superior to segmentectomy or wedge resection. For adeno- focal lung cancer should be discussed in a multidisciplinary
carcinoma, wedge resection was inferior to lobectomy, whereas tumour board [III, B].

by guest
on 05 February 2018
RCRI: at least 3 weighted

factors* or
1) Any cardiac condition History
requiring medications Physical examination
2) A newly suspected cardiac Baseline ECG
condition Calculate RCRI
3) Inability to climb two flights
of stairs
Yes No
Cardiac consulation with non-invasive

cardiac testing treatments as per
AHA/ACC guidelines
Need for coronary *RCRI weighted factors [138]:

intervention
(CABG or PCI) • High risk surgery (including
lobectomy or pneumonectomy)
Continue with ongoing cardiac care • Ischaemic heart disease
Institute any needed new medical (prior myocardial infarction,
angina pectoris)
interventions (i.e. beta-blockers,
anticoagulants or statins) • Heart failure
• Insulin-dependent diabetes
• Previous stroke or TIA
• Creatinine ≥ 2 mg.dL–1
Postpone surgery Lung function tests

for ≥ 6 weeks (see Figure 3)
Figure 5. Preoperative cardiac evaluation.

AHA/ACC, American Heart Association/American College of Cardiology; CABG, coronary artery bypass grafting; ECG, electrocardiogram; PCI,
percutaneous coronary intervention; RCRI, revised cardiac risk index; TIA, transient ischaemic attack.
Reprinted from [50], with permission from the European Respiratory Society.
Systemic therapy was found [68], a subgroup analysis indicated it was mainly due to
the outcome in patients with tumours > 4 cm [72, 73].
In a period of about two decades, it has become clear that adjuvant
Neoadjuvant ChT has not been evaluated as extensively as
ChT is of benefit for patients with N1 and N2 disease (stage II and
postoperative. However, comparing outcomes of both modalities
III), resulting overall in 4%–5% absolute survival improvement at
5 years [68]. These results were obtained by administering cisplatin- did not reveal a major difference in OS [74, 75]. Its use might be
based doublets, delivering at least 300 mg/m2 of cisplatin in three to beneficial as downstaging might be achieved [76], potentially
four cycles. Although for the accompanying drug, most data are avail- resulting in a less extensive resection.
able for the efficacy of vinorelbine, this does by no means exclude Predictive molecular markers have not been evaluated in pro-
newer agents, with at least comparable efficacy, such as docetaxel, spective studies. For cases with mutation in epidermal growth fac-
gemcitabine or pemetrexed. However, adding bevacizumab was not tor receptor (EGFR) there is limited evidence coming from a
beneficial [69, 70]. Patient selection criteria for these studies, such as meta-analysis [77], two major trials are currently recruiting to an-
proper recovery from surgery and the absence of major comorbid- swer this important question [78, 79]. Until these outcomes be-
ities, are essential. Although in most studies the interval between sur- come available, targeted agents should not be used in the adjuvant
gery and the start of ChT was restricted to 6 weeks, a recent analysis setting. Adjuvant immunotherapy trials using anti-PD-1 and anti
of the National Cancer Database showed a comparable outcome in PD-L1 checkpoint inhibitors in stage I-III adjuvant setting (trials
patients treated after a longer interval post-resection [71]. NCT NCT02504372 and NCT02273375) are ongoing. A neoadju-
Its value in lower stages is less clear. For stage IA, postoperative vant trial using anti-CTLA4 and anti-PD-1 in stage I-III neoadju-
ChT resulted in a worse outcome. In stage IB, a small overall benefit vant setting has been also recently initiated (NCT02998528).

by guest
on 05 February 2018
Recommendations: With the introduction of SABR for operable stage I tumours, a new
• Adjuvant ChT should be offered to patients with resected stage problem arises when recurrence of these tumours is detected during
II and III NSCLC [I, A] and can be considered in patients with follow-up. In those with proven recurrence (or a high suspicion), the
resected stage IB disease and a primary tumour > 4 cm [II, B]. possibility of salvage surgery should be considered [98–105].
Pre-existing comorbidity, time from surgery and postoperative The IASLC has defined ‘central tumours’ as tumours located
recovery need to be taken into account in this decision taken within 2 cm in all directions of any mediastinal critical structure,
in a multidisciplinary tumour board [V, A]. including the bronchial tree, oesophagus, heart, brachial plexus,
• For adjuvant ChT, a two-drug combination with cisplatin is pref- major vessels, spinal cord, phrenic nerve and recurrent laryngeal
erable [I, A]. In randomised studies, the attempted cumulative cis- nerve [106]. For tumours located in the hilar region, SABR using
platin dose was up to 300 mg/m2, delivered in three to four cycles. ‘risk-adapted’ fractionation schemes can achieve high local control
The most frequently studied regimen is cisplatin–vinorelbine. rates with limited toxicity [107]. However, care should be taken to
• At the present time, the choice of adjuvant therapy should distinguish moderately central tumours from so-called ‘ultracen-
not be guided by molecular analyses, e.g. ERCC1 mutation tral’ lesions, a term used to describe a planning target volume that
testing [IV, B]. overlaps the trachea or main bronchi [108]. SABR is not appropri-
• In the current state of knowledge, targeted agents should not ate for ultracentral tumours, as increased toxicity has already been
be used in the adjuvant setting [II, A]. reported for this subgroup, after conventional and hypo-
• In view of the equivalence of neoadjuvant and adjuvant ChT fractionated RT schemes. Data from a completed prospective
for OS, the consistent results and broad evidence base sup- Radiation Therapy Oncology Group (RTOG) study of SABR for
port adjuvant ChT as the timing of choice [II, C]. moderately central tumour are expected in the near future, and
• (Neo)adjuvant anti-PD(L)-1 checkpoint inhibitors are cur- until such time, a radical RT scheme using hypofractionated sched-
rently being evaluated in addition to current standard of care. ules can be considered an acceptable standard of care [89, 90].
Whether incorporating the new pathology classification [16], and
the possible pretherapy detection of less invasive types [25, 26],
Primary radiotherapy would change recommendations for subgroups remains to be seen.
For patients with comorbidities or other reasons for inoperabil-
ity, presenting with a peripherally located stage I NSCLC, or any Recommendations:
patient refusing surgery, stereotactic radiotherapy [SABR or ster- • The non-surgical treatment of choice for stage I NSCLC
eotactic body radiotherapy (SBRT)] is the preferred treatment, is SABR. The dose should be to a biologically equivalent tu-
with local control rates of 90% at 5 years [80, 81]. mour dose of 100 Gy, prescribed to the encompassing iso-
Current SABR practice generally utilises small planning mar- dose [III, A].
gins based on 4-dimensional CT (4DCT), multiple radiation • SABR for early-stage peripheral lung tumours is associated
beams or arcs, all of which reduce the risk of normal organ tox- with low toxicity in patients with chronic obstructive pul-
icity [82]. Acute treatment-related toxicity is uncommon, as de- monary disease (COPD) and the elderly [III, A].
terioration in quality of life [83]; however, the risk of high-grade • Salvage surgery, if feasible, may be offered to patients having
and fatal toxicity is high in patients with pre-existing interstitial complications post-SABR [V, B].
lung fibrosis and careful evaluation of the risks and benefits of the • Salvage surgery, if feasible, may be offered, using the same in-
procedure by an expert tumour board is advised [84, 85]. dications as for primary surgery in progressive disease after
Late toxicities reported in phase II trials include rib fractures SABR, but surgery may be more difficult with higher opera-
[86], dyspnoea and ventricular tachycardia [80, 87]. tive risk [V, B].
In elderly patients, the introduction of SABR led to an im- • For medically inoperable patients with tumours with a size > 5
provement in population-based survivals of patients with periph- cm and/or moderately central location, radical RT using more
erally located stage I, as well as a reduction of the number of conventional or accelerated schedules is recommended [III, A].
untreated patients [88]. When SABR is unavailable, radical RT
using hypofractionated schedules is preferred to the use of con-
ventionally fractionated RT [89, 90].
Radiofrequency ablation
Despite the available data on outcomes of SABR in patients with Fortunately, not many patients have contraindications for both
peripheral stage I tumours who are fit to undergo surgery [91, 92], surgery and SABR [85]. For these patients radiofrequency abla-
there is currently no evidence to routinely recommend SABR for pation (RFA) might be a reasonable alternative although the level of
tients who are at low risk for surgical complications. Three randomised evidence comes from observational studies only [109].
clinical trials in this population failed to complete accrual, and results
from four new trials will be forthcoming in the coming decade [93]. A Recommendation:
pooled analysis of two of the closed trials, the STARS and ROSEL stud- • Stage I NSCLC patients with strong contraindications for sur-
ies, revealed comparable recurrence-free survival at 3 years [94]. Given gery and/or SABR may be treated with RFA [V, C].
the differences in early toxicity and quality of life between surgery and
SABR, as well as the growing emphasis on patient reported endpoints
when evaluating new treatments [95], more attention should be given
Postoperative radiotherapy
towards developing tools for shared decision-making, as it may assist In a meta-analysis of rather old studies postoperative radiother-
operable patients and their clinicians to define a management plan apy (PORT) was found to be detrimental if given to patients with
that is consistent with a patient’s preferences and values [96, 97]. N0 and N1 disease [110]. The case for unexpected N2 disease

by guest
on 05 February 2018
discovered at surgery is less clear, and currently evaluated in a The treatment of resectable LA-NSCLC remains a matter of de-
large clinical trial, applying 54 Gy in 27–30 fractions [111]. The bate. There is only one trial comparing the two locoregional
use of PORT after an R1 resection appears reasonable, but it is modalities head-to-head, surgery and RT (60 Gy), in patients
not supported by high-quality evidence. with at least a minimal tumour response [113], no difference in
survival was found. In the Lung Intergroup Trial 0139, the induc-
Recommendations: tion regimen of chemoradiotherapy (CRT) (45 Gy), was followed
• PORT in completely resected early-stage NSCLC is not rec- by surgery or definitive RT to a dose of 61 Gy [114]. No signifi-
ommended [I, A]. cant difference in OS was found, but disease-free survival was sig-
• In case of R1 resection (positive resection margin, chest wall), nificantly better in the trimodality arm. An explanation for this
PORT should be considered [IV, B]. difference is the higher early toxic death rate in the surgery arm,
• Even if such patients were not included in randomised, clin- apparently due to the higher number of early postoperative
ical trials (RCTs), adjuvant ChT should be considered in pa- deaths in the group of patients undergoing right-sided pneumon-
tients with R1 resection of stage IB disease and a primary ectomy. Excluding pneumonectomy for an unplanned subgroup
tumour > 4 cm, stage II and III [V, A]. analysis of matched surgical patients treated by lobectomy, the
• In case both ChT and RT are administered post-surgery, RT surgical patients had a better survival. Two more recent studies
should be administered after ChT [V, C]. confirmed the outcomes with regard to disease-free survival and
OS after induction therapy followed by surgery. The SAKK study
failed to show benefit by adding relatively low doses of RT
(45 Gy) to ChT [115], whereas the ESPATUE trial confirmed that
Treatment of locally advanced stage (stage III) CRT (45 Gy) followed by surgery, is as good as CRT with defini-
Adequate staging through PET-CT imaging is indicated to rule tive RT (65–71 Gy) given as a boost in the last week of CRT [116].
out extracranial metastasis. Evaluation of the brain by MRI is As these studies showed no clear benefit for one of the local
indicated. therapies over the other, the choice of local treatment modality
Platinum-based ChT is an essential part of the treatment of lo- can vary across countries and centres.
cally advanced NSCLC (LA-NSCLC) as it improves survival in
Recommendations:
tumours considered resectable, as well in unresectable tumours.
• If, despite adequate mediastinal staging procedures, N2 dis-
Recommendations: ease is only documented intra-operatively, surgery should be
followed by adjuvant ChT [I, A]. In case of complete resec-
• All patients planned for definitive stage III NSCLC treatment
tion, addition of PORT is not routinely recommended, but
should undergo a diagnostic contrast-enhanced CT scan of the
may be an option following individual risk assessment [V, C].
chest and upper abdomen followed by a PET or a combined • If single station N2 disease can be demonstrated by preoperative
PET-CT with a CT technique with adequately high resolution
pathological nodal analysis, resection followed by adjuvant ChT,
for initial staging purposes [I, A] in order to rule out detectable
induction ChT followed by surgery or induction CRT followed
extrathoracic, extracranial metastasis, and to assess potential
by surgery are options. If induction ChT alone is given preopera-
mediastinal lymph node involvement, ideally within 4 weeks
tively, PORT is not standard treatment, but may be an option
before the start of treatment [III, B]. Single PET-positive distant
based on critical evaluation of locoregional relapse risks [IV, C].
lesions need pathological confirmation [V, B]. • In multistation N2 or N3, concurrent definitive CRT is preferred
• For patients with operable N2 disease, pathological staging of
[I, A]. An experienced multidisciplinary team is of paramount
the mediastinum is advised [III, C].
• All patients planned for curative stage III NSCLC treatment
importance in any complex multimodality treatment strategy de-
cision, including the role of surgery in these cases [IV, C].
should receive brain imaging for initial staging [III, B]. • In potentially resectable superior sulcus tumours, concurrent
Contrast-enhanced brain MRI is the preferred method for
CRT induction followed by definitive surgery is the treatment of
staging of the brain in stage III disease [III, A]. If it is not pos-
choice [III, A]. The same strategy may be applied for potentially
sible to perform MRI, dedicated contrast-enhanced brain CT
resectable T3 or T4 central tumours in highly selected cases and
scan is advised [III, B].
experienced centres [III, B]. In both situations, surgery should
be carried out within 4 weeks after the end of RT [III, B].
Resectable LA-NSCLC
Resectable in this situation usually refers to the following situations: Systemic therapy
• single station N2 disease where other nodal stations have
Which ChT is optimal has not been investigated extensively. In
been biopsied and proved to be benign. Postoperative ChT
fact, information coming from studies in stage IV has hardly been
should then be advised [112];
applied in this situation, probably the only exception being the
• T4N0 tumours where nodal disease had been excluded by inva-
PROCLAIM study, evaluating the use of pemetrexed-cisplatin
sive methods when a R0 resection is considered to be feasible;
versus standard cisplatin-etoposide, but failing to show any im-
• after induction therapy, when there has been nodal down-
provement except for less haematological toxicity [117].
staging and a pneumonectomy can be avoided.
Consolidation ChT after CRT failed to improve progression-free
All such cases should be evaluated within an experienced survival (PFS) [118]. There is no beneficial role for induction
multidisciplinary team. ChT before CRT [119], although in many centres for practical

by guest
on 05 February 2018
reasons, related to planning of RT, one cycle will be given prior to on cisplatin. There are no firm conclusions supporting single-
concurrent CRT. Adjuvant immunotherapy trials, using anti- agent carboplatin as a radiation sensitiser [I, A].
PD-1 and anti-PD-L1 checkpoint inhibitors in stage I-III adju- • Most comparative studies of concurrent CRT versus sequen-
vant setting, as well as the combination of anti-CTLA4 and anti- tial administration were using cisplatin þ etoposide or cis-
PD-1 in stage I-III neoadjuvant setting, are ongoing. A consolida- platin þ vinca alkaloid (typically: cisplatin þ vinorelbine), or
tion trial using an anti-PD-L1 drug in consolidation after CRT cisplatin þ pemetrexed if non-squamous histology. There are
will deliver results very soon (NCT NCT02125461). no comparative phase III trials using the paclitaxel/carbopla-
tin regimen. When delivered perioperatively cisplatin-based
Recommendations: combinations are considered the treatment of choice, in the
• For curative-intent management, patients should be able to absence of contraindications [I, A].
undergo platinum-based ChT (preferably cisplatin) [I, A]. • In the stage III disease CRT strategy, two to four cycles of
• (Neo)adjuvant anti PD(L)-1 checkpoint inhibitors are cur- concomitant ChT should be delivered [I, A]. There is no evi-
rently being evaluated in addition to current standard of care. dence for further induction or consolidation ChT. In the
Checkpoints are also being evaluated after CRT as consolida- perioperative setting, three to four cycles of cisplatin-based
tion therapy. ChT are recommended [I, A], aiming at a total cumulative
dose of at least 300 mg/m2 of cisplatin [II, B].
• 60–66 Gy in 30–33 daily fractions is recommended for con-
Unresectable LA-NSCLC
current CRT [I, A]. Maximum overall treatment time should
Unresectable in this situation refers to the situation that—even not exceed 7 weeks [III, B]. ‘Biological intensification’, such
after induction therapy—a complete resection (R0) would not be as treatment acceleration, is not standard practice in concur-
possible, based on evaluation within a multidisciplinary team, rent CRT schedules [III, B].
including an experienced thoracic surgeon. • In sequential approaches, RT delivered in a short overall
Sequential CRT (induction ChT followed by RT), usually given treatment time is recommended [I, A].
at a dose of 60–66 Gy in 30–33 fractions over 6–7 weeks, was com-
pared to concurrent CRT at comparable doses in several phase III
trials and in a meta-analysis [120].
Concurrent CRT is considered the preferred treatment for pa-
tients who are fit, as it leads to higher 5-year survival rates, albeit at Although proven to be beneficial in stage IV patients with driving
the cost of a higher rate of reversible oesophagitis. In recent phase mutations, such as in EGFR or translocation of anaplastic lym-
III trials delivering concurrent CRT to doses between 60 and 66 Gy, phoma kinase (ALK), the role of targeted agents in stage I, II and III
the incidence of grade 3 or higher oesophagitis ranged from 7% to has not been evaluated properly. From the meta-analysis [77], no
21%, with corresponding rates of grade 3 or higher radiation pneu- conclusion can be drawn for adjuvant use of targeted therapy in
monitis ranging from 2.5% to 7% [51, 118]. Another area of con- EGFR mutated stage I-III NSCLC. The only study in which more
cern is the early mortality rate of 10% following concurrent CRT. staging details are given included only 36 patients with stage III;
Tumour volume and pulmonary function were found to be risk fac- however, details on outcome of those patients were not given [122].
tors associated with mortality in the first 180-day post-treatment in
a multi-institutional analysis of 1245 patients [121]. The use of radi- Recommendations:
ation doses in excess of 66 Gy is not recommended outside trials, as • There is currently no role for targeted agents in stage III
delivery of 74 Gy with concurrent CRT led to a poorer survival [51]. NSCLC outside clinical trials [I, A].
For elderly and/or less fit patients with clinically relevant • Immunotherapy is being studied in early NSCLC as (neo)ad-
comorbidities, the sequential approach is a reasonable choice juvant therapy and as consolidation after CRT; data should
[50]. An individual patient data meta-analysis of trials conducted be awaited before any clinical use [I, A].
prior to 2006 found that accelerated RT schedules which are de-
livered in a shorter overall treatment time led to an absolute bene-
fit of 2.5% in 5-year OS [89]. Based on this, accelerated RT
schedules delivering once-daily fractions of 2.6–3 Gy, to a total
Follow-up, long-term implications and
dose of up to 60–66 Gy, are recommended in patients who receive survivorship
either sequential CRT or RT alone for stage III NSCLC. NSCLC patients treated with radical intent are at risk of develop-
ing new cancer related problems, with potentially considerable
Recommendations: consequences and different dynamics over time:
• Concurrent CRT is the treatment of choice in patients eval-
• treatment-related complications, treatment of existing
uated as unresectable in stage IIIA and IIIB [I, A]. If concur-
comorbidities;
rent CRT is not possible—for any reason—sequential ChT
• detection of treatable relapse;
followed by definitive RT represents a valid and effective al-
• detection of second primaries.
ternative [I, A].
• There is no role for prophylactic cranial irradiation in stage In the early phase after lung cancer resection, readmission for
III NSCLC [II, A]. complications is not rare; 12.8% of patients listed in a large
• In the absence of contraindications, the optimal ChT to be Surveillance, Epidemiology, and End Results (SEER) programme
combined with radiation in stage III NSCLC should be based database were readmitted within 30 days after discharge shortly

by guest
on 05 February 2018
Incidence/epidemiology
Screening with LDCT reduces lung cancer-related mortality [I, A]. It is not yet ready for large-scale implementation, mainly because the lung cancer mor-
tality reduction rate lacks definite proof of a second study result, and partly because of remaining questions regarding definition of the at-risk population,
timing, interval and method of CT (especially 2D versus 3D evaluation), how to handle (false-) positive findings and especially cost-effectiveness, notably
in relation to smoking cessation [I, A].
LDCT screening can be carried out outside a clinical trial provided it is offered within a dedicated programme with quality control, in a centre with ex-
perience in CT screening, a large volume of thoracic oncology activity and multidisciplinary management of suspicious findings [I, B]. Candidates are cur-
rent or former heavy smokers ( 30 pack-years or 15 years since smoking cessation) aged 55–74 years, who are well informed about potential benefits
and risks. Individuals offered LDCT screening should be referred to a smoking cessation programme.
LDCT screening should not be offered on an ad hoc individual basis, but patients requesting screening should be referred to a dedicated programme, as
recommended above [V, B].
Other screening methods, such as chest X-ray, sputum analysis or biomarkers are not recommended for clinical use [I, C].
Diagnosis
In patients with clinical stages I–III lesions, a pretreatment pathological diagnosis is recommended prior to any curative treatment.
Bronchoscopy is the recommended test to obtain a pathological diagnosis of centrally located tumours in stages I–III with biopsy of any visible lesion [III, A].
The pathological classification NOS should be used only in cases where it is impossible to obtain enough tissue for further classification, or when steps
to further classify the tumour are inconclusive [V, A].
An exception to the requirement for a pretreatment diagnosis can be made if an experienced multidisciplinary group decides that the risks of obtaining
pathology may be unacceptable in a patient in whom the likelihood of malignancy is high based on clinical and imaging findings [III, B].
A pretreatment pathological diagnosis is strongly recommended for all patients before SABR, unless a multidisciplinary tumour board is of the opinion
that the risk-benefit ratio of the procedure is unacceptable. In such a situation, the predicted likelihood of malignancy should preferably be at least 85%,
based upon accepted criteria [III, B] [25].
The descriptive element of the recent WHO classification of adenocarcinoma subtypes should be used to describe bronchoscopic and CT-guided biop-
sies whenever possible [III, A].
The revised adenocarcinoma classification may identify patient subtypes for whom an anatomical sublobar resection, rather than lobectomy, would be
sufficient [III, A].
FDG-PET may contribute for the selection of patients for anatomical sublobar resections as low SUVmax values of peripheral tumours indicate lack of me-
diastinal metastases [III, A]. This diagnosis may be made intra-operatively by video-assisted thoracoscopic biopsy and frozen section analysis.
In isolated cases a diagnostic anatomical sublobar resection may be acceptable.
Solitary pulmonary nodule
The diagnostic approach to non-calcified pulmonary nodules should be based on existing standard guidelines [III, A], although new evidence on nodule
management is emerging.
Likelihood of malignancy based upon risk calculation methods used in CT screening studies should be used only to guide the clinical assessment of pul-
monary nodules detected in the wider population [V, C].
In non-metastatic NSCLC, detailed locoregional staging according to the 8th TNM staging system and the cardiopulmonary fitness of the patient deter-
mine the choice of treatment [III, A].
Locoregional staging
For part-solid tumours, the size of the invasive component should be used to assign the T category for clinical staging [III, A]
Subsolid lesions need dedicated radiological expertise for evaluating the lung lesion composition [V, A].
If two lung lesions fulfil the criteria for two primaries these should be evaluated and treated accordingly [III, A].
For patients with abnormal mediastinal and/or hilar lymph nodes at CT and/or PET imaging, endosonography is recommended over surgical staging [I, A].
The preferred first technique for pathological confirmation of suspect nodes is needle aspiration under EBUS and/or EUS guidance [I, A].
If EBUS and/or EUS does not reveal nodal involvement in a situation of high clinical suspicion, mediastinoscopy is indicated [I, A].
Mediastinoscopy is the test with the highest negative predictive value to rule out mediastinal lymph node disease [I, A].
Screening for brain metastases by MRI might be useful in patients considered for curative therapy [III, B].
Pretreatment risk assessment
In non-metastatic NSCLC, the cardiopulmonary fitness of the patient will determine the choice of treatment [III, A].
The risk of postoperative morbidity and mortality can be estimated using risk-specific models, although none have been validated in a cancer population [III, B].
Before considering surgical resection, precise assessment of cardiac and pulmonary function is necessary to estimate risk of operative morbidity [III, A].
For cardiac assessment, use of recalibrated RCRI is recommended [III, A].
Formal lung function testing should be undertaken to estimate postoperative lung function. For patients with FEV1 and DLCO values > 80% of their pre-
dicted pulmonary function tests and no other major comorbidities, no further investigations are advised before surgical resection [III, A]. For others, exer-
cise testing and split lung function are recommended. In these patients, VO2max can be used to measure exercise capacity and predict postoperative
complications [III, A].
Comorbidities should be evaluated and optimised before surgery [III, A].
In patients with limited pulmonary function due to emphysema, a lung volume reduction effect may be observed by resection of the lung cancer within
emphysematous lung tissue [III, B].
Continued

by guest
on 05 February 2018
Table 4. Continued
Treatment of early stages (stages I and II)
Surgery
Surgery should be offered to all patients with stage I and II NSCLC as the preferred treatment to all who are willing to accept procedure-related risks [III, A].
For patients with a non-centrally located resectable tumour and absence of nodal metastasis on both CT and PET images, surgical resection is recom-
mended [I, A].
Anatomical resection is preferred over wedge resection [I, A].
Anatomical segmentectomy is generally considered acceptable for pure GGO lesions or adenocarcinomas in situ or with minimal invasion [III, B].
Lobectomy is still considered the standard surgical treatment of tumours 2 cm in size that have a solid appearance on CT [II, B].
Lymph node dissection should conform to IASLC specifications for staging [III, A].
Either open thoracotomy or VATS access can be carried out as appropriate to the expertise of the surgeon [III, A].
VATS should be the approach of choice in stage I tumours [V, C].
For patients with multifocal lung cancer, complete resection is recommended whenever possible. All patients with multifocal lung cancer should be dis-
cussed in a multidisciplinary tumour board [III, B].
Systemic therapy
Adjuvant ChT should be offered to patients with resected stage II and III NSCLC [I, A] and can be considered in patients with resected stage IB disease
and a primary tumour > 4 cm [II, B]. Pre-existing comorbidity, time from surgery and postoperative recovery need to be taken into account in this deci-
sion taken in a multidisciplinary tumour board [V, A].
For adjuvant ChT, a two-drug combination with cisplatin is preferable [I, A]. In randomised studies, the attempted cumulative cisplatin dose was up to
300 mg/m2, delivered in three to four cycles. The most frequently studied regimen is cisplatin–vinorelbine.
At the present time, the choice of adjuvant therapy should not be guided by molecular analyses, e.g. ERCC1 mutation testing [IV, B].
In the current state of knowledge, targeted agents should not be used in the adjuvant setting [II, A].
In view of the equivalence of neoadjuvant and adjuvant ChT for OS, the consistent results and broad evidence base support adjuvant ChT as the timing of choice [II, C].
(Neo)adjuvant anti-PD(L)-1 checkpoint inhibitors are currently being evaluated in addition to current standard of care.
Primary radiotherapy
The non-surgical treatment of choice for stage I NSCLC is SABR. The dose should be to a biologically equivalent tumour dose of 100 Gy, prescribed to
the encompassing isodose [III, A].
SABR for early-stage peripheral lung tumours is associated with low toxicity in patients with COPD and the elderly [III, A].
Salvage surgery, if feasible, may be offered to patients having complications post-SABR [V, B].
Salvage surgery, if feasible, may be offered, using the same indications as for primary surgery in progressive disease after SABR, but surgery may be more
difficult with higher operative risk [V, B].
For medically inoperable patients with tumours with a size > 5 cm and/or moderately central location, radical RT using more conventional or accelerated
schedules is recommended [III, A].
Radiofrequency ablation
Stage I NSCLC patients with strong contraindications for surgery and/or SABR may be treated with RFA [V, C].
PORT in completely resected early-stage NSCLC is not recommended [I, A].
In case of R1 resection (positive resection margin, chest wall), PORT should be considered [IV, B].
Even if such patients were not included in RCTs, adjuvant ChT should be considered in patients with R1 resection of stage IB disease and a primary tu-
mour > 4 cm, stage II and III [V, A].
In case both ChT and RT are administered post-surgery, RT should be administered after ChT [V, C].
Treatment of locally advanced stage (stage III)
All patients planned for definitive stage III NSCLC treatment should undergo a diagnostic contrast-enhanced CT scan of the chest and upper abdomen
followed by a PET or a combined PET-CT with a CT technique with adequately high resolution for initial staging purposes [I, A] in order to rule out detect-
able extrathoracic, extracranial metastasis, and to assess potential mediastinal lymph node involvement, ideally within 4 weeks before the start of treat-
ment [III, B]. Single PET-positive distant lesions need pathological confirmation [V, B].
For patients with operable N2 disease, pathological staging of the mediastinum is advised [III, C].
All patients planned for curative stage III NSCLC treatment should receive brain imaging for initial staging [III, B]. Contrast-enhanced brain MRI is the preferred
method for staging of the brain in stage III disease [III, A]. If it is not possible to perform MRI, dedicated contrast-enhanced brain CT scan is advised [III, B].
Resectable LA-NSCLC
If, despite adequate mediastinal staging procedures, N2 disease is only documented intra-operatively, surgery should be followed by adjuvant ChT [I, A].
In case of complete resection, addition of PORT is not routinely recommended, but may be an option following individual risk assessment [V, C].
If single station N2 disease can be demonstrated by preoperative pathological nodal analysis, resection followed by adjuvant ChT, induction ChT fol-
lowed by surgery or induction CRT followed by surgery are options. If induction ChT alone is given preoperatively, PORT is not standard treatment, but
may be an option based on critical evaluation of locoregional relapse risks [IV, C].
In multistation N2 or N3, concurrent definitive CRT is preferred [I, A]. An experienced multidisciplinary team is of paramount importance in any complex
multimodality treatment strategy decision, including the role of surgery in these cases [IV, C].
In potentially resectable superior sulcus tumours, concurrent CRT induction followed by definitive surgery is the treatment of choice [III, A]. The same
strategy may be applied for potentially resectable T3 or T4 central tumours in highly selected cases and experienced centres [III, B]. In both situations,
surgery should be carried out within 4 weeks after the end of RT [III, B].
Continued

by guest
on 05 February 2018
Table 4. Continued
Systemic therapy
For curative-intent management, patients should be able to undergo platinum-based ChT (preferably cisplatin) [I, A].
(Neo)adjuvant anti PD(L)-1 checkpoint inhibitors are currently being evaluated in addition to current standard of care.
Checkpoints are also being evaluated after CRT as consolidation therapy.
Unresectable LA-NSCLC
Concurrent CRT is the treatment of choice in patients evaluated as unresectable in stage IIIA and IIIB [I, A]. If concurrent CRT is not possible—for any
reason—sequential ChT followed by definitive RT represents a valid and effective alternative [I, A].
There is no role for prophylactic cranial irradiation in stage III NSCLC [II, A].
In the absence of contraindications, the optimal ChT to be combined with radiation in stage III NSCLC should be based on cisplatin. There are no firm
conclusions supporting single-agent carboplatin as a radiation sensitiser [I, A].
Most comparative studies of concurrent CRT versus sequential administration were using cisplatin þ etoposide or cisplatin þ vinca alkaloid (typically: cis-
platin þ vinorelbine), or cisplatin þ pemetrexed if non-squamous histology. There are no comparative phase III trials using the paclitaxel/carboplatin regi-
men. When delivered perioperatively cisplatin-based combinations are considered the treatment of choice, in the absence of contraindications [I, A].
In the stage III disease CRT strategy, two to four cycles of concomitant ChT should be delivered [I, A]. There is no evidence for further induction or con-
solidation ChT. In the perioperative setting, three to four cycles of cisplatin-based ChT are recommended [I, A], aiming at a total cumulative dose of at
least 300 mg/m2 of cisplatin [II, B].
60–66 Gy in 30–33 daily fractions is recommended for concurrent CRT [I, A]. Maximum overall treatment time should not exceed 7 weeks [III, B].
‘Biological intensification’, such as treatment acceleration, is not standard practice in concurrent CRT schedules [III, B].
In sequential approaches, RT delivered in a short overall treatment time is recommended [I, A].
There is currently no role for targeted agents in stage III NSCLC outside clinical trials [I, A].
Immunotherapy is being studied in early NSCLC as (neo)adjuvant therapy and as consolidation after CRT; data should be awaited before any clinical use [I, A].
NSCLC patients treated with radical intent should be followed for treatment-related complications, detection of treatable relapse or occurrence of se-
cond primary lung cancer [III, A].
Surveillance every 6 months for 2 years with a visit including history, physical examination and—preferably contrast-enhanced—volume chest CT scan
at least at 12 and 24 months is recommended, and thereafter an annual visit including history, physical examination and chest CT scan in order to detect
second primary tumours [III, B].
For individual patients, follow-up with six-monthly CT scans for 3 years is recommended for patients who are suitable for salvage treatment (e.g. sur-
gery, local ablative therapy) [III, B]. The frequency of the follow-up visits can be tailored to the individual patient for those not suitable for salvage treat-
ment [V, B].
The selective use of FDG–PET is recommended when recurrence after SABR is suspected based on serial spiral chest CT [III, B].
Due to a high number of false-positive findings on PET, patients suitable for salvage therapy should undergo a biopsy, whenever possible [III, B].
NSCLC patients should be offered smoking cessation, as this leads to superior treatment outcomes. Combining behaviour techniques with pharmaco-
therapy is the preferred approach [I, A].
2D, 2 dimensional; 3D, 3 dimensional; ChT, chemotherapy; COPD, chronic obstructive pulmonary disease; CRT, chemoradiotherapy; CT, computed tomography;
DLCO, diffusing capacity of the lungs for carbon monoxide; EBUS, endobronchial ultrasound; EUS, endoscopic ultrasound; FDG–PET, fluorodeoxyglucose posi-
tron emission tomography; FEV1, forced expiratory volume in 1 second; GGO, ground glass opacity; IASLC, International Association for the Study of Lung
Cancer; LA-NSCLC, locally advanced NSCLC; LDCT, low-dose CT; MRI, magnetic resonance imaging; NSCLC, non-small-cell lung cancer; NOS, not otherwise
specified; OS, overall survival; PET, positron emission tomography; PORT, postoperative radiotherapy; RCRI, revised cardiac risk index; RCT, randomised controlled
trial; RFA, radiofrequency ablation; RT, radiotherapy; SABR, stereotactic ablative radiotherapy; SUVmax, maximum standardised uptake value; TNM, tumour, node
and metastasis; VATS, video-assisted thoracoscopic surgery; VO2max, maximal oxygen consumption; WHO, World Health Organization.
after the resection; reasons were respiratory insufficiency, pneu- In a large group of resected patients, standardised follow-up re-
monia, pneumothorax and cardiac complications. Patient factors vealed that during the first 4 years after surgery, the risk of recur-
associated with readmission were resection type, age, prior induc- rence ranged from 6% to 10% per person per year, but decreased
tion CRT and preoperative comorbidities, including congestive thereafter to 2% [126]. Within this period a pattern can be recog-
heart failure and COPD. The 90-day mortality in those readmitted nised, during the first and second year recurrence is mainly local
at 30 days is 6-fold that of those not readmitted. This emphasises and rare thereafter, whereas at the end of the second year until the
the need for adequate care and more intense early follow-up in pa- end of the fourth year, recurrence is dominated by distant meta-
tients at risk of developing postoperative problems [123]. Overall, stases decreasing over time [127]. After 5 years, these are virtually
the 90-day mortality is nearly double the 30-day mortality, with a absent. The risk of developing a second primary lung cancer ex-
considerable difference between low and high-volume hospitals hibits a more uniform pattern over time, ranging from 1% to 6%
[124]. Overall, these patients have a significant excess conditional per person per year and did not diminish over time [126, 128].
mortality with an—increasing over time—relative contribution of This is not restricted to cancers developing in smokers but was
cardiovascular and respiratory co-morbidity [125]. observed at a comparable magnitude in non-smokers [129].

by guest
on 05 February 2018
Table 5. Table of levels of evidence and grades of recommendation (adapted from the Infectious Diseases Society of America-United States Public Health
Levels of evidence
C Insufficient evidence for efficacy or benefit does not outweigh the risk or the disadvantages (adverse events and costs, ...), optional
a
Surveillance after treatment with curative intent is only useful Recommendations:

if detection of a recurrence, locally or distant, or detection of a • NSCLC patients treated with radical intent should be followed
metachronous primary will result in potentially life-prolonging for treatment-related complications, detection of treatable re-
or preferable curative therapy. Curative therapy after a local relapse or occurrence of second primary lung cancer [III, A].
currence is often not possible, resulting only in 5-year survival • Surveillance every 6 months for 2 years with a visit including
rates of 15% [130, 131]. history, physical examination and—preferably contrast-
For second primaries, the outcome is better with 5-year sur- enhanced—volume chest CT scan at least at 12 and
vival rates ranging from 25% to 60% [132, 133]. 24 months is recommended, and thereafter an annual visit
This illustrates that detection of local recurrence or a meta- including history, physical examination and chest CT scan in
chronous primary may lead to therapy resulting in long-term disorder to detect second primary tumours [III, B].
ease-free survival. Therefore, regular screening for both is likely • For individual patients, follow-up with six-monthly CT scans
to be worthwhile. for 3 years is recommended for patients who are suitable for
There are no prospective trials evaluating what will be the most salvage treatment (e.g. surgery, local ablative therapy) [III, B].
optimal follow-up after surgery. As most local relapses will be The frequency of the follow-up visits can be tailored to the indi-
seen during the first two years after treatment, a follow-up visit vidual patient for those not suitable for salvage treatment [V, B].
every 6 months is recommended during that period, and annually • The selective use of FDG–PET is recommended when recur-
thereafter. A new finding detected through history, physical rence after SABR is suspected based on serial spiral chest CT
examination and/or imaging (preferably CT) usually needs to be scan [III, B].
discussed in an experienced multidisciplinary team taking into • Due to a high number of false-positive findings on PET, pa-
account that a new finding could be a treatment complication, a tients suitable for salvage therapy should undergo a biopsy,
metastasis or a new primary. whenever possible [III, B].
For patients initially treated with SABR, the late local recur- • NSCLC patients should be offered smoking cessation, as this leads
rences can be observed for up to 5 years post-treatment, and the to superior treatment outcomes. Combining behaviour techniques
incidence of second primary lung tumours appears to be similar with pharmacotherapy is the preferred approach [I, A].
to that post-surgery [81, 134]. As it may be sometimes difficult
to distinguish post-SABR recurrences from focal fibrosis, high-
risk radiological features have been identified [135] and the use
of such a scheme has recently been independently validated Methodology
[136]. These Clinical Practice Guidelines were developed in accordance
Patients who have undergone ChT and RT for stage III with the ESMO standard operating procedures for Clinical
NSCLC, are at high risk of developing progressive disease, either Practice Guidelines development http://www.esmo.org/
locally or at metastatic sites. Establishing locoregional disease Guidelines/ESMO-Guidelines-Methodology. The relevant litera-
progression is often a diagnostic challenge, but this is important ture has been selected by the expert authors. A summary of rec-
in patients who may be fit for salvage treatments [98–105]. ommendations is shown in Table 4. Levels of evidence and grades
Smoking cessation is crucial in all lung cancer patients treated of recommendation have been applied using the system shown in
with curative intent, and patients should be offered support to Table 5. Statements without grading were considered justified
achieve this goal. standard clinical practice by the experts and the ESMO Faculty.

by guest
on 05 February 2018
This manuscript has been subjected to an anonymous peer review 14. Novello S, Barlesi F, Califano R et al. Metastatic non-small-cell lung
process. cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and
follow-up. Ann Oncol 2016; 27 (Suppl 5): v1–v27.
15. Navani N, Brown JN, Nankivell M et al. Suitability of endobronchial
ultrasound-guided transbronchial needle aspiration specimens for sub-
Disclosure typing and genotyping of non-small cell lung cancer. Am J Respir Crit
Care Med 2012; 185: 1316–1322.
PEP has reported advisory boards for Bristol-Myers Squibb, 16. Travis WD, Brambilla E, Nicholson AG et al. The 2015 World Health
Boehringer Ingelheim, Novartis, AstraZeneca, Janssen Organization Classification of Lung Tumours: impact of genetic, clin-
Pharmaceuticals, Merck Sharp & Dohme and Roche; received hon- ical and radiologic advances since the 2004 classification. J Thor Oncol
oraria from Roche and travel grants from Merck Sharp & Dohme, 2015; 10: 1243–1260.
17. Warth A, Muley T, Meister M et al. The novel histologic International
Boehringer Ingelheim, Pfizer and Celgene; KMK has reported lec-
Association for the Study of Lung Cancer/American Thoracic Society/
ture honoraria and/or consultancy fees from AstraZeneca, Bristol- European Respiratory Society classification system of lung adenocarcin-
Myers Squibb, Boehringer Ingelheim, Eli Lilly, Merck KGaA, Merck oma is a stage-independent predictor of survival. J Clin Oncol 2012; 30:
Sharpe & Dohme, Novartis, Pfizer, Roche and Roche Diagnostics; 1438–1446.
SS has reported advisory boards for Lilly Oncology and research 18. Hung JJ, Yeh YC, Jeng WJ et al. Predictive value of the International
sponsored by Varian Medical Systems; JV has reported advisory Association for the Study of Lung Cancer/American Thoracic Society/
boards, consulting and honoraria from Merck Sharp & Dohme, European Respiratory Society classification of lung adenocarcinoma in
tumor recurrence and patient survival. J Clin Oncol 2014; 32: 2357–2364.
Boehringer, Eli Lilly, AstraZeneca and Novartis; MO, DW, CE and
19. Tsao MS, Marguet S, Le Teuff G et al. Subtype classification of lung
SP have reported no conflicts of interest. adenocarcinoma predicts benefit from adjuvant chemotherapy in pa-
tients undergoing complete resection. J Clin Oncol 2015; 33: 3439–3446.
20. Yu Y, Jian H, Shen L et al. Lymph node involvement influenced by lung
References adenocarcinoma subtypes in tumour size 3 cm disease: a study of
1. Centers for Disease Control and Prevention, Lung Cancer. https://www. 2268 cases. Eur J Surg Oncol 2016; 42: 1714–1719.
cdc.gov/cancer/lung/basic_info/ (18 May 2017, date last accessed). 21. Liu S, Wang R, Zhang Y et al. Precise diagnosis of intraoperative frozen
2. Sun S, Schiller JH, Gazdar AF. Lung cancer in never smokers—a differ- section is an effective method to guide resection strategy for peripheral
ent disease. Nat Rev Cancer 2007; 7: 778–790. small-sized lung adenocarcinoma. J Clin Oncol 2016; 34: 307–313.
3. Aberle DR, Adams AM, Berg CD et al. Reduced lung-cancer mortality with 22. Walts AE, Marchevsky AM. Root cause analysis of problems in the fro-
low-dose computed tomographic screening. N Engl J Med 2011; 365: 395–409. zen section diagnosis of in situ, minimally invasive, and invasive adeno-
4. Veronesi G, Maisonneuve P, Bellomi M et al. Estimating overdiagnosis carcinoma of the lung. Arch Pathol Lab Med 2012; 136: 1515–1521.
in low-dose computed tomography screening for lung cancer: a cohort 23. Yeh YC, Nitadori J, Kadota K et al. Using frozen section to identify
study. Ann Intern Med 2012; 157: 776–784. histological patterns in stage I lung adenocarcinoma of 3 cm: accur-
5. Patz EF Jr, Pinsky P, Gatsonis C et al. Overdiagnosis in low-dose com- acy and interobserver agreement. Histopathology 2015; 66: 922–938.
puted tomography screening for lung cancer. JAMA Intern Med 2014; 24. Donington JS. An additional step toward personalization of surgical care
174: 269–274. for early-stage non–small-cell lung cancer. J Clin Oncol 2016; 34: 295–296.
6. Travis WD, Brambilla E, Noguchi M et al. International Association for 25. Nakamura H, Saji H, Marushima H et al. Standardized uptake values in the
the Study of Lung Cancer/American Thoracic Society/European primary lesions of non-small-cell lung cancer in FDG-PET/CT can predict
Respiratory Society international multidisciplinary classification of lung regional lymph node metastases. Ann Surg Oncol 2015; 22: S1388–S1393.
adenocarcinoma. J Thorac Oncol 2011; 6: 244–285. 26. Matsuzawa R, Kirita K, Kuwata T et al. Factors influencing the concord-
7. Franklin WA, Merrick DT, Achcar RD, Aberle DR. Reclassification of ance of histological subtype diagnosis from biopsy and resected speci-
lung cancers detected by CT imaging in the American College of mens of lung adenocarcinoma. Lung Cancer 2016; 94: 1–6.
Radiology Imaging Network National Lung Screening Trial. J Thor 27. Callister ME, Baldwin DR, Akram AR et al. British Thoracic Society
Oncol 2015; 10 (9, Suppl 2): S220. oral24.05. Guidelines for the investigation and management of pulmonary nod-
8. Wiener RS, Gould MK, Arenberg DA et al. An official American Thoracic ules. Thorax 2015; 70: ii1–ii54.
Society/American College of Chest Physicians policy statement: imple- 28. MacMahon H, Naidich P, Goo JM et al. Guidelines for management of
mentation of low-dose computed tomography lung cancer screening pro- incidental pulmonary nodules detected on CT images: from the
grams in clinical practice. Am J Respir Crit Care Med 2015; 192: 881–891. Fleischner Society 2017. Radiology 2017; 284: 228–243.
9. Begnaud A, Hall T, Allen T. Lung cancer screening with low-dose CT: 29. Bai C, Choi CM, Chu CM et al. Evaluation of pulmonary nodules: clin-
implementation amid changing public policy at one health care system. ical practice consensus guidelines for Asia. Chest 2016; 150: 877–893.
Am Soc Clin Oncol Educ Book 2016; 35: e468–e475. 30. Brierley JD, Gospodarowicz MK and Wittekind C. (eds). TNM
10. van Iersel CA, de Koning HJ, Draisma G et al. Risk-based selection from Classification of Malignant Tumours, 8th edition. Oxford: John Wiley &
the general population in a screening trial: selection criteria, recruit- Sons, Inc. 2016.
ment and power for the Dutch-Belgian randomised lung cancer multi- 31. Rami-Porta R, Bolejack V, Crowley J et al. The IASLC lung cancer stag-
slice CT screening trial (NELSON). Int J Cancer 2007; 120: 868–874. ing project: proposals for the revisions of the T descriptors in the forth-
11. Horeweg N, van Rosmalen J, Heuvelmans MA et al. Lung cancer prob- coming eighth edition of the TNM classification for lung cancer. J Thor
ability in patients with CT-detected pulmonary nodules: a prespecified Oncol 2015; 10: 990–1003.
analysis of data from the NELSON trial of low-dose CT screening. 32. Travis WD, Asamura H, Bankier AA et al. The IASLC lung cancer stag-
Lancet Oncol 2014; 15: 1332–1341. ing project: proposals for coding T categories for subsolid nodules and
12. Walter JE, Heuvelmans MA, de Jong PA et al. Occurrence and lung can- assessment of tumor size in part-solid tumors in the forthcoming eighth
cer probability of new solid nodules at incidence screening with low- edition of the TNM classification of lung cancer. J Thorac Oncol 2016;
dose CT: analysis of data from the randomised, controlled NELSON 11: 1204–1223.
trial. Lancet Oncol 2016; 17: 907–916. 33. Detterbeck FC, Bolejack V, Arenberg DA et al. The IASLC lung cancer stag-
13. Yousaf-Khan U, van der Aalst C, de Joing PA et al. Risk stratification ing project: background data and proposals for the classification of lung
based on screening history: the NELSON lung cancer screening study. cancer with separate tumor nodules in the forthcoming edition of the
Thorax 2017; 72: 48–56. TNM classification for lung cancer. J Thor Oncol 2016; 11: 681–692.

by guest
on 05 February 2018
34. Detterbeck FC, Franklin WA, Nicholson AG et al. The IASLC lung can- 52. Hong JC, Salama JK. Dose escalation for unresectable locally advanced
cer staging project: background data and proposed criteria to distin- non-small cell lung cancer: end of the line?. Transl Lung Cancer Res
guish separate primary lung cancers from metastatic foci in patients 2016; 5: 126–133.
with two lung tumors in the forthcoming eighth edition of the TNM 53. Tucker SL, Liu A, Gomez D et al. Impact of heart and lung dose on early
classification for lung cancer. J Thor Oncol 2016; 11: 651–665. survival in patients with non-small cell lung cancer treated with chemo-
35. Detterbeck FC, Nicholson AG, Franklin WA et al. The IASLC lung can- radiation. Radiother Oncol 2016; 119: 495–500.
cer staging project: summary of proposals for revisions of the classifica- 54. Harada H, Yamashita Y, Misumi K et al. Multidisciplinary team-based
tion of lung cancers with multiple pulmonary sites of involvement in approach for comprehensive preoperative pulmonary rehabilitation
the forthcoming eighth edition of the TNM classification. J Thor Oncol including intensive nutritional support for lung cancer patients. PLoS
2016; 11: 639–650. One 2013; 8: e59566.
36. Detterbeck FC, Marom EM, Arenberg DA et al. The IASLC lung cancer 55. Rosen JE, Keshava HB, Yao X et al. The natural history of operable non-
staging project: background data and proposals for the application of small cell lung cancer in the National Cancer Database. Ann Thorac
TNM staging rules to lung cancer presenting as multiple nodules with Surg 2016; 101: 1850–1855.
ground glass or lepidic features or a pneumonic type of involvement in 56. Ginsberg RJ, Rubinstein LV. Randomized trial of lobectomy versus lim-
the forthcoming eighth edition of the TNM classification for lung can- ited resection for T1N0 non-small cell lung cancer. Lung Cancer Study
cer. J Thor Oncol 2016; 11: 666–680. Group. Ann Thorac Surg 1995; 60: 615–622.
37. Vilmann P, Clementsen PF, Colella S et al. Combined endobronchial and 57. Veluswamy RR, Ezer N, Mhango G et al. Limited resection versus lobec-
esophageal endosonography for the diagnosis and staging of lung cancer: tomy for older patients with early stage lung cancer: impact of histology.
European Society of Gastrointestinal Endoscopy (ESGE) guideline, in co- J Clin Oncol 2015; 33: 3447–3453.
operation with the European Respiratory Society (ERS) and the European 58. Koike T, Kitahara A, Sato S et al. Lobectomy versus segmentectomy in
Society of Thoracic Surgeons (ESTS). Endoscopy 2015; 47: 545–559. radiologically pure solid small-sized non-small cell lung cancer. Ann
38. Asamura H, Chansky C, Crowley J et al. The IASLC lung cancer staging Thorac Surg 2016; 101: 1354–1360.
project: proposals for the revisions of the N descriptors in the forthcom- 59. Blasberg JD, Pass HI, Donington JS. Sublobar resection: a movement
ing 8th edition of the TNM classification for lung cancer. J Thorac from the Lung Cancer Study Group. J Thorac Oncol 2010; 5:
Oncol 2015; 10: 1675–1684. 1583–1593.
39. Ettinger DS, Wood DE, Aisner DL et al. NCCN clinical practice guide- 60. Nakamura K, Saji H, Nakajima R et al. A phase III randomized trial of
lines in oncology: non–small cell lung cancer version 5.2017. March 16, lobectomy versus limited reection for for small-sized peripheral non-
2017. https://www.nccn.org/professionals/physician_gls/pdf/nscl.pdf. small cell lung cancer (JCOG0802/WJOG4607L). Jpn J Clin Oncol
(1 May 2017, date last accessed). 2010; 40: 271–274.
40. Baldwin DR, White B, Schmidt-Hansen M et al. Diagnosis and treat- 61. Petrella F, Spaggiari L. The smaller the better: a new concept in thoracic
ment of lung cancer: summary of updated NICE guidance. BMJ 2011; surgery? Lancet Oncol 2016; 17: 699–700.
342: d2110. 62. Zhang W, Wei Y, Jiang H et al. Video-assisted thoracoscopic surgery versus
41. Lim E, Baldwin D, Beckles M et al. Guidelines on the radical manage- thoracotomy lymph node dissection in clinical stage I lung cancer: a meta-
ment of patients with lung cancer. Thorax 2010; 65 (Suppl 3): iii1–iii27. analysis and system review. Ann Thorac Surg 2016; 101: 2417–2424.
42. Silvestri GA, Gonzalez AV, Jantz MA et al. Methods for staging non- 63. Bendixen M, Jørgensen OD, Kronborg C et al. Postoperative pain and
small cell lung cancer diagnosis and management of lung cancer, 3rd ed: quality of life after lobectomy via video-assisted thoracoscopic surgery
American College of Chest Physicians evidence-based clinical practice or anterolateral thoracotomy for early stage lung cancer: a randomised
guidelines. Chest 2013; 143(5 Suppl): e211S–e250S. controlled trial. Lancet Oncol 2016; 17: 836–844.
43. Vernon J, Andruszkiewicz N, Schneider L et al. Comprehensive clinical 64. Rami-Porta R, Wittekind C, Goldstraw P, International Association for the
staging for resectable lung cancer: clinicopathological correlations and Study of Lung Cancer (IASLC) Staging Committee. Complete resection in
the role of brain MRI. J Thor Oncol 2016; 11: 1970–1975. lung cancer surgery: proposed definition. Lung Cancer 2005; 49: 25–33.
44. Brunelli A, Postmus PE, Preoperative functional evaluation of the surgi- 65. Huang X, Wang J, Chen Q, Jiang J. Mediastinal lymph node dissection ver-
cal candidate. In HI Pass, D Ball, GV Scagliotti (eds), The IASLC sus mediastinal lymph node sampling for early stage non-small cell lung
Approach to Thoracic Oncology. Aurora, CO: IASLC 2014; 373–383. cancer: a systematic review and meta-analysis. PLoS One 2014; 9: e109979.
45. Yacoub WN, Meyers BF. Surgical resection in combination with lung 66. Chang JY, Liu YH, Zhu Z et al. Stereotactic ablative radiotherapy: a po-
volume reduction surgery for stage I non-small cell lung cancer. Semin tentially curable approach to early stage multiple primary lung cancer.
Thorac Cardiovasc Surg 2010; 22: 38–43. Cancer 2013; 119: 3402–3410.
46. Falcoz PE, Conti M, Brouchet L et al. The Thoracic Surgery Scoring 67. Griffioen GH, Lagerwaard FJ, Haasbeek CJ et al. Treatment of multiple
System (Thoracoscore): risk model for in-hospital death in 15,183 patients primary lung cancers using stereotactic radiotherapy, either with or
requiring thoracic surgery. J Thorac Cardiovasc Surg 2007; 133: 325–332. without surgery. Radiother Oncol 2013; 107: 403–408.
47. Brunelli A, Varela G, Salati M et al. Recalibration of the revised cardiac risk 68. Artal Cortés A, Calera Urquizu L, Hernando Cubero J. Adjuvant
index in lung resection candidates. Ann Thorac Surg 2010; 90: 199–203. chemotherapy in non-small cell lung cancer: state-of-the-art. Transl
48. Brunelli A, Cassivi SD, Fibla J et al. External validation of the recali- Lung Cancer Res 2015; 4: 191–197.
brated thoracic revised cardiac risk index for predicting the risk of 69. NSCLC Meta-analyses Collaborative Group; Arriagada R, Auperin A,
major cardiac complications after lung resection. Ann Thorac Surg Burdett S et al. Adjuvant chemotherapy, with or without postoperative
2011; 92: 445–448. radiotherapy, in operable non-small-cell lung cancer: two meta-
49. Lau KK, Rathinam S, Waller DA, Peake MD. The effects of increased analyses of individual patient data. Lancet 2010; 375: 1267–1277.
provision of thoracic surgical specialists on the variation in lung cancer 70. Wakelee HA, Dahlberg SE, Keller SM et al. E1505: adjuvant chemother-
resection rate in England. J Thor Oncol 2013; 8: 68–72. apy þ/- bevacizumab for early stage NSCLC—Outcomes based on
50. Brunelli A, Charloux A, Bolliger CT et al. ERS/ESTS clinical guidelines chemotherapy subsets. J Clin Oncol 2016; 34 (Suppl); abstr 8507.
on fitness for radical therapy in lung cancer patients (surgery and 71. Butts CA, Ding K, Seymour L et al. Randomized phase III trial of vinor-
chemo-radiotherapy). Eur Respir J 2009; 34: 17–41. elbine plus cisplatin compared with observation in completely resected
51. Bradley JD, Paulus R, Komaki R et al. Standard-dose versus high-dose stage IB and II non-small-cell lung cancer: updated survival analysis of
conformal radiotherapy with concurrent and consolidation carboplatin JBR-10. J Clin Oncol 2010; 28: 29–34.
plus paclitaxel with or without cetuximab for patients with stage IIIA or 72. Salazar MC, Rosen JE, Wang Z et al. Association of delayed adjuvant
IIIB non-small-cell lung cancer (RTOG 0617): a randomised, two-by- chemotherapy with survival after lung cancer surgery. JAMA Oncol
two factorial phase 3 study. Lancet Oncol 2015; 16: 187–199. 2017; 3: 610–619.

by guest
on 05 February 2018
73. Strauss GM, Herndon JE 2nd, Maddaus MA et al. Adjuvant paclitaxel plus 92. Lagerwaard FJ, Verstegen NE, Haasbeek CJ et al. Outcomes of stereotactic
carboplatin compared with observation in stage IB non-small-cell lung ablative radiotherapy in patients with potentially operable stage I non-
cancer: CALGB 9633 with the Cancer and Leukemia Group B, Radiation small cell lung cancer. Int J Radiat Oncol Biol Phys 2012; 83: 348–353.
Therapy Oncology Group, and North Central Cancer Treatment Group 93. Siva S, Ball D. Curing operable stage I non-small cell lung cancer with
Study Groups. J Clin Oncol 2008; 26: 5043–5051. stereotactic ablative body radiotherapy: the force awakens. Oncologist
74. Lim E, Harris G, Patel A et al. Preoperative versus postoperative chemo- 2016; 21: 393–398.
therapy in patients with resectable non-small cell lung cancer: system- 94. Chang JY, Senan S, Paul MA et al. Stereotactic ablative radiotherapy ver-
atic review and indirect comparison meta-analysis of randomized trials. sus lobectomy for operable stage I non-small-cell lung cancer: a pooled
J Thorac Oncol 2009; 4: 1380–1388. analysis of two randomised trials. Lancet Oncol 2015; 16: 630–637.
75. NSCLC Meta-analysis Collaborative Group. Preoperative chemother- 95. Cherny NI, Sullivan R, Dafni U et al. A standardised, generic, validated
apy for non-small-cell lung cancer: a systematic review and meta- approach to stratify the magnitude of clinical benefit that can be antici-
analysis of individual participant data. Lancet 2014; 383: 1561–1571. pated from anti-cancer therapies: the European Society for Medical
76. Gilligan D, Nicolson M, Smith I et al. Preoperative chemotherapy in pa- Oncology Magnitude of Clinical Benefit Scale (ESMO-MCBS). Ann
tients with resectable non-small cell lung cancer: results of the MRC Oncol 2015; 26: 1547–1573.
LU22/NVALT 2/EORTC 08012 multicentre randomised trial and up- 96. Samson P, Waters EA, Meyers B, Politi MC. Shared decision making and
date of systematic review. Lancet 2007; 369: 1929–1937. effective risk communication in the high-risk patient with operable stage
77. Huang O, Li J, Sun Y et al. Efficacy of EGFR tyrosine kinase inhibitors I non-small cell lung cancer. Ann Thorac Surg 2016; 101: 2049–2052.
in the adjuvant treatment for operable non-small cell lung cancer by a 97. Hopmans W, Damman OC, Senan S et al. A patient perspective on
meta-analysis. Chest 2016; 149: 1384–1392. shared decision making in stage I non-small cell lung cancer: a mixed
78. Tada H, Takeda K, Nakagawa K et al. Vinorelbine plus cisplatin versus methods study. BMC Cancer 2015; 15: 959.
gefitinib in resected non-small cell lung cancer harboring activating 98. Chen F, Matsuo Y, Yoshizawa A et al. Salvage lung resection for non-
EGFR mutation (WJOG6410L). J Clin Oncol 2012; 30 (Suppl); abstr small cell lung cancer after stereotactic body radiotherapy in initially
TPS7110. operable patients. J Thorac Oncol 2010; 5: 1999–2002.
79. National Cancer Institute, Lung Cancer. http://www.cancer.gov/types/ 99. Neri S, Takahashi Y, Terashi T et al. Surgical treatment of local recur-
lung/research/alchemist (18 May 2017, date last accessed). rence after stereotactic body radiotherapy for primary and metastatic
80. Lindberg K, Nyman J, Riesenfeld K€allskog V et al. Long-term results of lung cancers. J Thorac Oncol 2010; 5: 2003–2007.
a prospective phase II trial of medically inoperable stage I NSCLC 100. Allibhai Z, Cho BC, Taremi M et al. Surgical salvage following stereotactic
treated with SBRT – the Nordic experience. Acta Oncol 2015; 54: body radiotherapy for early-stage NSCLC. Eur Respir J 2012; 39: 1039–1042.
1096–1104. 101. Hamamoto Y, Kataoka M, Yamashita M et al. Lung-cancer related chest
81. Verstegen NE, Lagerwaard FJ, Hashemi SM et al. Patterns of disease re- events detected by periodical follow-up CT after stereotactic body
currence after SABR for early stage non-small-cell lung cancer: optimiz- radiotherapy for stage I primary lung cancer: retrospective analysis of
ing follow-up schedules for salvage therapy. J Thorac Oncol 2015; 10: incidence of lung-cancer related chest events and outcomes of salvage
1195–1200. treatment. Jpn J Radiol 2012; 30: 671–675.
82. Louie AV, Palma DA, Dahele M et al. Management of early-stage non- 102. Taira N, Kawabata T, Ichi T et al. Salvage operation for late recurrence
small cell lung cancer using stereotactic ablative radiotherapy: contro- after stereotactic body radiotherapy for lung cancer: two patients with
versies, insights, and changing horizons. Radiother Oncol 2015; 114: no viable cancer cells. Ann Thorac Surg 2014; 97: 2167–2171.
138–147. 103. Hamaji M, Chen F, Matsuo Y et al. Treatment and prognosis of isolated
83. Chen H, Louie AV, Boldt RG et al. Quality of life after stereotactic abla- local relapse after stereotactic body radiotherapy for clinical stage I
tive radiotherapy for early-stage lung cancer: a systematic review. Clin non-small-cell lung cancer: importance of salvage surgery. J Thorac
Lung Cancer 2016; 17: e141–e149. Oncol 2015; 10: 1616–1624.
84. Bahig H, Filion E, Vu T et al. Severe radiation pneumonitis after lung 104. Dickhoff C, Dahele M, Paul MA et al. Salvage surgery for locoregional re-
stereotactic ablative radiation therapy in patients with interstitial lung currence or persistent tumor after high dose chemoradiotherapy for locally
disease. Pract Radiat Oncol 2016; 6: 367–374. advanced non-small cell lung cancer. Lung Cancer 2016; 94: 108–113.
85. Chen H, Senan S, Nossent EJ et al. Treatment-related toxicity in pa- 105. Verstegen NE, Maat AP, Lagerwaard FJ et al. Salvage surgery for local
tients with early-stage non-small cell lung cancer and co-existing inter- failures after stereotactic ablative radiotherapy for early stage non-small
stitial lung disease: a systematic review. Int J Radiat Oncol Biol Phys cell lung cancer. Radiat Oncol 2016; 11: 131.
2017; 98: 245–246. 106. Chang JY, Bezjak A, Mornex F. IASLC Advanced Radiation Technology
86. Nambu A, Onishi H, Aoki S et al. Rib fracture after stereotactic radio- Committee. Stereotactic ablative radiotherapy for centrally located early
therapy for primary lung cancer: prevalence, degree of clinical symp- stage non–small-cell lung cancer: what we have learned. J Thorac Oncol
toms, and risk factors. BMC Cancer 2013; 13: 68. 2015; 10: 577–585.
87. Timmerman RD, Herman J, Cho LC. Emergence of stereotactic body 107. Senthi S, Haasbeek CJ, Slotman BJ, Senan S. Outcomes of stereotactic
radiation therapy and its impact on current and future clinical practice. ablative radiotherapy for central lung tumours: a systematic review.
J Clin Oncol 2014; 32: 2847–2854. Radiother Oncol 2013; 106: 276–282.
88. Haasbeek CJ, Palma D, Visser O et al. Early-stage lung cancer in elderly 108. Tekatli H, Haasbeek N, Dahele M et al. Outcomes of hypofractionated
patients: a population-based study of changes in treatment patterns and high-dose radiotherapy in poor-risk patients with “ultracentral” non-
survival in the Netherlands. Ann Oncol 2012; 23: 2743–2747. small cell lung cancer. J Thorac Oncol 2016; 11: 1081–1089.
89. Mauguen A, le Péchoux C, Saunders MI et al. Hyperfractionated or 109. Ambrogi MC, Fanucchi O, Dini P et al. Wedge resection and radiofre-
accelerated radiotherapy in lung cancer: an individual patient data quency ablation for stage I non small cell lung cancer. Eur Respir J 2015;
meta-analysis. J Clin Oncol 2012; 30: 2788–2797. 45: 1089–1097.
90. Cheung P, Faria S, Ahmed S et al. Phase II study of accelerated hypo- 110. PORT Meta-analysis Trialists Group. Postoperative radiotherapy for non-
fractionated three-dimensional conformal radiotherapy for stage T1-3 small cell lung cancer. Cochrane Database Syst Rev 2005; CD002142.
N0 M0 non-small cell lung cancer. NCIC CTG BR.25. J Natl Cancer 111. Le Péchoux C. Role of postoperative radiotherapy in resected non-small
Inst 2014; 106: dju164. cell lung cancer: a reassessment based on new data. Oncologist 2011; 16:
91. Onishi H, Shirato H, Nagata Y et al. Stereotactic body radiotherapy 672–681.
(SBRT) for operable stage I non-small-cell lung cancer: can SBRT be 112. Tsitsias T, Boulemden A, Ang K et al. The N2 paradox: similar out-
comparable to surgery?. Int J Radiat Oncol Biol Phys 2011; 81: comes of pre- and postoperatively identified single-zone N2a positive
1352–1358. non-small-cell lung cancer. Eur J Cardiothorac Surg 2014; 45: 882–887.

by guest
on 05 February 2018
113. van Meerbeeck JP, Kramer GW, Van Schil PE et al. Randomized con- cancer according to stage and time since diagnosis. Ann Oncol
trolled trial of resection versus radiotherapy after induction chemother- 2015; 26: 902–907.
apy in stage IIIA-N2 non-small-cell lung cancer. J Natl Cancer Inst 126. Lou F, Huang J, Sima CS et al. Patterns of recurrence and second pri-
2007; 99: 442–450. mary lung cancer in early-stage lung cancer survivors followed with
114. Albain KS, Swann RS, Rusch VW et al. Radiotherapy plus chemotherapy routine computed tomography surveillance. J Thorac Cardiovasc Surg
with or without surgical resection for stage III non-small-cell lung cancer: 2013; 145: 75–81. Discussion 81–82.
a phase III randomised controlled trial. Lancet 2009; 374: 379–386. 127. Demicheli R, Fornili M, Ambrogi F et al. Recurrence dynamics for non-
115. Pless M, Stupp R, Ris HR et al. Induction chemoradiation in stage IIIA/ small-cell-lung cancer: effect of surgery on the develoment of metasta-
N2 non-small-cell lung cancer: a phase 3 randomised trial. Lancet 2015; ses. J Thorac Oncol 2012; 7: 723–730.
386: 1049–1056. 128. Johnson BE. Second lung cancers in patients after treatment for an ini-
116. Eberhardt WE, Pöttgen C, Gauler TC et al. Phase III study of surgery tial lung cancer. J Natl Cancer Inst 1998; 90: 1335–1345.
versus definitive concurrent chemoradiotherapy boost in patients with 129. Ripley RT, McMillan RR, Sima CS et al. Second primary lung cancers:
resectable stage IIIA(N2) and selected IIIB non–small-cell lung cancer smokers versus nonsmokers after resection of stage I lung adenocarcin-
after induction chemotherapy and concurrent chemoradiotherapy oma. Ann Thorac Surg 2014; 98: 968–974.
(ESPATUE). J Clin Oncol 2015; 33: 4194–4201. 130. Voltolini L, Paladini P, Luzzi L et al. Iterative surgical resections for
117. Senan S, Brade A, Wang LH et al. PROCLAIM: randomized phase III trial local recurrent and second primary bronchogenic carcinoma. Eur J
of pemetrexed-cisplatin or etoposide-cisplatin plus thoracic radiation ther- Cardiothorac Surg 2000; 18: 529–534.
apy followed by consolidation chemotherapy in locally advanced nonsqua- 131. Hung JJ, Hsu WH, Hsieh CC et al. Post-recurrence survival in com-
mous non–small-cell lung cancer. J Clin Oncol 2016; 34: 953–962. pletely resected stage I non-small cell lung cancer with local recurrence.
118. Ahn JS, Ahn YC, Kim JH et al. Multinational randomized phase III trial Thorax 2009; 64: 192–196.
with or without consolidation chemotherapy using docetaxel and cisplatin 132. Hamaji M, Allen MS, Cassivi SD et al. Surgical treatment of meta-
after con-current chemoradiation in inoperable stage III non–small-cell chronous second primary lung cancer after complete resection of
lung cancer: KCSG-LU05-04. J Clin Oncol 2015; 33: 2660–2666. non-small cell lung cancer. J Thorac Cardiovasc Surg 2013; 145:
119. Vokes EE, Herndon JE II, Kelley MJ et al. Induction chemotherapy fol- 683–690.
lowed by chemoradiotherapy compared with chemoradiotherapy alone 133. Rosengart TK, Martini N, Ghosn P, Burt M. Multiple primary lung car-
for regionally advanced unresectable stage III nonsmall-cell lung cancer: cinomas: prognosis and treatment. Ann Thorac Surg 1991; 52: 773–778.
cancer and leukemia group B. J Clin Oncol 2007; 25: 1698–1704. 134. Spratt DE, Wu AJ, Adeseye V et al. Recurrence patterns and second pri-
120. Aupérin A, Le Péchoux C, Rolland E et al. Meta-analysis of concomitant mary lung cancers after stereotactic body radiation therapy for early-
versus sequential radiochemotherapy in locally advanced non–small- stage non small-cell lung cancer: implications for surveillance. Clin
cell lung cancer. J Clin Oncol 2010; 28: 2181–2190. Lung Cancer 2016; 17: 177–183.e2.
121. Warner A, Dahele M, Hu B et al. Factors associated with early mortality 135. Huang K, Senthi S, Palma DA et al. High-risk CT features for detection
in patients treated with concurrent chemoradiation therapy for locally of local recurrence after stereotactic ablative radiotherapy for lung can-
advanced non-small cell lung cancer. Int J Radiat Oncol Biol Phys 2016; cer. Radiother Oncol 2013; 109: 51–57.
94: 612–620. 136. Peulen H, Mantel F, Guckenberger M et al. Validation of high-risk com-
122. Kelly K, Altorki NK, Eberhardt WE et al. Adjuvant erlotinib versus pla- puted tomography features for detection of local recurrence after
cebo in patients with stage IB-IIIA non-small-cell lung cancer stereotactic body radiation therapy for early-stage non-small cell lung
(RADIANT): a randomized, double-blind, phase III trial. J Clin Oncol cancer. Int J Radiat Oncol Biol Phys 2016; 96: 134–141.
2015; 33: 4007–4014. 137. De Leyn P, Dooms C, Kuzdzal J et al. Revised ESTS guidelines for pre-
123. Hu Y, McMurry TL, Isbell JM et al. Readmission after lung resection is operative mediastinal lymph node staging for non-small-cell lung can-
associated with a 6-fold increase in 90-day postoperative mortality. cer. Eur J Cardiothorac Surg 2014; 3: 787–798.
J Thorac Cardiovasc Surg 2014; 148: 2261–2267. 138. Lee TH, Marcantonio ER, Mangione CM et al. Derivation and prospec-
124. Pezzi CM, Mallin K, Mendez AS et al. Ninety-day mortality after resec- tive validation of a simple index for prediction of cardiac risk of major
tion for lung cancer is nearly double 30-day mortality. J Thorac noncardiac surgery. Circulation 1999; 100: 1043–1049.
Cardiovasc Surg 2014; 148: 2269–2277. 139. Dykewicz CA. Summary of the guidelines for preventing opportunistic
125. Janssen-Heijnen ML, van Erning FN, de Ruysscher DK et al. infections among hematopoietic stem cell transplant recipients. Clin
Variation in causes of death in patients with non-small cell lung Infect Dis 2001; 33: 139–144.

by guest
on 05 February 2018
Metastatic non-small-cell lung cancer: ESMO Clinical

Practice Guidelines for diagnosis, treatment and
follow-up†
S. Novello1, F. Barlesi2, R. Califano3,4, T. Cufer5, S. Ekman6, M. Giaj Levra7, K. Kerr8, S. Popat9,
M. Reck10, S. Senan11, G. V. Simo12, J. Vansteenkiste13 & S. Peters14 on behalf of the ESMO
1
Oncology Department, University of Turin, AOU San Luigi-Orbassano, Orbassano, Italy; 2Assistance Publique Hôpitaux de Marseille, Multidisciplinary Oncology and
Therapeutic Innovations Department, Aix Marseille University, Marseille, France; 3Department of Medical Oncology, The Christie NHS Foundation Trust, Manchester;
4
Department of Medical Oncology, University Hospital of South Manchester NHS Foundation Trust, Manchester, UK; 5Medical Faculty Ljubljana, University Clinic Golnik,
Golnik, Slovenia; 6Department of Oncology, Karolinska University Hospital, Stockholm, Sweden; 7Thoracic Oncology Unit, Centre Hospitalier Universitaire Grenoble Alpes
(CHUGA), Grenoble, France; 8Department of Pathology, Aberdeen University Medical School, Aberdeen Royal Infirmary, Aberdeen; 9Department of Medicine, Royal
Marsden Hospital, London, UK; 10Department of Thoracic Oncology, LungenClinic Grosshansdorf, Airway Research Center North (ARCN), Member of the German Centre
for Lung Research (DZL), Grosshansdorf, Germany; 11Department of Radiation Oncology, VU University Medical Center, Amsterdam, The Netherlands; 12Thoracic Surgery
Service, Salamanca University Hospital, Salamanca, Spain; 13Respiratory Oncology Unit (Pulmonology), University Hospital KU Leuven, Leuven, Belgium; 14Oncology
Department, Centre Hospitalier Universitaire Vaudois (CHUV), Lausanne, Switzerland
clinical practice
guidelines
incidence and epidemiology cancer incidence from the mid-1980s in men and in the mid-
2000s in women [6].
Primary lung cancer remains the most common malignancy NSCLCs account for 85%–90% of lung cancers, while small-
after non-melanocytic skin cancer, and deaths from lung cancer cell lung cancer (SCLC) has been decreasing in frequency in
exceed those from any other malignancy worldwide [1]. many countries over the past two decades [1]. During the last 25
In 2012, lung cancer was the most frequently diagnosed years, the distribution of histological types of NSCLC has
cancer and the leading cause of cancer death in male popula- changed: in the USA, squamous cell carcinoma (SCC, which
tions. Among females, lung cancer was the leading cause of was formally the predominant histotype) decreased, while
cancer death in more developed countries, and the second adenocarcinoma has increased in both genders. In Europe,
leading cause of cancer death in less developed countries [2]. In similar trends have occurred in men, while in women, both SCC
2013 in the European Union, lung cancer mortality fell in men and adenocarcinoma are still increasing [8].
by 6% compared with 2009, while cancer death rates increased Tobacco smoking is still the main cause of lung cancer in
in women by 7%, thereby approaching male counterparts [3]. most of the patients, and the geographic and temporal patterns
A significantly higher proportion of lung cancer is diagnosed in of the disease largely reflect tobacco consumption during the
patients aged 65 and over [4], and the median age at diagnosis is previous decades. Both smoking prevention and smoking cessa-
around 70 years [5]. Data from 2012 revealed that in the USA, lung tion can lead to a reduction in a large fraction of human
cancer did represent the leading cause of cancer death in males cancers. In countries with effective tobacco control measures,
from the age of 40 years and in females from the age of 60 years the incidence of new lung cancer has begun to decline in men
[6]. A subset of patients with non-small-cell lung cancers and is reaching a plateau for women [3, 9, 10]. Several other
(NSCLCs) presents at a younger age (<40 years), but the incidence factors have been described, including exposure to asbestos,
in this population has decreased in the USA from 1978 to 2010 [7]. arsenic, radon and non-tobacco-related polycyclic aromatic
The number of cancer deaths expected to occur in 2016 in the hydrocarbons. There is evidence that lung cancer rates are
USA has been estimated, still reporting lung cancer as the higher in cities than in rural settings, but many confounding
leading cause of death in both genders, despite declines in lung factors other than outdoor air pollution may be responsible for
this pattern. Interesting hypotheses about indoor air pollution
(e.g. coal-fuelled stoves and cooking fumes) are available,
6962 Viganello-Lugano, Switzerland. showing a correlation with the relatively high burden of non-
E-mail: clinicalguidelines@esmo.org smoking-related lung cancer in women in some countries [11].
†
Evidence for a genetic predisposition to lung cancer has been
Approved by the ESMO Guidelines Committee: February 2002, last update August
2016. This publication supersedes the previously published version—Ann Oncol 2014; difficult to establish as it is confounded by environmental expo-
25 (Suppl. 3): iii27–iii39. sures, but there are emerging data suggesting that single-

by guest
on 05 February 2018
nucleotide polymorphisms in genes in certain loci—15q24-25 progression should be considered and is recommended in the
(CHRNA3, CHRNA5, CHRNAB4), 6p21.33, 5p15.23—have subgroups where it might guide subsequent treatment [IV, A].
some association with lung cancer risk [12, 13]. An example is the use of third-generation epidermal growth
The World Health Organization (WHO) estimates that lung factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) against
cancer is the cause of 1.37 million deaths globally per year. An T790M-mutated resistant tumours, the most common finding
estimated 71% of these deaths are caused by smoking, indicating in EGFR-mutated tumours relapsing after first-/second-
that ∼400 000 deaths annually are attributed to lung cancer in generation EGFR TKI therapy [19]. Pathologists should take
lifetime never smokers [1]. Prevalence of lung cancer in females steps to ensure that tissue handling and processing prioritises all
without a history of tobacco smoking is estimated to represent testing required for treatment selection.
19%, compared with 9% of male lung carcinoma in the USA [14, Genetic alterations, which are key oncogenic events (driver
15]. Especially in Asian countries, an increase in the proportion mutations), have been identified in NSCLC, with two of these—
of NSCLC in never smokers has been observed [16]. These new EGFR mutations and the anaplastic lymphoma kinase (ALK)
epidemiological data have resulted in ‘non-smoking-associated rearrangements—determining approved, selective pathway-
lung cancer’ being considered a distinct disease entity, where spe- directed systemic therapy. A personalised medicine synopsis
cific molecular and genetic tumour characteristics have been table is shown in Table 1.
identified [17]. However, other clinical series do not report differ- Activating (sensitising) EGFR mutations are predictive for re-
ences by sex after adjusting for age and for the higher number of sponse to the EGFR TKIs gefitinib, erlotinib and afatinib. Such
women >60 years who do not smoke compared with men. This treatments result in improved response rate (RR) and progression-
could justify the different incidence, which may not be due to a free survival (PFS), better tolerability and superior quality of life
real difference among genders specifically. (QoL) compared with platinum-based chemotherapy in the first-
line setting, as demonstrated in several randomised trials [21].
EGFR mutations are found in ∼10%–12% of Caucasians with
diagnosis and personalised medicine adenocarcinoma and are more frequent in never smokers, females
Pathological diagnosis of all sample types should be made and in patients of East Asian ethnicity. EGFR mutation testing is
according to the 2015 WHO classification. The classification recommended in all patients with advanced non-squamous cell
discusses the approach to surgically resected tumours, but also carcinoma (NSCC) [I, A] [20]. Molecular EGRF testing is not
has recommendations for small biopsy and cytology diagnosis, recommended in patients with an unequivocal diagnosis of SCC,
which are the only samples available for most patients. except in never/former light smokers (<15 pack years) [IV, A] [22].
Therapeutic decisions for NSCLC patients are dependent upon EGFR mutation testing should use validated methodology in a
specific tumour histological subtype, and this should be given laboratory participating in an external quality assurance scheme in
whenever possible. Immunohistochemistry (IHC) should be all such patient subgroups. The methodology used should provide
used to increase the specificity of diagnosis in the small sample the test sensitivity required for the tumour content of the sample
setting and reduce the NSCLC-NOS (not otherwise specified) and provide an adequate coverage of all clinically relevant muta-
rate to fewer than 10% of cases diagnosed [IV, A] [18]. Minimal tions. Test methodology should have adequate coverage of muta-
IHC should be used. Two markers only, p40 or p63 to predict tions in exons 18–21, including those associated with specific drug
squamous cell carcinoma and TTF1 to predict adenocarcinoma, resistance. At a minimum, when resources or material are limited,
are generally all that is required [18]. Excessive, unnecessary the most common activating mutations (exon 19 deletion and
IHC will consume tumour tissue and may prevent subsequent exon 21 L858R point mutation, including exon 20 T790M) should
molecular analysis. Obtaining adequate tissue material for histo- be determined [I, A].
logical diagnosis and molecular testing is important, to allow ALK fusion is encountered more frequently, but not exclu-
for individual treatment decisions. Tumour rebiopsy at disease sively, in never smokers, adenocarcinoma subtype and in
Table 1. A personalised medicine synopsis table for metastatic NSCLC [20]

Biomarker Method Use LOE,
GOR
EGFR mutation Any appropriate validated method, subject to external quality Used to select patients for EGFR TKI therapy, V, A
assurance identifying those, with sensitising mutations, most
likely to respond
ALK gene Any appropriate validated method, subject to external quality Used to select patients for ALK tyrosine kinase V, A
rearrangement assurance. Standard approach has been FISH, or less often, inhibitor therapy, identifying those, with a positive
multiplex PCR or RT-PCR. Certain IHC approaches may be used as test, most likely to respond
a substitute primary test. IHC may also be used to screen patients,
positive cases confirmed by an orthogonal method (FISH, PCR)
Adapted from [20] by permission of Oxford University Press.

NSCLC, non-small-cell lung cancer; LOE, level of evidence; GOR, grade of recommendation; EGFR, epidermal growth factor receptor; TKI, tyrosine kinase
inhibitor; ALK, anaplastic lymphoma kinase; FISH, fluorescence in situ hybridisation; PCR, polymerase chain reaction; RT-PCR, reverse transcription
polymerase chain reaction; IHC, immunohistochemistry.
v | Novello et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
younger patients, with a prevalence of around 5% in adenocarci- liver, kidneys and adrenal glands should be carried out. Imaging
nomas [23, 24]. ALK TKIs are effective therapies, and routine of the central nervous system (CNS) is most relevant in those
testing for ALK rearrangements is recommended in the same patients with neurological symptoms or signs, however, if avail-
patient groups tested for EGFR mutations, principally those with able, imaging of the CNS with magnetic resonance imaging
NSCC [II, A]. As ALK TKIs are now approved for first-line (MRI, preferably with gadolinium enhancement) or CT brain
therapy, ALK and EGFR testing should be carried out simulta- with iodine contrast should be carried out at diagnosis. MRI is
neously [25]. The break-apart fluorescence in situ hybridisation more sensitive than CT scan [32].
(FISH) test remains a core approach to detect ALK rearrange- If metastatic disease has been determined by CT scan of the
ments. Multiplex polymerase chain reaction (PCR) may be suc- chest and upper abdomen or by brain imaging, other imaging is
cessful but requires an adequate coverage of the many possible only necessary if it has an impact on treatment strategy.
fusion genes now recognised and is challenged by the availability If bone metastases are clinically suspected, bone imaging is
of adequate quality of nucleic acid from the tumour, and by vali- required. Positron emission tomography (PET), ideally coupled
dation of the methodology itself. IHC has a high positive and with CT, and bone scans are helpful for the systemic screening
negative predictive value for ALK fusion. It is widely used to for bone metastasis. PET/CT is the most sensitive modality in
screen patients for possible confirmatory ALK FISH testing but is detecting bone metastasis [33]. Fluorodeoxyglucose (FDG)–
rapidly being adopted in Europe as the primary test for prescribing PET or PET/CT has higher sensitivity and specificity than bone
ALK TKIs. Similar approaches may be taken for ROS1 fusion gene scintigraphy [34]. MRI can be useful to document and charac-
testing in those centres with access to drugs active in this setting. terise a localised bone metastasis.
Next-generation sequencing (NGS) is also used by many FDG–PET/CT scan offers the highest sensitivity for medias-
centres to facilitate testing for multiple gene mutations, as well tinal lymph nodes [35] and distant metastasis assessment.
as (less frequently) for gene fusions involving ALK, RET and NSCLC is staged according to the American Joint Committee
ROS1 [III, A]. NGS testing panels will also provide data on on Cancer (AJCC)/Union for International Cancer Control
HER2, BRAF and MET mutations, for example, which may (UICC) system (7th edition) and is grouped into the stage ca-
allow access to targeted treatment in late lines of therapy, often tegories shown in Tables 2 and 3. Measurement of lesions
in the context of a clinical trial, and, in addition, with a signifi- should follow Response Evaluation Criteria in Solid Tumours
cant survival outcome [26–28]. (RECIST) criteria v1.1 [36].
FISH quantitative analysis might allow for the documentation In the presence of a solitary metastatic lesion on imaging
of MET gene amplification, another strong tumour driver, studies, including pleural and pericardial effusion, efforts should
allowing for MET-directed therapy, mainly in the context of be made to obtain a cytological or histological confirmation of
clinical trials. stage IV disease.
Apart from a pivotal role in the subtyping of poorly differen- A proposal for an eighth staging edition has been made by the
tiated NSCLC in small samples, in ALK testing and probably International Association for the Study of Lung Cancer (IASLC)
also in ROS1 testing [29, 30], IHC is emerging as a useful tool in and will be submitted to the UICC and the AJCC for inclusion
lung cancer diagnostics in other settings, including rapid screen- in the new TNM classification for lung cancer, which is due to
ing for EGFR and BRAF mutations. The use of programmed be published in late 2016 and enacted in January 2017 [37].
death ligand 1 (PD-L1) IHC for selecting patients for anti-pro-
grammed death 1 (PD1) or anti-PD-L1 immunotherapy is con-
troversial, and is not yet considered as harbouring a sufficient management of advanced/metastatic
negative predictive value for checkpoint inhibitor-related treat-
ment decisions. It is, however, likely to emerge as a requirement
disease
for selected patients, at least in some treatment contexts to be The treatment strategy (Figures 1–4) should take into account
defined in ongoing clinical trials. factors like histology, molecular pathology, age, PS, co-morbi-
dities and the patient’s preferences. Treatment decisions should
ideally be discussed within a multidisciplinary tumour board,
staging and risk assessment who can evaluate and change management plans including
A complete medical history including smoking history and co- recommending additional investigations and changes in treat-
morbidities, weight loss, performance status (PS) and physical ment modality [38]. Systemic therapy should be offered to all
examination must be recorded. stage IV patients with PS 0-2 [I, A].
In any stage of NSCLC, smoking cessation should be highly
laboratory encouraged, because it improves outcome and because smoking
may interact with systemic therapy [II, A] [39]; for example,
Standard tests including routine haematology, renal and hepatic
smoking reduces erlotinib bioavailability [40]. Given the estab-
function, and bone biochemistry tests are required. The routine
lished relationship between smoking and lung cancer, patients who
use of serum markers, such as carcinoembryonic antigen (CEA),
have smoked may feel stigmatised or guilty after diagnosis and
is not recommended [31].
more pessimistic about their illness and likely outcomes, all of
which may have adverse implications for health-related QoL [41].
radiology For these reasons, healthcare professionals should give clear advice
A contrast-enhanced computed tomography (CT) scan of the about the adverse implications of continued smoking and include
chest and upper abdomen including complete assessment of smoking cessation programmes in the therapeutic algorithm.
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw326 | v

by guest
on 05 February 2018
Table 2. AJCC/UICC TNM staging system

Primary tumour (T)
TX Primary tumour cannot be assessed, or tumour proven by the presence of malignant cells in sputum or bronchial washings but not visualised
by imaging or bronchoscopy
T1 Tumour 3 cm or less in greatest dimension, surrounded by lung or visceral pleura, without bronchoscopic evidence of invasion more
proximal than the lobar bronchus (i.e. not in the main bronchus)a
T1a Tumour 2 cm or less in greatest dimension
T1b Tumour >2 cm but 3 cm or less in greatest dimension
T2 Tumour >3 cm but 7 cm or less or tumour with any of the following features (T2 tumours with these features are classified T2a if 5 cm or
less); involves main bronchus, 2 cm or more distal to the carina; invades visceral pleura (PL1 or PL2); associated with atelectasis or
obstructive pneumonitis that extends to the hilar region but does not involve the entire lung
T2a Tumour >3 cm but 5 cm or less in greatest dimension
T2b Tumour >5 cm but 7 cm or less in greatest dimension
T3 Tumour >7 cm or one that directly invades any of the following: parietal pleural (PL3) chest wall (including superior sulcus tumors),
diaphragm, phrenic nerve, mediastinal pleura, parietal pericardium or tumour in the main bronchus (<2 cm distal to the carinaa but
without involvement of the carina; or associated atelectasis or obstructive pneumonitis of the entire lung or separate tumour nodule(s) in
the same lobe)
T4 Tumour of any size that invades any of the following: mediastinum, heart, great vessels, trachea, recurrent laryngeal nerve, oesophagus,
vertebral body, carina, separate tumour nodule(s) in a different ipsilateral lobe

N0 No regional lymph node metastases
N1 Metastasis in ipsilateral peribronchial and/or ipsilateral hilar lymph nodes and intrapulmonary nodes, including involvement by direct
extension
N2 Metastasis in ipsilateral mediastinal and/or subcarinal lymph node(s)
N3 Metastasis in contralateral mediastinal, contralateral hilar, ipsilateral or contralateral scalene, or supraclavicular lymph node(s)
M1a Separate tumour nodule(s) in a contralateral tumour with pleural nodules or malignant pleural (or pericardial) effusionb
M1b Distant metastasis
AJCC, American Joint Committee on Cancer; UICC, Union for International Cancer Control; TNM, tumour-node-metastasis. Used with the permission
of the AJCC, Chicago, IL, USA. The original source for this material is the AJCC Cancer Staging Handbook, 7th edition (2010) published by Springer
Science and Business Media LLC, www.springerlink.com [183].
a
The uncommon superficial spreading tumour of any size with its invasive component limited to the bronchial wall, which may extend proximally to the
main bronchus, is also classified as T1a.
b
Most pleural (and pericardial) effusions with lung cancer are due to tumour. In a few patients, however, multiple cytopathologic examinations of pleural
(pericardial) fluid are negative for tumour, and the fluid is non-bloody and is not an exudate. Where these elements and clinical judgment dictate that the
effusion is not related to the tumour, the effusion should be excluded as a staging element and the patient should be classified as M0.
first-line treatment of EGFR and ALK-negative the risk of death at 1 year for platinum over non-platinum
disease combinations, without induction of unacceptable increase in tox-
Chemotherapy with platinum doublets should be considered in icity, platinum-based doublets are recommended in all patients
all stage IV NSCLC patients with EGFR- and ALK-negative with no contraindications to platinum compounds [I, A] [45].
disease, without major comorbidities and PS 0-2 [I, A]. Benefits Neither a large individual trial nor a meta-analysis found an
of chemotherapy versus best supportive care (BSC, namely a overall survival (OS) benefit of six versus fewer cycles of first-line
23% reduction of risk of death, a 1-year survival gain of 9% and platinum-based doublets, although a longer PFS coupled with
1.5-month absolute increase in median survival and improved significantly higher toxicity was reported in patients receiving six
QoL) are observed irrespective of age, sex, histology and PS in cycles [46, 47]. Therefore, four cycles of platinum-based doublets
two meta-analyses [42, 43]. The survival benefit of two-agent followed by less toxic maintenance monotherapy, or four up to a
over one-agent chemotherapy regimens was reported in a meta- maximum of six cycles in patients not suitable for maintenance
analysis in 2004, with no survival benefit seen for three-agent monotherapy, are currently recommended [I, A].
over two-agent regimens [44]. Based on a 2006 meta-analysis, Several platinum-based regimens with third-generation
revealing a statistically significant reduction (equal to 22%) in cytotoxics (cisplatin/paclitaxel, cisplatin/gemcitabine, cisplatin/
v | Novello et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Table 3. Anatomic stage/prognostic groups according to the AJCC/ doublets with the addition of a third-generation cytotoxic agent
UICC TNM staging system (gemcitabine, vinorelbine, taxanes) are recommended in advanced
SCC patients without major comorbidities and PS 0-2 [I, A].
Necitumumab, an immunoglobulin G1 (IgG1) monoclonal
Occult carcinoma TX N0 M0
antibody against EGFR, did not show any evidence that its
Stage 0 Tis N0 M0
addition to cisplatin/pemetrexed is able to increase survival in
Stage IA T1a,b N0 M0
first-line treatment of metastatic NSCC [51]. However, out-
Stage IB T2a N0 M0
comes were different when necitumumab was combined with
Stage IIA T2b N0 M0
T1a,b N1 M0
different chemotherapy regimens in SCC. In the SQUIRE trial,
T2a N1 M0 the addition of necitumumab to cisplatin/gemcitabine produced a
Stage IIB T2b N1 M0 significant OS improvement [11.5 versus 9.9 months, hazard ratio
T3 N0 M0 (HR) 0.84, 95% CI: 0.74–0.96; P = 0.01] and PFS improvement,
Stage IIIA T1a,b, T2a,b N2 M0 with a 1-year survival equal to 48% in the experimental arm versus
T3 N1, N2 M0 43% in the control arm [52]. In a retrospective analysis, tumours
T4 N0, N1 M0 without EGFR expression as assessed by IHC, did not show any
Stage IIIB T4 N2 M0 benefit from the addition of necitumumab. This improvement in
Any T N3 M0 OS and PFS was more pronounced in the group of patients with
Stage IV Any T Any N M1 EGFR-expressing tumours (median OS 11.7 months versus 10.0
months; HR 0.79, 95% CI: 0.69–0.92; P = 0.002; median PFS 5.7
AJCC, American Joint Committee on Cancer; UICC, Union for months versus 5.5 months, HR 0.84, 95% CI: 0.72–0.92; P = 0.018)
International Cancer Control; TNM, tumour-node-metastasis. Used [53]. On the basis of these results, necitumumab plus gemcitabine
with the permission of the AJCC, Chicago, Illinois. The original source and cisplatin represents a new first-line treatment option for
for this material is the AJCC Cancer Staging Handbook, 7th edition advanced SCC expressing EGFR by IHC [I, B; ESMO-MCBS
(2010) published by Springer Science and Business Media LLC, www. (Magnitude of Clinical Benefit Scale) v1.0 score: 1].
springerlink.com [183].
first-line treatment of NSCC. Any platinum-based doublets with

docetaxel, carboplatin/paclitaxel) have shown comparable effi- a third-generation agent including gemcitabine, vinorelbine or
cacy [48]. taxanes can be used in NSCC. The incorporation of pemetrexed
The expected toxicity profile should contribute to the selec- and bevacizumab into individual treatment schedules should be
tion of the chemotherapy regimen, taking into account that: considered based on the following:
- Meta-analyses have shown higher RRs for cisplatin combina- - Pemetrexed-based combination chemotherapy represents
tions compared with carboplatin combinations. a therapeutic option, based on the results of a recent meta-ana-
- One meta-analysis from individual patient data has shown lysis that showed a slight but significant survival benefit com-
slightly longer OS for cisplatin-based versus carboplatin- pared with gemcitabine- or docetaxel-based combinations and of
based doublet in patients with NSCC and treated with third- a pre-planned subgroup analysis of a large randomised phase III
generation regimens [I, B] [49]. trial [II, A] [54, 55]. Pemetrexed use should be restricted to
- Cisplatin-based chemotherapy is associated with more NSCC in any line of treatment in advanced disease [I, A] [56, 57].
gastrointestinal, neuro- and nephrotoxicity, while bone - The survival benefit of carboplatin in combination with
marrow toxicity is more common with carboplatin. pemetrexed has been investigated in a meta-analysis (explora-
- The albumin-bound paclitaxel (nab-paclitaxel)/carboplatin tory subgroup analysis); survival benefit for pemetrexed plus
(nab-PC) regimen has been shown in a large phase 3 trial to platinum held true for cisplatin-containing regimens but not
have a significantly higher RR compared with the solvent- for carboplatin-based regimens [54]; however, results from
based paclitaxel/carboplatin (sb-PC) and less neurotoxicity prospective randomised studies investigating this question are
[I, B] [50]. The benefits were observed in both SCC and not yet available.
NSCC, with a larger impact on response in SCC. Median - Findings of two randomised clinical trials revealed that
OS was 12.1 months [95% confidence interval (CI): 10.8– bevacizumab improves OS when combined with paclitaxel/car-
12.9 months] in the nab-PC arm compared with 11.2 boplatin regimens in patients with NSCC and PS 0-1, and,
months (95% CI: 10.3–12.6 months) in the sb-PC arm. For therefore, may be offered in the absence of contraindications in
this reason, the nab-PC regimen could be considered a che- eligible patients with advanced NSCC [I, A] [58, 59]. While
motherapeutic option in advanced NSCLC patients, par- one trial of non-taxane, gemcitabine/cisplatin combination
ticularly in patients with greater risk of neurotoxicity, pre- with or without bevacizumab demonstrated an objective RR
existing hypersensitivity to paclitaxel or contraindications (ORR) and modest PFS advantage, but no OS benefit [60], two
for standard paclitaxel premedication [I, B]. meta-analyses showed a consistent significant improvement of
RR, PFS and OS for the combination of bevacizumab and plat-
first-line treatment of SCC. Most individual trials and meta- inum-based chemotherapy, compared with platinum-based
analyses evaluating chemotherapy options in the first-line chemotherapy alone in eligible patients with NSCC [61, 62].
treatment of advanced NSCLC did not report any differential Treatment with bevacizumab also delayed the incidence of
efficacy in patients with SCC [43]. Therefore, platinum-based brain metastases in a retrospective analysis [63].
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw326 | v

by guest
on 05 February 2018
Stage IV SCC
Never or former light

smoker (<15 pack/year)
Molecular test Molecular test

(ALK/EGFR) negative
Molecular test positive Targeted therapy
I) Age
II) PS
<70 years and PS 0-1 <70 years and PS 2 PS 3-4

or
>70 years and PS 0-2
4-6 cycles:
4-6 cycles: BSC [II, B]
Cisplatin – gemcitabine [I, A]
Cisplatin – docetaxel [I, A] Carboplatin-based doublets [II, B]
Cisplatin – vinorelbine [I, A] Single-agent chemotherapy
Carboplatin – paclitaxel [I, A] (gemcitabine, vinorelbine or docetaxel) [I, A]
Carboplatin – nab-paclitaxel [I, B]
Cisplatin – gemcitabine – necitumumab
(if EGFR expression by IHC) [I, B; MCBS 1]
Disease progression PS 3-4 BSC
PS 0-2
Nivolumab [I, A; MCBS 5]

Pembrolizumab if PD-L1 >1%
[I, A; MCBS 3 if PD-L1 >1%; MCBS 5 if PD-L1 >50%]
Docetaxel [I, B]
Ramucirumab – docetaxel [I, B; MCBS 1]
Erlotinib [II, C]
Afatinib [II, C; MCBS 1]
Figure 1. Treatment algorithm for stage IV SCC. SCC, squamous cell carcinoma; ALK, anaplastic lymphoma kinase; EGFR, epidermal growth factor receptor;
PS, performance status; IHC, immunohistochemistry; BSC, best supportive care; MCBS, Magnitude of Clinical Benefit Scale.
revealed platinum-based regimens to be superior in terms of RR

performance status 2 and beyond and survival (74% higher probability of being alive after 1 year)
Chemotherapy prolongs survival and improves QoL in NSCLC despite an increase in toxicities (mainly haematological) [65]. In
patients with PS 2 compared with BSC [I, B] [64]. addition, the superiority of carboplatin-based combinations
A recently published meta-analysis of randomised trials com- over monotherapy in PS 2 patients has been identified in a sub-
paring the efficacy and safety of platinum-based doublets versus group analysis within large phase III trials, with an acceptable
single-agent regimens in the first-line therapy of PS 2 patients toxicity profile [66]. Moreover, combination chemotherapy with
v | Novello et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Stage IV NSCC
ALK translocation Molecular tests EGFR mutation
No EGFR/ALK Gefitinib [I, A]

Crizotinib [I, A] Erlotinib [I, A]
mutation
+/- bevacizumab [I, A; MCBS 2]
Afatinib [I, A]
<70 years and PS 2

<70 years and PS 0-1 PS 3-4
or
>70 years and PS 0-2
4-6 cycles: 4-6 cycles: BSC [II, B]

Cisplatin – pemetrexed [II, A] Carboplatin-based doublets [II, B]
Cisplatin – gemcitabine [I, B] Single-agent chemotherapy
Cisplatin – docetaxel [I, B] (gemcitabine, vinorelbine or docetaxel) [I, A]
Carboplatin – paclitaxel [I, B]
Carboplatin – nab-paclitaxel [I, B]
+/- bevacizumab
PS 0-1
Partial response
or stable disease
Maintenance treatment:
Pemetrexed (switch) [I, B]
Pemetrexed (continuation) [I, A]
Erlotinib (EGFR-activating mutation) [I, B]
+/- bevacizumab (if given before)
Disease progression PS 3-4 BSC
PS 0-2
Pemetrexed [I, B]
Docetaxel [I, B]
Nivolumab [I, B; MCBS 5]
Pembrolizumab if PD-L1 >1%
[I, A; MCBS 3 if PD-L1 >1%; MCBS 5 if PD-L1 >50%]
Ramucirumab – docetaxel [I, B; MCBS 1]
Nintedanib – docetaxel [II, B]
Erlotinib [II, C]
Figure 2. Treatment algorithm for stage IV NSCC. NSCC, non-squamous cell carcinoma; ALK, anaplastic lymphoma kinase; EGFR, epidermal growth factor
receptor; PS, performance status; BSC, best supportive care; MCBS, Magnitude of Clinical Benefit Scale.
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw326 | v

by guest
on 05 February 2018
Stage IIIB-IV lung carcinoma with EGFR-activating mutation
PS 0-2 [I, A] PS 3-4 [II, A]
Gefitinib [I, A]
Erlotinib [I, A]
+/- bevacizumab [I, A; MCBS 2]
Afatinib [I, A]
Oligoprogression Disease
progression
Local treatment (surgery or radiotherapy)

and continue targeted systemic treatment [IV, C] Systemic progression
Re-biopsy
Exon 20 T790M mutation-
Liquid biopsy
re-biopsy not feasible
Exon 20 T790M Platinum-based chemotherapy

mutation +
Osimertinib [III, A]
Figure 3. Treatment algorithm for stage IIIB–IV lung carcinoma with EGFR-activating mutation. EGFR, epidermal growth factor receptor; PS, performance status.
carboplatin significantly improved survival compared with expected during treatment. Therefore, clinicians and patients
monotherapy alone in patients with PS 2 [67]. Therefore, plat- should always discuss the risks and benefits of chemotherapy
inum-based ( preferably carboplatin) doublets should be consid- and should make a joint decision.
ered in eligible PS 2 patients [II, A] [68]. Single-agent Poor PS (3–4) patients should be offered BSC in the absence
chemotherapy with gemcitabine, vinorelbine and docetaxel of documented activating (sensitising) EGFR mutations or ALK
represents an alternative treatment option [I, B] [69]. rearrangements [II, B].
In this subgroup of patients, there are instances where poor A phase II randomised trial compared three treatment strat-
PS is attributable to a high tumour burden, and improved PS egies (gemcitabine, gefitinib or docetaxel) in chemotherapy-
may be expected in response to treatment. In other cases, poor naive patients with advanced NSCLC and PS 2-3, achieving
PS may be related to co-morbidity, and a worsening PS may be similar results in terms of PFS and median survival time.
v | Novello et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Stage IIIB-IV lung carcinoma with ALK translocation
Crizotinib [I, A]
Oligoprogression Disease progression
Local treatment (surgery or radiotherapy)

Systemic progression
and continue targeted systemic treatment
Re-biopsy
(recommended)
Ceritinib [III, A]
Alectinib [III, A]
Figure 4. Treatment algorithm for stage IIIB–IV lung carcinoma with ALK translocation. ALK, anaplastic lymphoma kinase.
Median survival times were 2.2 months, 95% CI: 1.9–3.4, 2.4 therapy [66]. Median OS was 10.3 months for doublet chemo-
months, 95% CI: 1.6–4.4 and 3.5 months, 95% CI: 1.8–6.6, for therapy and 6.2 months for monotherapy (HR 0.64, 95% CI:
gemcitabine, gefitinib and docetaxel, respectively, with docetaxel 0.52–0.78; P < 0.0001); 1-year survival was 44.5% (95% CI:
being associated with higher rates of adverse events (AEs) [70]. 37.9–50.9) and 25.4% (95% CI: 19.9–31.3), respectively.
Nevertheless, there are no data to support the use of front-line Benefit was observed across all subgroups, but increased tox-
cytotoxic treatment over BSC alone in PS >2 patients. icity (notably febrile neutropaenia and sepsis-related deaths)
was observed. With regard to particular platinum-based com-
binations, data show a good tolerability of nab-PC in patients
elderly patients aged ≥70, a tolerability comparable to that in younger counter-
Several phase III trials established single-agent therapy (doce- parts [74].
taxel, vinorelbine or gemcitabine) as the standard of care for Platinum-based chemotherapy is the preferred option for
first-line therapy for advanced NSCLC patients aged ≥70 [69, elderly patients with PS 0-1, as well as selected PS 2 and ad-
71]. However, several platinum-based and non-platinum-based equate organ function, while a single-agent approach (vinorel-
doublets have been tested in elderly patients in recent decades. bine, gemcitabine, docetaxel) might remain the recommended
Two meta-analyses reported a favourable RR but showed treatment of unfit or co-morbid patients, who are more likely to
increased toxicity (notably haematological toxicity) with develop a higher incidence of treatment-related AEs [I, B].
doublet- compared with single-agent therapy [72, 73], while a Comprehensive geriatric assessment (CGA) is based on a
statistically superior OS was observed in one of them, in which a multidisciplinary and global approach to elderly patients, cover-
subgroup analysis favoured platinum-based doublets. Among ing functional status, cognitive capacities, emotional status, co-
the largest prospective phase III trials in elderly patients, one morbidities, nutritional status, polypharmacy, social and envi-
study comparing monthly carboplatin plus weekly paclitaxel ronmental situations and a possible geriatric syndrome. CGA
versus single-agent vinorelbine or gemcitabine in patients aged can predict morbidity and mortality in elderly patients with
70–89 with PS 0-2 found a survival advantage for combination cancer [75] and can help to adapt cancer management to each
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw326 | v

by guest
on 05 February 2018
patient’s fitness or frailty [III, C], even if a recently published of progression after first-line chemotherapy and upon recovery
phase III trial did not demonstrate an improvement in survival from toxicities from the previous treatment [I, A].
outcomes of elderly patients with advanced NSCLC deriving Of note, three studies, one employing bevacizumab and the
from CGA-based allocation of chemotherapy [76]. other two using monoclonal antibodies against EGFR (cetuxi-
mab or necitumumab) administered concomitantly to chemo-
therapy and further continued as monotherapy until disease
maintenance progression, have demonstrated survival benefits, but the specif-
Decision-making about maintenance therapy must take into ic role of the maintenance phase cannot be appreciated in this
account histology, residual toxicity after first-line chemotherapy, context [52, 58, 84].
response to platinum doublet, PS and patient preference.
Several trials have investigated the role of maintenance treat-
ment in patients with good PS (0, 1) either as ‘continuation second-line treatment of EGFR- and ALK-negative
maintenance’ or as ‘switch maintenance’. ‘Continuation main- disease
tenance’ and ‘switch maintenance’ therapies refer, respectively, Patients clinically or radiologically progressing after first-line
to either the maintained use of an agent included in first-line chemotherapy, irrespective of administration of maintenance
treatment or the introduction of a new agent after four cycles of chemotherapy, and with PS 0-2, should be offered second-line
platinum-based chemotherapy. therapy [I, A]. Combination chemotherapy regimens failed to
Two randomised phase III switch maintenance trials have show any OS benefit over single-agent treatments [85]. Single
reported improvements in PFS and OS with pemetrexed [57] agents improve disease-related symptoms and OS. Comparable
and erlotinib [77] versus placebo following four cycles of plat- options in the second line consist of pemetrexed (for NSCC
inum-based chemotherapy. In the case of pemetrexed, this only) [86] or docetaxel [I, B] [87]. Erlotinib improved OS in
benefit was seen only in patients with NSCC [I, B]. second line or in third line in all NSCLC histological subtype
In the erlotinib trial, subgroup analyses revealed a benefit patients not eligible for further chemotherapy, including
restricted to patients with stable disease (SD) after induction patients with PS 3 [88]. Erlotinib was shown to be equivalent to
treatment, as opposed to patients with tumour response. These pemetrexed or docetaxel in refractory ( progression during the
results initially led to the label for switch maintenance with erlo- four cycles of a standard platinum-based chemotherapy
tinib in patients with SD after induction treatment [57, 77]. doublet), unselected patients in a randomised trial [89]. Finally,
However, this indication is no longer justified based on the a large randomised phase III trial showed comparable outcome
findings in the IUNO study, which failed to meet its primary with pemetrexed or erlotinib [90].
end point of OS. This phase III trial showed that in patients In a randomised trial including 222 EGFR WT NSCLC
with EGFR wild-type (WT) tumours who had not progressed patients, initially designed to assess selected biomarkers,
following four cycles of platinum-based chemotherapy, and second-line therapy with docetaxel was shown to be superior to
who had received ‘early erlotinib’ in the first-line maintenance erlotinib with respect to PFS, as well as to OS, but only using an
setting, OS was not superior to erlotinib treatment upon unplanned adjusted HR for primary end-point analysis [91].
disease progression (HR 1.02; 95% CI: 0.85–1.22; P = 0.8183) Subgroup analyses of a phase III trial of erlotinib versus doce-
and 1-year event-free rates were the same in both treatment taxel as second- or third-line therapy demonstrated superior
groups [78]. Maintenance treatment with erlotinib is, there- PFS but not OS for docetaxel treatment in WT EGFR [92].
fore, not recommended for NSCLC patients without an EGFR- Similar results have been reported in a meta-analysis carried
activating mutation [I, B]. out on six randomised, controlled trials with a total of 990
Randomised trials investigating continuation maintenance patients with WT EGFR. Results indicated that, in the second-
have shown an improvement of PFS and OS [79, 80]. A large line treatment of EGFR WT-advanced NSCLC, PFS was signifi-
phase III randomised trial of continuation maintenance with cantly inferior in the EGFR TKI group versus the chemotherapy
pemetrexed versus placebo after four induction cycles of cis- group (HR 1.37, 95% CI: 1.20–1.56, P < 0.00001). However, this
platin plus pemetrexed chemotherapy demonstrated a PFS and did not translate into an OS difference (HR 1.02, 95% CI: 0.87–
OS improvement in patients with a PS 0-1, confirmed at long- 1.20, P = 0.81) [93].
term follow-up [80, 81]. Median OS was 13.9 months (95% CI: In conclusion, erlotinib still represents a potential second-line
12.8–16.0 months) pemetrexed and 11.0 months (95% CI: 10.0– treatment option in pretreated patients with unknown or WT
12.5 months) placebo, with 1- and 2-year survival rates signifi- EGFR status and preferably in patients not suitable for chemo-
cantly longer for patients given pemetrexed (58% and 32%, re- therapy, with, however, limited efficacy in WT EGFR patients
spectively) than for those given placebo (45% and 21%). compared with chemotherapy [II, C].
Another phase III study comparing maintenance bevacizumab, Ramucirumab, a vascular endothelial growth factor receptor-
with or without pemetrexed, after first-line induction with beva- 2 (VEGFR2) inhibitor, was recently investigated as second-line
cizumab, cisplatin and pemetrexed showed a benefit in PFS for therapy for stage IV NSCLC [94]. The study compared doce-
the pemetrexed–bevacizumab combination but no improvement taxel with ramucirumab or placebo in patients who had pro-
in OS [82], although a trend towards improved OS was seen gressed after platinum-based chemotherapy. Median OS was
when analysing 58% of events of 253 patients randomised for longer (10.5 versus 9.1 months, HR 0.86, 95% CI: 0.75–0.98,
this study [83]. Continuing pemetrexed following completion of P = 0.023) in the ramucirumab arm compared with the placebo
four cycles of first-line cisplatin/pemetrexed chemotherapy is, arm. Median PFS was also superior in the ramucirumab arm
therefore, recommended in patients with NSCC, in the absence (4.5 versus 3 months, P < 0.0001). Ramucirumab combined
v | Novello et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
with docetaxel is, therefore, a treatment option in second-line improvement of 3.2 months [97, 98] leading to its approval by
treatment of advanced NSCLC with PS 0-2, regardless of his- the FDA in March 2015 and by the EMA in July 2015. At the
tology [I, B; ESMO-MCBS v1.0 score: 1]. data cut-off, median OS was 9.2 months (95% CI: 7.3–13.3
Bevacizumab administered every two weeks in association months) on nivolumab compared with 6.0 months (95% CI:
with weekly paclitaxel was recently investigated in patients with 5.1–7.3 months) on docetaxel, with a 41% reduction in the risk
NSCC. The study compared the association of paclitaxel-bevaci- of death in the nivolumab arm (HR 0.59, 95% CI: 0.44–0.49;
zumab versus docetaxel in second-third line of treatment; P < 0.001). An updated follow-up reported an 18-month OS of
median PFS was longer (5.4 versus 3.9 months, HR 0.62, 95% 28% and 13% in the nivolumab and docetaxel arms,
CI: 0.44–0-86, P = 0.005) in the paclitaxel- bevacizumab arm respectively, and an 18-month PFS equal to 17% for nivolumab
compared with docetaxel. No difference in median OS was and to 2.7% for docetaxel [99].
observed (9.9 versus 10.8 months, HR 1.15, P = 0.49) [95]. In the phase III trial, nivolumab was better tolerated than
While registration trials of pemetrexed, docetaxel and erloti- docetaxel, with 85% of nivolumab patients receiving at least 90%
nib did not limit therapy to a set number of treatment cycles, of their planned dose intensity, compared with 69% of docetaxel
second-line treatment duration should be individualised, and patients, together with a treatment discontinuation rate of 10%
treatment may be prolonged if disease is controlled and toxicity versus 3% of the patients treated with nivolumab and docetaxel,
acceptable [II, B]. respectively. The experimental arm also showed a positive
Lung cancer has been historically considered poorly immuno- impact on QoL and a longer time to symptom deterioration
genic, with no established benefit from cytokine modulation or compared with the standard arm [100]. The expression of PD-
vaccines. Nevertheless, the recent development of checkpoint L1 was neither prognostic nor predictive of clinical benefit in a
inhibitors provided a promising new approach for immunother- retrospective analysis using various cut points in this study.
apy in patients with NSCLC. Immune checkpoints are inhibitory Nivolumab at 3 mg/kg every 2 weeks is therefore recom-
pathways that maintain self-tolerance and protect peripheral mended in unselected pretreated patients with platinum pre-
tissues by restricting the immune responses. The two checkpoint treated advanced SCC [I, A; ESMO-MCBS v1.0 score: 5].
targets that have been studied more extensively in lung cancer are Based on the KEYNOTE-010 trial [96], pembrolizumab is
the cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) and another immunotherapy option in second-line but also in third-
the PD-1 receptor. Among the antibodies against PD-1, nivolu- line therapy in patients with SCC expressing PD-L1 [I, A;
mab, a fully IgG4 PD-1 immune checkpoint inhibitor, was the ESMO-MCBS v1.0 score: 3 if PD-L1 >1%; 5 if PD-L1 >50%].
first to be investigated in phase III trials, as reported below. Afatinib versus erlotinib was tested in a phase III trial on 795
Pembrolizumab is another anti-PD-1 monoclonal antibody advanced SCC patients. PFS at the primary analysis was signifi-
that has recently received European Medicines Agency (EMA) cantly longer with afatinib than with erlotinib, with a median of
approval for the treatment of any histological type of NSCLC 2.4 months (95% CI: 1.9–2.9) versus 1.9 months (95% CI: 1.9–
after failure of first-line therapy in patients with tumours expres- 2.2); HR 0.82, 95% CI: 0.68–1.00, P = 0.0427. OS was a median
sing PD-L1 [I, A; ESMO-MCBS v1.0 score: 3 if PD-L1 >1%; 5 if of 7.9 months (95% CI: 7.2–8.7) versus 6.8 months (95% CI:
PD-L1 >50%] (PD-L1 IHC 22C3 pharmDx is indicated as an 5.9–7.8); HR 0.81, 95% CI: 0.69–0.95, P = 0.0077.
aid in identifying NSCLC patients for treatment with pembroli- Afatinib could be an additional option for the treatment of
zumab). Eastern Cooperative Oncology Group PS 0-2 patients locally
The phase III KEYNOTE-010 trial randomised 1034 patients advanced or metastatic SCC progressing on or after platinum-
with previously treated NSCLC with PD-L1 expression on at based chemotherapy [II, C; ESMO-MCBS v1.0 score: 1] [101].
least 1% of tumour cells to receive pembrolizumab 2 mg/kg,
pembrolizumab 10 mg/kg or docetaxel 75 mg/m2 every 3 weeks. second-line treatment in NSCC
The primary end points were OS and PFS both in the total pemetrexed: A phase III second-line study demonstrated
population and in the patients with PD-L1 expression on at least non-inferiority for OS between pemetrexed and docetaxel (8.3
50% of tumour cells [96]. In the entire population, OS was sig- versus 7.9 months, HR 0.99, 95% CI: 0.8–1.2). However,
nificantly longer for pembrolizumab 2 mg/kg versus docetaxel pemetrexed showed a better toxicity profile with a significantly
(HR 0.71, 95% CI: 0.58–0.88; P = 0.0008) and for pembrolizu- lower rate of neutropaenia and alopaecia as well as lower rates of
mab 10 mg/kg versus docetaxel (HR 0.61, 95% CI: 0.49–0.75; gastrointestinal AEs [86]. In a retrospective analysis, a predictive
P < 0.0001), with median OS of 10.4, 12.7 and 8.5 months in the impact of histology on outcome by pemetrexed was reported
three arms, respectively. Pembrolizumab achieved a higher favouring those patients with NSCC (median OS: 8.0 versus 9.3
outcome for those patients with high PD-L1 expression (>50%). months, docetaxel versus pemetrexed, HR 0.78, 95% CI: 0.61–
Grade 3–5 treatment-related AEs were less common with pem- 1.0, P = 0.004) [56]. Pemetrexed is a treatment option for those
brolizumab than with docetaxel [43 (13%) of 339 patients given patients with NSCC who did not receive this drug as first-line
2 mg/kg, 55 (16%) of 343 given 10 mg/kg and 109 (35%) of 309 therapy [I, B].
given docetaxel]. The recommended dose and schedule of pem-
brolizumab is 2 mg/kg administered as an intravenous infusion docetaxel and nintedanib: The combination of docetaxel and
over 30 min every 3 weeks. the angiokinase inhibitor nintedanib has been investigated in
1314 patients with pretreated advanced NSCLC in the LUME
second-line treatment in SCC. A phase III trial (CheckMate Lung 1 trial. Compared with docetaxel, a significant prolongation
017) in patients previously treated for SCC compared 3 mg/kg of PFS, the primary end point, was observed in the intent-to-treat
of nivolumab given every 2 weeks with docetaxel, showing an (ITT) population (median PFS 3.4 versus 2.7 months, HR 0.79,
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw326 | v

by guest
on 05 February 2018
95% CI: 0.68–0.92; P = 0.0019), while a significant prolongation The recently presented phase IIb study LUX-LUNG 7 showed
of OS was reported for patients with adenocarcinoma histology that afatinib achieves a modestly higher RR and a longer PFS
(median OS 12.6 versus 10.3 months, HR 0.82, 95% CI: 0.7–0.99; [11 versus 10.9 months, HR (95% CI): 0.73 (0.57–0.95);
P = 0.0359). Additional unplanned analyses revealed a pronounced P = 0.0165] than gefitinib as first-line treatment of patients with
impact on OS in patients with fast-progressing disease and patients advanced NSCLC with common activating mutations (del19 or
who were refractory to first-line chemotherapy. Compared with L858R) [II, B]. Data on OS (co-primary end point) are still im-
docetaxel, the combination of nintedanib and docetaxel was mature and data on QoL have not been presented [114].
associated with a higher incidence of gastrointestinal AEs and EGFR TKIs represent the standard of care as first-line treat-
transient elevation of liver enzymes [102]. However, no impact on ment of advanced EGFR-mutated NSCLC [I, A].
QoL has been reported by the analysis of patient-reported Notably, none of the above studies have shown any benefit in
outcomes [103]. The combination of docetaxel and nintedanib OS for an EGFR TKI over platinum-based chemotherapy, likely
should be considered as a second-line option in patients with due to the high level of crossover.
adenocarcinoma, especially in those progressing within 9 months However, an unplanned pooled OS analysis of patients who
from the start of first-line chemotherapy [II, B]. have been recruited to either the LUX-Lung 3 or the LUX-Lung
6 trial revealed an OS benefit for afatinib compared with chemo-
nivolumab: Nivolumab led to a significant prolongation of OS therapy in patients with EGFR del-19 mutations (median OS:
compared with docetaxel in 582 pretreated patients with NSCC, 27.3 versus 24.3 months; HR 0.81, 95% CI: 0.66–0.99;
who were recruited to the Checkmate-057 trial (median OS 12.2 P = 0.0374), whereas this improvement was not observed in
versus 9.4 months, HR 0.73, 95% CI: 0.59–0.89; P = 0.002), patients with EGFR L858R mutations [II, A] [115].
although a small excess of early progression and/or death events Should the information on the presence of an EGFR-sensitising
were observed for nivolumab, compared with docetaxel. In mutation become available during first-line platinum-based
addition, RR (19% versus 12%, P = 0.021) and duration of chemotherapy, it is recommended to continue chemotherapy for
response (17.2 versus 5.6 months) were in favour of nivolumab, up to four cycles, and then to offer the EGFR TKI as maintenance
while no significant difference has been reported for PFS (median treatment in those patients achieving disease control [77], or as
PFS 2.3 versus 4.2 months, HR 0.92, 95% CI: 0.77–1.1). An second-line treatment at the time of progression [I, A].
exploratory retrospective analysis revealed an association of efficacy In a Japanese randomised trial, 154 EGFR-mutated patients
by nivolumab and the level of tumour membrane expression of were randomised to receive erlotinib and bevacizumab or
PD-L1. However, this analysis is limited by the retrospective and erlotinib alone (75 patients in the combination arm and 77
unplanned nature of this biomarker assessment. In the nivolumab in the erlotinib alone arm were included in the efficacy ana-
arm, compared with docetaxel, a lower frequency of both serious lyses) [116]. Median PFS was 16.0 months (95% CI: 13.9–
adverse events (SAEs; CTC grade 3/4 events: 10% versus 54%) and 18.1) with erlotinib plus bevacizumab and 9.7 months (95%
AEs leading to treatment discontinuation (5% versus 15%) were CI: 5.7–11.1) with erlotinib alone (HR 0.54, 95% CI: 0.36–
observed. The most frequent selected AEs were rash, pruritus, 0.79; log-rank test P = 0.0015). No new safety signals were
diarrhoea, hypothyroidism, elevation of liver enzymes and identified with the combination treatment, and the incidence
pneumonitis [104]. of AEs (any grade) and SAEs were similar between the treat-
Nivolumab at 3 mg/kg every 2 weeks represents a treatment ment arms. There was a higher frequency of grade 3 or worse
option in pretreated patients with advanced NSCC [I, B; ESMO- AEs with the combination and a relatively high incidence of bev-
MCBS v1.0 score: 5] and should be administered in second-line acizumab discontinuation due to AEs (41%); however, most of
NSCC patients. Patients with PD-L1-positive tumours extract these AEs were non-serious and reversible. A similar PFS was
an OS benefit compared with docetaxel [I, B], while in PD-L1- described in a European phase II trial that also evaluated the
negative tumours, both the use of nivolumab and docetaxel combination of erlotinib and bevacizumab, which represents a
resulted in similar OS outcomes, with a more favourable profile front-line treatment option in EGFR-mutated patients [I, A;
regarding nivolumab [II, A]. ESMO-MCBS v1.0 score: 2] [117].
The majority of patients will progress after 9–12 months of
pembrolizumab: Based on the KEYNOTE-010 trial [96], pem- treatment with an EGFR TKI, and various mechanisms of
brolizumab is another immunotherapy option in second-line acquired resistance to first-generation EGFR TKIs have been
but also in third-line therapy in patients with NSCC expressing described [118]. The most common (49%–60%) mechanism of
PD-L1 [I, A; ESMO-MCBS v1.0 score: 3 if PD-L1 <1%; 5 if PD- acquired resistance is the acquisition of a single recurrent mis-
L1 >50%]. sense mutation within exon 20, the T790M mutation [119, 120].
This mutation leads to the substitution of threonine by methio-
EGFR-mutated NSCLC patients nine at position 790, which encodes part of the kinase domain
Several studies have consistently demonstrated that the EGFR of the receptor and results in increased affinity for ATP, causing
TKIs (gefitinib, erlotinib and afatinib) produce higher RRs, resistance to competitive inhibition by reversible EGFR TKIs
longer PFS and improve QoL compared with standard plat- such as gefitinib and erlotinib [121, 122].
inum-based doublet chemotherapy in patients with good PS (PS A number of third-generation EGFR TKIs that are specifically
0-2), whose tumour harbours an activating (sensitising) EGFR designed to target EGFR T790M mutation have undergone clin-
mutation [105–112]. Patients with PS 3-4 may also be offered an ical development. Among these, osimertinib, an oral, selective,
EGFR TKI, as they are likely to receive a similar clinical benefit third-generation, irreversible EGFR TKI inhibitor with activity
to patients with good PS [II, A] [113]. against T790M mutation, is licensed for use in patients who
v | Novello et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
have developed the EGFR T790M resistance mutation and Subsequently, the phase III study PROFILE 1014 compared
should be the treatment of choice in this setting [123, 124]. crizotinib with cisplatin–pemetrexed without maintenance
Patients who progress after an EGFR TKI should undergo a pemetrexed as first-line treatment in ALK-positive advanced
rebiopsy to perform molecular analysis specifically looking for NSCC [137], and demonstrated a significantly longer PFS
EGFR T790M mutation. This approach could influence the next (median of 10.9 versus 7.0 months; HR for progression or death
therapeutic step or reveal alternative EGFR TKI resistance with crizotinib 0.45; 95% CI: 0.35–0.60; P < 0.001) and higher
mechanisms such as transformation to SCLC or bypass tracks ORR with crizotinib compared with chemotherapy (74% and
that could potentially be addressed in clinical trials. 45%, respectively; P < 0.001). Median OS was not reached in
In patients with clinically relevant progression after previous either group.
treatment with an EGFR TKI and confirmed T790M mutation, First-line treatment with crizotinib is the preferred treatment
treatment with osimertinib at a dose of 80 mg/day p.o. should of patients with ALK-rearranged NSCLC [I, A].
be considered [III, A]. As for the EGFR-mutated, ALK-rearranged NSCLC patients,
At the present time, when rebiopsy is not feasible or when the the combination of continuing the ALK TKI with local treat-
EGFR T790M mutation is not detected as a resistance mechanism, ment (radiotherapy or surgery) may represent a reasonable
the standard of care is represented by platinum-based chemother- option and could be considered on an individualised basis [III,
apy alone. There is no data to support continuation of the EGFR B] [132].
TKI with platinum-based chemotherapy [I, A] [125]. Despite improved outcome in patients with tumours harbour-
A valid alternative to tissue rebiopsy is represented by liquid ing ALK rearrangements and treated with crizotinib, all patients
biopsy, which has been validated [126], represents a surrogate will eventually experience disease progression through primary
source of DNA and is a new strategy for tumour genotyping, or acquired resistance. Furthermore, crizotinib penetration into
mainly at the time of progression for EGFR-mutated patients the cerebrospinal fluid (CSF) is negligible, and this pharmaco-
[III, A] [127–129]. In the event that a T790M mutation in per- logical limitation is extremely relevant in treatment decisions,
ipheral blood is observed, treatment with third-generation taking into account the high propensity of ALK-rearranged
EGFR TKIs is justified [130]. If a T790M-negative liquid biopsy NSCLC to metastasise to the brain [138]. Various resistance
is observed, a tissue rebiopsy is recommended if feasible and if mechanisms to ALK inhibitors have been identified, resulting in
accepted by the patient. the development of new therapeutic approaches and novel TKIs.
A phase II study has demonstrated benefit in PFS in patients Two phase I studies, including the multicentre open-label
who continued first-line erlotinib beyond radiological progres- ASCEND-1 study, showed a significant activity of ceritinib,
sion [131]; therefore, this strategy could be considered in based on an ORR of 56% and 6.9 months of PFS in patients
patients with asymptomatic progression. Evidence from retro- with ALK-rearranged NSCLC with crizotinib resistance [139].
spective series and case reports suggests that, in patients where The benefit also included intracranial responses in patients with
there is evidence of radiological progression in a single site (i.e. brain metastasis.
CNS metastasis or adrenal gland), but with ongoing dependence Based on this data, ceritinib can be recommended in patients
on the driver oncogene addiction and without rapid systemic with ALK-positive advanced NSCLC who progress on treatment
progression, the combination of continuing the EGFR TKI with with or are intolerant to crizotinib [III, A].
local treatment (radiotherapy or surgery) may represent a rea- Alectinib is another second-generation ALK inhibitor,
sonable option and could be considered on an individualised which has been approved in Japan for all patients with
basis [III, B] [132]. advanced ALK-positive NSCLC. Two phase II studies have
also demonstrated RR between 45% and 50% and PFS of
8.9 months [140, 141]. Alectinib was also effective for brain
ALK-rearranged NSCLC patients metastases. Furthermore, alectinib was tested in a phase III
The anti-tumour activity of crizotinib, a dual ALK and MET TKI, head-to-head trial comparing this molecule (300 mg b.i.d.)
was initially demonstrated in two multicentre single-arm studies, with crizotinib (250 mg b.i.d.) in untreated ALK-positive
with significant ORR and PFS advantages [133] as well as a sur- advanced NSCLC patients, demonstrating the superiority of
vival advantage compared with other treatment options [134]. alectinib as an initial targeted treatment [142]. Two hundred
The phase III PROFILE 1007 study confirmed the benefit of and seven patients were enrolled, when an independent data
crizotinib over chemotherapy, pemetrexed or docetaxel (inves- monitoring committee recommended the release of study data,
tigator’s choice), as second-line treatment with better ORR and because the superiority in the primary end-point PFS had been
PFS [135]. The median PFS, as determined by independent demonstrated at planned interim analysis. The PFS HR of the
radiological review, was 7.7 months (95% CI: 6.0–8.8) in the alectinib arm compared with the crizotinib arm was 0.34
crizotinib group, compared with 3.0 months (95% CI: 2.6–4.3) (99.6826% CI: 0.17–0.70, P < 0.0001). Median PFS was not
in the chemotherapy group (HR for disease progression or reached (95% CI: 20.3–NE) in the alectinib arm, while it was
death with crizotinib 0.49, 95% CI: 0.37–0.64, P < 0.001). 10.2 months (95% CI: 8.2–12.0) in the crizotinib arm. A
Crizotinib also showed an advantage over both pemetrexed similar global trial in ALK+ treatment-naïve patients has com-
and docetaxel with regards to the improvement in symptoms pleted accrual, and results are pending.
and QoL [136]. Several alternative ALK inhibitors are currently in clinical
Based on these data, any patient with NSCLC harbouring an development, with broader activity against a number of mutated
ALK fusion and previously treated should receive crizotinib in ALK genes and mainly characterised by higher brain activity
second line, if this was not previously administered [I, A]. [143].
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw326 | v

by guest
on 05 February 2018
role of radiotherapy in stage IV NSCLC In the case of a single metastasis, stereotactic radiosurgery
Radiotherapy plays a major role in the symptom control of me- (SRS) or resection is the recommended treatment [II, B] [149,
tastases, such as painful chest wall disease, superior vena cava 150]. For two to three metastases, SRS is recommended in
syndrome, soft tissue or neural invasion. Neurological symp- patients with RPA class I–II [II, B]. There is currently no evi-
toms from spinal cord compression can be relieved by early dence that adding upfront whole brain radiotherapy (WBRT) to
radiotherapy. surgery or to SRS has an impact on OS [I, A] [151].
Radiotherapy is indicated in cases of haemoptysis, symptom- Data from a prospective observational Japanese study sug-
atic airway obstruction and following surgery for CNS, and, gested that the use of SRS may have a role in patients with more
sometimes, bone surgery [II, B]. than three metastases [152]. The observational study enrolled
1194 eligible patients with 1–10 newly diagnosed brain metasta-
ses in a 3-year period, with the largest tumour <10 ml in volume
role of palliative surgery in stage IV NSCLC
and <3 cm in longest diameter; total cumulative volume ≤15 ml,
Recurrent pleural effusions can be managed by pleurodesis. The and a KPS score of 70 or higher, with all patients undergoing
preferred sclerosing agent is talc, which is more effective than standard SRS [152]. Median OS after SRS was 13.9 months
bleomycin or tetracycline [II, B] [144]; thoracoscopic insuffla- (95% CI: 12.0–15.6) in the 455 patients with a single metastasis,
tion with talc ( poudrage) is more effective than talc slurry scle- 10.8 months (95% CI: 9.4–12.4) in the 531 patients with 2–4
rosis [II, B] [145]. If pleurodesis is not possible due to bronchial tumours and 10.8 months (95% CI: 9.1–12.7) in the 208 patients
obstruction or trapped lung, or in the case of pleurodesis failure, with 5–10 tumours. OS was similar in patients with 2–4
recurrent pleural effusions may be managed by indwelling sub- tumours and in those with 5–10. In outpatients undergoing
cutaneous pleural catheters [146]. SRS, treatment-related side-effects occur in 8% of cases, findings
Surgery might be necessary in the case of significant local that indicate SRS as a valid alternative to WBRT in fit patients
complications related to primary tumour or metastasis, like [IV, C]. SRS, with or without WBRT, has recently been further
abscess, uncontrolled massive haemoptysis, spinal cord com- investigated in an individual patient data meta-analysis of three
pression or pathological bone fracture. phase III trials [153]. The age of the patient significantly influ-
enced survival (P = 0.04), with SRS alone favoured in patients
role of minimally invasive procedures in stage IV aged 50 or younger, and with no significant survival differences
NSCLC in patients aged >50. Patient age was also a significant factor for
Endoscopy has a role to play in palliative care, notably in case of brain failure outside of the radiation field(s) (P = 0.043), with
symptomatic major airway obstruction or postobstructive infec- similar failure rates in both arms for patients ≤50 years of age,
tion, where endoscopic debulking by laser, cryotherapy or stent while the risk was reduced with WBRT for patients aged >50.
placement may be helpful [III, C]. Endoscopy is useful in the When more than three brain metastases are diagnosed,
diagnosis and treatment (endobronchial or by guiding endovas- WBRT is recommended in patients with RPA class I–II [II, B],
cular embolisation) of haemoptysis [III, C]. Vascular stenting although the benefit of WBRT compared with supportive care
might be useful in NSCLC-related superior vena cava compres- alone has not been formally studied in randomised trials.
sion [II, B]. The role of WBRT has been questioned by data from a phase
III non-inferiority study, in which patients were randomised to
role of palliative care in stage IV NSCLC either BSC including dexamethasone plus WBRT 20 Gy in 4 Gy
fractions or to the same BSC without WBRT. This trial
Early palliative care intervention is recommended, in parallel (QUARTZ) demonstrated no difference between the treatment
with standard oncological care [II, A]. Evidence demonstrating arms regarding the relief of symptoms, steroid use, OS, QoL or
that palliative care interventions significantly improve QoL quality-adjusted life years, but full publication is awaited [154].
remains scarce. A randomised trial evaluating the impact of The WBRT most frequent schedules are 20 Gy in 5 fractions or
introducing specialised palliative care early after diagnosis of 30 Gy in 10 fractions, with no difference in outcome [I, A] [155].
stage IV disease on patient QoL in ambulatory patients was able In patients with asymptomatic brain metastases who have not
to show an improvement in QoL and mood, a reduction in ag- yet received prior systemic therapy (i.e. chemotherapy, TKIs)
gressive treatment and an improvement in median OS [147]. however, treatment with upfront systemic chemotherapy and
deferred WBRT should be considered [II, B] [156, 157].
focus on brain and bone metastases For most patients with symptomatic brain metastases and/or
brain metastases. The treatment of patients with brain significant oedema, a dose of dexamethasone of 4 mg/day or an
metastases, and no driver mutations, is dependent on the equivalent dose of another corticosteroid is recommended [II,
prognosis. Prognosis can be estimated based on the Radiation A] [158]. Tapering of the dose and, if possible, cessation after
Therapy Oncology Group recursive partitioning analysis (RPA) radiotherapy are recommended. Corticosteroids are not recom-
[148]: class I patients are those <65 years old, with a good PS mended in the case of asymptomatic brain metastases.
[Karnofsky Index (KI) ≥70%], and no other extracranial Among those patients with a druggable oncogene driver
metastases and a controlled primary tumour; class III patients (EGFR, ALK), between 44% and 60% develop brain metastases
have a KI <70%; and class II represents all other patients [148]. in the course of their disease [153]. In such patients with clini-
In RPA class III patients, radiotherapy is not recommended in cally asymptomatic brain metastases, the use of next-generation
view of the dismal prognosis [I, B]; only BSC is recommended, TKIs may restore control of brain disease, with the possibility to
as their median survival is <2 months. delay cranial radiotherapy [III, B]. In those oncogene-addicted
v | Novello et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
cases with symptomatic brain metastases, the indication and number of organs in which these metastases are present [165].
schedules are those indicated in the other NSCLC patients and The growing interest in oligometastases is based on the concept
already discussed above. that long-term disease control, or even cure, may be achieved in
In patients with EGFR mutation and brain metastases, the PFS some subgroups of these patients [166].
improvement with the irreversible inhibitor afatinib was similar Oligometastases can be either synchronous, when diagnosed
to that observed in patients without brain metastases [159]. The within 1 month before or after the primary tumour was identi-
PFS was significantly better with afatinib versus chemotherapy fied, or metachronous when they appear after treatment of the
(8.2 versus 5.4 months; HR 0.50; P = 0.0297). Results with third- primary tumour [165]. The biology and prognosis related to
generation TKIs, such as osimertinib, are awaited. synchronous and metachronous oligometastases may differ.
In ALK-positive patients progressing on crizotinib, treatment In patients receiving systemic therapy (mainly those onco-
with ceritinib or alectinib shows activity against CNS disease gene-addicted tumours treated with TKIs), the term oligopro-
[III, B]. ALK-positive patients often have brain progression gression can be also applied in the case of clonal progression of
while on crizotinib, as the CSF concentration of the latter is very a limited number of metastatic lesions, when all the other
low [138]. Retrospective review of 94 ALK-rearranged NSCLC lesions remain stable.
patients with brain metastases in a phase I expansion study of Stage IV patients with one to three synchronous metastases at
ceritinib, of which 75 were ALK inhibitor-pretreated, revealed diagnosis may experience long-term disease-free survival (DFS)
an intracranial disease control rate (DCR) of 65.3% in crizoti- following systemic therapy and radical local treatment (high-dose
nib-pretreated patients and 78.9% in TKI-naive patients [139]. radiotherapy or surgery) [III, B]. However, because of the limited
In a phase II trial, alectinib at crizotinib resistance has evidence available, inclusion in clinical trials is preferred. Stage
demonstrated activity for brain metastases, with an ORR of 57% IV patients with limited metachronous metastases may be treated
in patients with measurable lesions and complete response with a radical local treatment as some may experience long-term
observed in 20% of patients [140]. A CNS DCR of 83% was DFS [III, B]. However, this is based only on retrospective data.
reported in all 84 patients with baseline CNS metastases. In A systematic literature review identified 757 NSCLC patients
another small cohort of patients with measurable brain disease treated with 1–5 synchronous or metachronous metastases
being enrolled in another phase 2 trial, 75% of patients achieved [167]. These patients had a median age at diagnosis of 61 years,
an intracranial objective response [141]. 98% had a good PS and two-thirds of patients had early-stage
intrathoracic disease staged IA–IIB (after excluding metastatic
bone metastases. Given the incidence of bone metastases in disease). Surgery was the most common treatment modality for
NSCLC (30%–40% of patients with NSCLC develop bone both primary (n = 635, 83.9%) and metastases (n = 339, 62.3%).
metastases), it may be reasonable to evaluate for bone disease Predictive factors for OS were synchronous versus metachronous
upon disease diagnosis. metastases (P < 0.001), N-stage (P = 0.002) and adenocarcinoma
Zoledronic acid reduces skeletal-related events (SREs) ( patho- histology (P = 0.036). RPA for risk groups identified a good prog-
logical fracture, radiation or surgery to bone, or spinal cord nosis (low-risk) group presenting with metachronous metastases
compression) and is recommended in stage IV bone metastatic (5-year OS of 48%), an intermediate-risk group presenting with
disease [II, B] [160]. synchronous metastases and N0 disease (5-year OS of 36%) and,
Denosumab is not inferior to [I, B] and shows a trend towards finally, a high-risk group presenting with synchronous metastases
superiority to zoledronic acid in lung cancer in terms of SRE pre- and intrathoracic N1/N2 disease (5-year OS of 14%). Caution is
vention [II, B] [161]. In an exploratory analysis of a large phase warranted before concluding that positive outcomes in these
III trial, denosumab was associated with improved median OS in patients are due solely to the treatment intervention, rather than
the subgroup of 702 metastatic NSCLC patients [162]. population selection or other biases [165].
In the study of denosumab versus zoledronic acid in patients In this heterogeneous group of patients with oligometastases,
with advanced cancers, the time extent to which pain interfered the specific approach to oligometastases in the brain has been
with daily life (used as surrogate for QoL) was longer in patients discussed above.
treated with denosumab and with no pain or mild pain interfer- One further subgroup is that of patients with a solitary lesion
ence at baseline [163]. in the contralateral lung. The IASLC Staging and Prognostic
A systematic review analysed the use of radionuclide treat- Factors Committee carried out a systematic literature review,
ment in lung cancer and in the eligible trials, pain relief was aiming at distinguishing a second primary and a metastasis in
reported in 75% of the patients having the onset of pain relief patients who have more than one pulmonary site of cancer
within 1–5 weeks after treatment, lasting up to 6 months [164]. [168]. This review concluded that few features are definitive,
However, the methodology in the included trials was poor; only with many commonly used factors being suggestive, but carry a
two randomised trials were eligible, and neither compared substantial risk of misclassification as the majority of second
radionuclide treatments with placebo or best standard of care. primary lung tumours are of the same histology. For these cases,
Thus, further data are needed in this field. the IASLC recommended a careful review by a multidisciplinary
tumour board, and pursuit of radical therapy, such as that for a
treatment of oligometastatic NSCLC synchronous secondary primary tumour, when possible. Both
The term ‘oligometastases’ refers to a limited number of haema- surgery [169, 170] and SRS [171, 172] have been shown to result
togenous metastases, although there is no consensus on what in long-term survivors in this setting [IV, B].
‘limited’ means. Some groups propose a definition of up to Outcomes of radical approaches in patients with oligometas-
three, others up to five metastatic lesions, yet others limit the tases in other organs (such as bone, liver, adrenal glands or other)
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw326 | v

by guest
on 05 February 2018
Diagnosis and personalised medicine
• Adequate tissue material for histological diagnosis and molecular testing should be obtained to allow for individual treatment decisions.
• Pathological diagnosis should be made according to the 2015 WHO classification and the IASLC/ATS/ERS classification of adenocarcinoma.
• Specific subtyping of all NSCLCs is necessary for therapeutic decision-making and should be carried out wherever possible. IHC stains should be used to
reduce the NSCLC-NOS rate to fewer than 10% of cases diagnosed [IV, A].
• EGFR mutation status should be systematically analysed in advanced NSCC [I, A]. Test methodology should have adequate coverage of mutations in
exons 18–21, including those associated with specific drug resistance. At a minimum, when resources or material are limited, the most common
activating mutations (exon 19 deletion and exon 21 L858R point mutation, including exon 20 T790M) should be determined [I, A].
• Molecular EGFR testing is not recommended in patients with a confident diagnosis of SCC, except in never/former light smokers (<15 pack years) [IV, A].
• Testing for ALK rearrangement should be systematically carried out in advanced NSCC [II, A].
• Detection of the ALK translocation by FISH remains the standard, but IHC with high-performance ALK antibodies may be considered for screening.
• If possible, multiplex platforms for molecular testing are preferable [III, A]. Sequential testing may delay treatment.
• In NSCLC with EGFR-sensitising mutations or ALK translocations, a rebiopsy at the time of progression is encouraged [IV, A].
• A complete history including smoking history and co-morbidities, weight loss, PS and physical examination must be recorded.
• Laboratory: standard tests including routine haematology, renal and hepatic function and bone biochemistry tests are required. Routine use of serum
markers—such as CEA—is not recommended.
• Contrast-enhanced CT scan of the chest and upper abdomen including the liver and the adrenal glands should be carried out.
• Imaging of CNS is reserved for patients with neurological symptoms or signs. MRI is more sensitive than CT scan.
• If bone metastases are clinically suspected, bone imaging is required.
• PET, ideally coupled with CT, and bone scans are helpful for the systemic screening for bone metastasis. Moreover, PET-CT scan may demonstrate
unexpected metastases in 5%–10% of the patients with presumed non-metastatic stage based on conventional imaging.
• NSCLC is staged according to the AJCC/UICC system (7th edition) and is grouped into the stage categories shown in Tables 2 and 3.
• In the presence of a solitary metastatic site on imaging studies, efforts should be made to obtain a cytological or histological confirmation of stage IV
disease.
Management of advanced metastatic disease
• The treatment strategy should take into account the histology, molecular pathology, age, PS, co-morbidities and the patient’s preferences.
• Treatment decisions should be discussed within a multidisciplinary tumour board.
• Systemic therapy should be offered to all stage IV patients with PS 0-2 [I, A].
• In any stage of NSCLC, smoking cessation should be highly encouraged, because it improves the outcome [II, A].
First-line treatment of EGFR and ALK-negative disease (SCC and NSCC)
• Chemotherapy should be considered in all stage IV NSCLC patients with EGFR- and ALK-negative disease, without major co-morbidities and PS 0-2 [I, A].
• Platinum-based doublets are the recommended option in all stage IV NSCLC patients with no contraindications to platinum compounds [I, A].
• Four cycles of platinum-based doublets followed by less toxic maintenance monotherapy, or four up to a maximum of six cycles in patients not suitable
for maintenance monotherapy, are currently recommended [I, A].
• In non-squamous tumours and in patients treated with third-generation regimens, cisplatin should be the treatment of choice [I, B].
• The nab-PC regimen could be considered a chemotherapeutic option in advanced NSCLC patients, particularly in patients with greater risk of
neurotoxicity, pre-existing hypersensitivity to paclitaxel or contraindications for standard paclitaxel premedication [I, B].
• Platinum-based doublets with a third-generation cytotoxic agent (gemcitabine, vinorelbine, taxanes) are recommended in advanced SCC patients [I, A].
• Necitumumab plus gemcitabine and cisplatin represents a treatment option for advanced SCC expressing EGFR by IHC [I, B; ESMO-MCBS v1.0 score: 1].
• Pemetrexed is preferred to gemcitabine or docetaxel in patients with non-squamous tumours [II, A]. Pemetrexed use is restricted to NSCC in any line of
treatment [I, A].
• The combination of bevacizumab and other platinum-based chemotherapies may be considered in eligible patients with NSCC and PS 0-1 [I, A].
PS 2 and beyond
• In patients with PS 2, chemotherapy compared with BSC prolongs survival and improves QoL [I, B].
• Carboplatin-based combination chemotherapy should be considered in eligible PS 2 patients [II, A].
• Single-agent chemotherapy with gemcitabine, vinorelbine and docetaxel is an alternative treatment option [I, B].
• Poor PS (3–4) patients should be treated with BSC only [II, B].
Elderly patients
• Carboplatin-based doublet chemotherapy is recommended in eligible patients aged 70–89 with PS 0-2 and with adequate organ function [I, B].
• For those patients not eligible for doublet chemotherapy, single-agent chemotherapy remains the standard of care [I, B].
• CGA can predict morbidity and mortality in elderly patients with cancer and can help to adapt cancer management to each patient’s fitness or frailty [III, C].
Continued
v | Novello et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Table 4. Continued
Maintenance
• Maintenance chemotherapy should be offered only to patients with PS of 0-1 after first-line chemotherapy. Decisions about maintenance should
consider histology, response to platinum-doublet chemotherapy and remaining toxicity after first-line chemotherapy, PS and patient preference.
• In patients with NSCC and PS 0-1, pemetrexed switch maintenance should be considered in patients having disease control following four cycles of non-
pemetrexed containing platinum-based chemotherapy [I, B]. Pemetrexed continuation maintenance should be considered in patients having disease
control following four cycles of cisplatin-pemetrexed [I, A].
• Erlotinib is indicated for switch maintenance treatment, but limited to patients with locally advanced or metastatic NSCLC with EGFR-activating
mutations [I, B].
Second-line treatment of EGFR- and ALK-negative disease (SCC and NSCC)
• Patients clinically or radiologically progressing after first-line chemotherapy with PS 0-2 should be offered second-line chemotherapy [I, A].
• Treatment may be prolonged if disease is controlled and toxicity acceptable [II, B].
• Comparable options as second-line therapy consist of pemetrexed—for NSCC only—or docetaxel [I, B].
• Nivolumab at 3 mg/kg every 2 weeks is recommended in pretreated patients with advanced SCC [I, A; ESMO-MCBS v1.0 score: 5]. It represents a
treatment option in pretreated patients with advanced NSCC [I, B; ESMO-MCBS v1.0 score: 5]. PD-L1-positive tumour patients benefitted from the use
of nivolumab, compared with docetaxel [I, B]. In PD-L1-negative tumours, nivolumab and docetaxel showed similar results, with a more favourable
toxicity profile for nivolumab [II, A].
• Nintedanib combined with docetaxel is a treatment option in patients with adenocarcinoma, especially in those progressing within 9 months from the
start of first-line chemotherapy [II, B].
• Ramucirumab combined with docetaxel is a treatment option in patients with NSCLC progressing after first-line chemotherapy with PS 0-2 [I, B;
ESMO-MCBS v1.0 score: 1].
• Pembrolizumab at 2mg/kg every 3 weeks is recommended in pretreated patients with platinum-pretreated, advanced SCC or NSCC expressing PD-L1 [I,
A; ESMO-MCBS v1.0 score: 3 if PD-L1 >1%; 5 if PD-L1 >50%].
• In patients unfit for chemotherapy, erlotinib is a potential option in patients with unknown EGFR status, WT EGFR and unfit for chemotherapy [II, C].
• In patients with SCC unfit for chemotherapy, afatinib is a potential option in patients with unknown EGFR status or EGFR WT patients with PS 0-2 [II,
C; ESMO-MCBS v1.0 score: 1].
Tumours with an activating EGFR mutation
• First-line treatment with an EGFR TKI (erlotinib, gefitinib or afatinib) is the standard of care for tumours bearing an activating (sensitising) EGFR
mutation [I, A].
• Patients with EGFR mutation and PS 3-4 may also be offered an EGFR TKI [II, A].
• If information on an EGFR-sensitising mutation becomes available during first-line platinum-based chemotherapy, continue chemotherapy for up to four
cycles and offer the EGFR TKI as maintenance treatment in patients achieving disease control, or as second-line treatment at the time of progression [I, A].
• In patients who progress after an EGFR TKI, rebiopsy is strongly encouraged to look for EGFR T790M mutation, relevant for therapeutic strategy. An
alternative to tissue rebiopsy is represented by liquid biopsy [III, A].
• Osimertinib is recommended in patients who have developed the EGFR T790M resistance mutation after EGFR TKI treatment [III, A].
• When a rebiopsy is not feasible, or when the EGFR T790M mutation is not detected in patients who progress after an EGFR TKI, the standard of care is
platinum-based doublet chemotherapy. No data support the concurrent use of EGFR TKI and platinum-based doublet chemotherapy [I, A].
Tumours with ALK rearrangement
• First-line treatment with crizotinib is preferred for patients with ALK-rearranged NSCLC [I, A].
• Any patient with NSCLC harbouring an ALK fusion should receive crizotinib as next-line therapy, if not received previously [I, A].
• In patients who progress after an ALK TKI, second-generation ALK inhibitors such ceritinib are recommended [III, A]. Several alternative ALK
inhibitors, such as alectinib, are currently in clinical development.
Role of radiotherapy
• Radiotherapy can achieve symptom control for bone and brain metastases and is also effective in treating pain related to chest wall, soft tissue or neural
invasion.
• Neurological symptoms from spinal compression can be relieved by early radiotherapy.
• Radiotherapy is indicated in cases of haemoptysis, symptomatic airway obstruction and following surgery for CNS, and, sometimes, bone surgery [II, B].
Role of palliative surgery in stage IV NSCLC
• Recurrent pleural effusions can be managed by pleurodesis. The preferred sclerosing agent is talc, which is more effective than bleomycin or tetracycline
[II, B]; thoracoscopic insufflation with talc (poudrage) is more effective than talc slurry sclerosis [II, B].
Role of minimal invasive procedures in stage IV NSCLC
• In case of symptomatic major airways obstruction or postobstructive infection, endoscopy debulking by laser, cryotherapy or stent placement may be
helpful [III, C].
Continued
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw326 | v

by guest
on 05 February 2018
Table 4. Continued
• Endoscopy is useful in the diagnosis and treatment (endobronchial or by guiding endovascular embolisation) of haemoptysis [III, C].
• Vascular stenting might be useful in NSCLC-related superior vena cava compression [II, B].
Role of palliative care in stage IV NSCLC
• Early palliative care intervention is recommended, in parallel with standard oncological care [II, A].
Brain metastases
• Treatment is recommended in RPA class I patients (<65 years old, KI ≥70%, no other extracranial metastases and controlled primary tumour) or class II
patients (KI ≥70%, with other extracranial metastases and/or an uncontrolled primary tumour).
• In the case of a single metastasis, SRS or resection is the recommended treatment [II, B].
• For two to three metastases, SRS is recommended in patients with RPA class I–II [II, B]. When more than three brain metastases are diagnosed, WBRT is
recommended in patients with RPA class I–II [II, B].
• RPA class III patients (KI <70%) should not receive radiotherapy in view of the dismal prognosis [I, B]; only BSC is recommended.
• WBRT schedules of 20 Gy in 5 fractions or 30 Gy in 10 fractions have no difference in outcome [I, A].
• Systemic therapy is a reasonable option for patients with no or relatively minor symptoms from brain metastases. Radiotherapy is recommended in the
case of the development or progression of symptoms while on treatment [II, B].
• For symptomatic brain metastases and/or oedema, dexamethasone 4 mg/day or an equivalent dose of another corticosteroid is recommended [II, A].
• In fit patients, options other than WBRT for the treatment of brain metastases could be considered [IV, C].
• In patients with a druggable oncogene driver and clinically asymptomatic brain metastases, next-generation TKIs may restore control of brain disease
and delay cranial radiotherapy [III, B].
• In ALK-positive patients progressing on crizotinib, treatment with ceritinib or alectinib shows activity against CNS disease [III, B].
Bone metastases
• Zoledronic acid reduces SREs (pathological fracture, radiation/surgery to bone or spinal cord compression) and is recommended in stage IV bone
metastatic disease [II, B].
• Denosumab is not inferior to [I, B] and shows a trend towards superiority to zoledronic acid in lung cancer in terms of SRE prevention [II, B].
Treatment of oligometastatic disease
• Stage IV patients with one to three synchronous metastases at diagnosis may experience long-term DFS following systemic therapy and radical local
treatment (high-dose radiotherapy or surgery) [III, B]. Because of limited evidence, inclusion in clinical trials is preferred.
• Stage IV patients with limited metachronous metastases may be treated with a radical local treatment and may experience long-term DFS [III, B].
However, this is based only on retrospective data.
• Solitary lesions in the contralateral lung should, in most cases, be considered as synchronous secondary primary tumours and, if possible, treated with
radical intent [IV, B].
• In patients with driver mutations for whom active systemic therapies are available, the use of ablative therapies such as SABR or surgery is likely to
increase. However, there is limited prospective data to support this policy [IV, C].
Response evaluation
• Response evaluation is recommended after two to three cycles of chemotherapy using the same radiographic investigation that initially demonstrated
tumour lesions.
• Measurements and response assessment should follow RECIST criteria v1.1. However, the adequacy of RECIST in evaluating the response to EGFR or
ALK TKI in respective genetically driven NSCLC is debatable.
• In the case of immune checkpoint inhibitor therapy, RECIST criteria should be used, although irRC may have a role in the overall assessment of therapy.
Follow-up
• Close follow-up, at least every 6–12 weeks to allow for early initiation of second-line therapy, is advised, but should depend on individual retreatment
options [III, B].
• Follow-up with PET is not routinely recommended, due to its high sensitivity and relatively low specificity.
WHO, World Health Organisation; IASLC, International Association for the Study of Lung Cancer; ATS, American Thoracic Society; ERS, European
Respiratory Society; NSCLC, non-small-cell lung cancer; IHC, immunohistochemistry; NSCLC-NOS, non-small-cell lung cancer-not otherwise specified;
EGFR, epidermal growth factor receptor; NSCC, non-squamous cell carcinoma; SCC, squamous cell carcinoma; ALK, anaplastic lymphoma kinase; FISH,
fluorescence in situ hybridisation; PS, performance status; CEA, carcinoembryonic antigen; CT, computed tomography; CNS, central nervous system; MRI,
magnetic resonance imaging; PET, positron emission tomography; AJCC, American Joint Committee on Cancer; UICC, Union for International Cancer
Control; BSC, best supportive care; QoL, quality of life; CGA, comprehensive geriatric assessment; PD-L1, programmed death ligand 1; WT, wild-type;
TKI, tyrosine kinase inhibitor; RPA, recursive partitioning analysis; KI, Karnofsky Index; SRS, stereotactic radiosurgery; WBRT, whole brain radiotherapy;
SRE, skeletal-related event; DFS, disease-free survival; SABR, stereotactic ablative radiotherapy; RECIST, Response Evaluation Criteria in Solid Tumours;
irRC, immune-related response criteria.
v | Novello et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Annals of Oncology
Table 5. Magnitude of Clinical Benefit Scale (MCBS) table for new therapies/indications in non-small-cell lung cancer (NSCLC)a
Therapy Disease Trial Control Absolute HR (95% CI) QoL/toxicity MCBS scoreb
setting survival gain
Afatinib, an irreversible ErbB Advanced Afatinib versus erlotinib as second-line Erlotinib, as second-line OS gain: 1.1 OS: HR for death 0.81 Similar toxicity 1 (Form 2a)
family blocker treatment of patients with advanced treatment of patients with months (0.69–0.95) profile
squamous cell carcinoma of the lung advanced SCC of the lung. Improved
(LUX-Lung 8): an open-label Median OS 6.6 months overall
randomised controlled phase 3 trial health-related
[101] QoL
Phase III
NCT01523587
Bevacizumab, a humanised Advanced Erlotinib alone or with bevacizumab as Erlotinib alone as a first-line PFS gain: 6.3 PFS HR: 0.54 (0.36–0.79) Deteriorated 2 (Form 2b)
anti-VEGF monoclonal first-line therapy in patients with therapy until disease months toxicity
antibody, in combination advanced non-squamous non-small- progression or unacceptable profile.
with erlotinib cell lung cancer (NSCLC) harbouring toxicity. Median PFS 9.7 No
EGFR mutations (JO25567): an open- months improvement
label, randomised, multicentre, phase 2 in QoL
study [116]
Phase II
Japan Pharmaceutical
Information Center, number JapicCTI-

111390
Erlotinib, an EGFR TKI Advanced Erlotinib as maintenance treatment in Placebo, as maintenance PFS gain: 1.2 PFS: HR 0.71 (0.62–0.82) Deteriorated 1 (Form 2b)
advanced NSCLC: a multicentre, treatment in advanced weeks toxicity
randomised, placebo-controlled NSCLC. Control PFS 11.1 profile
phase 3 study [77] weeks
Phase III
NCT00556712
Necitumumab, a second- Advanced Necitumumab plus gemcitabine and Gemcitabine and cisplatin as OS gain: 1.6 OS: HR for death Deteriorated 1 (Form 2a)
generation, recombinant, cisplatin versus gemcitabine and first-line therapy in patients months 0.84 (0.74–0.96) toxicity
human IgG1 EGFR cisplatin alone as first-line therapy in with stage IV SCC. Control profile
antibody in combination patients with stage IV squamous OS 9.6 months
doi:10.1093/annonc/mdw326 | v
with gemcitabine and NSCLC (SQUIRE): an open-label,

cisplatin randomised, controlled phase 3 trial.
[52]
Phase III
NCT00981058
Nivolumab, a fully human Advanced Nivolumab versus docetaxel in Docetaxel in patients with OS gain: 3.2 OS: HR for death 0.59 Improved 5 (Form 2a)c
IgG4 PD-1 immune- advanced squamous cell NSCLC [98] advanced SCC who have months. 2- (0.44–0.79) toxicity
checkpoint–inhibitor Phase III disease progression during or year profile
antibody NCT01642004 after first-line chemotherapy. survival
Control OS 6 months gain 15%
Continued

by guest
on 05 February 2018
v | Novello et al.
Table 5. Continued
Therapy Disease Trial Control Absolute HR (95% CI) QoL/toxicity MCBS scoreb
setting survival gain
Nivolumab, a fully human Advanced Nivolumab versus docetaxel in Docetaxel in patients with OS gain: 2.8 OS: HR for death 0.73 Improved 5 (Form 2a)
IgG4 PD-1 immune- advanced non-squamous NSCLC NSCC that had progressed months. 2- (0.59–0.89) toxicity
checkpoint–inhibitor [104] during or after platinum- year profile
antibody Phase III based doublet chemotherapy. survival
NCT01673867 Control OS 9.4 months gain 16%
Ramucirumab, a human IgG1 Advanced Ramucirumab plus docetaxel versus Placebo plus docetaxel in OS gain: 1.4 OS: HR for death 0.86 Deteriorated 1 (Form 2a)
monoclonal antibody that placebo plus docetaxel for second- patients with SCC or NSCC months (0.75–0.98) toxicity
targets the extracellular line treatment of stage IV NSCLC who had progressed during or profile
domain of VEGFR2, in after disease progression on after a first-line platinum-
combination with docetaxel platinum-based therapy (REVEL): a based chemotherapy regimen.
multicentre, double-blind, Control OS 9.1 months
randomised phase 3 trial [94]
Phase III
NCT01168973
Pembrolizumab, an anti-PD-1 Advanced Pembrolizumab versus docetaxel for Docetaxel in patients with In PD-L1 In PD-L1 >1%:d Improved In PD-L1
monoclonal antibody previously treated, PD-L1-positive, previously treated, >1%:d OS: HR for death 0.71, toxicity >1%: 3
advanced non-small-cell lung cancer PD-L1-positive, advanced OS gain: (0.58–0.88) profile (Form 2a)
(KEYNOTE-010): a randomised NSCLC. Control OS 8.5 1.9 months
controlled trial [96] months In PD-L1 >50%: d In PD-L1
Phase III In PD- L1 OS: HR for death 0.54, >50%: 5
NCT01905657 >50%:d (0.38–0.77) (Form 2a)
OS gain:
6.7 months
HR, hazard ratio; CI, confidence interval; QoL, quality of life; OS, overall survival; VEGF, vascular endothelial growth factor; EGFR, endothelial growth factor receptor; TKI, tyrosine kinase inhibitor; PFS,
progression-free survival; NSCC, non-squamous cell carcinoma; SCC, squamous cell carcinoma; IgG1, immunoglobulin G1; PD-1, programmed death 1; VEGFR2, VEGF receptor 2.
a
EMA approvals in 2016 to end of August 2016.
b
ESMO-MCBS version 1.0 [181].
c
EMA approval, October 2015.
d
Co-primary end points (overall survival and progression-free survival both in the total population and in patients with PD-L1 expression on at least 50% of tumour cells)
Annals of Oncology
by guest
on 05 February 2018
are based on often small retrospective series. One prospective in patients treated with targeted therapies and/or immunother-
single-arm phase II trial was reported, in which the majority of apy. Follow-up with PET is not routinely recommended, due to
patients had a single metastatic lesion [173]. With systemic treat- its high sensitivity and relatively low specificity. Measurements
ment and radical local radiotherapy (brain SRS or high-dose frac- and response reporting should follow RECIST criteria v1.1 [36].
tionated radiotherapy) or surgery of all tumour lesions, the trial The adequacy of RECIST in evaluating response to EGFR or
reported that 13% of patients remained disease free at 3 years. ALK TKI in respective genetically driven NSCLC is still debat-
In a retrospective study of 37 patients with synchronous able even if this remains the standard method of evaluation for
NSCLC and brain metastasis, which were both surgically excised these patients. In these two subgroups of patients, treatment
[174], the 1- and 2-year OS rates were 62% and 24%, respectively. beyond RECIST progression is a common approach, pursuing
It is to note that in this series nodal status did not affect survival clinical benefit more than morphologic response. This approach
on univariate analysis. Nevertheless, lymph node involvement by differs from what was carried out historically in non-oncogene-
the primary tumour is usually considered a contraindication for addicted tumours treated with cytotoxic agents.
further surgical therapy, and thorough invasive assessment of the Criteria for response evaluation with immunostimulatory
mediastinum is recommended before any attempt of surgical monoclonal antibodies (imAbs) are currently the matter of
treatment of synchronous oligometastatic disease [169]. intense work and debate. The vast majority of trials in NSCLC
Analysis of the IASLC Lung Cancer Staging Project database evaluating anti-PD1/PD-L1 antibodies have traditionally used
for the eighth TNM has proposed the category M1b for a single RECIST v1.0/v1.1 criteria, which remain standard criteria in
metastatic lesion in a single distant organ, and M1c for either NSCLC. More recently, immune-related response criteria (irRC)
multiple metastases to a single organ or for multiple lesions to have been proposed and validated in malignant melanoma to
multiple organs [175]. This proposal will allow for prospective better assess the variety of responses that can be generated upon
collection and evaluation of staging and outcomes data for these imAbs [176, 177]. IrRC undoubtedly allows better taking into
subgroups of oligometastatic patients. account the potential for an initial ‘flare-up’ or pseudo-progres-
At present, it is recommended that patients with multiple syn- sion at the tumour site, for the appearance of new non-target
chronous metastases should be treated within prospective clinic- lesions as well as for the difference between kinetics of response
al trials. observed between imAbs and cytotoxic therapy; however, it is
With the advent of long-term survivors following treatment of still insufficient to describe all response profiles or clinical bene-
patients with druggable mutations, who may develop oligoprogres- fits observed, and further data are needed in this context. Also,
sion, use of ablative therapies such as stereotactic ablative radio- alternative end points for clinical trials evaluating imAbs, such
therapy (SABR) or surgery is likely to increase. Nevertheless, there as DCR and tumour growth rate, could be probably implemen-
is a paucity of prospective data to support this policy [IV, C]. ted, especially considering the highly variable timing of re-
sponse, ranging from 6 weeks to several months after treatment
initiation, or even after treatment cessation [178, 180].
response evaluation
Response evaluation is recommended after 2–3 cycles of chemo-
therapy, using the same initial radiographic investigation that
follow-up
demonstrated tumour lesions. The same procedure and timing The optimal approach to post-treatment management of
(every 6–9 weeks) should be applied for the response evaluation patients with NSCLC, including the role of radiological
Levels of evidence
II Small randomised trials or large randomised trials with a suspicion of bias (lower methodological quality) or meta-analyses of such trials or of
trials with demonstrated heterogeneity
a
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw326 | v

by guest
on 05 February 2018
evaluation, is controversial, with very limited literature available. reported an institutional research grant from AstraZeneca and
Due to the aggressive nature of this disease, generally close advisory role for Astra Zeneca, Novartis, MSD, Boehringer
follow-up, at least every 6–12 weeks after first-line therapy, is Ingelheim, Eli-Lilly and Roche. S.Pe. has provided consultation,
advised but should also depend on individual retreatment attended advisory boards and/or provided lectures for
options [III, B]. Given the clear benefits of second-line therapy F. Hoffmann–La Roche, Ltd; Eli Lilly, MSD, AstraZeneca, Pfizer,
in patients who presented an initial response to first-line chemo- Novartis, Boehringer Ingelheim, Bristol-Myers Squibb, Daiichi-
therapy and maintain good PS, radiological follow-up should be Sankyo, Morphotek, Merrimack, Merck Serono, Amgen, Clovis,
considered every 6–12 weeks to allow for early initiation of Tesaro, Celgene and Debiopharm.
second-line therapy. M.G.L. has declared no conflicts of interest.
methodology references
These clinical practice guidelines were developed in accordance 1. Jemal A, Bray F, Center MM et al. Global cancer statistics. CA Cancer J Clin
with the ESMO standard operating procedures for clinical prac- 2011: 61: 69–90.
tice guidelines development, http://www.esmo.org/Guidelines/ 2. Torre LA, Bray F, Siegel RL et al. Global cancer statistics, 2012. CA Cancer J Clin
2015; 65: 87–108.
ESMO-Guidelines-Methodology. The relevant literature has
3. Malvezzi M, Bertuccio P, Rosso T et al. European cancer mortality predictions for
the year 2015: does lung cancer have the highest death rate in EU women? Ann
dations is provided in Table 4. An MCBS table with ESMO- Oncol 2015; 26: 779–786.
MCBS scores is included in Table 5. ESMO-MCBS v1.0 [181] 4. GLOBOCAN. 2012 Cancer incidence, mortality and prevalence worldwide [database
was used to calculate scores for new therapies/indications online]. http://www-dep.iarc.fr/ (11 January 2016, date last accessed).
approved by the EMA since 1 January 2016. Levels of evidence 5. Howlader NNA, Krapcho M et al. SEER Cancer Statistics Review, 1975–2010.
and grades of recommendation have been applied using the Bethesda, MD: National Cancer Institute 2013. http://seer.cancer.gov/ (11
system shown in Table 6. Statements without grading were con- January 2016, date last accessed).
sidered justified standard clinical practice by the experts and the 6. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA Cancer J Clin 2016;
ESMO faculty. 66: 7–30.
7. Thomas A, Chen Y, Yu T et al. Trends and characteristics of young non-small cell
lung cancer patients in the United States. Front Oncol 2015; 5: 113.
acknowledgements 8. Forman D, Bray F, Brewster DH et al. Cancer Incidence in Five Continents. Lyon,
France: IARC Press 2013.
S.P. acknowledges the support of the National Health Service to 9. Jemal A, Ma J, Rosenberg PS et al. Increasing lung cancer death rates among young
the National Institute for Health Research Biomedical Research women in southern and midwestern states. J Clin Oncol 2012; 30: 2739–2744.
Centre at The Royal Marsden Hospital and the Institute of 10. International Agency for Research on Cancer (IARC). http://www.iarc.fr/ (13 Jan
Cancer Research. 2016, date last accessed).
11. Straif K, Cohen A, Samet J. Air Pollution and Cancer, IARC Scientific Publication
No. 161. Lyon, France: IARC Press 2013; 161.
conflict of interest 12. Hung RJ, McKay JD, Gaborieau V et al. A susceptibility locus for lung cancer
maps to nicotinic acetylcholine receptor subunit genes on 15q25. Nature 2008;
S.N. has reported speaker honoraria from Hoffmann-La Roche,
452: 633–637.
Eli Lilly, MSD, Boehringer Ingelheim, AstraZeneca and Bristol-
13. Wang Y, Broderick P, Webb E et al. Common 5p15.33 and 6p21.33 variants
Myers Squibb. F.B. has received honoraria from AstraZeneca, influence lung cancer risk. Nat Genet 2008; 40: 1407–1409.
Bristol-Myers Squibb, Boehringer Ingelheim, Clovis Oncology, 14. Novello S, Stabile LP, Siegfried JM. Gender-related differences in lung cancer. In
Eli Lilly Oncology, F. Hoffmann–La Roche Ltd, Novartis, Pass HI, Scagliotti GV, Ball D (eds), The IASLC Multidisciplinary Approach to
Merck, MSD, Pierre Fabre and Pfizer. R.C. has received honor- Thoracic Oncology. Aurora, CO: IASLC Press 2014; p. 45–67.
aria from Roche, Eli Lilly, MSD, Boehringer Ingelheim, 15. McCarthy WJ, Meza R, Jeon J, Moolgavkar SH. Chapter 6: lung cancer in never
AstraZeneca, Novartis, Clovis and Bristol-Myers Squibb. T.C. smokers: epidemiology and risk prediction models. Risk Anal 2012; 32(Suppl. 1):
has received consultant honoraria from Boehringer Ingelheim, S69–S84.
Pfizer and Bristol-Myers Squibb. S.E. is a consultant to Lilly, 16. Toh CK, Gao F, Lim WT et al. Never-smokers with lung cancer: epidemiologic
evidence of a distinct disease entity. J Clin Oncol 2006; 24: 2245–2251.
Novartis, Bristol-Myers Squibb, Roche and Boehringer
17. Couraud S, Souquet PJ, Paris C et al. BioCAST/IFCT-1002: epidemiological and
Ingelheim. K.K. has reported advisory and/or lecture fees molecular features of lung cancer in never-smokers. Eur Respir J 2015; 45:
from AstraZeneca, Roche, Lilly, Boehringer Ingelheim, Pfizer, 1403–1414.
Novartis, Bristol-Myers Squibb and MSD. S.Po. is an uncom- 18. Travis WD, Brambilla E, Burke AP et al. (eds). WHO Classification of Tumours of
pensated consultant to Ariad, AstraZeneca, Bristol-Myers the Lung, Pleura, Thymus and Heart, 4th edition. Lyon, France: IARC Press 2015.
Squibb, Clovis Oncology, MSD, Novartis and Pfizer and has 19. Sequist LV, Waltman BA, Dias-Santagata D et al. Genotypic and histological
received honoraria from Eli Lilly. M.R. has received honoraria evolution of lung cancers acquiring resistance to EGFR inhibitors. Sci Transl Med
for lectures and consultancy from Roche, Lilly, Bristol-Myers 2011; 3: 75ra26.
Squibb, MSD, AstraZeneca, Boehringer Ingelheim, Pfizer, 20. Kerr KM, Bubendorf L, Edelman MJ et al. Second ESMO consensus conference
on lung cancer: pathology and molecular biomarkers for non-small-cell lung
Novartis and Celgene. S.S. has received honoraria from Eli Lilly,
cancer. Ann Oncol 2014; 25: 1681–1690.
and honoraria and research support from Varian Medical 21. Lee JK, Hahn S, Kim DW et al. Epidermal growth factor receptor tyrosine kinase
Systems. G.V.S. has reported advisory role for Johnson & inhibitors vs conventional chemotherapy in non-small cell lung cancer harboring
Johnson, speaker’s honoraria from Bard, investigation grants wild-type epidermal growth factor receptor: a meta-analysis. JAMA 2014; 311:
from Baxter and team support from Medtronic. J.V. has 1430–1437.
v | Novello et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
22. Rekhtman N, Paik PK, Arcila ME et al. Clarifying the spectrum of driver oncogene 41. Rowland C, Danson SJ, Rowe R et al. Quality of life, support and smoking in
mutations in biomarker-verified squamous carcinoma of lung: lack of EGFR/KRAS advanced lung cancer patients: a qualitative study. BMJ Support Palliat Care
and presence of PIK3CA/AKT1 mutations. Clin Cancer Res 2012; 18: 2016; 6: 35–42.
1167–1176. 42. Non-Small Cell Lung Cancer Collaborative Group. Chemotherapy in non-small cell
23. Kwak EL, Bang YJ, Camidge DR et al. Anaplastic lymphoma kinase inhibition in lung cancer: a meta-analysis using updated data on individual patients from 52
non-small-cell lung cancer. N Engl J Med 2010; 363: 1693–1703. randomised clinical trials. BMJ 1995; 311: 899–909.
24. Shaw AT, Yeap BY, Mino-Kenudson M et al. Clinical features and outcome of 43. Burdett S, Stephens R, Stewart L et al. Chemotherapy in addition to supportive
patients with non-small-cell lung cancer who harbor EML4-ALK. J Clin Oncol care improves survival in advanced non-small-cell lung cancer: a systematic
2009; 27: 4247–4253. review and meta-analysis of individual patient data from 16 randomized
25. Lindeman NI, Cagle PT, Beasley MB et al. Molecular testing guideline for controlled trials. J Clin Oncol 2008; 26: 4617–4625.
selection of lung cancer patients for EGFR and ALK tyrosine kinase inhibitors: 44. Delbaldo C, Michiels S, Syz N et al. Benefits of adding a drug to a single-agent or
guideline from the College of American Pathologists, International Association for a 2-agent chemotherapy regimen in advanced non-small-cell lung cancer: a
the Study of Lung Cancer, and Association for Molecular Pathology. J Thorac meta-analysis. JAMA 2004; 292: 470–484.
Oncol 2013; 8: 823–859. 45. Pujol JL, Barlesi F, Daurès JP. Should chemotherapy combinations for advanced
26. McCourt CM, McArt DG, Mills K et al. Validation of next generation sequencing non-small cell lung cancer be platinum-based? A meta-analysis of phase III
technologies in comparison to current diagnostic gold standards for BRAF, EGFR randomized trials. Lung Cancer 2006; 51: 335–345.
and KRAS mutational analysis. PLoS ONE 2013; 8: e69604. 46. Park JO, Kim SW, Ahn JS et al. Phase III trial of two versus four additional cycles in
27. Frampton GM, Fichtenholtz A, Otto GA et al. Development and validation of a patients who are nonprogressive after two cycles of platinum-based chemotherapy
clinical cancer genomic profiling test based on massively parallel DNA in non small-cell lung cancer. J Clin Oncol 2007; 25: 5233–5239.
sequencing. Nat Biotechnol 2013; 31: 1023–1031. 47. Rossi A, Chiodini P, Sun JM et al. Six versus fewer planned cycles of first-line
28. Barlesi F, Mazieres J, Merlio J-P et al. Routine molecular profiling of patients with platinum-based chemotherapy for non-small-cell lung cancer: a systematic
advanced non-small-cell lung cancer: results of a 1-year nationwide programme review and meta-analysis of individual patient data. Lancet Oncol 2014; 15:
of the French Cooperative Thoracic Intergroup (IFCT). Lancet 2016; 387: 1254–1262.
1415–1426. 48. Schiller JH, Harrington D, Belani CP et al. Comparison of four chemotherapy regimens
29. Marchetti A, Di Lorito A, Pace MV et al. ALK protein analysis by IHC staining after for advanced non-small-cell lung cancer. N Engl J Med 2002; 346: 92–98.
recent regulatory changes: a comparison of two widely used approaches, revision 49. Ardizzoni A, Boni L, Tiseo M et al. Cisplatin- versus carboplatin-based
of the literature, and a new testing algorithm. J Thorac Oncol 2016; 11: chemotherapy in first-line treatment of advanced non-small-cell lung cancer: an
487–495. individual patient data meta-analysis. J Natl Cancer Inst 2007; 99: 847–857.
30. Viola P, Maurya M, Croud J et al. A validation study for the use of ROS1 50. Socinski MA, Bondarenko I, Karaseva NA et al. Weekly nab-paclitaxel in
immunohistochemical staining in screening for ROS1 translocations in lung combination with carboplatin versus solvent-based paclitaxel plus carboplatin as
cancer. J Thorac Oncol 2016; 11: 1029–1039. first-line therapy in patients with advanced non-small-cell lung cancer: final
31. Grunnet M, Sorensen JB. Carcinoembryonic antigen (CEA) as tumor marker in results of a phase III trial. J Clin Oncol 2012; 30: 2055–2062.
lung cancer. Lung Cancer 2012; 76: 138–143. 51. Paz-Ares L, Mezger J, Ciuleanu TE et al. Necitumumab plus pemetrexed and
32. Kuhn MJ, Hammer GM, Swenson LC et al. MRI evaluation of “solitary” brain cisplatin as first-line therapy in patients with stage IV non-squamous non-small-
metastases with triple-dose gadoteridol: comparison with contrast-enhanced CT cell lung cancer (INSPIRE): an open-label, randomised, controlled phase 3 study.
and conventional-dose gadopentetate dimeglumine MRI studies in the same Lancet Oncol 2015; 16: 328–337.
patients. Comput Med Imag Graph 1994; 18: 391–399. 52. Thatcher N, Hirsch FR, Luft AV et al. Necitumumab plus gemcitabine and
33. Wu Y, Li P, Zhang H et al. Diagnostic value of fluorine 18 fluorodeoxyglucose cisplatin versus gemcitabine and cisplatin alone as first-line therapy in patients
positron emission tomography/computed tomography for the detection of with stage IV squamous non-small-cell lung cancer (SQUIRE): an open-label,
metastases in non-small-cell lung cancer patients. Int J Cancer 2013; 132: randomised, controlled phase 3 trial. Lancet Oncol 2015; 16: 763–774.
E37–E47. 53. Paz-Ares L, Socinski MA, Shahidi J et al. 1320_PR: subgroup analyses of
34. Chang MC, Chen JH, Liang JA et al. Meta-analysis: comparison of F-18 patients with epidermal growth factor receptor (EGFR)-expressing tumors in
fluorodeoxyglucose-positron emission tomography and bone scintigraphy in the SQUIRE: a randomized, multicenter, open-label, phase III study of gemcitabine-
detection of bone metastasis in patients with lung cancer. Acad Radiol 2012; 19: cisplatin (GC) plus necitumumab (N) versus GC alone in the first-line treatment of
349–357. patients ( pts) with stage IV squamous non-small cell lung cancer (sq-NSCLC). J
35. Darling GE, Maziak DE, Inculet RI et al. Positron emission tomography-computed Thorac Oncol 2016; 11(4 Suppl): S153.
tomography compared with invasive mediastinal staging in non-small cell lung 54. Li M, Zhang Q, Fu P et al. Pemetrexed plus platinum as the first-line treatment
cancer: results of mediastinal staging in the early lung positron emission option for advanced non-small cell lung cancer: a meta-analysis of randomized
tomography trial. J Thorac Oncol 2011; 6: 1367–1372. controlled trials. PLoS ONE 2012; 7: e37229.
36. Eisenhauer EA, Therasse P, Bogaerts J et al. New response evaluation criteria in 55. Scagliotti GV, Parikh P, von Pawel J et al. Phase III study comparing cisplatin plus
solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009; 45: gemcitabine with cisplatin plus pemetrexed in chemotherapy-naive patients with
228–247. advanced-stage non-small-cell lung cancer. J Clin Oncol 2008; 26:
37. Goldstraw P, Chansky K, Crowley J et al. The IASLC Lung Cancer Staging Project: 3543–3551.
proposals for revision of the TNM stage groupings in the forthcoming (eighth) 56. Scagliotti G, Hanna N, Fossella F et al. The differential efficacy of pemetrexed
edition of the TNM classification for lung cancer. J Thorac Oncol 2016; 11: according to NSCLC histology: a review of two phase III studies. Oncologist 2009;
39–51. 14: 253–263.
38. Ung KA, Campbell BA, Duplan D et al. Impact of the lung oncology 57. Ciuleanu T, Brodowicz T, Zielinski C et al. Maintenance pemetrexed plus best
multidisciplinary team meetings on the management of patients with cancer. supportive care versus placebo plus best supportive care for non-small-cell lung
Asia Pac J Clin Oncol 2016; 12: e298–e304. cancer: a randomised, double-blind, phase 3 study. Lancet 2009; 374:
39. Baser S, Shannon VR, Eapen GA et al. Smoking cessation after diagnosis of lung 1432–1440.
cancer is associated with a beneficial effect on performance status. Chest 2006; 58. Sandler A, Gray R, Perry MC et al. Paclitaxel-carboplatin alone or with
130: 1784–1790. bevacizumab for non-small-cell lung cancer. N Engl J Med 2006; 355:
40. Hughes AN, O’Brien MER, Petty WJ et al. Overcoming CYP1A1/1A2 mediated 2542–2550.
induction of metabolism by escalating erlotinib dose in current smokers. J Clin 59. Zhou C, Wu YL, Chen G et al. BEYOND: a randomized, double-blind, placebo-
Oncol 2009; 27: 1220–1226. controlled, multicenter, phase III study of first-line carboplatin/paclitaxel plus
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw326 | v

by guest
on 05 February 2018
bevacizumab or placebo in Chinese patients with advanced or recurrent progressed following platinum. https://clinicaltrials.gov/ct2/show/NCT01328951
nonsquamous non-small-cell lung cancer. J Clin Oncol 2015; 33: 2197–2204. (5 August 2016, date last accessed)
60. Reck M, von Pawel J, Zatloukal P et al. Phase III trial of cisplatin plus gemcitabine 79. Zhang X, Zang J, Xu J et al. Maintenance therapy with continuous or switch
with either placebo or bevacizumab as first-line therapy for nonsquamous non- strategy in advanced non-small cell lung cancer: a systematic review and meta-
small-cell lung cancer: AVAil. J Clin Oncol 2009; 27: 1227–1234. analysis. Chest 2011; 140: 117–126.
61. Lima ABC, Macedo LT, Sasse AD. Addition of bevacizumab to chemotherapy in 80. Paz-Ares LG, de Marinis F, Dediu M et al. PARAMOUNT: final overall survival
advanced non-small cell lung cancer: a systematic review and meta-analysis. results of the phase III study of maintenance pemetrexed versus placebo
PLoS ONE 2011; 6: e22681. immediately after induction treatment with pemetrexed plus cisplatin for advanced
62. Soria JC, Mauguen A, Reck M et al. Systematic review and meta-analysis of nonsquamous non-small-cell lung cancer. J Clin Oncol 2013; 31: 2895–2902.
randomised, phase II/III trials adding bevacizumab to platinum-based 81. Paz-Ares L, de Marinis F, Dediu M et al. Maintenance therapy with pemetrexed
chemotherapy as first-line treatment in patients with advanced non-small-cell plus best supportive care versus placebo plus best supportive care after induction
lung cancer. Ann Oncol 2013; 24: 20–30. therapy with pemetrexed plus cisplatin for advanced non-squamous non-small-
63. Ilhan-Mutlu A, Osswald M, Liao Y et al. Bevacizumab prevents brain metastases cell lung cancer (PARAMOUNT): a double-blind, phase 3, randomised controlled
formation in lung adenocarcinoma. Mol Cancer Ther 2016; 15: 702–770. trial. Lancet Oncol 2012; 13: 247–255.
64. Gridelli C, Ardizzoni A, Le Chevalier T et al. Treatment of advanced non-small-cell 82. Barlesi F, Scherpereel A, Rittmeyer A et al. Randomized phase III trial of
lung cancer patients with ECOG performance status 2: results of an European maintenance bevacizumab with or without pemetrexed after first-line induction
experts panel. Ann Oncol 2004; 15: 419–426. with bevacizumab, cisplatin, and pemetrexed in advanced nonsquamous non-
65. Bronte G, Rolfo C, Passiglia F et al. What can platinum offer yet in the treatment small-cell lung cancer: AVAPERL (MO22089). J Clin Oncol 2013; 31:
of PS2 NSCLC patients? A systematic review and meta-analysis. Crit Rev Oncol 3004–3011.
Hematol 2015; 95: 306–317. 83. Barlesi F, Scherpereel A, Gorbunova V et al. Maintenance bevacizumab-
66. Quoix E, Zalcman G, Oster JP et al. Carboplatin and weekly paclitaxel doublet pemetrexed after first-line cisplatin-pemetrexed-bevacizumab for advanced
chemotherapy compared with monotherapy in elderly patients with advanced nonsquamous nonsmall-cell lung cancer: updated survival analysis of the
non-small-cell lung cancer: IFCT-0501 randomised, phase 3 trial. Lancet 2011; AVAPERL (MO22089) randomized phase III trial. Ann Oncol 2014; 25:
378: 1079–1088. 1044–1052.
67. Lilenbaum R, Villaflor VM, Langer C et al. Single-agent versus combination 84. Pirker R, Pereira JR, Szczesna A et al. Cetuximab plus chemotherapy in patients
chemotherapy in patients with advanced non-small cell lung cancer and a with advanced non-small-cell lung cancer (FLEX): an open-label randomised
performance status of 2: prognostic factors and treatment selection based on phase III trial. Lancet 2009; 373: 1525–1531.
two large randomized clinical trials. J Thorac Oncol 2009; 4: 869–874. 85. Di Maio M, Chiodini P, Georgoulias V et al. Meta-analysis of single-agent
68. Zukin M, Barrios CH, Pereira JR et al. Randomized phase III trial of single-agent chemotherapy compared with combination chemotherapy as second-line
pemetrexed versus carboplatin and pemetrexed in patients with advanced non- treatment of advanced non-small-cell lung cancer. J Clin Oncol 2009; 27:
small-cell lung cancer and Eastern Cooperative Oncology Group performance 1836–1843.
status of 2. J Clin Oncol 2013; 31: 2849–2853. 86. Hanna N, Shepherd FA, Fossella FV et al. Randomized phase III trial of
69. Gridelli C, Perrone F, Gallo C et al. Chemotherapy for elderly patients with pemetrexed versus docetaxel in patients with non-small-cell lung cancer
advanced non-small-cell lung cancer: the Multicenter Italian Lung Cancer in the previously treated with chemotherapy. J Clin Oncol 2004; 22: 1589–1597.
Elderly Study (MILES) phase III randomized trial. J Natl Cancer Inst 2003; 95: 87. Shepherd FA, Dancey J, Ramlau R et al. Prospective randomized trial of
362–372. docetaxel versus best supportive care in patients with non-small-cell lung cancer
70. Morère J-F, Bréchot J-M, Westeel V et al. Randomized phase II trial of gefitinib or previously treated with platinum-based chemotherapy. J Clin Oncol 2000; 18:
gemcitabine or docetaxel chemotherapy in patients with advanced non-small-cell 2095–2103.
lung cancer and a performance status of 2 or 3 (IFCT-0301 study). Lung Cancer 88. Shepherd FA, Rodrigues Pereira J, Ciuleanu T et al. Erlotinib in previously treated
2010; 70: 301–307. non-small-cell lung cancer. N Engl J Med 2005; 353: 123–132.
71. Kudoh S, Takeda K, Nakagawa K et al. Phase III study of docetaxel compared 89. Ciuleanu T, Stelmakh L, Cicenas S et al. Efficacy and safety of erlotinib versus
with vinorelbine in elderly patients with advanced non-small-cell lung cancer: chemotherapy in second-line treatment of patients with advanced, non-small-cell
results of the West Japan Thoracic Oncology Group Trial (WJTOG 9904). J Clin lung cancer with poor prognosis (TITAN): a randomised multicentre, open-label,
Oncol 2006; 24: 3657–3663. phase 3 study. Lancet Oncol 2012; 13: 300–308.
72. Des Guetz G, Uzzan B, Nicolas P et al. Comparison of the efficacy and safety 90. Karampeazis A, Voutsina A, Souglakos J et al. Pemetrexed versus erlotinib in
of single-agent and doublet chemotherapy in advanced non-small cell lung pretreated patients with advanced non-small cell lung cancer: a Hellenic
cancer in the elderly: a meta-analysis. Crit Rev Oncol Hematol 2012; 84: Oncology Research Group (HORG) randomized phase 3 study. Cancer 2013;
340–349. 119: 2754–2764.
73. Qi WX, Tang L, He AN et al. Doublet versus single cytotoxic agent as first-line 91. Garassino MC, Martelli O, Broggini M et al. Erlotinib versus docetaxel as second-
treatment for elderly patients with advanced non-small-cell lung cancer: a line treatment of patients with advanced non-small-cell lung cancer and wild-
systematic review and meta-analysis. Lung 2012; 190: 477–485. type EGFR tumours (TAILOR): a randomised controlled trial. Lancet Oncol 2013;
74. Socinski MA, Langer CJ, Okamoto I et al. Safety and efficacy of weekly nab®- 14: 981–988.
paclitaxel in combination with carboplatin as first-line therapy in elderly patients 92. Kawaguchi T, Ando M, Asami K et al. Randomized phase III trial of erlotinib
with advanced non-small-cell lung cancer. Ann Oncol 2013; 24: 314–321. versus docetaxel as second- or third-line therapy in patients with advanced non-
75. Extermann M, Hurria A. Comprehensive geriatric assessment for older patients small-cell lung cancer: Docetaxel and Erlotinib Lung Cancer Trial (DELTA). J Clin
with cancer. J Clin Oncol 2007; 25: 1824–1831. Oncol 2014; 32: 1902–1908.
76. Corre R, Greillier L, Le Caër H et al. Use of a comprehensive geriatric assessment 93. Zhao N, Zhang XC, Yan HH et al. Efficacy of epidermal growth factor receptor
for the management of elderly patients with advanced non-small-cell lung inhibitors versus chemotherapy as second-line treatment in advanced non-small-
cancer: the phase III randomized ESOGIA-GFPC-GECP 08-02 study. J Clin Oncol cell lung cancer with wild-type EGFR: a meta-analysis of randomized controlled
2016; 34: 1476–1483. clinical trials. Lung Cancer 2014; 85: 66–73.
77. Cappuzzo F, Ciuleanu T, Stelmakh L et al. Erlotinib as maintenance treatment in 94. Garon EB, Ciuleanu TE, Arrieta O et al. Ramucirumab plus docetaxel versus
advanced non-small-cell lung cancer: a multicentre, randomised, placebo- placebo plus docetaxel for second-line treatment of stage IV non-small-cell lung
controlled phase 3 study. Lancet Oncol 2010; 11: 521–529. cancer after disease progression on platinum-based therapy (REVEL): a
78. A study of first-line maintenance erlotinib versus erlotinib at disease progression multicentre, double-blind, randomised phase 3 trial. Lancet 2014; 384:
in participants with advanced non-small cell lung cancer (NSCLC) who have not 665–673.
v | Novello et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
95. Cortot A.B., Audiger-Valette C., Molinier O. et al. Weekly paclitaxel plus 114. Park K, Tan EH, O’Byrne K et al. Afatinib versus gefitinib as first-line treatment of
bevacizumab versus docetaxel as second or third line treatment in advanced patients with EGFR mutation-positive non-small-cell lung cancer (LUX-Lung 7): a
non-squamous non-small cell lung cancer (NSCLC): results from the phase III phase 2B, open-label, randomised controlled trial. Lancet Oncol 2016; 17:
study IFCT-1103 ULTIMATE. J Clin Oncol 2016; 34 (May 20 Suppl.), 2016: abstr 577–589.
9005. 115. Yang JCH, Wu YL, Schuler M et al. Afatinib versus cisplatin-based chemotherapy
96. Herbst RS, Baas P, Kim DW et al. Pembrolizumab versus docetaxel for previously for EGFR mutation-positive lung adenocarcinoma (LUX-Lung 3 and LUX-Lung 6):
treated, PD-L1-positive, advanced non-small-cell lung cancer (KEYNOTE-010): a analysis of overall survival data from two randomised, phase 3 trials. Lancet
randomised controlled trial. Lancet 2016; 387: 1540–1550. Oncol 2015; 16: 141–151.
97. Rizvi NA, Mazières J, Planchard D et al. Activity and safety of nivolumab, an anti- 116. Seto T, Kato T, Nishio M et al. Erlotinib alone or with bevacizumab as first-line
PD-1 immune checkpoint inhibitor, for patients with advanced, refractory therapy in patients with advanced non-squamous non-small-cell lung cancer
squamous non-small-cell lung cancer (CheckMate 063): a phase 2, single-arm harbouring EGFR mutations (JO25567): an open-label, randomised, multicentre,
trial. Lancet Oncol 2015; 16: 257–265. phase 2 study. Lancet Oncol 2014; 15: 1236–1244.
98. Brahmer J, Reckamp KL, Baas P et al. Nivolumab versus docetaxel in advanced 117. Stahel R, Dafni U, Gautschi O et al. A phase II trial of erlotinib (E) and
squamous-cell non-small-cell lung cancer. N Engl J Med 2015; 373: 123–135. bevacizumab (B) in patients with advanced non-small-cell lung cancer (NSCLC)
99. Reckamp KL, Brahmer JR, Spigel DR et al. Phase 3, randomized trial with activating epidermal growth factor receptor (EGFR) mutations with and
(CheckMate 017) of nivolumab (NIVO) vs docetaxel in advanced squamous (SQ) without T790 M mutation.The Spanish Lung Cancer Group (SLCG) and the
cell non-small cell lung cancer (NSCLC). J Thorac Oncol 2015; 10(Suppl. 2): European Thoracic Oncology Platform (ETOP) BELIEF trial. Ann Oncol 2015; 26
S174. (Suppl. 6): abstr 3BA.
100. Gralla RJ, Coon C, Taylor F et al. Evaluation of disease-related symptoms in 118. Cortot AB, Jänne PA. Molecular mechanisms of resistance in epidermal growth
patients with advanced squamous non-small cell lung cancer treated with factor receptor-mutant lung adenocarcinomas. Eur Respir Rev 2014; 23:
nivolumab or docetaxel. J Thorac Oncol 2015; 10(Suppl. 2): S233. 356–366.
101. Soria JC, Felip E, Cobo M et al. Afatinib versus erlotinib as second-line treatment 119. Yu HA, Arcila ME, Rekhtman N et al. Analysis of tumor specimens at the time of
of patients with advanced squamous cell carcinoma of the lung (LUX-Lung 8): an acquired resistance to EGFR-TKI therapy in 155 patients with EGFR-mutant lung
open-label randomised controlled phase 3 trial. Lancet Oncol 2015; 16: cancers. Clin Cancer Res 2013; 19: 2240–2247.
897–907. 120. Blakely CM, Bivona TG. Resiliency of lung cancers to EGFR inhibitor treatment
102. Reck M, Kaiser R, Mellemgaard A et al. Docetaxel plus nintedanib versus unveiled, offering opportunities to divide and conquer EGFR inhibitor resistance.
docetaxel plus placebo in patients with previously treated non-small-cell lung Cancer Discov 2012; 2: 872–875.
cancer (LUME-Lung 1): a phase 3, double-blind, randomised controlled trial. 121. Pao W, Miller VA, Politi KA et al. Acquired resistance of lung adenocarcinomas to
Lancet Oncol 2014; 15: 143–155. gefitinib or erlotinib is associated with a second mutation in the EGFR kinase
103. Novello S, Kaiser R, Mellemgaard A et al. Analysis of patient-reported outcomes domain. PLoS Med 2005; 2: e73.
from the LUME-Lung 1 trial: a randomised, double-blind, placebo-controlled, 122. Yun CH, Mengwasser KE, Toms AV et al. The T790M mutation in EGFR kinase
phase III study of second-line nintedanib in patients with advanced non-small cell causes drug resistance by increasing the affinity for ATP. Proc Natl Acad Sci USA
lung cancer. Eur J Cancer 2015; 51: 317–326. 2008; 105: 2070–2075.
104. Borghaei H, Paz-Ares L, Horn L et al. Nivolumab versus docetaxel in advanced 123. Jänne PA, Yang JCH, Kim DW et al. AZD9291 in EGFR inhibitor-resistant non-
nonsquamous non-small-cell lung cancer. N Engl J Med 2015; 373: small-cell lung cancer. N Engl J Med 2015; 372: 1689–1699.
1627–1639. 124. Mitsudomi T, Tsai CM, Shepherd FA et al. AZD9291 in pre-treated T790M
105. Mok TS, Wu YL, Thongprasert S et al. Gefitinib or carboplatin-paclitaxel in positive advanced NSCLC: AURA2 phase II study. J Thorac Oncol 2015; 10
pulmonary adenocarcinoma. N Engl J Med 2009; 361: 947–957. (Suppl. 2): S320.
106. Han JY, Park K, Kim SW et al. First-SIGNAL: first-line single-agent iressa versus 125. Soria JC, Wu YL, Nakagawa K et al. Gefitinib plus chemotherapy versus placebo
gemcitabine and cisplatin trial in never-smokers with adenocarcinoma of the plus chemotherapy in EGFR-mutation-positive non-small-cell lung cancer after
lung. J Clin Oncol 2012; 30: 1122–1128. progression on first-line gefitinib (IMPRESS): a phase 3 randomised trial. Lancet
107. Inoue A, Kobayashi K, Maemondo M et al. Updated overall survival results from a Oncol 2015; 16: 990–998.
randomized phase III trial comparing gefitinib with carboplatin-paclitaxel for 126. Douillard JY, Ostoros G, Cobo M et al. First-line gefitinib in Caucasian EGFR
chemo-naïve non-small cell lung cancer with sensitive EGFR gene mutations mutation-positive NSCLC patients: a phase-IV, open-label, single-arm study. Br J
(NEJ002). Ann Oncol 2013; 24: 54–59. Cancer 2014; 110: 55–62.
108. Maemondo M, Inoue A, Kobayashi K et al. Gefitinib or chemotherapy for non- 127. Marchetti A, Del Grammastro M, Felicioni L et al. Assessment of EGFR mutations
small-cell lung cancer with mutated EGFR. N Engl J Med 2010; 362: in circulating tumor cell preparations from NSCLC patients by next generation
2380–2388. sequencing: toward a real-time liquid biopsy for treatment. PLoS ONE 2014; 9:
109. Mitsudomi T, Morita S, Yatabe Y et al. Gefitinib versus cisplatin plus docetaxel in e103883.
patients with non-small-cell lung cancer harbouring mutations of the epidermal 128. Marchetti A, Palma JF, Felicioni L et al. Early prediction of response to tyrosine
growth factor receptor (WJTOG3405): an open label, randomised phase 3 trial. kinase inhibitors by quantification of EGFR mutations in plasma of NSCLC
Lancet Oncol 2010; 11: 121–128. patients. J Thorac Oncol 2015; 10: 1437–1443.
110. Sequist LV, Yang JCH, Yamamoto N et al. Phase III study of afatinib or cisplatin 129. Sueoka-Aragane N, Katakami N, Satouchi M et al. Monitoring EGFR T790 M with
plus pemetrexed in patients with metastatic lung adenocarcinoma with EGFR plasma DNA from lung cancer patients in a prospective observational study.
mutations. J Clin Oncol 2013; 31: 3327–3334. Cancer Sci 2016; 107: 162–167.
111. Wu YL, Zhou C, Hu CP et al. Afatinib versus cisplatin plus gemcitabine for first- 130. Oxnard GR, Thress KS, Alden RS et al. 135O_PR: plasma genotyping for
line treatment of Asian patients with advanced non-small-cell lung cancer predicting benefit from osimertinib in patients ( pts) with advanced NSCLC. J
harbouring EGFR mutations (LUX-Lung 6): an open-label, randomised phase 3 Thorac Oncol 2016; 11(Suppl. 4S): S154.
trial. Lancet Oncol 2014; 15: 213–222. 131. Park K, Yu CJ, Kim SW et al. First-line erlotinib therapy until and beyond
112. Rosell R, Carcereny E, Gervais R et al. Erlotinib versus standard chemotherapy as response evaluation criteria in solid tumors progression in Asian patients with
first-line treatment for European patients with advanced EGFR mutation-positive epidermal growth factor receptor mutation-positive non-small-cell lung cancer:
non-small-cell lung cancer (EURTAC): a multicentre, open-label, randomised the ASPIRATION study. JAMA Oncol 2016; 2: 305–312.
phase 3 trial. Lancet Oncol 2012; 13: 239–246. 132. Weickhardt AJ, Scheier B, Burke JM et al. Local ablative therapy of
113. Inoue A, Kobayashi K, Usui K et al. First-line gefitinib for patients with advanced oligoprogressive disease prolongs disease control by tyrosine kinase inhibitors in
non-small-cell lung cancer harboring epidermal growth factor receptor mutations oncogene-addicted non-small-cell lung cancer. J Thorac Oncol 2012; 7:
without indication for chemotherapy. J Clin Oncol 2009; 27: 1394–1400. 1807–1814.
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw326 | v

by guest
on 05 February 2018
133. Camidge DR, Bang YJ, Kwak EL et al. Activity and safety of crizotinib in patients (OS) results from the UK Medical Research Council QUARTZ randomised clinical
with ALK-positive non-small-cell lung cancer: updated results from a phase 1 trial (ISRCTN 3826061). J Clin Oncol 2015; 33(20 Suppl); abstr 8005.
study. Lancet Oncol 2012; 13: 1011–1019. 155. Tsao MN, Lloyd N, Wong RK et al. Whole brain radiotherapy for the treatment of
134. Shaw AT, Yeap BY, Solomon BJ et al. Effect of crizotinib on overall survival in newly diagnosed multiple brain metastases. Cochrane Database Syst Rev 2012;
patients with advanced non-small-cell lung cancer harbouring ALK gene 4: CD003869.
rearrangement: a retrospective analysis. Lancet Oncol 2011; 12: 1004–1012. 156. Robinet G, Thomas P, Breton JL et al. Results of a phase III study of early versus
135. Shaw AT, Kim DW, Nakagawa K et al. Crizotinib versus chemotherapy in delayed whole brain radiotherapy with concurrent cisplatin and vinorelbine
advanced ALK-positive lung cancer. N Engl J Med 2013; 368: 2385–2394. combination in inoperable brain metastasis of non-small-cell lung cancer: Groupe
136. Blackhall F, Kim DW, Besse B et al. Patient-reported outcomes and quality of life Français de Pneumo-Cancérologie (GFPC) Protocol 95-1. Ann Oncol 2001; 12:
in PROFILE 1007: a randomized trial of crizotinib compared with chemotherapy in 59–67.
previously treated patients with ALK-positive advanced non-small-cell lung 157. Lim SH, Lee JY, Lee MY et al. A randomized phase III trial of stereotactic
cancer. J Thorac Oncol 2014; 9: 1625–1633. radiosurgery (SRS) versus observation for patients with asymptomatic cerebral
137. Solomon BJ, Mok T, Kim DW et al. First-line crizotinib versus chemotherapy in oligo-metastases in non-small-cell lung cancer. Ann Oncol 2015; 26: 762–768.
ALK-positive lung cancer. N Engl J Med 2014; 371: 2167–2177. 158. Vecht CJ, Hovestadt A, Verbiest HB et al. Dose-effect relationship of
138. Costa DB, Kobayashi S, Pandya SS et al. CSF concentration of the anaplastic dexamethasone on Karnofsky performance in metastatic brain tumors: a
lymphoma kinase inhibitor crizotinib. J Clin Oncol 2011; 29: e443–e445. randomized study of doses of 4, 8, and 16 mg per day. Neurology 1994; 44:
139. Kim DW, Mehra R, Tan DS et al. Activity and safety of ceritinib in patients with 675–680.
ALK-rearranged non-small-cell lung cancer (ASCEND-1): updated results from 159. Schuler M, Wu YL, Hirsh V et al. First-line afatinib versus chemotherapy in
the multicentre, open-label, phase 1 trial. Lancet Oncol 2016; 17: 452–463. patients with non-small cell lung cancer and common epidermal growth factor
140. Ou S-HI, Ahn JS, De Petris L et al. Alectinib in crizotinib-refractory ALK- receptor gene mutations and brain metastases. J Thorac Oncol 2016; 11:
rearranged non-small-cell lung cancer: a phase II global study. J Clin Oncol 380–390.
2016; 34: 661–668. 160. Rosen LS, Gordon D, Tchekmedyian NS et al. Long-term efficacy and safety of
141. Shaw AT, Gandhi L, Gadgeel S et al. Alectinib in ALK-positive, crizotinib-resistant, zoledronic acid in the treatment of skeletal metastases in patients with nonsmall
non-small-cell lung cancer: a single-group, multicentre, phase 2 trial. Lancet cell lung carcinoma and other solid tumors: a randomized, phase III, double-
Oncol 2016; 17: 234–242. blind, placebo-controlled trial. Cancer 2004; 100: 2613–2621.
142. ALEX Study. A randomized, phase III study comparing alectinib with crizotinib in 161. Henry DH, Costa L, Goldwasser F et al. Randomized, double-blind study of
treatment-naive anaplastic lymphoma kinase-positive advanced non-small cell denosumab versus zoledronic acid in the treatment of bone metastases in
lung cancer participants. https://clinicaltrials.gov (5 May 2016, date last patients with advanced cancer (excluding breast and prostate cancer) or multiple
accessed). myeloma. J Clin Oncol 2011; 29: 1125–1132.
143. Katayama R, Lovly CM, Shaw AT. Therapeutic targeting of anaplastic lymphoma 162. Scagliotti GV, Hirsh V, Siena S et al. Overall survival improvement in patients with
kinase in lung cancer: a paradigm for precision cancer medicine. Clin Cancer lung cancer and bone metastases treated with denosumab versus zoledronic
Res 2015; 21: 2227–2235. acid: subgroup analysis from a randomized phase 3 study. J Thorac Oncol 2012;
144. Shaw P, Agarwal R. Pleurodesis for malignant pleural effusions. Cochrane 7: 1823–1829.
Database Syst Rev 2004; 1: CD002916. 163. Henry D, Vadhan-Raj S, Hirsh V et al. Delaying skeletal-related events in a
145. Dresler CM, Olak J, Herndon JE et al. Phase III intergroup study of talc poudrage randomized phase 3 study of denosumab versus zoledronic acid in patients with
vs talc slurry sclerosis for malignant pleural effusion. Chest 2005; 127: advanced cancer: an analysis of data from patients with solid tumors. Supp Care
909–915. Cancer 2014; 22: 679–687.
146. Davies HE, Mishra EK, Kahan BC et al. Effect of an indwelling pleural catheter vs 164. Zacho HD, Karthigaseu NN, Fonager RF, Petersen LJ. Treatment with bone-
chest tube and talc pleurodesis for relieving dyspnea in patients with malignant seeking radionuclides for painful bone metastases in patients with lung cancer: a
pleural effusion: the TIME2 randomized controlled trial. JAMA 2012; 307: systematic review. BMJ Support Palliat Care 2016; doi:10.1136/bmjspcare-
2383–2389. 2015-000957.
147. Temel JS, Greer JA, Muzikansky A et al. Early palliative care for patients with 165. Palma DA, Salama JK, Lo SS et al. The oligometastatic state - separating truth
metastatic non-small-cell lung cancer. N Engl J Med 2010; 363: 733–742. from wishful thinking. Nat Rev Clin Oncol 2014; 11: 549–557.
148. Sperduto PW, Kased N, Roberge D et al. Summary report on the graded 166. Hellman S, Weichselbaum RR. Oligometastases. J Clin Oncol 1995; 13: 8–10.
prognostic assessment: an accurate and facile diagnosis-specific tool to estimate 167. Ashworth AB, Senan S, Palma DA et al. An individual patient data metaanalysis
survival for patients with brain metastases. J Clin Oncol 2012; 30(4): 419–425. of outcomes and prognostic factors after treatment of oligometastatic non-small-
149. Patil CG, Pricola K, Sarmiento JM et al. Whole brain radiation therapy (WBRT) cell lung cancer. Clin Lung Cancer 2014; 15: 346–355.
alone versus WBRT and radiosurgery for the treatment of brain metastases. 168. Detterbeck FC, Franklin WA, Nicholson AG et al. The IASLC Lung Cancer Staging
Cochrane Database Syst Rev 2012; 9: CD006121. Project: background data and proposed criteria to distinguish separate primary
150. Andrews DW, Scott CB, Sperduto PW et al. Whole brain radiation therapy with or lung cancers from metastatic foci in patients with two lung tumors in the
without stereotactic radiosurgery boost for patients with one to three brain forthcoming eighth edition of the TNM classification for lung cancer. J Thorac
metastases: phase III results of the RTOG 9508 randomised trial. Lancet 2004; Oncol 2016; 11: 651–665.
363: 1665–1672. 169. Kozower BD, Larner JM, Detterbeck FC, Jones DR. Special treatment issues in
151. Soon YY, Tham IW, Lim KH et al. Surgery or radiosurgery plus whole brain non-small cell lung cancer: diagnosis and management of lung cancer, 3rd ed:
radiotherapy versus surgery or radiosurgery alone for brain metastases. Cochrane American College of Chest Physicians evidence-based clinical practice
Database Syst Rev 2014; 3: CD009454. guidelines. Chest 2013; 143(Suppl. 5): e369S–e399S.
152. Yamamoto M, Serizawa T, Shuto T et al. Stereotactic radiosurgery for patients 170. Tönnies M, Pfannschmidt J, Bauer TT et al. Metastasectomy for synchronous
with multiple brain metastases (JLGK0901): a multi-institutional prospective solitary non-small cell lung cancer metastases. Ann Thorac Surg 2014; 98:
observational study. Lancet Oncol 2014; 15: 387–395. 249–256.
153. Sahgal A, Aoyama H, Kocher M et al. Phase 3 trials of stereotactic radiosurgery 171. Griffioen GHMJ, Lagerwaard FJ, Haasbeek CJA et al. Treatment of multiple
with or without whole-brain radiation therapy for 1 to 4 brain metastases: primary lung cancers using stereotactic radiotherapy, either with or without
individual patient data meta-analysis. Int J Radiat Oncol Biol Phys 2015; 91: surgery. Radiother Oncol 2013; 107: 403–408.
710–717. 152. 172. Chang JY, Liu YH, Zhu Z et al. Stereotactic ablative radiotherapy: a potentially
154. Mulvenna PM, Nankivell MG, Barton R et al. Whole brain radiotherapy for brain curable approach to early stage multiple primary lung cancer. Cancer 2013; 119:
metastases from non-small lung cancer: quality of life (QoL) and overall survival 3402–3410.
v | Novello et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
173. De Ruysscher D, Wanders R, van Baardwijk A et al. Radical treatment of non- 178. Weber J. Review: anti-CTLA-4 antibody ipilimumab: case studies of clinical
small-cell lung cancer patients with synchronous oligometastases: long-term response and immune-related adverse events. Oncologist 2007; 12: 864–872.
results of a prospective phase II trial (Nct01282450). J Thorac Oncol 2012; 7: 179. Saenger YM, Wolchok JD. The heterogeneity of the kinetics of response to
1547–1555. ipilimumab in metastatic melanoma: patient cases. Cancer Immun 2008; 8: 1.
174. Cheufou DH, Welter S, Chalvatzoulis E et al. Surgery of primary lung cancer with 180. Postel-Vinay S, Aspeslagh S, Lanoy E et al. Challenges of phase 1 clinical trials
oligometastatic m1b synchronous single brain metastasis: analysis of 37 cases. evaluating immune checkpoint-targeted antibodies. Ann Oncol 2016; 27:
Thorac Cardiovasc Surg 2014; 62: 612–615. 214–224.
175. Eberhardt WEE, Mitchell A, Crowley J et al. The IASLC Lung Cancer Staging 181. Cherny NI, Sullivan R, Dafni U et al. A standardised, generic, validated approach
Project: proposals for the revision of the m descriptors in the forthcoming eighth to stratify the magnitude of clinical benefit that can be anticipated from anti-
edition of the TNM classification of lung cancer. J Thorac Oncol 2015; 10: cancer therapies: the European Society for Medical Oncology Magnitude of
1515–1522. Clinical Benefit Scale (ESMO-MCBS). Ann Oncol 2015; 26: 1547–1573.
176. Nishino M, Giobbie-Hurder A, Gargano M et al. Developing a common language 182. Dykewicz CA, Centers for Disease Control and Prevention (U.S.), Infectious
for tumor response to immunotherapy: immune-related response criteria using Diseases Society of America, American Society of Blood and Marrow
unidimensional measurements. Clin Cancer Res 2013; 19: 3936–3943. Transplantation. Summary of the guidelines for preventing opportunistic
177. Nishino M, Gargano M, Suda M et al. Optimizing immune-related tumor response infections among hematopoietic stem cell transplant recipients. Clin Infect Dis
assessment: does reducing the number of lesions impact response assessment 2001; 33(2): 139–144.
in melanoma patients treated with ipilimumab? J Immunother Cancer 2014; 183. Edge SB, Byrd DR, Compton CC (eds). AJCC Cancer Staging Handbook, 7th
2: 17. edition. New York, NY: Springer 2010
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw326 | v

by guest
on 05 February 2018
Published online 27 June 2013
Small-cell lung cancer (SCLC): ESMO Clinical Practice

M. Früh1, D. De Ruysscher2, S. Popat3, L. Crinò4, S. Peters5 & E. Felip6, on behalf of the ESMO
Guidelines Working Group*
1
Department of Medical Oncology and Hematology, Kantonsspital St Gallen, Switzerland; 2Radiation Oncology, University Hospitals Leuven/KU Leuven, Belgium; 3Royal
Marsden Hospital, London, UK; 4Department of Oncology, Hospital Santa Maria della Misericordia, Sant Andrea delle Fratte, Perugia, Italy; 5Département d’Oncologie,
Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland; 6Medical Oncology Service, Vall d’Hebron University Hospital, Barcelona, Spain;
incidence and epidemiology diagnosis and pathology/molecular

An estimated 1.6 million new lung cancers are diagnosed biology
worldwide each year. The highest incidence rates in males are Pathological diagnosis should be made according to the
observed in Central/Eastern and Southern Europe (57 and 49 World Health Organisation (WHO) classification using
per 100 000, respectively), whereas in women the highest morphology (uniform round to spindled-shaped small cells,
rates are found in Northern Europe (36 per 100 000) [1]. sparse cytoplasm, high mitotic index, necrotic areas).
clinical practice
Five-year survival rates of lung cancer patients have only Immunohistochemistry to confirm the diagnosis of SCLC
guidelines
slightly improved during the past decade but remain low at (synaptophysin, chromogranin A, CD56, thyroid
10% [2]. transcription factor 1 and MIB-1) is not mandatory, but
Small-cell lung cancer (SCLC) originates from should be used in case of any doubt (e.g. in case of
neuroendocrine-cell precursors and is characterised by its pronounced crush artefacts). Due to its frequent central
rapid growth, its high response rates to both chemotherapy localisation within the chest, biopsies may best be obtained by
and radiotherapy and development of treatment resistance in bronchoscopy. Other methods include mediastinoscopy,
patients with metastatic disease. In the Western world, the endobronchial ultrasound (EBUS), endoscopic ultrasound,
proportion of patients with SCLC has decreased to 13% [3]. transthoracic needle aspiration or even thoracoscopy if
Virtually all patients have a history of tobacco use. Therefore, necessary. A biopsy from a metastatic lesion may be the
smoking habits are closely linked to incidence, which varies preferred option if the location of the metastasis is easily and
across different populations. In addition, the new description safely accessible to biopsy, as this will also pathologically stage
of large-cell neuroendocrine tumours in the 1990s, which the patient (e.g. liver, skin).
may have been summarised previously as SCLC, possibly has
contributed to the decline. Smoking cessation not only
reduces the risk of developing SCLC but also has been shown staging and risk assessment
to decrease the risk of death of patients with localised SCLC
by almost 50% [4]. Only one-third of the patients are The prognosis of SCLC strongly depends on the tumour stage.
diagnosed with localised disease, where cure is the treatment The new tumour-node-metastasis (TNM) version 7 staging
goal. Due to the aggressive natural course, screening by system according to the Union for International Cancer Control
radiological imaging is unlikely to lead to a reduction of (UICC) as adopted for non-small-cell lung cancer should also
mortality, and smoking prevention will undoubtedly remain be used for SCLC [I, A] [6,7] (See Tables 1 and 2). This
the primary and most important intervention to further classification should replace the former 1989 International
decrease mortality [5]. Association for the Study of Lung Cancer (IASLC) staging
system, which defined limited stage as tumour being confined to
one hemithorax with regional lymph node metastasis including
both ipsilateral and contralateral hilar, supraclavicular and
*Correspondence to: ESMO Guidelines Working Group, ESMO Head Office, Via,
L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland;
mediastinal nodes, as well as ipsilateral pleural effusion. The
E-mail: clinicalguidelines@esmo.org current TNM staging system is based on 8088 SCLC patients
and provides better prognostic information and more precise
†
Approved by the ESMO Guidelines Working Group: February 2002, last update May
nodal staging, which is required for conformal radiation
2013. This publication supersedes the previously published version—Ann Oncol 2010;
21 (Suppl. 5): v120–v125. techniques and intensity-modulated radiation therapy. The

by guest
on 05 February 2018
Table 1. Tumour node metastasis classification. Initial assessment should encompass medical history
including smoking history, physical examination, complete
TX Positive cytology only blood count including differential count, liver enzymes, sodium,
T1 ≤3 cm potassium, calcium, glucose, lactate dehydrogenase levels and
T1a ≤2 cm renal function tests, and in the case of localised disease, lung
T1b >2 to 3 cm function tests. An initial computed tomography (CT) scan with
T2 Main bronchus ≥2 cm from carina invades visceral pleura, partial contrast of the chest and abdomen is recommended. If the
atelectasis metastatic stage is not obvious on the CT scan or clinical
T2a >3–5 cm findings suggest bone or brain involvement, further imaging
T2b >5–7 cm with bone scintigraphy and CT or magnetic resonance imaging
T3 >7 cm; chest wall, diaphragm, pericardium, mediastinal pleura, (MRI) of the brain are recommended. In case of abnormal
main bronchus <2 cm from carina, total atelectasis, separate blood count or signs of blood–bone marrow barrier rupture
nodule(s) in the same lobe (e.g. peripheral blood erythroblasts), a bone marrow aspiration
T4 Mediastinum, heart, great vessels, carina, trachea, esophagus, and biopsy may be indicated, particularly in patients with
vertebra; separate tumour nodule(s) in a different ipsilateral lobe otherwise absent metastases [V, C]. Alternatively to CT and
N1 Ipsilateral peribronchial, ipsilateral hilar
bone scintigraphy, a 2-fluor-2-deoxy-D-glucose positron-
N2 Subcarinal, ipsilateral mediastinal
emission-tomography (FDG-PET) CT scan can be carried out.
N3 Contralateral mediastinal or hilar, scalene or supraclavicular
A recent review has suggested that with PET-CT 9% of the
patients are up- and 4% downstaged [8]; however, individual
M1a Separate tumour nodule(s) in a contralateral lobe; pleural nodules
studies in this analysis were non-randomised and were either
or malignant pleural, or pericardial effusion
retrospective or small and frequently lacked histological
confirmation. Thus, PET-CT findings which could impact
Lababede O, Meziane M, Rice T. Seventh Edition of the Cancer Staging treatment decisions should be pathologically confirmed [III, C].
Manual and Stage Grouping of Lung Cancer. Chest 2011; 139: 183–189. In patients with a solitary metastasis, its pathological
Reproduced with permission from the American College of Chest confirmation should not delay treatment start. In this case, the
Physicians. solitary metastatic lesion’s size should be re-evaluated after two
cycles, allowing further judgement as to whether it is a true
metastatic site [V, C]. Alternatively, an initial second
radiological method (e.g. MRI if solitary small liver or bone
Table 2. Tumour stage grouping. lesion) is recommended [V, C]. If a pleural or pericardial
effusion is the only site of M1, no malignant cells are identified
Occult carcinoma TX N0 M0 in the pleural fluid and a plausible explanation other than
Stage 0 Tis N0 M0 tumour involvement is clinically suspected, treatment should be
Stage IA T1a,b N0 M0 according to an M0 status [V, B].
Stage IB T2a N0 M0
Stage IIA T2b N0 M0
T1a,b N1 M0
T2a N1 M0
management of localised disease
Stage IIB T2b N1 M0 (t1-4, n0-3 m0)
T3 N0 M0 In localised disease, median survival and 2-year survival rates
Stage IIIA T1a,b, T2a,b N2 M0
have been reported to be 15–20 months and 20%–40%
T3 N1, N2 M0
respectively [9]. Importantly, the proportion of patients who
T4 N0, N1 M0
survive for 5 years has been reported to be 20%–25% [10].
Stage IIIB T4 N2 M0
Approximately 5% of patients with SCLC present as T1, 2 N0,
Any T N3 M0
Stage IV Any T Any N M1
1 M0 tumours (Figure 1). These patients have more favourable
outcomes with 5-year survival rates in the order of 50% [11, 12].
Lababede O, Meziane M, Rice T. Seventh Edition of the Cancer Staging Most series report on patients having been treated with surgery
Manual and Stage Grouping of Lung Cancer. Chest 2011; 139: 183–189. for a coin lesion without pathological diagnosis. A surgical
Reproduced with permission from the American College of Chest approach in this group of patients is justified after ruling out
Physicians. mediastinal lymph node involvement (i.e. negative lymph nodes
on CT scan, PET-CT scan or EBUS and/or mediastinoscopy if
enlarged) [V, C]. Postoperatively, four cycles of adjuvant
chemotherapy should be administered [III, C]. In the case of
former term limited stage would now include T1-4, N0-3 M0 unforeseen N2 or N1 or in patients who have not undergone
tumours, whereas metastatic tumours encompass former systematic nodal dissection, postoperative radiotherapy should
extensive stage patients. In addition, T1 or T2 N0 or N1 M0 be considered [V, C]. There is no role for surgery after induction
tumours ( previously described as ‘very limited stage’) were chemotherapy in N2 disease [II, B]. In the absence of
identified as a group with a more favourable outcome compared randomised trials, due to frequent early dissemination and
with patients with N2 or N3 disease. because total gross tumour volume has shown to be an
vi | Früh et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
Figure 1. Small-cell lung cancer (SCLC) treatment algorithm.
independent prognostic factor leading to improved outcomes radiotherapy was <30 days [hazard ratio (HR): 0.62, 95%
irrespective of the local treatment modality, patients with T1, 2 confidence interval (CI) 0.49–0.80, P = 0.0003] [16]. An update
N0, 1 M0 may alternatively be treated with combined of an American trial reached the same conclusion [17]. On the
concurrent chemoradiotherapy [III, C] [13]. This treatment is other hand, a recent randomised trial did not show any survival
recommended as the first option in patients who are at difference when radiotherapy was administered with the third as
increased risk for perioperative complications (e.g. significant opposed to the first cycle with less toxicity in the late arm in an
concomitant medical illnesses) [II, C]. All patients with T1, 2 Asian population [18]. Starting chest radiotherapy within 30
N0, 1 M0 should be considered for prophylactic cranial days after the beginning of chemotherapy is preferred [II, B].
irradiation (PCI) if they have responded to initial treatment When the general condition of the patient does not allow for the
using the same dose and fractionation as for patients with stage immediate administration of concurrent treatment or lung
III SCLC. constraints preclude the target radiotherapy dose, chest
All other patients with T1-4, N0-3 M0 tumours who are in a irradiation may be postponed until the start of the third cycle of
good performance status (PS) should be treated with concurrent chemotherapy [II, B].
chemotherapy and thoracic radiotherapy [I, A]. Several The optimal target volume remains to be defined. Omission
radiotherapy schedules have been studied. One phase III trial of of elective node irradiation based on CT scans should be used
471 patients reported a superior 5-year overall survival (OS) with caution as this strategy may result in nodal failures [III, C].
with twice-daily radiotherapy (1.5 Gy twice-daily, 30 fractions) Whether selective node irradiation based on pre-treatment
compared with once-daily (1.8 Gy, 25 fractions) of 26% versus PET-CT scans can replace elective node irradiation has been
16% (P = 0.04) [10]. The inconvenience of the twice-daily addressed in two small studies [19, 20]. Both studies, one
administration and the significantly increased rate of transient prospective and the other one retrospective, have shown
grade 3 oesophagitis were, however, the main reasons why this promisingly low nodal recurrence rates. This strategy, however,
regimen was not widely adopted. This current accelerated needs further prospective evaluation although it has been
standard schedule is being compared with 70 Gy in daily adopted already in some national guidelines [III, D]. Elective
fractions as an experimental arm in ongoing North American nodal volumes are not well-defined but may include the
and European phase III trials in patients in which the lung dose involved lymph node regions and one adjacent region and
can be kept within safe limits. Outside of a clinical trial, a twice- supraclavicular regions depending on the location of the
daily 1.5 Gy in 30-fraction regimen should be considered in fit primary tumour and the N2 or N3 nodes.
patients who are willing to accept temporarily increased toxicity RECIST criteria are not well-suited to determine tumour
[I, B]. The chemotherapy schedule consists of four cycles of response after radiotherapy. Patients in a reasonably good PS
cisplatin–etoposide or 4–6 cycles if a once-daily radiotherapy without progression should be offered PCI. The recommended
schedule is used [I, B]. dose is 25 Gy in 10 daily fractions [I, A]. Although PCI
The optimal timing of the concurrent radiotherapy has been increases long-term survival, patients >65 years and/or with
studied extensively. Seven older trials assessing the timing of important vascular disease have a slightly elevated risk (HR
thoracic radiotherapy were analysed in two meta-analyses, with 1.04) of developing neurocognitive side-effects [21, 22].
the conclusion that thoracic radiotherapy should be initiated as
early as possible beginning with the first or second cycle when
cisplatin-based chemotherapy was used [14, 15]. In addition, an management of metastatic disease
analysis of four of these studies which reported 5-year survival
rates and used two concurrent arms with cisplatin–etoposide first-line treatment
treatment found improved 5-year survival rates if the time Treatment of stage IV SCLC is palliative, and combination
between the first day of chemotherapy and the last day of chemotherapy has been the main treatment option for more
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt178 | vi

by guest
on 05 February 2018
than three decades. Despite response rates (RRs) close to 70%, lacking. In addition, there is a considerable risk of increased
outcomes remain poor with a median progression-free toxicity with prolonged platinum-based chemotherapy.
survival (PFS) of only 5.5 months and a median OS of <10 Continuing chemotherapy beyond 4–6 cycles of first-line
months [22, 23]. treatment is not recommended [II, B].
A meta-analysis of 19 randomised trials with a total of 4054 PCI significantly decreases the risk of symptomatic brain
patients demonstrated prolonged OS of patients receiving a metastases (from 40.4% to 14.6% at 1 year) and increases OS
cisplatin-containing regimen compared with older (HR 0.68; 95% CI, 0.52–0.88) [36]. Of note, in this trial initial
chemotherapy combinations [25]. Another meta-analysis of 36 pre-treatment brain imaging was not required. PCI is associated
trials reported an OS benefit in favour of etoposide alone or in with adverse effects such as fatigue and hair loss, and health-
combination with cisplatin compared with regimens that did related quality of life may be negatively affected as well [37].
not contain one of the two drugs [26]. These results led to the Patients with any response to first-line treatment and who have
adoption of etoposide–cisplatin as a standard treatment a reasonably good PS should be evaluated for PCI [II, B]. The
regimen. A recent individual patient data meta-analysis PCI dose may be 25 Gy in 10 daily fractions or 20 Gy in 5
including four randomised clinical trials comparing cisplatin fractions.
versus carboplatin-based combination chemotherapy Due to the often centrally located primary tumours,
demonstrated no difference in efficacy outcomes including RR, symptoms such as dyspnoea, infections due to atelectasis, chest
PFS and OS [24]. In the carboplatin group, increased pain or superior vena cava syndrome are frequent and make the
haematological toxicity rates were observed, whereas higher incorporation of thoracic radiotherapy into the initial treatment
renal and neurotoxicity was seen with cisplatin. According to algorithm an appealing concept. A four-arm randomised phase
these results, cisplatin can be substituted by carboplatin in III trial has demonstrated a survival benefit of concurrent
patients with metastatic SCLC [I, B]. Due to the limited number thoracic radiotherapy in patients whose primary tumours have
of only 663 patients included in this analysis, there was limited responded after three cycles of cisplatin–etoposide and whose
statistical power to draw conclusions in important subgroups metastatic sites were in complete remission (OS: 17 versus 11
such as patients with localised disease and young patients. In months, P = 0.041) [38]. This single centre trial was however
these subgroups, etoposide–cisplatin is recommended [II, B]. small (54 patients per arm), and the concurrent
Studies with 3-drug regimens and the administration of chemoradiotherapy treatment used does not correspond to the
increased dose intensity regimens, using increased dose or non- current standard approach. The routine use of thoracic
cross-resistant regimens, have not consistently reported irradiation in patients with metastatic SCLC is not
improvement in OS. In addition, they have frequently been recommended and the results of the Dutch phase III trial
associated with significant toxicity in this usually co-morbid (CREST study) testing this concept should be awaited [II, C].
patient population [27]. Such regimens are not recommended as
first-line treatment [II, C].
A recent literature-based meta-analysis of seven randomised second-line treatment
studies showed an improved OS, but not PFS with irinotecan– RRs to second-line treatment depend on the treatment-free
platinum compared with etoposide–platinum. Irinotecan led to interval and are usually in the order of 10% in resistant disease
more gastrointestinal toxic effects, while more haematological (i.e. progression-free interval <3 months) and 20% in sensitive
toxic effects were observed with etoposide [28]. The results, disease (i.e. interval >3 months). In refractory patients (i.e.
however, were primarily driven by Asian studies, and patients not responding or progressing during chemotherapy)
pharmacogenomic differences between Asian and Western and resistant patients with early relapse (<6 weeks), outcomes
populations possibly contributing to these differential outcomes are poor and the clinical benefit of further systemic therapy is
have previously been described [29]. No chemotherapy doublet uncertain. For these patients, participation in a clinical trial or
has yet been shown to be superior to i.v. etoposide–platinum in best supportive care is recommended [II, C]. Oral topotecan led
a Western population. Randomised phase III trials which to better symptom control including slower time to quality of
compared irinotecan–cisplatin, gemcitabine–carboplatin (in life deterioration and improved survival compared with best
poor prognostic patients only) or i.v. or oral topotecan–cisplatin supportive care in a study in which half of the patients had
to etoposide–platinum have demonstrated non-inferiority for resistant disease [39]. Prior to topotecan development,
survival [30–33]. These regimens are recommended as anthracycline-based regimes have been commonly used,
alternative treatment options in the case of contraindications to including cyclophosphamide, doxorubicin and vincristine
etoposide [II, C]. (CAV). In 1999, a trial of i.v. topotecan and CAV demonstrated
Continuation of chemotherapy beyond 4–6 cycles has been equal efficacy, with similar RRs, time-to-progression, and OS,
assessed in at least 14 randomised, controlled trials. Although a and better tolerance when compared with CAV [40]. Oral and
significant OS benefit was reported in a literature-based review i.v. topotecan have shown to be equally effective [41], but with
including 11 trials (HR 0.89, 95% CI: 0.81–0.92; P = 0.02), the differing toxicity profiles. Either oral or i.v. topotecan are
benefit was small and high heterogeneity among the included recommended for patients having resistant or sensitive relapse
trials was observed [34]. Similarly, a previous meta-analysis with CAV being an alternative option [II, B]. Only patients with
found a small OS benefit of 4% at 2 years with maintenance sensitive disease derive benefit from rechallenge with first-line
therapy [35]. However, the majority of the randomised, therapy (usually platinum–etoposide) [V, C].
controlled trials did not show any significant OS benefit, and a A recent randomised, phase III trial failed to show a survival
properly designed large clinical trial to address this question is benefit of amrubicin versus topotecan, despite a higher RR and
vi | Früh et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
Table 3. Summary of recommendations.
Diagnosis • Pathological diagnosis should be made according to the World Health Organisation (WHO) classification
• Biopsies are best obtained by bronchoscopy. A biopsy from a metastatic lesion is preferred if the location of the
metastasis can be easily and safely accessed to biopsy (e.g. liver, skin)
• No predictive molecular marker for treatment selection is currently available
Staging and risk assessment • Initial assessment should include smoking history, physical examination, complete blood count, liver enzymes, sodium,
potassium, calcium, glucose, lactate dehydrogenase levels and lung (if localised disease) and renal function tests
• A computed tomography (CT) scan with contrast of the chest and abdomen is recommended
• In localised disease or if symptoms or clinical findings suggest involvement, additional bone scintigraphy and CT or MRI
of the brain are recommended
• 2-fluor-2-desoxy-D-glucose positron-emission-tomography (FDG-PET CT) scan is optional in localised disease. PET
findings, which modify treatment decisions, should be pathologically confirmed [III, C]
• A bone marrow aspiration and biopsy should be carried out in the case of abnormal blood counts suggesting
involvement, particularly in localised disease [V, C]
• Version 7 of the TNM staging system according to the Union for International Cancer Control (UICC) should be used
(Tables 1 and 2) [I, A]
Treatment strategy • Figure 1 summarises the treatment algorithm of patients with SCLC
• In localised disease, a bimodality treatment approach is curative and chemotherapy plus radiotherapy result in 5-year
survival rates of 20%–25%
• Treatment of stage IV SCLC is palliative and various combination chemotherapy regimens demonstrate similarly high
response rates (RRs) of 60%–70%. Due to frequent rapid relapse and limited activity of second-line treatment, overall
survival (OS) remains poor (<10 months)
• All SCLC patients responding to first-line treatment should be evaluated for prophylactic cranial irradiation (PCI)
Treatment of localised • A small subset of patients who present with T1, 2 N0, 1 M0 tumours have a more favourable outcome and 5-year
disease survival rates of 50% have been reported with surgery. These patients should receive four cycles of adjuvant chemotherapy
[III, C] and postoperative thoracic radiotherapy if staged pN1 or pN2 [V, C]
• All other patients with T1-4, N0-3 M0 tumours who are in a good performance status (PS) should be treated with
concurrent chemotherapy and thoracic radiotherapy [I, A]
• The best OS rates in fit patients were demonstrated with twice-daily 1.5 Gy in 30 fractions given concurrently with four
cycles of cisplatin and etoposide [I, B]
• Patients who are not fit enough for twice-daily radiotherapy or are unwilling to accept increased toxic effects may be
treated with a once-daily radiotherapy schedule with 4–6 cycles of concurrent etoposide–cisplatin [I, B]
• In good PS patients, thoracic radiotherapy should be initiated with the first or second cycle (i.e. within 30 days) of
chemotherapy [II, B]
• All patients with T1-4, N0-3 M0 disease without disease progression after treatment and a reasonably good PS should be
offered PCI [I, A]
First-line treatment of • 4–6 cycles of etoposide plus cisplatin or carboplatin are recommended [I, B]
metastatic disease • In young patients and patients with localised disease, etoposide–cisplatin is recommended [II, B]
• Irinotecan–cisplatin, gemcitabine–carboplatin (in poor prognostic patients only) and i.v. or oral topotecan–cisplatin are
alternative options if etoposide is contraindicated [II, C]
• Patients in a reasonably good PS with any response to first-line treatment should be evaluated for PCI [II, B]
• The routine use of thoracic irradiation in patients with metastatic SCLC is not recommended [II, C]
Second-line treatment of • For refractory patients and resistant patients with early relapse (<6 weeks), participation in a clinical trial or best
metastatic disease supportive care is recommended [II, C]
• Oral or i.v. topotecan are recommended for patients having resistant or sensitive relapse with CAV being an alternative
option [II, B]
• Patients with sensitive relapse may derive benefit from reintroduction of the first-line regimen (usually platinum–
etoposide) [V, C]
Follow-up and long-term • The occurrence of second malignancies, particularly if smoking is continued, is of concern in survivours and smoking
implications cessation counselling is essential
• Two to three-monthly CT scans are recommended in patients with metastatic disease potentially qualifying for further
treatments [V, C]
• Six-monthly CT scans for 2 years with lengthening of intervals thereafter are recommended for patients with non-
metastatic disease who have received potentially curative treatment [V, C]
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt178 | vi

by guest
on 05 February 2018
Table 4. LOE and GOR adapted from the Infectious Diseases Society of America-United States Public Health Service Grading System†
Levels of evidence
I Evidence from at least one large randomised control trial of good methodological quality (low potential for bias) or meta-analyses of well-conducted
IV Retrospective cohort studies or case-control studies
Summary of recommendations
C Insufficient evidence for efficacy or benefit does not outweigh the risk or the disadvantages (adverse events, costs,.), optional
†Dykewicz CA. Summary of the guidelines for preventing opportunistic infections among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001;
33: 139-144. By permission of the Infectious Diseases Society of America.
improved quality of life with amrubicin [42]. The subgroup of conflict of interest
refractory patients derived a small survival benefit from
amrubicin. Amrubicin is currently not available in Western Dr Peters has reported consultancy/honoraria from Roche, Eli
countries. Lilly, AstraZeneca, Pfizer, Boehringer-Ingelheim, Bristol-Myers
Squibb, Daiichi-Sankyo, and Tesaro. Dr Felip has reported
consultancy/honoraria from Lilly, GlaxoSmithKline, Pfizer,
personalised medicine Roche, Boehringer Ingelheim. The other authors have declared
no potential conflicts of interest.
In this disease setting, more research is needed to identify
molecular markers which could lead to advances in
personalised medicine. references
1. Jemal A, Bray F, Center MM et al. Global cancer statistics. CA Cancer J Clin 2011;
61: 69–90.
follow-up and long-term implications 2. Sant M, Allemani C, Santiaquilani M et al. EUROCARE-4. Survival of cancer
patients diagnosed in 1995-1999. Results and commentary. Eur J Cancer 2009;
All patients with metastatic SCLC and approximately three- 45: 931–991.
quarters of patients with localised disease will progress. In 3. Govindan R, Page N, Morgensztern D et al. Changing epidemiology of small-cell
survivors, the occurrence of second malignancies, particularly if lung cancer in the United States over the last 30 years: analysis of the
smoking is continued, is of concern and smoking cessation surveillance, epidemiologic, and end results database. J Clin Oncol 2006; 24:
counselling is essential. The main goal of regular follow-up is to 4539–4544.
detect recurrence early, while the patient is still in a good PS 4. Parsons A, Daley A, Begh R et al. Influence of smoking cessation after diagnosis of
early stage lung cancer on prognosis: systematic review of observational studies
[43]. The frequency of follow-up visits depends on the
with meta-analysis. BMJ 2010; 340: b5569.
availability of treatment options. Although there is no clinical 5. Cuffe S, Moua T, Summerfield R et al. Characteristics and outcomes of small
trial evaluating the benefit of regular follow up, 2–3-monthly cell lung cancer patients diagnosed during two lung cancer computed
CT scans are recommended in patients with metastatic disease tomographic screening programs in heavy smokers. J Thorac Oncol 2011; 6:
potentially qualifying for further treatments. Patients with 818–822.
localised disease who have received potentially curative 6. Lababede O, Meziane M, Rice T. Seventh edition of cancer staging manual and
treatment should undergo 3–6-monthly CT scans for two years stage grouping of lung cancer: quick reference chart and diagrams. Chest 2011;
with lengthening of intervals thereafter. Due to the high risk of 139: 183–189.
secondary primary lung cancer, annual low-dose CT scans after 7. Shepherd FA, Crowley J, Van Houtte P et al. The International Association for the
study of lung cancer lung cancer staging project: proposals regarding the clinical
5 years might be considered [V, C]. Summary of staging of small cell lung cancer in the forthcoming (seventh) edition of the tumor,
recommendations is provided in Table 3. node, metastasis classification for lung cancer. J Thorac Oncol 2007; 2:
1067–1077.
8. Thomson D, Hulse P, Lorigan P et al. The role of positron emission tomography
note in management of small cell lung cancer. Lung Cancer 2011; 73: 121–126.
9. van Meerbeeck JP, Fennell DA, De Ruysscher DK. Small-cell lung cancer. Lancet
Levels of evidence and grades of recommendation have been 2011; 378: 1741–1755.
applied using the system shown in Table 4. Statements without 10. Turrisi AT, 3rd, Kim K, Blum R et al. Twice-daily compared with once-daily thoracic
grading were considered justified standard clinical practice by radiotherapy in limited small-cell lung cancer treated concurrently with cisplatin
the experts and the ESMO faculty. and etoposide. N Engl J Med 1999; 340: 265–271.
vi | Früh et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
11. Yu JB, Decker RH, Detterbeck FC et al. Surveillance epidemiology and end results with methodology assessment and meta-analysis. Lung Cancer 2000; 30:
evaluation of the role of surgery for stage I small cell lung cancer. J Thorac Oncol 23–26.
2010; 5: 215–219. 27. Popat S, O’Brien M. Chemotherapy strategies in the treatment of small cell lung
12. Schreiber D, Rineer J, Weedon J et al. Survival outcomes with the use of surgery cancer. Anticancer Drugs 2005; 16: 361–372.
in limited-stage small cell lung cancer: should its role be re-evaluated? Cancer 28. Shao N, Jin S, Zhu W. An updated meta-analysis of randomized controlled trials
2010; 116: 1350–1357. comparing irinotecan/platinum with etoposide/platinum in patients with previously
13. Reymen B, Van Loon J, van Baardwijk A et al. Total gross tumor volume is an untreated extensive-stage small cell lung cancer. J Thorac Oncol 2012; 7: 470–472.
independent prognostic factor in patients treated with selective nodal irradiation for 29. Lara PN, Jr, Natale R, Crowley J et al. Phase III trial of irinotecan/cisplatin
stage I to III small cell lung cancer. Int J Radiat Oncol Biol Phys 2013; 85: compared with etoposide/cisplatin in extensive-stage small-cell lung cancer:
1319–1324. clinical and pharmacogenomic results from SWOG S0124. J Clin Oncol 2009; 27:
14. Fried DB, Morris DE, Poole C et al. Systematic review evaluating the timing of 2530–2535.
thoracic radiation therapy in combined modality therapy for limited-stage small-cell 30. Zatloukal P, Cardenal F, Szczesna A et al. A multicenter international randomized
lung cancer. J Clin Oncol 2004; 22: 4837–4845. phase III study comparing cisplatin in combination with irinotecan or etoposide in
15. Pijls-Johannesma M, De Ruysscher D, Vansteenkiste J et al. Timing of chest previously untreated small-cell lung cancer patients with extensive disease. Ann
radiotherapy in patients with limited stage small cell lung cancer: a systematic Oncol 2010; 21: 1810–1816.
review and meta-analysis of randomised controlled trials. Cancer Treat Rev 2007; 31. Fink TH, Huber RM, Heigener DF et al. Topotecan/cisplatin compared with
33: 461–473. cisplatin/etoposide as first-line treatment for patients with extensive disease small-
16. De Ruysscher D, Pijls-Johannesma M, Bentzen SM et al. Time between the first cell lung cancer: final results of a randomized phase III trial. J Thorac Oncol 2012;
day of chemotherapy and the last day of chest radiation is the most important 7: 1432–1439.
predictor of survival in limited-disease small-cell lung cancer. J Clin Oncol 2006; 32. Eckardt JR, von Pawel J, Papai Z et al. Open-label, multicenter, randomized,
24: 1057–1063. phase III study comparing oral topotecan/cisplatin versus etoposide/cisplatin as
17. Blackstock AW, Bogart JA, Matthews C et al. Split-course versus continuous treatment for chemotherapy-naive patients with extensive-disease small-cell lung
thoracic radiation therapy for limited-stage small-cell lung cancer: final report of a cancer. J Clin Oncol 2006; 24: 2044–2051.
randomized phase III trial. Clin Lung Cancer 2005; 6: 287–292. 33. Lee SM, James LE, Qian W et al. Comparison of gemcitabine and carboplatin
18. Sun JM, Ahn YC, Choi EK et al. Phase III trial of concurrent thoracic radiotherapy versus cisplatin and etoposide for patients with poor-prognosis small cell lung
with either first- or third-cycle chemotherapy for limited-disease small-cell lung cancer. Thorax 2009; 64: 75–80.
cancer. Ann Oncol 2013; 24: 2088–2092. 34. Rossi A, Garassino MC, Cinquini M et al. Maintenance or consolidation therapy in
19. van Loon J, De Ruysscher D, Wanders R et al. Selective nodal irradiation on small-cell lung cancer: a systematic review and meta-analysis. Lung Cancer 2010;
basis of (18)FDG-PET scans in limited-disease small-cell lung cancer: 70: 119–128.
a prospective study. Int J Radiat Oncol Biol Phys 2010; 77: 35. Bozcuk H, Artac M, Ozdogan M et al. Does maintenance/consolidation
329–336. chemotherapy have a role in the management of small cell lung cancer (SCLC)?
20. Shirvani SM, Komaki R, Heymach JV et al. Positron emission tomography/ A meta-analysis of the published controlled trials. Cancer 2005; 104:
computed tomography-guided intensity-modulated radiotherapy for limited- 2650–2657.
stage small-cell lung cancer. Int J Radiat Oncol Biol Phys 2012; 82: 36. Slotman BJ, Faivre-Finn C, Kramer GW et al. Prophylactic cranial irradiation in
e91–e97. small-cell lung cancer. N Engl J Med 2007; 357: 664–672.
21. Le Péchoux C, Laplanche A, Faivre-Finn C et al. Clinical neurological outcome and 37. Slotman BJ, Mauer ME, Bottomley A et al. Prophylactic cranial irradiation in
quality of life among patients with limited small-cell cancer treated with two extensive disease small-cell lung cancer: short-term health-related quality of life
different doses of prophylactic cranial irradiation in the intergroup phase III trial and patient reported symptoms: results of an international phase III randomized
(PCI99-01, EORTC 22003-08004, RTOG 0212 and IFCT 99-01). Ann Oncol controlled trial by the EORTC Radiation Oncology and Lung Cancer Groups. J Clin
2011; 22: 1154–1163. Oncol 2009; 27: 78–84.
22. Wolfson AH, Bae K, Komak R et al. Primary analysis of a phase II randomized trial 38. Jeremic B, Shibamoto Y, Nikolic N et al. Role of radiation therapy in the combined-
Radiation Therapy Oncology Group (RTOG) 0212: impact of different total doses modality treatment of patients with extensive disease small-cell lung cancer: a
and schedules of prophylactic cranial irradiation on chronic neurotoxicity and randomized study. J Clin Oncol 1999; 17: 2092–2099.
quality of life for patients with limited-disease small-cell lung cancer. Int J Radiat 39. O’Brien ME, Ciuleanu TE, Tsekov H et al. Phase III trial comparing supportive care
Oncol Biol Phys 2011; 81: 77–84. alone with supportive care with oral topotecan in patients with relapsed small-cell
23. Foster NR, Qi Y, Shi Q et al. Tumor response and progression-free survival as lung cancer. J Clin Oncol 2006; 24: 5441–5447.
potential surrogate endpoints for overall survival in extensive stage small-cell lung 40. von Pawel J, Schiller JH, Shepherd FA et al. Topotecan versus cyclophosphamide,
cancer: findings on the basis of North Central Cancer Treatment Group trials. doxorubicin, and vincristine for the treatment of recurrent small-cell lung cancer. J
Cancer 2011; 117: 1262–1271. Clin Oncol 1999; 1 and 427: 658–667.
24. Rossi A, Di Maio M, Chiodini P et al. Carboplatin- or cisplatin-based chemotherapy 41. Eckardt JR, von Pawel J, Pujol JL et al. Phase III study of oral compared with
in first-line treatment of small-cell lung cancer: the COCIS meta-analysis of intravenous topotecan as second-line therapy in small-cell lung cancer. Clin Oncol
individual patient data. J Clin Oncol 2012; 30: 1692–1698. 2007; 25: 2086–2092.
25. Pujol JL, Carestia L, Daurès JP. Is there a case for cisplatin in the treatment of 42. Jotte R, Von Pawel J, Spigel D et al. Randomized phase III trial of amrubicin versus
small-cell lung cancer? A meta-analysis of randomized trials of a cisplatin- topotecan (Topo) as second-line treatment for small cell lung cancer (SCLC). J Clin
containing regimen versus a regimen without this alkylating agent. Br J Cancer Oncol 2012; 29 (Suppl.): abstr 7000.
2000; 83: 8–15. 43. Sugiyama T, Hirose T, Hosaka T et al. Effectiveness of intensive follow-up after
26. Mascaux C, Paesmans M, Berghmans T et al. A systematic review of the response in patients with small cell lung cancer. Lung Cancer 2008; 59:
role of etoposide and cisplatin in the chemotherapy of small cell lung cancer 255–261.
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt178 | vi

by guest
on 05 February 2018
Thymic epithelial tumours: ESMO Clinical Practice

N. Girard1, E. Ruffini2, A. Marx3, C. Faivre-Finn4 & S. Peters5, on behalf of the ESMO Guidelines
Committee*
1
Department of Respiratory Medicine, Expert Centre for Thymic Malignancies, Reference Centre for Orphan Pulmonary Diseases, Hôpital Louis Pradel, Hospices Civils de
Lyon, Lyon, France; 2Department of Thoracic Surgery, University of Torino, Turin, Italy; 3Institute of Pathology, University Medical Centre Mannheim, University of
Heidelberg, Mannheim, Germany; 4Institute of Cancer Sciences, The University of Manchester, Manchester Academic Health Science Centre, The Christie NHS Foundation
Trust, Manchester, UK; 5Department of Medical Oncology, Centre Hospitalier Universitaire Vaudois (CHUV), Lausanne, Switzerland
incidence and epidemiology overall, even if a slight female preponderance has been reported
for type A, AB and B1 subtypes in most studies, and a male pre-
Thymic epithelial tumours represent a heterogeneous group of dominance in carcinomas [2–7].
rare thoracic cancers, with reported annual incidence ranging No environmental or infectious factors have been demon-
from 1.3 to 3.2 per million [1]. Thymic epithelial tumours are strated to play a role in the pathogenesis of thymic epithelial
classified according to the World Health Organization (WHO) tumours. Reports on development of thymoma after radiation,
histopathological classification, which distinguishes thymomas solid-organ transplantation and immunosuppression, including
from thymic carcinomas. the context of human immunodeficiency virus infection, are
clinical practice
rare; differential diagnosis with thymic rebound hyperplasia

guidelines
thymomas
may be discussed in this setting (see below).
Thymomas are further subdivided into different types (called Genetic risk factors, such as multiple endocrine neoplasia
A, AB, B1, B2, B3 and rare others) based upon the morphology of 1 (MEN1), may influence the development of thymomas, as well
epithelial tumour cells, the relative proportion of the non-tumoural as thymic carcinoids, given their reported familial occurrence as
lymphocytic component (decreasing from type B1 to B3) and well as their association with cancer susceptibility syndromes [8].
resemblance to normal thymic architecture (Table 1) [2, 3]. The Moreover, extrathymic haematopoietic cancers (mostly diffuse
term ‘benign thymoma’ should be avoided. Thymomas are far large B-cell lymphoma and leukaemia) and a broad spectrum of
more frequent than thymic carcinomas, which have an incidence solid cancers (stomach, pancreas, colon and thyroid) have been
of 0.2 to 0.5 per million [3]. reported to occur more frequently in thymoma patients, particu-
larly subsequently [9]. This might be related to a shared unknown
thymic carcinomas
oncogenic trigger, a thymoma-associated immune deficiency or
Thymic carcinomas are similar to their extrathymic counterpart, (less likely) to adverse effects of treatments.
the most frequent subtype being squamous cell carcinoma.
Neuroendocrine tumours may occur in the thymus, and will not
be discussed in these guidelines; while localised primary thymic diagnosis
neuroendocrine tumours may benefit from surgical resection,
similar to other thymic carcinomas, the prognosis is poor given imaging and laboratory tests
frequent recurrences; for recurrent, advanced and metastatic Standard imaging for thymic tumours is i.v. contrast-enhanced
tumours, the management actually follows that of extra-thoracic computed tomography (CT) scan of the thorax, allowing a com-
neuroendocrine tumours. plete exploration of the mediastinum and the pleura from the
apex to the costodiaphragmatic recesses [IV, A]. CT is equal or
epidemiology superior to magnetic resonance imaging (MRI) for the diagnosis
Mean age at diagnosis is 50–60 years of age, but thymic tumours of mediastinal anterior masses, except in the setting of cystic
may actually be diagnosed in children as well as in elderly lesions [IV, B] [10].
patients. There is no consistent gender predilection in thymomas One-third of patients with thymoma present with autoimmune
disorders (Table 2), mainly myasthenia gravis which is particular-
ly common in type AB, B1 and B2 thymomas and almost always
*Correspondence to: ESMO Guidelines Committee, ESMO Head Office, Via L. Taddei 4, associated with anti-acetylcholine receptor antibodies (Table 1)
CH-6962 Viganello-Lugano, Switzerland. [11]. Other frequent disorders include pure red cell aplasia (5% of
cases) and hypogammaglobulinaemia (Good syndrome: 5% of
†
Approved by the ESMO Guidelines Committee: July 2015. cases) [12].

by guest
on 05 February 2018
Table 1. Histological subtypes of thymic epithelial tumours: relative frequency, frequency of myasthenia gravis and correlation with stage
Relative frequency Myasthenia gravis Masaoka stage
I II III IVA IVB
Type A 12% (3%–26%) 15% (0%–35%) 60% 31% 8% <1% <1%

Type AB 28% (15%–43%) 20% (5%–42%) 67% 26% 6% 1% 1%
Type B1 18% (6%–53%) 40% (5%–69%) 50% 37% 9% 3% 1%
Type B2 26% (8%–41%) 50% (23%–73%) 32% 29% 28% 8% 3%
Type B3 16% (3%–35%) 50% (25%–65%) 19% 36% 27% 15% 3%
Carcinoma 18% (1%–28%) <5% 10% 10% 45% 15% 20%
Data are based on references [5–8].
In addition to recording a complete history and conducting a Table 2. Autoimmune disorders associated with thymoma [11, 12]
full clinical examination (looking especially at neurological
signs), systematic immunological check-up is recommended Neuromuscular Myasthenia gravis
when a diagnosis of thymic epithelial tumour is suspected, in- Myotonic dystrophy
cluding complete blood cells count with reticulocytes and serum Limbic encephalitis
protein electrophoresis, as well as anti-acetylcholine receptor Peripheral neuropathy
and anti-nuclear antibodies tests [V, A]. Indeed, frequent Autonomic neuropathy
Acquired neuromyotonia
immune disorders associated with thymoma may impact the
Morvan syndrome (neuromyotonia and
course of all therapeutic interventions including surgery, radio-
encephalitis)
therapy as well as chemotherapy.
Stiff person syndrome
Cerebellar degeneration
diagnosis approach Polymyositis (carcinomas)
The diagnosis of any thymic epithelial tumour relies on making Haematological disorders Red cell aplasia
the differential diagnosis with other anterior mediastinal tumours Pernicious anaemia
and non-malignant thymic lesions [13]. CT is the imaging mo- Erythrocytosis
dality of choice. The need for pretreatment biopsy depends on Pancytopoenia
the resectability of the tumour [14–16]. Haemolytic anaemia
Thymic epithelial tumours are the most frequent cause of an- Leukaemia
terior mediastinal mass, accounting for 35% of cases; the most Multiple myeloma
relevant differential diagnoses include lymphomas (Hodgkin’s Collagen and autoimmune Systemic lupus erythematosus
or non-Hodgkin’s) in ∼25% of cases and germ-cell tumours disorders Rheumatoid arthritis
(teratoma or seminoma/non-seminomatous tumours) in ∼20% Sjogren’s syndrome
of cases [13]. Thymic carcinoma must be differentiated from Scleroderma
lung carcinoma, as well as from rarer entities, such as NUT car- Interstitial pneumonitis
cinomas [17].
Immune deficiency Hypogammaglobulinaemia (Good
Clinical judgement based on a complete history and physical,
disorders syndrome)
especially neurological, examination, correlated with laboratory T-cell deficiency syndrome
tests and radiological features, helps to develop a presumptive
diagnosis. Thymoma is the most likely diagnosis when facing a Endocrine disorders Multiple endocrine neoplasia
mediastinal mass associated with one of the above autoimmune Cushing’s syndrome
diseases, while thymic carcinoma patients typically have unspe- Thyroiditis
cific local symptoms [IV, A]. Lymphoma may be considered in Dermatological disorders Pemphigus
case of rapid onset of B-signs, coexistent lymphadenopathy or Lichen planus
elevated lactate dehydrogenase. Teratoma usually shows a het- Chronic mucosal candidiasis
erogeneous morphology on imaging, with fat and cystic pattern Alopecia areata
[18]. Seminomas and non-seminomatous germ-cell tumours may
Miscellaneous Giant cell myocarditis
be large and have a fulminant onset. Elevated serum β-human
Nephrotic syndrome
chorionic gonadotropin may be observed in seminomas, along
Ulcerative colitis
with elevated alphafetoprotein in non-seminomatous germ-cells
Hypertrophic osteoarthropathy
tumours.
Differentiating thymic malignancy from hyperplasia or
non-involuted thymus may be challenging. Thymic rebound corticosteroids. Thymic lymphoid hyperplasia is most common-
hyperplasia should be considered after stress, injuries, chemo- ly observed in myasthenia gravis, but also in the setting of
therapy, radiotherapy, anti-hormonal treatment or hyperthyroidism, connective tissue or vascular disease. CT
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv277 | v

by guest
on 05 February 2018
features include low-attenuation, symmetric and fatty pattern carcinoma component should always be mentioned first [V, C].
maintaining the bi-pyramidal shape of the thymus [18]. In In case of difficult diagnosis, it is recommended to consult a
equivocal cases at CT, chemical-shift MRI may detect microscop- second pathologist or refer the case to a thymic tumour path-
ic fatty infiltration by showing homogeneous signal decrease on ology panel.
opposed phase images relative to in-phase images, which is not
observed in thymoma [IV, B] [19]. Therapeutic intervention is
usually not required if the lesion is <30 mm, given a low risk of staging and risk assessment
progression or thymic malignancy [III, D] [20].
18-Fluorodeoxyglucose positron emission tomography (PET) staging
scan is generally not recommended to assess thymic masses [IV, C]. Thymic epithelial tumours are routinely staged according to the
Standard uptake values may be higher in type B3 thymomas Masaoka-Koga staging system (Table 3) [III, A] [25–27], which is
and thymic carcinomas; however, thymic hyperplasia may also correlated with overall survival (OS) [4, 28, 29]. Masaoka-Koga
present with hypermetabolism [21]. PET scan is optional in the staging is a surgical pathology system that is assessable only after
case of tumours with aggressive histology and an advanced stage surgical resection of the tumour. A typical feature of thymic epi-
to complete the staging work-up or further characterise lesions thelial tumours is the correlation between the WHO classification
suspicious for recurrences. and stage at diagnosis (Table 1), which may explain its reported
prognostic value [4–6] (Figures 1–3).
need for biopsy The International Association for the Study of Lung Cancer
(IASLC) Staging Prognostic Factors Committee, together with
Pretreatment biopsy is not required if the diagnosis of thymic the International Thymic Malignancy Interest Group (ITMIG),
tumour is highly probable and upfront surgical resection is recently proposed a Tumour–Node–Metastasis (TNM)-based
achievable (see below, definition of resectability) [IV, E]. Biopsy is staging system for thymic malignancies, based on OS analyses of
required in all other clinical situations [IV, A]: approaches may a retrospective international database of more than 10 000 cases
consist of percutaneous core-needle biopsy or incisional surgi- (Table 4) [30]. The TNM-based approach has the advantage of
cal biopsy through mediastinotomy or mini-thoracotomy, with being more appropriate both for thymoma and thymic carcin-
sensitivity rates ranging from 40% to 93% [22]. Biopsies that omas, which present with a higher propensity toward nodal and
are deep and multiple are preferred. Pleural spaces should be distant metastatic invasion. The IASLC/ITMIG TNM system of
respected to avoid tumour cell seeding. Fine-needle aspiration thymic tumours will be incorporated as the official thymic tumour
is generally not recommended [IV, D]. staging system into the 8th edition of the TNM staging system of
thoracic malignancies expected in 2016–2017. From our stand-
thymomas. Although designed for surgical resection specimens, point, the Masaoka-Koga staging should remain the standard for
the WHO classification may be used for small biopsies [V, A]. the routine management of patients, pending the approval of the
However, thymoma subtyping on small biopsies is usually not American Joint Committee on Cancer (AJCC) and Union for
needed for the therapeutically relevant distinction between International Cancer Control (UICC) [III, A]. Moreover, given the
lymphoma and solid tumour. In any case, diagnostic discrepancies major switch that the TNM system represents and the limited
between core-needle and resection specimen histology can be amount of fair level of evidence data to support our current treat-
anticipated, given the frequent occurrence of histological ment strategies (especially postoperative radiotherapy), the value of
tumour heterogeneity that may be missed due to sampling error the TNM system to drive the therapeutic strategy has to be assessed.
[23]. The recent proposal of major and minor morphological Correlative clinical data based on this system may be encouraged in
and immunohistochemical criteria to better individualise each a research setting.
thymic epithelial tumour entity aims at addressing those issues, The assessment of resectability is mostly based on the surgeon’s
and has been integrated in the revised WHO classification expertise; it is recommended to discuss indications for surgery in a
[3, 24]. Immunohistochemical markers may be helpful, multidisciplinary tumour board setting [V, B]. There is no recog-
including cytokeratins and p63 expression for normal and nised clinical staging system, and the treatment strategy for thymic
neoplastic epithelial cells, and terminal deoxynucleotidyl epithelial tumours is primarily based on whether the tumour may
transferase expression in immature T cells (usually observed be resected upfront or not [IV, A], as complete resection has been
in types AB, B1, B2 and B3 thymomas, and absent in identified as the most consistent and significant prognostic factor of
carcinomas and type A thymomas) [3]. disease-free survival and OS [5, 6, 29]. Correlation between clinical
and surgical pathology stage is higher in advanced stages, given the
thymic carcinomas. Immunohistochemistry with anti-CD117/ identification of vessel invasion, enlarged lymph nodes, pleural/
KIT and anti-CD5 antibodies helps to establish the thymic pericardial lesions or even systemic metastases [28]. Preoperative
origin in ∼80% of mediastinal carcinomas [V, A]. Since these CT findings reported to be associated with tumour invasiveness
markers are not absolutely specific, correlation with the clinical and/or completeness of resection include: tumour size (>5/7/8 cm,
setting is always recommended, and is mandatory in the subset depending on studies), lobulated or irregular contours, calcifica-
of 20% of thymic carcinomas without expression of CD117/KIT tions, infiltration of surrounding fat, lung infiltration, great vessel
and CD5 [3]. invasion or encirclement [31–33]. The new TNM staging may even
In thymic tumours showing more than one histological provide more help in formalising resectability: T1–3 level of inva-
pattern, each component should be listed (starting with the pre- sion refers to structures amenable to surgical resection, while T4
dominant one) and be quantified in 10% increments; a thymic level of invasion includes unresectable structures (Table 4).
v | Girard et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Annals of Oncology
Table 3. Staging of thymic epithelial tumours: Masaoka-Koga-based staging system [25, 26], International Thymic Malignancy Interest Group refinements [27] and overall survival and recurrence-free
survival (range)a [28]
Masaoka-Koga, 1994 International Thymic Malignancy Interest Group, 2011 10-year overall 10-year cumulative incidence
survival of recurrence
Thymoma Thymic carcinoma
Stage I Grossly and microscopically completely encapsulated tumour – Invasion into but not through the capsule 84% (81%–86%)
– In the absence of capsule, absence of invasion into
surrounding tissues
Stage IIA Microscopic transcapsular invasion – Microscopic transcapsular invasion (<3 mm) 83% (79%–87%) 8% (7%–8%) 25% (22%–29%)
Stage IIB Macroscopic invasion into thymic or surrounding fatty tissue, – Gross extension into normal thymus or perithymic fat
or grossly adherent to but not breaking through the surrounding the tumour (microscopically confirmed)
mediastinal pleura or pericardium – Adherence to pleura or pericardium, with microscopic
confirmation of perithymic invasion
Stage III Macroscopic invasion into neighbouring organ (i.e. – Microscopic invasion of the mediastinal pleura (either 70% (64%–75%) 29% (27%–31%) 59% (44%–76%)
pericardium, great vessel or lung) partial or penetrating the elastin layer)
– Microscopic invasion of the pericardium (either partial in

the fibrous layer or penetrating through to the serosal layer)
– Microscopically confirmed direct penetration into the outer
elastin layer of the visceral pleura or into the lung
parenchyma
– Invasion into the phrenic or vagus nerves (microscopically
confirmed)
– Invasion into or penetration through major vascular
structures (microscopically confirmed)
– Adherence (i.e. fibrous attachment) of lung or adjacent
organs only if there is mediastinal pleural or pericardial
invasion (microscopically confirmed)
doi:10.1093/annonc/mdv277 | v
Stage IVA Pleural or pericardial metastasis – Microscopically confirmed separate nodules in the visceral 42% (26%–58%) 71% (34%–100%) 76% (58%–100%)
or parietal pleural, pericardial or epicardial surfaces
Stage IVB Lymphogenous or haematogenous metastasis – Lymphogenous or haematogenous metastasis 53% (32%–73%) 57% (24%–90%) 54% (37%–67%)
a
Information reprinted from [27] with permission of John Wiley & Sons, Inc.

by guest
on 05 February 2018
Figure 1. Treatment algorithm for resectable thymic tumour (Masaoka-Koga stage I–III, TNM stage I–IIIA).
risk assessment management of resectable disease

Prognostic assessment is challenging, as the impact of tumour The treatment strategy is based on the resectability of the tumour.
stage and histology on OS is superseded by the achievement of a If complete resection is deemed to be achievable upfront, as it is
complete resection in reported series [4–7, 28–30]. Moreover, the case in Masaoka-Koga stage I/II and some stage III tumours
depending on stage, up to 50%–60% of patients actually do not (classified as stage I, II, IIIA/T3 in the IASLC/ITMIG TNM pro-
die of progression of the thymic tumour [34]. Autoimmune posed system), surgery represents the first step of the treatment
disorders have been reported as the cause of death in 25% of [IV, A], possibly followed by postoperative radiotherapy and, less
thymomas, especially those with early-stage tumours. The evo- frequently, chemotherapy (Table 5).
lution of these alterations, which are related to the abnormal
intrathymomatous selection of constitutively autoreactive lym-
phocytes, does not parallel that of the tumour. This is contrary surgical principles
to what is observed in paraneoplastic syndromes that are The standard approach is median sternotomy [IV, A], which
caused by tumour cell-derived cytokines or hormones or by allows the wide opening of the mediastinum and both pleural
cross-reactive antibodies. The management of autoimmune cavities, followed by evaluation of macroscopic capsular inva-
syndromes will not be discussed in these guidelines, but should sion, infiltration of perithymic and mediastinal fat, peritumoural
be integrated in the oncological management of these patients and pleural adherences and involvement of surrounding struc-
[V, A]. tures [14, 15, 35, 36]. Generally, complete thymectomy
v | Girard et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Figure 2. Treatment algorithm for unresectable thymic tumour (Masaoka-Koga stage III–IVA, TNM stage IIIA–IIIB–IVA).
including the tumour, the residual thymus gland and perithymic areas are also designated on the resection specimen, as dis-
fat is preferred because local recurrences have been observed cussed below. Phrenic nerve preservation does not affect OS
after partial thymectomy when part of the thymus gland is left but increases the risk of local recurrence [IV, C], and should be
behind [IV, B]. Thymomectomy—leaving residual thymic tissue balanced with the achievement of a complete resection, espe-
and perithymic fat behind—alone is an option in stage I cially in patients with severe myasthenia gravis [38, 39]. Frozen
tumours in non-myasthenic patients [IV, C] [14, 37]. If the sections to assess tumour involvement of resection margins are
tumour is widely invasive (stage III/IV), en bloc removal of all not recommended [V, D], given the high risk of false-negative
affected structures, including lung parenchyma (usually through results [36].
limited resection), pericardium, great vessels, nerves and pleural Minimally invasive surgery is an option for presumed stage I
implants, should be carried out [IV, A]. Resection of venous and possibly stage II tumours in the hands of appropriately
vascular structures (innominate vein(s) and superior vena trained thoracic surgeons [IV, C] [14, 35, 40]. This includes
cava) include partial resection with suturing or complete resec- transcervical, extended transcervical, video-assisted thoraco-
tion and vessel reconstruction using vascular prosthesis. Areas scopy (VATS) and robotic approaches (right or left, right and
of uncertain resection margins are marked with clips to allow left, right and cervical, left and cervical, subxiphoid and right
precise delivery of postoperative radiotherapy [IV, B]; those and left, cervical and subxiphoid); robotic surgery may allow a
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv277 | v

by guest
on 05 February 2018
better visualisation of the tumour when compared with VATS. jeopardise or change the principles that are deemed appropri-
The choice for minimally invasive resection should not ate for an open approach, especially the achievement of com-
plete resection that may ultimately require switching to an
open procedure [V, A]. Minimally invasive surgery is not
recommended for stage III tumours, given the absence of long-
term follow-up data [IV, D].
Lymphadenectomy has historically rarely been carried out
after resection of thymic tumours. The new IASLC/ITMIG
TNM staging system of thymic tumours, however, leads to the
recommendation that locoregional lymphoadenectomy should
be carried out during resection of all types of thymic tumours. A
proposed nodal map is available from ITMIG [41]. The pro-
posed N descriptor in the staging system includes:
• anterior region (N1), which involves the anterior mediastinal
nodes ( prevascular, para-aortic, ascending aorta, superior and
inferior phrenic and supradiaphragmatic) and the anterior
cervical nodes (low anterior cervical); and
• the deep region (N2), which includes the middle mediastinal
(internal mammary, upper and lower paratracheal, subaortic,
subcarinal and hilar) and the deep cervical (lower jugular and
supraclavicular).
Routine removal of anterior mediastinal nodes and anterior cer-

vical nodes is recommended [IV, A]. Systematic sampling of
other intrathoracic sites is encouraged (i.e. paratracheal, aorto-
pulmonary window and subcarinal areas, depending on tumour
location) in stage III/IV tumours [V, B] [36]. Systematic lym-
Figure 3. Treatment algorithm for metastatic thymic tumour (Masaoka-
Koga stage IVB, TNM stage IVB).
phadenectomy (N1 + N2) is strongly recommended in case of
Table 4. Proposed Tumour–Node–Metastasis staging (International Association for the Study of Lung Cancer Prognostic Factors
Committee- International Thymic Malignancy Interest Group) [30] and corresponding Masaoka-Koga stage
Stage Descriptors
Tumour
T1 T1a Encapsulated or unencapsulated, with or without extension into the mediastinal fat
T1b Extension into the mediastinal pleura
T2 Direct invasion of the pericardium (partial or full-thickness)
T3 Direct invasion of the lung, the brachiocephalic vein, the superior vena cava, the chest wall, the phrenic nerve
and/or hilar (extrapericardial) pulmonary vessels
T4 Direct invasion of the aorta, arch vessels, the main pulmonary artery, the myocardium, the trachea or the
oesophagus
Node
N0 N0 No nodal involvement
N1 N1 Anterior (perithymic) nodes (IASLC levels 1, 3a, 6 and/or supradiaphragmatic/inferior phrenics/
pericardial)
N2 N2 Deep intrathoracic or cervical nodes (IASLC levels 2, 4, 5, 7, 10 and/or internal mammary nodes)
Metastasis
M0 No metastatic pleural, pericardial or distant sites
M1 M1a Separate pleural or pericardial nodule(s)
M1b Pulmonary intraparenchymal nodule or distant organ metastasis
Stage grouping Corresponding Masaoka-Koga stage
I T1N0M0 I, IIA, IIB, III

II T2N0M0 III
IIIA T3N0M0 III
IIIB T4N0M0 III
IVA T any N0,1 M0,1a IVA, IVB
IVB T any N0-2 M0-1b IVB
v | Girard et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Table 5. Stage-matched therapeutic strategy
Masaoka-Koga Thymoma Thymic carcinoma
stage
Stage I Upfront surgery [IV, A] Upfront surgery [IV, A]

No biopsy [IV, E] No biopsy [IV, E]
If complete resection (R0): no postoperative radiotherapy [II, E] If resection complete (R0): consider postoperative radiotherapy
If incomplete resection (R1): postoperative radiotherapy (45–50 Gy) [V, C]
(50–54 Gy) [IV, B] If incomplete resection (R1): postoperative radiotherapy
(50–54 Gy) [IV, B]
Stage IIA Upfront surgery [IV, A] Upfront surgery [IV, A]

If complete resection (R0): If complete resection (R0):
– Type A–B2: no postoperative radiotherapy [IV, C] – Consider postoperative radiotherapy (45–50 Gy) [IV, B]
– Type B3: consider postoperative radiotherapy If incomplete resection (R1):
(45–50 Gy) [IV, C] – Postoperative radiotherapy (50–54 Gy) [IV, B]
If incomplete resection (R1): – Consider postoperative chemotherapy
– Postoperative radiotherapy (50–54 Gy) [IV, B]
Stage IIB Upfront surgery [IV, A] Upfront surgery [IV, A]

If complete resection (R0): If complete resection (R0):
– Type A–B1: no postoperative radiotherapy [IV, C] – Consider postoperative radiotherapy (45–50 Gy) [IV, B]
– Type B2–B3: consider postoperative radiotherapy If incomplete resection (R1):
(45–50 Gy) [IV, C] – Postoperative radiotherapy (50–54 Gy) [IV, B]
If incomplete resection (R1): – Consider postoperative chemotherapy
– Postoperative radiotherapy (50–54 Gy) [IV, B]
Stage III–IVA Resectable tumour (TNM I–IIIA, i.e. T1–3): Resectable tumour (TNM I–IIIA, i.e. T1–3):
– Upfront surgery [IV, A] – Upfront surgery [IV, A]
– Postoperative radiotherapy (45–50 Gy), with boost on areas of – Postoperative radiotherapy (40–50 Gy), with boost on areas of
concern [IV, B] concern [IV, B]
Unresectable tumour (TNM IIIA-B, i.e. T3-T4, IVA): – Consider postoperative chemotherapy
– Biopsy Unresectable tumour (TNM IIIA-B, i.e. T3-T4, IVA):
– Primary chemotherapy (prefer anthracycline-based) [III, A] – Biopsy
– If the tumour becomes resectable: – Primary chemotherapy (prefer anthracycline-based) [III, A]
– Surgery [III, A] – If the tumour becomes resectable:
– Postoperative radiotherapy (45–50 Gy), with boost on areas – Surgery [III, A]
of concern (R0, R1 resection) [IV, B] – Postoperative radiotherapy (45–50 Gy), with boost on areas
– If the tumour remains unresectable or R2: of concern (R0, R1 resection) [IV, B]
– Definitive radiotherapy (60 Gy) [IV, B] – Consider postoperative chemotherapy (R0, R1 resection)
– Option: chemoradiotherapy – If the tumour remains unresectable or R2:
– Option: concurrent chemoradiotherapy (platin and etoposide, – Definitive radiotherapy (60 Gy) [IV, B]
60 Gy) [III, B] – Option: chemoradiotherapy
– Option: concurrent chemoradiotherapy (platin and etoposide,
60 Gy)
Stage IVB Definitive chemotherapy [III, A] Definitive chemotherapy [III, A]

– If the tumour becomes resectable, consider:
– Surgery and postoperative radiotherapy
– Definitive radiotherapy
thymic carcinoma due to the high rate of lymphatic spread (20% involved structures, organs or areas of likely residual microscopic
versus 3% in thymomas) [V, B]. or macroscopic disease are the primary responsibility of the oper-
ating surgeon, and may be done using a mediastinal board [V, B].
The final pathological examination leads to a final histological
surgical pathology principles diagnosis and staging of the tumour, based on the WHO classi-
Communication between surgeons and pathologists is required to fication and the Masaoka-Koga system, respectively (Tables 1
accurately stage thymic epithelial tumours [V, A] [36]. The and 3). Staging according to the proposed IASLC/ITMIG TNM
proper orientation of the specimen and the designation of system is optional [V, C]. Given the potential heterogeneity of
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv277 | v

by guest
on 05 February 2018
thymic epithelial tumours, a sufficient number of representative to define the target volume) [IV, B]; however, the optimal
sections should be examined regardless of the tumour diameter postoperative dose/fractionation is still to be defined; and
(at least five sections up to a diameter of 5 cm, with one add- • conventional fractionation scheme consisting of daily doses of
itional block per additional centimetre of maximal diameter); if 1.8–2 Gy over a 4- to 6-week period [IV, A] [48]. The field
the margin is <1 mm, at least three additional sections through may encompass involved nodes [IV, B] and the site of a
this area should be obtained [V, B]. Completeness of resection resected pleural implant [V, C].
should be assessed, making the distinction between tissues that
have been cut or dissected, and a surface bounded by a space Prophylactic irradiation of supraclavicular nodes is not recom-
(such the mediastinal pleura, pericardium or endothelium of the mended [V, E]. Low-dose entire hemithoracic radiotherapy is
innominate veins), which should not be designated as a positive also not recommended [IV, C] [49]. Ideally, postoperative radio-
margin [V, A]. therapy should start within 3 months of the surgical procedure
[V, B].
postoperative radiotherapy thymomas. Postoperative radiotherapy is not indicated after

Current practices for postoperative mediastinal radiotherapy are complete resection of Masaoka-Koga stage I thymoma [II, E]. One
highly variable and there is paucity of prospective, multicentre randomised trial including 29 patients compared postoperative
evidence. The global trend over the past years has been towards radiotherapy versus surgery alone in this setting, and failed to show
a less frequent use of postoperative radiotherapy in thymoma, any differences in patient outcome [50].
and to keep it in reserve for high-risk cases (Table 5) [14–16]. Postoperative radiotherapy is not recommended after com-
This is based on recent reports from large databases [6, 7, 42–45], plete resection of stage II thymoma [IV, C]. In the ITMIG data-
as well as pooled analyses of retrospective studies [46], indicating: base, cumulative incidence of mediastinal or extramediastinal
recurrence was only 8% at 10 years (Table 2) [28]. Postoperative
• the absence of survival benefit after radiotherapy in stage I radiotherapy may be considered in case of aggressive histology
thymoma, or after R0/1 resection of stage II–III thymoma (type B2, B3) or extensive transcapsular invasion (stage IIB) [IV,
[7, 42, 43]; C] [51–53].
• a similar rate of recurrence in patients who received post- Postoperative radiotherapy is recommended after complete
operative radiotherapy or not, after complete resection of resection of stage III/IVA thymoma, in an effort to prolong RFS
thymoma [46]; and and OS [IV, B] [54].
• a recurrence-free survival (RFS) and OS benefit with postoperative
radiotherapy after resection of thymic carcinoma [7, 44, 45].
thymic carcinoma. After complete resection of thymic carcinoma,
Stage and completeness of resection are thus the most relevant postoperative radiotherapy is optional for stage I tumours [V, C],
criteria in the decision making, followed by histology [IV, B]. should be considered for stage II tumours [IV, B] and is recom-
Those factors are the most significant predictors of RFS [4–7, mended for stage III/IVA tumours [IV, B] [7, 44, 45]. Postoperati-
28–29]; however, one must take into account that retrospective ve radiotherapy is recommended in case of microscopically
analyses are likely to be biased, since postoperative radiotherapy (R1) or macroscopically incomplete (R2) resection [IV, B], to a
is most likely administered in patients with incomplete resection total dose of 50–54 and 60 Gy, respectively, with a 10-Gy boost
or high-grade tumours. Therefore, the absence of survival differ- directed to areas of likely residual disease.
ences may then suggest that postoperative radiotherapy reduced
or overcame the risk of recurrence in those patients. Another postoperative chemotherapy
point to consider is that recurrences of thymic epithelial tumours thymomas. Postoperative chemotherapy is not recommended
occur outside the mediastinum in more than 60% of cases [47]. after R0–R1 resection of a thymoma [III, E] [6, 14–16, 55].
The development of the TNM system leads to redefinition of
subsets of patients, based especially on the T descriptors, that
thymic carcinomas. Since thymic carcinomas do present with
may help to clarify which patients benefit from postoperative
frequent and early locoregional and systemic recurrences after
radiotherapy. Meanwhile, the grouping of Masaoka-Koga stage I,
incomplete surgery [3, 7, 28, 44, 45], postoperative chemotherapy
IIA, IIB and some stage III tumours in one single T1 category
may be considered as an option in stage II/III/IV thymic
must be assessed in dedicated studies.
carcinomas, especially if not delivered as induction treatment
Current evidence for postoperative radiotherapy in thymic
(Table 5) [14–16].
epithelial tumours support the use of:
• 3D conformal radiotherapy or intensity-modulated radiation
therapy targeted to the tumour bed, keeping at-risk thoracic
management of advanced disease
organs within accepted safe constraints [IV, A];—clinical If complete resection is deemed not to be achievable upfront on
target volume includes the whole thymic space, the tumour the basis of imaging studies, as it is frequently the case in
and its extensions and the anterior, superior and middle Masaoka-Koga stage III/IVA tumours (classified as stage IIIA/
mediastinum [IV, A]. T3, IIIB/T4, /IVA in the IASLC/ITMIG TNM proposed system),
• a total dose of 45–50 Gy after complete resection, 50–54 Gy a biopsy should be carried out, followed by primary/induction
after R1 resection, with a boost to areas of likely residual chemotherapy as part of a curative-intent sequential strategy
disease (as mentioned above, surgical clips may then be useful that integrates subsequent surgery or radiotherapy [14–16].
v | Girard et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Table 6. Selected chemotherapy regimens for advanced thymic performance status or co-existing medical conditions—definitive
epithelial tumours assessed in phase II trials (adapted from [60] with radiotherapy is recommended as part of a sequential chemora-
permission of Informa Healthcare) diotherapy strategy [III, A] [56]. Combination with chemother-
Regimen Agents Doses apy (including cisplatin, etoposide chemotherapy and a total dose
of radiation of 60–66 Gy in 30–33 fractions) may be considered
ADOC Doxorubicin 40 mg/m2/3 weeks [V, C].
Cisplatin 50 mg/m2/3 weeks
Vincristine 0.6 mg/m2/3 weeks
Cyclophosphamide 700 mg/m2/3 weeks definitive chemotherapy
2 Chemotherapy should be offered as single modality treatment in
CAP Cisplatin 50 mg/m /3 weeks
Doxorubicin 50 mg/m2/3 weeks
advanced, non-resectable, non-irradiable or metastatic (stage
Cyclophosphamide 500 mg/m2/3 weeks IVB) thymic epithelial tumours. The aim is to relieve tumour-
related symptoms by eliciting tumour shrinkage, while pro-
PE Cisplatin 60 mg/m2/3 weeks longed survival is uncertain [III, A]. Cisplatin-based combin-
Etoposide 120 mg/m2 × 3 days/3 weeks ation regimens should be administered [III, A] (Table 6) [66–
VIP Etoposide 75 mg/m2 × 4 days/3 weeks 71]. No randomised studies have been conducted and which
Ifosfamide 1.2 g/m2 × 4 days/3 weeks regimen should be considered standard remains unknown.
Cisplatin 20 mg/m2 × 4 days/3 weeks Multiagent combination regimens and anthracycline-based regi-
mens appear to have improved response rates compared with
CODE Cisplatin 25 mg/m2/1 week
etoposide-based regimens [59, 60, 72]. Combination of cispla-
Vincristin 1 mg/m2/2 weeks
tin, doxorubicin and cyclophosphamide is preferred [III, B].
Doxorubicin 40 mg/m2/2 weeks
Combination of carboplatin and paclitaxel is an option for
Etoposide 80 mg/m2 × 3 days/2 weeks
thymic carcinoma [69, 71] [III, B]. Surgery or radiotherapy is
Carbo-Px Carboplatin AUC 5–6/3 weeks possible in rare and selected metastatic cases, without proven
Paclitaxel 200–225 mg/m2/3 weeks outcome benefit [IV, C].
Response Evaluation Criteria in Solid Tumours (RECIST)
CAP-GEM Capecitabine 650 mg/m2 b.i.d. 14 days/3 weeks
Gemcitabine 1000 mg/m2 × 2 days/3 weeks
v1.1 criteria should be used to assess response to chemotherapy
[V, A]; adapted criteria for pleural lesions include the use of the
short axis as the measurement plane and the unidimensional meas-
Patients not eligible for local treatment should receive palliative urement of two pleural tumour sites at three different levels [73].
chemotherapy only.
primary chemotherapy recurrences

Primary/induction chemotherapy is standard in non-resectable Recurrences of thymic epithelial tumours are not uncommon
advanced thymic epithelial tumours [III, A] [56–58]. Cisplatin- (∼10%–15% of all-stage resected tumours) and should be managed
based combination regimens should be administered; combinations according to the same strategy as newly diagnosed tumours [IV,
of cisplatin, doxorubicin and cyclophosphamide, and cisplatin A]. The average time-to-recurrence in completely resected thymic
and etoposide, are the recommended options (Table 6) [III, A] [59, tumours is 5 years (with a range of 3–7 years). Complete resection
60]. Primary chemoradiotherapy with platin and etoposide is an of recurrent lesions represents a major predictor of favourable
option, especially for thymic carcinomas [III, B] [33, 61]. outcome [74–76], and surgery is then recommended in the case
Usually, two to four cycles are administered before imaging is of resectable lesion. Of note, histological switch from lymphocytic
carried out to reassess resectability of the tumour [III, A]. Surgery lesions to more epithelial tumours has been reported, and may be
should be offered to patients for whom complete resection is related to tumour heterogeneity, as well as the effect of previous
deemed achievable, according to principles discussed above [III, A]; corticosteroid and chemotherapy treatment [74].
extended resection may be required [62]. Hyperthermic intra- In non-resectable recurrences, several consecutive lines of
pleural chemotherapy, as well as extra-pleural pneumonectomy, chemotherapy may be administered when the patient presents
may be discussed in case of stage IVA tumour [IV, C] [63, 64]. with tumour progression (Table 6). The re-administration of a
Postoperative radiotherapy delivery should follow. previously effective regimen should be considered [IV, B], espe-
Subtotal resection, so-called debulking resection, is an option cially in case of previous response, late occurring recurrence
in selected cases of thymoma, aiming at facilitating subsequent and, for anthracyclines, a patient in a good medical condition
definitive radiotherapy [IV, C] [65]. Debulking is not recom- who has not received cumulative doses precluding the safe deliv-
mended in thymic carcinoma [V, D]. Postoperative chemora- ery of at least three additional cycles [77]. Of note, the risk of
diotherapy (including cisplatin, etoposide chemotherapy and a cardiac toxicity is further increased in patients having received
total dose of radiation of 60 Gy in 30 fractions) may be consid- previous mediastinal radiotherapy. Participation to clinical trials
ered after debulking/R2 resection [IV, B]. is recommended.
Preferred regimens for second-line treatment include carbo-
definitive radiotherapy platin plus paclitaxel [69], and platin plus etoposide [67] [III,
If the patient is not deemed a surgical candidate—either because B]; capecitabine plus gemcitabine is an option (Table 6) [III, B].
R0 resection is not thought to be achievable, or because of poor These regimens were evaluated in dedicated phase II trials.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv277 | v

by guest
on 05 February 2018
Diagnosis
– Thymic epithelial tumours are classified according to the WHO histopathological classification.
– Although designed for surgical resection specimen, the WHO classification may be used for small biopsies [V, A].
– Immunohistochemistry with anti-CD117/KIT and anti-CD5 antibodies is useful to establish the thymic primary nature of a mediastinal carcinoma [V, A].
– Each component of heterogeneous tumours may be quantified by 10% increments [V, C].
– Consultation with a second pathologist or referral of the case to a thymic tumour pathology panel is recommended whenever there is any diagnostic difficulty.
– Oncogenetic assessment should be carried out in case of familial thymic epithelial tumour, looking especially at MEN1.
Imaging and diagnostic tests
– Thymoma is the first diagnosis to consider when facing a mediastinal mass associated with autoimmune disease [IV, A].
– The diagnosis of any thymic epithelial tumour relies on making the differential diagnosis with other anterior mediastinal tumours and non-malignant
thymic lesions.
– Standard imaging for thymic tumours is i.v. contrast-enhanced (CT) scan of the thorax [IV, A].
– MRI is recommended to differentiate thymic tumour from hyperplasia whenever CT scan is doubtful, or in case of cystic lesion [IV, B].
– PET scan is generally not recommended to assess thymic masses [IV, C].
– Therapeutic intervention is usually not required if the lesion is <30 mm, given a low risk of progression or thymic malignancy [III, D].
– Systematic immunological check-up is recommended, including complete blood cells count with reticulocytes and serum protein electrophoresis, as well as
anti-acetylcholine receptor and anti-nuclear antibodies tests [V, A].
Need for a biopsy
– Pretreatment biopsy is not required if the diagnosis of thymic epithelial tumour is highly suspected and upfront surgical resection is achievable [IV, E].
– Biopsy is required in all other clinical situations [IV, A]; approaches may consist of percutaneous core-needle biopsy or incisional surgical biopsy through
mediastinotomy or mini-thoracotomy. Fine-needle aspiration is not recommended [IV, D].
Staging
– Thymic epithelial tumours are routinely staged according to the Masaoka-Koga staging system [III, A]. Masaoka-Koga staging is a surgical pathology
system that is assessable only after surgical resection of the tumour.
– Staging according to proposed IASLC/ITMIG TNM system is optional [V, C].
– The Masaoka-Koga staging system should remain the standard for the routine management of patients, pending the approval of the AJCC and UICC [III, A].
Risk assessment
– The management of autoimmune syndromes should be integrated in the oncological management of these patients [V, A].
Management of resectable disease
– The treatment strategy for thymic epithelial tumour is primarily based on whether the tumour may be resected upfront or not [IV, A].
– The assessment of resectability is mostly based on the surgeon’s expertise; it is recommended to discuss indications for surgery in a multidisciplinary
tumour board setting [V, B].
– If complete resection is deemed to be achievable upfront, surgery represents the first step of the treatment [IV, A].
Surgery principles
– Standard approach is median sternotomy [IV, A].

– Complete thymectomy including the tumour, the residual thymus gland and perithymic fat, is preferred [IV, B].
– Thymomectomy alone—leaving residual thymic tissue and perithymic fat behind—is an option in stage I tumours in non-myasthenic patients [IV, C].
– If the tumour is widely extensive invasive (stage III/IV), en bloc removal of all affected structures, including lung parenchyma (usually through limited
resection), pericardium, venous great vessels, nerves and pleural implants, should be carried out [IV, A].
– Areas of uncertain margins are marked with clips to allow precise delivery of postoperative radiotherapy [IV, B]: those areas are also designated on the
resection specimen.
– Phrenic nerve preservation does not affect OS but increases the risk of local recurrence [IV, C].
– Frozen sections to assess tumour involvement of resection margins are not recommended [V, D].
– Minimally invasive surgery is an option for presumed stage I–II tumours in the hands of appropriately trained thoracic surgeons [IV, C].
– The choice for minimally invasive resection should not jeopardise or change the principles that are deemed appropriate for an open approach, especially
the achievement of complete resection that may ultimately require switching to an open procedure [V, A].
– Minimally invasive surgery is not recommended for stage III tumours, given the absence of long-term follow-up data [IV, D].
– Routine removal of anterior mediastinal and anterior cervical nodes is recommended [IV, A].
– Systematic sampling of intrathoracic sites is encouraged in stage III/IV tumours [V, B].
– Systematic lymphadenectomy (N1 + N2) is strongly recommended in case of thymic carcinoma due to the high rate of lymphatic spread [V, B].
Continued
v | Girard et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Table 7. Continued
Surgical pathology principles
– Communication between surgeons and pathologists is required to accurately stage thymic epithelial tumours [V, A].
– The proper orientation of the specimen and the designation of involved structures, organs or areas of likely residual microscopic or macroscopic disease are
the primary responsibility of the operating surgeon and may be done using a mediastinal board [V, B].
– A sufficient number of representative sections should be examined regardless of the tumour diameter; if the margin is <1 mm, at least three additional
sections through this area should be obtained [V, B].
– Completeness of resection should be assessed, making the distinction between tissues that have been cut or dissected, and a surface bounded by a space,
which should not be designated as a positive margin [V, A].
– Postoperative radiotherapy should start within 3 months of the surgical procedure [V, B].
– The use of 3D conformal radiotherapy or intensity-modulated radiation therapy targeted to the tumour bed is recommended [IV, A].
– Clinical target volume includes the whole thymic space, the tumour and its extensions and the anterior, superior and middle mediastinum [IV, A].
– Standard dose constraints for thoracic radiotherapy should be used.
– Total dose of 45–50 Gy after complete resection, 50–54 Gy after R1 resection, with a boost to areas of likely residual disease are recommended (surgical
clips may then be useful to define the target volume) [IV, B].
– The use of a conventional fractionation scheme is recommended in daily doses from 1.8 to 2 Gy over a 4- to 6-week period [IV, A].
– The field may encompass involved nodes [IV, B], and the site of a resected pleural implant [V, C].
– Prophylactic irradiation of supraclavicular nodes is not recommended [V, E].
– Low-dose entire hemithoracic radiotherapy is not recommended [IV, C].
– After complete resection of thymoma:
• Postoperative radiotherapy is not indicated after complete resection of Masaoka-Koga stage I thymoma [II, E].
• Postoperative radiotherapy is not recommended after complete resection of stage II thymoma [IV, C], but may be considered in case of aggressive
histology (type B2, B3) or transcapsular invasion (stage IIB) [IV, C].
• Postoperative radiotherapy is recommended after complete resection of stage III/IVA thymoma [IV, B].
– After complete resection of thymic carcinoma:
• postoperative radiotherapy is optional for stage I tumours [V, C],
• it should be considered for stage II tumours [IV, B] and
• it is recommended for stage III/IVA tumours [IV, B].
– Postoperative radiotherapy is recommended in case of microscopically (R1) or macroscopically incomplete (R2) resection [IV, B], to a total dose of 50–54
and 60 Gy, respectively, with a 10-Gy boost to areas of likely residual disease.
Postoperative chemotherapy
– Postoperative chemotherapy is not recommended after R0–R1 resection of a thymoma [III, E].
– Postoperative chemotherapy may be considered as an option in stage II/III/IV thymic carcinomas, especially if not delivered as induction treatment.
Management of advanced disease
– If complete resection is deemed not to be achievable upfront, primary/induction chemotherapy is administered, part of curative-intent sequential strategy
integrating subsequent surgery or radiotherapy. Cases not eligible for local treatment receive palliative chemotherapy only.
Primary/induction chemotherapy
– Primary/induction chemotherapy is the standard in non-resectable locally advanced thymic epithelial tumours [III, A].
– Cisplatin-based combination regimens should be administered; combinations of cisplatin, doxorubicin and cyclophosphamide, and cisplatin and etoposide
are recommended options [III, A].
– Primary chemoradiotherapy with platin and etoposide chemotherapy is an option for thymic carcinomas [III, B].
– Usually, two to four cycles are administered before imaging is carried out to reassess resectability of the tumour [III, A].
– Surgery should be offered to patients for whom complete resection is deemed achievable after primary chemotherapy, according to principles discussed in
the text [III, A].
– Hyperthermic intrapleural chemotherapy, as well as extra-pleural pneumonectomy, may be discussed in case of stage IVA tumour [IV, C].
– Postoperative radiotherapy is recommended.
– Subtotal resection, so-called debulking resection, is an option in selected cases of thymoma, aiming at facilitating subsequent definitive radiotherapy [IV, C].
– Debulking is not recommended in thymic carcinoma [V, D].
– Postoperative chemoradiotherapy (including cisplatin, etoposide chemotherapy and a total dose of radiation of 60 Gy) may be considered after debulking/
R2 resection [IV, B].
Definitive radiotherapy
– When the patient is not deemed to be a surgical candidate, definitive radiotherapy is recommended as part of a sequential chemoradiotherapy strategy [III, A].
– Combination with chemotherapy (including cisplatin, etoposide chemotherapy and a total dose of radiation of 60–66 Gy) may be considered [V, C].
Continued
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv277 | v

by guest
on 05 February 2018
Table 7. Continued
Definitive chemotherapy
– Chemotherapy should be offered as the single modality treatment in advanced, non-resectable, non-irradiable or metastatic (stage IVB) thymic epithelial
tumour [III, A].
– Cisplatin-based multiagent combination regimens should be administered [III, A].
– Combination of cisplatin, doxorubicin and cyclophosphamide is preferred [III, B].
– Combination of carboplatin and paclitaxel is an option for thymic carcinoma [III, B].
– Surgery or radiotherapy is possible in rare and selected metastatic cases without proven outcome benefit [IV, C].
– RECIST v1.1 criteria should be used to assess response to chemotherapy [V, A].
Recurrences
– Recurrences of thymic epithelial tumours should be managed according to the same strategy as newly diagnosed tumours [IV, A].
– Complete resection of recurrent lesions, when achievable, is recommended.
– Several consecutive lines of chemotherapy may be administered when the patient presents with tumour progression. The re-administration of a previously
effective regimen should be considered [IV, B].
– Preferred regimens for second-line treatment include carboplatin plus paclitaxel, and platin plus etoposide [III, B]; capecitabine plus gemcitabine is an
option [III, B].
– Options for subsequent lines include pemetrexed [III, B] and oral etoposide.
– In patients with octreoscan-positive thymoma not eligible to receive additional chemotherapy, octreotide alone or with prednisone may represent a
valuable option [III, B].
Targeted agents
– KIT sequencing (exons 9–17) is an option for refractory thymic carcinomas in the setting of potential access to specific inhibitors, particularly in the
context of clinical trials [IV, B].
– It is not recommended to administer imatinib in the absence of a KIT-sensitising mutation [III, E].
– Sunitinib is an option as second-line treatment of thymic carcinomas independently from KIT status [III, A].
– Everolimus may represent an option for refractory tumours [III, B].
Follow-up
– Baseline thoracic CT scan should be carried out 3–4 months after surgery [V, C].
– For completely resected stage I/II thymomas: CT scan should be done every year for 5 years, then every 2 years [V, C].
– For stage III/IV thymomas, thymic carcinoma or after R1–2 resection: CT scan should be done every 6 months for 2 years, then annually [V, C].
– Follow-up may be continued for 10–15 years [V, C].
– Patients with clinical myasthenia gravis, or even isolated positive anti-acetyl choline receptor antibodies, should be informed and educated about the risks
of myasthenic crisis in specific situations such as stress or the administration of certain drugs [V, A].
WHO, World Health Organization; MEN1, multiple endocrine neoplasia 1; CT, computed tomography; MRI, magnetic resonance imaging; PET scan,
positron emission tomography scan; AJCC, American Joint Committee on Cancer; UICC, Union for International Cancer Control; OS, overall survival;
RECIST, Response Evaluation Criteria in Solid Tumours; IASLC, International Association for the Study of Lung Cancer; ITMIG, International Thymic
Malignancy Interest Group; TNM, tumour node metastasis.
Options for subsequent lines include pemetrexed [78] [III, B] or predictive value. However, the recent identification of mo-
and oral etoposide. In patients with octreoscan-positive thymoma lecular alterations occurring in the KIT, vascular endothelial
not eligible to receive additional chemotherapy, octreotide alone or growth factor receptors (VEGFRs) and mammalian target of
with prednisone may represent a valuable option [III, B] [79, 80]. rapamycin (mTOR) signalling pathways, may lead to consider-
ation—in an off-label setting—of the use of targeted agents for
the treatment of refractory thymic malignancies.
Molecular characterisation of thymic epithelial tumours indi- targeting of KIT
cates the occurrence of chromosomal aberrations, altered DNA While KIT is overexpressed in 80% of thymic carcinomas, KIT
methylation and deregulated expression of cancer-related genes, gene mutations are found only in 9% of cases, consisting of
such as CDKN2, MGMT, FOXC1, IGF-1R [81]; some of those mutations observed in gastrointestinal stromal tumours or mela-
alterations and gene expression signatures were reported to asso- nomas (V560del, L576P), or restricted to thymic carcinomas
ciate preclinically with the efficacy of some targeted agents, or to (H697Y, D820E) [81]. Responses were reported with the use of
predict recurrence or survival of patients [81, 82]. KIT tyrosine kinase inhibitors (TKIs) imatinib, sunitinib or sor-
Personalised medicine approaches in thymic malignancies are afenib, mostly in single-case observations [59]. KIT-mutant
ultimately hampered by the limited amount of available research tumours are not uniformly sensitive to imatinib, based on the
to identify reliable, validated molecular markers with prognostic clinical and/or the preclinical evidence in thymic carcinoma
v | Girard et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
and/or other KIT-mutant malignancies. KIT sequencing (exons [87]. Everolimus was evaluated in thymic epithelial tumours in
9–17) is an option for refractory thymic carcinomas in the a recently reported phase II trial reporting on a 22% response
setting of potential access to such inhibitors, particularly in the rate, as well as a 93% DCR [88]. Everolimus may, therefore, re-
context of clinical trials [IV, B]. present an off-label option for refractory tumours [III, B].
It is not recommended to administer imatinib in the absence
of a KIT-sensitising mutation, following the report showing the
absence of activity in two phase II trials conducted in this follow-up and long-term implications
setting [III, E] [83, 84]. No prospective data are available to build recommendations
The use of KIT inhibitors is off-label in thymic malignancies. about post-treatment oncological follow-up of patients. While a
relapse might still be treatable in a curative-intent, patients should
targeting of angiogenesis benefit from a regular radiological assessment. Based on expert
Available KIT TKIs also potently inhibit other kinases, including consensus, and data of cumulative incidence of recurrences over
VEGFRs and platelet-derived growth factor receptors activated in time [28], the proposals are the following [V, C]:
thymic malignancies [85, 86]. A phase II trial recently demon- • baseline thoracic CT scan 3–4 months after surgery
strated the efficacy of sunitinib in terms of response and disease • for completely resected stage I/II thymomas: CT scan every
control rate (DCR) in thymic epithelial tumours, including thymic year for 5 years, then every 2 years
carcinomas [objective response rate (ORR) 26%; DCR: 91%] and, • for stage III/IV thymomas, thymic carcinoma or after R1–2
to a lesser extent, thymomas (ORR: 6%; DCR: 81%) [82]. resection: CT scan every 6 months for 2 years, then annually
Sunitinib may then represent an off-label option as second-line • continuation of follow-up for 10–15 years
treatment of thymic carcinomas, independently from KIT status
[III, A]. There is no reliable clinical data reporting on anti-tumour Secondary tumours may occur. Besides oncological follow-up,
efficacy of other anti-angiogenic drugs, including aflibercept and clinicians should be aware of the possible late onset of new auto-
bevacizumab. immune disorders. Patients with clinical myasthenia gravis, or
even isolated positive anti-acetyl choline receptor antibodies,
targeting mammalian target of rapamycin should be informed and educated about the risks of myasthenic
mTOR is emerging as a potential target in thymic epithelial crisis in specific situations such as stress or the administration of
tumours, following tumour responses observed in phase I trials certain drugs [V, A].
Participation in collaborative research initiatives, including
Table 8. Levels of evidence and grades of recommendation regional and international databases, and clinical trials when
(adapted from the Infectious Diseases Society of America-United available, is recommended.
Levels of evidence methodology
I Evidence from at least one large randomised, controlled trial These clinical practice guidelines were developed in accordance
of good methodological quality (low potential for bias) or with the ESMO standard operating procedures for clinical prac-
meta-analyses of well-conducted randomised trials without tice guidelines development. The relevant literature has been
heterogeneity selected by the expert authors, using the Medline database, as
II Small randomised trials or large randomised trials with a well as abstracts lists from major surgery, medical oncology and
suspicion of bias (lower methodological quality) or meta- thoracic oncology meetings, over the past 20 years. A summary
analyses of such trials or of trials with demonstrated of recommendations is presented in Table 7. Levels of evidence
heterogeneity and grades of recommendation have been applied using the
system shown in Table 8 [89]. Statements without grading were
considered justified standard clinical practice by the experts and
the ESMO faculty. This manuscript has been subjected to an an-
Grades of recommendation onymous peer review process.
strongly recommended acknowledgements
clinical benefit, generally recommended The authors thank Benjamin Besse (Gustave Roussy, Villejuif,
C Insufficient evidence for efficacy or benefit does not outweigh France) for reviewing the manuscript. NG thanks all investiga-
the risk or the disadvantages (adverse events, costs, …), tors of the RYTHMIC network for their input on the French
optional recommendations, which was valuable for the subsequent devel-
D Moderate evidence against efficacy or for adverse outcome, opment of the present clinical practice guidelines.
conflict of interest
recommended
NG has reported consultancy from Amgen, Astra-Zeneca, Bristol-
a
By permission of the Infectious Diseases Society of America [89]. Myers Squibb, Boehringer Ingelheim, Eli-Lilly, Hoffmann-La
Roche, Novartis, Pfizer, Teva. SP has provided consultancy, attended
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv277 | v

by guest
on 05 February 2018
advisory boards and/or provided lectures for: F. Hoffmann-La 21. Kaira K, Endo M, Abe M et al. Biologic correlation of 2-[18F]-fluoro-2-deoxy-D-
Roche, Ltd, Eli-Lilly and Company Oncology, Astra-Zeneca, glucose uptake on positron emission tomography in thymic epithelial tumors. J
Clin Oncol 2010; 28: 3746–3753.
Pfizer, Boehringer Ingelheim, Bristol-Myers Squibb, Daiichi-
22. Marchevsky A, Marx A, Ströbel P et al. Policies and reporting guidelines for small
Sankyo, Morphotek, Merrimack, Merck Serono, MSD, Amgen,
biopsy specimens of mediastinal masses. J Thorac Oncol 2011; 6(Suppl 3):
Clovis, Astellas and Tesaro. The other authors have declared no S1724–S1729.
potential conflicts of interest. 23. Wang H, Sima CS, Beasley MB et al. Classification of thymic epithelial neoplasms
is still a challenge to thoracic pathologists: a reproducibility study using digital
microscopy. Arch Pathol Lab Med 2014; 138: 658–663.
24. den Bakker MA, Roden AC, Marx A, Marino M. Histologic classification of thymoma:
references a practical guide for routine cases. J Thorac Oncol 2014; 9(Suppl 2): S125–S130.
1. de Jong WK, Blaauwgeers JL, Schaapveld M et al. Thymic epithelial tumours: a 25. Masaoka A, Monden Y, Nakahara K, Tanioka T. Follow-up study of thymomas with
population-based study of the incidence, diagnostic procedures and therapy. Eur J special reference to their clinical stages. Cancer 1981; 48: 2485–2492.
Cancer 2008; 44: 123–130. 26. Koga K, Matsuno Y, Noguchi M et al. A review of 79 thymomas: modification of
2. Travis WB, Brambilla A, Muller-Hermelinck HK, Marx A. Pathology and genetics of staging system and reappraisal of conventional division into invasive and non-
tumours of the lung, pleura, thymus and heart. In: Travis WB (ed), World Health invasive thymoma. Pathol Int 1994; 44: 359–367.
Organization Classification of Tumours. Lyon: IARC Press 2004; p. 146. 27. Detterbeck FC, Nicholson AG, Kondo K et al. The Masaoka-Koga stage
3. Marx A, Ströbel P, Badve SS et al. ITMIG consensus statement on the use of the classification for thymic malignancies: clarification and definition of terms. J
WHO histological classification of thymoma and thymic carcinoma: refined Thorac Oncol 2011; 6(Suppl 3): S1710–S1716.
definitions, histological criteria, and reporting. J Thorac Oncol 2014; 9: 596–611. 28. Detterbeck F. Towards a TNM based prognostic classification for thymic tumours.
4. Kondo K, Monden Y. Therapy for thymic epithelial tumors: a clinical study of 1,320 J Thorac Oncol 2013; 8(Suppl 2): S68 (abst. MS16.2).
patients from Japan. Ann Thorac Surg 2003; 76: 878–884. 29. Detterbeck F, Youssef S, Ruffini E, Okumura M. A review of prognostic factors in
5. Weis CA, Yao X, Deng Y et al. The impact of thymoma histotype on prognosis in a thymic malignancies. J Thorac Oncol 2011; 6(Suppl 3): S1698–S1704.
worldwide database. J Thorac Oncol 2015; 10: 367–372. 30. Detterbeck FC, Stratton K, Giroux D et al. The IASLC/ITMIG Thymic Epithelial
6. Ruffini E, Detterbeck F, Van Raemdonck D et al. Tumours of the thymus: a cohort Tumors Staging Project: proposal for an evidence-based stage classification
study of prognostic factors from the European Society of Thoracic Surgeons system for the forthcoming (8th) edition of the TNM classification of malignant
database. Eur J Cardiothorac Surg 2014; 46: 361–368. tumors. J Thorac Oncol 2014; 9(Suppl 2): S65–S72.
7. Omasa M, Date H, Sozu T et al. Postoperative radiotherapy is effective for thymic 31. Marom EM, Milito MA, Moran CA et al. Computed tomography findings predicting
carcinoma but not for thymoma in stage II and III thymic epithelial tumors: The Japanese invasiveness of thymoma. J Thorac Oncol 2011; 6: 1274–1281.
Association for Research on the Thymus Database Study. Cancer 2015; 121: 32. Zhao Y, Chen H, Shi J et al. The correlation of morphological features of chest
1008–1016. computed tomographic scans with clinical characteristics of thymoma. Eur J
8. Kojima Y, Ito H, Hasegawa S et al. Resected invasive thymoma with multiple Cardiothorac Surg 2014 [Epub ahead of print].
endocrine neoplasia type 1. Jpn J Thorac Cardiovasc Surg 2006; 54: 171–173. 33. Korst RJ, Bezjak A, Blackmon S et al. Neoadjuvant chemoradiotherapy for locally
9. Engels EA, Pfeiffer RM. Malignant thymoma in the United States: demographic advanced thymic tumors: a phase II, multi-institutional clinical trial. J Thorac
patterns in incidence and associations with subsequent malignancies. Int J Cancer Cardiovasc Surg 2014; 147: 36–44.
2003; 105: 546–551. 34. Huang J, Detterbeck FC, Wang Z, Loehrer PJ, Sr. Standard outcome measures for
10. Seki S, Koyama H, Ohno Y et al. Diffusion-weighted MR imaging vs. multi-detector thymic malignancies. J Thorac Oncol 2011; 6(Suppl 3): S1691–S1697.
row CT: direct comparison of capability for assessment of management needs for 35. Davenport E, Malthaner RA. The role of surgery in the management of thymoma: a
anterior mediastinal solitary tumors. Eur J Radiol 2014; 83: 835–842. systematic review. Ann Thorac Surg 2008; 86: 673–684.
11. Evoli A, Lancaster E. Paraneoplastic disorders in thymoma patients. J Thorac 36. Detterbeck FC, Moran C, Huang J et al. Which way is up? Policies and procedures
Oncol 2014; 9(Suppl 2): S143–S147. for surgeons and pathologists regarding resection specimens of thymic
12. Marx A, Hohenberger P, Hoffmann H et al. The autoimmune regulator AIRE in malignancy. J Thoracic Oncol 2011; 6(Suppl 3): S1730–S1738.
thymoma biology: autoimmunity and beyond. J Thorac Oncol 2010; 5(Suppl 4): 37. Nakagawa K, Asamura H, Sakurai H et al. Does the mode of surgical resection
S266–S272. affect the prognosis/recurrence in patients with thymoma? J Surg Oncol 2014;
13. Carter BW, Marom EM, Detterbeck FC. Approaching the patient with an anterior 109: 179–183.
mediastinal mass: a guide for clinicians. J Thorac Oncol 2014; 9(Suppl 2): 38. Yano M, Sasaki H, Moriyama S et al. Preservation of phrenic nerve involved by
S102–S109. stage III thymoma. Ann Thorac Surg 2010; 89: 1612–1619.
14. Falkson CB, Bezjak A, Darling G et al. The management of thymoma: a systematic 39. Hamdi S, Mercier O, Fadel E et al. Is sacrifying the phrenic nerve during thymoma
review and practice guideline. J Thorac Oncol 2009; 4: 911–919. resection worthwhile? Eur J Cardiothorac Surg 2014; 45: e151–e155.
15. Girard N, Mornex F, Van Houtte P et al. Thymoma: a focus on current therapeutic 40. Toker A, Sonett J, Zielinski M et al. Standard terms, definitions, and policies for
management. J Thorac Oncol 2009; 4: 119–126. minimally invasive resection of thymoma. J Thorac Oncol 2011; 6(Suppl 3):
16. NCCN Clinical Practice Guidelines in Oncology. Thymic malignancies. V.1.2015. S1739–S1742.
www.nccn.org (7 July 2015, date last accessed). 41. Bhora FY, Chen DJ, Detterbeck FC et al. The ITMIG/IASLC Thymic Epithelial
17. Evans AG, French CA, Cameron MJ et al. Pathologic characteristics of NUT Tumors Staging Project: a Proposed Lymph Node Map for Thymic Epithelial
midline carcinoma arising in the mediastinum. Am J Surg Pathol 2012; 36: Tumors in the Forthcoming 8th Edition of the TNM Classification of Malignant
1222–1227. Tumors. J Thorac Oncol 2014; 9(Suppl 2): S88–S96.
18. Ackman JB, Verzosa S, Kovach AE et al. High rate of unnecessary thymectomy 42. Forquer JA, Rong N, Fakiris AJ et al. Postoperative radiotherapy after surgical
and its cause. Can computed tomography distinguish thymoma, lymphoma, resection of thymoma: differing roles in localized and regional disease. Int J Radiat
thymic hyperplasia, and thymic cysts? Eur J Radiol 2015; 84: 524–533. Oncol Biol Phys 2010; 76: 440–445.
19. Priola AM, Priola SM, Ciccone G et al. Differentiation of rebound and lymphoid 43. Patel S, Macdonald OK, Nagda S et al. Evaluation of the role of radiation therapy in
thymic hyperplasia from anterior mediastinal tumors with dual-echo chemical-shift the management of malignant thymoma. Int J Radiat Oncol Biol Phys 2012; 82:
MR imaging in adulthood: reliability of the chemical-shift ratio and signal intensity 1797–1801.
index. Radiology 2015; 274: 238–249. 44. Ahmad U, Yao X, Detterbeck F et al. Thymic carcinoma outcomes and prognosis:
20. Henschke CI, Lee IJ, Wu N et al. CT screening for lung cancer: prevalence and results of an international analysis. J Thorac Cardiovasc Surg 2015; 149:
incidence of mediastinal masses. Radiology 2006; 239: 586–590. 95–100.
v | Girard et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
45. Ruffini E, Detterbeck F, Van Raemdonck D et al. Thymic carcinoma: a cohort study European Organization for Research and Treatment of Cancer Lung Cancer
of patients from the European Society of Thoracic Surgeons Database. J Thorac Cooperative Group. J Clin Oncol 1996; 14: 814–820.
Oncol 2014; 9: 541–548. 68. Loehrer PJ, Sr, Jiroutek M, Aisner S et al. Combined etoposide, ifosfamide, and
46. Korst RJ, Kansler AL, Christos PJ, Mandal S. Adjuvant radiotherapy for thymic cisplatin in the treatment of patients with advanced thymoma and thymic
epithelial tumors: a systematic review and meta-analysis. Ann Thorac Surg 2009; carcinoma: an intergroup trial. Cancer 2001; 91: 2010–2015.
87: 1641–1647. 69. Lemma GL, Lee JW, Aisner SC et al. Phase II study of carboplatin and paclitaxel in
47. Rimner A, Gomez DR, Wu AJ et al. Failure patterns relative to radiation treatment advanced thymoma and thymic carcinoma. J Clin Oncol 2011; 29: 2060–2065.
fields for stage II-IV thymoma. J Thorac Oncol 2014; 9: 403–409. 70. Palmieri G, Merola G, Federico P et al. Preliminary results of phase II study of
48. Gomez D, Komaki R, Yu J et al. Radiation therapy definitions and reporting capecitabine and gemcitabine (CAP-GEM) in patients with metastatic pretreated
guidelines for thymic malignancies. J Thorac Oncol 2011; 6(Suppl 3): S1743–8. thymic epithelial tumors (TETs). Ann Oncol 2010; 21: 1168–1172.
49. Sugie C, Shibamoto Y, Ikeya-Hashizume C et al. Invasive thymoma: postoperative 71. Hirai F, Yamanaka T, Taguchi K et al. A multicenter phase II study of carboplatin
mediastinal irradiation, and low-dose entire hemithorax irradiation in patients with and paclitaxel for advanced thymic carcinoma: WJOG4207L. Ann Oncol 2015;
pleural dissemination. J Thorac Oncol 2008; 3: 75–81. 26: 363–368.
50. Zhang H, Lu N, Wang M et al. Postoperative radiotherapy for stage I thymoma: a 72. Okuma Y, Saito M, Hosomi Y et al. Key components of chemotherapy for
prospective randomized trial in 29 cases. Chin Med J (Engl) 1999; 112: 136–138. thymic malignancies: a systematic review and pooled analysis for anthracycline-,
51. Gao L, Wang C, Fang W et al. Outcome of multimodality treatment for 188 cases carboplatin- or cisplatin-based chemotherapy. J Cancer Res Clin Oncol 2015;
of type B3 thymoma. J Thorac Oncol 2013; 8: 1329–1334. 141: 323–331.
52. Utsumi T, Shiono H, Kadota Y et al. Postoperative radiation therapy after complete 73. Benveniste MF, Korst RJ, Rajan A et al. A practical guide from the International
resection of thymoma has little impact on survival. Cancer 2009; 115: 5413–5420. Thymic Malignancy Interest Group (ITMIG) regarding the radiographic assessment
53. Chen YD, Feng QF, Lu HZ et al. Role of adjuvant radiotherapy for stage II thymoma of treatment response of thymic epithelial tumors using modified RECIST criteria.
after complete tumor resection. Int J Radiat Oncol Biol Phys 2010; 78: J Thorac Oncol 2014; 9: S119–S124.
1400–1406. 74. Sandri A, Cusumano G, Lococo F et al. Long-term results after treatment for
54. Weksler B, Shende M, Nason KS et al. The role of adjuvant radiation therapy for recurrent thymoma: a multicenter analysis. J Thorac Oncol 2014; 9: 1796–1804.
resected stage III thymoma: a population-based study. Ann Thorac Surg 2012; 93: 75. Hamaji M, Allen MS, Cassivi SD et al. The role of surgical management in
1822–1828. recurrent thymic tumors. Ann Thorac Surg 2012; 94: 247–254.
55. Attaran S, McCormack D, Pilling J, Harrison-Phipps K. Which stages of thymoma 76. Mizuno T, Okumura M, Asamura H et al. Surgical management of recurrent thymic
benefit from adjuvant chemotherapy post-thymectomy? Interact Cardiovasc Thorac epithelial tumors: a retrospective analysis based on the Japanese nationwide
Surg 2012; 15: 273–275. database. J Thorac Oncol 2015; 10: 199–205.
56. Loehrer PJ, Sr, Chen M, Kim K et al. Cisplatin, doxorubicin, and cyclophosphamide 77. Lara PN, Jr, Bonomi PD, Faber LP. Retreatment of recurrent invasive thymoma with
plus thoracic radiation therapy for limited-stage unresectable thymoma: an intergroup platinum, doxorubicin, and cyclophosphamide. Chest 1996; 110: 1115–1117.
trial. J Clin Oncol 1997; 15: 3093–3099. 78. Loehrer PJ, Yiannoutsos CT, Dropcho S et al. A phase II trial of pemetrexed in
57. Kim ES, Putnam JB, Komaki R et al. Phase II study of a multidisciplinary approach patients with recurrent thymoma or thymic carcinoma. J Clin Oncol 2006;
with induction chemotherapy, followed by surgical resection, radiation therapy, and 24(Suppl 18): abstr 7079.
consolidation chemotherapy for unresectable malignant thymomas: final report. 79. Loehrer PJ, Sr, Wang W, Johnson DH et al. Octreotide alone or with prednisone in
Lung Cancer 2004; 44: 369–379. patients with advanced thymoma and thymic carcinoma: an Eastern Cooperative
58. Berruti A, Borasio P, Gerbino A et al. Primary chemotherapy with adriamycin, Oncology Group Phase II Trial. J Clin Oncol 2004; 22: 293–299.
cisplatin, vincristine and cyclophosphamide in locally advanced thymomas: a 80. Palmieri G, Montella L, Martignetti A et al. Somatostatin analogs and prednisone in
single institution experience. Br J Cancer 1999; 81: 841–845. advanced refractory thymic tumors. Cancer 2002; 94: 1414–1420.
59. Girard M, Lal R, Wakelee H et al. Chemotherapy definitions and policies for thymic 81. Rajan A, Girard N, Marx A. State of the art of genetic alterations in thymic
malignancies. J Thorac Oncol 2011; 6(7 Suppl 3): S1749–S1755. epithelial tumors. J Thorac Oncol 2014; 9(Suppl 2): S131–S136.
60. Girard N. Chemotherapy and targeted agents for thymic malignancies. Expert Rev 82. Thomas A, Rajan A, Berman A et al. Sunitinib in patients with chemotherapy-
Anticancer Ther 2012; 12: 685–695. refractory thymoma and thymic carcinoma: an open-label phase 2 trial. Lancet
61. Wright CD, Choi NC, Wain JC et al. Induction chemoradiotherapy followed by Oncol 2015; 16: 177–186.
resection for locally advanced Masaoka stage III and IVA thymic tumors. Ann 83. Giaccone G, Rajan A, Ruijter R et al. Imatinib mesylate in patients with WHO B3
Thorac Surg 2008; 85: 385–389. thymomas and thymic carcinomas. J Thorac Oncol 2009; 4: 1270–1273.
62. Wright CD. Extended resections for thymic malignancies. J Thorac Oncol 2010; 84. Salter JT, Lewis D, Yiannoutsos C et al. Imatinib for the treatment of thymic
5(10 Suppl 3): S344–S347. carcinoma. J Clin Oncol 2008; 26(15s suppl): abstr 8116.
63. Yellin A, Simansky DA, Ben-Avi R et al. Resection and heated pleural chemoperfusion 85. Ströbel P, Bargou R, Wolff A et al. Sunitinib in metastatic thymic carcinomas:
in patients with thymic epithelial malignant disease and pleural spread: a single- laboratory findings and initial clinical experience. Br J Cancer 2010; 103: 196–200.
institution experience. J Thorac Cardiovasc Surg 2013; 145: 83–87. 86. Lattanzio R, La Sorda R, Facciolo F et al. Thymic epithelial tumors express vascular
64. Wright CD. Pleuropneumonectomy for the treatment of Masaoka stage IVA endothelial growth factors and their receptors as potential targets of antiangiogenic
thymoma. Ann Thorac Surg 2006; 82: 1234–1239. therapy: a tissue micro array-based multicenter study. Lung Cancer 2014; 85:
65. Hamaji M, Kojima F, Omasa M et al. A meta-analysis of debulking surgery versus 191–196.
surgical biopsy for unresectable thymoma. Eur J Cardiothorac Surg 2015; 47: 87. Wheler J, Hong D, Swisher SG et al. Thymoma patients treated in a phase I clinic
602–607. at MD Anderson Cancer Center: responses to mTOR inhibitors and molecular
66. Loehrer PJ, Sr, Kim K, Aisner SC et al. Cisplatin plus doxorubicin plus analyses. Oncotarget 2013; 4: 890–898.
cyclophosphamide in metastatic or recurrent thymoma: final results of an intergroup 88. Zucali PA, De Pas TM, Palmieri G et al. Phase II study of everolimus in patients
trial. The Eastern Cooperative Oncology Group, Southwest Oncology Group, and with thymoma and thymic carcinoma previously treated with cisplatin-based
Southeastern Cancer Study Group. J Clin Oncol 1994; 12: 1164–1168. chemotherapy. J Clin Oncol 2014; 32 (5s suppl): abstr 7527.
67. Giaccone G, Ardizzoni A, Kirkpatrick A et al. Cisplatin and etoposide combination 89. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections among
chemotherapy for locally advanced or metastatic thymoma. A phase II study of the hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33: 139–144.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv277 | v

by guest
on 05 February 2018
Malignant pleural mesothelioma: ESMO Clinical Practice

P. Baas1,2, D. Fennell3, K. M. Kerr4, P. E. Van Schil5, R. L. Haas6 & S. Peters7, on behalf of the ESMO
1
Department of Thoracic Oncology, The Netherlands Cancer Institute, Amsterdam; 2The Academic Medical Center, Amsterdam, The Netherlands; 3Department of Medical
Oncology, University of Leicester, Leicester; 4Department of Pathology, University of Aberdeen, Aberdeen, UK; 5Department of Thoracic and Vascular Surgery, University of
Antwerp, Antwerp, Belgium; 6Department of Radiation Oncology, The Netherlands Cancer Institute, Amsterdam, The Netherlands; 7Department of Medical Oncology,
Centre Hospitalier Universitaire Vaudois (CHUV), Lausanne, Switzerland
epidemiology Plain chest radiography lacks sufficient sensitivity for routine

staging. Significant volumes of pleural effusions can mask
incidence pleural/chest lesions and make small malignant pleural effusions
Malignant pleural mesothelioma (MPM) is considered to be a undetectable.
relatively rare tumour. In Great Britain, the incidence in males is When an occupational history indicates considerable asbestos
3.4/100 000, in France it is 2.3/100 000 and in the Netherlands it exposure, or the radiology is suggestive of mesothelioma, cy-
is 3.2/100 000. In the last 10 years, the incidence of MPM has tology can be used to detect malignant cells but histological spe-
increased slightly, mainly due to the lag time of 30–50 years after cimens must often be obtained (see ‘pathology’ section).
exposure to asbestos and the banning of handling and importing A thoracoscopy is recommended to obtain adequate histology,
this product in the late twentieth century. The World Health to optimally stage, and to allow pleural fluid evacuation (with or
clinical practice
Organisation (WHO) estimates asbestos-related disease (ARD) without pleurodesis) [3, 4]. This can be performed as a pleuro-
guidelines
accounts for 92 250 deaths per year globally [1]. Occupational scopy or as video-assisted thoracic surgery (VATS). MPM can be
exposure to asbestos accounts for more than 80% of the cases and difficult to identify and it is therefore recommended to obtain bi-
makes MPM a preventable disease. Although the Western world opsies from tissue of both abnormal and normal appearance.
is moving towards a levelling-off of ARD incidence, the continued When a thoracoscopy is not feasible or contra-indicated, ultra-
use of asbestos in the developing world could lead to a global epi- sound-guided true-cut biopsies are a good alternative. Besides a
demic of MPM. Even though asbestos is banned in Europe, other clinical reason to obtain a diagnosis, there are medico-legal
developed countries have only controlled the import, but not reasons to confirm the diagnosis of MPM. In Europe, the require-
abolished handling of asbestos products. Recently, a germline ments for this vary between countries. To date, there are no
mutation in the BAP1 gene has been linked to predisposition in studies that recommend screening of patients who have had any
some cases of MPM [2]. Somatic mutations may also play a role (occupational) history of asbestos exposure.
in the development of MPM. Circulating tumour markers have been tested to a great extent
and only a few have been able to facilitate the diagnostic process:
diagnosis cyfra 21.1, Fibulin-3 and Mesothelin all lack specificity and
Patients typically present with complaints of shortness of should not be used as specific markers for mesothelioma [5, 6].
breath, pain and weight loss. These symptoms can occur over a Carcinoembryonic antigen (CEA) is a negative marker and is
period of many months. During physical examination, unilateral not increased in MPM [7]. It can therefore be used to rule out
effusions are often observed. It is of great importance that a MPM if cytological/histological analysis is inconclusive.
detailed occupational history is obtained.
Standard work-up includes:
Recommendation 1
• Chest X-ray Diagnostic procedures in MPM should encompass at least
• Computed tomography (CT) scan of chest and upper abdomen
• Occupational history with emphasis on asbestos exposure [II, A]
• Thoracentesis, with examination of the pleural effusion
• CT scanning of the thorax [II, A]
• General laboratory blood tests • In all patients who have a unilateral pleural thickening, with or
without fluid and/or calcified asbestos plaques, efforts should be
*Correspondence to: ESMO Guidelines Committee, ESMO Head Office, via L. Taddei 4, made to obtain a pathological specimen, as there are no specific
CH-6962 Viganello-Lugano, Switzerland.
clinical features of MPM [II, A]
• There is no place for screening of persons exposed to asbestos [IV, B]
†
Approved by the ESMO Guidelines Committee: December 2004, last update July 2015. • Tumour markers cannot distinguish MPM [II, B]
This publication supersedes the previously published version. Ann Oncol 2010; 21
(Suppl. 5): v126–v128.

by guest
on 05 February 2018
pathology from organising fibrinous exudates (fibrinous/fibrous pleurisy)

requires a full-thickness biopsy sample, with correct orientation
The pathological diagnosis of MPM may be difficult for a of histological sections, perpendicular to the pleural surface.
number of reasons: Pathological details of these differential diagnoses are discussed
• MPMs are a heterogeneous group of tumours, with the ability elsewhere [9, 10].
to mimic almost any other form of malignant tumour. The most commonly used mesothelial markers are calretinin,
• The three main subtypes (epithelioid, biphasic and sarcoma- cytokeratin 5/6, WT1 and podoplanin (D240). For (adeno)carcin-
toid) have numerous variants, as described in the 2004 WHO oma, the most useful markers are TTF1, CEA and EP4 [9, 10].
classification [8]. Some markers have been advocated for, due to their distinction of
• The pleura is a common site for metastatic disease and react- benign (desmin) versus malignant mesothelial cells (EMA, p53,
ive changes in the pleura may be confused with MPM. GLUT1, IMP3), but these methods lack reliability and are gener-
• There are other uncommon benign and malignant pleural ally not recommended. Other immunohistochemical markers
tumours. may be appropriate, depending on the differential diagnosis in a
particular case. It is worth noting that although pan-cytokeratin
Samples for diagnosis may vary widely: pleural effusions, small markers are not specific in any way for mesothelial cells, or malig-
(closed) pleural biopsies, image-guided needle core biopsies, nancy, in the appropriate context, they can be extremely useful in
larger open or VATS surgical biopsy samples or debulking speci- the diagnosis of sarcomatoid mesothelioma, which often does not
mens. Surgical resection (extrapleural pneumonectomy) is express the usual markers mentioned above.
rarely performed. In some cases, samples may also be obtained The use of in situ hybridisation (e.g., FISH) to detect homozy-
through autopsy. gous deletion of p16 is strongly associated with malignancy; it is not
Significant sampling errors can occur in effusion cytology and specific to MPM, but may aid diagnosis [11, 12]. The role of this
small biopsy samples, but also with larger surgical samples technique has yet to be defined and established. Comprehensive
(though less common). Blind biopsies are not recommended analysis of the genomic landscape of mesothelioma has not yet
because of risk of complications and are no longer indicated since been completely defined, however the Tumour Genome Atlas study
the introduction of thoracic ultrasonography. is currently underway (www.cancergenome.nih.gov).
Cytological features in effusions may permit a diagnosis of
malignancy but reported sensitivities vary widely. When a Recommendation 2
biopsy is not possible, appropriate clinical and radiological fea- A. Definitive diagnosis of MPM on effusion cytology specimens
tures may assist in suggesting a diagnosis of MPM. Many
mesotheliomas lack significant cytological atypia and it is im- • Effusion cytology for definitive diagnosis of MPM remains a
possible to distinguish between benign, reactive mesothelial pro- controversial topic and is still generally not recommended [IV, C].
liferations and MPM. Cytology sample cells may show variable • If effusion cytology is frankly malignant, the diagnosis may be
atypia (usually low grade) and exhibit a mesothelial immune strongly suggested but confirmation by biopsy, if possible, is
phenotype, but malignancy cannot be confirmed. The term recommended [A, no level of evidence].
‘atypical mesothelial proliferation’ is useful in this context, but is • IHC is invaluable to characterise the nature of atypical effusion cells
and sample preparation to facilitate IHC should be carried out if at
insufficient for a diagnosis of MPM. This does not confirm the
all possible [A, no level of evidence].
diagnosis of MPM, but leaves the possibility open [see below for
fluorescence in situ hybridisation (FISH) testing]. B. Definitive diagnosis of MPM on tissue biopsy specimens
In the vast majority of cases, it is necessary to have adequate
tissue biopsies and the use of appropriate immunohistochemistry • The recognition of tissue invasion is required for definitive diagnosis
of MPM [IV, A].
(IHC) for definitive, primary diagnosis of MPM. Consequently,
definitive diagnosis of mesothelioma by frozen section is not • Larger and directly targeted biopsy samples facilitate definitive
diagnosis. Surgical-type samples are preferred for diagnosis [IV, A].
recommended.
• A major subtype diagnosis (epithelioid, biphasic, sarcomatoid)
Tissue biopsy samples the abnormal (mesothelial) cell popu-
should be given in all cases of MPM [IV, A].
lation and permits micro-anatomical assessment of the location
of these cells. This is crucial to identify the extent of invasion. C. IHC in the diagnosis of MPM
IHC is pivotal in confirming the mesothelial nature of cells, but
cannot confirm their biological potential (see below). The larger • IHC is recommended for all primary diagnoses of MPM [IV, A].
• At least two ‘mesothelial’ markers and at least two ‘(adeno)
the tissue biopsy and the more targeted the sampling approach
carcinoma’ markers should be used [V, A].
[radiological or surgical (VATS or open procedure)], the more
• Sarcomatoid MPM often does not express usual ‘mesothelial’
reliable and definitive the diagnosis.
markers [IV, A].
Invasion may be difficult to recognise, especially when tissue
sampling is limited, but identification may be assisted by IHC
(see below). Early invasive mesothelioma is particularly difficult,
often disguised by cutting artefacts or the malorientation of sec-
staging
tions, but may be suspected if there is nodular mesothelial cell Staging procedures are standard in all tumours. Staging not only
proliferation. If definitive invasion cannot be recognised, the describes the anatomical extent of a tumour, but it also corre-
diagnosis of ‘atypical mesothelial proliferation’ is appropriate, lates with prognosis and helps in treatment decision-making. At
and further sampling may be indicated. Distinguishing MPM least five staging systems for MPM have been reported. The first
v | Baas et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Table 1. TNM staging according to the International Mesothelioma patients with mesothelioma, and the TNM system is therefore
Interest Group (IMIG)/Union for International Cancer Control the preferred system.
(UICC) [14] For decision-making, magnetic resonance imaging (MRI),
Stage TNM Comments using gadolinium, may improve delineation of the tumour with
regard to the surrounding tissues, especially when surgical re-
Ia T1a N0 M0 Primary tumour limited to ipsilateral parietal section is considered to be a part of the treatment plan. This will
pleura also help to visualise foci that may be present in the diaphragm,
Ib T1b N0 M0 As stage Ia plus focal involvement of visceral pericardium or chest wall [18].
pleura The use of positron emission tomography (PET) scanning is
still under debate because MPM tends to only grow locally and
II T2 N0 M0 As stage Ia or Ib plus confluent involvement of
metastases occur solely in patients with advanced disease.
diaphragm or visceral pleura or involvement
Discrimination of involved lymph nodes is difficult, due to
of the lung
anatomy and the limited spatial resolution of current PET scans
III Any T3 M0 Locally advanced tumour [19]. PET scanning can be used in the diagnostic work-up when
Any N1 M0 Ipsilateral, bronchopulmonary or hilar lymph PET-avid sites in the thoracic cavity need to be identified to
node involvement obtain representative tissue. Some studies use repeated PET
Any N2 M0 Subcarinal or ipsilateral mediastinal lymph node scanning as a response criterion in addition to the CT scan.
involvement However, no randomised prospective studies have yet been pub-
IV Any T4 Locally advanced, technically unresectable lished on this. One of the caveats in the evaluation of PET scan-
tumour ning is the false-positive outcome after pleurodesis. This can
Any N3 Contralateral mediastinal, internal mammary, result in high activity for a period of more than 6 months after
and ipsilateral or contralateral supraclavicular pleurodesis.
lymph node involvement
Any M1 Distant metastases Recommendation 3
Staging for every patient with a confirmed diagnosis of MPM
Reproduced with permission from the American College of Chest
Physicians. • In the absence of a uniform, robust and validated staging system,
experts advocate the use of the most recent TNM-based IMIG/UICC
classification [III, B].
• The use of MRI is only recommended in special situations when
staging system was introduced by Butchart, consisted of four tumour delineation is necessary [II, B].
stages and was based on observations from only 29 patients • The use of PET scanning is limited and can be used for localisation
[13]. This system was succeeded by others developed by of tumour sites, distant metastases or early response to treatment, as
part of a study protocol [III, B].
Mattson, Boutin and Sugarbaker who based their system on
their own experiences. Most of these staging systems had limita-
tions, being based on small numbers of patients. The most
recent system was developed in 1995; it was presented by the
International Mesothelioma Interest Group (IMIG) and is front-line therapy for mesothelioma
approved by the Union for International Cancer Control
(UICC) (Table 1) [14]. The limitation of most classifications is Front-line chemotherapy improves survival of patients with unre-
their inaccuracy in describing tumour (T-) and node (N-)extent. sectable MPM. Combination doublet chemotherapy of cisplatin,
Most staging systems are based on surgical interventions, as with either pemetrexed or raltitrexed, has shown a longer survival
current imaging techniques have limited resolution. The IMIG compared with cisplatin alone in randomised phase III trials [20,
is currently validating a new tumour-node-metastases (TNM) 21]. Carboplatin is an acceptable alternative to cisplatin and may
staging system, using a large retrospective dataset obtained from be better tolerated in the elderly population [22, 23].
centres all over the world. Several phase II clinical trials are investigating the addition of
Although the IMIG staging system could predict prognosis novel agents to pemetrexed/cisplatin therapy. To date, no agent
[15], it failed to be an independent prognostic factor when ana- has demonstrated superior efficacy. Although the agent CBP501
lysed in the clinical setting using multivariate analysis [16]. After (a G2 checkpoint abrogator) met its primary end point, it was
the first analysis of an IMIG/International Association for the not considered to improve upon the efficacy of standard chemo-
Study of Lung Cancer (IASLC) database with data from 3101 therapy. Trials of anti-angiogenic agents such as bevacizumab
patients with MPM, several areas of the current staging system or sunitinib [24, 25] have so far failed to demonstrate improve-
have been defined as requiring modification [17]. Multivariable ment over standard treatment.
analyses showed significant differences in overall survival (OS)
for most T stages, but not for T2 versus T1. Although a negative
node status was of prognostic importance, no difference between
maintenance therapy for mesothelioma
N1 and N2 was noted. The use of continuation or switch maintenance therapy with
Disease stage according to the TNM system, when assessed by pemetrexed monotherapy has changed practice in the manage-
surgical staging, is an important predictor of prognosis in ment of non-small-cell lung cancer, but is yet to be evaluated in
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv199 | v

by guest
on 05 February 2018
the mesothelioma setting. However, a phase II trial addressing radiotherapy

this question [NCT01085630], led by the Cancer and Leukemia
Group B (CALGB), is currently underway. Switch maintenance, Radiotherapy (RT) can be used for different indications in
with the focal adhesion kinase inhibitor defactinib (VS6063) mesothelioma: as palliation, as preventive treatment and as part
versus placebo, is currently under evaluation in the COMMAND of a multimodality treatment.
trial [NCT01870609]. Another phase III, switch maintenance, For patients suffering from pain (e.g., by chest wall invasion),
study of gemcitabine versus observation is currently on-going in RT, prescribing usually short course regimens, can be considered
the Netherlands (NVALT 19). A recent phase III study evaluating although the systematic review by Macleod et al. [37, 38] sug-
switch maintenance to thalidomide was negative [26]. gested that no high-quality evidence currently exists to support
RT in treating pain in MPM.
In the case of palliation, the aim of RT is to relieve pain and it
is recommended in cases of infiltration of the chest wall or per-
second-line therapy for mesothelioma
meation nodules by MPM. The treatment is usually given in
There is currently no second-line standard of care. Phase III short courses such as 1 × 10 or 3 × 8 Gy.
evaluation of pemetrexed monotherapy in previously treated There is much debate whether a scar after thoracoscopy and/
patients was not associated with longer survival when compared or drainage procedures should be irradiated prophylactically in
with best supportive care (BSC). Post-study chemotherapy has order to reduce the likelihood of seeding metastases. It is prob-
been shown to be associated with significantly longer survival, ably best to recommend refraining from this procedure unless in
with an adjusted hazard ratio of 0.56 [27]. Single agent vinorel- the setting of a clinical trial [39], such as the United Kingdom
bine has shown useful activity in phase II trials [28, 29], demon- ‘PIT’ study (ClinicalTrials.gov Identifier NCT01604005).
strating a trend towards longer survival as was seen in the first- One randomised trial compared immediate drain site RT (21
line study (MSO1) [30]. Gy in three fractions) to observation in 61 patients treated
There are promising developments in the novel agent arena, for between 1998 and 2004 [40]. The authors concluded that
example, anti-mesothelin immunotoxin [31]. Immunotherapy prophylactic drain site RT in MPM did not reduce the incidence
targeting CTLA4 with tremelimumab [32] is under evaluation in a of tumour seeding, as indicated by previous studies conducted
large global phase III trial [NCT01843374]. Recent data suggest in the 1990s. Quality control of RT, the use of first-line therapy
that the PDL1, a putative biomarker for PD1/PDL1 therapy, is sig- and patient selection can probably explain the discrepancy of
nificantly expressed in mesotheliomas, particularly the sarcoma- these results. Puncture points or thoracoscopy scars should be
toid subtype. In the absence of standard second-line or further- identified and checked for early irradiation as soon as the diag-
line therapy, it is recommended that patients are enrolled into nosis of MPM is confirmed (expert advice). A randomised study
clinical trials. of post-intervention radiation of the tract is accruing in the UK
(PIT trial).
pre- and postoperative RT
Individual patient meta-analyses have shown that response to
chemotherapy or progression-free survival (PFS) [33] is corre- Data from the literature are limited and come from retrospective
lated with a longer survival. Personalised therapy is therefore studies only. In general, it is not recommended that RT is admi-
warranted, although currently it is in its infancy. nistered pre- or postoperatively with large fields (hemi-thoracic
Mesotheliomas harbouring epigenetically silenced argininosuc- RT) outside the setting of a clinical trial. The results are poor, in
cinate synthase (AS) are sensitive to arginine-degrading enzymes terms of local control, because of the complex growth patterns
pre-clinically. Open label, randomised clinical evaluation of ADI- in the diaphragmal gutters and in the lobar fissures. The field
PEG 30, in AS-negative patients, has confirmed efficacy with size and neighbouring vital organs contribute considerably to
increased PFS compared with BSC alone. A study to evaluate toxicity. Radiation-induced lung toxicity is especially high when
ADI-PEG20, in combination with chemotherapy, in AS-negative the lung remains in situ after decortication. Improved 3D plan-
mesothelioma [NCT02029690] is underway. Mutation of NF2 ning and the introduction of intensity-modulated RT (IMRT)
occurs in around 50% of mesotheliomas, sensitising inhibition of seem to overcome most of these issues and allow the remaining
FAK [34–36]. Accordingly, this trial is stratifying patients by tumour tissue to be properly irradiated. Currently, the ‘SMART’
Merlin expression. study is investigating a short accelerated course of high-dose
hemithoracic IMRT followed by extrapleural pneumonectomy
(EPP) [41].
Recommendation 4
In the absence of phase III randomised trials, the establish-
The first- and second-line treatment of unresectable mesothelioma
ment of a prospective controlled study evaluating the efficacy
• Anti-folate/platinum doublet is the only approved standard of care and tolerability of adjuvant RT post-EPP is recommended. In
[I, A]. this study, a minimum recommended dose of 50 Gy, with a
• Maintenance therapy (switch or continuation) has not yet improved daily fraction size of 1.8–2 Gy should be given. In one study,
the OS and patients should be included in these studies [II, A]. hemithoracic irradiation (54 Gy) was given as adjuvant therapy
• Patients in good condition should be recommended to join studies in after EPP [42]. The local recurrence rate was 13%, with a 4%
second line [II, A]. local-only recurrence rate. In two other studies, hemithoracic
irradiation, in lower doses, was given as part of a trimodality
v | Baas et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
therapy [43, 44]. The local recurrence rate was 50%–60%, with can be examined. Pleural effusions can be drained and, if
the highest rate within 12 months after completion of treatment. required, a decortication or pleurodesis can be carried out.
One paper addressed the pattern of failure after trimodality
treatment with 3D conformal RT (3DCRT) and highly con-
formal RT (HCRT) in 39 patients. It was concluded that HCRT surgery for staging and palliation
was superior to 3DCRT and in-field recurrences occurred only As most of the staging systems involve the extension of the
in those treated with 3DCRT (16%). It remains unclear whether tumour on the pleural lining and invasion of muscle layers,
technical issues of surgery, the irradiation or other issues (such thoracoscopic inspection of the pleural cavity is required.
as patient selection) are the reason for this observation. Higher Besides this intervention, the staging procedure can also be used
doses of radiation have resulted in better local control [45]. It is to control pleural effusion; perform a talc poudrage or even de-
therefore recommended that this is carried out only in specia- cortication in the case of a captured lung. One study compared
lised centres (expert advice). VATS ( partial) pleurectomy versus standard talc poudrage in
196 patients [47]. OS did not improve in the experimental arm,
choice of radiation type after EPP but control of pleural effusion and quality of life were signifi-
Preliminary results of IMRT in the adjuvant setting after EPP cantly better at 6 and 12 months.
seemed particularly promising. IMRT may provide good local
control and protect at risk organs such as the heart or liver.
Even after removal of one entire lung, fatal pulmonary toxicity surgery with radical intent
remained a problem, with six out of 13 patients developing fatal Due to the intricate location and relation to other normal tissues,
pneumonitis [46]. To predict the risk of pneumonitis, the pul- it is virtually impossible to obtain free resection margins.
monary dosimetric values V20, V5 and mean lung dose (MLD) Therefore, the aim of this procedure is to obtain a macroscopic re-
should be specified: V20 [volume of both lungs minus the plan- section by removing as much visible tumour as possible, using
ning target volume (PTV)] should be less than 15% and MLD different surgical procedures.
should be less than 10 Gy. These dosimetric constraints can also Initially, terms like ‘radical’ pleurectomy and decortication
be used for conformal RT; dose–volume histograms of all target were used without proper description, making comparison
volumes [clinical target volumes (CTV) and PTV] and of all between reported studies difficult. The IASLC established a
critical organs (contralateral lung, cardiac volume, spinal cord, working group to recommend uniform definitions for surgical
oesophagus, liver, right and left kidney) should be clearly stated. procedures dealing with mesothelioma [48]. Currently, a clear
With more dosimetric constraints on the residual contralateral distinction is made between EPP and pleurectomy/decortication
lung, the risk of pneumonitis could be reduced to a minimum (P/D) with different subcategories:
after EPP.
Further studies are needed to better establish the role of RT. • EPP implies a complete en bloc removal of the involved parietal
Recent studies have underlined the importance of RT technique, and visceral pleura including the whole ipsilateral lung. If
both in terms of local control and toxicity. It is therefore recom- required, the diaphragm and pericardium can also be resected.
mended that RT is delivered in specialised centres (expert • Extended P/D is the same procedure but the lung is left in situ:
advice). macroscopic complete resection is still the goal.
• P/D refers to removal of all gross tumour, without resection of
the diaphragm or the pericardium.
Recommendation 5
• A partial pleurectomy entails partial removal of parietal and/
RT can be considered in the following cases
or visceral pleura leaving gross tumour behind.
• For palliation of pain related to tumour growth, RT can be
In a retrospective analysis of data from three large institutions,
considered [II, A].
• The use of RT to prevent growth in drainage tracts is not proved to
663 patients who underwent an EPP or P/D were examined for
be useful [III, A]. survival outcome and toxicity [49]. The operative mortality was
• RT can be given in an adjuvant setting after surgery or slightly higher (7%) for EPP compared with P/D (4%), with a
chemo-surgery to reduce the local failure rate. However, no higher OS of 16 months for P/D versus 12 months for EPP.
evidence is available for its use as a standard treatment [II, A]. Some studies reported on a trimodality approach in order to
• When postoperative RT is applied, strict constraints must be adhered obtain cure. Different combined-modality regimens have been
to in order to avoid toxicity to neighbouring organs, and special, investigated. Similar to locally advanced lung cancer, induction
tissue sparing, techniques should be used [II, A]. chemotherapy was considered to increase the complete resection
rate of early-stage mesothelioma.
A Swiss multicentre trial reported on the additive effect of radi-
ation therapy after a combination of three cycles of cisplatin and
gemcitabine as induction therapy followed by EPP in patients
surgery with resectable MPM [50]. Macroscopic complete resection was
Surgery is used for staging procedures or with palliative or cura- obtained in 37/61 (61%) patients and 36 patients received post-
tive intent. Using VATS or thoracoscopy, large biopsy samples operative RT. The 90-day mortality was 3.2%, but 62% of patients
can be obtained for proper pathological, molecular and IHC experienced one or more complications [empyema (16%) and
analyses. During this procedure, the local extent of the tumour bronchopleural fistula (9.5%)].
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv199 | v

by guest
on 05 February 2018
Diagnosis Recommendation 1
Diagnostic procedures in MPM should encompass at least
• Occupational history with emphasis on asbestos exposure [II, A]
• CT scanning of the thorax [II, A]
• In all patients who have a unilateral pleural thickening, with or without fluid and/or calcified asbestos plaques, efforts should be made
to obtain a pathological specimen, as there are no specific clinical features of MPM [II, A]
• There is no place for screening of persons exposed to asbestos [IV, B]
• Tumour markers cannot distinguish MPM [II, B]
Pathology Recommendation 2
A. Definitive diagnosis of MPM on effusion cytology specimens
• Effusion cytology for definitive diagnosis of MPM remains a controversial topic and is still generally not recommended [IV, C]
• If effusion cytology is frankly malignant, the diagnosis may be strongly suggested but confirmation by biopsy, if possible,
is recommended [A, no level of evidence]
• IHC is invaluable to characterise the nature of atypical effusion cells and sample preparation to facilitate IHC should be carried
out if at all possible [A, no level of evidence]
B. Definitive diagnosis of MPM on tissue biopsy specimens
• The recognition of tissue invasion is required for definitive diagnosis of MPM [IV, A]
• Larger and directly targeted biopsy samples facilitate definitive diagnosis. Surgical-type samples are preferred for diagnosis [IV, A]
• A major subtype diagnosis (epithelioid, biphasic, sarcomatoid) should be given in all cases of MPM [IV, A]
C. IHC in the diagnosis of MPM
• IHC is recommended for all primary diagnoses of MPM [IV, A]
• At least two ‘mesothelial’ markers and at least two ‘(adeno)carcinoma’ markers should be used [V, A]
• Sarcomatoid MPM often does not express usual ‘mesothelial’ markers [IV, A]
Staging Recommendation 3
Staging for every patient with a confirmed diagnosis of MPM
• In the absence of a uniform, robust and validated staging system, experts advocate the use of the most recent TNM-based IMIG/UICC
classification [III, B]
• The use of MRI is only recommended in special situations when tumour delineation is necessary [II, B]
• The use of PET scanning is limited and can be used for localisation of tumour sites, distant metastases or early response to treatment,
as part of a study protocol [III, B]
Personalised Recommendation 4
medicine The first- and second-line treatment of unresectable mesothelioma
• Anti-folate/platinum doublet is the only approved standard of care [I, A]
• Maintenance therapy (switch or continuation) has not yet improved the OS and patients should be included in these studies [II, A]
• Patients in good condition should be recommended to join studies in second line [II, A]
Radiotherapy Recommendation 5
RT can be considered in the following cases
• For palliation of pain related to tumour growth, RT can be considered [II, A]
• The use of RT to prevent growth in drainage tracts is not proved to be useful [III, A]
• RT can be given in an adjuvant setting after surgery or chemo-surgery to reduce the local failure rate. However, no evidence is available
for its use as a standard treatment [II, A]
• When postoperative RT is applied, strict constraints must be adhered to in order to avoid toxicity to neighbouring organs and special,
tissue sparing, techniques should be used [II, A]
Surgery Recommendation 6
The indications for surgery are
• For palliation of pleural effusions when chest tube drainage is not successful [II, A]
• To obtain diagnostic samples of tumour tissue and to stage the patient [II, A]
• To be part of a multimodality treatment, preferably as part of a study [II, A]
• To perform a macroscopic complete resection by means of P/D or EPP [III, C]
MPM, malignant pleural mesothelioma; CT, computed tomography; IHC, immunohistochemistry; TNM, tumour-node-metastasis; IMIG, International
Mesothelioma Interest Group; UICC, Union for International Cancer Control; MRI, magnetic resonance imaging; PET, positron emission tomography;
OS, overall survival; RT, radiotherapy; P/D, pleurectomy/decortication; EPP, extrapleural pneumonectomy.
v | Baas et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
For the patients undergoing EPP, an encouraging median
Recommendation 6
survival time of 23 months was obtained.
The indications for surgery are
In another retrospective study of trimodality therapy, 60
patients underwent induction chemotherapy (cisplatin/anti- • For palliation of pleural effusions when chest tube drainage is not
folate in 30 patients), followed by EPP and postoperative RT to successful [II, A].
50 Gy [51]. The full treatment protocol could be applied in half • To obtain diagnostic samples of tumour tissue and to stage the
of the patients. patient [II, A].
The European Organisation for Research and Treatment of • To be part of a multimodality treatment, preferably as part of a
Cancer (EORTC) studied the feasibility of trimodality therapy study [II, A].
in a phase II trial (EORTC 08031) with clearly defined timelines • To perform a macroscopic complete resection by means of P/D or
[52]. Patients with pathologically proven mesothelioma (up to EPP [III, C].
stage cT3N1M0) received induction chemotherapy (cisplatin
and pemetrexed × 3) followed by EPP within 21–56 days after
the last dose of chemotherapy, in the absence of progressive
disease and unacceptable toxicity. response evaluation and follow-up
A ‘success of treatment’ was defined as a patient who had
received the full protocol and was alive after 90 days without It is advised that response evaluation is performed with CT
progressive disease and without grade 3 or 4 toxicity. Of the 57 scanning and the examinations performed at presentation.
patients included, 42 had EPP (73.7%) after induction therapy. The follow-up of a patient will depend on the local recom-
The 90-day mortality was 6.5%, with an OS of 18.4 months and mendations or as dictated by the protocol in case of study
progression-free median survival of 13.9 months. Only 24 participation.
patients (42.1%) met the definition of success, and therefore the
primary end point was not reached. methodology
A similar phase II trial in the USA, but without predefined
time limits, included 77 patients from nine institutions. The op- These clinical practice guidelines were developed in accordance
erative mortality was 7% and median OS was 16.8 months [53]. with the ESMO standard operating procedures for clinical practice
Although trimodal therapy seemed feasible in selected guidelines development. The relevant literature has been selected
patients with promising results, this concept was tested in the by the expert authors. A summary of recommendations is shown
UK with the Mesothelioma and Radical Surgery 1 (MARS 1) in Table 2. Levels of evidence and grades of recommendation have
trial. MARS 1 was designed as a randomised trial between EPP
and no EPP after induction chemotherapy. In the feasibility
study, 112 patients were entered in 11 centres, during a 3-year
(adapted from the Infectious Diseases Society of America-United
period. Only 50 patients (45%) could be randomised after in-
duction therapy and 16 patients were randomly assigned to
receive EPP. In this small group, there were three case fatalities, Levels of evidence
giving a mortality rate of 18.8% [54]. The median OS for I Evidence from at least one large randomised, controlled trial
patients undergoing EPP was 14 months, compared with 19 of good methodological quality (low potential for bias) or
months for those not having EPP. It was concluded by the meta-analyses of well-conducted randomised trials without
authors that trimodality therapy offers no benefit and in fact heterogeneity
may harm patients. Therefore, a further study, MARS 2, was II Small randomised trials or large randomised trials with a suspicion
designed to assess the feasibility of randomisation into P/D and of bias (lower methodological quality) or meta-analyses of such
not EPP. trials or of trials with demonstrated heterogeneity
In a systematic review of EPP for mesothelioma, carried out in III Prospective cohort studies
2010, 34 studies from 26 institutions were evaluated [55]. Median IV Retrospective cohort studies or case–control studies
OS after EPP varied from 9.4 to 27.5 months, with a 5-year sur- V Studies without control group, case reports, experts opinions
vival rate from 0% to 24%. Overall mortality ranged from 0% to Grades of recommendation
11.8% and morbidity from 22% to 82%. The conclusion of this sys-
tematic review was that selected patients might benefit from EPP,
strongly recommended
especially when combined with induction or adjuvant therapy.
B Strong or moderate evidence for efficacy but with a limited clinical
The safety and efficacy of trimodality treatment was assessed benefit, generally recommended
in a systematic review encompassing 16 studies (including 5 C Insufficient evidence for efficacy or benefit does not outweigh the
prospective trials). The median OS ranged from 12.8 to 46.9 risk or the disadvantages (adverse events, costs, etc.), optional
months with perioperative mortality from 0% to 12.5% [56]. D Moderate evidence against efficacy or for adverse outcome,
The authors concluded that trimodality therapy may offer ac- generally not recommended
ceptable perioperative outcomes and long-term survival in E Strong evidence against efficacy or for adverse outcome, never
selected patients treated in specialised centres. recommended
A multidisciplinary team with sufficient experience should
a
provide recommendations on the suitability of patients for By permission of the Infectious Diseases Society of America [57].
trimodality therapy.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv199 | v

by guest
on 05 February 2018
been applied using the system shown in Table 3. Statements 15. Nowak AK, Armato SG, III, Ceresoli GL et al. Imaging in pleural mesothelioma: a
without grading were considered justified standard clinical practice review of imaging research presented at the 9th International Meeting of the
International Mesothelioma Interest Group. Lung Cancer 2010; 70: 1–6.
by the experts and the ESMO faculty. This manuscript has been
16. Tammilehto L, Kivisaari L, Salminen US et al. Evaluation of the clinical TNM
subjected to an anonymous peer-review process.
staging system for malignant pleural mesothelioma: an assessment in 88 patients.
Lung Cancer 1995; 12: 25–34.
17. Rusch VW, Giroux D, Kennedy C et al. Initial analysis of the International
conflict of interest Association for the Study of Lung Cancer Mesothelioma Database. J Thorac Oncol
PB has reported: consultant for Bristol-Myers Squibb, Merck 2012; 7: 1631–1639.
Sharp & Dohme, Pfizer and Verastem; research funding from 18. Heelan RT, Rusch VW, Begg CB et al. Staging of malignant pleural mesothelioma:
Bristol-Myers Squibb, Merck Sharp & Dohme. DF has reported: comparison of CT and MR imaging. AJR Am J Roentgenol 1999; 172:
1039–1047.
honoraria and advisory board for Lilly; research funding and ad-
19. Armato SG, III, Labby ZE, Coolen J et al. Imaging in pleural mesothelioma: a review
visory board for Pierre Fabre. SP has reported: consultancy, ad- of the 11th International Conference of the International Mesothelioma Interest
visory boards and/or lectures for F. Hoffmann-La Roche Ltd, Eli Group. Lung Cancer 2013; 82: 190–196.
Lilly and Company Oncology, AstraZeneca, Pfizer, Boehringer- 20. Vogelzang NJ, Rusthoven JJ, Symanowski J et al. Phase III study of pemetrexed in
Ingelheim, Bristol-Myers Squibb, Daiichi-Sankyo, Morphotek, combination with cisplatin versus cisplatin alone in patients with malignant pleural
Merrimack, Merck Serono, Merck Sharp & Dohme, Amgen, mesothelioma. J Clin Oncol 2003; 21: 2636–2644.
Clovis, Astellas and Tesaro. The other authors have reported no 21. van Meerbeeck JP, Gaafar R, Manegold C et al. Randomized phase III study of
potential conflicts of interest. cisplatin with or without raltitrexed in patients with malignant pleural
mesothelioma: an intergroup study of the European Organisation for Research and
Treatment of Cancer Lung Cancer Group and the National Cancer Institute of
Canada. J Clin Oncol 2005; 23: 6881–6889.
references
22. Santoro A, O’Brien ME, Stahel RA et al. Pemetrexed plus cisplatin or pemetrexed
1. Delgermaa V, Takahashi K, Park EK et al. Global mesothelioma deaths reported to plus carboplatin for chemonaive patients with malignant pleural mesothelioma:
the World Health Organization between 1994 and 2008. Bull World Health Organ results of the International Expanded Access Program. J Thorac Oncol 2008; 3:
2011; 89: 716–724. 756–763.
2. Testa JR, Cheung M, Pei J et al. Germline BAP1 mutations predispose to 23. Ceresoli GL, Castagneto B, Zucali PA et al. Pemetrexed plus carboplatin in elderly
malignant mesothelioma. Nat Genet 2011; 43: 1022–1025. patients with malignant pleural mesothelioma: combined analysis of two phase II
3. Maskell NA, Gleeson FV, Davies RJ. Standard pleural biopsy versus CT-guided trials. Br J Cancer 2008; 99: 51–56.
cutting-needle biopsy for diagnosis of malignant disease in pleural effusions: a 24. Kindler HL, Karrison TG, Gandara DR et al. Multicenter, double-blind, placebo-
randomised controlled trial. Lancet 2003; 361: 1326–1330. controlled, randomized phase II trial of gemcitabine/cisplatin plus bevacizumab or
4. Greillier L, Cavailles A, Fraticelli A et al. Accuracy of pleural biopsy using placebo in patients with malignant mesothelioma. J Clin Oncol 2012; 30:
thoracoscopy for the diagnosis of histologic subtype in patients with malignant 2509–2515.
pleural mesothelioma. Cancer 2007; 110: 2248–2252. 25. Nowak AK, Millward MJ, Creaney J et al. A phase II trial of intermittent sunitinib
5. Greillier L, Baas P, Welch JJ et al. Biomarkers for malignant pleural mesothelioma: maleate as second-line therapy in progressive malignant pleural mesothelioma. J
current status. Mol Diagn Ther 2008; 12: 375–390. Thorac Oncol 2012; 7: 1449–1456.
6. Creaney J, Dick IM, Meniawy TM et al. Comparison of fibulin-3 and mesothelin as 26. Buikhuisen WA, Burgers JA, Vincent AD et al. Thalidomide versus active
markers in malignant mesothelioma. Thorax 2014; 69: 895–902. supportive care for maintenance in patients with malignant mesothelioma after
7. van den Heuvel MM, Korse CM, Bonfrer JM, Baas P. Non-invasive diagnosis of first-line chemotherapy (NVALT 5): an open-label, multicentre, randomised phase 3
pleural malignancies: the role of tumour markers. Lung Cancer 2008; 59: 350–354. study. Lancet Oncol 2013; 14: 543–551.
8. Churg A, Roggli VL, Galateau-Salle F et al. Tumours of the pleura: mesothelial 27. Manegold C, Symanowski J, Gatzemeier U et al. Second-line ( post-study)
tumours. In: Travis WD, Brambilla E, Muller-Hermelink HK, Harris CC (eds): chemotherapy received by patients treated in the phase III trial of pemetrexed plus
Pathology and Genetics of Tumours of the Lung, Pleura, Thymus and Heart. Lyon, cisplatin versus cisplatin alone in malignant pleural mesothelioma. Ann Oncol
France: IARC, 2004, World Health Organization Classification of Tumours 10: 2005; 16: 923–927.
128–136. 28. Stebbing J, Powles T, McPherson K et al. The efficacy and safety of weekly
9. Husain AN, Colby T, Ordonez N et al. Guidelines for pathologic diagnosis of vinorelbine in relapsed malignant pleural mesothelioma. Lung Cancer 2009; 63:
malignant mesothelioma: 2012 update of the consensus statement from the 94–97.
International Mesothelioma Interest Group. Arch Pathol Lab Med 2013; 137: 29. Zauderer MG, Kass SL, Woo K et al. Vinorelbine and gemcitabine as second- or
647–667. third-line therapy for malignant pleural mesothelioma. Lung Cancer 2014; 84:
10. Henderson DW, Reid G, Kao SC et al. Challenges and controversies in the 271–274.
diagnosis of mesothelioma: Part 1. Cytology-only diagnosis, biopsies, 30. Muers MF, Stephens RJ, Fisher P et al. Active symptom control with or without
immunohistochemistry, discrimination between mesothelioma and reactive chemotherapy in the treatment of patients with malignant pleural mesothelioma
mesothelial hyperplasia, and biomarkers. J Clin Pathol 2013; 66: 847–853. (MS01): a multicentre randomised trial. Lancet 2008; 371: 1685–1694.
11. King J, Thatcher N, Pickering C, Hasleton P. Sensitivity and specificity of 31. Hassan R, Miller AC, Sharon E et al. Major cancer regressions in mesothelioma
immunohistochemical antibodies used to distinguish between benign and after treatment with an anti-mesothelin immunotoxin and immune suppression.
malignant pleural disease: a systematic review of published reports. Histopathology Sci Transl Med 2013; 5: 208ra147.
2006; 49: 561–568. 32. Calabro L, Morra A, Fonsatti E et al. Tremelimumab for patients with
12. Chiosea S, Krasinskas A, Cagle PT et al. Diagnostic importance of 9p21 chemotherapy-resistant advanced malignant mesothelioma: an open-label, single-
homozygous deletion in malignant mesotheliomas. Mod Pathol 2008; 21: 742–747. arm, phase 2 trial. Lancet Oncol 2013; 14: 1104–1111.
13. Butchart EG, Gibbs AR. Pleural mesothelioma. Curr Opin Oncol 1990; 2: 33. Blayney JK, Ceresoli GL, Castagneto B et al. Response to chemotherapy is
352–358. predictive in relation to longer overall survival in an individual patient combined-
14. Rusch VW. A proposed new international TNM staging system for malignant pleural analysis with pleural mesothelioma. Eur J Cancer 2012; 48: 2983–2992.
mesothelioma. From the International Mesothelioma Interest Group. Chest 1995; 34. Hasan B, Greillier L, Pallis A et al. Progression free survival rate at 9 and 18 weeks
108: 1122–1128. predict overall survival in patients with malignant pleural mesothelioma: an
v | Baas et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
individual patient pooled analysis of 10 European Organisation for Research and 47. Rintoul RC, Ritchie AJ, Edwards JG et al. Efficacy and cost of video-assisted
Treatment of Cancer Lung Cancer Group studies and an independent study thoracoscopic partial pleurectomy versus talc pleurodesis in patients with
validation. Eur J Cancer 2014; 50: 2771–2782. malignant pleural mesothelioma (MesoVATS): an open-label, randomised controlled
35. Poulikakos PI, Xiao GH, Gallagher R et al. Re-expression of the tumor suppressor trial. Lancet 2014; 384: 1118–1127.
NF2/merlin inhibits invasiveness in mesothelioma cells and negatively regulates 48. Rice D, Rusch V, Pass H et al. Recommendations for uniform definitions of surgical
FAK. Oncogene 2006; 25: 5960–5968. techniques for malignant pleural mesothelioma: a consensus report of the
36. Shapiro IM, Kolev VN, Vidal CM et al. Merlin deficiency predicts FAK international association for the study of lung cancer international staging
inhibitor sensitivity: a synthetic lethal relationship. Sci Transl Med 2014; 6: committee and the international mesothelioma interest group. J Thorac Oncol
237ra68. 2011; 6: 1304–1312.
37. MacLeod N, Chalmers A, O’Rourke N et al. Is radiotherapy useful for treating pain 49. Flores RM, Pass HI, Seshan VE et al. Extrapleural pneumonectomy versus
in mesothelioma? A phase II trial. J Thorac Oncol 2015; 10: 944–950. pleurectomy/decortication in the surgical management of malignant pleural
38. Macleod N, Price A, O’Rourke N et al. Radiotherapy for the treatment of pain in mesothelioma: results in 663 patients. J Thorac Cardiovasc Surg 2008; 135:
malignant pleural mesothelioma: a systematic review. Lung Cancer 2014; 83: 620–626.
133–138. 50. Weder W, Stahel RA, Bernhard J et al. Multicenter trial of neo-adjuvant
39. Davies HE, Musk AW, Lee YC. Prophylactic radiotherapy for pleural puncture sites chemotherapy followed by extrapleural pneumonectomy in malignant pleural
in mesothelioma: the controversy continues. Curr Opin Pulm Med 2008; 14: mesothelioma. Ann Oncol 2007; 18: 1196–1202.
326–330. 51. de Perrot M, Feld R, Cho BC et al. Trimodality therapy with induction chemotherapy
40. O’Rourke N, Garcia JC, Paul J et al. A randomised controlled trial of intervention followed by extrapleural pneumonectomy and adjuvant high-dose hemithoracic
site radiotherapy in malignant pleural mesothelioma. Radiother Oncol 2007; 84: radiation for malignant pleural mesothelioma. J Clin Oncol 2009; 27: 1413–1418.
18–22. 52. Van Schil PE, Baas P, Gafaar R et al. Trimodality therapy for malignant pleural
41. Cho BC, Feld R, Leighl N et al. A feasibility study evaluating surgery for mesothelioma: results from an EORTC phase II multicentre trial. Eur Resp J 2010;
mesothelioma after radiation therapy: the “SMART” approach for resectable 36: 1362–1369.
malignant pleural mesothelioma. J Thorac Oncol 2014; 9: 397–402. 53. Krug LM, Pass HI, Rusch VW et al. Multicenter phase II trial of neoadjuvant
42. Rusch VW, Rosenzweig K, Venkatraman E et al. A phase II trial of surgical pemetrexed plus cisplatin followed by extrapleural pneumonectomy and radiation
resection and adjuvant high-dose hemithoracic radiation for malignant pleural for malignant pleural mesothelioma. J Clin Oncol 2009; 27: 3007–3013.
mesothelioma. J Thorac Cardiovasc Surg 2001; 122: 788–795. 54. Treasure T, Lang-Lazdunski L, Waller D et al. Extra-pleural pneumonectomy versus
43. Baldini EH, Recht A, Strauss GM et al. Patterns of failure after trimodality therapy no extra-pleural pneumonectomy for patients with malignant pleural
for malignant pleural mesothelioma. Ann Thorac Surg 1997; 63: 334–338. mesothelioma: clinical outcomes of the Mesothelioma and Radical Surgery (MARS)
44. Okubo K, Sonobe M, Fujinaga T et al. Survival and relapse pattern after trimodality randomised feasibility study. Lancet Oncol 2011; 12: 763–772.
therapy for malignant pleural mesothelioma. Gen Thorac Cardiovasc Surg 2009; 55. Cao CQ, Yan TD, Bannon PG, McCaughan BC. A systematic review of extrapleural
57: 585–590. pneumonectomy for malignant pleural mesothelioma. J Thorac Oncol 2010; 5:
45. Gomez DR, Hong DS, Allen PK et al. Patterns of failure, toxicity, and survival after 1692–1703.
extrapleural pneumonectomy and hemithoracic intensity-modulated radiation 56. Cao C, Tian D, Manganas C et al. Systematic review of trimodality therapy for patients
therapy for malignant pleural mesothelioma. J Thorac Oncol 2013; 8: 238–245. with malignant pleural mesothelioma. Ann Cardiothorac Surg 2012; 1: 428–437.
46. Allen AM, Czerminska M, Jänne PA et al. Fatal pneumonitis associated with 57. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
intensity-modulated radiation therapy for mesothelioma. Int J Radiat Oncol Biol among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
Phys 2006; 65: 640–645. 139–144.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv199 | v

by guest
on 05 February 2018
Primary breast cancer: ESMO Clinical Practice

E. Senkus1, S. Kyriakides2, S. Ohno3, F. Penault-Llorca4,5, P. Poortmans6, E. Rutgers7,
S. Zackrisson8 & F. Cardoso9, on behalf of the ESMO Guidelines Committee*
1
Department of Oncology and Radiotherapy, Medical University of Gdańsk, Gdańsk, Poland; 2Europa Donna Cyprus, Nicosia, Cyprus; 3Breast Oncology Center, Cancer
Institute Hospital, Tokyo, Japan; 4Department of Pathology, Centre Jean Perrin, Clermont-Ferrand; 5EA 4677 Université d’Auvergne, Clermont-Ferrand, France; 6Radboud
University Medical Center, Nijmegen, The Netherlands; 7Department of Surgery, Netherlands Cancer Institute, Amsterdam, The Netherlands; 8Department of Diagnostic
Radiology, Lund University, Malmö, Sweden; 9Breast Unit, Champalimaud Clinical Center, Lisbon, Portugal
incidence and epidemiology mortality reduction benefit in the age group of 50–69 years and is
recommended by the European Union and numerous individual
In 2012, the estimated age-adjusted annual incidence of breast countries [8]. The evidence for effectiveness of mammography
cancer in 40 European countries was 94.2/100 000 and the mor- screening in women aged 40–49 years is limited [9]. This was also
tality 23.1/100 000 [1]. The incidence increased after the intro- the conclusion in the recent breast cancer screening report from
duction of mammography screening, and continues to grow the International Agency for Research on Cancer [10]. There is no
with the ageing of the population. The most important risk consensus about the exact effect of mammography screening on
factors include: genetic predisposition, exposure to oestrogens breast cancer mortality reduction, as the reported estimates vary.
(endogenous and exogenous), ionising radiation, low parity and In a UK review of the randomised, controlled mammography
a history of atypical hyperplasia. The Western-style diet, obesity trials, a 20% relative breast cancer mortality reduction was esti-
clinical practice
and the consumption of alcohol also contribute to the rising in- mated in women aged between 50 and 70 years old [11]. It must
guidelines
cidence of breast cancer [2]. There is a steep age gradient, with be noted that the review stresses the importance of taking into
about a quarter of breast cancers occurring before age 50, and account the risk of over-diagnosis and over-treatment, as well as
<5% before age 35. The estimated 5-year prevalence of breast false-positive screening, when balancing the benefits and harms of
cancer in Europe in 2012 was 1 814 572 cases [1]. Prevalence is screening. Screening programmes carry the risk of false-negative
increasing, as a consequence of increased incidence and due to results, consequently a false feeling of security among patients and
improvements in treatment outcomes. In most Western coun- doctors may be instilled. Nevertheless, mammography screening
tries, the mortality rate has decreased in recent years, especially and population-based awareness programmes, together with im-
in younger age groups, because of improved treatment and proved treatment, may contribute to mortality reduction in breast
earlier detection [3, 4]. However, breast cancer is still the leading cancer. Therefore, we recommend (after a discussion of the bene-
cause of cancer-related deaths in European women. fits and risks with the woman who is to be screened) regular mam-
Breast cancer in males is rare, contributing to ∼1% of cases. mography in women aged 50–69 years [I, A]. There is controversy
The major risk factors include clinical disorders carrying hor- and no consensus regarding the role of screening in women aged
monal imbalances (especially gynaecomastia and cirrhosis), 40–49 years, or for the use of ultrasound.
radiation exposure and, in particular, a positive family history In women with familial breast cancer, with or without proven
and genetic predisposition [5]. BRCA mutations, annual screening with magnetic resonance
imaging (MRI) of the breast, in combination with mammography,
breast cancer screening can detect the disease at a more favourable stage compared with
mammography screening alone (70% lower risk of being diagnosed
Eighteen European countries have established national or re- with breast cancer stage II or higher). However, it is not known
gional population-based mammography screening programmes, whether breast cancer mortality is lowered [12]. We recommend
to detect breast cancers at a pre-clinical stage [6]. The European annual MRI concomitantly or alternating every 6 months with
Guidelines for quality assurance in breast cancer screening and mammography, starting 10 years younger than the youngest case in
diagnosis recommend performance parameters and indicators the family [III,A]. There is no consensus for the use of ultrasound.
that should be monitored in any screening programme [7].
Mammography screening, every 2 years, has shown the greatest
biology
CH-6962 Viganello-Lugano, Switzerland. The diagnosis of breast cancer is based on clinical examination
†
in combination with imaging, and confirmed by pathological as-
Approved by the ESMO Guidelines Committee: August 2003, last update July 2015.
This publication supersedes the previously published version – Ann Oncol 2013; 24 sessment (Table 1). Clinical examination includes bimanual pal-
(Suppl. 6): vi7–vi23. pation of the breasts and locoregional lymph nodes and

by guest
on 05 February 2018
Table 1. Diagnostic workup for early breast cancer
Assessment of general health status History

Menopausal status
Physical examination
Full blood count
Liver, renal and cardiac (in patients planned for anthracycline and/or trastuzumab treatment)
function tests, alkaline phosphatase and calcium
Assessment of primary tumour Physical examination
Mammography
Breast ultrasound
Breast MRI
Core biopsy with pathology determination of histology, grade, ER, PgR, HER-2 and Ki67
Assessment of regional lymph nodes Physical examination
Ultrasound
Ultrasound-guided biopsy if suspicious
Assessment of metastatic disease Physical examination
Other tests are not routinely recommended, unless locally advanced or when symptoms suggestive of
metastases are present
MRI, magnetic resonance imaging; ER, oestrogen receptor; PgR, progesterone receptor; HER2, human epidermal growth factor 2 receptor.
assessment for distant metastases (bones, liver and lungs; a aspiration indicating carcinoma) must be obtained before any type
neurological examination is only required when symptoms are of treatment is initiated. If preoperative systemic therapy is
present). Imaging includes bilateral mammography and ultra- planned, a core needle biopsy is mandatory to ensure a diagnosis
sound of the breast and regional lymph nodes [7]. An MRI of the of invasive disease and assess biomarkers [III, A]. A marker (e.g.
breast is not routinely recommended, but should be considered surgical clip, carbon) should be placed into the tumour at biopsy,
in cases of: familial breast cancer associated with BRCA muta- to ensure surgical resection of the correct site [V, A]. As a
tions, breast implants, lobular cancers, suspicion of multifocality/ minimum, ultrasound-guided fine needle aspiration or core
multicentricity (particularly in lobular breast cancer), or large biopsy of suspicious lymph nodes should be carried out [III, A].
discrepancies between conventional imaging and clinical examin- In patients with clinically and imaging negative axilla, the best
ation [III, B]. MRI may also be recommended before neoadjuvant timing to carry out sentinel lymph node biopsy (SLNB), i.e. before
chemotherapy, when evaluating the response to primary systemic or after preoperative systemic therapy, remains controversial [II,
therapy or when the findings of conventional imaging are incon- C]. The recently published SENTINA and ACOSOG Z1071
clusive (such as a positive axillary lymph node status with an studies demonstrated lower detection rates and higher rates of
occult primary tumour in the breast) [III, A] [13]. Several new false-negatives when SLNB is carried out after systemic therapy,
techniques are being tested for screening and diagnostic imaging, compared with SNLB that is carried out before neoadjuvant
such as: 3D mammography (breast tomosynthesis), 3D ultra- chemotherapy [14, 15]. However, if the axilla is negative on ultra-
sound, shear wave elastography and contrast-enhanced mam- sound and/or positron emission tomography (PET)/computed
mography/spectral mammography. None of these are routinely tomography (CT) scanning, carried out before the start of systemic
implemented as yet, but they have the potential to increase diag- therapy, a post-systemic therapy SNLB can be considered [V, B].
nostic accuracy, especially in women with dense breasts. Final pathological diagnosis should be made according to
Apart from imaging, pre-treatment disease evaluation includes the World Health Organization (WHO) classification [16]
pathological examination of the primary tumour and cytology/ and the tumour–node–metastases (TNM) staging system. The
histology of the axillary nodes, if involvement is suspected. Other pathological report should include the histological type, grade,
assessments include: complete personal medical history, family immunohistochemical (IHC) evaluation of oestrogen receptor
history relating to breast/ovarian and other cancers, physical (ER) status (using a standardised assessment methodology, e.g.
examination, a full blood count, liver and renal function tests, al- Allred or H-score) and, for invasive cancer, IHC evaluation of
kaline phosphatase and calcium levels. Assessing the menopausal progesterone receptor (PgR) and human epidermal growth
status of the patient is imperative, if in doubt measure serum factor 2 receptor (HER2) gene expression. HER2 gene amplifica-
oestradiol and follicle-stimulating hormone levels. In patients tion status may be determined directly from all invasive
planned for (neo)adjuvant treatment, with anthracyclines and/or tumours using in situ hybridisation (fluorescent, chromogenic
trastuzumab, evaluation of cardiac function with a cardiac ultra- or silver), replacing IHC or only for tumours with an ambiguous
sound or a multigated acquisition scan is essential [I, A]. (2+) IHC score [II, B] [17]. The guidelines for HER2 testing
Pathological diagnosis should be based on a core needle biopsy, have recently been updated by the American Society of Clinical
obtained preferably by ultrasound or stereotactic guidance. A core Oncology–College of American Pathologists (ASCO-CAP)
needle biopsy (if this is not possible, at least a fine needle group. There is a change in the definition of HER2 positivity by
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv298 | v

by guest
on 05 February 2018
IHC (3+ when more than 10% of the cells, instead of 30%, Asymptomatic distant metastases are very rare and patients do
harbour a complete membrane staining), and by in situ hybrid- not benefit from comprehensive laboratory (including tumour
isation ( positive if the number of HER2 gene copies is ≥6 or the markers [25]) and radiological staging [III, D]. Minimum blood
ratio HER2/chromosome 17 is ≥2, instead of 2.2). The defin- workup (a full blood count, liver and renal function tests, alka-
ition of equivocal cases is broader; if a case is defined as equivo- line phosphatase and calcium levels) is recommended before
cal after two tests it is eligible for trastuzumab, and should be surgery and systemic (neo)adjuvant therapy [V, B]. A CT scan
discussed in multidisciplinary tumour boards [V, B] [18]. of the chest, an abdominal ultrasound or CT scan and a bone
Proliferation markers such as the Ki67 labelling index may scan can be considered for patients with: clinically positive axil-
supply additional useful information, particularly if the assay lary nodes, large tumours (e.g. ≥5 cm), aggressive biology or
can be standardised [V, A] [19, 20]. Alternatively, these bio- clinical signs, symptoms or laboratory values suggesting the
logical markers can be assessed in the definitive surgical speci- presence of metastases [III, B]. Dual imaging methods combin-
men if primary systemic therapy is not planned. However, ing functional and anatomical information such as fluorodeoxy-
fixation is better controlled for core biopsies, allowing safer glucose positron emission tomography (FDG-PET)/CT may be
antigen preservation for IHC [21]. In case of negativity of ER/ useful when conventional methods are inconclusive [V, A].
PgR and HER2 in the biopsy specimen, it is advisable to retest PET/CT scanning can replace traditional imaging for staging in
for them in the surgical specimen to account for the putative high-risk patients who are candidates for neoadjuvant chemo-
tumour heterogeneity [III, A] [22]. therapy, as well as those with locally advanced and/or inflamma-
For the purpose of prognostication and treatment decision tory disease due to their high risk of having metastatic disease
making, tumours should be grouped into surrogate intrinsic [V, B] [26]. Current evidence does not support the use of FDG-
subtypes, defined by routine histology and IHC data [III, A] PET/CT in the staging of local/regional disease, due to its
(Table 2) [23]. limited specificity when compared with the gold standard,
SLNB and axillary lymph node dissection [27].
The postoperative pathological assessment of the surgical
staging and risk assessment specimens should be made according to the pathological TNM
Disease stage should be assessed according to the TNM system system (Tables 3 and 4). This assessment should include: the
(Tables 3 and 4). In early breast cancer, routine staging evalua- number, location and maximum diameter of the tumours
tions are directed at locoregional disease (Figure 1). removed, the total number of removed and positive lymph
Table 2. Surrogate definitions of intrinsic subtypes of breast cancer according to the 2015 St Gallen Consensus Conference [23] and also
recommended by the ESMO Clinical Practice Guidelines
Intrinsic subtype Clinicopathologic surrogate definition Notes
Luminal A ‘Luminal A-like’ *Ki-67 scores should be interpreted in the light of local laboratory values: as an
ER-positive example, if a laboratory has a median Ki-67 score in receptor-positive disease of
HER2-negative 20%, values of 30% or above could be considered clearly high; those of 10% or
Ki67 low* less clearly low,
PgR high** **Suggested cut-off value is 20%; quality assurance programmes are essential for
low-risk molecular signature (if available) laboratories reporting these results.
Luminal B ‘Luminal B-like (HER2-negative)’
ER-positive
HER2-negative
and either
Ki67 high or
PgR low
high-risk molecular signature (if
available)
‘Luminal B-like (HER2-positive)’
ER-positive
HER2-positive
any Ki67
any PgR
HER2 overexpression ‘HER2-positive (non-luminal)’
HER2-positive
ER and PgR absent
‘Basal-like’ ‘Triple-negative (ductal)’ There is ∼80% overlap between ‘triple-negative’ and intrinsic ‘basal-like’ subtype,
ER and PgR absent but ‘triple-negative’ also includes some special histological types such as (typical)
HER2-negative medullary and adenoid cystic carcinoma with low risks of distant recurrence.
ER, oestrogen receptor; HER2, human epidermal growth factor 2 receptor; PgR, progesterone receptor.
v | Senkus et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Table 3. Tumour–node–metastases (TNM) staging system for carcinoma of the breast
Primary tumour (T)a,b,c,d

Tis (DCIS) Ductal carcinoma in situ
Tis (LCIS) Lobular carcinoma in situ
Tis (Paget’s) Paget’s disease (Paget disease) of the nipple NOT associated with invasive carcinoma and/or carcinoma in situ (DCIS and/or LCIS) in
the underlying breast parenchyma. Carcinomas in the breast parenchyma associated with Paget’s disease are categorised based on
the size and characteristics of the parenchymal disease, although the presence of Paget’s disease should still be noted.
T1 Tumour ≤20 mm in greatest dimension
T1mi Tumour ≤1 mm in greatest dimension
T1a Tumour >1 mm but ≤5 mm in greatest dimension
T1b Tumour >5 mm but ≤10 mm in greatest dimension
T1c Tumour >10 mm but ≤20 mm in greatest dimension
T2 Tumour >20 mm but ≤50 mm in greatest dimension
T3 Tumour >50 mm in greatest dimension
T4 Tumour of any size with direct extension to the chest wall and/or to the skin (ulceration or skin nodules)e
T4a Extension to the chest wall, not including only pectoralis muscle adherence/invasion
T4b Ulceration and/or ipsilateral satellite nodules and/or oedema (including peau d’orange) of the skin, which do not meet the criteria for
inflammatory carcinoma
T4c Both T4a and T4b
T4d Inflammatory carcinomaf
Clinical (cN)g,h,i,j,k
NX Regional lymph nodes cannot be assessed (e.g. previously removed)
N1 Metastases to movable ipsilateral level I, II axillary lymph node(s)
N2 Metastases in ipsilateral level I, II axillary lymph nodes that are clinically fixed or matted; or in clinically detectedk ipsilateral internal
mammary nodes in the absence of clinically evident axillary lymph node metastases
N2a Metastases in ipsilateral level I, II axillary lymph nodes fixed to one another (matted) or to other structures
N2b Metastases only in clinically detectedk ipsilateral internal mammary nodes and in the absence of clinically evident level I, II axillary
lymph node metastases
N3 Metastases in ipsilateral infraclavicular (level III axillary) lymph node(s) with or without level I, II axillary lymph node involvement;
or in clinically detectedk ipsilateral internal mammary lymph node(s) with clinically evident level I, II axillary lymph node
metastases; or metastases in ipsilateral supraclavicular lymph node(s) with or without axillary or internal mammary lymph node
involvement
N3a Metastases in ipsilateral infraclavicular lymph node(s)
N3b Metastases in ipsilateral internal mammary lymph node(s) and axillary lymph node(s)
N3c Metastases in ipsilateral supraclavicular lymph node(s)
Pathological (pN)h,i,j,k
pNX Regional lymph nodes cannot be assessed (e.g. previously removed or not removed for pathological study)
pN0 No regional lymph node metastasis identified histologically
pN0(i−) No regional lymph node metastases histologically, negative immunohistochemistry (IHC)
pN0(i+) Malignant cells in regional lymph node(s) not >0.2 mm [detected by haematoxylin and eosin (H&E) staining or IHC including
isolated tumour cell clusters (ITCs)]
pN0(mol−) No regional lymph node metastases histologically, negative molecular findings (RT-PCR)
pN0(mol+) Positive molecular findings (RT-PCR), but no regional lymph node metastases detected by histology or IHC
pN1 Micrometastases; or metastases in one to three axillary lymph nodes; and/or in internal mammary nodes with metastases detected by
SLNB but not clinically detectedl
pN1mi Micrometastases (>0.2 mm and/or >200 cells, but none >2.0 mm)
pN1a Metastases in one to three axillary lymph nodes, at least one metastasis >2.0 mm
pN1b Metastases in internal mammary nodes with micrometastases or macrometastases detected by SLNB but not clinically detectedl
pN1c Metastases in one to three axillary lymph nodes and in internal mammary lymph nodes with micrometastases or macrometastases
detected by SLNB but not clinically detectedl
pN2 Metastases in four to nine axillary lymph nodes; or in clinically detectedk internal mammary lymph nodes in the absence of axillary
lymph node metastases
pN2a Metastases in four to nine axillary lymph nodes (at least one tumour deposit >2.0 mm)
Continued
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv298 | v

by guest
on 05 February 2018
Table 3. Continued
pN2b Metastases in clinically detectedk internal mammary lymph nodes in the absence of axillary lymph node metastases
pN3 Metastases in ≥10 axillary lymph nodes; or in infraclavicular (level III axillary) lymph nodes; or in clinically detectedk ipsilateral
internal mammary lymph nodes in the presence of one or more positive level I, II axillary lymph nodes; or in more than three
axillary lymph nodes and in internal mammary lymph nodes with micrometastases or macrometastases detected by SLNB but not
clinically detectedl; or in ipsilateral supraclavicular lymph nodes
pN3a Metastases in ≥10 axillary lymph nodes (at least one tumour deposit >2.0 mm); or metastases to the infraclavicular (level III axillary
lymph) nodes
pN3b Metastases in clinically detectedk ipsilateral internal mammary lymph nodes in the presence of one or more positive axillary lymph
nodes; or in more than three axillary lymph nodes and in internal mammary lymph nodes with micrometastases or
macrometastases detected by SLNB but not clinically detectedl
pN3c Metastases in ipsilateral supraclavicular lymph nodes
M0 No clinical or radiographic evidence of distant metastases

cM0(i+) No clinical or radiographic evidence of distant metastases, but deposits of molecularly or microscopically detected tumour cells in
circulating blood, bone marrow or other non-regional nodal tissue that are not >0.2 mm in a patient without symptoms or signs of
metastases
M1 Distant detectable metastases as determined by classic clinical and radiographic means and/or histologically proven >0.2 mm
a
DCIS; LCIS. Post-treatment ypT: The use of neoadjuvant therapy does not change the clinical (pre-treatment) stage. Clinical (pre-treatment) T will be
defined by clinical and radiographic findings, while y pathological (post-treatment) T will be determined by pathological size and extension. The ypT will
be measured as the largest single focus of invasive tumour, with the modifier ‘m’ indicating multiple foci. The measurement of the largest tumour focus
should not include areas of fibrosis within the tumour bed.
b
The T classification of the primary tumour is the same regardless of whether it is based on clinical or pathological criteria, or both. Designation should be
made with the subscript ‘c’ or ‘p’ modifier to indicate whether the T classification was determined by clinical (physical examination or radiological) or
pathological measurements, respectively. In general, pathological determination should take precedence over clinical determination of T size.
c
Size should be measured to the nearest millimetre.
d
Multiple simultaneous ipsilateral primary carcinomas are defined as infiltrating carcinomas in the same breast, which are grossly or macroscopically distinct
and measurable. T stage is based only on the largest tumour. The presence and sizes of the smaller tumour(s) should be recorded using the ‘(m)’ modifier.
e
Invasion of the dermis alone does not qualify as T4; dimpling of the skin, nipple retraction or any other skin change except those described under T4b and T4d
may occur in T1, T2 or T3 without changing the classification. The chest wall includes ribs, intercostal muscles and serratus anterior muscle, but not the
pectoralis muscles.
f
Inflammatory carcinoma is a clinical–pathological entity characterised by diffuse erythema and oedema (peau d’orange) involving a third or more of the skin of
the breast. These skin changes are due to lymphoedema caused by tumour emboli within dermal lymphatics. Although dermal lymphatic involvement supports
the diagnosis of inflammatory breast cancer, it is neither necessary nor sufficient, in the absence of classical clinical findings, for the diagnosis of inflammatory
breast cancer.
g
Classification is based on axillary lymph node dissection with or without SLNB. Classification based solely on SLNB without subsequent axillary lymph node
dissection is designated (sn) for ‘sentinel node’, e.g. pN0(sn).
h
Isolated tumour cell clusters (ITCs) are defined as small clusters of cells not >0.2 mm, or single tumour cells, or a cluster of <200 cells in a single histological
cross section. ITCs may be detected by routine histology or by immunohistochemical (IHC) methods. Nodes containing only ITCs are excluded from the total
positive node count for purposes of N classification but should be included in the total number of nodes evaluated.
i
Post-treatment yp ‘N’ should be evaluated as for pre-treatment ‘N’. The modifier ‘sn’ is used if a sentinel node evaluation was carried out. If no subscript is
attached, it is assumed that the axillary nodal evaluation was by axillary node dissection.
j
ypN categories are the same as those for pN.
k
Clinically detected is defined as detected by imaging studies (excluding lymphoscintigraphy) or by clinical examination and having characteristics highly
suspicious for malignancy or a presumed pathological macrometastasis based on fine needle aspiration biopsy with cytological examination. Confirmation of
clinically detected metastatic disease by fine needle aspiration without excision biopsy is designated with an (f) suffix, e.g. cN3a(f). Excisional biopsy of a lymph
node or biopsy of a sentinel node, in the absence of assignment of a pT, is classified as a clinical N, e.g. cN1. Information regarding the confirmation of the nodal
status will be designated in site-specific factors as clinical, fine needle aspiration, core biopsy or sentinel lymph node biopsy. Pathological classification (pN) is
used for excision or SLNB only in conjunction with a pathological T assignment.
l
‘Not clinically detected’ is defined as not detected by imaging studies (excluding lymphoscintigraphy) or not detected by clinical examination.
From [24]. Used with the permission of the American Joint Committee on Cancer (AJCC), Chicago, IL, USA. The original source for this material is the AJCC
Cancer Staging Handbook, Seventh Edition (2010) published by Springer Science and Business Media LLC, www.springer.com.
SLNB, sentinel lymph node biopsy; RT-PCR, reverse transcription polymerase chain reaction; DCIS, ductal carcinoma in situ; LCIS, called lobular carcinoma
in situ.
v | Senkus et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Table 4. Stage grouping system for carcinoma of the breast number of involved regional lymph nodes, tumour histology,
a the size, grade and the presence of peri-tumoural vascular inva-
sion. Additionally, in patients undergoing breast-conserving
Stage 0 therapy (BCT), the ipsilateral breast recurrence risk is related to
Tis N0 M0 the status of the surgical margins.
Stage IA Clinical parameters (age, tumour stage, ER expression and
T1b N0 M0 histological grade) have been integrated into scoring systems
Stage IB allowing a relatively accurate estimation of the probability of re-
T0 N1mi M0
currence and death from breast cancer; examples include the
T1b N1mi M0
Nottingham Prognostic Index (NPI), Adjuvant! Online (www.
Stage IIA
adjuvantonline.com) or the PREDICT score [28–30]. Urokinase
T0 N1c M0
plasminogen activator–plasminogen activator inhibitor 1 (uPA-
T1b N1c M0
PAI1) (FEMTELLE, Sekisui Diagnostics) is an Elisa test evaluat-
T2 N0 M0
Stage IIB
ing the metastatic potential of a breast tumour. Despite its level
T2 N1 M0 IA prognostic value, this test is not extensively used outside
T3 N0 M0 Germany, probably due to the requirement for a substantial
Stage IIIA amount of fresh-frozen tissue [31].
T0 N2 M0 Gene expression profiles, such as MammaPrint (Agendia,
T1b N2 M0 Amsterdam, the Netherlands), Oncotype DX Recurrence Score
T2 N2 M0 (Genomic Health, Redwood City, CA), Prosigna (Nanostring
T3 N1 M0 technologies, Seattle, WA) and Endopredict (Myriad Genetics),
T3 N2 M0 may be used to gain additional prognostic and/or predictive in-
Stage IIIB formation to complement pathology assessment and to predict
T4 N0 M0 the benefit of adjuvant chemotherapy. The three latter tests are
T4 N1 M0 designed for patients with ER-positive early breast cancer only.
T4 N2 M0 The clinical utility of Mammaprint and Oncotype DX is still
Stage IIIC being prospectively evaluated in large randomised clinical trials
Any T N3 M0 such as MINDACT for Mammaprint, WSG PLAN B trial,
Stage IV TAILORx and RxPONDER for Oncotype DX. A IB evidence
Any T Any N M1 level has been achieved from retrospective analyses of data from
a
prospective trials regarding the prognostic value of MammaPrint,
Anatomic stage: M0 includes M0(i+). The designation pM0 is not valid;
Oncotype DX, Prosigna and Endopredict, in ER-positive breast
any M0 should be clinical. If a patient presents with M1 before
cancers [31, 32]. In addition, the prognostic value of MammaPrint
neoadjuvant systemic therapy, the stage is considered stage IV and
has been validated in the Raster trial, a prospective but non-
remains stage IV regardless of response to neoadjuvant therapy. Stage
designation may be changed if postsurgical imaging studies reveal the
randomised, clinical trial [33].
presence of distant metastases, provided that the studies are carried out ER/PgR and HER2 are the only validated predictive factors,
within 4 months of diagnosis in the absence of disease progression and allowing the selection of patients for endocrine therapies (ETs)
provided that the patient has not received neoadjuvant therapy. Post- and anti-HER2 treatments, respectively. High ER expression is
neoadjuvant assessment is designated with a ‘yc’ or ‘yp’ prefix. Of note, usually associated with lesser absolute benefit of chemotherapy.
no stage group is assigned if there is a complete pathological response After neoadjuvant systemic treatment, the response to treat-
(pCR) to neoadjuvant therapy, e.g. ypT0ypN0cM0. ment and the amount of residual disease are important prognos-
b
T1 includes T1mi. tic factors, but need as much standardisation as any of the other
c
T0 and T1 tumours with nodal micrometastases only are excluded from biological markers. A multidisciplinary international working
stage IIA and are classified stage IB. group developed practical recommendations for the systematic,
From [24]. Used with the permission of the American Joint Committee standardised evaluation of the post-neoadjuvant surgical breast
on Cancer (AJCC), Chicago, Illinois. The original source for this material cancer specimen in clinical trials [34]. Systematic sampling of
is the AJCC Cancer Staging Handbook, 7th Edition (2010) published by areas identified by intelligent mapping of the specimen and
Springer Science and Business Media LLC, www.springer.com. close correlation with radiological findings is preferable to
overly exhaustive sampling, and permits the collection of tissue
samples for translational research. If a pathologic complete re-
nodes, as well as the extent of metastases in the lymph nodes sponse ( pCR) was achieved, both in the breast and axilla, this
[isolated tumour cells, micrometastases (0.2–2 mm), macrome- must be clearly stated. In addition, the presence or absence of re-
tastases]. The report should also include: the histological type sidual ductal carcinoma in situ (DCIS) must be described. In
and grade of the tumour(s) using a standard grading system, case of residual invasive carcinoma, a comment must be made
evaluation of the resection margins, including the location and as to the presence or absence of chemotherapy effect in the
minimum distance of the margin, vascular invasion, and a bio- breast and the lymph nodes. The Residual Cancer Burden is the
marker analysis, as described above [III, A]. preferred method for quantifying residual disease in clinical
The most important prognostic factors in early breast cancer trials; other methods can be used according to regional prefer-
are: expression of ER/PgR, HER2 and proliferation markers, the ence. Post-treatment tumour staging, using the TNM system,
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv298 | v

by guest
on 05 February 2018
Figure 1. Early breast cancer treatment algorithm. Cht, chemotherapy; BCS, breast-conserving surgery; ET, endocrine therapy; RT, radiotherapy.
should also be included. Some guidance is also provided by the Depending on the local situation and availability, other
United States Food and Drug Administration recommendation members of the breast team may include plastic/reconstructive
for accelerated drug approval in neoadjuvant treatment of breast surgeons, psychologists, physiotherapists and geneticists. Following
cancer [35]. a diagnosis of breast cancer, a patient finds herself in a new and
unfamiliar landscape. This creates different levels of stress that
vary from patient to patient, and needs to be addressed indi-
management of local/locoregional vidually and tailored to each patient’s needs. Most will remem-
ber the information provided to them in a fragmented way.
disease They will need space and time to process and comprehend their
Treatment should be carried out in ‘breast units’ defined as spe- diagnosis, so they can cope psychologically with the diagnosis
cialised institutions/departments that care for a high volume of and treatment plan. To accommodate for this, information on
breast cancer patients (a minimum of 150 early breast cancer diagnosis and treatment choice should be given repeatedly (both
cases per year). Treatment should be provided by a multidiscip- verbally and in writing) in a comprehensive and easily under-
linary team, which includes: at least one surgeon, radiation on- standable form. The use of reliable, patient-centred websites or
cologist, medical oncologist, radiologist and pathologist, who similar sources of information, is important and very useful.
are specialised in breast cancer [IV, A] [36–38]. A breast nurse, The choice of treatment strategy must be extensively dis-
or a similarly trained and specialised health care practitioner, cussed with the patient and take into account the patient’s pre-
acting as a patient navigator, is also important [III, B] [36, 37]. ferences. It should be based on the tumour burden/location
v | Senkus et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
(size and location of primary tumour, number of lesions, extent reconstruction in most women can make the prospect of losing
of lymph node involvement) and biology ( pathology, including a breast easier to accept, but not all women will be suitable for
biomarkers and gene expression), as well as the age and general immediate reconstruction. Some women may decline or defer
health status of the patient. Age should be taken into consider- reconstruction because of personal preference. Others will be
ation in conjunction with other factors and should not be the advised against immediate reconstruction for oncological
sole determinant for withholding or recommending a treatment. reasons, particularly in the case of inflammatory breast cancer.
Age is a continuous variable and its cut-offs in clinical trials are The autologous tissue-based techniques generally tolerate
always arbitrarily chosen. Overall, we strongly recommend that postoperative RT well. Implant-based reconstruction can result
‘younger’ patients should not be over-treated because they are in an unfavourable aesthetic outcome, following postoperative
‘young’, just as ‘older’ patients should not be under-treated, RT [49, 50]. Skin-sparing mastectomy allows the skin envelope
because they are deemed to be old. Patients should be actively to be conserved for use in the breast reconstruction. If post-
involved in all management decisions. The possibility of heredi- mastectomy radiotherapy (PMRT) is indicated, a temporary
tary cancer should be explored and, if needed, prophylactic pro- implant should be positioned before RT.
cedures discussed following appropriate genetic counselling and For women undergoing breast reconstruction, whether imme-
testing of the patient [IV, D] [39]. In younger premenopausal diate or delayed, a wide range of surgical options are available.
patients, possible fertility issues should be discussed and guid- The best technique for each patient should be discussed indi-
ance about fertility-preservation techniques should be provided, vidually taking into account anatomic, treatment and patient
before the initiation of treatment [40–43]. preference. Silicone gel implants are safe and acceptable compo-
nents of the reconstructive armamentarium [III, A]. Advances
local treatment in gel cross-linking have reduced silicone bleed, and cohesive gel
implants are likely to have fewer problems relating to capsular
surgery. Arguably the major change in the surgical treatment
rupture.
of primary breast cancer has been a shift towards breast-
Autologous tissue flaps can replace relatively large volumes of
conservation treatment, which started >30 years ago. Currently,
breast tissue. Tissue can be taken from the latissimus dorsi
in Western Europe, 60%–80% of newly diagnosed cancers are
muscle from the back, transverse rectus abdominis muscle, the
amenable to breast conservation [wide local excision and
free deep inferior epigastric perforator flap from the lower
radiation therapy (RT)]. In some patients, mastectomy is still
abdomen, superior gluteal artery-based perforator flap or free
carried out due to: tumour size (relative to breast size), tumour
gracilis-based flap.
multicentricity, inability to achieve negative surgical margins
There is no evidence that reconstruction makes detection of
after multiple resections, prior radiation to the chest wall/breast
local recurrence more difficult, and no basis for the outdated
or other contraindications to RT, or patient choice [44].
view that patients should wait 1–2 years after mastectomy before
being offered reconstruction.
breast-conservation surgery: For patients undergoing wide
local excision, greater emphasis is now placed on achieving advances in axillary staging: Regional lymph node status
acceptable cosmesis, and breast surgeons are trained to remains one of the strongest predictors of long-term prognosis
undertake oncoplastic approaches to reduce the impact of local in primary breast cancer. Axillary clearance is associated with
tumour excision on cosmesis, often using tissue displacement lymphoedema affecting the upper limb in up to 25% of women
techniques. Oncoplastic procedures can result in better cosmetic following surgery (up to 15% following axillary RT without
outcomes, especially in patients with: large breasts, a less surgical clearance and below 10% following SLNB) [51, 52]. The
favourable tumour/breast size ratio or a cosmetically challenging incidence of lymphoedema rises significantly (to 40%) when
(central or inferior) location of the tumour within the breast. A axillary clearance is combined with RT to the axilla. SLNB,
careful histological assessment of resection margins is essential. rather than full nodal clearance, is now accepted as the standard
Margin status should be reported according to the of care for axillary staging in early, clinically node-negative
recommendations of the CAP; for example a margin is positive, breast cancer [II, A], unless axillary node involvement is proven
and should be reported as such, when there is ink touching on ultrasound-guided biopsy [53]. With appropriate training in
invasive cancer or DCIS, the anatomic location of the positive the dual radiocolloid/blue dye or indocyanine green
margin should be specified in oriented specimens. For negative fluorescence technique, high identification rates (over 97%), low
margins (i.e. ink not touching invasive cancer or DCIS), the false-negative rates and favourable axillary recurrence rates
distance of invasive cancer and/or DCIS from the margin(s) following SLNB are achievable [54]. SLNB delivers less
should be reported [45]. No tumour at the inked margin is morbidity in terms of shoulder stiffness and arm swelling and
required and >2 mm (for in situ disease) is preferred [46, 47]. allows for a reduced hospital stay [I, A].
Marking the tumour bed with clips in a standardised way There is no definite consensus for the pathologic assessment
facilitates accurate planning of the radiation boost field, if it is of SLNB. The significance of occult micrometastases in terms
indicated. Currently, achievable low local recurrence rates of surgical management and patient outcome appears to be
[<0.5% per year (with a target of <0.25%) and ≤10% overall at negligible [55]. Thus, the authors of this manuscript agree
very long-term follow-up] should be maintained [48]. with the guidelines published by ASCO [53], the National
Comprehensive Cancer Network [56] and others [57, 58], that
mastectomy: Breast reconstruction should be available to routine IHC or polymerase chain reaction is not recommended
those women requiring mastectomy [37]. Immediate for the evaluation of sentinel lymph nodes.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv298 | v

by guest
on 05 February 2018
The optimal management of micrometastatic spread and iso- BRCA1 mutation carrier varies between 65% and 90%, with a
lated tumour cells is the subject of ongoing research. Based on 10-year actuarial risk of contralateral breast cancer ranging from
the results of the IBCSG 23–01 trial, further axillary treatment 25% to 31% [65–67]. With bilateral mastectomy, the risk for
does not seem to be required when a sentinel node (SN) has both subsequent breast cancer incidence and mortality is reduced
micrometastasis (0.2–2 mm) [59]. The presence of macrometa- by 90%–95% [III, A]. Careful genetic assessment and
static spread in the SN traditionally mandated conventional axil- psychological counselling are mandatory before undertaking
lary lymph node clearance. Recent results of a randomised, such surgery.
controlled trial (6.3 years of median follow-up) for patients with Despite the overall trend towards breast conservation, increas-
clinical T1–T2 cN0 invasive breast cancer and one to two ing numbers of breast cancer patients are opting for bilateral
sentinel lymph nodes containing metastases [treated with mastectomy (incorporating contralateral risk-reducing surgery)
breast-conservation surgery (BCS) and tangential adjuvant RT], rather than the preferred breast conservation and mammographic
reported non-inferior rates of overall survival (OS), disease-free surveillance of the irradiated breast. These patients should be
survival (DFS) and locoregional recurrence-free survival [60]. properly counselled and informed that patients with early-stage
Another option in patients with cN0 and sentinel lymph node breast cancer, who opt for BCT, might have an even better
metastases (irrespective of the risk factors) is axillary irradiation, outcome compared with those who have a mastectomy [68].
as demonstrated by the AMAROS study [51]. Therefore, all
patients with micrometastatic spread or isolated tumour cells surgery after primary systemic therapy: Primary systemic
(<0.2 mm) in the SN and patients with limited involvement of therapy should be followed by surgery, according to the
the sentinel lymph node, undergoing tangential breast irradiation principles outlined above. Downsizing of a large unifocal
and adjuvant systemic treatment, may not need to have any primary tumour with neoadjuvant therapy will allow BCT to be
further axillary procedure [II, B]. However, these results need to undertaken in a substantial proportion of patients, even in
be confirmed and cannot be extended to patients with different tumours that were unresectable at diagnosis. This includes those
characteristics than those of the trial’s patient population. with HER2 overexpression and triple-negative breast cancers
who, at presentation, would have otherwise required
surgery for in situ malignancy (intraepithelial neoplasia): mastectomy. With multifocal disease, or where reduction of the
DCIS may be treated with total mastectomy or BCT, provided primary tumour size is more limited, mastectomy will still be
clear resection margins can be achieved. There is no general required. Breast MRI is the most accurate modality for assessing
consensus on what is considered an adequate margin; however, the extent of residual disease following neoadjuvant treatment.
circumferential margins <2 mm are considered less adequate Breast MRI should also be carried out before the start of systemic
and no DCIS on inked margins is a minimal requirement [47]. therapy for proper comparative evaluation. When a breast-
Axillary node evaluation with SLNB is not required with in situ conserving procedure is anticipated, it is necessary to mark the
malignancy, but may be reasonable in large and/or high-grade primary site (using a marker clip or carbon localisation, under
tumours, especially when mastectomy is required (in case an ultrasound guidance) to facilitate accurate surgery.
incidental invasive cancer is subsequently identified in the
surgical specimen). Lobular neoplasia (formerly called lobular radiation therapy
invasive carcinoma:
carcinoma in situ), unlike DCIS, is considered a non-obligate
RT after BCS:
precursor to invasive cancer. It is regarded as a risk factor for
whole breast radiation therapy: Postoperative RT is strongly
future development of invasive cancer in both breasts [relative
recommended after BCS [I, A] [69]. Whole breast radiation
risk (RR) 5.4–12] and does not require active treatment. The
therapy (WBRT) alone reduces the 10-year risk of any first
pleomorphic variant of lobular neoplasia may behave similarly
recurrence (including locoregional and distant) by 15% and the
to DCIS and should be treated accordingly, after
15-year risk of breast cancer-related mortality by 4% [69]. Boost
multidisciplinary discussion.
irradiation gives a further 50% RR reduction. It is indicated for
The risk of a positive SN with pure DCIS is small (7%–9%)
patients who have unfavourable risk factors for local control
and most of the metastases found are micrometastases or iso-
such as: age <50 years, grade 3 tumours, extensive DCIS,
lated tumour cells, detected by immunohistochemistry [61, 62].
vascular invasion or non-radical tumour excision (focally—
The decision to perform an SN procedure should be based on
otherwise further surgery should be advocated) [I, A] [70, 71].
the underlying risk of invasion. This invasive breast cancer
underestimation rate is reported to be 20%–38%, and increases accelerated partial breast irradiation only: The concept of
with: the presence of a palpable mass, an associated density on accelerated partial breast irradiation (APBI) is an appealing
the mammogram, poorly differentiated DCIS in the biopsy, approach to shorten the overall treatment time substantially.
younger age, and a larger extent of microcalcifications [63, 64]. The rationale for APBI is that the majority of local failures occur
Generally, an SN procedure is usually done if there is an asso- in the vicinity of the primary tumour site, while so-called
ciated mass, or if an ablative treatment is chosen. elsewhere in-breast failures may represent a new primary
tumour. However, in the ELIOT (single dose of electrons) and
risk-reducing mastectomy: Risk-reducing surgery (with TARGIT (single intra-operative dose 50 kV X-rays) randomised
prophylactic bilateral mastectomy and reconstruction) may be trials, the ipsilateral breast recurrence rate was significantly
offered to women at very high risk, such as those who have had higher in the APBI groups, compared with the WBRT trial [72,
previous chest irradiation for lymphoma or BRCA1 or BRCA2 73]. Notwithstanding these negative results, APBI might be
gene mutation carriers. The lifetime risk of breast cancer in a considered an acceptable treatment option in patients with a
v | Senkus et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
low risk for local recurrence, for example those who are at least outcomes in those patients. Further hypofractionation (to five
50 years old with: unicentric, unifocal, node-negative, non- fractions) is currently the subject of ongoing prospective clinical
lobular breast cancer, up to 3 cm without the presence of trials.
extensive intraductal components or vascular invasion, and with
negative margins, especially if they will receive adjuvant patients with unresectable disease: Most patients who
hormonal treatment [III, C] [74]. In the meantime, long-term present with unresectable non-metastatic disease will first be
results of several past and still on-going prospective randomised treated with primary systemic therapy. If rendered resectable,
APBI trials are eagerly awaited. this should be followed by surgery and RT, according to the
principles outlined for locoregional advanced disease.
radiation after mastectomy: PMRT in node-positive patients
If the disease remains unresectable, RT should be considered
reduces the 10-year risk of any recurrence (including
to treat all sites of the original tumour extension, with a boost to
locoregional and distant) by 10% and the 20-year risk of breast
residual disease. Most durable remissions can be expected with
cancer-related mortality by 8% [75]. The benefits of PMRT are
an elective dose up to an equivalent of 50 Gy to regions with a
independent from the number of involved axillary lymph nodes
high likelihood of bearing subclinical disease and a boost up to
and the administration of adjuvant systemic treatment.
60–76 Gy (depending on the dose to the organs at risk) to all
Therefore, while PMRT was always recommended for high-risk
sites of macroscopic disease. Regular evaluation during the
patients, including involved resection margins, four or more
course of RT is advised, to select patients that might become
involved axillary lymph nodes [I, A], and T3–T4 tumours
amenable for resection after 45–50 Gy.
independent of the nodal status [II, B], we should now also
Interesting, but early, reports are published on combined ra-
consider routine use of PMRT for patients with one to three
diation and chemotherapy, which should be further evaluated in
positive axillary lymph nodes [I, A] [75, 76].
prospective trials.
regional irradiation: Older randomised trials have used It is advisable that patients are seen by the radiation oncolo-
large comprehensive locoregional RT encompassing the chest gist preceding initiation of primary systemic therapy. If judged
wall and all regional lymph nodes. Recently presented results relevant, a CT scan should be taken in the treatment position,
support this approach, especially for patients with involved for later image co-registration to improve localisation of the
axillary lymph nodes [77–79]. Therefore, although clinically target volumes (e.g. enlarged lymph nodes that might not be
apparent lymph node relapses (especially axillary and internal resectable).
mammary) are rare, nodal irradiation remains indicated for
patients with involved lymph nodes [I, B] [80]. We cannot non-invasive carcinoma (intraepithelial neoplasia): WBRT
discriminate which part of the regional lymph nodes is most after BCS for DCIS decreases the risk of local recurrence, with
important to irradiate. The recent Danish population-based survival equal to that after mastectomy [I, A] [86]. The decrease
study, in which left-sided node-positive breast cancer patients in the risk of local recurrence by RT is evident in all subtypes of
received medial supraclavicular RT, while right-sided patients DCIS. WBRT is recommended in the majority of women with
received the same including the internal mammary nodes, DCIS, on the basis of the substantial reduction in disease
points to the importance of including the internal mammary recurrence, and the inability to define subsets of women who do
lymph nodes in the regional target volume. Regarding the not benefit from radiotherapy [56, 87]. However, in some
supraclavicular part of the target volume, the authors of this patients with low-risk DCIS (tumour size <10 mm, low/
manuscript agree with the new European Society for intermediate nuclear grade, adequate surgical margins), the risk
Radiotherapy and Oncology (ESTRO) guidelines for target of local recurrence following excision only is low, and omitting
volume delineation in breast cancer. The ESTRO guidelines radiation may be an option. Randomised data on additional
advise to include only the most caudal lymph nodes dose to the tumour bed (boost) are lacking, but a boost can be
surrounding the sub-clavicular arch and the base of the jugular considered for patients at higher risk for local failure [III, B].
vein [81]. After axillary lymph node dissection, the resected part APBI should only be carried out within a clinical trial. Total
of the axilla should not be irradiated, except in cases of residual mastectomy with clear margins in DCIS is curative, and PMRT
disease after surgery. is not recommended. Lobular neoplasia is a risk factor for future
development of invasive cancer in both breasts; RT is not
RT doses and fractionation: Doses used for local and/or warranted, with exception of the pleomorphic subtype.
regional adjuvant irradiation have traditionally been 45–50 Gy
in 25–28 fractions of 1.8–2.0 Gy with a typical boost dose of 10–
16 Gy in 2 Gy single doses. Shorter fractionation schemes (e.g. adjuvant systemic treatment
15–16 fractions with 2.5–2.67 Gy single dose) have shown The decision on systemic adjuvant treatment should be based
similar effectiveness and comparable side-effects [I, A] [82–85]. on the predicted sensitivity to particular treatment types, the
Strictly speaking, these data are not fully validated in young benefit from their use and an individual’s risk of relapse. The
patients and in patients with mastectomy and/or additional final decision should also incorporate the predicted treatment
regional irradiation. These patients were either not included or sequelae, the patient’s biological age, general health status, co-
under-represented in the relevant trials. As hypofractionation in morbidities and preferences. Treatment should start preferably
many places is being introduced for all patient subgroups, and within 2–6 weeks after surgery. The data show an important de-
in the unlikelihood of prospective, randomised trials that will crease in systemic therapy efficacy when it is administered more
test this, we advise to carefully monitor, evaluate and compare than 12 weeks after surgery [88].
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv298 | v

by guest
on 05 February 2018
The most recent publication of the Early Breast Cancer The authors of this manuscript agree with the 2013 and
Trialists’ Collaborative Group (EBCTCG) overview states the 2015 St Gallen guidelines, which recommend, for luminal
relative benefit of chemotherapy is similar in all the subgroups cases with unclear chemotherapy indications, that the decision
independent of age, stage, histopathological grade and ER status on systemic adjuvant therapies should be based on the surro-
[89]. This seems to be in contradiction with the results from in- gate intrinsic phenotype (Figure 2), that is determined by
dividual trials, both in the adjuvant and neoadjuvant settings, as ER/PgR, HER2 and Ki67 assessment (Tables 2 and 5), with the
well as knowledge of breast cancer biology. One also needs to selective help of genomic tests when available ( particularly
take into account that many trials included in the EBCTCG MammaPrint, Oncotype DX, Prosigna ROR and Endopredict)
overview have incomplete data on ER expression, in particular [23, 33].
quantitative immunohistochemistry. Furthermore, these trials It must be stressed that IHC/fluorescence in situ hybridisation
have included patients with generally higher risk of recurrence determination of intrinsic phenotype, is not fully accurate. The
than those seen today and often used suboptimal ETs (by current prerequisite for using such a surrogate assessment is the use of
standards). However, these views can be conciliated when ac- standardised assays and a meticulous quality control.
knowledging that, even if the relative benefit would be similar, the All luminal cancers should be treated with ET. Most luminal
absolute benefit derived from adjuvant chemotherapy varies sub- A tumours, except those with the highest risk of relapse (ex-
stantially. The risk to the individual patient is determined by the tensive nodal involvement), require no chemotherapy [I, A]
biology and the burden of the disease, for example the absolute whereas luminal B HER2-negative cancers constitute a popula-
benefit of adjuvant chemotherapy for a low burden luminal- tion of the highest uncertainty regarding chemotherapy indica-
A-like breast cancer, is extremely small and needs to be balanced tions [I, C]. Indications for chemotherapy within this subtype
against the known short- and long-term side-effects. depend on the individual’s risk of relapse, taking into account
Figure 2. (Neo)adjuvant systemic treatment choice by biomarker expression and intrinsic phenotype. ER, oestrogen receptor; HER2, human epidermal
growth factor 2 receptor; ChT, chemotherapy; ET, endocrine therapy; T, trastuzumab.
v | Senkus et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Table 5. Systemic treatment recommendations for early breast cancer subtypes
Subtype Recommended therapy Comments
Luminal A-like ET alone in the majority of cases Consider ChT if:

high tumour burden (four or more positive LN, T3 or higher)
grade 3
Luminal B-like (HER2-negative) ET + ChT for the majority of cases
Luminal B-like (HER2-positive) ChT + anti-HER2 + ET for all patients If contraindications for the use of ChT, one may consider ET + anti-HER2
therapy, although no randomised data exist.
HER2-positive (non-luminal) ChT + anti-HER2
Triple-negative (ductal) ChT
For special histological types, we recommend following the St Gallen 2013 recommendations [23] that propose ET for endocrine responsive histologies
(cribriform, tubular and mucinous), ChT for high-risk endocrine nonresponsive (medullary, metaplastic) and no systemic therapy for low-risk endocrine
nonresponsive (secretory juvenile, adenoid cystic and apocrine).
ET, endocrine therapy; ChT, chemotherapy; LN, lymph node; HER2, human epidermal growth factor 2 receptor.
the tumour extent and features suggestive of its aggressiveness taxane-based chemotherapy [III, B]. If chemotherapy and RT
(grade, proliferation, vascular invasion), presumed responsive- are to be used separately, chemotherapy usually precedes RT.
ness to ET and patient preferences. Features associated with
lower endocrine responsiveness include, low steroid receptor ex- endocrine therapy. ET is indicated in all patients with
pression, lack of PgR expression, high tumour grade and high detectable ER expression (defined as ≥1% of invasive cancer
expression of proliferation markers. Several decision-making cells) irrespective of the use of chemotherapy and/or targeted
tools such as Adjuvant! Online, PREDICT and the Nottingham therapy [I, A] [97, 98]. The choice of agent is primarily
Prognostic Index exist to help in predicting recurrence risks determined by the patient’s menopausal status. Other factors
and benefits from particular treatments [28–30]. uPA-PAI1 include differences in efficacy and side-effect profiles.
tumour markers have level I evidence as prognostic factors and
can be used to aid treatment decision making in early breast premenopausal patients: Tamoxifen 20 mg/day for 5–10
cancer [I, A] [90]. In cases of uncertainty regarding indications years is a standard [I, A]. In patients becoming postmenopausal
for adjuvant chemotherapy (after consideration of other tests), during the first 5 years of tamoxifen, a switch to letrozole, an
gene expression assays, such as MammaPrint, Oncotype DX, aromatase inhibitor (AI), seems to be particularly beneficial
Prosigna and Endopredict, may be used, where available. [99]. The value of adding ovarian suppression [by
These assays can determine the individuals recurrence risk as gonadotropin-releasing hormone (GnRH) agonists or ovarian
well as, potentially predict the benefit of chemotherapy [IV, A] ablation] has long been a matter of controversy, particularly in
[23, 33, 91–94]. Luminal B HER2-positive tumours are treated chemotherapy-treated patients, who frequently develop ovarian
with chemotherapy, ET and trastuzumab [I, A]. No randomised failure as a consequence of cytotoxic treatment [II, B] [100,
data exist to support omission of chemotherapy in this group. 101]. Chemotherapy-induced amenorrhoea has been
However, in small, node-negative tumours, combination of demonstrated to be related to improved long-term outcomes
single-agent paclitaxel and trastuzumab provides excellent [97, 102, 103].
results [95]. Additionally, in cases of contraindications for The SOFT trial demonstrated no significant overall DFS
chemotherapy or patient refusal, in selected cases it may be ac- improvement in patients undergoing ovarian suppression.
ceptable to offer the combination of targeted agents (ET and Treatment effect was most pronounced among those treated
trastuzumab) [V, A]. Triple-negative tumours benefit from adju- with adjuvant chemotherapy. The data on OS is not yet mature.
vant chemotherapy, with the possible exception of low-risk Not all clinical situations have been adequately represented in
‘special histological subtypes’ such as, secretory juvenile, apo- the SOFT trial and a case-by-case decision should be taken, after
crine or adenoid cystic carcinomas [I, A]. HER2 (non-luminal) careful discussion with the patient regarding potential benefits
cancers are treated with chemotherapy plus trastuzumab, apart and substantially different side-effect profiles [I, B] [104]. Less
from selected cases with very low risk, such as T1aN0 [I, A]. clear is the value of combining ovarian suppression and AI, as
In general, chemotherapy should not be used concomitantly contradictory results were obtained in the ABCSG-12 and com-
with ET [II, D] [96]. Trastuzumab may routinely be combined bined SOFT-TEXT trials [I, C] [105, 106]. Combination of
with non-anthracycline-based chemotherapy and ET [I, A]. ovarian ablation and tamoxifen in ER-positive patients is at least
Concomitant use with anthracyclines is not routinely recom- as effective as cyclophosphamide/methotrexate/fluorouracil
mended outside of clinical trials. For most patients, the use of a (CMF)-type chemotherapy and may be used as an alternative
sequential anthracycline-based followed by taxane–trastuzu- [II, A] [100, 107]. The optimal duration of ovarian suppression
mab-based regimen is the preferred choice. RT may be delivered is not known, although it is usually administered for 2–5 years
safely during trastuzumab, ET and non-anthracycline–non- [V, B].
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv298 | v

by guest
on 05 February 2018
For patients with contraindications to tamoxifen, a GnRH analysis suggested that, although it may have some predictive
agonist in combination with an AI should be used. In rare cases value and its use may be related to a small clinical benefit, it
where both tamoxifen and AI are not tolerated, a GnRH agonist cannot be recommended for clinical practice [I, C] [121].
alone may be considered. The role of GnRH agonists in prevent- The addition of taxanes improves the efficacy of chemother-
ing chemotherapy-related ovarian failure has been recently sup- apy, independently of age, nodal status, tumour size or grade,
ported by the efficacy data (less premature ovarian failures and steroid receptor expression or tamoxifen use, but at the cost of
more pregnancies) from the POEMS trial (ER-negative patients) increased non-cardiac toxicity [I, A] [89, 122]. Sequential use of
and safety data from TEXT trial (ER-positive patients) [II, B] anthracyclines and taxanes is superior to concomitant use [I, B]
[106, 108]. However, due to contradictory results from previous [123]. Some data suggest that a taxane-anthracycline sequence
trials, the decision must be taken in a case-by-case manner and may be more effective than the traditionally used anthracycline-
after careful discussion with the patient regarding benefits and taxane order [124]. Overall, chemotherapy regimens based on
risks of such an approach. anthracyclines and taxanes reduce breast cancer mortality by
about one-third [89, 98]. Non-anthracycline, taxane-based regi-
postmenopausal patients: AIs (both non-steroidal and mens, such as four cycles of docetaxel and cyclophosphamide
steroidal) and tamoxifen are valid options. AIs allow for (TC), may be used as an alternative to four cycles of anthracy-
prolongation of the DFS, with no significant impact on OS (1%– cline-based chemotherapy in selected patients (i.e. those at risk
2%, depending on the choice of an upfront or sequential of cardiac complications) [I, A] [125]. No robust, prospective
strategy) [I, B] [109–112]. They can be used upfront (non- randomised data exist on the use of platinum compounds in the
steroidal AI and exemestane), after 2–3 years of tamoxifen adjuvant setting, either in unselected triple-negative tumours or
(non-steroidal AI and exemestane) or as extended adjuvant BRCA 1/2 mutation carriers. Therefore, its use cannot be recom-
therapy, after 5 years of tamoxifen (letrozole and anastrozole) mended for routine use.
[113, 114]. There is no proven benefit for the routine use of AIs Chemotherapy is usually administered for 12–24 weeks (four
for >5 years. The recently published ATLAS study demonstrated to eight cycles), depending on the individual recurrence risk and
an advantage of 10 years rather than 5 years of tamoxifen. With the selected regimen. The use of dose-dense schedules (with
this in mind, extended adjuvant therapy should be discussed granulocyte colony-stimulating factor [G-CSF] support) should
with all patients, except the ones with a very low risk, although be considered, particularly in highly proliferative tumours [I, B]
the optimal duration and regimen of adjuvant ET is currently [126]. High-dose chemotherapy with stem cell support should
unknown [I, C] [115]. not be used [I, E].
The use of tamoxifen is associated with an increased risk of
thromboembolic complications and endometrial hyperplasia HER2-directed therapy. Trastuzumab combined with
(including endometrial cancer). Caution should be exercised chemotherapy in patients with HER2 overexpression/
in patients with conditions predisposing to these sequela. amplification approximately halves the recurrence risk,
Appropriate diagnostic tests should be carried out in those pre- compared with chemotherapy alone, translating into a 10%
senting with symptoms that are suggestive of these complica- absolute improvement in long-term DFS and 9% increase in 10-
tions. Patients on tamoxifen should be advised to avoid the use year OS [I, A] [127–129]. Trastuzumab is approved in patients
of strong and moderate CYP2D6 inhibitors (although there are with node-positive disease and in N0 patients with tumours >1
no unequivocal data on their detrimental effects). If such drugs cm. Due to the relatively high failure risk, even in patients with
cannot be replaced, a switch to alternative treatment, i.e. AIs, N0 tumours <1 cm, it should also be considered in this patient
should be considered [IV, B] [116, 117]. Patients undergoing group, particularly in ER-negative disease [IV, B] [130]. In line
ovarian suppression and those taking AIs, are at an increased with the new ASCO HER2 guidelines, if an HER2 test result is
risk of bone loss and should be advised to have adequate ultimately deemed to be equivocal, even after reflex testing with
calcium and vitamin D3 intake. In addition, periodic assessment an alternative assay, HER2-targeted therapy may also be
of their bone mineral density (by dual energy X-ray absorption considered [18].
[DEXA] scan) should be undertaken [I, A]. In most studies, trastuzumab was administered for 1 year,
although in the FinHER trial a similar improvement was obtained
chemotherapy. Chemotherapy is recommended in the vast with only 9 weeks of treatment [II, A] [131]. No additional
majority of triple-negative, HER2-positive breast cancers and in benefit was demonstrated for 2-year trastuzumab administration
high-risk luminal HER2-negative tumours [I, A]. The absolute in the HERA trial [132]. The PHARE trial compared 6 and 12
benefit from chemotherapy is more pronounced in ER-negative months of trastuzumab, where the non-inferiority of 6 months
tumours [118, 119]. In ER-positive tumours, chemotherapy at of trastuzumab could not be demonstrated. Therefore, a dur-
least partially exerts its effect by induction of ovarian failure [97, ation of one year remains the standard [133]. Trastuzumab is
102]. The most frequently used regimens contain anthracyclines usually well-tolerated, although (usually reversible) cardiac dys-
and/or taxanes, although in selected patients CMF may still be function may occur. Patient selection should be founded on the
used. Four cycles of doxorubicin and cyclophosphamide (AC) baseline cardiac function (expressed by the left ventricular ejec-
are considered equal to six cycles of CMF. The added value of tion fraction). Periodic (usually every 3–4 months) monitoring
six cycles of three-drug anthracycline-based regimens is of cardiac function during treatment is necessary.
controversial [I, C] [89, 120]. Data on topoisomerase IIα as a Due to its cardiotoxicity, trastuzumab should not be rou-
predictive factor for anthracycline-based chemotherapy have tinely administered concomitantly with anthracyclines [I, B].
not been confirmed in prospective studies. A large meta- Combination with taxanes is safe and has been demonstrated
v | Senkus et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
to be more effective than sequential treatment [I, A] [128]. C] [122, 146]. All modalities (chemotherapy, ET and targeted
Trastuzumab may also be safely combined with RT and ET. therapy) used in adjuvant treatment may also be used
In the neoadjuvant setting, dual anti-HER2 blockade associated preoperatively. If chemotherapy is used, it is recommended to
with chemotherapy (trastuzumab + lapatinib, trastuzumab + deliver all planned treatment without unnecessary breaks, i.e.
pertuzumab) has led to improvements in the pCR rate when com- without dividing it into preoperative and postoperative periods,
pared with chemotherapy associated with one anti-HER2 agent. irrespective of the magnitude of tumour response [V, B]. This
However, this did not translate into improvement in long-term will increase the probability of achieving a pCR, which is a
outcomes for the combination of trastuzumab + lapatinib, and proven factor for a good prognosis. For the same reason, in
such a treatment cannot be recommended [134–137]. For the HER2-positive breast cancer, trastuzumab therapy should be
trastuzumab + pertuzumab combination, the results of the large started in the neoadjuvant setting, in association with the taxane
adjuvant Aphinity trial are needed before this can be recom- part of the chemotherapy regimen. This increases the
mended for routine use. However, after reviewing potential risks probability of achieving a pCR. The role of dual HER2 blockade
and benefits (including the financial impact), in selected higher (including a combination of trastuzumab and pertuzumab) is
risk cases it can be considered an acceptable option as neoadju- not well proven and such treatment is not recommended for
vant therapy. routine use, although it may be discussed on a case-by-case
basis. The addition of platinum compound (carboplatin) to
bisphosphonates. Prophylactic use of bisphosphonates, neoadjuvant chemotherapy allows for an increase in the pCR
although not formally approved in most countries, may be rate in triple-negative tumours, particularly in those carrying
discussed in women with a low-oestrogen status (undergoing deleterious BRCA 1/2 or RAD mutations or in patients with a
ovarian suppression or postmenopausal), as prolongation of DFS family history of breast/ovarian cancer. But the effect of those
and breast cancer specific survival was demonstrated in these compounds on long-term outcomes is unknown (I, B) [147–
populations [I, B] [105, 138, 139]. In patients with treatment- 149]. The chemotherapy regimens to be used in the neoadjuvant
related bone loss, bisphosphonates decrease the risk of skeletal setting are the same ones used in the adjuvant setting.
complications [I, A] [140, 141]. Unfortunately, there are no validated predictive markers to
allow the tailoring of the regimen to the individual patient.
treatment of elderly patients. Due to the limited data from Therefore, it is recommended that a sequential regimen of
randomised studies, strong recommendations cannot be made anthracyclines and taxanes is used for the vast majority of
regarding the use of adjuvant systemic therapies in this patients [I, B]. For BRCA 1/2 or RAD mutations it is acceptable
population. In general, treatment decisions should be based on to add a platinum compound, after discussion with the patient.
biological rather than formal age, and ‘fit’ elderly patients should After delivery of the standard 4 to 8 cycles of anthracyclines
receive identical treatments to their younger counterparts. Full and taxanes, even in the absence of pCR, no additional chemo-
doses of drugs should be used, whenever feasible [V, A]. In therapy should be administered in the adjuvant setting, since
patients suitable for standard chemotherapy, single-agent such an approach has no proven benefit.
capecitabine or docetaxel have been demonstrated to be inferior ER-positive, HER2-negative carcinomas, especially of the
to the standard multidrug regimen (AC or CMF) and therefore, lobular subtype, are generally less responsive to primary chemo-
a standard multidrug regimen should be used [II, D] [142, 143]. therapy than ER-negative and HER2-positive tumours, and may
In frail elderly patients, the use of a single-agent pegylated benefit more from primary ET [150]. In post-menopausal
liposomal doxorubicin and metronomic cyclophosphamide plus patients ET is usually given for 4–8 months before surgery or
methotrexate is feasible and demonstrates similar activity, until maximum response, and continued postoperatively. AIs
although their efficacy in comparison to standard chemotherapy are more effective than tamoxifen in decreasing the tumour size
remains unknown [II, B] [144]. and facilitating less extensive surgery [I, A] [151–153]. Due to
paucity of data from randomised trials, preoperative ET is not
systemic adjuvant therapy for DCIS. In patients treated routinely recommended in premenopausal patients outside clin-
conservatively for ER-positive DCIS, tamoxifen decreases the ical trials.
risk of both invasive and non-invasive recurrences, and reduces The vast majority of breast cancer cases in male patients are
the incidence of second primary (contralateral) breast cancer, ductal invasive carcinoma of the luminal type. Tamoxifen is the
without an effect on OS [I, B] [145]. Following mastectomy, standard adjuvant systemic therapy, AIs should not be used
tamoxifen might be considered to decrease the risk of alone in this setting, as they are less effective [154, 155].
contralateral breast cancer in patients who are at a high risk of Chemotherapy indications and regimens should, for the moment,
new breast tumours [II, B]. AIs are being investigated for use in follow the same recommendations as those for luminal-like breast
adjuvant therapy for DCIS, but they should not be used in cancer in female patients [155–157].
routine care at this time.
primary (neoadjuvant) systemic therapy. In locally advanced personalised medicine

and large ‘operable’ cancers, in particular when mastectomy is Breast cancer is the pioneer of personalised medicine in on-
required due to tumour size, primary systemic therapy (used cology. ER and/or PgR and HER2 status have been used for
before local treatment) may decrease the extent of surgery many years as predictive factors to select patients for targeted
needed [I, A]. In operable cases, the timing of treatment ( pre- ET or anti-HER2 treatment. In recent years, surrogate intrinsic
versus postoperative) has no effect on long-term outcomes [II, tumour phenotypes, based on biomarker expression, have also
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv298 | v

by guest
on 05 February 2018
Table 6. Summary of biomarkers used in treatment decision making

Biomarker Prognostic Predictive Technical Clinical validation Test and scoring Patient selection
validation recommendations
ER ++ +++ Yes LOE IB Yes IHC Hormonal treatment

PgR +++ + Yes LOE IB No IHC If negative, chemotherapy in
some cases
HER2 ++ +++ Yes LOE IB Yes IHC ≥10% cells with complete Anti-HER2 treatment
membrane staining
ISH: number of HER2 gene
copies ≥6 or the ratio HER2/
chromosome 17 ≥2
Ki67 ++ + No No IHC no final consensus on cut- Chemotherapy if elevated
off but values below 10% are
considered low and above
30% are high
Intrinsic subtypes ++ ++ Yes Yes Gene expression profile (not for Different responses to
IHC surrogates) neoadjuvant chemotherapy
according to the subtype
First generation +++ ++ Yes Validated retrospectively in Gene expression profile, RT- Chemotherapy if high risk
signatures prospective clinical trials, PCR or high score
(MammaPrint, prospective clinical
Oncotype Dx) validation ongoing
Second generation ++ ++ Yes Validated retrospectively in N-Counter TM technology, Prognosis, chemotherapy if
signatures prospective clinical trials RT-PCR high risk or high score
(Prosigna®,
Endopredict®)
ER, oestrogen receptor; IHC, immunohistochemistry; RT-PCR, reverse transcription polymerase chain reaction; LOE, level of evidence; PgR, progesterone
receptor; ISH, in situ hybridisation; HER2, human epidermal growth factor 2 receptor.
been used for treatment individualisation. Additionally, uPA- • To provide psychological support and information in order to
PAI1, a marker of tumour invasiveness, has been validated in enable a return to normal life after breast cancer.
prospective clinical trials as a prognostic marker for both node-
negative and node-positive breast cancer [I, A] [92] and can Ten-year survival of breast cancer exceeds 70% in most
be used in treatment decision making for early breast cancer. In European regions, with 89% survival for local and 62% for re-
cases when decisions might be challenging, such as luminal gional disease [158]. The annual hazard of recurrence peaks in
B HER2-negative and node-negative breast cancer, commercial- the second year after diagnosis but remains at 2%–5% in years
ly available molecular signatures for ER-positive breast cancer, 5–20. Patients with node-positive disease tend to have higher
such Oncotype DX, EndoPredict, Prosigna, and for all types of annual hazards of recurrence than patients with node-negative
breast cancer ( pN0–1), such as MammaPrint and Genomic cancers. In the first few years, the risk of recurrence is higher in
Grade Index, may be used in conjunction with all clinicopatho- patients with ER-negative cancers, but 5–8 years after diagnosis,
logical factors, to help in treatment decision making [23, 92]. their annual hazard of recurrence drops below the level of ER-
Results from large phase III prospective clinical trials positive tumours [III, B] [159]. Relapse of breast cancer may
(MINDACT, Plan B, TAILORx and RxPONDER) are eagerly occur as late as >20 years after the initial diagnosis, particularly
awaited for an optimal and accurate use of these new tools in in patients with ER/PgR-positive disease.
clinical practice. A biomarker summary is shown in Table 6. Despite the fact that no randomised data exist to support any
particular follow-up sequence or protocol, balancing patient
needs and follow-up costs, we recommend regular visits every
follow-up and long-term implications 3–4 months in the first 2 years, every 6 months from years 3–5
The aims of follow-up are: and annually thereafter [V, A]. Every visit should include a thor-
ough history, eliciting of symptoms and a physical examination.
• To detect early local recurrences or contralateral breast cancer. Annual ipsilateral (after BCT) and/or a contralateral mammog-
• To evaluate and treat therapy-related complications (such as raphy with ultrasound is recommended [II, A]. An MRI of the
menopausal symptoms, osteoporosis and second cancers). breast may be indicated for young patients, especially in cases of
• To motivate patients continuing ET. dense breast tissue and genetic or familial predispositions.
v | Senkus et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Screening and diagnosis

• Mammography screening in the 50- to 70-year age group reduces breast cancer mortality.
• In women with familial breast cancer, with or without proven BRCA mutations, annual screening with MRI of the breast, in combination with
mammography is recommended [III, A].
• Diagnosis and treatment should be carried out in ‘breast units’: specialised institutions caring for a high volume of breast cancer patients, and provided
by a multidisciplinary team including at least a surgeon, radiation oncologist, medical oncologist, radiologist, pathologist and a breast nurse (or another
trained and specialised health care practitioner)—all specialised in and dedicated to breast cancer [IV, A]. The patients should be provided with full,
preferably written, culturally adapted information about their disease and treatment [V, A].
• The diagnosis of breast cancer is based on clinical examination in combination with imaging, and confirmed by pathological assessment. Other
assessments include complete personal and family medical history, including evaluation of menopausal status, a physical examination, a full blood count,
liver and renal function tests, alkaline phosphatase and calcium levels.
• Imaging includes bilateral mammography and ultrasound of the breast and regional lymph nodes. An MRI of the breast is not routinely recommended,
but should be considered in cases of: familial breast cancer associated with BRCA mutations, breast implants, lobular cancers, suspicion of multifocality/
multicentricity (particularly in lobular breast cancer) or large discrepancies between conventional imaging and clinical examination, and before and
during neoadjuvant chemotherapy [III, A].
• Pathological diagnosis should be based on core needle biopsy obtained (preferably) by ultrasound or stereotactic guidance. If preoperative systemic
therapy is planned, a core needle biopsy is mandatory to ensure a diagnosis of invasive disease and assess biomarkers [III, A].
• The pathological report should include the histological type, grade, ER status and, for invasive cancer, PgR status, HER2 status and a proliferation
measure such as Ki67 [III, A]. For the purpose of prognostication and treatment decision making, tumours should be grouped into surrogate intrinsic
subtypes, defined by routine histology and IHC data [III, A]. In case of negativity of ER/PgR and HER2 in the biopsy specimen, it is advisable to retest
for them in the surgical specimen to account for the putative tumour heterogeneity [III, A].
• A marker (e.g. surgical clip, carbon) should be placed into the tumour at biopsy, to ensure surgical resection of the correct site [V, A].

• Lymph nodes should be assessed by clinical examination and ultrasound, supplemented by ultrasound-guided fine needle aspiration or core biopsy of
suspicious lymph nodes [III, A].
• Routine staging evaluations are directed at locoregional disease, as in early breast cancer asymptomatic distant metastases are very rare and patients do
not profit from comprehensive laboratory and radiological staging.
• Asymptomatic distant metastases are very rare and most patients do not benefit from comprehensive laboratory (including tumour markers [25]) and
radiological staging [III, D].
• Additional investigations such as chest CT abdominal ultrasound or CT scan and bone scan should be considered for patients with clinically positive
axillary nodes, large tumours (e.g. ≥5 cm), aggressive biology or clinical signs, symptoms or laboratory values suggesting the presence of metastases
[III, B]. Dual imaging methods combining functional and anatomical information such as FDG-PET/CT may be useful when conventional methods are
inconclusive [V, A]. PET/CT scanning can replace traditional imaging for staging in high-risk patients who are candidates for neoadjuvant
chemotherapy, as well as those with locally advanced and/or inflammatory disease due to their high risk of having metastatic disease [V, B].
• In patients planned for (neo)adjuvant treatment, with anthracyclines and/or trastuzumab, evaluation of cardiac function with a cardiac ultrasound or a
multigated acquisition scan is essential [I, A].
• The postoperative pathological assessment of the surgical specimen should be made according to the pTNM system to include: number, location and
maximum diameter of tumour(s) removed, histological type and grade of the tumour(s), vascular invasion, biomarker analysis, evaluation of the
resection margins, the total number of removed and number of positive lymph nodes and the extent of metastases in the lymph nodes [III, A].
• HER2 gene amplification status may be determined directly from all invasive tumours using in situ hybridisation (fluorescent, chromogenic or silver),
replacing IHC or only for tumours with an ambiguous (2+) IHC score [II, B].
• Proliferation markers such as the Ki67 labelling index may supply additional useful information, particularly if the assay can be standardised [V, A].
Treatment
• The choice of treatment strategy is based on biology (pathology including biomarkers, gene expression) and tumour extent/location (size and location of
primary tumour, number of lesions, number and extent of lymph node involvement) as well as on the age, body habitus and general health status of the
patient and her/his preferences.
• The possibility of hereditary cancer should be explored and, if needed, prophylactic procedures discussed, following appropriate genetic counselling and
testing [IV, D]. Risk-reducing surgery with prophylactic bilateral mastectomy and reconstruction may be offered to women with a very high risk of breast
cancer, such as those carrying the BRCA1 or BRCA2 gene mutations or those with previous chest irradiation for lymphoma. With bilateral mastectomy,
the risk for both subsequent breast cancer incidence and mortality is reduced by 90%–95% [III, A].
• DCIS may be treated with BCS, provided clear resection margins can be achieved, or with mastectomy.
• WBRT after BCS for DCIS decreases the risk of local recurrence with survival equal to that after mastectomy [I, A].
• Following mastectomy for DCIS, tamoxifen might be considered to decrease the risk of contralateral breast cancer in patients who are at a high risk of
new breast tumours [II, B].
Continued
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv298 | v

by guest
on 05 February 2018
Table 7. Continued
• Randomised data on additional dose to the tumour bed (boost) in DCIS patients are lacking, but a boost can be considered for patients at higher risk for
local failure [III, B].
• Breast conservation (wide local excision and RT) is the local treatment of choice in the majority of patients with invasive cancer. In some circumstances,
mastectomy may still be carried out because of tumour size (relative to breast size), tumour multicentricity, prior radiation to the chest or breast, or
patient choice.
• Oncoplastic procedures can achieve better cosmetic outcomes, especially in patients with large breasts, with a less favourable tumour/breast size ratio or
with a cosmetically difficult location of the tumour in the breast.
• Breast reconstruction, preferably immediate, should be available to women requiring mastectomy.
• Silicone gel implants are safe and acceptable components of the reconstructive armamentarium [III, A].
• SLNB rather than full axillary nodal clearance, is now the standard of care, unless axillary node involvement is proven [II, A].
• Patients with isolated tumour cells (<0.2 mm) in the sentinel node and patients with limited involvement of the sentinel lymph nodes undergoing
tangential breast irradiation may not need to have any further axillary procedure [II, B].
• In patients undergoing preoperative systemic therapy, SLNB carried out after systemic therapy demonstrated lower detection rates and higher rates of
false-negatives. However, if the axilla is negative on ultrasound and/or PET/CT scanning, carried out before the start of systemic therapy, a post-systemic
therapy SNLB can be considered [V, B].
• Postoperative RT is strongly recommended after BCS [I, A]. Boost irradiation gives a further 50% risk reduction and is indicated for patients with
unfavourable risk factors for local control [I, A].
• Partial breast irradiation may be considered carefully as an acceptable treatment option in selected patients at least 50 years old, with unicentric, unifocal,
node-negative, non-lobular breast cancer up to 3 cm in size without the presence of an extensive intraductal component or vascular invasion, and with
negative margins [III, C].
• Post-mastectomy RT is recommended for patients with involved axillary nodes and/or with T3–T4 tumours, especially in the presence of additional risk
factors [I, A].
• Although clinically apparent lymph node relapses (especially axillary and internal mammary) are rare, nodal irradiation remains indicated for patients
with involved lymph nodes [I, B].
• Shorter fractionation schemes (e.g. 15–16 fractions with 2.5–2.67 Gy single dose) have been validated in large prospective studies and are generally
recommended [I, A].
• The decision on systemic adjuvant therapies is based on the surrogate intrinsic phenotype determined by ER/PgR, HER2 and Ki67 assessment or on the
genomic-based intrinsic subtype.
• All patients with detectable ER expression, defined as ≥1% of invasive cancer cells, should be offered ET [I, A]. For premenopausal patients, tamoxifen is
a standard [I, A] and ovarian suppression may improve DFS in patients remaining premenopausal after having received chemotherapy. For some
premenopausal patients, the combination of an AI and ovarian suppression can be an option, although long-term follow-up and survival data are still
lacking. For postmenopausal patients, AIs (both non-steroidal and steroidal) and tamoxifen are valid options [I, B].
• Extended adjuvant therapy should be discussed with all postmenopausal patients, except the ones with a very low risk, although the optimal duration and
regimen of adjuvant ET is currently unknown [I, C].
• Patients on tamoxifen should be advised to avoid the use of strong and moderate CYP2D6 inhibitors. If such drugs cannot be replaced, a switch to AIs
(in premenopausal patients in combination with ovarian suppression) should be considered [IV, B].
• Patients undergoing ovarian suppression and those taking AIs are at an increased risk of bone loss and should be advised to have adequate calcium and
vitamin D3 intake. In addition, periodic assessment of their bone mineral density should be undertaken [I, A].
• Chemotherapy is recommended in the vast majority of triple-negative, HER2-positive breast cancers and in high-risk luminal HER2-negative tumours
[I, A].
• The added value of six cycles of three-drug anthracycline-based regiments is controversial [I, C]. The addition of taxanes improves the efficacy of
chemotherapy, independently of age, nodal status, tumour size or grade, steroid receptor expression or tamoxifen use, but at the cost of increased non-
cardiac toxicity [I, A]. Non-anthracycline, taxane-based regimens, such as four cycles of docetaxel and cyclophosphamide, may be used as an alternative
to four cycles of anthracycline-based chemotherapy in selected patients (i.e. those at risk of cardiac complications) [I, A].
• Most luminal A tumours, except those with the highest risk of relapse (extensive nodal involvement), require no chemotherapy [I, A].
• For luminal HER2(−) cancers, the indications for chemotherapy depend on the individual risk of relapse presumed responsiveness to ET and patient
preferences. In general, chemotherapy should not be used concomitantly with ET [II, D].
• Luminal B HER2(+) tumours are treated with chemotherapy, ET and trastuzumab [I, A]; no data exist to support omission of chemotherapy in this
group. In cases of contraindications for chemotherapy or patient refusal, in selected cases it may be acceptable to offer the combination of targeted agents
(ET and trastuzumab) [V, A].
• uPA-PAI1 tumour markers have level I evidence as prognostic factors and can be used to aid treatment decision making in early breast cancer [I, A].
• In cases of uncertainty regarding indications for adjuvant chemotherapy (after consideration of other tests), gene expression assays, such as
MammaPrint, Oncotype DX, Prosigna and Endopredict, may be used, where available. These assays can determine the individual’s recurrence risk as
well as potentially predict the benefit of chemotherapy [IV, A].
• HER2(+) (non-luminal) cancers should be treated with chemotherapy plus trastuzumab [I, A].
• Triple-negative tumours benefit from adjuvant chemotherapy, with possible exclusion of low-risk ‘special histological subtypes’ such as secretory
juvenile, apocrine or adenoid cystic carcinomas [I, A].
Continued
v | Senkus et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Table 7. Continued
• Chemotherapy usually consists of four to eight cycles of anthracycline- and/or taxane-based regimen. Sequential use of anthracyclines and taxanes,
instead of concomitant, is recommended [I, B].
• The use of dose-dense schedules (with G-CSF support) should be considered particularly in highly proliferative tumours [I, B].
• High-dose chemotherapy with stem cell support should not be used [I, E].
• Trastuzumab combined with chemotherapy in patients with HER2 overexpression/amplification approximately halves the recurrence risk and improves
overall survival, compared with chemotherapy alone [I, A].
• Trastuzumab is approved in patients with node-positive disease and in N0 patients with tumours >1 cm. Due to the relatively high failure risk, even in
patients with N0 tumours <1 cm, it should also be considered in this patient group, particularly in ER-negative disease [IV, B].
• Due to its cardiotoxicity, trastuzumab should not be routinely administered concomitantly with anthracyclines [I, B]. Combination with taxanes is safe
and has been demonstrated to be more effective than sequential treatment [I, A].
• RT may be delivered safely during trastuzumab, ET and non-anthracycline–non-taxane-based chemotherapy [III, B].
• Prophylactic use of bisphosphonates may be discussed in women with a low-oestrogen status (undergoing ovarian suppression or postmenopausal) [I,
B]. In patients with treatment-related bone loss, bisphosphonates decrease the risk of skeletal complications [I, A].
• In elderly patients, full doses of drugs should be used, whenever feasible [V, A]. In patients suitable for standard chemotherapy a standard multidrug
regimen should be used [II, D].
• In locally advanced and large ‘operable’ cancers, in particular when mastectomy is required due to tumour size, primary systemic therapy (used before
local treatment) may allow for achieving operability or decreasing the extent of surgery [I, A]. All modalities (chemotherapy, ET and targeted therapy)
used in adjuvant treatment may also be used preoperatively. In operable cases, the timing of treatment (pre- versus postoperative) has no effect on long-
term outcomes [II, C]. If chemotherapy is used, it is recommended to deliver all planned treatment without unnecessary breaks, irrespective of the
magnitude of tumour response [V, B].
Follow-up and survivorship

• The aims of follow-up are to detect early local recurrences or contralateral breast cancer, to evaluate and treat therapy-related complications, to motivate
patients continuing hormonal treatments and to provide psychological support and information in order to enable a return to normal life.
• Regular visits every 3–4 months in the first 2 years, every 6 months from years 3–5 and annually thereafter are recommended [V, A].
• Annual ipsilateral (after BCT) and/or contralateral mammography with ultrasound is recommended [II, A]. In asymptomatic patients, there are no data
to indicate that other laboratory or imaging tests produce a survival benefit but available data come from old studies and new trials are needed.
• Ultrasound can be considered in the follow-up of lobular invasive carcinomas [III, B].
• Routine blood tests are usually indicated to follow-up patients on ET due to the potential side-effects of these drugs, namely in the lipid profile [V, A].
• For patients on tamoxifen, an annual gynaecological examination, possibly with a gynaecological ultrasound, by an experienced gynaecologist is
recommended [V, B].
• Regular bone density evaluation is recommended for patients on AIs [I, A].
• Regular exercise should be recommended to all suitable patients after treatment of breast cancer [II, B].
• Nutritional counselling should be recommended as part of the survivor care for all obese patients [III, B].
• The use of hormone replacement therapy increases the risk of recurrence and should be discouraged [I, A].
MRI, magnetic resonance imaging; ER, oestrogen receptor; PgR, progesterone receptor; HER2, human epidermal growth factor 2 receptor; CT, computed
tomography; FDG, fluorodeoxyglucose; PET, positron emission tomography; IHC, immunohistochemistry; DCIS, ductal carcinoma in situ; BCS, breast-
conserving surgery; WBRT, whole breast radiation therapy; G-CSF, granulocyte colony-stimulating factor; BCT, breast-conserving therapy; RT,
radiotherapy; SLNB, sentinel lymph node biopsy; DFS, disease-free survival; AI, aromatase inhibitor; ET, endocrine therapy; uPA-PAI1, urokinase
plasminogen activator-plasminogen activator inhibitor 1.
Ultrasound can also be considered in the follow-up of lobular sophisticated diagnostic procedures and less efficacious treat-
invasive carcinomas [III, B]. In asymptomatic patients, there are ment of advanced disease, and new trials are urgently needed to
no data to indicate that other laboratory or imaging tests (e.g. reassess this question. In symptomatic patients or in the case of
blood counts, routine chemistry tests, chest X-rays, bone scans, abnormal findings on examination, appropriate tests should be
liver ultrasound exams, CT scans, PET/FDG CT or any tumour carried out immediately.
markers such as CA15–3 or CEA) produce a survival benefit [I, A]. In addition to adequate local and systemic treatments, epi-
However, routine blood tests are usually indicated to follow-up demiological evidence points towards lifestyle factors having an
patients on ET due to the potential side-effects of these drugs, effect on the prognosis of patients with breast cancer, for
namely in the lipid profile [V, A]. For patients on tamoxifen, an example regular exercise provides functional and psychological
annual gynaecological examination, possibly with a gynaecological benefits [II, B] and possibly reduces the risk of recurrence.
ultrasound, by an experienced gynaecologist is recommended Regular exercise is therefore a relatively simple and effective rec-
[V, B]. Regular bone density evaluation is recommended for ommendation that should be made to all suitable patients after
patients on AIs [I, A]. Very importantly, most available data for treatment of breast cancer [II, B] [160]. Weight gain and obesity
follow-up recommendations come from an era of less are likely to adversely affect the prognosis of breast cancer [161].
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv298 | v

by guest
on 05 February 2018
Table 8. Levels of evidence and grades of recommendation adjuvant chemotherapy and/or RT. This is also aggravated by
(adapted from the Infectious Diseases Society of America-United long-term survivorship issues involving work, family and sexual-
States Public Health Service Grading Systema) ity, which are often not closely addressed during follow-up and
result in some women not being able to cope effectively. Long-
Levels of evidence
term survivorship needs to be addressed as a different set of chal-
I Evidence from at least one large, randomised, controlled trial lenges and realities, to encompass the psychosocial needs of
of good methodological quality (low potential for bias) or women once treatment ends. Follow-up clinics should focus not
meta-analyses of well-conducted, randomised trials only on late side-effects but also on issues that deal with the
without heterogeneity long-term implications of living with breast cancer. Assessing
II Small, randomised trials or large, randomised trials with a the various quality-of-life issues, particularly for women under
suspicion of bias (lower methodological quality) or meta- long-term ET, is an important aspect of follow-up care. The role
analyses of such trials or of trials with demonstrated of a specialised breast nurse (or equivalent dedicated health pro-
heterogeneity fessional acting as a patient navigator) throughout a patient’s
III Prospective cohort studies diagnosis, treatment and follow-up is crucial. All countries
IV Retrospective cohort studies or case–control studies should develop the necessary educational structure and infra-
V Studies without the control group, case reports, experts
structure required to provide the help of specialised breast nurses
opinions
within the multidisciplinary team, to all breast cancer patients.
A Strong evidence for efficacy with a substantial clinical benefit, methodology

clinical benefit, generally recommended with the ESMO standard operating procedures for clinical prac-
C Insufficient evidence for efficacy or benefit does not outweigh tice guidelines development. The relevant literature has been
the risk or the disadvantages (adverse events, costs, …), selected by the expert authors. A summary of recommendations
optional is shown in Table 7. Levels of evidence and grades of recom-
D Moderate evidence against the efficacy or for adverse mendation have been applied using the system shown in
outcomes, generally not recommended Table 8. Statements without grading were considered justified
E Strong evidence against the efficacy or for adverse outcomes, standard clinical practice by the experts and the ESMO faculty.
never recommended This manuscript has been subjected to an anonymous peer
review process.
a
ES has reported honoraria from Astellas, Janssen, Bayer, Roche,
Nutritional counselling should be recommended as part of the AstraZeneca, GlaxoSmithKline and Amgen; advisory board
survivor care for all obese patients [III, B]. The use of hormone for Astellas, Pierre Fabre, Amgen, Janssen, Roche and Teva;
replacement therapy increases the risk of recurrence and should travel support from Roche, Novartis, Janssen, Astellas, Amgen
be discouraged [I, A] [162]. and Sandoz. FP-L has reported consultancy/honoraria from
Patients should have unlimited access to specialised rehabili- Roche, GlaxoSmithKline, Genomic Health and Nanostring.
tation facilities and services, to decrease the physical, psycho- SZ has reported travel support from Siemens AG; speaker’s
logical and social sequela of breast cancer treatment. The main fees from Siemens AG and AstraZeneca. FC has reported con-
aims of physiotherapy should include the prevention and treat- sultancy/research grants from Astellas/Medivation, AstraZeneca,
ment of lymphoedema, assuring full range of movements of arm Celgene, Daiichi-Sankyo, Eisai, GE Oncology, Genentech,
and shoulder, and prevention or correction of postural defects GlaxoSmithKline, Merck-Sharp, Merus, Novartis, Pfizer, Roche
resulting from mastectomy. There are no data indicating that and Sanofi. The other authors have declared no potential con-
any type of physiotherapy may increase the risk of recurrence. flicts of interest.
When indicated, patients should not be denied access to re-
habilitation services.
It is uncertain whether women who have undergone axillary references
clearance should be advised to avoid cannulation, venesection 1. http://eco.iarc.fr/EUCAN/CancerOne.aspx?Cancer=46&Gender=2#block-table-f
and blood pressure monitoring in the ipsilateral arm [V, D]. (30 July 2015, date last accessed).
Prompt initiation of antibiotic treatment of potentially infected 2. McTiernan A. Behavioral risk factors in breast cancer: can risk be modified?
wounds on the ipsilateral arm is advised, in particular after axil- Oncologist 2003; 8: 326–334.
lary lymph node dissection. 3. Autier P, Boniol M, La Vecchia C et al. Disparities in breast cancer mortality
trends between 30 European countries: retrospective trend analysis of WHO
Follow-up cannot and should not be seen exclusively from the
mortality database. BMJ 2010; 341: c3620.
physical perspective. Women often have increased levels of 4. Allemani C, Weir HK, Carreira H et al. Global surveillance of cancer survival
anxiety after the completion of treatment, when close contact 1995–2009: analysis of individual data for 25,676,887 patients from 279
with the treatment team decreases. Depression and intense population-based registries in 67 countries (CONCORD-2). Lancet 2015; 385:
fatigue frequently occur in the months following the end of 977–1010.
v | Senkus et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
5. Ottini L, Palli D, Rizzo S et al. Male breast cancer. Crit Rev Oncol Hematol 2010; 28. Blamey RW, Pinder SE, Ball GR et al. Reading the prognosis of the individual with
73: 141–155. breast cancer. Eur J Cancer 2007; 43: 1545–1547.
6. Giordano L, von Karsa L, Tomatis M et al. Mammographic screening 29. Ravdin PM, Siminoff LA, Davis GJ et al. Computer program to assist in making
programmes in Europe: organization, coverage and participation. J Med Screen decisions about adjuvant therapy for women with early breast cancer. J Clin
2012; 19(Suppl 1): 72–82. Oncol 2001; 19: 980–991.
7. Perry N, Broeders M, deWolf C et al. European Guidelines for Quality Assurance 30. Wishart GC, Bajdik CD, Azzato EM et al. A population-based validation of the
in Breast Cancer Screening and Diagnosis, 4th edition. Luxembourg: European prognostic model PREDICT for early breast cancer. Eur J Surg Oncol 2011; 37:
Commission Office for Official Publications of the European Communities 2006. 411–417.
8. Association of European Cancer Leagues. European Union Council 31. Harbeck N, Sotlar K, Wuerstlein R, Doisneau-Sixou S. Molecular and protein
Recommendation on Cancer Screening. http://www.europeancancerleagues.org/ markers for clinical decision making in breast cancer: today and tomorrow.
cancer-in-europe/resources-on-cancer-in-europe/82-eu-council-recommendation- Cancer Treat Rev 2014; 40: 434–444.
on-cancer-screening.html (30 July 2015, date last accessed). 32. Wazir U, Mokbel K. Emerging gene-based prognostic tools in early breast cancer:
9. Gøtzsche PC, Nielsen M. Screening for breast cancer with mammography. first steps to personalised medicine. World J Clin Oncol 2014; 5: 795–799.
Cochrane Database Syst Rev 2011; 1: CD001877. 33. Drukker CA, Bueno-de-Mesquita JM, Retèl VP et al. A prospective evaluation of a
10. Lauby-Secretan B, Scoccianti C, Loomis D et al. Breast-cancer screening— breast cancer prognosis signature in the observational RASTER study. Int J
viewpoint of the IARC Working Group. N Engl J Med 2015; 372: 2353–2358. Cancer 2013; 133: 929–936.
11. Independent UK Panel on Breast Cancer Screening. The benefits and harms of 34. Bossuyt V, Provenzano E, Symmans WF et al. Recommendations for standardized
breast cancer screening: an independent review. Lancet 2012; 380: 1778–1786. pathological characterization of residual disease for neoadjuvant clinical trials
12. Warner E, Messersmith H, Causer P et al. Systematic review: using magnetic of breast cancer by the BIG-NABCG collaboration. Ann Oncol 2015; 26:
resonance imaging to screen women at high risk for breast cancer. Ann Intern 1280–1291.
Med 2008; 148: 671–679. 35. http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/
13. Sardanelli F, Boetes C, Borisch B et al. Magnetic resonance imaging of the Guidances/UCM305501.pdf (30 July 2015, date last accessed).
breast: recommendations from the EUSOMA working group. Eur J Cancer 2010; 36. Del Turco MR, Ponti A, Bick U et al. Quality indicators in breast cancer care. Eur
46: 1296–1316. J Cancer 2010; 46: 2344–2356.
14. Kuehn T, Bauerfeind I, Fehm T et al. Sentinel-lymph-node biopsy in patients with 37. EUSOMA. The requirements of a specialist breast unit. Eur J Cancer 2000; 36:
breast cancer before and after neoadjuvant chemotherapy (SENTINA): a 2288–2293.
prospective, multicentre cohort study. Lancet Oncol 2013; 14: 609–618. 38. Wilson AR, Marotti L, Bianchi S et al. The requirements of a specialist Breast
15. Boughey JC, Suman VJ, Mittendorf EA et al. Sentinel lymph node surgery after Centre. Eur J Cancer 2013; 49: 3579–3587.
neoadjuvant chemotherapy in patients with node-positive breast cancer: the 39. Pruthi S, Gostout BS, Lindor NM. Identification and management of women with
ACOSOG Z1071 (Alliance) clinical trial. JAMA 2013; 310: 1455–1461. BRCA mutations or hereditary predisposition for breast and ovarian cancer. Mayo
16. Lakhani SR, Ellis IO, Schnitt SJ et al. WHO Classification of Tumours, 4th edition. Clin Proc 2010; 85: 1111–1120.
Lyon: IARC WHO Classification of Tumours, IARC Press 2012. 40. Senkus E, Gomez H, Dirix L et al. Young breast cancer patients’ attitudes towards
17. Hammond ME. American Society of Clinical Oncology-College of American the risk of loss of fertility related to adjuvant therapies. EORTC study 10002 BIG
Pathologists guidelines for breast predictive factor testing: an update. Appl 3–98. Psychooncology 2014; 23: 173–182.
Immunohistochem Mol Morphol 2011; 19: 499–500. 41. Lee SJ, Schover LR, Partridge AH et al. American Society of Clinical Oncology
18. Wolff AC, Hammond ME, Hicks DG et al. Recommendations for human epidermal recommendations on fertility preservation in cancer patients. J Clin Oncol 2006;
growth factor receptor 2 testing in breast cancer: American Society of Clinical 24: 2917–2931.
Oncology/College of American Pathologists clinical practice guideline update. 42. Partridge AH, Pagani O, Abulkhair O et al. First international consensus guidelines
J Clin Oncol 2013; 31: 3997–4013. for breast cancer in young women (BCY1). Breast 2014; 23: 209–220.
19. Dowsett M, Nielsen TO, A’Hern R et al. Assessment of Ki67 in breast cancer: 43. Cardoso F, Loibl S, Pagani O et al. The European Society of Breast Cancer
recommendations from the International Ki-67 Breast Cancer Working Group. Specialists recommendations for the management of young women with breast
J Natl Cancer Inst 2011; 103: 1656–1664. cancer. Eur J Cancer 2012; 48: 3355–3377.
20. Guiu S, Michiels S, André F et al. Molecular subclasses of breast cancer: how do 44. Association of Breast Surgery at Baso 2009. Surgical guidelines for the
we define them? The IMPAKT 2012 Working Group Statement. Ann Oncol 2012; management of breast cancer. Eur J Surg Oncol 2009; 35(Suppl 1): 1–22.
23: 2997–3006. 45. Schnitt SJ, Moran MS, Houssami N, Morrow M. The Society of Surgical
21. Mann GB, Fahey VD, Feleppa F, Buchanan MR. Reliance on hormone receptor Oncology-American Society for Radiation Oncology Consensus Guideline on
assays of surgical specimens may compromise outcome in patients with breast margins for breast-conserving surgery with whole-breast irradiation in stages I
cancer. J Clin Oncol 2005; 23: 5148–5154. and II invasive breast cancer: perspectives for pathologists. Arch Pathol Lab Med
22. Chen X, Yuan Y, Gu Z, Shen K. Accuracy of estrogen receptor, progesterone 2015; 139: 575–577.
receptor, and HER2 status between core needle and open excision biopsy in 46. Houssami N, Macaskill P, Marinovich ML et al. Meta-analysis of the impact of
breast cancer: a meta-analysis. Breast Cancer Res Treat 2012; 134: 957–967. surgical margins on local recurrence in women with early-stage invasive breast
23. Coates AS, Winer EP, Goldhirsch A et al. Tailoring therapies—improving the cancer treated with breast-conserving therapy. Eur J Cancer 2010; 46: 3219–3232.
management of early breast cancer: St Gallen International Expert Consensus on 47. Moran MS, Schnitt SJ, Giuliano AE et al. Society of Surgical Oncology-American
the Primary Therapy of Early Breast Cancer 2015. Ann Oncol 2015; 26: Society for Radiation Oncology consensus guideline on margins for breast-
1533–1546. conserving surgery with whole-breast irradiation in stages I and II invasive breast
24. Edge SB, Byrd DR, Compton CC (eds). AJCC Cancer Staging Handbook, 7th cancer. J Clin Oncol 2014; 32: 1507–1515.
edition. New York, NY: Springer 2010. 48. Poortmans P, Aznar M, Bartelink H. Quality indicators for breast cancer: revisiting
25. Harris L, Fritsche H, Mennel R et al. American Society of Clinical Oncology 2007 historical evidence in the context of technology changes. Semin Radiat Oncol
update of recommendations for the use of tumor markers in breast cancer. J Clin 2012; 22: 29–39.
Oncol 2007; 25: 5287–5312. 49. Chatterjee JS, Lee A, Anderson W et al. Effect of postoperative radiotherapy on
26. Koolen BB, Vrancken Peeters MJ, Aukema TS et al. 18F-FDG PET/CT as a staging autologous deep inferior epigastric perforator flap volume after immediate breast
procedure in primary stage II and III breast cancer: comparison with conventional reconstruction. Br J Surg 2009; 96: 1135–1140.
imaging techniques. Breast Cancer Res Treat 2012; 131: 117–126. 50. Senkus-Konefka E, Wełnicka-Jaśkiewicz M, Jaśkiewicz J, Jassem J.
27. Robertson IJ, Hand F, Kell MR. FDG-PET/CT in the staging of local/regional Radiotherapy for breast cancer in patients undergoing breast reconstruction or
metastases in breast cancer. Breast 2011; 20: 491–494. augmentation. Cancer Treat Rev 2004; 30: 671–682.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv298 | v

by guest
on 05 February 2018
51. Donker M, van Tienhoven G, Straver ME et al. Radiotherapy or surgery of the 71. Bartelink H, Maingon P, Poortmans P et al. Whole-breast irradiation with or
axilla after a positive sentinel node in breast cancer (EORTC 10981–22023 without a boost for patients treated with breast-conserving surgery for early
AMAROS): a randomised, multicentre, open-label, phase 3 non-inferiority trial. breast cancer: 20-year follow-up of a randomised phase 3 trial. Lancet Oncol
Lancet Oncol 2014; 15: 1303–1310. 2015; 16: 47–56.
52. Gebruers N, Verbelen H, De Vrieze T et al. Incidence and time path of lymphedema 72. Veronesi U, Orecchia R, Maisonneuve P et al. Intraoperative radiotherapy versus
in sentinel node negative breast cancer patients: a systematic review. Arch Phys external radiotherapy for early breast cancer (ELIOT): a randomised controlled
Med Rehabil 2015; 96: 1131–1139. equivalence trial. Lancet Oncol 2013; 14: 1269–1277.
53. Lyman GH, Temin S, Edge SB et al.; American Society of Clinical Oncology 73. Vaidya JS, Wenz F, Bulsara M et al. Risk-adapted targeted intraoperative
Clinical Practice. Sentinel lymph node biopsy for patients with early-stage breast radiotherapy versus whole-breast radiotherapy for breast cancer: 5-year results
cancer: American Society of Clinical Oncology clinical practice guideline update. for local control and overall survival from the TARGIT-A randomised trial. Lancet
J Clin Oncol 2014; 32: 1365–1383. 2014; 383: 603–613.
54. Krag DN, Anderson SJ, Julian TB et al. Sentinel-lymph-node resection compared 74. Polgár C, Van Limbergen E, Potter R et al. Patient selection for accelerated partial
with conventional axillary-lymph-node dissection in clinically node-negative breast irradiation (APBI) after breast-conserving surgery: recommendations of the
patients with breast cancer: overall survival findings from the NSABP B-32 Groupe Européen de Curiethérapie-European Society for Therapeutic Radiology
randomised phase 3 trial. Lancet Oncol 2010; 11: 927–933. and Oncology (GEC-ESTRO) breast cancer working group based on clinical
55. Giuliano AE, Hawes D, Ballman KV et al. Association of occult metastases in evidence (2009). Radiother Oncol 2010; 94: 264–273.
sentinel lymph nodes and bone marrow with survival among women with early- 75. Early Breast Cancer Trialists’ Collaborative Group (EBCTCG). Effect of radiotherapy
stage invasive breast cancer. JAMA 2011; 306: 385–393. after mastectomy and axillary surgery on 10-year recurrence and 20-year breast
56. NCCN Clinical Practice Guidelines in Oncology. Breast Cancer Version 3.2014, cancer mortality: meta-analysis of individual patient data for 8135 women in 22
http://www.nccn.org/professionals/physician_gls/pdf/breast.pdf (30 July 2015, randomised trials. Lancet 2014; 383: 2127–2135.
date last accessed). 76. Kyndi M, Overgaard M, Nielsen HM et al. High local recurrence risk is not
57. Kaufmann M, Morrow M, von Minckwitz G et al. Locoregional treatment of associated with large survival reduction after postmastectomy radiotherapy in
primary breast cancer: consensus recommendations from an International expert high-risk breast cancer: a subgroup analysis of DBCG 82 b&c. Radiother Oncol
panel. Cancer 2010; 116: 1184–1191. 2009; 90: 74–79.
58. Weaver DL. Pathology evaluation of sentinel lymph nodes in breast cancer: protocol 77. Poortmans PM, Collette S, Kirkove C, et al. Internal mammary and medial
recommendations and rationale. Mod Pathol 2010; 23(Suppl 2): S26–S32. supraclavicular irradiation in breast cancer. N Engl J Med 2015; 373: 317–327.
59. Galimberti V, Cole BF, Zurrida S et al. Axillary dissection versus no axillary 78. Whelan TJ, Olivotto IA, Parulekar WR, et al. Regional nodal irradiation in early-
dissection in patients with sentinel-node micrometastases (IBCSG 23–01): a stage breast cancer. N Engl J Med 2015; 373: 307–316.
phase 3 randomised controlled trial. Lancet Oncol 2013; 14: 297–305. 79. Thorsen LBJ, Berg M, Brodersen HJ et al. DBCG-IM: improved survival with
60. Giuliano AE, Hunt KK, Ballman KV et al. Axillary dissection versus no axillary internal mammary node irradiation: a prospective study on 3,072 breast cancer
dissection in women with invasive breast cancer and sentinel node metastasis: a patients. Radiother Oncol 2014; (Suppl 1): abstr OC-0148.
randomized clinical trial. JAMA 2011; 305: 569–575. 80. Poortmans P. Postmastectomy radiation in breast cancer with one to three
61. Leidenius M, Salmenkivi K, von Smitten K, Heikkila P. Tumour-positive sentinel involved lymph nodes: ending the debate. Lancet 2014; 383: 2104–2106.
node findings in patients with ductal carcinoma in situ. J Surg Oncol 2006; 94: 81. Offersen BV, Boersma LJ, Kirkove C et al. ESTRO consensus guideline on target
380–384. volume delineation for elective radiation therapy of early stage breast cancer.
62. Moore KH, Sweeney KJ, Wilson ME et al. Outcomes for women with ductal Radiother Oncol 2015; 114: 3–10.
carcinoma-in-situ and a positive sentinel node: a multi-institutional audit. Ann 82. Whelan TJ, Pignol JP, Levine MN et al. Long-term results of hypofractionated
Surg Oncol 2007; 14: 2911–2917. radiation therapy for breast cancer. N Engl J Med 2010; 362: 513–520.
63. Meijnen P, Oldenburg HS, Loo CE et al. Risk of invasion and axillary lymph node 83. Bentzen SM, Agrawal RK et al. The UK Standardisation of Breast Radiotherapy
metastasis in ductal carcinoma in situ diagnosed by core-needle biopsy. Br J (START) trial B of radiotherapy hypofractionation for treatment of early breast
Surg 2007; 94: 952–956. cancer: a randomised trial. Lancet 2008; 371: 1098–1107.
64. Yen TW, Hunt KK, Ross MI et al. Predictors of invasive breast cancer in patients 84. Bentzen SM, Agrawal RK et al. The UK Standardisation of Breast Radiotherapy
with an initial diagnosis of ductal carcinoma in situ: a guide to selective use of (START) trial A of radiotherapy hypofractionation for treatment of early breast
sentinel lymph node biopsy in management of ductal carcinoma in situ. J Am cancer: a randomised trial. Lancet Oncol 2008; 9: 331–341.
Coll Surg 2005; 200: 516–526. 85. Adams S, Chakravarthy AB, Donach M et al. Preoperative concurrent paclitaxel-
65. Risch HA, McLaughlin JR, Cole DE et al. Population BRCA1 and BRCA2 mutation radiation in locally advanced breast cancer: pathologic response correlates with
frequencies and cancer penetrances: a kin-cohort study in Ontario, Canada. J five-year overall survival. Breast Cancer Res Treat 2010; 124: 723–732.
Natl Cancer Inst 2006; 98: 1694–1706. 86. Goodwin A, Parker S, Ghersi D, Wilcken N. Post-operative radiotherapy for ductal
66. Antoniou A, Pharoah PD, Narod S et al. Average risks of breast and ovarian carcinoma in situ of the breast—a systematic review of the randomized trials.
cancer associated with BRCA1 or BRCA2 mutations detected in case series Breast 2009; 18: 143–149.
unselected for family history: a combined analysis of 22 studies. Am J Hum 87. Morrow M, Schnitt SJ, Norton L. Current management of lesions associated with
Genet 2003; 72: 1117–1130. an increased risk of breast cancer. Current management of lesions associated
67. Liebens FP, Carly B, Pastijn A, Rozenberg S. Management of BRCA1/2 with an increased risk of breast cancer. Nat Rev Clin Oncol 2015; 12: 227–238.
associated breast cancer: a systematic qualitative review of the state of 88. Lohrisch C, Paltiel C, Gelmon K et al. Impact on survival of time from definitive
knowledge in 2006. Eur J Cancer 2007; 43: 238–257. surgery to initiation of adjuvant chemotherapy for early-stage breast cancer. J Clin
68. Hwang ES, Lichtensztajn DY, Gomez SL et al. Survival after lumpectomy and Oncol 2006; 24: 4888–4894.
mastectomy for early stage invasive breast cancer: the effect of age and hormone 89. Peto R, Davies C et al. Comparisons between different polychemotherapy
receptor status. Cancer 2013; 119: 1402–1411. regimens for early breast cancer: meta-analyses of long-term outcome among
69. Darby S, McGale P et al. Effect of radiotherapy after breast-conserving surgery on 100,000 women in 123 randomised trials. Lancet 2012; 379: 432–444.
10-year recurrence and 15-year breast cancer death: meta-analysis of individual 90. Harbeck N, Kates RE, Look MP et al. Enhanced benefit from adjuvant
patient data for 10,801 women in 17 randomised trials. Lancet 2011; 378: chemotherapy in breast cancer patients classified high-risk according to
1707–1716. urokinase-type plasminogen activator (uPA) and plasminogen activator inhibitor
70. Werkhoven EV, Hart G, Tinteren HV et al. Nomogram to predict ipsilateral breast type 1 (n=3424). Cancer Res 2002; 62: 4617–4622.
relapse based on pathology review from the EORTC 22881–10882 boost versus 91. Paik S, Shak S, Tang G et al. A multigene assay to predict recurrence of tamoxifen-
no boost trial. Radiother Oncol 2011; 100: 101–107. treated, node-negative breast cancer. N Engl J Med 2004; 351: 2817–2826.
v | Senkus et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
92. Azim HA, Jr, Michiels S, Zagouri F et al. Utility of prognostic genomic tests in women with hormone receptor-positive breast cancer. J Clin Oncol 2010; 28:
breast cancer practice: the IMPAKT 2012 working group consensus statement. 3784–3796.
Ann Oncol 2013; 24: 647–654. 114. Dowsett M, Cuzick J, Ingle J et al. Meta-analysis of breast cancer outcomes in
93. Dowsett M, Sestak I, Lopez-Knowles E et al. Comparison of PAM50 risk of adjuvant trials of aromatase inhibitors versus tamoxifen. J Clin Oncol 2010; 28:
recurrence score with oncotype DX and IHC4 for predicting risk of distant 509–518.
recurrence after endocrine therapy. J Clin Oncol 2013; 31: 2783–2790. 115. Davies C, Pan H, Godwin J et al. Long-term effects of continuing adjuvant
94. Dubsky P, Brase JC, Jakesz R et al. The EndoPredict score provides prognostic tamoxifen to 10 years versus stopping at 5 years after diagnosis of oestrogen
information on late distant metastases in ER+/HER2- breast cancer patients. Br J receptor-positive breast cancer: ATLAS, a randomised trial. Lancet 2013; 381:
Cancer 2013; 109: 2959–2964. 805–816.
95. Tolaney SM, Barry WT, Dang CT et al. Adjuvant paclitaxel and trastuzumab for 116. Regan MM, Leyland-Jones B, Bouzyk M, et al. CYP2D6 genotype and tamoxifen
node-negative, HER2-positive breast cancer. N Engl J Med 2015; 372: 134–141. response in postmenopausal women with endocrine-responsive breast cancer:
96. Albain KS, Barlow WE, Ravdin PM et al. Adjuvant chemotherapy and timing of the breast international group 1–98 trial. J Natl Cancer Inst 2012; 104: 441–451.
tamoxifen in postmenopausal patients with endocrine-responsive, node-positive 117. Sideras K, Ingle JN, Ames MM et al. Coprescription of tamoxifen and medications
breast cancer: a phase 3, open-label, randomised controlled trial. Lancet 2009; that inhibit CYP2D6. J Clin Oncol 2010; 28: 2768–2776.
374: 2055–2063. 118. Berry DA, Cirrincione C, Henderson IC et al. Estrogen-receptor status and
97. Colleoni M, Gelber S et al. Tamoxifen after adjuvant chemotherapy for outcomes of modern chemotherapy for patients with node-positive breast cancer.
premenopausal women with lymph node-positive breast cancer: International JAMA 2006; 295: 1658–1667.
Breast Cancer Study Group trial 13–93. J Clin Oncol 2006; 24: 1332–1341. 119. Clarke M, Coates AS, Darby SC et al. Adjuvant chemotherapy in oestrogen
98. Early Breast Cancer Trialists’ Collaborative Group. Effects of chemotherapy and receptor-poor breast cancer: patient-level meta-analysis of randomised trials.
hormonal therapy for early breast cancer on recurrence and 15-year survival: an Lancet 2008; 371: 29–40.
overview of the randomised trials. Lancet 2005; 365: 1687–1717. 120. Geyer CE, Jr, Samuel JA, Wilson JW et al. A randomized phase III trial comparing
99. Goss PE, Ingle JN, Martino S et al. A randomized trial of letrozole in six cycles of 5-fluorouracil (5-FU), epirubicin, and cyclophosphamide (FEC) to
postmenopausal women after five years of tamoxifen therapy for early-stage four cycles of adriamycin and cyclophosphamide (AC) in patients ( pts) with node-
breast cancer. N Engl J Med 2003; 349: 1793–1802. negative breast cancer. SABCS 2014; S3-02.
100. Cuzick J, Ambroisine L et al. Use of luteinizing-hormone-releasing hormone 121. Di Leo A, Desmedt C, Bartlett JM et al. HER2 And TOP2A as predictive markers
agonists as adjuvant treatment in premenopausal patients with hormone- for anthracycline-containing chemotherapy regimens as adjuvant treatment of
receptor-positive breast cancer: a meta-analysis of individual patient data from breast cancer: a meta-analysis of individual patient data. Lancet Oncol 2011; 12:
randomised adjuvant trials. Lancet 2007; 369: 1711–1723. 1134–1142.
101. Davidson NE, O’Neill AM, Vukov AM et al. Chemoendocrine therapy for 122. Gianni L, Baselga J, Eiermann W et al. Phase III trial evaluating the addition of
premenopausal women with axillary lymph node-positive, steroid hormone paclitaxel to doxorubicin followed by cyclophosphamide, methotrexate, and
receptor-positive breast cancer: results from INT 0101 (E5188). J Clin Oncol fluorouracil, as adjuvant or primary systemic therapy: European Cooperative Trial
2005; 23: 5973–5982. in Operable Breast Cancer. J Clin Oncol 2009; 27: 2474–2481.
102. Swain SM, Jeong JH, Wolmark N. Amenorrhea from breast cancer therapy—not 123. Shao N, Wang S, Yao C et al. Sequential versus concurrent anthracyclines and
a matter of dose. N Engl J Med 2010; 363: 2268–2270. taxanes as adjuvant chemotherapy of early breast cancer: a meta-analysis of
103. Pagani O, O’Neill A, Castiglione M et al. Prognostic impact of amenorrhoea after phase III randomized control trials. Breast 2012; 21: 389–393.
adjuvant chemotherapy in premenopausal breast cancer patients with axillary 124. Earl HM, Vallier AL, Hiller L et al. Effects of the addition of gemcitabine, and
node involvement: results of the International Breast Cancer Study Group (IBCSG) paclitaxel-first sequencing, in neoadjuvant sequential epirubicin, cyclophosphamide,
Trial VI. Eur J Cancer 1998; 34: 632–640. and paclitaxel for women with high-risk early breast cancer (Neo-tAnGo): an open-
104. Francis PA, Regan MM, Fleming GF et al. Adjuvant ovarian suppression in label, 2×2 factorial randomised phase 3 trial. Lancet Oncol 2014; 15: 201–212.
premenopausal breast cancer. N Engl J Med 2015; 372: 436–446. 125. Jones S, Holmes FA, O’Shaughnessy J et al. Docetaxel with cyclophosphamide
105. Gnant M, Mlineritsch B, Schippinger W et al. Endocrine therapy plus zoledronic is associated with an overall survival benefit compared with doxorubicin and
acid in premenopausal breast cancer. N Engl J Med 2009; 360: 679–691. cyclophosphamide: 7-year follow-up of US Oncology Research Trial 9735. J Clin
106. Pagani O, Regan MM, Walley BA et al. Adjuvant exemestane with ovarian Oncol 2009; 27: 1177–1183.
suppression in premenopausal breast cancer. N Engl J Med 2014; 371: 107–118. 126. Citron ML, Berry DA, Cirrincione C et al. Randomized trial of dose-dense versus
107. Jonat W, Kaufmann M, Sauerbrei W et al. Goserelin versus cyclophosphamide, conventionally scheduled and sequential versus concurrent combination
methotrexate, and fluorouracil as adjuvant therapy in premenopausal patients chemotherapy as postoperative adjuvant treatment of node-positive primary
with node-positive breast cancer: the Zoladex Early Breast Cancer Research breast cancer: first report of Intergroup Trial C9741/Cancer and Leukemia Group
Association study. J Clin Oncol 2002; 20: 4628–4635. B Trial 9741. J Clin Oncol 2003; 21: 1431–1439.
108. Moore HC, Unger JM, Phillips KA et al. Goserelin for ovarian protection during 127. Gianni L, Dafni U, Gelber RD et al. Treatment with trastuzumab for 1 year after
breast-cancer adjuvant chemotherapy. N Engl J Med 2015; 372: 923–932. adjuvant chemotherapy in patients with HER2-positive early breast cancer: a 4-?
109. Bliss JM, Kilburn LS, Coleman RE et al. Disease-related outcomes with long-term year follow-up of a randomised controlled trial. Lancet Oncol 2011; 12:
follow-up: an updated analysis of the intergroup exemestane study. J Clin Oncol 236–244.
2012; 30: 709–717. 128. Perez EA, Romond EH, Suman VJ et al. Trastuzumab plus adjuvant
110. Regan MM, Neven P, Giobbie-Hurder A et al. Assessment of letrozole and chemotherapy for human epidermal growth factor receptor 2-positive breast
tamoxifen alone and in sequence for postmenopausal women with steroid cancer: planned joint analysis of overall survival from NSABP B-31 and NCCTG
hormone receptor-positive breast cancer: the BIG 1–98 randomised clinical trial N9831. J Clin Oncol 2014; 32: 3744–3752.
at 8.1 years median follow-up. Lancet Oncol 2011; 12: 1101–1108. 129. Slamon D, Eiermann W, Robert N et al. Adjuvant trastuzumab in HER2-positive
111. Cuzick J, Sestak I, Baum M et al. Effect of anastrozole and tamoxifen as adjuvant breast cancer. N Engl J Med 2011; 365: 1273–1283.
treatment for early-stage breast cancer: 10-year analysis of the ATAC trial. Lancet 130. Gonzalez-Angulo AM, Litton JK, Broglio KR et al. High risk of recurrence for patients
Oncol 2010; 11: 1135–1141. with breast cancer who have human epidermal growth factor receptor 2-positive,
112. Goss PE, Ingle JN, Pater JL et al. Late extended adjuvant treatment with letrozole node-negative tumors 1 cm or smaller. J Clin Oncol 2009; 27: 5700–5706.
improves outcome in women with early-stage breast cancer who complete 5 131. Joensuu H, Bono P, Kataja V et al. Fluorouracil, epirubicin, and
years of tamoxifen. J Clin Oncol 2008; 26: 1948–1955. cyclophosphamide with either docetaxel or vinorelbine, with or without
113. Burstein HJ, Prestrud AA, Seidenfeld J et al. American Society of Clinical trastuzumab, as adjuvant treatments of breast cancer: final results of the FinHer
Oncology clinical practice guideline: update on adjuvant endocrine therapy for Trial. J Clin Oncol 2009; 27: 5685–5692.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv298 | v

by guest
on 05 February 2018
132. Goldhirsch A, Gelber RD, Piccart-Gebhart MJ et al. 2 years versus 1 year of 148. Sikov WM, Berry DA, Perou CM et al. Impact of the addition of carboplatin and/or
adjuvant trastuzumab for HER2-positive breast cancer (HERA): an open-label, bevacizumab to neoadjuvant once-per-week paclitaxel followed by dose-dense
randomised controlled trial. Lancet 2013; 382: 1021–1028. doxorubicin and cyclophosphamide on pathologic complete response rates in
133. Pivot X, Romieu G, Debled M et al. 6 months versus 12 months of adjuvant stage II to III triple-negative breast cancer: CALGB 40603 (Alliance). J Clin Oncol
trastuzumab for patients with HER2-positive early breast cancer (PHARE): a 2015; 33: 13–21.
randomised phase 3 trial. Lancet Oncol 2013; 14: 741–748. 149. Rugo HS, Olopade O, DeMichele A et al. Veliparib/carboplatin plus standard
134. Baselga J, Bradbury I, Eidtmann H et al. Lapatinib with trastuzumab for HER2- neoadjuvant therapy for high-risk breast cancer: first efficacy results from the I-
positive early breast cancer (NeoALTTO): a randomised, open-label, multicentre, SPY 2 TRIAL. SABCS 2013; S5-02.
phase 3 trial. Lancet 2012; 379: 633–640. 150. von Minckwitz G, Untch M, Nüesch E et al. Impact of treatment characteristics on
135. Guarneri V, Frassoldati A, Bottini A et al. Preoperative chemotherapy plus response of different breast cancer phenotypes: pooled analysis of the German
trastuzumab, lapatinib, or both in human epidermal growth factor receptor 2- neo-adjuvant chemotherapy trials. Breast Cancer Res Treat 2011; 125:
positive operable breast cancer: results of the randomized phase II CHER-LOB 145–156.
study. J Clin Oncol 2012; 30: 1989–1995. 151. Cataliotti L, Buzdar AU, Noguchi S et al. Comparison of anastrozole versus
136. Gianni L, Pienkowski T, Im YH et al. Efficacy and safety of neoadjuvant tamoxifen as preoperative therapy in postmenopausal women with hormone
pertuzumab and trastuzumab in women with locally advanced, inflammatory, or receptor-positive breast cancer: the Pre-Operative ‘Arimidex’ compared to
early HER2-positive breast cancer (NeoSphere): a randomised multicentre, open- tamoxifen (PROACT) trial. Cancer 2006; 106: 2095–2103.
label, phase 2 trial. Lancet Oncol 2012; 13: 25–32. 152. Smith IE, Dowsett M, Ebbs SR et al. Neoadjuvant treatment of postmenopausal
137. Piccart-Gebhart MJ, Holmes AP, Baselga J et al. First results from the phase III breast cancer with anastrozole, tamoxifen, or both in combination: the
ALTTO trial (BIG 2–06; NCCTG [Alliance] N063D) comparing one year of anti- immediate preoperative anastrozole, tamoxifen, or combined with tamoxifen
HER2 therapy with lapatinib alone (L), trastuzumab alone (T), their sequence (IMPACT) multicenter double-blind randomized trial. J Clin Oncol 2005; 23:
(T→L), or their combination (T+L) in the adjuvant treatment of HER2-positive 5108–5116.
early breast cancer (EBC). J Clin Oncol 2014; 32(suppl 5s); abstr LBA4. 153. Eiermann W, Paepke S, Appfelstaedt J et al. Preoperative treatment of
138. Coleman R, Cameron D, Dodwell D et al. Adjuvant zoledronic acid in patients with postmenopausal breast cancer patients with letrozole: a randomized double blind
early breast cancer: final efficacy analysis of the AZURE (BIG 01/04) randomized multicenter study. Ann Oncol 2001; 12: 1527–1532.
open-label phase 3 trial. Lancet Oncol 2014; 15: 997–1006. 154. Eggemann H, Ignatov A, Smith BJ et al. Adjuvant therapy with tamoxifen
139. Coleman R, Gnant M, Paterson A et al. Effects of bisphosphonate treatment on compared to aromatase inhibitors for 257 male breast cancer patients. Breast
recurrence and cause-specific mortality in women with early breast cancer: a meta- Cancer Res Treat 2013; 137: 465–470.
analysis of individual patient data from randomised trials. SABCS 2013; S4-07. 155. Korde LA, Zujewski JA, Kamin L et al. Multidisciplinary meeting on male breast
140. Eidtmann H, de Boer R, Bundred N et al. Efficacy of zoledronic acid in cancer: summary and research recommendations. J Clin Oncol 2010; 28:
postmenopausal women with early breast cancer receiving adjuvant letrozole: 2114–2122.
36-month results of the ZO-FAST study. Ann Oncol 2010; 21: 2188–2194. 156. Sousa B, Moser E, Cardoso F. An update on male breast cancer and future
141. Reid DM, Doughty J, Eastell R et al. Guidance for the management of breast directions for research and treatment. Eur J Pharmacol 2013; 717: 71–83.
cancer treatment-induced bone loss: a consensus position statement from a UK 157. Cardoso F, Bartlett J, Slaets L et al. Characterization of male breast cancer: first
Expert Group. Cancer Treat Rev 2008; 34(Suppl 1): S3–S18. results of the EORTC 10085/TBCRC/BIG/NABCG International Male BC Program’.
142. Muss HB, Berry DA, Cirrincione CT et al. Adjuvant chemotherapy in older women SABCS 2014; S6-05.
with early-stage breast cancer. N Engl J Med 2009; 360: 2055–2065. 158. Allemani C, Minicozzi P, Berrino F et al. Predictions of survival up to 10 years
143. Perrone F, Nuzzo F, Di Rella F et al. Weekly docetaxel versus CMF as adjuvant after diagnosis for European women with breast cancer in 2000–2002. Int J
chemotherapy for older women with early breast cancer: final results of the Cancer 2013; 132: 2404–2412.
randomized phase III ELDA trial. Ann Oncol 2015; 26: 675–682. 159. Park S, Koo JS, Kim MS et al. Characteristics and outcomes according to
144. Crivellari D, Gray KP, Dellapasqua S et al. Adjuvant pegylated liposomal molecular subtypes of breast cancer as classified by a panel of four biomarkers
doxorubicin for older women with endocrine nonresponsive breast cancer who using immunohistochemistry. Breast 2012; 21: 50–57.
are NOT suitable for a ‘standard chemotherapy regimen’: the CASA randomized 160. Holmes MD, Chen WY, Feskanich D et al. Physical activity and survival after
trial. Breast 2013; 22: 130–137. breast cancer diagnosis. JAMA 2005; 293: 2479–2486.
145. Staley H, McCallum I, Bruce J. Postoperative tamoxifen for ductal carcinoma in 161. Chlebowski RT, Aiello E, McTiernan A. Weight loss in breast cancer patient
situ. Cochrane Database Syst Rev 2012; 10: CD007847. management. J Clin Oncol 2002; 20: 1128–1143.
146. Rastogi P, Anderson SJ, Bear HD et al. Preoperative chemotherapy: updates of? 162. Holmberg L, Iversen OE, Rudenstam CM et al. Increased risk of recurrence after
National Surgical Adjuvant Breast and Bowel Project Protocols B-18 and B-27. J hormone replacement therapy in breast cancer survivors. J Natl Cancer Inst
Clin Oncol 2008; 26: 778–785. 2008; 100: 475–482.
147. von Minckwitz G, Schneeweiss A, Loibl S et al. Neoadjuvant carboplatin in 163. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
patients with triple-negative and HER2-positive early breast cancer (GeparSixto; among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
GBG 66): a randomised phase 2 trial. Lancet Oncol 2014; 15: 747–756. 139–144.
v | Senkus et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Published online 5 December 2016
SPECIAL ARTICLE
3rd ESO–ESMO International Consensus Guidelines

for Advanced Breast Cancer (ABC 3)
F. Cardoso1*, A. Costa2, E. Senkus3, M. Aapro4, F. André5, C. H. Barrios6, J. Bergh7, G. Bhattacharyya8,

L. Biganzoli9, M. J. Cardoso10, L. Carey11, D. Corneliussen-James12, G. Curigliano13, V. Dieras14,
N. El Saghir15, A. Eniu16, L. Fallowfield17, D. Fenech18, P. Francis19, K. Gelmon20, A. Gennari21,
N. Harbeck22, C. Hudis23, B. Kaufman24, I. Krop25, M. Mayer26, H. Meijer27, S. Mertz28, S. Ohno29,
O. Pagani30, E. Papadopoulos31, F. Peccatori32, F. Penault-Llorca33, M. J. Piccart34, J. Y. Pierga35,
H. Rugo36, L. Shockney37, G. Sledge38, S. Swain39, C. Thomssen40, A. Tutt41, D. Vorobiof42, B. Xu43,
L. Norton44 & E. Winer45
1
European School of Oncology & Breast Unit, Champalimaud Clinical Center, Lisbon, Portugal; 2European School of Oncology, Milan, Italy and European School of
Oncology, Bellinzona, Switzerland; 3Department of Oncology and Radiotherapy, Medical University of Gdansk, Gdansk, Poland; 4Breast Center, Genolier Cancer
Center, Genolier, Switzerland; 5Department of Medical Oncology, Gustave Roussy Institute, Villejuif, France; 6Department of Medicine, PUCRS School of Medicine,
Porto Alegre, Brazil; 7Department of Oncology/Radiumhemmet, Karolinska Institutet & Cancer Center Karolinska and Karolinska University Hospital, Stockholm,
Sweden; 8Department of Medical Oncology, Fortis Hospital, Kolkata, India; 9Medical Oncology Department, Hospital of Prato, Prato, Italy; 10Breast Unit,
Champalimaud Clinical Center, Lisbon, Portugal; 11Department of Hematology and Oncology, UNC Lineberger Comprehensive Cancer Center; 12METAvivor
Research and Support, Annapolis, USA; 13Division of Experimental Therapeutics, European Institute of Oncology, Milan, Italy; 14Department of Medical Oncology,
Institut Curie, Paris, France; 15NK Basile Cancer Institute Breast Center of Excellence, American University of Beirut, Beirut, Lebanon; 16Department of Breast
Tumors, Cancer Institute ‘I. Chiricuta’, Cluj-Napoca, Romania; 17Brighton & Sussex Medical School, University of Sussex, Falmer, UK; 18Breast Care Support Group,
Europa Donna Malta, Mtarfa, Malta; 19Division of Cancer Medicine, Peter MacCallum Cancer Centre, Melbourne, Australia; 20BC Cancer Agency, Vancouver Cancer
Centre, Vancouver, Canada; 21Department of Medical Oncology, Galliera Hospital, Genoa, Italy; 22Brustzentrum der Universitat München, Munich, Germany;
23
Breast Medicine Service, Memorial Sloan-Kettering Cancer Centre, New York, USA; 24Sheba Medical Center, Tel Hashomer, Israel; 25Department of Medical
Oncology, Dana-Farber Cancer Institute, Boston; 26Advanced Breast Cancer.org, New York, USA; 27Department of Radiation Oncology, Radvoud University Medical
Center, Nijmegen, The Netherlands; 28Metastatic Breast Cancer Network US, Inversness, USA; 29Breast Oncology Centre, Cancer Institute Hospital, Tokyo, Japan;
30
Oncology Institute of Southern Switzerland and Breast Unit of Southern Switzerland, Bellinzona, Switzerland; 31Europa Donna, Nicosia, Cyprus; 32European
School of Oncology, Milan, Italy and Bellinzona, Switzerland; 33Jean Perrin Centre, Comprehensive Cancer Centre, Clermont Ferrand, France; 34Department of
Medicine, Institut Jules Bordet, Université Libre de Bruxelles, Brussels, Belgium; 35Department of Medical Oncology, Institut Curie-Université Paris Descartes, Paris,
France; 36Department of Medicine, Breast Oncology Program, UCSF Helen Diller Family Comprehensive Cancer Center, San Francisco; 37Department of Surgery
and Oncology, Johns Hopkins Breast Center, Baltimore; 38Indiana University Medical CTR, Indianapolis; 39Lombardi Comprehensive Cancer Center, Georgetown
University, Washington, USA; 40Department of Gynaecology, Martin-Luther-University Halle-Wittenberg, Halle an der Saale, Germany; 41Breakthrough Breast
Cancer Research Unit, King’s College London and Guy’s and St Thomas’s NHS Foundation Trust, London, UK; 42Sandton Oncology Centre, Johannesburg, South
Africa; 43Department of Medical Oncology, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, China; 44Breast Cancer Program,
Memorial Sloan-Kettering Cancer Centre, New York; 45Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, USA
*Correspondence to: Fatima Cardoso, Dr Breast Unit, Champalimaud Clinical Center, Av. De Brasılia s/n, 1400-038 Lisbon, Portugal. Tel: þ351-210-480-004; E-mail:
fatimacardoso@fundacaochampalimaud.pt
Note: These Guidelines were developed by ESO and ESMO and are published simultaneously in Annals of Oncology (2016; doi:10.1093/annonc/mdw544) and
The Breast and should both be cited.
Introduction improved outcomes, quality of life, awareness and information

Advanced Breast Cancer (ABC) comprises both locally advanced regarding ABC.
(LABC) and metastatic breast cancer (MBC) [1]. Although treat- The level of evidence used to base many recommendations re-
able, MBC remains an incurable disease with a median overall mains low, and more and better designed trials are needed to ad-
survival of 2–3 years and a 5-year survival of only 25% [2–4]. dress clinically important questions. An improved understanding
Some more recent series seem to indicate an improvement in me- of the biology of ABC, its heterogeneity, and of the mechanisms
dian overall survival [5, 6]. of resistance to the different types of therapies is being acquired
A recent comprehensive report [2] of the advances in this field and it is anticipated that the application of new technologies,
in the last decade shows that progress has been slow in terms of such as next generation sequencing, patient xenographs, systems
V
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs licence (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-
commercial reproduction and distribution of the work, in any medium, provided the original work is not altered or transformed in any way, and that the work properly cited. For commercial re-use, please
contact journals.permissions@oup.com
by guest
on 05 February 2018
Annals of Oncology Special article
Table 1. Grading system [7]
Grade of recommendation/ Benefit versus risk and burdens Methodological quality of sup- Implications
description porting evidence
1A/Strong recommendation, high Benefits clearly outweigh risk and RCTs without important limitations Strong recommendation, can
quality evidence burdens, or vice versa or overwhelming evidence from apply to most patients in most
observational studies circumstances without
reservation
1B/Strong recommendation, mod- Benefits clearly outweigh risk and RCTs with important limitations (in- Strong recommendation, can
erate quality evidence burdens, or vice versa consistent results, methodological apply to most patients in most
flaws, indirect, or imprecise) or ex- circumstances without
ceptionally strong evidence from reservation
observational studies
1C/Strong recommendation, low Benefits clearly outweigh risk and Observational studies or case series Strong recommendation, but may
quality evidence burdens, or vice versa change when higher quality evi-
dence becomes available
2A/Weak recommendation, high Benefits closely balanced with risks RCTs without important limitations Weak recommendation, best ac-
quality evidence and burden or overwhelming evidence from tion may differ depending on
observational studies circumstances or patients’ or so-
cietal values
2B/Weak recommendation, mod- Benefits closely balanced with risks RCTs with important limitations (in- Weak recommendation, best ac-
erate quality evidence and burden consistent results, methodological tion may differ depending on
flaws, indirect, or imprecise) or ex- circumstances or patients’ or so-
ceptionally strong evidence from cietal values
observational studies
2C/Weak recommendation, low Benefits closely balanced with risks Observational studies or case series Very weak recommendation, other
quality evidence and burden alternatives may be equally
reasonable
SECTION I. GENERAL RECOMMENDATIONS
GUIDELINE STATEMENT LoE Consensus
The ABC community strongly calls for clinical trials addressing important unanswered clinical questions in this set- Expert opinion Voters: 43
ting, and not just for regulatory purposes. Clinical trials should continue to be performed, even after approval Yes: 100%
of a new treatment, providing real world performance of the therapy.
Every advanced breast cancer patient must have access to optimal cancer treatment and supportive care accord- Expert opinion Voters: 44
ing to the highest standards of patient centered care, as defined by: Yes: 100%
Open communication between patients and their cancer care teams as a primary goal.
Educating patients about treatment options and supportive care, through development and dissemination of
evidence-based information in a clear, culturally appropriate form.
Encouraging patients to be proactive in their care and to share decision-making with their health care
providers.
Empowering patients to develop the capability of improving their own quality of life within their cancer
experience.
Always taking into account patient preferences, values and needs as essential to optimal cancer care.
We strongly recommend the use of objective scales, such as the ESMO Magnitude of Clinical Benefit Scale or the Expert opinion Voters: 40
ASCO Value Framework, to evaluate the real magnitude of benefit provided by a new treatment and help pri- Yes: 87.5% (35)
oritize funding, particularly in countries with limited resources. Abstain: 5% (2)
The use of telemedicine oncology to help management of patients with ABC living in remote places, is an import- Expert opinion Voters: 42
ant option to consider when geographic distances are a problem and provided that issues of connectivity are Yes: 92.8% (39)
solved. Abstain: 4.7% (2)
Continued
Volume 28 | Issue 1 | 2017 doi:10.1093/annonc/mdw544 | 17

by guest
on 05 February 2018
Special article Annals of Oncology
SECTION I Continued
Strong consideration should be given to the use of validated PROMs (patient-reported outcome measures) for pa- 1C Voters: 39
tients to record the symptoms of disease and side effects of treatment experienced as a regular part of clinical Yes: 87.1% (34)
care. These PROMs should be simple, and user-friendly to facilitate their use in clinical practice, and thought Abstain: 5.1% (2)
needs to be given to the easiest collection platform, e.g. tablets or smartphones. Systematic monitoring would
facilitate communication between patients and their treatment teams by better characterizing the toxicities of
all anticancer therapies. This would permit early intervention of supportive care services enhancing quality of
life
As survival is improving in many patients with ABC, consideration of survivorship issues should be part of the rou- Expert opinion Voters: 40
tine care of these patients. Health professionals should therefore be ready to change and adapt treatment strat- Yes: 95% (38)
egies to disease status, treatment adverse effects and quality of life, patients’ priorities and life plans. Abstain: 5% (2)
Attention to chronic needs for home and family care, job and social requirements, should be incorporated in the
treatment planning and periodically updated.
ABC patients who desire to work or need to work for financial reasons should have the opportunity to do so, with Expert opinion Voters: 42
needed and reasonable flexibility in their working schedules to accommodate continuous treatment and hos- Yes: 100%
pital visits.
ABC patients with stable disease, being treated as a ‘chronic condition’, should have the option to undergo breast Expert opinion Voters: 39
reconstruction. Yes: 82% (32)
Abstain: 7.6% (3)
In ABC patients with long-standing stable disease, screening breast imaging should be an option. Expert opinion Voters: 40
Yes: 52.5% (21)
N: 47.5% (19)
Breast imaging should also be performed when there is a suspicion of loco-regional progression. Expert opinion Voters: 40
Yes: 100%
A biopsy (preferably providing histology) of a metastatic lesion should be performed, if easily accessible, to con- 1B Voters: 43
firm diagnosis particularly when metastasis is diagnosed for the first time. Yes: 98% (42)
Biological markers (especially HR and HER-2) should be reassessed at least once in the metastatic setting, if clinic- 1B Voters: 44
ally feasible. Depending on the metastatic site (e.g. bone tissue), technical considerations need to be discussed Yes: 98% (43)
with the pathologist.
If the results of tumour biology in the metastatic lesion differ from the primary tumor, it is currently unknown Expert Opinion 87%
which result should be used for treatment-decision making. Since a clinical trial addressing this issue is difficult
to undertake, we recommend considering the use of targeted therapy (ET and/or anti-HER-2 therapy) when re-
ceptors are positive in at least one biopsy, regardless of timing.
To date, the removal of the primary tumor in patients with de novo stage IV breast cancer has not been associated 2B Voters: 44
with prolongation of survival, with the possible exception of the subset of patients with bone only disease. Yes: 70.4% (31)
However, it can be considered in selected patients, particularly to improve quality of life, always taking into ac-
count the patient’s preferences. Of note, some studies suggest that surgery is only valuable if performed with the
same attention to detail (e.g. complete removal of the disease) as in patients with early stage disease.
Additional prospective clinical trials evaluating the value of this approach, the best candidates and best timing are
currently ongoing
A small but very important subset of patients with ABC, for example those with oligo-metastatic disease or low vol- Expert opinion Voters: 43
ume metastatic disease that is highly sensitive to systemic therapy, can achieve complete remission and a long Yes: 91% (39)
survival.
A multimodal approach, including local-regional treatments with curative intent, should be considered for these
selected patients.
LoE, available level of evidence; consensus, percentage of panel members in agreement with the statement.
biology, and computer modelling, among others, will accelerate The ABC guidelines are developed as a joint effort from ESO
advances. (European School of Oncology) and ESMO (European Society of
Aiming at providing clinically oriented guidelines on how to Medical Oncology), and are endorsed by EUSOMA (European
best manage ABC, the 3rd International Consensus Conference Society of Breast Cancer Specialists), ESTRO (European Society
for Advanced Breast Cancer (ABC 3) took place in Lisbon, of Radiation Oncology), UICC (Union for International Cancer
Portugal on November 5th–7th, 2015, bringing together over Control), SIS (Senologic International Society) and FLAM
1100 participants from 84 countries, including health profes- (Federation LatinoAmericana de Mastologia). There was also of-
sionals, patient advocates and journalists. ficial representation of ASCO (American Society of Clinical
18 | Cardoso et al. Volume 28 | Issue 1 | 2017

by guest
on 05 February 2018
Table 2. Other ABC1 [10] and ABC2 [1] statements with only minor updates or with no updates
Recommendations LoE % Consensus
ABC IMPORTANT DEFINITIONS

VISCERAL CRISIS is defined as severe organ dysfunction as assessed by signs and symptoms, laboratory studies, Expert opinion 95
and rapid progression of disease. Visceral crisis is not the mere presence of visceral metastases but implies import-
ant visceral compromise leading to a clinical indication for a more rapidly efficacious therapy, particularly since an-
other treatment option at progression will probably not be possible.
PRIMARY ENDOCRINE RESISTANCE is defined as: Relapse while on the first 2 years of adjuvant ET, or PD within Expert opinion 67
first 6 months of 1st line ET for MBC, while on ET.
SECONDARY (ACQUIRED) ENDOCRINE RESISTANCE is defined as: Relapse while on adjuvant ET but after the first
2 years, or Relapse within 12 months of completing adjuvant ET, or PD 6 months after initiating ET for MBC, while
on ET.
Note: resistance is a continuum and these definitions help mainly clinical trials and not necessarily clinical practice
GENERAL STATEMENTS
The management of ABC is complex and, therefore, involvement of all appropriate specialties in a multidisciplinary Expert opinion 100
team (including but not restricted to medical, radiation, surgical oncologists, imaging experts, pathologists, gyne-
cologists, psycho-oncologists, social workers, nurses and palliative care specialists), is crucial.
From the time of diagnosis of ABC, patients should be offered appropriate psychosocial care, supportive care, and Expert opinion 100
symptom-related interventions as a routine part of their care. The approach must be personalized to meet the
needs of the individual patient.
Following a thorough assessment and confirmation of MBC, the potential treatment goals of care should be dis- Expert opinion 97
cussed. Patients should be told that MBC is incurable but treatable, and that some patients can live with MBC for ex-
tended periods of time (many years in some circumstances).
This conversation should be conducted in accessible language, respecting patient privacy and cultural differences,
and whenever possible, written information should be provided.
Patients (and their families, caregivers or support network, if the patient agrees) should be invited to participate in Expert opinion 100
the decision-making process at all times. When possible, patients should be encouraged to be accompanied by
persons who can support them and share treatment decisions (e.g. family members, caregivers, support network).
There are few proven standards of care in ABC management. After appropriate informed consent, inclusion of pa- Expert opinion 100
tients in well-designed, prospective, independent trials must be a priority whenever such trials are available and
the patient is willing to participate.
The medical community is aware of the problems raised by the cost of ABC treatment. Balanced decisions should Expert opinion 100
be made in all instances; patients’ well-being, length of life and preferences should always guide decisions.
Specialized oncology nurses (if possible specialized breast nurses) should be part of the multidisciplinary team Expert opinion 92
managing ABC pts. In some countries this role may be played by a physician assistant or another trained and speci-
alized health care practitioner.
All ABC patients should be offered comprehensive, culturally sensitive, up-to-date and easy to understand informa- 1B 97
tion about their disease and its management.
The age of the patient should not be the sole reason to withhold effective therapy (in elderly patients) nor to over- 1B 100
treat (in young patients). Age alone should not determine the intensity of treatment.
ASSESSMENT GUIDELINES
Minimal staging workup for MBC includes a history and physical examination, hematology and biochemistry tests, 2C 67
and imaging of chest, abdomen and bone.
Brain imaging should not be routinely performed in asymptomatic patients. This approach is applicable to all pa- Expert opinion 94
tients with MBC including those patients with HER-2þ and/or TNBC MBC.
The clinical value of tumor markers is not well established for diagnosis or follow-up after adjuvant therapy, but 2C 89
their use is reasonable (if elevated) as an aid to evaluate response to treatment, particularly in patients with non-
measurable metastatic disease. A change in tumor markers alone should not be used to initiate a change in
treatment.
Evaluation of response to therapy should generally occur every 2–4 months for ET or after two to four cycles for CT, Expert opinion 81
depending on the dynamics of the disease, the location and extent of metastatic involvement, and type of
treatment.
Imaging of target lesions may be sufficient in many patients. In certain patients, such as those with indolent disease,
less frequent monitoring is acceptable.
Additional testing should be performed in a timely manner, irrespective of the planned intervals, if PD is suspected
or new symptoms appear. Thorough history and physical examination must always be performed.
Continued

by guest
on 05 February 2018
Table 2 Continued
TREATMENT GENERAL GUIDELINES

Treatment choice should take into account at least these factors: HR and HER-2 status, previous therapies and tox- Expert opinion 100
icities, disease-free interval, tumour burden (defined as number and site of metastases), biological age, perform-
ance status, co-morbidities (including organ dysfunctions), menopausal status (for ET), need for a rapid disease/
symptom control, socio-economic and psychological factors, available therapies in the patient’s country and pa-
tient preference.
ER 1/HER-2 NEGATIVE ABC
Endocrine treatment after CT (maintenance ET) to maintain benefit is a reasonable option, although this approach 1C 88
has not been assessed in randomized trials.
Concomitant CTþET has not shown a survival benefit and should not be performed outside of a clinical trial. 1B 100
CHEMOTHERAPY AND BIOLOGICAL THERAPY
Both combination and sequential single agent CT are reasonable options. Based on the available data, we recom- 1B 96
mend sequential monotherapy as the preferred choice for MBC. Combination CT should be reserved for patients
with rapid clinical progression, life-threatening visceral metastases, or need for rapid symptom and/or disease
control
In the absence of medical contraindications or patient concerns, anthracycline or taxane based regimens, prefer- 1A 71
ably as single agents, would usually be considered as first line CT for HER-2 negative MBC, in those patients who
have not received these regimens as (neo)adjuvant treatment and for whom chemotherapy is appropriate. Other
options are, however, available and effective, such as capecitabine and vinorelbine, particularly if avoiding alopecia
is a priority for the patient.
In patients with taxane-naive and anthracycline-resistant MBC or with anthracycline maximum cumulative dose or 1A 59
toxicity (i.e. cardiac) who are being considered for further CT, taxane-based therapy, preferably as single agents,
would usually be considered as treatment of choice. Other options are, however, available and effective, such as
capecitabine and vinorelbine, particularly if avoiding alopecia is a priority for the patient.
In patients pre-treated (in the adjuvant and/or metastatic setting) with an anthracycline and a taxane, and who do 1B 77
not need combination CT, single agent capecitabine, vinorelbine or eribulin are the preferred choices. Additional
choices include gemcitabine, platinum agents, taxanes, and liposomal anthracyclines. The decision should be indi-
vidualized and take into account different toxicity profiles, previous exposure, patient preferences, and country
availability.
If given in the adjuvant setting, a taxane can be re-used as 1st line therapy, particularly if there has been at least 1A 92
1 year of disease-free survival.
Duration of each regimen and the number of regimens should be tailored to each individual patient. Expert opinion 96
Usually each regimen (except anthracyclines) should be given until progression of disease or unacceptable toxicity. 1B 72
What is considered unacceptable should be defined together with the patient.
OTHER AGENTS
Bevacizumab combined with a chemotherapy as 1st or 2nd line therapy for MBC provides only a moderate benefit 1A 74
in PFS and no benefit in OS. The absence of known predictive factors for bevacizumab efficacy renders recommen-
dations on its use difficult. Bevacizumab can only therefore be considered as an option in selected cases in these
settings and is not recommended after 1st/2nd line.
SPECIFIC POPULATIONS: TREATMENT OF METASTATIC MALE MBC
For ERþ Male MBC, which represents the majority of the cases, ET is the preferred option, unless there is concern Expert opinion 100
or proof of endocrine resistance or rapidly progressive disease needing a fast response.
For ERþ Male MBC tamoxifen is the preferred option. Expert opinion 83
For male patients with MBC who need to receive an AI, a concomitant LHRH agonist or orchidectomy is the pre- Expert opinion 86
ferred option. AI monotherapy may also be considered, with close monitoring of response.
Clinical trials are needed in this patient population.
SPECIFIC SITES OF METASTASES
BONE METASTASES
Radiological assessments are required in patients with persistent and localized pain due to bone metastases to de- 1A 96
termine whether there are impending or actual pathological fractures. If a fracture of a long bone is likely or has
occurred, an orthopaedic assessment is required as the treatment of choice may be surgical stabilization, which is
generally followed by RT. In the absence of a clear fracture risk, RT is the treatment of choice.
Neurological symptoms and signs which suggest the possibility of spinal cord compression must be investigated 1B 100
as a matter of urgency. This requires a full radiological assessment of potentially affected area as well as adjacent
areas of the spine. MRI is the method of choice. An emergency surgical opinion (neurosurgical or orthopaedic) may
be required for surgical decompression. If no decompression/stabilization is feasible, emergency radiotherapy is
the treatment of choice and vertebroplasty is also an option.
Continued

by guest
on 05 February 2018
Table 2 Continued
BRAIN METASTASES
Patients with a single or small number of potentially resectable brain metastases should be treated with surgery or 1B 92
radiosurgery. Radiosurgery is also an option for some unresectable brain metastases.
If surgery/radiosurgery is performed it may be followed by whole brain radiotherapy but this should be discussed 1B 72
in detail with the patient, balancing the longer duration of intracranial disease control and the risk of neurocogni-
tive effects.
Because patients with HER2þve MBC and brain metastases can live for several years, consideration of long-term 1C 89
toxicity is important and less toxic local therapy options (e.g. stereotactic RT) should be preferred to whole brain
RT, when available and appropriate (e.g. in the setting of a limited number of brain metastases).
LIVER METASTASES
Prospective randomized clinical trials of local therapy for BC liver metastases are urgently needed, since available Expert opinion 83
evidence comes only from series in highly selected patients. Since there are no randomized data supporting the ef-
fect of local therapy on survival, every patient must be informed of this when discussing a potential local therapy
technique. Local therapy should only be proposed in very selected cases of good performance status, with limited
liver involvement, no extra-hepatic lesions, after adequate systemic therapy has demonstrated control of the dis-
ease. Currently, there are no data to select the best technique for the individual patient (surgery, stereotactic RT,
intra-hepatic CT. . .).
MALIGNANT PLEURAL EFFUSIONS
Malignant pleural effusions require systemic treatment with/without local management. Thoracentesis for diagno- 2B 86
sis should be performed if it is likely that this will change clinical management. False negative results are common.
Drainage is recommended in patients with symptomatic, clinically significant pleural effusion. Use of an intrapleural
catheter or intrapleural administration of talc or drugs (e.g. bleomycin, biological response modifiers) can be help-
ful. Clinical trials evaluating the best technique are needed.
CHEST WALL AND REGIONAL (NODAL) RECURRENCES
Due to the high risk of concomitant distant metastases, patients with chest wall or regional (nodal) recurrence Expert opinion 100
should undergo full restaging, including assessment of chest, abdomen and bone.
Chest wall and regional recurrences should be treated with surgical excision when feasible with limited risk of 1B 97
morbidity.
Locoregional radiotherapy is indicated for patients not previously irradiated. 1B 97
For patients previously irradiated, re-irradiation of all or part of the chest wall may be considered in selected cases. Expert opinion 97
In addition to local therapy (surgery and/or RT), in the absence of distant metastases, the use of systemic therapy 1B 95
(CT, ET and/or anti-HER-2 therapy) should be considered.
CT after first local or regional recurrence improves long-term outcomes primarily in ER negative disease. ET in this
setting improves long-term outcomes for ER positive disease.
The choice of systemic treatment depends on tumor biology, previous treatments, length of disease free interval,
and patient-related factors (co-morbidities and preferences).
In patients with disease not amenable to radical local treatment, the choice of palliative systemic therapy should be Expert opinion 97
made according to principles previously defined for metastatic BC.
These patients may still be considered for palliative local therapy.
SUPPORTIVE AND PALLIATIVE CARE
Supportive care allowing safer and more tolerable delivery of appropriate treatments should always be part of the 1A 100
treatment plan.
Early introduction of expert palliative care, including effective control of pain and other symptoms, should be a 1A 100
priority.
Access to effective pain treatment (including morphine, which is inexpensive) is necessary for all patients in need 1A 100
of pain relief.
Optimally, discussions about patient preferences at the end of life should begin early in the course of metastatic Expert 96
disease. However, when active treatment no longer is able to control widespread and life-threatening disease, and Opinion
the toxicities of remaining options outweigh benefits, physicians and other members of the healthcare team
should initiate discussions with the patient (and family members/friends, if the patient agrees) about end-of-life
care.
ABC STATEMENTS FOR LABC (Note: For the purpose of these recommendations, LABC means inoperable, non-metastatic locally advanced
breast cancer)
Before starting any therapy, a core biopsy providing histology and biomarker (ER, PR, HER-2, proliferation/grade) ex- 1B 97
pression is indispensable to guide treatment decisions.
1B 100
Continued

by guest
on 05 February 2018
Table 2 Continued
Since LABC patients have a significant risk of metastatic disease, a full staging workup, including a complete history,
physical examination, lab tests and imaging of chest and abdomen (preferably CT) and bone, prior to initiation of
systemic therapy is highly recommended.
PET-CT, if available, may be used (instead of and not on top of CTs and bone scan). 2B 100
Systemic therapy (not surgery or RT) should be the initial treatment. Expert opinion 100
If LABC remains inoperable after systemic therapy and eventual radiation, ‘palliative’ mastectomy should not be
done, unless the surgery is likely to result in an overall improvement in quality of life.
A combined treatment modality based on a multidisciplinary approach (systemic therapy, surgery and radiother- 1A 100
apy) is strongly indicated in the vast majority of cases.
For Triple Negative LABC, Anthracycline- and-taxane-based chemotherapy is recommended as initial treatment. 1A 85
For HER-21 LABC, concurrent taxane and anti-HER-2 therapy is recommended since it increases the rate of pCR. 1A 92
For HER-21 LABC, anthracycline-based chemotherapy should be incorporated in the treatment regimen. 1A 72
When an anthracycline is given, it should be administered sequentially with the anti-HER-2 therapy. 1A 87
Options for HR1 LABC include an anthracycline- and taxane-based chemotherapy regimen, or endocrine therapy. 1A 85
The choice of CT versus ET, as initial treatment, will depend on tumor (grade, biomarker expression) and patient Expert 85
(menopausal status, performance status, comorbidities, preference) considerations. Opinion
Following effective neoadjuvant systemic therapy with or without radiotherapy, surgery will be possible in many 2B 98
patients. This will consist of mastectomy with axillary dissection in the vast majority of cases, but in selected pa-
tients with a good response, breast conserving surgery may be possible.
INFLAMMATORY LABC
For inflammatory LABC, overall treatment recommendations are similar to those for non-inflammatory LABC, with 1B 93
systemic therapy as first treatment.
Mastectomy with axillary dissection is recommended in almost all cases, even when there is good response to pri- IB 95
mary systemic therapy.
Immediate reconstruction is generally not recommended in patients with inflammatory LABC. Expert opinion 95
Loco-regional radiotherapy (chest wall and lymph nodes) is required, even when a pCR is achieved with systemic 1B 98
therapy.
Oncology) in the consensus panel. The ABC Conference was also section, were not voted on during the consensus session, but
organized under the auspices of OECI (Organization of discussed and unanimously agreed by email, and are con-
European Cancer Institutes), and with the support of the BCRF sidered to have 100% agreement.
(Breast Cancer Research Foundation) and the Susan G Komen Supplementary Table S1, available at Annals of Oncology on-
for the Cure. line, lists all members of the ABC 3 consensus panel and their
The present article summarizes the guidelines developed at disclosures of any relationships with the pharmaceutical indus-
ABC3 and is supported with the level of evidence, the percentage try that could be perceived as a potential conflict of interest.
of consensus reached at the Conference, and supporting Table 1 describes the grading system used [7]. ABC1 [10] and
references. ABC2 [1] statements with only minor updates or with no updates
are listed in Table 2.
Methodology
Prior to the ABC 3 Conference, a set of preliminary recom-
mendation statements on the management of ABC were pre- General recommendations
pared, based on available published data and following the The continuous increase in cancer care costs has inevitably led to
ESMO guidelines methodology. These recommendations were inequalities in access both between countries and within each
circulated to all 44 panel members by email for comments and country. Cost, value and access are now central discussion points
corrections on content and wording. A final set of recommen- and important factors in treatment-decision making. Both
dations was presented, discussed and voted upon during the ESMO and ASCO have put considerable effort into the develop-
consensus session of ABC 3. All panel members were in- ment of validated objective scales aiming at evaluating the real
structed to vote on all questions, with members with a poten- magnitude of benefit provided by each new treatment, including
tial conflict of interest or who did not feel comfortable efficacy measures (e.g. impact on DFS, OS or PFS) and toxicity/
answering the question (e.g. due to lack of expertise in a par- quality of life measures. The ESMO Magnitude of Clinical Benefit
ticular field) instructed to vote ‘abstain’. Additional changes in Scale [8] and the ASCO Value Framework [9] are user-friendly
the wording of statements were made during the session. The tools that can greatly assist decision-makers at the country and/or
statements related to management of side effects and difficult hospital level in the difficult decisions regarding approval and
symptoms, included under the Supportive and Palliative care reimbursement.

by guest
on 05 February 2018
SECTION 2. ABC IMPORTANT DEFINITIONS
OLIGO-METASTATIC DISEASE is defined as low volume metastatic disease with limited number and size of Expert opinion Voters: 36
metastatic lesions (up to five and not necessarily in the same organ), potentially amenable for local treat- Yes: 78% (28)
ment, aimed at achieving a complete remission status. Abstain: 6% (2)
PATIENTS WITH MULTIPLE CHRONIC CONDITIONS are defined as patients with additional comorbidities Expert opinion Voters: 42
(e.g. cardiovascular, impaired renal or liver function, autoimmune disease) making it difficult to account for Yes:100%
all of the possible extrapolations to develop specific recommendations for care.
LoE, available level of evidence; consensus, percentage of panel members in agreement with the statement.
The ABC3 experts also emphasize the responsibility of the aca- imaging should be performed in case of suspicion of disease pro-
demic and medical communities to advance the knowledge on gression in the breast.
breast cancer and other relevant unanswered issues, by involve- Regarding the need to biopsy metastatic disease and re-evaluate
ment in clinical research aimed at addressing important clinical the common biomarkers, the ABC recommendations had only
questions, and not only in studies conducted for regulatory minor changes. There are situations where the need for a biopsy in
purposes. the metastatic setting is very clear, such as single lesions, history of
The importance of providing patients with full information two or more malignancies, suspicion of benign histology or doubt
in appropriate, understandable and culturally sensitive way, as between progression or post-treatment necrosis. There is also con-
well as involving them in sharing the decision-making regard- sensus regarding the importance of such biopsy in situations where
ing all aspects of their management has been repeatedly stressed when a change in biomarkers would impact the treatment choice,
in all ABC guidelines [1, 10]. A high standard of patient centred which would mainly occur when biomarkers were negative in the
care includes the following elements: appropriate information, primary tumor. There is some controversy about the benefits of a
good communication with health professionals, patient educa- biopsy in situations where there is no doubt about the nature of
tion, proactive advocacy, sensitivity to the patient’s prefer- the lesion(s) and where all receptors were positive in the primary
ences, values and needs, and providing patients with the tumor, since the clinical implementation of new technologies such
capabilities to improve their own quality of life [11]. as next generation sequencing for management decision-making s
Although the overall survival of ABC has remained stable, for not yet validated. However, the exact nature of a lesion is hard to
some subtypes, and in particular HER-2-positive metastatic ascertain without the confirmation by a biopsy as shown in some
breast cancer, prolonged survival, well beyond the median 2– retrospective and prospective studies [13–15]. There is also an
3 years, has become a frequent reality. For these long-term sur- undisputable importance of collection of material for research
vivors, survivorship issues which are specific for advanced cancer purposes, both ongoing and future.
patients, have emerged and need appropriate attention, research Technical issues should be discussed with the breast patholo-
and management. Work-related issues are central and solutions gist, in particular in case of bone biopsies with the inherent decal-
not easy to implement. A recently published survey [12], found cification problems, which may interfere with the biomarker
that approximately half of the women in employment had to analysis [16, 17], as experienced in Safir01/UNICANCER trial
change their work situation due to ABC and that 37% of them [18]. For that reason, decalcification using EDTA is recom-
had to give up work temporarily or permanently. Due to these in- mended for bone biopsies, when it is the only metastatic site [17].
come problems and those related to the cost of care, the same sur- Adding to the complexity of this issue is the fact that negative bio-
vey found that 56% of ABC patients experienced a decline in marker results may limit the eligibility for reimbursement of
household income as a result of their disease. The ABC commu- therapies dedicated to specific subtypes, in some countries.
nity strongly advocates for the right of ABC patients to return or A number of prospective randomized trials have assessed or are
maintain their work, since a substantial proportion of these pa- assessing the role of removing the primary tumor in patients with
tients are in their most productive years. Furthermore, in some de novo metastatic disease. So far only two small studies have been
countries, health coverage is dependent on being employed. For published/presented [19, 20]. A subgroup analysis of the Turkish
that to occur, we need flexibility of working schedules, new com- study suggested a potential benefit in patients with ER/PgRþ,
munication technologies and home-based work which the ABC HER-2 negative, solitary bone metastasis, who are younger than 55
community supports. In many countries this may imply a change years of age, while patients with multiple pulmonary and liver me-
in the current labour-related laws. tastasis did worse with an overall 3-year survival of 31% in the sur-
Survivorship issues also include the potential discussion of gery group versus 67% for the systemic therapy group [20]. In the
breast reconstruction, in those cases where the metastatic disease Indian trial, a decrease in distant progression-free survival was
is either in complete remission or in a durable stable situation. observed in patients allocated to surgery. Results of larger, pro-
No consensus could be reached regarding the use of breast imag- spective studies are awaited. Until then, the recommendation is to
ing to follow-up the unaffected breast, but the experts agreed that discuss surgery on a case-by-case basis and importantly, only

by guest
on 05 February 2018
SECTION 3. HER-2 POSITIVE ABC
Anti-HER-2 therapy should be offered early (as 1st line) to all patients with HER-2þ ABC, except in the pres- 1A Voters: 43
ence of contra-indications to the use of such therapy Yes: 98% (42)
For highly selected patients* with ERþ/HER-2þ MBC, for whom ET is chosen over CT, ET should be given in 1A Voters: 43
combination with anti-HER-2 therapy (either trastuzumab or lapatinib) since the combination provides PFS Yes: 72% (31)
benefit (i.e. ‘time without CT’) compared to ET alone. The addition of anti-HER-2 therapy to ET in the 1st line Abstain: 9% (4)
setting has not led to a survival benefit but long-term follow-up was not collected in the available trials. In
addition, this strategy is currently being directly compared with CTþanti-HER2 therapy. (*see definition in
text)
For patients with ERþ/HER-2þ MBC, for whom CTþanti-HER2 therapy was chosen as 1st line therapy and pro- 1C Voters: 39
vided a benefit, it is reasonable to use ETþanti-HER2 therapy as maintenance therapy, after stopping CT, al- Yes: 79% (31)
though this strategy has not been studied in randomized trials. Abstain: 10% (4)
Patients progressing on an anti-HER-2 therapy combined with a cytotoxic or endocrine agent should be 1B Voters: 43
offered additional anti-HER-2 therapy with subsequent treatment since it is beneficial to continue suppres- Yes: 91% (39)
sion of the HER-2 pathway. The optimal duration of anti-HER-2 therapy for MBC (i.e. when to stop these Abstain: 7% (3)
agents) is currently unknown.
In patients achieving a complete remission, the optimal duration of maintenance anti-HER2 therapy is un- Expert Opinion Voters: 42
known and needs to be balanced against treatment toxicity, logistical burden and cost. Stopping anti-HER2 Yes: 93% (39)
therapy after several years of sustained complete remission may be considered in some patients, particu- No: 7% (3)
larly if treatment re-challenge is available in case of progression.
Patients who have received any type of (neo)adjuvant anti-HER-2 therapy should not be excluded from clinical 1B Voters: 40
trials for HER-2þ MBC. These patients remain candidates for anti-HER-2 therapies. Yes: 100%
In the 1st line setting, for HER-2þ MBC previously treated (in the adjuvant setting with DFI >12 months) or un- 1A Voters: 44
treated with trastuzumab, combinations of CTþtrastuzumab are superior to combinations of CTþlapatinib Yes: 95% (42)
in terms of PFS and OS. Abstain: 5% (2)
The standard 1st line therapy for patients previously untreated with anti-HER-2 therapy is the combination of 1A Voters: 42
CTþtrastuzumab and pertuzumab, because it has proven to be superior to CTþtrastuzumab in terms of OS Yes: 86% (36)
in this population. Abstain: 12% (5)
For patients previously treated (in the (neo)adjuvant setting) with anti-HER-2 therapy, the combination of 1A Voters: 41
CTþtrastuzumab and pertuzumab is an important option for 1st line therapy. Few (88) of these patients Yes: 76% (31)
were treated in the Cleopatra trial and all with trastuzumab-free interval >12 months. Abstain: 22% (9)
There are currently no data supporting the use of dual blockade with trastuzumabþpertuzumab and CT be- 1 A (against its use) Voters: 43
yond progression (i.e. continuing dual blockade beyond progression) and therefore this 3 drug regimen Yes: 86% (37)
should not be given beyond progression outside clinical trials. Abstain: 9% (4)
In a HER-2þ MBC patient, previously untreated with the combination of CTþtrastuzumabþpertuzumab, it is Expert Opinion Voters: 37
acceptable to use this treatment after 1st line. Yes: 76% (28)
Abstain: 16% (6)
After 1st line trastuzumab-based therapy, T-DM1 provides superior efficacy relative to other HER-2-based thera- 1A Voters: 42
pies in the 2nd line (versus lapatinibþcapecitabine) and beyond (versus treatment of physician’s choice). Yes: 88% (37)
T-DM1 should be preferred in patients who have progressed through at least 1 line of trastuzumab-based ther- Abstain: 129% (5)
apy, because it provides an OS benefit.
However, there are no data on the use of T-DM1 after dual blockade with trastuzumabþpertuzumab.
In case of progression on trastuzumab-based therapy, the combination trastuzumabþlapatinib is a reasonable 1B Voters: 43
treatment option for some patients. There are however, no data on the use of this combination after pro- Yes: 84% (36)
gression on pertuzumab or T-DM1. Abstain: 12% (5)
All patients with HER-2þ MBC who relapse after adjuvant or any line metastatic anti-HER-2 therapy should be 1B Voters: 40
considered for further anti-HER-2 therapy, except in the presence of contraindications. The choice of the Yes: 86% (36)
anti-HER-2 agent will depend on country-specific availability, the specific anti-HER-2 therapy previously ad- Abstain: 12.5% (5)
ministered, and the relapse free interval. The optimal sequence of all available anti-HER-2 therapies is cur-
rently unknown.
Regarding the CT component of HER-2 positive MBC treatment:
When pertuzumab is not given, 1st line regimens for HER-2 MBC can include trastuzumab combined with 1A Voters: 41
vinorelbine or a taxane. Differences in toxicity between these regimens should be considered and discussed Yes: 88% (36)
with the patient in making a final decision. Abstain: 10% (4)
Other CT agents can be administered with trastuzumab but are not as well studied and are not preferred.
For later lines of therapy, trastuzumab can be administered with several CT agents, including but not limited 2A Voters: 43
to, vinorelbine (if not given in 1st line), taxanes (if not given in 1st line), capecitabine, eribulin, liposomal Yes: 91% (39)
Continued

by guest
on 05 February 2018
SECTION 3 Continued
anthracyclines, platinum, gemcitabine, or metronomic CM. The decision should be individualized and take Abstain: 9% (4)
into account different toxicity profiles, previous exposure, patient preferences, and country availability.
CT agents to combine with a dual blockade of trastuzumabþpertuzumab are docetaxel (LoE: 1A) or paclitaxel See in statement Voters: 43
(LoE: 1B). Also possible are vinorelbine (LoE: 2 A), nab-paclitaxel (LoE: 2B) and capecitabine (LoE: 2A). Yes: 86% (37)
Abstain: 11.6% (5)
HER-21 ABC and BRAIN METASTASES
In patients with HER-2-positive ABC with brain metastases and stable extracranial disease, systemic therapy 1C Voters: 42
should not be changed. Yes: 95% (40)
Abstain: 5% (2)
For patients with HER-2-positive cancers where brain metastases are the only site of recurrence, the addition 1C Voters: 42
of CT to local therapy is not known to alter the course of the disease. It is recommended to re-start the Y: 83% (35)
anti-HER-2 therapy (trastuzumab) if this had been stopped. A: 7% (3)
LoE, available level of evidence; consensus, percentage of panel members in agreement with the statement; ET, endocrine therapy; CT, chemotherapy;
DFI, disease-free interval, CM, cyclophosphamide þ methotrexate.
consider surgery if it can be performed with a high quality proced- HER-2 positive ABC
ure [21].
Among all breast cancer subtypes, HER2-positive ABC has had
The definition of oligometastatic disease (see next section) has
the largest progress over the last decade. The introduction of new
been enlarged to encompass low volume metastatic disease, i.e.
anti-HER2 therapies, such as pertuzumab and T-DM1 [23–27],
limited number and size of metastatic lesions (up to five and not
was a significant step forward but also created a number of new
necessarily in the same organ) and potentially amenable for local
uncertainties related to optimal combination/sequence of all
treatment which is aimed at achieving a complete remission. The
available treatments.
development of minimally invasive surgical techniques and
In view of the overall survival (OS) results obtained with most
highly conformal ablative radiotherapy allow for safe and effect-
combinations of chemotherapy plus anti-HER-2 agents, the role
ive ablation of metastatic lesions in most locations. Although
of endocrine therapy plus anti-HER-2 agents for the subgroup of
some retrospective studies have suggested that achieving a sus-
patients with ERþ/HER-2þ disease has been questioned.
tained complete remission seems to be associated with a longer
Although published studies have not demonstrated an OS benefit
survival [22], the true impact of these local-regional therapies on
of this combination, long-term data were not collected in these
long-term outcome remains unknown, and prospective and if
trials. Of note, the OS analysis of the TAnDEM trial, excluding
possible randomized trials are needed.
patients who crossed over to trastuzumab, demonstrated a bor-
derline OS benefit for the combination arm [28]. In the absence
of valuable biomarkers, this approach should be reserved for
ABC important definitions highly selected patients, including those with contraindications
Most clinical situations occur as a continuum and dividing them to chemotherapy, patient’s with a strong preference against
into categories of stage, grade, risk group, or other factors is al- chemotherapy or those with a long disease-free interval, minimal
ways artificial and based on oversimplification of thresholds. disease burden, in particular in terms of visceral involvement,
Such a categorization is, however, useful to guide treatment and/or strong ER/PgR expression. Trials directly comparing
choices, to help assure adherence to guidelines and recommenda- chemotherapy plus anti-HER2 therapy versus endocrine therapy
tions, and to facilitate clinical research. Following the effort of plus anti-HER2 therapy are currently ongoing (Detect
previous editions, ABC provides two additional definitions: ‘oli- V/CHEVENDO (NCT02344472), SYSUCC-002 (NCT01950182)
gometastatic disease’ discussed above and the complex clinical and PERNETTA trials) and their results will allow for better
situation of ‘multiple chronic conditions’. The latter is becoming recommendations. In addition, in several countries anti-HER2
increasingly important and more frequent in view of the aging of therapy, namely trastuzumab, can only be used once in the meta-
the population in general and of cancer patients in particular. static setting since its use beyond progression is either not
Managing advanced cancer, the consequences of the disease and approved or not reimbursed; in those cases, preference should be
of the rapidly increasing number and type of pharmacologic and given to a combination of chemotherapy plus anti-HER-2
non-pharmacologic interventions in patients with several coex- therapy.
isting conditions is a major challenge. Furthermore, these pa- The combination of endocrine therapy plus anti-HER2 therapy
tients are systematically excluded from clinical trials and hence is particularly useful as maintenance therapy for ERþ/
available data, in particular regarding the use of new agents in HER2þ ABC, after initial cycles of chemotherapy plus anti-HER-
these situations, are scarce and eagerly needed. 2 therapy. Despite the absence of randomized trials, clinical

by guest
on 05 February 2018
SECTION 4. ER POSITIVE/HER-2 NEGATIVE (LUMINAL) ABC
Endocrine therapy (ET) is the preferred option for hormone receptor positive disease, even in the presence of 1A Voters: 41
visceral disease, unless there is visceral crisis or concern/proof of endocrine resistance. Yes: 93% (38)
Abstain: 7% (3)
The preferred 1st line ET for postmenopausal patients depends on type and duration of adjuvant ET as well as 1A Voters: 44
time elapsed from the end of adjuvant ET; it can be an aromatase inhibitor, tamoxifen or fulvestrant. Yes: 84% (37)
Abstain: 7% (3)
The combination of a nonsteroidal AI and fulvestrant as first-line therapy for postmenopausal patients resulted 2B Voters: 43
in significant improvement in both PFS and OS compared to AI alone in one phase III trial and no benefit in Yes: 33% (14)
a second trial with a similar design. Subset analysis suggested that the benefit was limited to patients with- No: 53% (23)
out prior exposure to adjuvant ET (tamoxifen). Based on these data, combination ET may be offered to Abstain: 14% (6)
some patients with MBC without prior exposure to adjuvant ET.
The addition of everolimus to an AI is a valid option for some postmenopausal patients with disease progres- 1B Voters: 40
sion after a non-steroidal AI, since it significantly prolongs PFS, albeit without OS benefit. The decision to treat Yes: 84% (34)
must take into account the individual relevant toxicities associated with this combination and should be Abstain: 13% (5)
made on a case by case basis.
Tamoxifen can also be combined with everolimus. 2B
The addition of the CDK4/6 inhibitor palbociclib to an aromatase inhibitor, as 1st line therapy, for postmeno- 1A Voters: 37
pausal patients (except patients relapsing <12 months from the end of adjuvant AI), provided a significant im- Yes: 92% (34)
provement in PFS (10 months), with an acceptable toxicity profile, and is therefore one of the preferred Abstain: 3% (1)
treatment options, where available. OS results are still awaited.
ESMO MCBS: 3*
The addition of CDK4/6 inhibitor palbociclib to Fulvestrant, beyond 1st line therapy, for pre/peri/postmeno- 1A Voters: 42
pausal patients, provided significant improvement in PFS (5 months) as well as improvement of QoL, and is Yes: 86% (36)
a treatment option. OS results are awaited. Abstain: 10% (4)
For pre/peri-menopausal pts, an LHRH-agonist must also be used.
At present, no predictive biomarker other than hormone receptor status exists to identify patients who will
benefit from these type of agents and research efforts must continue.
ESMO MCBS: 4*
The optimal sequence of endocrine agents after 1st line ET is uncertain. It depends on which agents were used 1A Voters: 40
in the (neo)adjuvant and 1st line ABC settings. Available options include AI, tamoxifen, fulvestrantþpalbociclib, Yes: 93% (37)
AIþeverolimus, tamoxifenþeverolimus, fulvestrant, megestrol acetate and estradiol. Abstain: 5% (2)
It is currently unknown how the different combinations of endocrineþbiological agents compare with each
other, and with single agent CT. Several trials are ongoing.
For pre-menopausal women, for whom ET was decided, ovarian suppression/ablation combined with add- 1B Voters: 43
itional endocrine therapy is the preferred choice. Yes: 93% (40)
Abstain: 5% (2)
Ovarian ablation by laparoscopic bilateral oophorectomy ensures definitive estrogen suppression and contra- Expert Opinion Voters: 43
ception, avoids potential initial tumor flare with LHRH agonist, and may increase eligibility for clinical trials. Yes: 91% (39)
Patients should be informed on the options of OS/OA and decision should be made on a case by case. Abstain: 7% (3)
For pre-menopausal women, the additional endocrine agent can be AI or tamoxifen, according to type and 1B Voters: 42
duration of prior adjuvant endocrine therapy but AI absolutely mandates the use of ovarian suppression/ Y: 95% (40)
ablation. 1C Abstain: 5% (2)
Fulvestrant is also a valuable option, but for the moment also mandates the use of ovarian suppression/
ablation.
LoE, available level of evidence; consensus, percentage of panel members in agreement with the statement; ET, endocrine therapy; CT, chemotherapy;
QoL, quality-of-life.
ESMO MBCS ¼ ESMO Magnitude of Clinical Benefit Scale; * 5 very important explanation in text.
experience and low toxicity (in particular if trastuzumab is used), are sufficient data [29, 30] to recommend continuing trastuzu-
makes this a reasonable option, most probably delaying disease mab beyond progression, but the optimal duration of this treat-
progression and the consequent need for chemotherapy. ment and how many lines beyond progression should it be used is
The issue of duration of anti-HER-2 therapy in the metastatic currently unknown. Data are very scarce related to the use beyond
setting is of crucial importance, in view of the potential benefits progression of other anti-HER2 agents and no data exist support-
as well as the substantial costs associated with these agents. There ing the use of dual blockade beyond progression.

by guest
on 05 February 2018
A particularly difficult situation, albeit also a fortunate one, re- studies, such as the SystHERs Registry Study [31] and
lates to the optimal duration of trastuzumab therapy in patients registHER, as well as collection of treatment and outcome data
achieving long-term complete remission. This needs to be bal- beyond progression in all HER-2-positive ABC clinical trials,
anced against toxicity, logistical burden and cost. Currently no are of great importance.
data exist to support therapeutic decisions in this setting, and the In ABC3, the optimal chemotherapy component for the
panel supported a cautious statement approving consideration of treatment of HER-2þ disease was discussed. The panel has
stopping trastuzumab in these circumstances in some patients, stressed the importance of treatment decisions that are based
particularly if treatment re-challenge is available in case of pro- not only on efficacy, but also on toxicity profile, and patients’
gression, which is not the case in all countries. preferences.
Dual blockade with trastuzumab and pertuzumab in combin- For 1st line therapy, when trastuzumab is used as sole anti-
ation with chemotherapy as 1st line therapy, provides substantial Her2 agent, the preferred agents are vinorelbine or a taxane.
benefit in terms of OS and PFS [23]. It is therefore considered by Importantly, single agent vinorelbine in association with trastu-
the panel as the standard of care for patients previously untreated zumab has shown superior or equal efficacy compared to either
with trastuzumab, in the (neo)adjuvant setting, and an important paclitaxel or docetaxel, in the TRAVIOTA and HERNATA trials,
treatment option for patients previously treated with trastuzu- and has a better tolerability [32, 33]. For later lines of therapy,
mab. The difference in the strength of recommendation is due to trastuzumab can be administered with almost all chemotherapy
the fact that very few patients (only 88) who were previously agents, including but not limited to, vinorelbine (if not given in
treated with trastuzumab were enrolled in the Cleopatra trial. In 1st line), taxanes (if not given in 1st line), capecitabine, eribulin,
addition, in the Marianne trial [26] the dual blockade strategy liposomal anthracyclines, platinum, gemcitabine, or metronomic
did not prove to be superior to chemotherapy and trastuzumab, CM (low dose, oral, cyclophosphamide and methotrexate). The
albeit with a different combination of agents—T-DM1 and decision should be individualized and take into account different
Pertuzumab. The reasons for this lack of benefit are currently un- toxicity profiles, previous exposure, patient preferences, and
known and could be related to the different patient populations country availability. Combinations of other anti-HER2 agents,
enrolled in both trials (more (30%) patients in Marianne had namely TKIs, with chemotherapy are more limited due to tox-
been previously treated with trastuzumab), the choice of agents icity. There are currently no data to decide on the best sequence
with the presence or absence of synergistic effects, the absence of for each individual patient.
standard chemotherapy agents (DM1 being a cytotoxic agent not When dual blockade with trastuzumab and pertuzumab is
used as single agent) or other factors. used, possible agents to combine are docetaxel [23], weekly pacli-
After the discussion and voting during ABC3, the Pherexa [27] taxel [34], vinorelbine [35] and nab-paclitaxel [36]. After the vot-
study was presented, evaluating the role of dual blockade with ing that took place in ABC3, the Pherexa trial [27], presented at
trastuzumab þ pertuzumab þ capecitabine for patients previ- ASCO 2016, provided some evidence regarding the combination
ously treated with a taxane and trastuzumab in the metastatic set- of dual blockade with capecitabine.
ting. Surprisingly, a non-significant benefit of only 2 months was
seen in the primary endpoint PFS, while an 8-month benefit was
observed in OS albeit non-statistically significant (in view of the
lack of significant PFS benefit). ER positive/HER-2 negative (luminal) ABC
Many questions remain unanswered in the management of One of the most important recommendations relates to the pre-
HER-2þ ABC. We have no data on the role of dual blockade for ferred treatment for luminal ABC, which should be endocrine
patients relapsing during and within 12 months of adjuvant tras- therapy in the majority of cases, excluding those with visceral cri-
tuzumab, since these patients have been excluded from clinical sis and concern or proof of endocrine resistance. All breast cancer
trials. This aggressive situation is a clear unmet need for which guidelines concur with this recommendation but unfortunately
data must be generated. Following the approval, both by FDA real life data studies show that most of these patients still receive
and EMA, of pertuzumab use in the neoadjuvant setting, there is chemotherapy as their first treatment, despite the lower efficacy
an urgent need to evaluate the best treatment options for the [37].
patients who relapse after receiving chemother- Visceral crisis and endocrine resistance have been defined dur-
apy þ trastuzumab þ pertuzumab in the early setting. It is also ing ABC 2 and published [1]. However, better predictive factors
currently unknown how trastuzumab þ pertuzumab þ chemo- are urgently needed to clearly identify those patients whose
therapy compares to T-DM1, as 1st or later lines of therapy. We tumors have primary endocrine resistance and are responsible for
also have no data on the best treatment option after progression the early and rapid progression seen in 20–25% of luminal ABC
on dual blockade with pertuzumab þ trastuzumab, namely how patients treated with endocrine therapy [38]. Possible reasons
T-DM1 performs in this setting. may include ER loss [39] or ER mutations [40].
While trastuzumab þ lapatinib (without chemotherapy) is a The most important advance in the management of luminal
valuable option for some patients, after progression on chemo- ABC over the last 2 years has undoubtedly been the introduction
therapy þ trastuzumab, there are no data on the use of this com- of a new class of agents, the CDK4/6 inhibitors, in combination
bination after progression on pertuzumab or T-DM1. with an endocrine agent.
All these unanswered questions and the definition of the best The value of the CDK4/6 inhibitor palbociclib, combined with
sequence of therapies for the individual patient may prove dif- an aromatase inhibitor as 1st line therapy was evaluated initially
ficult to evaluate in prospective, randomized trials, with the in a randomized phase II study, the PALOMA 1 trial [41], which
absence of specific biomarkers. In this scenario, registry showed a substantial 10-month benefit in progression-free-

by guest
on 05 February 2018
SECTION 5. TRIPLE NEGATIVE ABC
For non-BRCA-associated triple negative ABC, there are no data supporting different or specific CT recommen- 1A Voters: 44
dations. Therefore, all CT recommendations for HER-2 negative disease also apply for triple negative ABC. Yes: 98% (43)
Abstain: 2% (1)
In triple-negative ABC patients (regardless of BRCA status), previously treated with anthracyclines with or with- 1A Voters: 43
out taxanes in the (neo)adjuvant setting, carboplatin demonstrated comparable efficacy and a more favor- Yes: 91% (39)
able toxicity profile, compared to docetaxel, and is therefore an important treatment option. Abstain: 5% (2)
LoE, available level of evidence; consensus, percentage of panel members in agreement with the statement; CT, chemotherapy.
survival (PFS) coupled with a favorable toxicity profile (main Another possible therapy is the combination of endocrine ther-
toxicity being neutropenia). Based on these results, FDA granted apy with the mTOR inhibitor, everolimus. This combination has
accelerated approval, which resulted in the drug being commer- shown a PFS benefit of 6 months, without a significant OS
cially available in USA. At the 2016 ASCO meeting, the phase III benefit, and with significant toxicity [44, 45]. However, as with
PALOMA 2 trial was presented and confirmed the 10-month many agents, as more experience is gained regarding the use of
benefit in PFS, with the main toxicities being hematological everolimus and the management of its toxicities, its clinical use
(mainly neutropenia) and fatigue [41]. OS results are still becomes easier. In addition, patient education is fundamental for
awaited. In view of these results, the initial statement developed prevention and early management of associated side effects. Of
at ABC3 was modified and re-voted by email and considers this particular attention is the possibility of an excess mortality of this
option as one of the preferred treatment options, where available. combination in elderly patients (>70 years of age) [44, 46].
Very recently (September 2016) EMA also started the approval Currently, and in spite of intensive research, no predictive bio-
process of Palbociclib. However, its approval/reimbursement in marker, other than hormone receptor status, exists to identify pa-
all individual countries is still pending and the issue of cost is of tients who will benefit the most from either m-TOR or CDK4-6
crucial importance for its implementation in clinical practice, as inhibitors and research efforts must continue.
it is for many targeted agents namely anti-HER-2 agents. The panel did not support (53.4% against) the 1st line combin-
Beyond 1st line endocrine therapy, addition of palbociclib to ation of non-steroidal aromatase inhibitor and fulvestrant based
fulvestrant resulted in significant albeit lower 5-month PFS pro- on the results of the SWOG S0226 trial [47]. There may be a bene-
longation in the PALOMA 3 phase III trial [42]. The quality of fit for the minority of postmenopausal patients who are endo-
life substudy has shown both an overall improvement and a crine-naı̈ve.
delayed deterioration of this important endpoint, with greater The definition of the best 1st line approach for postmeno-
improvement in baseline pain, in the palbociclib arm [43]. pausal patients will soon have additional data through the phase
Importantly, the PALOMA-3 study accrued both postmeno- III FALCON data that will be presented this year.
pausal and pre/perimenopausal (in combination with ovarian The optimal sequence of single endocrine agents and combin-
function suppression) patients, allowing for assessment of the ations with targeted agents is currently unknown and is a research
drug efficacy in a breast cancer population usually excluded from priority. It is crucial to collect data from clinical trials beyond
ABC endocrine therapy trials. OS results are still awaited. In view progression to better understand the efficacy of each class of agent
of available results, the ABC panel considers this as a treatment when given after the other (e.g. CDK4-6 inhibitors after m-TOR
option, where available. inhibitors and vice-versa).
The ESMO Magnitude of Clinical Benefit Scale (MCBS) was
calculated for the recently approved Palbociclib, for use in 1st
line and in 2nd line. As a reminder, the MCBS scores a given
treatment in a given setting, and based on published trials. At the Triple negative ABC
time of publishing the ABC3 guidelines, PALOMA 2 main results The treatment of triple-negative breast cancer (TN-ABC) still re-
and the accompanying quality of life substudy have been pre- mains the largest unmet need within ABC. In spite of extensive re-
sented but not yet published. For this reason, the MCBS for the search, no treatments apart from chemotherapy have so far proven
use of palbociclib in 1st line was calculated using the PALOMA 1 to be effective for this population. For this reason, no specific recom-
trial efficacy data, which scores a 3 for efficacy. Once the mendations can be made for this ABC subtype, with the possible ex-
PALOMA 2 data is published the MCBS will be updated an e-up- ception of platinum compounds for BRCA-mutated patients.
date made available through the ESMO guidelines website. For Probably the largest achievement of the last 2 years was the
the use of palbociclib as 2nd line therapy, data from PALOMA 3, TNT study, comparing ‘standard’ docetaxel to carboplatin in
both efficacy and quality of life, were used. The MCBS was 3 for unselected TNBC patients (with pre-specified subgroup analysis
efficacy, and due to the improvement in quality of life upgraded of BRCA-mutation carriers). The superiority of carboplatin was
to 4, which is the final score for this setting. demonstrated only among BRCA-positive patients, while in the

by guest
on 05 February 2018
SECTION 6. OTHER RECOMMENDATIONS
CHEMOTHERAPY OTHER
Metronomic chemotherapy is a reasonable treatment option, for patients not requiring rapid tumor re- 1B Voters: 43
sponse. The better studied regimen is CM (low dose oral cyclophosphamide and methotrexate); other regi- Yes: 88% (38)
mens are being evaluated (including capecitabine and vinorelbine). Randomized trials are needed to Abstain: 5% (2)
accurately compare metronomic CT with standard dosing regimens.
Even if given in the adjuvant setting, provided that cumulative dose has not been achieved and that there are 1C Voters: 44
no cardiac contra-indications, anthracyclines can be re-used in MBC, particularly if there has been at least Yes: 93% (41)
1 year of disease-free survival. Abstain: 5% (2)
BRCA-ASSOCIATED ABC
In patients with BRCA-associated triple negative or endocrine-resistant MBC previously treated with an 1A Voters: 44
anthracycline with or without a taxane (in the adjuvant and/or metastatic setting), a platinum regimen is Yes: 86% (38)
the preferred option, if not previously administered and no suitable clinical trial is available. Abstain: 9% (4)
In patients with TN or Luminal MBC, genetic counseling and possibly BRCA testing should be discussed Expert Opinion Voters: 43
with the patient, if the results can impact on treatment decisions and/or on clinical trials entry. Yes: 91% (39)
Abstain: 7% (3)
BONE METASTASES
A bone modifying agent (bisphosphonate, denosumab) should be routinely used in combination with other 1A Voters: 44
systemic therapy in patients with MBC and bone metastases. Yes: 95% (42)
Three-monthly zolendronic acid seems to be not inferior to standard monthly schedule. 1B Abstain: 5% (2)
Supplementation of calcium and vitamin D3 is mandatory, unless contraindications exist. 1C
OTHERÅBIOMARKERS
Multigene panels, such as those obtained using next generation sequencing (NGS) or other technology, re- 1C Voters: 44
garding evolving molecular changes in ABC tumors has not yet proven beneficial in clinical trials, their im- Yes: 95% (42)
pact on outcome remains undefined and should only be considered investigational. Abstain: 5% (2)
LoE, available level of evidence; consensus, percentage of panel members in agreement with the statement; MBC, metastatic breast cancer.
unselected TN-ABC population docetaxel and carboplatin seem on their efficacy and toxicity profile, are capecitabine, vinorelbine
to have a similar efficacy [48], although the study was not de- and eribulin. The latter is one of the few agents to provide a survival
signed as a non-inferiority study. Of note, in this study, 15% of gain, albeit small (2.5 months) in a heavily pretreated population of
patients had no prior adjuvant chemotherapy and only 35% had ABC patients [51]. In a head-to-head comparison between eribulin
received (neo)adjuvant taxanes. Importantly, due to the signifi- and capecitabine, as first or second line therapy, there were no major
cantly better toxicity profile of carboplatin, it remains an at- differences between the drugs in efficacy but a different toxicity pro-
tractive treatment choice even for unselected TN-ABC patients. file [52].
Unfortunately, other putative predictive factors of increased It is also possible to re-challenge with anthracyclines, particu-
sensitivity to platinum, such as homologous recombination def- larly if there has been at least 1 year of disease-free survival, and if
icit (HRD) and the basal-like Prosigna PAM50 signature were the cumulative dose has not been reached, a common situation
not proven of value for making treatment decisions in this nowadays because of the lower doses of anthracyclines used in
setting. the adjuvant setting. Re-challenge with taxanes is also possible,
The future of TN-ABC treatment seems to lie in a better biolo- provided that there has been at least 1 year of disease-free
gical characterization of this breast cancer subtype into further survival.
subgroups, followed by the development of specific therapies for Another very attractive option is the use of metronomic
each of the subgroups. An example is the Luminal AR subtype, chemotherapy, defined as the use of low doses and short intervals,
characterized by the expression of the androgen receptor; antian- which has been evaluated in the advanced setting with interesting
drogens have recently demonstrated some activity and are being efficacy results and an excellent toxicity profile [53]. The best
further evaluated, and where a potential predictive marker, the evaluated regimen is oral cyclophosphamide and oral methotrex-
Predict AR assay, is also being tested [49, 50]. ate but other agents are being studied such as vinorelbine and
capecitabine.
In view of the lack of substantial efficacy differences among the
Other recommendations different available options, their toxicity profile must be discussed
Several options exist for chemotherapy both for first and subsequent with the patient and her/his preferences taken into account.
lines of therapy. The ABC panel maintains that for patients pre- ABC3 also further endorsed the use of bone-modifying
treated with anthracyclines and taxanes the preferred agents, based agents (bisphosphonate, denosumab) in combination with

by guest
on 05 February 2018
SECTION 7. SUPPORTIVE AND PALLIATIVE CARE
Management of CANCER RELATED FATIGUE 100%

Cancer related fatigue is frequently experienced by patients with ABC, exerts a deleterious impact on QoL and
limits physical, functional, psychological and social well-being. The etiology of this fatigue is complex so effect- as in the text
ive management needs to be multidimensional. It is important to assess it using appropriate PRO measures
before implementing various non-pharmacological (such as exercise—LoE: 1 A) and if needed pharmaco-
logical interventions (LoE: 2 B).
Management of CDK Inhibitor Induced Neutropenia 2A 100%
Neutropenia is the most common toxicity associated with CDK 4/6 inhibition and is not generally associated
with febrile neutropenia although an increase in infections has been reported. Treatment should be delayed
until neutrophils have recovered to at least 1000/ll; dose reduction can also be considered.
Management of Non-Infectious Pneumonitis (NIP) 2A 100%
NIP is an uncommon complication of mTOR inhibition. Patient education is critical to ensure early reporting of
respiratory symptoms.
Treatment interruption and dose reduction are generally effective for grade 2 symptomatic NIP with use of sys-
temic steroids and treatment discontinuation for grade 3 or greater toxicity.
Management of MUCOSITIS/STOMATITIS
Mild toothpaste and gentle hygiene are recommended for the treatment of stomatitis. Early intervention is rec- Expert opinion 100%
ommended. For grade 2 or higher stomatitis, delaying treatment until the toxicity resolves and considering
lowering the dose of the targeted agent are also recommended. Consider adding steroid dental paste to treat
developing ulcerations.
Steroid mouthwash can be used for prevention of stomatitis (suggested schedule: 0.5 mg/5 ml dexamethasone, 1B
10 ml to swish2 min then spit out qid).
Management of DYSPNEA
Treatable causes like pleural effusion, pulmonary emboli, cardiac insufficiency, anemia or drug toxicity must be 1 A, 100%
ruled out. Patient support is essential. Oxygen is of no use in non-hypoxic patients. Opioids are the drugs of 2 A,
choice in the palliation of dyspnea (LoE: 1 A). Benzodiazepines can be used in patients experiencing anxiety Expert opinion
(LoE: 2A). Steroids can be effective in dyspnea caused by lymphangitis carcinomatosis, radiation or drug-
induced pneumonitis, superior vena cava syndrome, an inflammatory component, or in (cancer-induced) ob-
struction of the airways (in which case laser/stent is to be considered).
Management of NAUSEA and VOMITING
ESMO/MASCC GUIDELINES are available for management of chemotherapy-induced and morphine-induced
nausea and vomiting, and these are endorsed by ABC3.
There is a need to study nausea and vomiting related to chronic use of anticancer drugs. Expert Opinion 100%
Management of endocrine toxicities of mTOR inhibition 2A 100%
Hyperglycemia and hyperlipidemia are common sub-acute complications of mTOR inhibition. Evaluation of
preexisting diabetes or hyperglycemia at baseline is essential. Regular careful monitoring of glycemia and lipid
panel is needed to identify these toxicities.
Management of grade 1 and 2 hyperglycemia include treatment with oral antidiabetics and basal insulin, in ac-
cordance with international recommendation for diabetes mellitus treatment. Statins are indicated to treat
grade 2 and 3 hypercholesterolemia, and fibrates should be introduced if triglyceride level >500 mg/dl (with
attention to possible drug–drug interaction between everolimus and fibrates). Treatment interruption and
dose reduction are generally effective for grade 2 and 3. Treatment should be discontinued for grade 4
toxicity.
LoE, available level of evidence; consensus, percentage of panel members in agreement with the statement; QoL, quality of life.
NOTE: The statements of this section were not voted during the ABC Consensus panel but were developed and agreed upon by email, by all
panel members.
calcium þ vitamin D3 supplementation as a routine component of 4 weekly administration of intravenous bisphosphonates by 3-

management of patients with bone metastases. Denosumab has monthly zoledronic acid after an initial period of monthly use [55,
demonstrated slightly better efficacy and better tolerability, com- 56]. Early 3-monthly use seems associated with increased need for
pared to zoledronic acid [54], having the advantage of a subcutane- major surgeries [57], so a reasonable compromise may be to start
ous route of administration and the disadvantage of a substantially with the monthly schedule for the first year and then change to 3-
higher cost in most countries; where available, it can be considered a monthly regimen. No data exist on the optimal overall treatment
preferred option. Currently available data support replacing routine duration of bone modifying agents, and their efficacy must be

by guest
on 05 February 2018
weighed against long-term toxicity (such as osteonecrosis of the jaw Conference, the ESMO Committees will apply the MCBS and the
and atypical fractures). result will be made available as an e-update to the present
When a bone modifying agent is given, supplements of calcium guidelines.
and vitamin D are mandatory, except in the presence of contra- Following on the success of the ABC Consensus Conference,
indications. the ABC community has come together to create the ABC Global
Unfortunately, no multigene testing technology has been proven Alliance. This Alliance will function as a platform where all
to be beneficial in supporting treatment choices in ABC patients involved partners (advocacy groups, pharma, cooperative
[18] and the panel strongly discourages their use in clinical prac- groups, societies, individuals) will be able to work together, in
tice. They should continue to be considered investigational. projects designed to improve the lives of ABC patients. The
Global Status of ABC Decade Report [2] has highlighted several
areas of unmet needs. Based on these findings, a global Call-To-
Supportive and palliative care Action is being developed, with tangible objectives that need to
be achieved within the next decade to meaningfully impact the
The ABC panel decided to dedicate several recommendations to outcomes of ABC patients.
the management of disease and treatment-related symptoms, a
problem faced daily by patients and every practicing oncologist,
that can significantly affect a patient’s quality of life Funding
Unfortunately, little high-quality data exist in many areas of
symptom management, probably due to difficulties in conduct- None declared.
ing research in this field, including the lack of well-defined end-
points, of patient-reported symptoms and side effects, and of
optimal tools to evaluate impact on quality of life for advanced Disclosure
cancer patients. New classes of drugs introduced into breast can- Detailed CoI for all panel members are described in online
cer management have brought into the clinical practice new tox- supplement.
icities, poorly understood in the beginning and unfamiliar to
most oncologists. Undoubtedly this is an area of unmet need,
which should be a research priority. References
The ABC3 guidelines provide guidance on the management of 1. Cardoso F, Costa A, Norton L et al. ESO-ESMO 2nd International
drug-induced pneumonitis, mucositis [58, 59], endocrine and Consensus Guidelines for Advanced Breast Cancer (ABC2).
metabolic disorders and CDK4/6 inhibitor-related neutropenia. Simultaneous publication in The Breast 2014, 23: 489–502; http://dx.doi.
For nausea and vomiting ABC fully endorses the guidelines de- org/10.1016/j.breast.2014.08.009) and Annals of Oncology (Ann Oncol
2014; 25: 1871–1888; doi:10.1093/annonc/mdu385).
veloped by ESMO/MASCC [60].
2. Global Status of Advanced/Metastatic Breast Cancer 2005–2015 Decade
The ABC panel continues to discuss and provide guidance on Report. www.breastcancervision.com and www.abc-lisbon.org (27
the management of frequent and difficult to manage cancer- October 2016, date last accessed).
associated symptoms. In this edition, dyspnea and fatigue were 3. Howlader N, Noone AM, Krapcho M et al. (eds). SEER Cancer Statistics
discussed. Cancer related fatigue is frequently experienced by Review, 1975–2013. Bethesda, MD: National Cancer Institute. http://
advanced cancer patients, exerts a deleterious impact on their seer.cancer.gov/csr/1975_2013/, based on November 2015 SEER data
submission, posted to the SEER web site, April 2016 (updated September
quality of life and limits physical, functional, psychological and
12 2016, 27 October 2016, date last accessed).
social well-being. Its etiology is complex and therefore effective 4. Sundquist M, Eriksson Z, Tejler G, Brudin L. Trends in survival in meta-
management needs to be multidimensional [61–63]. It is import- static breast. Cancer Eur J Cancer 2010; 8(3): 191. Abstract 453.
ant to assess cancer related fatigue using appropriate patient- 5. Kobayashi K, Ito Y, Matsuura M et al. Impact of immunohistological
reported outcome measures before implementing various subtypes on the long-term prognosis of patients with metastatic breast
pharmacological and non-pharmacological interventions. cancer. Surg Today 2016; 46: 821–826.
Randomized studies have suggested improvement of fatigue by 6. Chia SK, Speers CH, D’yachkova Y et al. The impact of new chemothera-
peutic and hormone agents on survival in a population-based cohort of
various types of exercise quite convincingly [64], and meditation
women with metastatic breast cancer. Cancer 2007; 110: 973–979.
and some pharmacologic interventions are under evaluation. The 7. Guyatt G, Gutterman D, Baumann MH et al. Grading strength of recom-
use of good evidence-based algorithms for management of cancer mendations and quality of evidence in clinical guidelines: report from an
related fatigue can also be helpful [65]. american college of chest physicians task force. Chest 2006; 129:
174–181.
8. Cherny NI, Sullivan R, Dafni U et al. A standardised, generic, validated ap-
Conclusions proach to stratify the magnitude of clinical benefit that can be anticipated
from anti-cancer therapies: the European Society for Medical Oncology
Since the ABC3 Conference two important initiatives have al- Magnitude of Clinical Benefit Scale (ESMO-MCBS). Annals of Oncology
ready been initiated. 2015; 26: 1547–1573.
The ESMO Magnitude of Clinical Benefit Scale (MCBS) [8] 9. Schnipper LE, Davidson NE, Wollins DS et al. American Society of
Clinical Oncology Statement: a conceptual framework to assess the value
has been published and is being applied to all new anticancer
of cancer treatment options. J Clin Oncol 2015; 33: 2563–2577.
treatments approved by EMA. The latest drug for which EMA 10. Cardoso F, Costa A, Norton L et al. 1st International consensus guide-
started the approval process was Palbociclib in September 2016 lines for advanced breast cancer (ABC1). The Breast 2012; 21: 242–252.
and its MCBS evaluation is included in the present article. Should 11. Key components of Patient Centered Care, adapted from Levit L et al.
another agent be approved before the next ABC Consensus Delivering high-quality cancer care: charting a new course for a system in

by guest
on 05 February 2018
crisis. National Academy of Science: Institute of Medicine, The National 30. Rayson D, Lutes S, Walsh G et al. Trastuzumab beyond progression for
Academies, 2013; 20,1. 3 p.3.4–3.7. HER2 positive metastatic breast cancer: progression-free survival on first-
12. Cardoso F, Harbeck N, Mertz S, Fenech D. Evolving psychosocial, emo- line therapy predicts overall survival impact. Breast J 2014; 20: 408–413.
tional, functional, and support needs of women with advanced breast 31. Tripathy D, Rugo HS, Kaufman PA et al. The SystHERs registry: an observa-
cancer: results from the count us, know us, join us and here & now sur- tional cohort study of treatment patterns and outcomes in patients with
veys. The Breast 2016; 28: 5–12. human epidermal growth factor receptor 2-positive metastatic breast cancer.
13. Karlsson E, Appelgren J, Solterbeck A et al. Breast cancer during follow- BMC Cancer 2014; 14: 307.
up and progression – a population based cohort on new cancers and 32. Andersson M, Lidbrink E, Bjerre K et al. Phase III randomized study
changed biology. Eur J Cancer 2014; 50: 2916–2924. comparing docetaxel plus trastuzumab with vinorelbine plus trastuzu-
14. Thompson AM, Jordan LB, Quinlan P et al. Prospective comparison of mab as first-line therapy of metastatic or locally advanced human epider-
switches in biomarker status between primary and recurrent breast can- mal growth factor receptor 2-positive breast cancer: the HERNATA
cer: the Breast Recurrence In Tissues Study (BRITS). Breast Cancer Res study. J Clin Oncol 2011; 29: 264–271.
2010; 12: R92. 33. Burstein HJ, Keshaviah A, Baron AD et al. Trastuzumab plus vinorelbine
15. Amir E, Miller N, Geddie W et al. Prospective study evaluating the im- or taxane chemotherapy for HER2-overexpressing metastatic breast can-
pact of tissue confirmation of metastatic disease in patients with breast cer: the trastuzumab and vinorelbine or taxane study. Cancer 2007; 110:
cancer. J Clin Oncol 2012; 30: 587–592. 965–972.
16. Penault-Llorca F, Coudry RA, Hanna WM et al. Recommendations for 34. Smyth LM, Iyengar NM, Chen MF et al. Weekly paclitaxel with trastuzu-
retesting breast cancer metastases for HER2 and hormone receptor sta- mab and pertuzumab in patients with HER2-overexpressing metastatic
tus. Breast 2013; 22: 200–202. breast cancer: overall survival and updated progression-free survival re-
17. Schrijver WA, van der Groep P, Hoefnagel LD et al. Influence of decalci- sults from a phase II study. Breast Cancer Res Treat 2016; 158: 91–97.
fication procedures on immunohistochemistry and molecular pathology 35. Andersson M, L opez-Vega JM, Petit T et al. The co-administration of per-
in breast cancer. Mod Pathol 2016; doi:10.1038/modpathol.2016.116.
tuzumab (P) and trastuzumab (T) as a single infusion, followed by vinorel-
18. André F, Bachelot T, Commo F et al. Comparative genomic hybridisa-
bine (V), in first-line (1L) treatment of HER2-positive locally advanced or
tion array and DNA sequencing to direct treatment of metastatic breast
metastatic breast cancer (MBC) patients (pts): VELVET study interim ana-
cancer: a multicentre, prospective trial (SAFIR01/UNICANCER). Lancet
lysis. J Clin Oncol 2015; 33(15_suppl): 586.
Oncol 2014; 15: 267–274.
36. Untch M, Jackisch C, Schneeweiss A et al. Nab-paclitaxel versus solvent-
19. Badwe R, Hawaldar R, Nair N et al. Locoregional treatment versus no
based paclitaxel in neoadjuvant chemotherapy for early breast cancer
treatment of the primary tumour in metastatic breast cancer: an open-
(GeparSepto-GBG 69): a randomised, phase 3 trial. Lancet Oncol 2016;
label randomised controlled trial. Lancet Oncol 2015; 16: 1380–1388.
20. Soran A, Ozmen V, Ozbas S et al. A randomized controlled trial evaluat- 17: 345–356.
ing resection of the primary breast tumor in women presenting with de 37. Lobbezoo DJ, van Kampen RJ, Voogd AC et al. In real life, one-quarter
novo stage IV breast cancer: Turkish Study (Protocol MF07-01). J Clin of patients with hormone receptor-positive metastatic breast cancer re-
Oncol 2016; 34(suppl): abstr 1005. ceive chemotherapy as initial palliative therapy: a study of the Southeast
21. Thomas A, Khan SA, Chrischilles EA, Schroeder MC. Initial surgery and Netherlands Breast Cancer Consortium. Ann Oncol 2016; 27: 256–262.
survival in stage IV breast cancer in the United States, 1988-2011. JAMA 38. Finn RS, Martin M, Rugo HS et al. PALOMA-2: primary results from a
Surg 2016; 151: 424–431. phase III trial of palbociclib (P) with letrozole (L) compared with letro-
zole alone in postmenopausal women with ERþ/HER2– advanced breast
22. Greenberg PA, Hortobagyi GN, Smith TL et al. Long-term follow-up of
cancer (ABC). J Clin Oncol 2016; 34 (suppl): abstr 507.
patients with complete remission following combination chemotherapy
39. Weigel RJ, deConinck EC. Transcriptional control of estrogen receptor
for metastatic breast cancer. J Clin Oncol 1996; 14: 2197–2205.
in estrogen receptor-negative breast carcinoma. Cancer Res 1993; 53:
23. Swain SM, Kim SB, Cortés J et al. Pertuzumab, trastuzumab, and doce-
3472–3474.
taxel for HER2-positive metastatic breast cancer (CLEOPATRA study):
40. Fribbens C, O’Leary B, Kilburn L et al. Plasma ESR1 mutations and the
overall survival results from a randomised, double-blind, placebo-con-
treatment of estrogen receptor-positive advanced breast cancer. J Clin
trolled, phase 3 study. Lancet Oncol 2013; 14: 461–471.
Oncol 2016; 34: 2961–2968.
24. Verma S, Miles D, Gianni L et al. Trastuzumab emtansine for HER2-
41. Finn RS, Crown JP, Lang I et al. The cyclin-dependent kinase 4/6 inhibi-
positive advanced breast cancer. N Engl J Med 2012; 367: 1783–1791.
tor palbociclib in combination with letrozole versus letrozole alone as
25. Krop IE, Kim SB, Gonzalez-Martın A et al. Trastuzumab emtansine versus first-line treatment of estrogen receptor-positive, HER2-negative,
treatment of physician’s choice for pretreated HER2-positive advanced advanced breast cancer (PALOMA-1/TRIO-18): a randomised phase 2
breast cancer (TH3RESA): a randomised, open-label, phase 3 trial. Lancet study. Lancet Oncol 2015; 16: 25–35.
Oncol 2014; 1: 689–699. 42. Turner NC, Ro J, André F et al. Palbociclib in hormone-receptor-
26. Ellis P, Barrios C, Eiermann W et al. Phase III, randomized study of tras- positive advanced breast cancer. N Engl J Med 2015; 373: 209–219.
tuzumab emtansine (T-DM1) 6 pertuzumab (P) vs trastuzumab þ tax- 43. Harbeck N, Iyer S, Turner N et al. Quality of life with palbociclib plus
ane (HT) for first-line treatment of HER2-positive MBC: primary results fulvestrant in previously treated hormone receptor-positive, HER2-nega-
from the MARIANNE study. J Clin Oncol 2015; 33(suppl): abstr 507. tive metastatic breast cancer: patient-reported outcomes from the
27. Urruticoechea A, Rizwanullah M, Im SA et al. PHEREXA: a phase III PALOMA-3 trial. Ann Oncol 2016; 27: 1047–1054.
study of trastuzumab (H) þ capecitabine (X) 6 pertuzumab (P) for pa- 44. Piccart M, Hortobagyi GN, Campone M et al. Everolimus plus exemes-
tients (pts) who progressed during/after one line of H-based therapy in tane for hormone-receptor-positive, human epidermal growth factor
the HER2-positive metastatic breast cancer (MBC) setting. J Clin Oncol receptor-2-negative advanced breast cancer: overall survival results from
2016; 34(15_suppl): abstr. 504. BOLERO-2. Ann Oncol 2014; 25: 2357–2362.
28. Kaufman B, Mackey JR, Clemens MR et al. Trastuzumab plus anastrozole 45. Bachelot T, Bourgier C, Cropet C et al. Randomized phase II trial of
versus anastrozole alone for the treatment of postmenopausal women with everolimus in combination with tamoxifen in patients with hormone
human epidermal growth factor receptor 2-positive, hormone receptor- receptor-positive, human epidermal growth factor receptor 2-negative
positive metastatic breast cancer: results from the randomized phase III metastatic breast cancer with prior exposure to aromatase inhibitors: a
TAnDEM study. J Clin Oncol 2009; 27: 5529–5537. GINECO study. J Clin Oncol 2012; 30: 2718–2724.
29. von Minckwitz G, du Bois A, Schmidt M et al. Trastuzumab beyond pro- 46. Pritchard KI, Burris HA III, Ito Y et al. Safety and efficacy of everolimus
gression in human epidermal growth factor receptor 2-positive advanced with exemestane vs. exemestane alone in elderly patients with HER2-
breast cancer: a German Breast Group 26/breast international group 03- negative, hormone receptor-positive breast cancer in BOLERO-2. Clin
05 study. J Clin Oncol 2009; 27: 1999–2006. Breast Cancer 2013; 13: 421–432.

by guest
on 05 February 2018
47. Mehta RS, Barlow WE, Albain KS et al. Combination anastrozole and bone metastases from breast cancer: results of the OPTIMIZE-2 trial. J
fulvestrant in metastatic breast cancer. N Engl J Med 2012; 367: 435–444. Clin Oncol 2014; 32: 5s (suppl): abstr LBA9500̂.
48. Tutt A, Ellis P, Kilbum L et al. TNT: a randomized phase III trial of car- 57. Himelstein AL, Qin R, Novotny PJ et al. CALGB 70604 (Alliance): a
boplatin compared to docetaxel for patients with metastatic or recurrent randomized phase III study of standard dosing vs. longer interval dosing
locally advanced triple-negative or BRCA1/2 breast cancer. Cancer Res of zoledronic acid in metastatic cancer. J Clin Oncol 2015; 33: (suppl):
2015; 75 (9 Suppl): S3–01. abstr 9501.
49. Cortes J, Crown J, Awada A et al. Overall survival (OS) from the phase 2 58. Jones VE, Jensen LL, McIntyre KJ et al. Evaluation of miracle mouthwash
study of enzalutamide, an androgen receptor (AR) signaling inhibitor, in (MMW) plus hydrocortisone versus prednisolone mouth rinses as
arþ advanced triple-negative breast cancer (aTNBC). Eur J Cancer 2015; prophylaxis for everolimus-associated stomatitis: preliminary results of a
51(Suppl. S3): S265. abstr. 1802. randomized phase II study. 2015 San Antonio Breast Cancer
50. Bonnefoi H, Grellety T, Tredan O et al. A phase II trial of abiraterone Symposium; December 5–8, 2015; San Antonio, TX. Abstract P1-15-06.
acetate plus prednisone in patients with triple-negative androgen recep- 59. Rugo HS, Seneviratne L, Beck JT et al. Prevention of everolimus/exemes-
tor positive locally advanced or metastatic breast cancer (UCBG 12-1). tane (EVE/EXE) stomatitis in postmenopausal (PM) women with hor-
Ann Oncol 2016; 27: 812–818. mone receptor-positive (HRþ) metastatic breast cancer (MBC) using a
51. Cortes J, O’Shaughnessy J, Loesch D et al. Eribulin monotherapy versus dexamethasone-based mouthwash (MW): results of the SWISH trial. J
treatment of physician’s choice in patients with metastatic breast cancer Clin Oncol 2016; 34 (suppl); abstr 525.
(EMBRACE): a phase 3 open-label randomized study. Lancet 2011; 377: 60. Roila F, Molassiotis A, Herrstedt J et al. MASCC and ESMO Consensus
914–923. Guidelines for the prevention of chemotherapy and radiotherapy-
52. Kaufman PA, Awada A, Twelves C et al. Phase III open-label randomized induced nausea and vomiting: ESMO clinical practice guidelines. Ann
study of eribulin mesylate versus capecitabine in patients with locally Oncol 2016; 27(suppl 5): v119–v133.
advanced or metastatic breast cancer previously treated with an anthra- 61. Payne C, Wiffen PJ, Martin S. Interventions for fatigue and weight loss in
cycline and a taxane. J Clin Oncol 2015; 33: 594–601. adults with advanced progressive illness. Cochrane Database Syst Rev
53. Munzone E, Colleoni M. Clinical overview of metronomic chemotherapy 2012; 1: CD008427.
in breast cancer. Nat Rev Clin Oncol 2015; 12: 631–644. 62. Berger A, Mitchell SA, Jacobsen PB, Pirl WF. Screening, evaluation and
54. Stopeck AT, Lipton A, Body JJ et al. Denosumab compared with zole- management of cancer-related fatigue: ready for implementation to prac-
dronic acid for the treatment of bone metastases in patients with tice. CA Cancer J Clin 2015; 65: 190–211.
advanced breast cancer: a randomized, double-blind study. J Clin Oncol 63. Aapro M, Scotte F, Bouillet T et al. A practical approach to fatigue man-
2010; 28: 5132–5139. agement in colorectal cancer. Clin Colorectal Cancer 2016; doi: 10.1016/
55. Amadori D, Aglietta M, Alessi B et al. Efficacy and safety of 12-weekly j.clcc.2016.04.010.
versus 4-weekly zoledronic acid for prolonged treatment of patients with 64. Travier N, Velthuis MJ, Steins Bisschop CN et al. Effects of an 18-week
bone metastases from breast cancer (ZOOM): a phase 3, open-label, exercise programme started early during breast cancer treatment: a
randomised, non-inferiority trial. Lancet Oncol 2013; 14: 663–670. randomized controlled trial. BMC Med 2015, 8: 121.
56. Hortobagyi GH, Lipton A, Chew HK et al. Efficacy and safety of contin- 65. Koornstra RH, Peters M, Donofrio S et al. Management of fatigue in patients
ued zoledronic acid every 4 weeks versus every 12 weeks in women with with cancer – a practical overview. Cancer Treat Rev 2014; 40: 791–799.

by guest
on 05 February 2018
Cancers of unknown primary site: ESMO Clinical

Practice Guidelines for diagnosis, treatment
and follow-up†
K. Fizazi1, F. A. Greco2, N. Pavlidis3, G. Daugaard4, K. Oien5 & G. Pentheroudakis3, on behalf of the
1
Department of Cancer Medicine, Institut Gustave Roussy, University of Paris Sud, Villejuif, France; 2Tennessee Oncology, Centennial Medical Center, Nashville, USA;
3
Department of Medical Oncology, University of Ioannina, Ioannina, Greece; 4Department of Oncology 5073, Rigshospitalet, Copenhagen University Hospital,
Copenhagen, Denmark; 5University of Glasgow, Institute of Cancer Sciences, Glasgow, UK
definition, incidence and biology progesterone receptors in females with axillary node metastases
is advisable to rule out hormone-sensitive tumours amenable to
Cancers of unknown primary site (CUPs) represent a heteroge- specific therapy. Staining for keratins CK7 and CK20 may
neous group of metastatic tumours for which a standardised provide indications of a possible primary site, and staining for
diagnostic work-up fails to identify the site of origin at the time chromogranin A and synaptophysin is needed to profile neu-
of diagnosis. CUPs account for 3%–5% of all malignancies. The roendocrine differentiation (Figure 1). Examples of stainings
unique biology of these tumours remains almost unknown [1]. that are rather specific include CK7+, WT-1+, PAX8+, CK20−
Nonetheless, current data suggest that metastatic dissemination (ovarian cancer) and RCC+, PAX8+ (renal cancer).
clinical practice
can occur in the absence of growth of a primary tumour by
guidelines
virtue of inherent metastatic aggressiveness of cancer cells.
Chromosomal instability was recently suggested to account for personalised medicine
part of the uncommon clinical presentation, chemoresistance Several gene expression profiling assays have become commer-
and poor outcome in patients with CUP [2]. cially available, claiming to blindly and correctly identify a
known primary cancer and a likely tissue of origin in patients
diagnosis, pathology and molecular with CUP in ∼80% [6, 7] [III]. These assays are based on
mRNA or miRNA RT-PCR or oligonucleotide microarray tech-
biology
nologies [8–10]. No significant differences in the tumour
Diagnosis of CUP requires pathology evaluation of a good microRNA expression profile were evident when CUP metasta-
quality tissue sample. These tumours are categorised by path- ses biologically assigned to a primary tissue of origin were com-
ology into: pared with metastases from typical solid tumours of known
origin [11]. These tests may aid in the diagnosis of the putative
• well- and moderately differentiated adenocarcinomas;
primary tumour site in some patients [12]. However, their
• squamous cell carcinomas;
impact on patient outcome via administration of primary site-
• carcinomas with neuroendocrine differentiation;
specific therapy remains questionable and unproven in rando-
• poorly differentiated carcinomas (including poorly differen-
mised trials [13] [IV, C]. A large prospective non-randomised
tiated adenocarcinomas);
phase II study of 252 patients suggested that survival may be
• undifferentiated neoplasms.
improved by site-specific therapy determined by a gene expres-
Immunohistochemistry should be applied meticulously [3, 4] sion profile assay of the biopsy specimen, particularly for
in order to identify the tissue of origin and to exclude chemo- patients with a tissue of origin diagnosis of more responsive
sensitive and potentially curable tumours (i.e. lymphomas and tumour types [7]. A prospective randomised phase III trial
germ-cell tumours) (Table 1) [III, A]. If the diagnosis is carcin- testing such a precision medicine strategy versus empirical
oma or adenocarcinoma, immunostaining for prostate-specific chemotherapy is currently on-going in Europe (NCT01540058).
antigen (PSA) in male patients and for oestrogen and
CH-6962 Viganello-Lugano, Switzerland. CUPs are by definition metastatic cancers, and the prognosis for
E-mail: clinicalguidelines@esmo.org patients with CUP is generally poor. However, an appropriate
†
diagnostic work-up can help to identify a minority of CUP
Approved by the ESMO Guidelines Committee: April 2002, last update July 2015. This
publication supersedes the previously published version — Ann Oncol 2011; 22 (Suppl. patients who can expect to benefit from directed therapy. The
6): vi64–vi68. following recommendations epitomise the standard and

by guest
on 05 February 2018
Table 1. Immunohistochemical work-up in patients with cancers of unknown primary site (CUPs)
Cytokeratins PSA ER, CDX2+, TTF1, Thyroglobulin, NSE, AFP, LCA S100, Vimentin,
PgR CK20+, NapsinA, calcitonin chromogranin, OCT4, HMB45 desmin
CK7– CK7+ synaptophysin hCG,
PLAP
Undifferentiated + − ± − − − − − − − ±
carcinoma
Prostate cancer + + − − − − − − − − −
Breast cancer + − ± − − − − − − − ±
Colorectal cancer + − − + − − − − − − −
Lung adenocarcinoma + − − − + − − − − − −
Thyroid cancer + − − − ± + ± − − − −
Neuroendocrine + − − − ± ± + − − − −
Germ-cell cancer + − − − − − − + − − ±
Lymphoma − − − − − − − − + − −
Melanoma − − − − − − − − − + ±
Sarcoma − − − − − − − − − ± +
The table shows general staining patterns but exceptions exist, especially for S100 and vimentin
Thyroid and neuroendocrine cancers often positive with CK7 and TTF1 but not with NapsinA.
PSA, prostate specific antigen; ER, oestrogen receptor; PgR, progesterone receptor; CK, cytokeratin; TTF1, thyroid transrcription factor 1; NSE, neuron-
specific enolase; AFP, alpha fetoprotein; hCG, human chorionic gonadotropin; PLAP, placental alkaline phosphatase; LCA, leukocyte common antigen.
Figure 1. Basic immunohistochemical work-up of cancers of unknown primary. Reproduced with permission: [5]. CK, cyrokeratin; CEA, carcinoembryonic
antigen; TTF1, thyroid transcription factor 1; ER, oestrogen receptor; PgR, progesterone receptor; GCDFP-15, gross cystic disease fluid protein-15; WT-1,
Wilms tumour gene 1; PSA, prostate specific antigen.
optional assessments suggested. Diagnostic and staging guide- extragonadal germ-cell tumours, neuroendocrine tumours and
lines for patients with an anticipatory CUP diagnosis are sum- prostate cancers amenable to hormonal treatment.
marised in Table 2. Whole-body 2-deoxy-2-[18F]fluoro-D-glucose-positron emis-
Thorough physical examination (including head and neck, sion tomography (FDG–PET)/CT may contribute to the man-
rectal, pelvic and breast examination), basic blood and biochem- agement of patients with cervical adenopathies from CUP and
ical analyses, and computed tomography (CT) scans of thorax, those with a single CUP metastasis [IV, B]. For other CUPs, the
abdomen and pelvis constitute a minimal basic work-up [IV, B]. role of FDG–PET is limited [14, 15], making this imaging pro-
Endoscopies should be sign-, symptom- or laboratory cedure not mandatory in the systematic work-up [III, C].
abnormality-guided. Serum assessment of α-fetoprotein, human For patients with predominant midline lymph node involve-
chorionic gonadotropin, plasma chromogranin A and PSA is ment, the diagnosis of lymphoma or extragonadal germ-cell
suggested in male patients to exclude potentially curable tumours should be ruled out.
v | Fizazi et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Table 2. Diagnostic and staging guidelines for cancers of unknown Distinct subsets of patients with CUP have been defined
primary site (CUPs) based on clinical and pathological criteria [2] (Table 3). An add-
Assessment suggested Target patient population
itional subset of CUP with a colorectal IHC or molecular profile
also seems to have a better prognosis, likely thanks to more
Thorough medical history and All patients active systemic treatments developed over the last two decades
physical examination for colon cancer [16]. A minority of patients (15%–20%) belong
Basic blood and biochemistry All patients to clinico-pathological subsets with a more favourable progno-
analyses sis. These favourable-risk CUP patients harbour chemosensitive
CT scans of thorax, abdomen and All patients and potentially curable tumours and may experience long-term
pelvis disease control with appropriate multidisciplinary management.
The majority of patients (80%–85%) do not belong to specific
Mammography Female patients
subsets. Sensitivity to therapy is only modest and median overall
Work-up for CUP subsets survival is generally <1 year (6–10 months). Two prognostic
Breast MRI Females with axillary groups can be identified among patients with CUP:
adenocarcinoma
• those with a good performance status (0–1) and a normal
Serum α-fetoprotein and human Patients with midline metastatic
lactate dehydrogenase (LDH) value, with a median life expect-
chorionic gonadotropin disease
ancy of 1 year, and
Serum prostate-specific antigen Males with adenocarcinomatous • those with either one or both of these prognostic factors
bone metastases ( poor performance status and elevated serum LDH), with a
Head and neck CT/PET scan Cervical squamous cell carcinoma median overall survival of only ∼4 months [17].
(optional)
A proposal for the practical management of patients with
Endoscopies Sign/symptom/laboratory-
CUP, including recognition of specific subsets, exclusion of
oriented
non-CUP neoplasms and the use of prognostic parameters in
Octreoscan and plasma Patients with neuroendocrine clinical practice, is summarised in Figure 2.
chromogranin A tumour CUP
Additional diagnostic pathology Sign/symptom/laboratory-
treatment
oriented
Therapy should be individually tailored according to the
CT, computed tomography; MRI, magnetic resonance imaging; PET, clinico-pathological subset with a distinct prognosis to which
positron emission tomography. the patient belongs [III, B]. Referral to specialised centres is
strongly encouraged. The 10%–15% of CUP patients in the
Table 3. Therapy for patients with favourable-risk cancers of unknown primary site (CUPs)
CUP subtype Proposed treatment Potential equivalent tumour
Poorly differentiated neuroendocrine carcinomas Platinum + etoposide combination chemotherapy Poorly differentiated neuroendocrine
of an unknown primary carcinomas with a known primary
Well-differentiated neuroendocrine tumour of Somatostatin analogues, streptozocin+5-FU, Well-differentiated neuroendocrine tumour
unknown primary sunitinib, everolimus of a known primary site
Peritoneal adenocarcinomatosis of a serous Optimal surgical debulking followed by platinum– Ovarian cancer
papillary histological type in females taxane-based chemotherapy
Isolated axillary nodal metastases in females Axillary nodal dissection, mastectomy or breast Breast cancer (found in 50%–70% when
irradiation and adjuvant chemohormonotherapy breast MRI is performed)
Squamous cell carcinoma involving non- Neck dissection and/or irradiation of bilateral neck Head and neck squamous cell cancer
supraclavicular cervical lymph nodes and head–neck axis. For advanced stages
induction chemotherapy with platinum-based
combination or chemoradiation
CUP with a colorectal IHC (CK20+ CDX2+ Systemic treatment used for colorectal cancer Metastatic colorectal cancer
CK7−) or molecular profile
Single metastatic deposit from unknown primary Resection and/or RT ± systemic therapy Single metastasis
Men with blastic bone metastases or IHC/serum Androgen deprivation therapy ± RT Prostate cancer
PSA expression
5-FU, 5-fluorouracil; MRI, magnetic resonance imaging; IHC, immunohistochemistry; PSA, prostate-specific antigen; RT, radiotherapy; CK, cytokeratin.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv305 | v

by guest
on 05 February 2018
favourable-risk subsets should be treated similarly to patients

with equivalent known primary tumours with metastatic dis-
semination [IV, B]. These patients achieve long-term disease
control in 30%–60% of cases, and optimal management is
pivotal for long-term survival (Table 3). Retrospective analyses
confirm that the clinical behaviour, biology, response to treat-
ment and outcome of patients with favourable-risk CUP are not
different from similar factors relative to metastatic tumours
from a known primary site [18–22].
Patients with poor-risk CUP have a dismal prognosis
despite management with a variety of chemotherapeutic com-
binations in small clinical studies [23]. A review conducted in
the 2000s showed no evidence of superior efficacy of any of
the administered regimens comprising platinum salts, taxanes
or new-generation cytotoxic compounds (gemcitabine, vinca
alkaloids or irinotecan) [24]. A randomised prospective phase
III study of 198 patients compared gemcitabine/irinotecan
with paclitaxel/carboplatin/oral etoposide in fit poor-risk
patients and reported significantly less toxicity with the two-
drug regimen and equal survival rates [II, A] [25]. On the
other hand, the efficacy/toxicity ratio of the cisplatin–gemcita-
bine combination was found to be better than that of the
cisplatin–irinotecan regimen in a randomised phase II trial
[I, A] [26]. Finally, better outcome was reported with the
two-drug cisplatin–gemcitabine regimen when compared
with cisplatin alone, although this was not assessed in a large
and adequately powered randomised phase III trial [27] (II).
Modest survival prolongation and symptom palliation with
preservation of quality of life are currently the only realistic
aims of therapy for these patients [II, A], although rare cases
of cure have been reported [28]. Consequently, low-toxicity
Figure 2. Clinical management of patients presenting with CUPs. IHC,
patient-convenient chemotherapy regimens should be admi-
immunohistochemistry; PS, performance status; LDH, lactate dehydrogen-
nistered to reasonably fit poor-risk CUP patients (Table 4). If
ase; OS, overall survival.
evaluation of patient demographics, metastatic pattern, results
Table 4. Commonly used low-toxicity palliative chemotherapy regimens for poor-risk patients with cancers of unknown primary site (CUPs)
Chemotherapy (mg/m2) Time Interval Comments
Cisplatin 60–75 + Day 1 Q3 weeks Fit patients, adequate hydration

Gemcitabine 1000 Day 1+8
Cisplatin 75 + Day 1 Q3 weeks Fit patients with neuroendocrine-feature CUP, adequate hydration
Etoposide 100 Day 1–3
Paclitaxel 175 + Convenient outpatient regimen, monitor neurotoxicity

Carboplatin AUC 5 Day 1 Q3 weeks
Docetaxel 75 + Convenient outpatient regimen, monitor neurotoxicity

Carboplatin AUC 5 Day 1 Q3 weeks
Irinotecan 160 + Outpatient regimen, monitor for neurotoxicity and diarrhoea

Oxaliplatin 80 Day 1 Q3 weeks
Oral capecitabine 2000 ± Days 1–14 Outpatient regimen, risk for diarrhoea and neurotoxicity
Oxaliplatin 85–130 Day 1 Q3 weeks
Gemcitabine 1000/irinotecan 100 Day 1+8 Q3 weeks Convenient outpatient regimen, monitor diarrhoea
AUC, area under the curve.
v | Fizazi et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Table 5. Levels of evidence and grades of recommendation are particularly relevant for patients with poor-risk CUP for whom
(adapted from the Infectious Diseases Society of America-United excessive treatment-related toxicity is not justified [IV, B].
follow-up and long-term implications
Levels of evidence
There is no evidence that follow-up of asymptomatic patients is
I Evidence from at least one large randomised, controlled trial of
needed. Specific examinations as clinically indicated.
heterogeneity
methodology
II Small randomised trials or large randomised trials with a suspicion These clinical practice guidelines were developed in accordance
of bias (lower methodological quality) or meta-analyses of such with the ESMO standard operating procedures for clinical prac-
trials or of trials with demonstrated heterogeneity tice guidelines development. The relevant literature has been
III Prospective cohort studies selected by the expert authors. Levels of evidence and grades of
IV Retrospective cohort studies or case–control studies recommendation have been applied using the system shown in
Grades of recommendation standard clinical practice by the experts and the ESMO faculty.
This manuscript has been subjected to an anonymous peer
strongly recommended review process.
benefit, generally recommended acknowledgements
C Insufficient evidence for efficacy or benefit does not outweigh the The authors thank Lorna Saint Ange for editing the manuscript.
risk or the disadvantages (adverse events, costs, … ), optional
generally not recommended conflict of interest
The authors have declared no conflicts of interest.
recommended
a
By permission of the Infectious Diseases Society of America [32]. references
1. Massard C, Loriot Y, Fizazi K. Carcinomas of an unknown primary origin—
diagnosis and treatment. Nat Rev Clin Oncol 2011; 8: 701–710.
2. Vikeså J, Møller AK, Kaczkowski B et al. Cancers of unknown primary origin (CUP)
of clinical and laboratory tests, imaging data, pathological
are characterized by chromosomal instability (CIN) compared to metastasis of
evaluations and gene expression is relatively unambiguous, a know origin. BMC Cancer 2015; 15: 151.
site-specific treatment may be considered, though prospective 3. Oien KA. Pathologic evaluation of unknown primary cancer. Semin Oncol 2009;
evidence that this is better than empirical chemotherapy is 36: 8–37.
lacking so far. 4. Abbruzzese JL, Abbruzzese MC, Lenzi R et al. Analysis of a diagnostic strategy for
Whether targeted agents should be used or not in patients patients with suspected tumors of unknown origin. J Clin Oncol 1995; 13:
with CUPs is still unknown [29]. Although only a few non- 2094–2103.
chemotherapy drugs have been tested in patients with CUP, 5. Varadhachary GR. Carcinoma of unknown primary origin. Gastrointest Cancer Res
belinostat was randomly assessed and it did not improve the 2007; 1: 229–235.
results of the carboplatin-paclitaxel regimen [30]. Preliminary 6. Greco FA, Lennington WJ, Spigel DR, Hainsworth JD. Molecular profiling diagnosis
in unknown primary cancer: accuracy and ability to complement standard
retrospective data suggest that CUP patients with immunohisto-
pathology. J Natl Cancer Inst 2013; 105: 782–790.
chemical and/or molecular profile assay diagnoses of ‘colorectal’ 7. Hainsworth JD, Rubin MS, Spigel DR et al. Molecular gene expression profiling to
carcinomas have response rates and survival after colorectal site- predict the tissue of origin and direct site-specific therapy in patients with
specific therapies (i.e. FOLFOX or FOLFIRI) that are similar to carcinoma of unknown primary site: a prospective trial of the Sarah Cannon
those of patients with known advanced colorectal carcinomas Research institute. J Clin Oncol 2013; 31: 217–223.
[16, 31] [IV, B]. These data are from small numbers of patients, 8. Pillai R, Deeter R, Rigl CT et al. Validation and reproducibility of a microarray-based
and additional prospective validation is necessary to substantiate gene expression test for tumor identification in formalin-fixed, paraffin-embedded
these preliminary findings. specimens. J Mol Diagn 2011; 13: 48–56.
9. Kerr SE, Schnabel CA, Sullivan PS et al. Multisite validation study to determine
Participation in clinical trials evaluating combinations of
performance characteristics of a 92‐gene molecular cancer classifier. Clin Cancer
cytotoxic compounds with targeted agents or site-specific Res 2012; 18: 3952–3960.
therapy in patients with putative primary tumour sites highly 10. Meiri E, Mueller WC, Rosenwald S et al. A second-generation microRNA-
suspected from immunohistochemical or microarray studies based assay for diagnosing tumor tissue origin. Oncologist 2012; 17:
should be strongly encouraged. 801–812.
11. Pentheroudakis G, Spector Y, Krikelis D et al. Global microRNA profiling in
favorable prognosis subgroups of cancer of unknown primary (CUP) demonstrates
response evaluation no significant expression differences with metastases of matched known primary
tumors. Clin Exp Metastasis 2013; 30: 431–439.
Response evaluation is recommended after two or three chemo- 12. Gross-Goupil M, Massard C, Lesimple T et al. Identifying the primary site
therapy cycles by individually adequate tests. Quality-of-life issues using gene expression profiling in patients with carcinoma of an unknown
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv305 | v

by guest
on 05 February 2018
primary (CUP): a feasibility study from the GEFCAPI. Onkologie 2012; 35: 23. Bugat R, Bataillard A, Lesimple T et al. Summary of the standards, options and
54–55. recommendations for the management of patients with carcinoma of unknown
13. Hainsworth JD, Greco FA. Gene expression profiling in patients with carcinoma of primary site (2002). Br J Cancer 2003; 89(Suppl 1): S59–S66.
unknown primary site: from translational research to standard of care. Virchows 24. Golfinopoulos V, Pentheroudakis G, Salanti G et al. Comparative survival with
Arch 2014; 464: 393–402. diverse chemotherapy regimens for cancer of unknown primary site: multiple-
14. Moller AK, Loft A, Berthelsen AK et al. A prospective comparison of 18F-FDG PET/ treatments meta-analysis. Cancer Treat Rev 2009; 35: 570–573.
CT and CT as diagnostic tools to identify the primary tumor site in patients with 25. Hainsworth JD, Spigel DR, Clark BL et al. Paclitaxel/carboplatin/etopside versus
extracervical carcinoma of unknown primary site. Oncologist 2012; 17: gemcitabine/irinotecan in the first-line treatment of patients with carcinoma of
1146–1154. unknown primary site: a randomised phase III Sarah Cannon Research Consortium
15. Seve P, Billotey C, Broussolle C et al. The role of 2-deoxy-2-[F-18]fluoro-d-glucose Trial. Cancer J 2010; 16: 70–75.
positron emission tomography in disseminated carcinoma of unknown primary 26. Culine S, Lortholary A, Voigt JJ et al. Cisplatin in combination with either
site. Cancer 2007; 109: 292–299. gemcitabine or irinotecan in carcinomas of unknown primary site: results of a
16. Hainsworth JD, Schnabel CA, Erlander MG et al. A retrospective study of treatment randomised phase II study—Trial for the French Study Group on Carcinomas of
outcomes in patients with carcinoma of unknown primary site and a colorectal Unknown Primary (GEFCAPI 01). J Clin Oncol 2003; 21: 3479–3482.
cancer molecular profile. Clin Colorectal Cancer 2012; 11: 112–118. 27. Gross-Goupil M, Fourcade A, Blot E et al. Cisplatin alone or combined with
17. Culine S, Kramar A, Saghatchian M et al. Development and validation of a gemcitabine in carcinomas of unknown primary: results of the randomized
prognostic model to predict the length of survival in patients with carcinomas of an GEFCAPI 02 trial. Eur J Cancer 2012; 48: 721–727.
unknown primary site. J Clin Oncol 2002; 20: 4679–4683. 28. Levy A, Massard C, Gross-Goupil M, Fizazi K. Carcinomas of an unknown primary
18. Spigel DR, Hainsworth JD, Greco FA. Neuroendocrine carcinoma of unknown site: a curable disease? Ann Oncol 2008; 19: 1657–1658.
primary site. Semin Oncol 2009; 36: 52–59. 29. Massard C, Voigt JJ, Laplanche A et al. Carcinoma of an unknown primary: are
19. Pentheroudakis G, Pavlidis N. Serous papillary peritoneal carcinoma: unknown EGF receptor, Her-2/neu, and c-Kit tyrosine kinases potential targets for therapy?
primary tumour, ovarian cancer counterpart or a distinct entity? A systematic Br J Cancer 2007; 97: 857–861.
review. Crit Rev Oncol Hematol 2010; 75: 27–42. 30. Hainsworth JD, Daugaard G, Lesimple T et al. Paclitaxel/carboplatin with or without
20. Pentheroudakis G, Lazaridis G, Pavlidis N. Axillary nodal metastases from belinostat as empiric first-line treatment for patients with carcinoma of unknown
carcinoma of unknown primary (CUPAx): a systematic review of published primary site: a randomized, phase 2 trial. Cancer 2015; 121: 1654–1661.
evidence. Breast Cancer Res Treat 2010; 119: 1–11. 31. Varadhachary GR, Raber MN, Matamorous A, Abbruzzese JL. Carcinoma of
21. Pavlidis N, Pentheroudakis G, Plataniotis G. Cervical lymph node metastases of unknown primary with colon-cancer profile-changing paradigm and emerging
squamous cell carcinoma from an unknown primary site: a favourable prognosis definitions. Lancet Oncol 2008; 9: 596–599.
subset of patients with CUP. Clin Transl Oncol 2009; 11: 340–348. 32. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
22. Hainsworth JD, Fizazi K. Treatment for patients with unknown primary cancer and among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
favorable prognostic factors. Semin Oncol 2009; 36: 44–51. 139–144.
v | Fizazi et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Published online 7 April 2016
Acute lymphoblastic leukaemia in adult patients: ESMO

and follow-up†
D. Hoelzer1, R. Bassan2, H. Dombret3, A. Fielding4, J. M. Ribera5 & C. Buske6 on behalf of the
1
ONKOLOGIKUM Frankfurt am Museumsufer, Frankfurt, Germany; 2Hematology Unit, Ospedale dell’Angelo e Ospedale SS. Giovanni e Paolo, Mestre-Venezia, Italy;
3
Institut Universitaire d‘Hematologie Hopital St Louis, Paris, France; 4Cancer Institute, University College London, London, UK; 5Department of Clinical Hematology,
ICO-Hospital Germans Trias i Pujol, Jose Carreras Research Institute, Universitat Autonoma de Barcelona, Barcelona, Spain; 6CCC Ulm, Institut für Experimentelle
Tumorforschung, Universitätsklinikum Ulm, Ulm, Germany
incidence and epidemiology • distinguish B-cell precursor (BCP) ALL from T-cell ALL (T-
ALL),
The estimated overall incidence of acute lymphoblastic leukaemia • distinguish Burkitt leukaemia (B-ALL) from BCP-ALL (differ-
(ALL) and lymphoblastic lymphoma in Europe is 1.28 per 100 000 ent treatment required),
individuals annually, with significant age-related variations (0.53 at • distinguish Philadelphia (Ph) chromosome-positive (Ph+)
45–54 years, ∼1.0 at 55–74 years and 1.45 at 75–99 years) and that ALL from Ph-negative (Ph−) ALL (different treatment
clinical practice
of Burkitt leukaemia/lymphoma is between 0.17 and 0.33 in the required), and
guidelines
same age groups [1]. These figures qualify ALL as a rare disease in • shorten time to treatment start.
adults, making assessment and care at qualified centres highly de-
sirable. Predisposing risk factors for adult ALL are not known, con- Aspiration of bone marrow is a standard procedure, with a core
trary to childhood ALL [2]. Therapeutic progress is undeniable as marrow biopsy being necessary only in case of insufficient cell
shown by large registry data. In Europe, 5-year overall survival yield. The bone marrow must contain at least 20% blast cells, as
(OS) improved from 29.8% in the years 1997–1999 to 41.1% in a criterion to differentiate ALL from lymphoblastic lymphoma
2006–2008 (P < 0.0001), still as a function of age. Compared with with/without marrow involvement, even if therapeutic conse-
the reference group (age 15–54 years: OS >50%), OS was <30% in quences are very limited. The proportion of circulating blasts is
the 55–64 years age group (hazard ratio 2.05) and <20% in the highly variable. ALL blasts are atypical lymphoid or undifferen-
≥65 years age group (hazard ratios 2.71 and 3.75) [3]. tiated cells. Once minimally differentiated acute myelogenous
leukaemia (AML) has been ruled out, the morphological ana-
diagnosis and pathology/molecular lysis is uninformative in ALL, if not for the common association
between FAB L3 morphology and B-ALL [7]. The immunophe-
biology
notype study plays the key diagnostic role, demonstrating com-
A comprehensive diagnostic approach requires the study of cell mitment of the blast cell population to the B- or T-cell lineage.
morphology, immunophenotype, genetics/cytogenetics and The European Group for the Immunological Characterization
genomics, as detailed in the 2008 World Health Organization of Leukemias (EGIL) recognised distinct BCP/T-ALL subsets,
(WHO) classification [4] and recently reviewed [I, A] [5, 6]. providing a rational immunological classification along with cri-
teria for differential diagnosis [8]. Original EGIL standards and
morphology/immunophenotype/molecular definitions of mixed-lineage leukaemias (MLLs) variously expres-
screening sing B-, T- and myeloid-associated antigens were updated and
The initial diagnostic work-up (Table 1) must be carried out improved [9, 10]. Further indications on how to best perform diag-
expeditiously and before any chemotherapy (within 1–2 nostic flow cytometry were presented by a panel of experts [11].
working days) to: The early diagnostic step is completed by a rapid molecular screen-
ing by means of reverse transcriptase polymerase chain reaction
• confirm ALL diagnosis, (RT-PCR) or fluorescence in situ hybridisation (FISH) assays pri-
marily for the detection of BCR-ABL1 gene rearrangements, de-
*Correspondence to: ESMO Guidelines Committee, ESMO Head Office, Via L. Taddei 4, noting an underlying t(9;22)(q34;q11)/BCR-ABL1 chromosomal
translocation typical of Ph+ ALL and sensitive to targeted therapy
with tyrosine kinase inhibitors (TKIs) [I, A] [12].
†
Approved by the ESMO Guidelines Committee: February 2016.

by guest
on 05 February 2018
Table 1. Diagnostic work-up in adult ALL

Diagnostic step Results/ALL subsets Recommendations
Morphology
– Bone marrow and peripheral blood – Lymphoid/undifferentiated blasts Mandatory
– Cerebro-spinal fluid (≥20% bone marrow involvement)
– FAB L3 morphology in Burkitt leukaemia Recommended
– CNS involvement Mandatory
Immunophenotype
– MPO (differential diagnosis versus AML) – MPO negative; B/T markers >20% Mandatory
– B-lineage markers: CD19, CD79a, cCD22 (at least 2); others: TdT, CD10, (CD3, CD79a >10%)
CD20, CD24, cIgM, sIg (kappa or lambda) – B-lineage ALL: Mandatory
– T-lineage markers: cCD3; others: TdT, CD1a, CD2, CD5, CD7 CD4, Pro-B/B-I (CD19/CD79a/cCD22+)
CD8, TCR α/β or γ/δ Common/B-II (CD10+/cIgM−)
– Stem/myeloid cell markers (variable): CD34, CD13, CD33, CD117 Pre-B/B-III (cIgM+/sIg−)
Mature-B/B-IV (sIg+)
– T-lineage ALL: Mandatory

Pro-T/T-I (cCD3/CD7+)
Pre-T/T-II (CD2/CD5)
Cortical-T/T-III (CD1a+)
Mature-T/T-IV (CD3+/CD1a−)
Cytogenetics/genetics
– Cytogenetics/FISH/RT-PCR – ALL with adverse clinico-biological features: Mandatory
Ph+ ALL (rapid detection, to TKI therapy)
t(4;11)+ ALL
t(1;19)+ ALL
other high-risk cytogenetics
– CGH/SNP/GEP/NGS – ALL with adverse clinico-biological features: Recommended for

Ph-like ALL new clinical trials
ETP ALL
NOTCH1/FBW7-unmutated/RAS/
PTEN-altered T-ALL
IKZF1, CLRF2, MLL, TP53, CREBBP,
RAS alterations
MRD study
– MRD marker(s): LAIP (immunophenotype)/molecular probe (PCR) – MRD-based risk classification Mandatory
Storage of diagnostic material

– Cell banking/storage of DNA/RNA/protein lysates – Additional/future studies Highly recommended
HLA typing
– Patient/siblings – Early application of SCT if required Recommended
ALL, acute lymphoblastic leukaemia; CNS, central nervous system; MPO, myeloperoxidase; AML, acute myelogenous leukaemia; c, cytoplasmic; IgM,
immunoglobulin M; s, surface; Ig, immunoglobulin; FISH, fluorescence in situ hybridisation; RT-PCR, reverse transcriptase polymerase chain reaction;
Ph+, Philadelphia-positive; TKI, tyrosine kinase inhibitor; CGH, comparative genomic hybridisation; SNP, single nucleotide polymorphism; GEP, gene
expression profiling; NGS, next-generation sequencing; Ph, Philadelphia; ETP, early T-cell precursor; T-ALL, T-cell ALL; MRD, minimal residual disease;
LAIP, leukaemia-associated immunophenotype; PCR, polymerase chain reaction; HLA, human leucocyte antigen; SCT, stem-cell transplantation.
cytogenetics/genetics • t(4;11)(q21;q23)/MLL-AFA4, abn11q23/MLL, t(1;19)(q23;

Results from cytogenetics, genetics and genomics are available at p13)/PBX-E2A, t(8;14) or other abn14q32 in non-Burkitt ALL,
a later stage, allowing the recognition of several ALL syndromes • del(6q), del(7p), del(17p), −7, +8, low hypodiploidy, i.e.
with prognostic and/or therapeutic implications (reviewed in with 30–39 chromosomes/near triploidy with 60–78 chromo-
references [2] and [6]). Standard cytogenetics/FISH and especial- somes,
ly RT-PCR are routinely performed to obtain a rapid diagnosis of • complex (≥5 unrelated clonal abnormalities), and
Ph+ ALL and identify other intermediate/high- and high-risk • T-ALL lacking NOTCH1/FBXW7 mutations and/or with
karyotypes/gene rearrangements, mainly: RAS/PTEN abnormalities [I, A] [13–17].
v | Hoelzer et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
The more prognostically favourable cytogenetic/genetic subsets risk assessment and prognostic factors
are t(12;21)( p13;q22)/TEL-AML1 + ALL (rare in adults) and
hyperdiploid ALL, and NOTCH-1/FBXW7-mutated T-ALL. While the suggested diagnostic work-up permits the identifica-
tion of some high-risk (HR) subsets, clinical risk groups are
further defined by several disease-related factors and some hos-
new genetics/genomics t-related factors [20, 21], and the individual prognosis is highly
refined by ALL response dynamics (Table 2). Patients presenting
The integration of above studies with new genetics/genomics, i.e.
with no risk factors are defined as standard risk (SR). Older age,
array-comparative genomic hybridisation, gene expression profiling,
reduced tolerability to treatments and higher white blood cell
single-nucleotide polymorphism array analysis and next-generation
(WBC) count on presentation (reflecting higher tumoural
sequencing, led to the recognition of highly specific poor-risk condi-
burden) are universally recognised as independent risk variables
tions, whose global incidence is ∼30%. These are: Ph-like ALL,
predicting for lower complete remission (CR) rate and shorter
characterised by a gene expression profile similar to Ph+ ALL and
CR duration. The kinetics of response to early treatment steps is
associated with IKZF1 deletion, CLRF2 overexpression and tyrosine
also well recognised and increasingly sought-for prognostic in-
kinase-activating rearrangements involving ABL1, JAK2, PDGFRB
formation. This can be obtained through different methodolo-
and several other genes [16]; and early T-cell precursor (ETP) ALL,
gies and at different treatment times, ranging from pre-phase
characterised by lack of CD1a and CD8, weak CD5 expression, at
therapy ( prednisone response) to induction day 8–15 (marrow
least one myeloid/stem cell marker, a specific transcriptional profile
blast cell clearance), end of induction (time to CR, MRD) and
and the possible involvement of several critical genes [18]. Other
post-induction phase (MRD) [III, A].
genetic aberrations that impart an inferior outlook are other MLL
gene rearrangements, TP53 and CREBBP mutations, and deregula-
tion of RAS signalling components (NRAS, KRAS, FLT3, NF1). minimal residual disease
Although these assays are still investigational and not regularly Quantification of MRD is a major and well-established risk factor
carried out in the clinical practice, they are recommended for new and should be obtained whenever possible for all patients also
clinical trials to improve the risk classification and support targeted outside of clinical trials. Methods for MRD evaluation and stand-
therapies [III, B]. ardisation of MRD quantifıcation have been intensively described
[22–24]. Molecular response can be evaluated only for patients in
complete cytologic remission (Table 3), with one marker or more
other. The diagnostic phase is completed by the search for a for MRD analysis and samples available at the respective time
sensitive molecular marker or an aberrant leukaemia-associated points. Definition of responses are summarised in Table 3. If
immunophenotype (LAIP) for the detection and monitoring of MRD is measured by flow cytometry, a good MRD response is
minimal residual disease (MRD) [III, B] [19]. Human leucocyte often defıned as less than 10−3, although MRD levels less than
antigen (HLA) typing of patients and relatives is recommended 10−4 can be achieved with the 8–12 colour flow cytometers.
at this stage, to facilitate subsequent application of an early Achievement of complete molecular remission (molCR)/
stem-cell transplantation (SCT), according to study/treatment molecular remission is the most relevant independent prognos-
indications [V, B]. tic factor for disease-free survival (DFS) and OS. Patients with
Table 2. High-risk factors in adult ALL

Risk factors Risk subsets (notes) Recommendations
Patient-related
– Age (years) – >40/55/65 Mandatory
– Performance status (ECOG score) – >1 Highly recommended
Disease-related
– WBC (×109/l) – >30 (B-lineage)/>100 (T-lineage) Mandatory
– Immunophenotype (B-T-subsets) – Pro-B/early and mature-T Mandatory
– Cytogenetics (karyotype) – Ph+/t(4;11)+/other adverse Mandatory
– Genetics – BCR-ABL1+/MLL+/PBX-E2A+/ Mandatory
Ph-like/IKZF1del/ETP/unmutated NOTCH1 Recommended for new clinical trials
– Miscellaneous – Central nervous system involvement Mandatory
Response dynamics
– corticosteroid sensitivity (blast count after pre-phase) – Poor prednisone response (≥1 × 109/l) Recommended
– early blast cell response (BM morphology) – Day 8–15 blasts ≥5% Recommended
– time to CR (no. of courses) – >1 cycle (late CR) Mandatory
– MRD (molecular/LAIP) – MRD+ (post-induction) Mandatory
ALL, acute lymphoblastic leukaemia; ECOG, Eastern Cooperative Oncology Group; WBC, white blood cells; Ph+, Philadelphia-positive; Ph, Philadelphia;
ETP, early T-cell precursor; BM, bone marrow; CR, complete remission; MRD, minimal residual disease; LAIP, leukaemia-associated immunophenotype.
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw025 | v

by guest
on 05 February 2018
Table 3. Response parameters according to MRD

Terminology Definitions
CR (complete haematological remission) – Leukaemic cells not detectable by light microscopy in BM/PB/CSF (BM < 5% blasts)
MolCR (complete molecular remission/MRD negativity) – Patient in CR
– MRD not detectable by sensitive molecular probe(s) (sensitivity ≥10−4)
MolR (molecular/MRD response, less than molCR) – Patient in CR, not in molCR
– Low-level non-quantifiable MRD (<10−4/0.01%, i.e. <1 leukaemic cell in 10 000)
– Assessable by MFC (lower detection limit, between 10−3 and 10−4, higher sensitivity
with 8–12 colour techniques)
MolFail (molecular failure/MRD positivity) – Patient in CR, not in molCR/molR
– Quantifiable MRD (≥10−4/0.01%, i.e. ≥1 leukaemic cell in 10 000)
– Assessable by MFC (lower detection limit, between 10−3 and 10−4)
MolRel (molecular/MRD relapse) – Patient still in CR, prior molCR/molR

– Loss of molCR/molR status (≥10−4/0.01%, i.e. >1 leukaemic cell in 10 000)
– Assessable by MFC (lower detection limit, between 10−3 and 10−4)
Relapse – Loss of CR status
– Haematological relapse (BM ALL blasts >5%)
– Extramedullary relapse (CNS, other site)
MRD, minimal residual disease; BM/PB/CSF, bone marrow/peripheral blood/cerebro-spinal fluid; MFC, multiparameter flow cytometry; ALL, acute
lymphoblastic leukaemia; CNS, central nervous system.
molCR after induction therapy in several studies had significant- or dexamethasone 6–16 mg/day, both i.v. or p.o.) alone, or in
ly superior outcomes, with a DFS of 54%–74%, compared combination with another drug (e.g. vincristine, cyclophospha-
with 17%–40% for MRD-positive patients [25–31]. Patients mide), is often given together with allopurinol and hydration for
with molecular failure (molFail) after induction proceeded to ∼5–7 days. The first intra-thecal therapy for central nervous
allogeneic haematopoietic SCT, and their outcome was thereby system (CNS) prophylaxis is administered in this period in
substantially improved, compared with the chemotherapy-only some studies. The pre-phase therapy allows a safe tumour re-
arm [29, 32, 33]. duction, avoiding in most cases a tumour lysis syndrome (TLS)
The question arises as to whether the evaluation of MRD over- [35]. In some cases, rasburicase may be given to prevent TLS. In
comes all of the pre-therapeutic risk factors, or whether MRD cases with a very high WBC count (e.g. >100 000/µl), either
should be combined with the pre-therapeutic factors [27, 34, 35]. measure is sufficient, and a leukapheresis is needed only in very
A practical approach is to bring the conventional prognostic factors rare cases. The time needed for pre-phase therapy will also allow
and MRD into a decision algorithm. At diagnosis, patients are to obtain the results of the diagnostic work-up, e.g. cytogenetics,
stratifıed into SR and HR groups, since HR patients are potential molecular genetics. The response to pre-phase therapy defines
candidates for SCT in first complete remission (CR1), and an early the chemosensitivity of the disease, and is included in some
donor search is warranted. However, it is not clear how to proceed studies for risk assessment, since good responders to prednisone
with HR patients in molCR/molecular remission, although some may have a better outcome [36].
studies suggest a lack of benefit from SCT in these patients. Also, Supportive therapy should be initiated whenever necessary
MRD is not available for all patients, and the risk stratification in early on, e.g. to treat infections or to substitute platelets/erythro-
those patients should rely on conventional risk factors. Overall, a cytes. Severe neutropaenia (<500/µl) is often seen at diagnosis
rapid yet comprehensive diagnostic approach is essential for accur- and is most frequent (>80%) during induction therapy, causing
ate risk definitions and appropriate risk-related treatment choices infections and infection-related death. A joint analysis of five
(Figure 1). randomised trials revealed a shorter duration of neutropaenia,
and reduction in the rate of febrile neutropaenia in some but not
all cases [37], and based on that, prophylactic granulocyte
treatment of newly diagnosed ALL colony-stimulating factor should be considered during induc-
tion therapy [II, B].
pre-phase therapy and supportive measures
When the diagnosis is established, treatment should start imme-
diately, preferably in a specialised hospital; that is, physicians treatment: remission induction therapy
with experience in the treatment of acute leukaemia, a well- and consolidation
trained nursing staff, sufficient supportive care (e.g. platelet sub- induction of complete remission. The goal of induction therapy
stitution) and access to an intensive care unit. A pre-phase is the achievement of a CR, or even better, a molCR/good
therapy with corticosteroids (usually prednisone 20–60 mg/day molecular response, usually evaluated within 6–16 weeks from
v | Hoelzer et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Figure 1. Diagnosis and risk assessment in adult ALL for achievement of CR and risk-oriented post-remission therapy. Major diagnostic subsets are identified
within 1–2 days to allow start of pre-phase therapy, identify cases eligible to targeted therapy (TKI in Ph+ ALL), and set up the MRD study. Pre-phase therapy
allows for management/prevention of metabolic/infectious/haemorrhagic complications before start of induction therapy, and checks HLA identity between patient
and siblings. Induction/early consolidation therapy (incorporating CNS prophylaxis) aims to induce CR with a deep MRD response, to support subsequent risk-
and MRD-oriented therapy with/without allogeneic SCT. ALL, acute lymphoblastic leukaemia; RT-PCR, reverse transcriptase polymerase chain reaction; MRD,
minimal residual disease; LAIP, leukaemia-associated immunophenotype; WBC, white blood cells; CR, complete remission; CNS, central nervous system;
SR, standard risk; HR, high risk; SCT, stem-cell transplantation; TKI, tyrosine kinase inhibitor; Ph+, Philadelphia-positive; HLA, human leucocyte antigen.
start of chemotherapy, after which time the achievement of molCR asparaginase (PEG-Asp) has the advantage of a significantly longer
is rather uncommon. Most regimens are centred on vincristine, period of asparagine depletion. Dexamethasone is often preferred to
corticosteroids, and anthracycline (daunorubicin, doxorubicin, prednisone, since it penetrates the blood–brain barrier and also acts
rubidazone, idarubicin), with or without cyclophosphamide or on resting leukaemic blast cells (LBCs).
cytarabine. L-Asparaginase is the only ALL-specific drug that
depletes the asparagine levels and has been particularly explored in results of induction therapy. There are no randomised trials
paediatric trials. It is now more intensively used in adults. Pegylated comparing different induction regimens; however, there is a
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw025 | v

by guest
on 05 February 2018
substantial number of large (>100 patients) prospective non- until blast clearance in the spinal fluid. Their OS is identical to the
randomised trials. In 6617 patients from 14 studies, the weighted CNS-negative cohort of patients, or slightly inferior [41].
mean for the CR rate was 83% (62%–92%) [35]. Using current
approaches, the CR rate had increased to 80%–90%, higher for SR age-adapted protocols
patients at ≥90%, and less in HR patients at ∼75%. There is only one
The outcome of ALL is strictly related to the age of a patient, with
randomised study for induction therapy; this compares prednisone to
cure rates from 80% to 90% in childhood ALL, decreasing to
dexamethasone [38], demonstrating equal outcome [I, C].
<10% in elderly/frail ALL patients. Therefore, age-adapted proto-
cols have emerged, where the age limits are mainly directed by the
treatment principles. There are two chemotherapy regimens; one
haematological and non-haematological toxicities. Although there
is a widespread schema patterned after the paediatric BFM (Berlin–
is no uniform consensus, the following age groups are separated:
Frankfurt–Münster) protocols with Induction I, Induction II,
Consolidation cycles, sometimes an intermittent re-induction • Adolescents and young adults (AYA), differently defined
cycle, and is mostly used in European adult ALL trials. from 15/18 to 35/40 years,
A schematic treatment algorithm in adult ALL, including • Adult ALL, age range 35/40 up to ≤55/60 years,
diagnosis and risk assessment for achievement of CR and risk- • Elderly ALL protocols for patients above the age of >55/60
oriented post-remission therapy, is given in Figure 1. years, and
Another approach is to repeat two different alternating • Frail patients not suitable for any intensive therapy, usually
intensive chemotherapy cycles, identical for Induction and considered above the age of 70/75 years.
Consolidation, accounting for a total of eight cycles, such as the
hyper-CVAD (cyclophosphamide, vincristine, doxorubicin, dexa-
methasone) protocol, preferentially used in the United States, but adolescents and young adults. Paediatric-inspired therapy provides
also in other parts of the world. an increased drug intensity at several treatment steps, including
larger cumulative doses of drugs such as corticosteroids, vincristine,
L-asparaginase and consequent CNS-directed therapy, which
post-remission consolidation. The rationale to use systemic
should be strictly adhered to, with a reduced role of SCT. In a
high-dose (HD) therapy is particularly to reach sufficient drug
systemic review and meta-analysis in 2012, in 11 trials including
levels in sanctuary sites, such as the CNS. Most protocols employ
2489 AYA patients, paediatric-inspired regimens were superior
6–8 courses which contain either HD methotrexate or HD
to conventional adult chemotherapy [42]. None of the trials
cytarabine ± asparaginase. HD cytarabine is usually administered
were a randomised comparison. In recent studies for AYAs [43–
for 4–12 doses at 1–3 g/m2 and methotrexate at 1–1.5 g/m2 and
45], survival rates at 5 years were 67%–78%, compared with
up to 3 g/m2.
34%–41% with the former protocols.
maintenance therapy
adult ALL. The treatment results for adult ALL patients have
Maintenance therapy usually consists of daily 6-mercaptopurine also improved. In the above-mentioned 14 studies, the weighted
and weekly methotrexate. In some treatment regimens, repeated mean for DFS was 34% (25% at 5 years, 48% at 3 years) and the
cycles of vincristine, dexamethasone or other drugs in monthly OS 38% (27% at 9 years, 54% at 5 years). Currently, the OS rates
or longer intervals are given. In one randomised study, the for SR adult ALL patients is 50%–70% with chemotherapy alone.
maintenance arm with reinforcement cycles was not superior to The outcome for HR patients has also improved, from 20%–30%
conventional maintenance therapy (37% versus 38% at 8 years) to ∼50% when they receive an allogeneic SCT in CR1. Prospective
[36]. A treatment duration of 2.5–3 years is optimal and is adult studies applying the same drugs and time–dose intensity,
usually recommended. using or not using the term ‘paediatric-inspired’, or some using
Omission of maintenance worsens outcome significantly in the term ‘paediatric-derived’, achieved identical results compared
BCP-ALL, but less so in T-ALL [39], and not in B-ALL [40]. with AYAs, with survival rates of 60%–70% or more [46–49].
CNS prophylaxis elderly ALL. The incidence of ALL is increasing after the age of
Effective prophylaxis to prevent CNS relapse is an essential part of 50 years [50]. Different approaches have been applied in this
ALL therapy. Treatment modalities of CNS prophylaxis are CNS patient cohort [51]. Older patients (55–91 years) with a palliative
irradiation, intra-thecal (i.th.) methotrexate, mono- or i.th. triple treatment had a CR rate of 43% (34%–53%), an early death rate
(usually methotrexate, steroids, cytarabine) and systemic HD of 24% (18%–42%) and an OS of only 7 months (3–10 months).
therapy with either methotrexate and/or cytarabine. With a com- In contrast, those with an intensive chemotherapy designed for
bination of these CNS prophylactic measures, the CNS relapse adult ALL had a CR rate of 56% (40%–81%), but still an early
rate in recent adult ALL trials could be reduced from 10% to <5%. death rate of 23% (6%–42%), and an OS of 14 months (3–29
CNS irradiation is also effective to eradicate residual LBCs in the months). In recent decades, several elderly specific ALL protocols
CNS; however, in most studies, it is either omitted or restricted to have been initiated. Their principle is a less intensive therapy,
HR patients. Furthermore, it is given only as an irradiation of the based on corticosteroids, vincristine and asparaginase, and largely
skull (mostly 24 Gy), and no longer of the whole neuroaxis, since avoiding anthracyclines and alkylating agents, to reduce early
this aggravates cytopaenias. Patients with CNS involvement treatment-related death. In nine prospective studies for older ALL
(mostly of the leptomeninges) at diagnosis are treated with the patients (55–81 years), with this less intensive protocol, the CR
standard chemotherapy regimen, and additional i.th. applications rate was 71% (43%–90%), early death decreased to 15% (0%–36%)
v | Hoelzer et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
and OS was significant at 33 months (16–71 months). Thus, all SCT relapse after MAC-SCT [70]. This remains to be studied
patients, irrespective of age, should be offered a treatment. after RIC-SCT.
Whether a subset of good-risk patients may be treated with
targeted therapies continuous TKI/chemotherapy and be proposed allogeneic SCT
in first CR is still under evaluation, as is the optimal TKI/
There is still a need to improve the outcome in adult and elderly chemotherapy combination that could be offered to such
ALL that is achieved with chemotherapy/SCT alone. Currently, patients. Achievement of a good early MRD response (here eval-
there are two major new approaches, particularly for B-lineage uated on BCR-ABL1 transcript levels) might be of great help in
ALL patients; either therapies with antibodies, or, for Ph+ ALL, defining these good-risk patients [71]. In older Ph+ ALL
targeted therapies with TKIs. patients, usually not candidates for allogeneic SCT, a poor MRD
response but also the presence of additional chromosomal ab-
antibodies. The anti-CD20 MoAb rituximab has substantially normalities at diagnosis were both associated with a worse
improved the outcome in Burkitt leukaemia/lymphoma, as outcome [72]. It has also been shown that the presence of IKZF1
demonstrated by a single randomised trial [52]. Repeated short gene deletion, a frequent event in Ph+ ALL, may be of poor
cycles of intensive chemotherapy, combined with rituximab prognostic value, as also observed in Ph− ALL patients [73, 74].
increased OS from 62% to 83% (reviewed in [40]). The anti- The last issue relies on which is the best TKI/chemotherapy
CD20 antibody is also being applied in CD20-positive de novo B- front-line combination. Usually, TKI therapy is initiated in front-
lineage ALL, with encouraging results [53], and randomised trials line therapy together with the first chemotherapy cycle. A con-
are ongoing. The monoclonal antibodies directed against CD22, tinuous TKI exposure should be favoured, even if a few weeks off
linked to cytotoxic agents, either to calicheamicin (inotuzumab may be needed to limit myelosuppression [75]. To date, there is
ozogamicin) or to plant or bacterial toxins (epratuzumab), are no comparative study evaluating second-generation TKIs (niloti-
explored in refractory/relapsed childhood and adult ALL [54]. nib, dasatinib) versus imatinib as first-line treatment. The use
Targeting CD19 is of great interest, as it is expressed in all B- of nilotinib and dasatinib may result in achieving a good MRD
lineage cells, most likely including early lymphoid precursor cells. response more quickly, which could be of interest before SCT. On
The bispecific antibody blinatumomab combines single chain the other hand, more potent TKIs could induce a higher inci-
antibodies to CD19 and CD3, and thereby T cells lyse the CD19- dence of rate of resistance mutation, as was observed with T315I
bearing B cells. It is effective in patients with positive MRD [55] mutations at relapse in older patients receiving front-line dasati-
or refractory/relapsed ALL [56]. The CD19 antigen is also the nib [72]. Combination of TKIs with dose-reduced chemotherapy
target for a promising new approach, the use of chimaeric antigen should probably be preferred, compared with standard intensive
receptor-modified T cells (CAR T cells) [57, 58]. chemotherapy/TKI combinations [76]. This has been randomly
In T-ALL, specific antibodies as in B-lineage ALL are not demonstrated in the GRAAPH-2005 trial, with lower early mor-
available. The few new drugs under investigation are nelarabine, tality and higher CR rate in patients receiving imatinib, combined
which is active in advanced disease [59, 60] (currently evaluated with less intensive chemotherapy compared with those receiving
in first-line therapy), and γ-secretase inhibitors blocking Notch1 Hyper-CVAD/imatinib [69]. GIMEMA (the Italian Group for
signalling. Haematological Diseases in Adults) has also reported very good
response rates and short-term outcomes in older patients treated
tyrosine kinase inhibitors in Ph+ ALL. When compared with almost exclusively with front-line dasatinib [77]. Once CR has
the pre-imatinib era [61, 62], CR rates improved from 60%–70% been reached, autologous SCT might also be a good option, at
to 80%–90% or even higher and short-term outcome was much least in patients who have reached a good MRD response, or in
better in relatively small non-randomised studies, which mostly those who cannot tolerate allogeneic SCT [69, 78, 79].
simply added imatinib to their previous standard chemotherapy In patients with persistent MRD or progressive disease, the
regimens in Ph+ ALL patients [63, 64]. These marked recommendation is to switch to another TKI while screening for
improvements were then confirmed in the long term [65–68], TKI resistance mutations and then to adapt TKI choice accord-
with survival reaching ∼50%, compared with ≤20% in the pre- ing to the resistance profile. The third-generation TKI ponatinib
imatinib era, making combined imatinib/chemotherapy the is currently the only option in patients progressing with the
standard treatment of Ph+ ALL. Early enthusiasm was such that T315I mutation.
even the place of allogeneic SCT in first CR (which was
considered as the only curative option for Ph+ ALL patients) was tyrosine kinase inhibitors in Ph-like ALL. TKIs might be also
challenged. Nevertheless, a recent prospective trial from the used as targeted treatment in some patients with Ph− ALL. The
GRAALL (Group for Research on Adult Acute Lymphoblastic Ph-like entity has recently been described as associated with a
Leukemia) suggests that allogeneic SCT is still associated with a gene expression signature similar to Ph+ ALL, but with no Ph
better relapse-free survival in younger Ph+ ALL patients [69]. chromosome or BCR-ABL1 rearrangement. Kinase-activating
These younger patients may receive standard myeloablative events, including ABL1 itself, PDGFR-beta, JAK2 or other kinases
conditioning (MAC), but the role of reduced-intensity are frequently found in this poor-prognosis ALL subset [16], and
conditioning (RIC)-SCT in older patients remains to be some remarkable cases of TKI treatment success have been
prospectively evaluated. After SCT, a recent randomised study reported in these patients [80, 81]. Imatinib, dasatinib or even
from the GMALL (German Multicenter Study Group for Adult ruxolitinib could thus be evaluated in these patients, who
Acute Lymphoblastic Leukemia) suggests that prophylactic frequently have primary refractory ALL or very early relapse. See
imatinib maintenance is probably the best option to prevent post- Table 4.
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw025 | v

by guest
on 05 February 2018
Table 4. Recommendations for TKI/chemotherapy combinations in adults with Ph+ ALL

Recommendation
Adults with Ph+ ALL should be treated front line with a combination of imatinib or second-generation TKI and chemotherapy. Mandatory
Reduced-intensity chemotherapy may be used in combination with TKI during the first treatment cycles, to minimise early toxicity Recommended
and mortality.
TKI should be administered as continuously as possible, with respect to haematological tolerance. Recommended
There is no standard post-remission treatment in patients not receiving allogeneic SCT because of no donor or advanced age. Recommended
Prolonged administration of imatinib/consolidation chemotherapy followed by imatinib maintenance should be applied.
Allogeneic SCT in first CR with a standard myeloablative conditioning probably remains the best post-remission option in younger Recommended
patients with a donor. The role of reduced-intensity conditioning allogeneic SCT remains to be evaluated in this ALL subset.
Post-SCT imatinib maintenance is recommended for 1–2 years of duration. Recommended
Prolonged monitoring of BCR-ABL1 MRD levels is recommended, associated with resistance mutation screening in patients with Recommended
persistent MRD detection or re-increasing MRD levels. Results should be used to guide the switch towards a second- or third-
generation TKI in these higher risk patients.
TKI, tyrosine kinase inhibitor; Ph+, Philadelphia-positive; ALL, acute lymphoblastic leukaemia; SCT, stem-cell transplantation; CR, complete remission;
MRD, minimal residual disease.
stem-cell transplantation SCT can be offered to patients lacking a sibling donor [IV, A].
Despite the existence of studies and meta-analyses comparing Umbilical cord blood can be an alternative source when
chemotherapy and SCT [82], the issue of the indications of SCT an HSCT is needed urgently or when the search for a very
in adult ALL patients in first CR is not defined in a satisfactory well-matched, unrelated donor is unsuccessful [88–90].
way and requires continuous update. This is due to the improv- Haploidentical SCT could be an option in patients without a
ing results with conventional and targeted chemotherapy regi- matched sibling or MUD, but prospective comparative studies
mens on one hand and to the decreasing mortality and broader are lacking. In turn, autologous SCT is considered inferior to
chemotherapy and to allogeneic SCT [91] [I, A], but could be
availability of SCT on the other. Several attempts have been
reconsidered in MRD-negative patients [92] unfit for allogenic
made to provide evidence-based guidelines for indication of
SCT [IV, D], as has been shown in patients with Ph+ ALL [71].
SCT (Table 5, ref. [83]) [84–86] [V, A]. From these guidelines,
4) The intensity of the conditioning. There is no standard MAC
the OS for SCT was superior to chemotherapy in HR patients
regimen, but total body irradiation-based regimens seem to
[I, A], but left the role of SCT in SR open [I, C]. On the
have better anti-leukaemic activity than busulfan-based pre-
other hand, there is a general agreement that SCT is clearly
parative regimens [93] [IV, B]. RIC regimens are increasingly
the best therapeutic option for patients in second or later CR
considered as an option for elderly HR patients or patients with
[III, A]. contraindications for MAC-SCT [84, 94] [IV, B], but no pro-
Some important issues emerged from the results of recent spective comparative studies between these two types of pre-
studies: parative regimens have been conducted in young, fit patients.
1) Whether or not SCT should be offered to AYA with SR factors 5) The need for SCT in specific genetically defined groups of
treated with paediatric-based or -inspired protocols that ALL, such as BCR-ABL1-positive (as previously reviewed) or
provide long-term OS rates ∼70% [87]. In view of these MLL-positive cases. Allogeneic SCT is currently carried out
results, most groups skip SCT in these patients to avoid acute for MLL-rearranged ALL in most trials and, in the largest
mortality and long-term effects [III, B]. study conducted to date, better results have been observed
2) The use of MRD (the most important prognostic factor compared with chemotherapy [95] [IV, A].
in ALL) to guide the decision of chemotherapy or SCT after
consolidation. Data from recent studies have shown that SCT treatment of relapsed or refractory ALL
offers better results than chemotherapy in patients with high Relapsed ALL in adults is still a major clinical challenge. There
MRD levels after consolidation, regardless of the convention- is no universally accepted treatment protocol and a lack of evi-
al risk factors at baseline [29] [III, A]. The question remains dence based on randomised, controlled trials. However, there is
whether SCT is justified in patients with conventional HR consensus on the general approach to managing these patients.
features but low or negative MRD after consolidation, for
whom OS rates >50% are expected with chemotherapy [29, diagnostic work-up
31–33]. Phase III studies addressing this point are desirable, Therapy-related AML should be excluded. Enumeration of
because the trials included in the most recent meta-analysis CD19, CD20 and CD22 expression on blast cells is important as
[82] did not incorporate the MRD analysis as a decision tool. it may have therapeutic relevance. Cytogenetic evaluation should
3) The indication for the different types of SCT. Regarding allogen- take into account fusion proteins that may indicate a BCR-ABL-
eic SCT, there is increasing evidence that sibling and very well- like phenotype [16, 81]. If allogeneic SCT is a possible therapeutic
matched, unrelated donors (MUD) SCT can be considered option, and if this was not done at diagnosis, the HLA profiling
equivalent options in terms of results and, therefore, MUD of the patient and siblings should be carried out urgently, and an
v | Hoelzer et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Table 5. Recommendations for SCT in adult ALL (data from [83])
Question Recommendations
CR1. AutoSCT versus non- – Not recommended outside a clinical trial

transplantation – Maintenance therapy, biological therapy, or TKIs may improve outcomes in selected patients
CR1. AlloSCT versus non- – AlloSCT recommended in all patients with poor early MRD response
transplantation – AlloSCT not recommended in SR patients with sustained molecular response
– Indication unclear in HR patients with sustained molecular response
CR ≥2. AlloSCT versus – AlloSCT superior
non-transplantation
AlloSCT versus autoSCT – Advantage for alloSCT
– Insufficient data in patients with negative MRD levels, including Ph+ ALL
Sibling donor versus MUD – Similar, and possibly equivalent survival outcomes
UD CBT versus UD BMT – Consider CBT if no HLA-well-matched donor or need for urgent SCT
– Haploidentical SCT should also be considered in this setting
Conditioning regimens – Benefit of TBI regimens for myeloablative SCT
– RIC regimens appropriate only for adults in remission unfit for myeloablative conditioning and elderly fit
patients
Areas of research needed Comments
AlloSCT versus more intensive/specific – Re-evaluate, especially in the context of biological therapies and TKIs
CHT regimens
MRD to guide therapy in ALL – Increasing importance of the monitoring of MRD during initial treatment to guide individual patient
eligibility and timing of allogeneic SCT
MRD monitoring after SCT – To detect early post-SCT relapse for pre-emptive therapy
– Effective therapies are under development
RIC versus MAC regimens – Further studies needed, adjusted for a patient’s tolerance of conditioning toxicity balanced against the
risk of relapse
CBT techniques – Single unit, double unit, ex vivo expansion, third-party donor. Larger multicentre experience needed to
evaluate the broader applicability of CB grafting for adults with ALL
Haploidentical SCT – Comparative studies with SCT from other sources needed (non-randomised comparisons show similar
results)
QoL and functional status after – Evaluation and measures for improvement needed
successful SCT
SCT, stem-cell transplantation; ALL, acute lymphoblastic leukaemia; CR1, first complete remission; autoSCT, autologous stem-cell transplantation; TKIs,
tyrosine kinase inhibitors; alloSCT, allogeneic stem-cell transplantation; MRD, minimal residual disease; SR, standard risk; HR, high risk; CR ≥2, second or
later complete remission; Ph+, Philadelphia-positive; MUD, matched unrelated donor; UD, unrelated donor; CBT, cord blood transplantation; BMT, bone
marrow transplantation; HLA, human leucocyte antigen; TBI, total body irradiation; RIC, reduced-intensity conditioning; CHT, chemotherapy; MAC,
myeloablative conditioning; CB, cord blood; QoL, quality of life.
unrelated donor search should be initiated if a sibling match is treatment principles

not available. In the case of Ph+ ALL, BCR-ABL1 tyrosine kinase Treatment with a curative aim involves achievement of CR fol-
domain mutations should be evaluated [96]. lowed by allogeneic SCT. In four large trials, the outcome was
Overall evaluation of the clinical situation should take into very similar [59, 97–99]. The rate of second CR achieved was
account the disease-specific factors (BCP-ALL or T-ALL, BCR- 44%–45%, the median OS 4.5–8.4 months (24% at 3 years in one
ABL1 status), patient factors (age, performance status, organ study). Long duration of first CR (>2 years), then re-induction
function and presence of extramedullary disease, in particular with a standard induction regimen—such as that used for original
CNS), previous therapy (with particular reference to prior allo- treatment—may be used [59, 97–99]. Short first CR or primary
graft, anthracycline dose) and specific toxicities of prior treatment refractory disease is a very high-risk situation, and consideration
which might guide therapeutic selection (e.g. osteonecrosis, vinca should immediately be given to the availability of trials of novel
alkaloid neuropathy and specific infectious complications such as agents that may be non-cross-resistant with chemotherapy. For
fungal infections). BCP-ALL, such agents are now more widely available. Both
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw025 | v

by guest
on 05 February 2018
blinatumomab [56] and inotuzumab [96] have shown promising granulocyte colony-stimulating factor and idarubicin). Despite its
results in phase II studies and are being evaluated in randomised, common use and inclusion as ‘standard of care’ arm in current
controlled trials where the comparator arm is ‘standard of care’ randomised, controlled trials of relapsed ALL, there is remarkably
chemotherapy. A clinical trial involving immunotherapy with little published on FLAG-Ida in relapsed ALL [97]. Clofarabine-
CD19 CAR T-cell therapy [58] is also a possibility. based regimens including cytarabine, cyclophosphamide or
etoposide are also commonly used based mostly on data in
chemotherapy for relapsed ALL. The most commonly used childhood ALL [98]. Liposomal vincristine [99] is licensed for the
regimens in Europe are fludarabine- and anthracycline-containing treatment of relapsed ALL. These standard chemotherapy
regimens, for example, FLAG-Ida (fludarabine, high-dose ara-C, approaches are applicable in BCP-ALL as well as in T-ALL.
Table 6. Summary of recommendations for adult ALL
Diagnostic work-up of ALL
• Morphology, immunophenotype and cytogenetics to confirm the diagnosis and ALL subsets are mandatory
• New genetics and molecular genetics are recommended to detect rare subtypes, such as Ph-like ALL, ETP ALL
• Targets for therapy with TKIs or antibodies have to be identified
• Minimal residual disease by immunophenotype or molecular probe at diagnosis, for MRD-based risk classification and treatment algorithm, mandatory
Risk assessment and prognostic factors
• It is essential to stratify patients as standard-risk or high-risk patients

• Risk stratification is currently determined by a combination of prognostic factors at diagnosis and treatment-related parameters, preferentially MRD
• MRD during therapy is now the most relevant prognostic parameter for treatment decisions
Treatment
Treatment algorithm
• Chemotherapy includes induction therapy 1–2 months, consolidation cycles (alternating) 6–8 months and maintenance therapy 2–2.5 years
• Ongoing chemotherapy protocols for AYAs use paediatric-type regimens
• Prophylactic treatment to prevent CNS relapse is mandatory
Antibody therapy
• Anti-CD20 rituximab in combination with a chemotherapy is strongly recommended for Burkitt leukaemia/lymphoma
• Anti-CD22 immunoconjugates directed against CD22 currently under investigation
• Anti-CD19; activation of patients’ own T cells directed against CD19
• Bispecific (CD3/CD19) blinatumomab under investigation
• Chimaeric antigen receptor-modified T cells directed against CD19 in early phase
Targeted therapy with TKIs in Ph+ ALL
• A TKI should be combined with chemotherapy in front-line therapy

• The TKI imatinib (400–800 mg/day) should be administered continuously, also post-SCT
• Prolonged monitoring of BCR-ABL-1 MRD is recommended, as well as resistance mutation screening. In case of persisting MRD, increasing MRD
level, or resistance mutation, switch to a second- or third-generation TKI
SCT
• AlloSCT in CR1 significantly improves OS and EFS in high-risk patients/MRD+ patients and is the best post-remission option for Ph+ ALL and
MLL-rearranged ALL
• Conditioning regimens are age-adapted with full allo versus RIC for elderly patients or patients unfit for full conditioning
• The role of autoSCT should be investigated for MRD-negative patients, in the setting of clinical trials
• All patients in CR ≥2 are candidates for alloSCT
Approach for relapsed/ refractory ALL
• Full diagnostic work-up necessary to exclude/reveal clonal aberrations, and to provide bases for targeted therapies
• Different treatment for patients with short versus long first remission duration (>18/24 months) where re-induction is considered
• Treatment; there is no standard re-induction therapy established, most often used new drugs
ALL, acute lymphoblastic leukaemia; Ph, Philadelphia; ETP, early T-cell precursor; MRD, minimal residual disease; AYAs, adolescents and young adults;
CNS, central nervous system; TKI, tyrosine kinase inhibitor; Ph+, Philadelphia-positive; SCT, stem-cell transplantation; alloSCT, allogeneic SCT; CR1, first
complete remission; OS, overall survival; EFS, event-free survival; RIC, reduced-intensity conditioning; autoSCT, autologous SCT; CR ≥2, second or later
complete remission.
v | Hoelzer et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Table 7. Levels of evidence and grades of recommendation personalised medicine
Progress in the diagnosis of ALL with identification of genomic-
defined sub-entities, the evaluation of MRD, and new targeted
Levels of evidence therapies have led to a substantial realisation of personalised
medicine in adult ALL. Current options such as less intensive
chemotherapy, new modalities of SCT, incorporation of targeted
therapies and optimal combinations of treatments require pro-
heterogeneity
spective, cooperative research, hereby further refining the indivi-
dualised approach to each patient.
heterogeneity follow-up and long-term implications
III Prospective cohort studies The follow-up of asymptomatic patients should include blood
IV Retrospective cohort studies or case–control studies cell counts and routine chemistry during maintenance therapy;
usually every 2 weeks during the first 2 years to adjust treatment
Grades of recommendation accordingly. Thereafter, follow-up should be 3-monthly in years
1, 2 and 3, since the majority of relapses occur within the first
2.5 years after initiation of treatment; then half-yearly in the 4th
and 5th year. For evaluation of MRD, which is now the most im-
portant prognostic parameter, bone marrow aspiration is
required 3-monthly. It is also desirable in Ph+ MRD to search
C Insufficient evidence for efficacy or benefit does not outweigh
for MRD (BCR-ABL) and, if possible, for mutations to switch to
optional
another TKI inhibitor.
D Moderate evidence against efficacy or for adverse outcome, In adults, adverse long-term effects are fewer compared with
generally not recommended children with ALL, and most adult ALL patients are in good
E Strong evidence against efficacy or for adverse outcome, never clinical conditions. Relevant late toxicities are: endocrinological
recommended disorders (thyroid, gonadal), osteonecrosis/osteoporosis, skin
and mucosal disorders, cataract, cardiovascular disorders, infec-
a
By permission of the Infectious Diseases Society of America [100]. tion, graft versus host disease/sicca syndrome, fatigue and cogni-
tive disorders. Second malignancies can also occur but with a
low frequency (<3%) after chemotherapy as well as SCT. Long-
term observation including quality-of-life assessment of cured
Additionally, nelarabine is licensed for this indication, and a ALL patients is an essential part of treatment optimisation
response rate of about 50% is noted [59]. Myelotoxicity is mild to studies.
moderate, but the neurotoxicity can be severe and irreversible.
Co-administration with agents used to treat CNS disease can
increase the risk. methodology
with the ESMO standard operating procedures for clinical prac-
Ph+ ALL. Patients with relapsed Ph+ ALL should be offered the
new generations of TKIs, according to the results of mutational
analysis of their BCR-ABL1 transcripts. Patients who have lost
response to imatinib may respond to nilotinib or dasatinib and
dations is presented in Table 6. Levels of evidence and grades of
there is even an option, ponatinib, for patients with the T315I
recommendation have been applied using the system is provided
mutation. Although TKIs are not without adverse events
in Table 7. Statements without grading were considered justified
(ponatinib, in particular, carries a risk of cardiovascular events),
they are nonetheless a vastly superior option compared with
repetitive treatment with myelosuppressive chemotherapy, as they
review process.
preserve performance status and are better tolerated by elderly
patients. There is no evidence of long-term survival induced by
TKIs post-relapse and the majority of patients will have to receive
allogeneic SCT. Second allografts are being reported, and there are
case reports of good outcomes, although of uncertain long-term DH has reported advisory boards for Amgen, Pfizer, Erytech
benefit. and Bristol-Myers Squibb. AF has reported advisory boards for
Even in a palliative setting BCR-ABL1, kinase domain muta- Amgen and Pfizer. CB has reported honoraria from Roche,
tional analysis should be carried out and used to guide therapy Pfizer, Celgene, Pharmacyclics and Janssen and research grants
with TKIs and to monitor treatment response and impending from Roche and Janssen. RB, JR and HD have reported no po-
relapse. tential conflicts of interest.
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw025 | v

by guest
on 05 February 2018
references 22. Brüggemann M, Schrauder A, Raff T et al. Standardized MRD quantification in

European ALL trials. In: Proceedings of the Second International Symposium on
1. Sant M, Allemani C, Tereanu C et al. Incidence of hematologic malignancies in MRD assessment in Kiel, Germany, 18–20 September 2008. Leukemia 2010;
Europe by morphologic subtype: results of the HAEMACARE project. Blood 2010; 24: 521–535.
116: 3724–3734. 23. Campana D. Minimal residual disease in acute lymphoblastic leukemia.
2. Inaba H, Greaves M, Mullighan CG. Acute lymphoblastic leukaemia. Lancet Hematology Am Soc Hematol Educ Program 2010; 2010: 7–12.
2013; 381: 1943–1955. 24. Garand R, Beldjord K, Cavé H et al. Flow cytometry and IG/TCR quantitative PCR
3. Sant M, Minicozzi P, Mounier M et al. Survival for haematological malignancies in for minimal residual disease quantitation in acute lymphoblastic leukemia: a
Europe between 1997 and 2008 by region and age: results of EUROCARE-5, a French multicenter prospective study on behalf of the FRALLE, EORTC and
population-based study. Lancet Oncol 2014; 15: 931–942. GRAALL. Leukemia 2013; 27: 370–376.
4. Jaffe ES, Harris NL, Stein H et al. Introduction and overview of the classification 25. Holowiecki J, Krawczyk-Kulis M, Giebel S et al. Status of minimal residual
of the lymphoid neoplasms. In Swerdlow SH, Campo E, Harris NL et al. (eds), disease after induction predicts outcome in both standard and high-risk Ph-
WHO Classification of Tumours of Haematopoietic and Lymphoid Tissue. Lyon: negative adult acute lymphoblastic leukaemia: the Polish Adult Leukemia Group
IARC, 2008; 158–166. ALL 4-2002 MRD Study. Br J Haematol 2008; 142: 227–237.
5. Foà R, Guarini A, Vitale A. Approach to the patient with a suspect of ALL. In 26. Patel B, Rai L, Buck G et al. Minimal residual disease is a significant predictor of
Goekbuget N, Bassan R, Dombret H et al. (eds), Recommendations of the treatment failure in non T-lineage adult acute lymphoblastic leukaemia: final results
European Working Group for Adult ALL. Bremen-London-Boston: UNI-MED of the international trial UKALL XII/ECOG2993. Br J Haematol 2010; 148: 80–89.
Verlag AG, 2011; 24–31. 27. Beldjord K, Chevret S, Asnafi V et al. Oncogenetics and minimal residual disease
6. Chiaretti S, Zini G, Bassan R. Diagnosis and subclassification of acute are independent outcome predictors in adult patients with acute lymphoblastic
lymphoblastic leukemia. Mediterr J Hematol Infect Dis 2014; 6: e2014073. leukemia. Blood 2014; 123: 3739–3749.
7. Bennett JM, Catovsky D, Daniel MT et al. The morphological classification of 28. Brüggemann M, Raff T, Flohr T et al. Clinical significance of minimal residual
acute lymphoblastic leukaemia: concordance among observers and clinical disease quantification in adult patients with standard-risk acute lymphoblastic
correlations. Br J Haematol 1981; 47: 553–561. leukemia. Blood 2006; 107: 1116–1123.
8. Béné MC, Castoldi G, Knapp W et al. Proposals for the immunological classification 29. Gökbuget N, Kneba M, Raff T et al. Adult patients with acute lymphoblastic leukemia
of acute leukemias. European Group for the Immunological Characterization of and molecular failure display a poor prognosis and are candidates for stem cell
Leukemias (EGIL). Leukemia 1995; 9: 1783–1786. transplantation and targeted therapies. Blood 2012; 120: 1868–1876.
9. Béné MC, Nebe T, Bettelheim P et al. Immunophenotyping of acute leukemia 30. Bassan R, Spinelli O, Oldani E et al. Improved risk classification for risk-specific
and lymphoproliferative disorders: a consensus proposal of the European therapy based on the molecular study of minimal residual disease (MRD) in adult
LeukemiaNet Work Package 10. Leukemia 2011; 25: 567–574. acute lymphoblastic leukemia (ALL). Blood 2009; 113: 4153–4162.
10. Matutes E, Pickl WF, van’t Veer M et al. Mixed-phenotype acute leukemia: 31. Ribera JM, Oriol A, Morgades M et al. Treatment of high-risk Philadelphia
clinical and laboratory features and outcome in 100 patients defined according to chromosome-negative acute lymphoblastic leukemia in adolescents and adults
the WHO 2008 classification. Blood 2011; 117: 3163–3171. according to early cytologic response and minimal residual disease after
11. van Dongen JM, Lhermitte L, Böttcher S et al. EuroFlow antibody panels for consolidation assessed by flow cytometry: final results of the PETHEMA ALL-AR-
standardized n-dimensional flow cytometric immunophenotyping of normal, 03 trial. J Clin Oncol 2014; 32: 1595–1604.
reactive and malignant leukocytes. Leukemia 2012; 26: 1908–1975. 32. Bassan R, Spinelli O, Oldani E et al. Different molecular levels of post-induction
12. Gleissner B, Gökbuget N, Bartram CR et al. Leading prognostic relevance of the minimal residual disease may predict hematopoietic stem cell transplantation
BCR-ABL translocation in adult acute B-lineage lymphoblastic leukemia: a outcome in adult Philadelphia-negative acute lymphoblastic leukemia. Blood
prospective study of the German Multicenter Trial Group and confirmed Cancer J 2014; 4: e225.
polymerase chain reaction analysis. Blood 2002; 99: 1536–1543. 33. Dhédin N, Huynh A, Maury S et al. Role of allogeneic stem cell transplantation in
13. Moorman AV, Harrison CJ, Buck GA et al. Karyotype is an independent prognostic adult patients with Ph-negative acute lymphoblastic leukemia. Blood 2015; 125:
factor in adult acute lymphoblastic leukemia (ALL): analysis of cytogenetic 2486–2496.
data from patients treated on the Medical Research Council (MRC) UKALLXII/ 34. Hoelzer D, Gökbuget N. Change in prognostic factors. Leukemia 2012; 1(Suppl. 1):
Eastern Cooperative Oncology Group (ECOG) 2993 trial. Blood 2007; 109: S1–S2.
3189–3197. 35. Bassan R, Hoelzer D. Modern therapy of acute lymphoblastic leukemia. J Clin
14. Pullarkat V, Slovak ML, Kopecky KJ et al. Impact of cytogenetics on the outcome Oncol 2011; 29: 532–543.
of adult acute lymphoblastic leukemia: results of Southwest Oncology Group 36. Annino L, Vegna ML, Camera A et al. Treatment of adult acute lymphoblastic
9400 study. Blood 2008; 111: 2563–2572. leukemia (ALL): long-term follow-up of the GIMEMA ALL 0288 randomized study.
15. Mancini M, Scappaticci D, Cimino G et al. A comprehensive genetic classification Blood 2002; 99: 863–871.
of adult acute lymphoblastic leukemia (ALL): analysis of the GIMEMA0496 37. Gökbuget N, Bassan R, Dombret H et al. Recommendations of the European
protocol. Blood 2005; 105: 3434–3441. Working Group for Adult ALL (EWALL), 1st edition. Bremen: UNI-MED Science
16. Roberts KG, Li Y, Payne-Turner D et al. Targetable kinase-activating lesions in Verlag, 2011.
Ph-like acute lymphoblastic leukemia. N Engl J Med 2014; 371: 1005–1015. 38. Labar B, Suciu S, Willemze R et al. Dexamethasone compared to prednisolone
17. Trinquand A, Tanguy-Schmidt A, Ben Abdelali R et al. Toward a NOTCH1/ for adults with acute lymphoblastic leukemia or lymphoblastic lymphoma: final
FBXW7/RAS/PTEN-based oncogenetic risk classification of adult T-cell acute results of the ALL-4 randomized, phase III trial of the EORTC Leukemia Group.
lymphoblastic leukemia: a Group for Research in Adult Acute Lymphoblastic Haematologica 2010; 95: 1489–1495.
Leukemia study. J Clin Oncol 2013; 31: 4333–4342. 39. Marks DI, Paietta EM, Moorman AV et al. T-cell acute lymphoblastic leukemia in
18. Coustan-Smith E, Mullighan CG, Onciu M et al. Early T-cell precursor leukaemia: adults: clinical features, immunophenotype, cytogenetics, and outcome from the
a subtype of very high-risk acute lymphoblastic leukaemia. Lancet Oncol 2009; large randomized prospective trial (UKALL XII/ECOG 2993). Blood 2009; 114:
10: 147–156. 5136–5145.
19. Brüggemann M, Raff T, Kneba M. Has MRD monitoring superseded other 40. Hoelzer D, Walewski J, Döhner H et al. Improved outcome of adult Burkitt
prognostic factors in adult ALL? Blood 2012; 120: 4470–4481. lymphoma/leukemia with rituximab and chemotherapy: report of a large
20. Hoelzer D, Thiel E, Löffler H et al. Prognostic factors in a multicenter study for prospective multicenter trial. Blood 2014; 124: 3870–3879.
treatment of acute lymphoblastic leukemia in adults. Blood 1988; 71: 123–131. 41. Lazarus HM, Richards SM, Chopra R et al. Central nervous system involvement in
21. Rowe JM. Prognostic factors in adult acute lymphoblastic leukaemia. Br J adult acute lymphoblastic leukemia at diagnosis: results from the international
Haematol 2010; 150: 389–405. ALL trial MRC UKALL XII/ECOG E2993. Blood 2006; 108: 465–472.
v | Hoelzer et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
42. Ram R, Wolach O, Vidal L et al. Adolescents and young adults with acute 63. Thomas DA, Faderl S, Cortes J et al. Treatment of Philadelphia chromosome-
lymphoblastic leukemia have a better outcome when treated with pediatric-inspired positive acute lymphocytic leukemia with hyper-CVAD and imatinib mesylate.
regimens: systematic review and meta-analysis. Am J Hematol 2012; 87: 472–478. Blood 2004; 103: 4396–4407.
43. Ribera JM, Ribera J, Genescà E. Treatment of adolescent and young adults with 64. de Labarthe A, Rousselot P, Huguet-Rigal F et al. Imatinib combined with
acute lymphoblastic leukemia. Mediterr J Hematol Infect Dis 2014; 6: e2014052. induction or consolidation chemotherapy in patients with de novo Philadelphia
44. Burke PW, Douer D. Acute lymphoblastic leukemia in adolescents and young chromosome-positive acute lymphoblastic leukemia: results of the GRAAPH-
adults. Acta Haematol 2014; 132: 264–273. 2003 study. Blood 2007; 109: 1408–1413.
45. McNeer JL, Raetz EA. Acute lymphoblastic leukemia in young adults: which 65. Ribera JM, Oriol A, González M et al. Concurrent intensive chemotherapy and
treatment? Curr Opin Oncol 2012; 24: 487–494. imatinib before and after stem cell transplantation in newly diagnosed
46. Rytting ME, Thomas DA, O’Brien SM et al. Augmented Berlin-Frankfurt-Münster Philadelphia chromosome-positive acute lymphoblastic leukemia. Final results of
therapy in adolescents and young adults (AYAs) with acute lymphoblastic the CSTIBES02 trial. Haematologica 2010; 95: 87–95.
leukemia (ALL). Cancer 2014; 120: 3660–3668. 66. Fielding AK, Rowe JM, Buck G et al. UKALLXII/ECOG2993: addition of imatinib to
47. Gökbuget N, Beck J, Brandt K et al. Significant improvement of outcome in a standard treatment regimen enhances long-term outcomes in Philadelphia
adolescents and young adults (AYAs) aged 15–35 years with acute lymphoblastic positive acute lymphoblastic leukemia. Blood 2014; 123: 843–850.
leukemia (ALL) with a pediatric derived adult ALL protocol; results of 1529 AYAs 67. Bassan R, Rossi G, Pogliani EM et al. Chemotherapy-phased imatinib pulses
in 2 consecutive trials of the German Multicenter Study Group For Adult ALL improve long-term outcome of adult patients with Philadelphia chromosome-
(GMALL). Blood (ASH Annual Meeting Abstracts) 2013; 122: abstr 839. positive acute lymphoblastic leukemia: Northern Italy Leukemia Group protocol
48. Huguet F, Leguay T, Raffoux E et al. Pediatric-inspired therapy in adults with 09/00. J Clin Oncol 2010; 28: 3644–3652.
Philadelphia chromosome-negative acute lymphoblastic leukemia: the GRAALL- 68. Pfeifer H, Goekbuget N, Völp C et al. Long-term outcome of 335 adult patients
2003 study. J Clin Oncol 2009; 27: 911–918. receiving different schedules of imatinib and chemotherapy as front-line
49. DeAngelo DJ, Stevenson KE, Dahlberg SE et al. Long-term outcome of a treatment for Philadelphia-positive acute lymphoblastic leukemia (Ph+ ALL).
pediatric-inspired regimen used for adults aged 18–50 years with newly Blood (ASH Annual Meeting Abstracts) 2010; 116: abstr 173.
diagnosed acute lymphoblastic leukemia. Leukemia 2015; 29: 526–534. 69. Chalandon Y, Thomas X, Hayette S et al. Randomized study of reduced-intensity
50. Guru Murthy GS, Venkitachalam R, Mehta P. Trends in survival outcomes of B- chemotherapy combined with imatinib in adults with Ph-positive acute
lineage acute lymphoblastic leukemia in elderly patients: analysis of Surveillance, lymphoblastic leukemia. Blood 2015; 125: 3711–3719.
Epidemiology, and End Results database. Leuk Lymphoma 2015; 56: 2296–2300. 70. Pfeifer H, Wassmann B, Bethge W et al. Randomized comparison of prophylactic
51. Gökbuget N. How I treat older patients with ALL. Blood 2013; 122: 1366–1375. and minimal residual disease-triggered imatinib after allogeneic stem cell
52. Ribrag V, Koscielny S, Bouabdallah K et al. Addition of rituximab improves outcome transplantation for BCR-ABL1-positive acute lymphoblastic leukemia. Leukemia
of HIV negative patients with Burkitt Lymphoma treated with the Lmba protocol: 2013; 27: 1254–1262.
results of the randomized intergroup (GRAALL-Lysa) LMBA02 protocol. (IGR 71. Ravandi F, Jorgensen JL, Thomas DA et al. Detection of MRD may predict the
sponsored LMBA02, NCT00180882). Blood (ASH Annual Meeting Abstracts) 2012; outcome of patients with Philadelphia chromosome-positive ALL treated with
120: abstr 685. tyrosine kinase inhibitors plus chemotherapy. Blood 2013; 122: 1214–1221.
53. Thomas DA, O’Brien S, Faderl S et al. Chemoimmunotherapy with a modified 72. Rousselot P, Coudé MM, Huguet F et al. Dasatinib (Sprycel®) and low intensity
hyper-CVAD and rituximab regimen improves outcome in de novo Philadelphia chemotherapy for first-line treatment in patients with de novo Philadelphia
chromosome-negative precursor B-lineage acute lymphoblastic leukemia. J Clin positive ALL aged 55 and over: final results of the EWALL-Ph-01 Study. Blood
Oncol 2010; 28: 3880–3889. (ASH Annual Meeting Abstracts) 2012; 120: abstr 666.
54. Kantarjian H, Thomas D, Jorgensen J et al. Inotuzumab ozogamicin, an anti- 73. Martinelli G, Iacobucci I, Storlazzi CT et al. IKZF1 (Ikaros) deletions in BCR-ABL1-
CD22-calecheamicin conjugate, for refractory and relapsed acute lymphocytic positive acute lymphoblastic leukemia are associated with short disease-free
leukaemia: a phase 2 study. Lancet Oncol 2012; 13: 403–411. survival and high rate of cumulative incidence of relapse: a GIMEMA AL WP
55. Topp MS, Kufer P, Gökbuget N et al. Targeted therapy with the T-cell-engaging report. J Clin Oncol 2009; 27: 5202–5207.
antibody blinatumomab of chemotherapy-refractory minimal residual disease in 74. van der Veer A, Zaliova M, Mottadelli F et al. IKZF1 status as a prognostic feature
B-lineage acute lymphoblastic leukemia patients results in high response rate in BCR-ABL1-positive childhood ALL. Blood 2014; 123: 1691–1698.
and prolonged leukemia-free survival. J Clin Oncol 2011; 29: 2493–2498. 75. Wassmann B, Pfeifer H, Goekbuget N et al. Alternating versus concurrent
56. Topp MS, Gökbuget N, Stein AS et al. Safety and activity of blinatumomab for adult schedules of imatinib and chemotherapy as front-line therapy for Philadelphia-
patients with relapsed or refractory B-precursor acute lymphoblastic leukaemia: a positive acute lymphoblastic leukemia (Ph+ ALL). Blood 2006; 108: 1469–1477.
multicentre, single-arm, phase 2 study. Lancet Oncol 2015; 16: 57–66. 76. Ribera JM, García O, Montesinos P et al. Treatment of young patients with
57. Brentjens RJ, Davila ML, Riviere I et al. CD19-targeted T cells rapidly induce Philadelphia chromosome-positive acute lymphoblastic leukaemia using
molecular remissions in adults with chemotherapy-refractory acute lymphoblastic increased dose of imatinib and deintensified chemotherapy before allogeneic
leukemia. Sci Transl Med 2013; 5: 177ra38. stem cell transplantation. Br J Haematol 2012; 159: 78–81.
58. Grupp SA, Kalos M, Barrett D et al. Chimeric antigen receptor-modified T cells 77. Foà R, Vitale A, Vignetti M et al. Dasatinib as first-line treatment for adult patients
for acute lymphoid leukemia. N Engl J Med 2013; 368: 1509–1518. with Philadelphia chromosome-positive acute lymphoblastic leukemia. Blood
59. Gökbuget N, Basara N, Baurmann H et al. High single-drug activity of nelarabine 2011; 118: 6521–6528.
in relapsed T-lymphoblastic leukemia/lymphoma offers curative option with 78. Giebel S, Labopin M, Gorin NC et al. Improving results of autologous stem cell
subsequent stem cell transplantation. Blood 2011; 118: 3504–3511. transplantation for Philadelphia-positive acute lymphoblastic leukaemia in the era
60. DeAngelo DJ, Yu D, Johnson JL et al. Nelarabine induces complete remissions in of tyrosine kinase inhibitors: a report from the Acute Leukaemia Working Party of
adults with relapsed or refractory T-lineage acute lymphoblastic leukemia or the European Group for Blood and Marrow Transplantation. Eur J Cancer 2014;
lymphoblastic lymphoma: Cancer and Leukemia Group B study 19801. Blood 50: 411–417.
2007; 109: 5136–5142. 79. Wetzler M, Watson D, Stock W et al. Autologous transplantation for Philadelphia
61. Dombret H, Gabert J, Boiron JM et al. Outcome of treatment in adults with chromosome-positive acute lymphoblastic leukemia achieves outcomes similar
Philadelphia chromosome-positive acute lymphoblastic leukemia—results of the to allogeneic transplantation: results of CALGB Study 10001 (Alliance).
prospective multicenter LALA-94 trial. Blood 2002; 100: 2357–2366. Haematologica 2014; 99: 111–115.
62. Fielding AK, Rowe JM, Richards SM et al. Prospective outcome data on 267 80. Ljungman P, Bregni M, Brune M et al. Allogeneic and autologous transplantation
unselected adult patients with Philadelphia chromosome-positive acute for haematological diseases, solid tumours and immune disorders: current
lymphoblastic leukemia confirms superiority of allogeneic transplantation over practice in Europe 2009. Bone Marrow Transplant 2010; 45: 219–234.
chemotherapy in the pre-imatinib era: results from the International ALL Trial 81. Dombret H, Cluzeau T, Huguet F, Boissel N. Pediatric-like therapy for adults with
MRC UKALLXII/ECOG2993. Blood 2009; 113: 4489–4496. ALL. Curr Hematol Malig Rep 2014; 9: 158–164.
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw025 | v

by guest
on 05 February 2018
82. Nishiwaki S, Miyamura K, Ohashi K et al. Impact of a donor source on adult 90. Soverini S, Gnani A, Colarossi S et al. Philadelphia-positive patients who already
Philadelphia chromosome-negative acute lymphoblastic leukemia: a retrospective harbor imatinib-resistant Bcr-Abl kinase domain mutations have a higher
analysis from the Adult Acute Lymphoblastic Leukemia Working Group of the likelihood of developing additional mutations associated with resistance to
Japan Society for Hematopoietic Cell Transplantation. Ann Oncol 2013; 24: second- or third-line tyrosine kinase inhibitors. Blood 2009; 114: 2168–2171.
1594–1602. 91. Fielding AK, Richards SM, Chopra R et al. Outcome of 609 adults after relapse of
83. Oliansky DM, Camitta B, Gaynon P et al. The role of cytotoxic therapy with acute lymphoblastic leukemia (ALL); an MRC UKALL12/ECOG 2993 study. Blood
hematopoietic stem cell transplantation in the treatment of pediatric acute 2007; 109: 944–950.
lymphoblastic leukemia: update of the 2005 evidence-based review. ASBMT 92. Weston BW, Hayden MA, Roberts KG et al. Tyrosine kinase inhibitor therapy
Position Statement. Biol Blood Marrow Transplant 2012; 18: 979–981. induces remission in a patient with refractory EBF1-PDGFRB-positive acute
84. Marks DI, Woo KA, Zhong X et al. Unrelated umbilical cord blood transplant for lymphoblastic leukemia. J Clin Oncol 2013; 31: e413–e416.
adult acute lymphoblastic leukemia in first and second complete remission: a 93. Oriol A, Vives S, Hernández-Rivas JM et al. Outcome after relapse of acute
comparison with allografts from adult unrelated donors. Haematologica 2014; lymphoblastic leukemia in adult patients included in four consecutive risk-
99: 322–328. adapted trials by the PETHEMA Study Group. Haematologica 2010; 95:
85. Atsuta Y, Suzuki R, Nagamura-Inoue T et al. Disease-specific analyses of unrelated 589–596.
cord blood transplantation compared with unrelated bone marrow transplantation in 94. Gökbuget N, Stanze D, Beck J et al. Outcome of relapsed adult lymphoblastic
adult patients with acute leukemia. Blood 2009; 113: 1631–1638. leukemia depends on response to salvage chemotherapy, prognostic factors, and
86. Goldstone AH, Richards SM, Lazarus HM et al. In adults with standard-risk acute performance of stem cell transplantation. Blood 2012; 120: 2032–2041.
lymphoblastic leukemia, the greatest benefit is achieved from a matched sibling 95. Tavernier E, Boiron JM, Huguet F et al. Outcome of treatment after first relapse in
allogeneic transplantation in first complete remission, and an autologous adults with acute lymphoblastic leukemia initially treated by the LALA-94 trial.
transplantation is less effective than conventional consolidation/maintenance Leukemia 2007; 21: 1907–1914.
chemotherapy in all patients: final results of the International ALL Trial (MRC 96. Kantarjian H, Thomas D, Jorgensen J et al. Results of inotuzumab ozogamicin, a
UKALL XII/ECOG E2993). Blood 2008; 111: 1827–1833. CD22 monoclonal antibody, in refractory and relapsed acute lymphocytic
87. Yanada M, Naoe T, Iida H et al. Myeloablative allogeneic hematopoietic stem cell leukemia. Cancer 2013; 119: 2728–2736.
transplantation for Philadelphia chromosome-positive acute lymphoblastic 97. Specchia G, Pastore D, Carluccio P et al. FLAG-IDA in the treatment of refractory/
leukemia in adults: significant roles of total body irradiation and chronic graft- relapsed adult acute lymphoblastic leukemia. Ann Hematol 2005; 84: 792–795.
versus-host disease. Bone Marrow Transplant 2005; 36: 867–872. 98. Hijiya N, Thomson B, Isakoff MS et al. Phase 2 trial of clofarabine in combination
88. Mohty M, Labopin M, Volin L et al. Reduced-intensity versus conventional with etoposide and cyclophosphamide in pediatric patients with refractory or
myeloablative conditioning allogeneic stem cell transplantation for patients with relapsed acute lymphoblastic leukemia. Blood 2011; 118: 6043–6049.
acute lymphoblastic leukemia: a retrospective study from the European Group for 99. O’Brien S, Schiller G, Lister J et al. High-dose vincristine sulfate liposome
Blood and Marrow Transplantation. Blood 2010; 116: 4439–4443. injection for advanced, relapsed, and refractory adult Philadelphia chromosome-
89. Marks DI, Moorman AV, Chilton L et al. The clinical characteristics, therapy and negative acute lymphoblastic leukemia. J Clin Oncol 2013; 31: 676–683.
outcome of 85 adults with acute lymphoblastic leukemia and t(4;11)(q21;q23)/ 100. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
MLL-AFF1 prospectively treated in the UKALLXII/ECOG2993 trial. Haematologica among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
2013; 98: 945–952. 139–144.
v | Hoelzer et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Acute myeloblastic leukaemias in adult patients: ESMO

and follow-up†
M. F. Fey1 & C. Buske2, on behalf of the ESMO Guidelines Working Group*
1
Department of Medical Oncology, Inselspital and University of Bern, Bern, Switzerland; 2Comprehensive Cancer Center Ulm, Institute of Experimental Cancer Research,
University Hospital Ulm, Ulm, Germany
incidence and epidemiology risk assessment and prognostic factors

The yearly incidence of acute myeloblastic leukaemia (AML) in Patient age, initial leukocyte counts and co-morbidity are
European adults is five to eight cases per 100 000 individuals important risk factors. AML having evolved from previously
with a steep increase in the population aged over 70 years where documented MDS generally have an adverse prognosis.
the incidence reaches 15–25/100 000 per annum. The yearly Molecular and genetic risk stratification are the key principles to
mortality figure in AML is four to six cases per 100 000. guide the therapy of AML, and prognosis is chiefly governed by
AML subtypes or entities defined through their karyotype or
diagnosis and pathology/molecular specific molecular features [I, A] [5–11].
biology
clinical practice
karyotype/cytogenetics
guidelines
The diagnosis of AML requires the examination of peripheral

The most favourable types of AML are acute promyelocytic
blood and bone marrow specimens. The work-up of these
leukaemia (APL) with the chromosomal translocation t(15;17)
specimens should include morphology, cytochemistry,
(q22;q12), and AML with the t(8;21)(q22;q22), inv(16)
immunophenotyping, cytogenetics and molecular genetics
( p13.1q22) or t(16;16)( p13.1;q22) (mostly myelomonocytic
[chiefly polymerase chain reaction (PCR) and fluorescence
leukaemia with preponderance of eosinophil granulocytes in the
in situ hybridisation (FISH) techniques]. See Table 1.
bone marrow), termed core binding factor acute myeloblastic
Whilst historically sorted by the largely descriptive French-
leukaemia (CBF-AML). Patients with normal karyotype AML
American-British (FAB) criteria [1], AML are now classified
are in an intermediate risk group, and AML with complex
according to the World Health Organisation (WHO)
karyotype abnormalities and/or chromosomal monosomies fare
classification from 2001, revised in 2008 [2–4]. The WHO
poorly [I, A]. See Table 2.
classification incorporates, in addition to morphological criteria,
cytogenetic data, molecular genetics, immunophenotype data molecular genetics
and clinical information into a diagnostic algorithm to delineate
Good-risk translocations in AML defined above are all
clinically significant disease entities. In the WHO classification
amenable to detection with molecular techniques (PCR or
the term ‘myeloid’ includes all cells belonging to the
FISH) which may be faster than classical cytogenetics, and are
granulocytic, monocyte/macrophage, erythroid, megakaryocytic
therefore recommended. In cytogenetically normal AML,
and mast cell lineage. The percentage of blast cells in the bone
somatic mutations of the genes FLT3 (a receptor tyrosine
marrow is a practical tool for categorising myeloid neoplasms
kinase), NPM1 (nucleophosmin) or CEBPα (a transcription
into AML or myelodysplastic syndromes (MDS), respectively,
factor) have been identified as important prognostic factors.
where myeloid neoplasms with more than 20% blasts in the
NPM1 and bi-allelic CEBPα mutations are favourable when
peripheral blood or bone marrow are considered AML, either de
present as single molecular aberrations. FLT3 alterations present
novo, or having evolved from a pre-existing MDS. Blasts are
as a single molecular abnormality or with a high allelic ration
defined using the criteria recently proposed by the International
predict for a high (and early) relapse rate [10–13]. Patients with
Working Group on Morphology of MDS.
abnormalities of the chromosomal region 11q23 representing
the mixed lineage leukaemia (MLL) gene fare poorly [I, A]. The
*Correspondence to: ESMO Guidelines Working Group, ESMO Head Office,
list of genes with diagnostic and prognostic value will certainly
Via L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland.
E-mail: clinicalguidelines@esmo.org increase, for example to include gene mutations in the isocitrate
dehydrogenase (IDH) genes or in the tet oncogene family
†
Approved by the ESMO Guidelines Working Group: August 2002, last update July
member 2 (TET2) genes; in addition, the genes for DNA
2013. This publication supersedes the previously published version—Ann Oncol 2010;
21 (Suppl. 5): v158–v161. (cytosine-5)-methyltransferase 3 α (DNMT3A), Runt-related

by guest
on 05 February 2018
Table 1. Diagnostic work-up in AML other pre-treatment investigations
Patients potentially suitable for allogeneic stem cell
• Bone marrow aspirate and biopsy as well as peripheral blood films transplantation (alloSCT) should be human leukocyte antigen
• Immunophenotyping of peripheral blood and bone marrow aspirates (HLA) typed at diagnosis, as should their available first- and
• Cytogenetics and molecular genetics (PCR and FISH techniques) second-degree family members. In high-risk disease
• Routine chemistry including liver and kidney parameters (e.g. poor-risk karyotype), early matched unrelated donor
• Coagulation profile (MUD) allogeneic transplantation must be considered, and
• Blood group and HLA typing of patient and family members therefore, a donor search should be carried out as early as
• Radiology to include dental survey as well as CT scan of chest and possible [I, A].
abdomen (or chest X-ray and abdominal ultrasound)
• Sperm preservation in men (according to patient preference)
• Serum pregnancy test in female patients treatment
Whenever possible, AML treatment should be offered in clinical
trials, and given only in experienced centres offering an
Table 2. AML risk factors adequate multidisciplinary infrastructure as well as a suitably
high case load [14, 15]. Treatment should be planned with
Favourable APL with t(15;17) curative intent whenever possible. Intensive chemotherapy of
CBF-AML with t(8;21) or inv 16 AML is divided into an induction phase, consolidation and
Biallelic CEBPα mutation with normal cytogenetics (rarely) maintenance. Potential candidates for alloSCT
Normal karyotype with NPM mutation and no FLT3 ITD (scheduled for the consolidation phase) must be identified early
Intermediate AML with normal cytogenetics and no adverse molecular at diagnosis or during induction chemotherapy [I, A].
features Treatment of APL differs in several important aspects from
FLT3 ITD with normal karyotype
therapy of all other AML types and is, therefore, discussed in a
Adverse Complex karyotype abnormalities (>3)
separate section.
Monosomal karyotype
intensive treatment of non-APL AML

transcription factor 1 (RUNX1) and additional sex combs-like
1 (ASXL1) may be listed. However, these ‘newer’ markers do not All patients undergoing intensive chemotherapy need a central
have an established use in routine practice at present. Gene intravenous line inserted, if necessary, under platelet
expression profiles assessed by microarray technology have been transfusion. Induction chemotherapy should only be started
reported to split AML into defined sub-/categories, but these (if possible) when all material needed for diagnostic testing has
techniques are not yet ready for widespread routine use. The been satisfactorily sampled. Patients with excessive
same applies to the techniques of next-generation sequencing of leukocytosis at presentation and with clinical signs of
AML cases. leukostasis may require emergency leukapheresis coordinated
with the start of chemotherapy. These patients are at particular
risk of a tumour lysis syndrome under induction
co-morbidity and other host factors chemotherapy and need appropriate monitoring. In these cases
Patients aged ≥60–65 years are more susceptible to treatment a single injection of rasburicase may be considered to prevent
complications ( particularly severe infections) than younger hyperuricaemia and hence renal failure, but data are
patients, which contributes to higher risk of an unfavourable insufficient to support a firm recommendation in this respect.
outcome. Pre-existing medical conditions such as diabetes, In most patients with AML, the start of treatment can safely be
coronary heart disease, or chronic pulmonary obstructive postponed for several days until all diagnostic material has
disease must also be recognised as contributing to poor risk. It is been collected and the results of analyses such as molecular
recommended to assess cardiac risk factors at diagnosis, in typing are available.
addition to clinical examination and cardiac echocardiography. Induction chemotherapy should include an anthracycline and
At diagnosis, patients should be investigated for the presence cytarabine with the particularly well-known and time-honoured
of active infection, particularly those planned for intensive ‘3 + 7’ regimen [I, A]. Data on dose escalation of daunorubicin
treatment. Careful clinical and haematological assessment is to improve AML outcome look promising, but longer follow-up
required to identify such patients where the start of is required to support a firm recommendation [II, C].
chemotherapy could or should be delayed, until active infection Haematopoietic growth factors are an optional adjunct to
has been treated. In addition to clinical examination, additional intensive induction chemotherapy; however, evidence on their
techniques recommended are computed tomography (CT) role in reducing the incidence and/or the severity of infectious
scans of the chest and the abdomen, and radiological imaging of complications during bone marrow aplasia, and evidence on
teeth and jaws to identify infectious foci such as dental root their putative benefit conferred through priming of leukaemic
granulomas and caries. In addition to haematological and cells to increase sensitivity to cytostatic agents, is not convincing
chemistry laboratory tests, a coagulation status must be obtained [II, C] [13, 15–20].
to detect leukaemia-related coagulopathy, particularly in APL; Consolidation therapy in AML is warranted once patients
such tests must be carried out before the insertion of central have reached clinical and haematological remission [I, A]. There
intravenous lines. is no consensus on a single ‘best’ post-remission treatment
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt320 | vi

by guest
on 05 February 2018
schedule. In good-risk AML patients in first remission, who therapy of refractory or relapsed AML
have a relapse risk of 35% or less, alloSCT is not justified Resistance to therapy (refractory or relapsed AML) is the major
because its toxic effect and/or its risk of transplantation-related cause of treatment failure, rather than mortality due to
mortality exceed the benefit. Also, these patients may receive infections and other treatment-related complications. Patients
salvage therapy including alloSCT in second remission. Good- failing to respond to one or two cycles of induction treatment
risk AML patients (including NPM-mutated AML with absence are considered refractory and are at very high risk of ultimate
of internal tandem duplications of FLT3 (FLT3-ITD), CBF treatment failure. Carefully selected patients with an HLA-
AML, and bi-allelic mutant CEBPα AML) as well as patients matched donor may be offered alloSCT, albeit with limited
who are unsuitable for alloSCT for other reasons should receive chances of success and at the cost of considerable morbidity
at least one cycle of intensive consolidation chemotherapy from this procedure [II, B]. For patients unsuited to this
preferably incorporating intermediate or high-dose cytarabine approach, BSC or palliative systemic treatment is often a
[I, A]. Patients with AML in intermediate- and poor-risk reasonable option with, at least, limited toxic effect. The
groups with an HLA-identical sibling may be candidates for prognosis of such patients is often dismal regardless of
alloSCT, provided their age and performance status allow for treatment attempts.
such treatment [21–32]. Newer data suggest that alloSCT may Patients presenting with relapse after a first remission may be
no longer be mandatory in intermediate risk patients, but offered intensive re-induction, for which chances of success are
these data need to be confirmed [III, C] [15]. Patients in these better after longer duration of first remission. Patients in second
risk groups without a family donor may qualify for alloSCT or subsequent remission may still qualify for alloSCT with a
with an HLA-matched unrelated donor identified through an family or unrelated HLA-matched donor, or with cord blood-
international donor registry. In fact, peripheral stem cells derived stem cells.
harvested from unrelated HLA-matched donors have become
the most frequently used source of stem cells. If a killer-
immunoglobulin-like receptor (KIR) mismatch is present, treatment of APL
haploidentical transplants may be considered. Conditioning Suspicion of or established diagnosis of APL must trigger a
regimens for alloSCT with dose-reduced chemotherapy distinctive therapy programme [36–39]. If in doubt and/or if
intensity (RIC) may be used for patients in the upper age APL is a diagnostic possibility at presentation, oral all-trans
range ( particularly those >50 years of age), but there is some retinoic acid (ATRA) should immediately be started, and only
evidence that RIC may also be used in adults at a younger age discontinued when APL has been specifically excluded in the
[II, B]. Infectious disease complications contracted during diagnostic work-up of newly diagnosed AML [I, A]. Risk
induction should be under suitable control before an alloSCT assessment of APL is chiefly based on white blood cell (WBC)
is enacted. The role of high-dose chemotherapy with count at presentation, where patients with a WBC count
autologous stem cell re-transfusion in AML is still >10 000/mm3 fare worse. APL induction chemotherapy consists
controversial. Recent data suggest that it may be a good option of ATRA as a differentiating agent and an anthracycline given
(and thus an alternative to alloSCT) in patients in an simultaneously, but the role of cytarabine in the treatment of
intermediate risk group [19]. Whilst it may prolong time to APL is controversial. The use of arsenic trioxide (ATO) in first-
relapse or remission duration, its potential to prolong overall line APL therapy is promising, but long-term results are not yet
survival is uncertain [I, C] [23, 25–27]. available. However, the results of ATRA-ATO therapy without
chemotherapy look promising, particularly in good risk APL [II,
C] [40]. The need for daily i.v. application of ATO over a
prolonged period of time, electrolyte as well as cardiac problems
non-intensive treatment of AML (including potentially fatal torsade-de-pointe ventricular
Patients with significant co-morbidity and the elderly are often arrhythmias) and case reports on secondary cancers after ATO
not eligible for intensive treatment. They should receive best must all be considered. If the results of ASO-ATRA induction
supportive care (BSC) or palliative systemic treatment, which (without chemotherapy) hold true over time, this regime might
may incorporate either low-dose cytarabine or a demethylating well become a new future standard, at least in low- to
agent such as decitabine or azacytidine [II, B]. Excessive intermediate-risk APL [II, C] [40]. Primary resistance of APL is
leukocytosis due to spilling of malignant blasts into the virtually unheard of and must infer a revision of the initial
periphery may be reduced with cytoreductive agents such as diagnosis. APL patients must be followed very closely for the
hydroxyurea or low-dose cytarabine, which, however, also lower development of leukaemia-associated coagulopathy and
normal blood cell counts such as red cells, neutrophils or haemorrhage at presentation and/or under induction. Their
platelets. Treatment of infections due to neutropenia and platelet counts should be kept at a minimum of 30–50 G/l.
transfusions to cover anaemia or thrombocytopenia are Fibrinogen should be kept in the normal range (100–150 mg/dl)
important additional measures. Erythropoietin is of throughout induction, whenever possible until signs of
questionable value in patients with anaemia due to extensive coagulopathy subside. The use of heparin is controversial and is
infiltration of the marrow with leukaemia. In severely not recommended as a routine procedure.
neutropenic patients haematopoietic growth factors may be The optimal strategy of consolidation in APL is less clear, but
tried when neutropenic fever or infections are a problem; there is consensus on giving two to three anthracycline-
however, there is no evidence to support their continuous use containing chemotherapy cycles [II, B]. A high proportion of
[II, D] [13, 17, 23, 32–35]. patients reach molecular remission after this treatment series,
vi | Fey and Buske Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
where the t(15;17) is no longer PCR-detectable. The small note
fraction of patients with persistent molecular disease (i.e. with
the molecular equivalent of the t(15;17) still detectable with A summary of recommendations is provided in Table 3. Levels
sensitive quantitative PCR assays) may need maintenance of evidence and grades of recommendation have been applied
therapy including non-marrow ablative long-term using the system shown in Table 4. Statements without grading
chemotherapy and ATRA. There is no role for alloSCT in were considered justified standard clinical practice by the
patients with APL in first remission. experts and the ESMO faculty.
In relapsed APL, ATO can induce remissions, even in patients
having turned refractory to ATRA [II, B]. It should be given
until remission is documented. Patients at particularly high risk
for later additional APL relapses may be candidates for alloSCT
or for high-dose chemotherapy consolidation with re-
transfusion of autologous stem cells. • Diagnostic work-up of AML must in include morphology of peripheral
blood and bone marrow, cytogenetics and molecular genetics assessed
before start of therapy
• HLA-typing of patient and family members to plan for allogeneic stem
personalised medicine cell transplantation where indicated
The impact of molecular AML typing has been demonstrated • AML should only be treated in specialised and experienced centres
extensively for diagnostic purposes (to define AML entities or offering a multidisciplinary approach, and the possibility of clinical trials
subtypes) and for AML risk assessment (see above section) [5, • AML therapy with curative intent includes induction chemotherapy
6, 9, 12, 13]. However, AML molecular genotyping still has (incorporating an anthracycline and cytarabine), consolidation which in
relatively little practical use in predicting specific drug intermediate to high-risk patients may incorporate allogeneic stem cell
treatment. APL is the only type of AML for which a targeted transplantation
agent (ATRA) has become mandatory in routine practice [I, A]. • APL needs a specific therapy approach. ATRA must be started whenever
in a case of leukaemia the differential diagnosis of APL is considered, and
For all other AML entities, targeted agents are either not
combined with anthracycline-based chemotherapy once the diagnosis of
available, e.g. for core-binding factor leukaemias, or are still
APL is confirmed
experimental, e.g. the treatment of FLT-3-positive AML with
the tyrosine kinase inhibitors sorafenib or midostaurin [41]. A
monoclonal antibody targeting the CD33 antigen often present
on AML blasts has shown discordant results in clinical trials;
hence, no firm recommendation for its routine application can Table 4. Levels of evidence and grades of recommendation (adapted from
be given [II, D] [42]. the Infectious Diseases Society of America-United States Public Health
response evaluation and follow-up Levels of evidence

I Evidence from at least one large randomised, controlled trial of good
Response of AML to treatment is monitored clinically, with methodological quality (low potential for bias) or meta-analyses
serial peripheral blood counts and repeat bone marrow of well-conducted randomised trials without heterogeneity
examinations. During intensive chemotherapy, bone marrow II Small randomised trials or large randomised trials with a suspicion
should be examined in the aplastic phase to monitor blast of bias (lower methodological quality) or meta-analyses of such
clearance, persistence or early relapse. The usually accepted trials or of trials with demonstrated heterogeneity
criteria of response in AML are blast clearance in the bone III Prospective cohort studies
marrow to <5% of all nucleated cells, morphologically normal IV Retrospective cohort studies or case–control studies
haematopoiesis and return of peripheral blood cell counts to V Studies without control group, case reports, experts opinions
normal levels. Clearance of infections contracted during Grades of recommendation
therapy-induced aplasia should also be documented. A Strong evidence for efficacy with a substantial clinical benefit,
Patients having concluded treatment should be followed strongly recommended
clinically and with repeated haematological examinations. Serial B Strong or moderate evidence for efficacy but with a limited clinical
bone marrow examinations of patients in remission are of benefit, generally recommended
uncertain value, and cannot therefore be generally C Insufficient evidence for efficacy or benefit does not outweigh the
recommended. Although sensitive PCR methods as well as risk or the disadvantages (adverse events, costs, …), optional
immunophenotyping are available, permitting molecular D Moderate evidence against efficacy or for adverse outcome, generally
follow-up and detection of minimal residual disease in patients not recommended
E Strong evidence against efficacy or for adverse outcome, never
with suitable markers (mostly specific chromosomal
recommended
translocations, or typical antigen expression profiles,
respectively), the early detection of molecular relapse in the a
Dykewicz CA. Summary of the guidelines for preventing opportunistic
absence of morphological evidence for recurrent AML is of infections among hematopoietic stem cell transplant recipients. Clin Infect
uncertain therapeutic consequence. Specifically, evidence that Dis 2001; 33: 139–144. By permission of the Infectious Diseases Society of
early re-induction treatment of such patients still in America.
haematological remission would be of any benefit is lacking.
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt320 | vi

by guest
on 05 February 2018
acknowledgements 18. Löwenberg B, van Putten W, Theobald M et al. Effect of priming with granulocyte
colony-stimulating factor on the outcome of chemotherapy for acute myeloid
We are indebted to Alan K. Burnett for helpful critical leukemia. N Engl J Med 2003; 349: 743–752.
comments on the text. 19. Pfirrmann M, Ehninger G, Thiede C et al. Prediction of post-remission survival in
acute myeloid leukaemia: a post-hoc analysis of the AML96 trial. Lancet Oncol
2012; 13: 207–214.
conflict of interest 20. Pabst T, Vellenga E, van Putten W et al. Favorable effect of priming with granulocyte
colony-stimulating factor in remission induction of acute myeloid leukemia restricted
Prof. Buske has reported consultancy/honoraria from Celgene, to dose escalation of cytarabine. Blood 2012; 119: 5367–5373.
Pfizer and Roche. Prof. Fey has reported no potential conflicts of 21. Burnett AK, Hills RK, Milligan DW et al. Attempts to optimise induction and
interest. consolidation treatment in acute myeloid leukaemia: results of the MRC AML 12
trial. J Clin Oncol 2010; 28: 586–595.
22. Koreth J, Schlenk R, Kopecky KJ et al. Allogeneic stem cell transplantation for
acute myeloid leukaemia in first complete remission: systematic review, and meta-
references analysis of prospective clinical trials. JAMA 2009; 301: 2349–2361.
23. Stone RM. Acute myeloid leukaemia in first remission: to choose transplantation or
1. Bennett JM, Catovsky D, Daniel MT et al. Proposals for the classification of the not? J Clin Oncol 2013; 31: 1262–1266.
acute leukaemias: French-American-British Cooperative Group. Br J Haematol 24. Cornelissen JJ, van Putten WL, Verdonck LF et al. Results of a HOVON/SAKK
1976; 33: 451–458. donor versus no-donor analysis of myeloablative HLA-identical sibling stem cell
2. Jaffe ES, Harris NL, Stein H, Vardiman JW. (eds): World Health Organisation (WHO) transplantation in first remission acute myeloid leukemia in young and middle aged
Classification of Tumours: Pathology and Genetics of Haematopoietic and adults: benefits for whom? Blood 2007; 109: 3658–3666.
Lymphatic Tissues, 3rd edition. Lyon, France: IARC Press, 2001. 25. Burnett AK, Goldstone AH, Stevens RM et al. Randomised comparison of addition
3. Vardiman J, Harris NL, Brunning RD. The World Health Organisation (WHO) of autologous bone-marrow transplantation to intensive chemotherapy for acute
classification of the myeloid neoplasms. Blood 2002; 100: 2292–2302. myeloid leukaemia in first remission: results of MRC AML 10 trial. Lancet 1998;
4. Swerdlow SH, Campo E, Harris NL et al. (eds): World Health Organisation (WHO) 351: 700–708.
classification of tumours: pathology and genetics of haematopoietic and lymphatic 26. Cassileth PA, Harrington DP, Appelbaum FR et al. Chemotherapy compared with
tissues, 4th edition. Lyon, France: IARC Press 2008. autologous or allogeneic bone marrow transplantation in the management of acute
5. Schlenk RF, Döhner H, Krauter J et al. Mutations and treatment outcome in myeloid leukemia in first remission. N Engl J Med 1998; 339: 1649–1656.
cytogenetically normal acute myeloid leukemia. N Engl J Med 2008; 358: 27. Zittoun R, Mandelli F, Willemze R et al. Autologous or allogeneic bone marrow
1909–1918. transplantation compared with intensive chemotherapy in acute myelogenous
6. Wouters BJ, Löwenberg B, Erpelinck-Verschueren CA et al. Double CEBPA leukaemia. N Engl J Med 1995; 332: 217–223.
mutations, but not single CEBPA mutations, define a subgroup of acute myeloid 28. Mayer RJ, Davis RB, Schiffer CA et al. Intensive post-remission chemotherapy in
leukaemia with a distinctive gene expression profile that is uniquely associated adults with acute myeloid leukaemia. N Engl J Med 1994; 331: 896–903.
with a favorable outcome. Blood 2009; 113: 3088–3091. 29. Basara N, Schulze A, Wedding U et al. Early related or unrelated haematopoietic
7. Falini B, Meducci C, Tiacci E et al. Cytoplasmic nucleophosmin in acute cell transplantation results in higher overall survival and leukaemia-free survival
myelogneous leukaemia with a normal karyotype. N Engl J Med 2005; 352: compared with conventional chemotherapy in high-risk acute myeloid leukaemia
254–266. patients in first complete remission. Leukemia 2009; 23: 635–640.
8. Green CL, Koo KK, Hills RK et al. Prognostic significance of CEBPA mutations 30. Bradstock KF, Matthews JP, Lowenthal RM et al. A randomized trial of high-versus
in a large cohort of younger adult patients with acute myeloid leukemia: impact of conventional-dose cytarabine in consolidation chemotherapy for adult de novo
double CEBPA mutations and the interaction with FLT3 and NPM1 mutations. acute myeloid leukemia in first remission after induction therapy containing high-
J Clin Oncol 2010; 28: 2739–2747. dose cytarabine. Blood 2005; 105: 481–488.
9. Marcucci G, Haferlach T, Döhner H. Molecular genetics of adult acute myeloid 31. Breems DA, Löwenberg B. Acute myeloid leukemia and the position of autologous
leukemia: prognostic and therapeutic implications. J Clin Oncol 2011; 29: stem cell transplantation. Semin Hematol 2007; 44: 259–266.
475–486. 32. Thomas X, Suciu S, Rio B et al. Autologous stem cell transplantation after
10. Kühnl A, Grimwade D. Molecular markers in acute myeloid leukaemia. Int J complete remission and first consolidation in acute myeloid leukemia patients aged
Hematol 2012; 96: 153–163. 61–70 years: results of the prospective EORTC-GIMEMA AML-13 study.
11. Mrozek K, Marcucci G, Paschka P et al. Clinical relevance of mutations and gene- Haematologica 2007; 92: 389–396.
expression changes in adult acute myeloid leukemia with normal cytogenetics: are 33. Kantarjian HM, Thomas XG, Dmoszynska A et al. Multicenter randomized open-label
we ready for a prognostically prioritized molecular classification? Blood 2007; 109: phase III trial of decitabine versus patient choice, with physician advice, of either
431–448. supportive care or low-dose cytarabine for the treatment of older patients with newly
12. Pulsoni A, Iacobelli S, Bernardi M et al. M4 acute myeloid leukemia: the role of diagnosed acute myeloid leukaemia. J Clin Oncol 2012; 30: 2670–2677.
eosinophilia and cytogenetics in treatment response and survival. The GIMEMA 34. Wheatley K, Goldstone AH, Littlewood T et al. Randomised placebo-controlled trial
experience. Haematologica 2008; 93: 1025–1032. of granulocyte colony stimulating factor (G-CSF) as supportive care after induction
13. Estey EH. Acute myeloid leukaemia: 2013 update on risk-stratification and chemotherapy in adult patients with acute myeloid leukaemia: a study of the
management. Am J Hematol 2013; 88: 318–327. United Kingdom Medical Research Council Adult Leukaemia Working Party. Br J
14. Mengis C, Aebi S, Tobler A et al. Assessment of differences in patient populations Haematol 2009; 146: 54–63.
selected for or excluded from participation in clinical phase III acute myelogenous 35. Amadori S, Suciu S, Jehn U et al. Use of glycosylated recombinant human G-CSF
leukemia trials. J Clin Oncol 2003; 21: 3933–3939. (lenograstim) during and/or after induction chemotherapy in patients 61 years of
15. Burnett AK, Goldstone A, Hills RK et al. Curability of patients with acute myeloid age and older with acute myeloid leukemia: final results of AML-13, a randomized
leukaemia who did not undergo transplantation in first remission. J Clin Oncol phase-3 study. Blood 2005; 106: 27–34.
2013; 31: 1293–1301. 36. Sanz MA, Grimwade D, Tallman MS et al. Management of acute promyelocytic
16. Roboz GJ. Current treatment of acute myeloid leukemia. Curr Opin Oncol 2012; leukemia: recommendations from an expert panel on behalf of the European
24: 711–719. LeukemiaNet. Blood 2009; 113: 1875–1891.
17. Döhner H, Estey EH, Amadori S et al. Diagnosis and management of acute myeloid 37. Fenaux P, Chastang C, Chevret S et al. A randomized comparison of all
leukemia in adults: recommendations from an international expert panel, on behalf transretinoic acid (ATRA) followed by chemotherapy and ATRA plus chemotherapy,
of the European LeukemiaNet. Blood 2010; 115: 453–474.
vi | Fey and Buske Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
and the role of maintenance therapy in newly diagnosed acute promyelocytic 40. Lo-Coco F, Avvisati G, Vignetti M et al. Retinoic acid and arsenic trioxide for acute
leukemia. Blood 1999; 94: 1192–1200. promyelocytic leukemia. N Engl J Med 2013; 369: 111–121.
38. Sanz MA, Lo-Coco F. Modern approaches to treating acute promyelocytic 41. Ravandi F, Alattar ML, Grunwald MR et al. Phase II study of azacytidine plus
leukemia. J Clin Oncol 2011; 29: 495–503. sorafenib in patients with acute myeloid leukemia and FLT-3 internal tandem
39. de Botton S, Fawaz A, Chevret S et al. Autologous and allogeneic stem-cell duplication mutation. Blood 2013; 121: 4655–4662.
transplantation as salvage treatment of acute promyelocytic leukaemia initially 42. Burnett AK, Hills RK, Milligan D et al. Identification of patients with acute
treated with all-trans-retinoic acid: a retrospective analysis of the European Acute myeloblastic leukaemia who benefit from the addition of gemtuzumab ozogamicin:
Promyelocytic Leukaemia Groups. J Clin Oncol 2005; 23: 120–126. results of the MRC AML 15 trial. J Clin Oncol 2011; 29: 369–377.
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt320 | vi

by guest
on 05 February 2018
Multiple myeloma: ESMO Clinical Practice Guidelines

P. Moreau1, J. San Miguel2, P. Sonneveld3, M. V. Mateos4, E. Zamagni5, H. Avet-Loiseau6, R. Hajek7,

M. A. Dimopoulos8, H. Ludwig9, H. Einsele10, S. Zweegman11, T. Facon12, M. Cavo5, E. Terpos8,
H. Goldschmidt13, M. Attal6 & C. Buske14, on behalf of the ESMO Guidelines Committee*
1
Haematology Department, University Hospital Hôtel-Dieu, Nantes, France; 2Clinica Universidad de Navarra, CIMA, IDISNA, Pamplona, Spain; 3Erasmus Medical
Center Institute, Rotterdam, The Netherlands; 4University Hospital of Salamanca, IBSAL, Salamanca, Spain; 5Seragnoli Institute of Hematology, School of Medicine,
Bologna University, Bologna, Italy; 6The Cancer Research Center of Toulouse, CRCT, INSERM U 1037, Toulouse, France; 7Faculty of Medicine, University Hospital
Ostrava, Ostrava, Czech Republic; 8Department of Clinical Therapeutics, School of Medicine, University of Athens, Athens, Greece; 9Wilhelminen Cancer Research
Institute, Wilhelminenspital, Vienna, Austria; 10Department of Internal Medicine II, University Hospital Wuerzburg, Wuerzburg, Germany; 11Department of
Hematology, VU University Medical Center, Amsterdam, The Netherlands; 12University Hospital Huriez, Lille, France; 13Department of Medicine, University of
Heidelberg, Heidelberg; 14Comprehensive Cancer Center Ulm and Department of Internal Medicine III, Institute of Experimental Cancer Research, University
Hospital, Ulm, Germany
†
Approved by the ESMO Guidelines Committee: August 2003, last update January 2017. This publication supersedes the previously published version—Ann Oncol
2013; 24 (Suppl 6): vi133–vi137.
*Correspondence to: ESMO Guidelines Committee, ESMO Head Office, Via, L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland. E-mail: clinicalguidelines@esmo.org
Incidence and epidemiology • Evaluation of bone marrow (BM) plasma cell infiltration: BM
aspiration and/or biopsies are the standard options to
Multiple myeloma (MM) accounts for 1% of all cancers evaluate the number and characteristics of plasma cells in
and 10% of all haematological malignancies. The incidence BM. Moreover, the BM sample should be used for cytogen-
in Europe is 4.5–6.0/100 000/year with a median age at diagnosis etic/flurorescent in situ hybridisation (FISH) studies on im-
of 72 years; the mortality is 4.1/100 000/year [1]. Almost all pa- munologically recognised or sorted plasma cells and also
tients with MM evolve from an asymptomatic pre-malignant has the potential for immunophenotypic and molecular
stage termed monoclonal gammopathy of undetermined signifi- investigations.
cance (MGUS). MGUS progresses to MM at a rate of 1% per • Evaluation of lytic bone lesions: whole-body low-dose com-
year. In some patients, an intermediate asymptomatic but more puted tomography (WBLD-CT) is the new standard for the
advanced pre-malignant stage termed smouldering (or indolent) diagnosis of lytic disease. Conventional radiography can also
MM (SMM) can be recognised. SMM progresses to myeloma at be used if WBLD-CT is not available. Magnetic resonance
a rate of 10% per year over the first 5 years following diagnosis, imaging (MRI) provides greater details and is recommended
3% per year over the following 5 years, and 1.5% per year there- whenever spinal cord compression is suspected. Either
after [2]. whole-body MRI or MRI of the spine and the pelvis may be
used, according to their availability, to assess the BM plasma
cell infiltration, in particular the presence of bone focal le-
sions. 18F-fluorodeoxyglucose positron emission tomography
with CT (PET-CT) can be done to evaluate bone lesions, ac-
Diagnosis of MM should be based on the following tests [3, 4]: cording to availability and resources.
• Detection and evaluation of the monoclonal (M) component
• Complete blood cell count, with differential serum creatinine,
by serum and/or urine protein electrophoresis (concentrate creatinine clearance and calcium level.
of 24h urine collection); nephelometric quantification of IgG, These tests can allow for the differential diagnosis between
IgA and IgM immunoglobulins; characterisation of the heavy MM, SMM and MGUS.
and light chains by immunofixation; and serum-free light- The criteria for diagnosis of MM were updated in 2014 by the
chain (FLC) measurement. International Myeloma Working Group (IMWG) [2]. The
V

by guest
on 05 February 2018
diagnosis requires 10% clonal BM plasma cells or biopsy- It has recently been demonstrated that combining FISH and
proven bony or extra-medullary plasmacytoma and any of the LDH, along with the ISS stage, could significantly improve the
following myeloma-defining events (Table 1): prognostic assessment in terms of progression-free survival (PFS)
and overall survival (OS), according to this new and revised ISS
• Evidence of end-organ damage (the so-called CRAB criteria:
(R-ISS) (Table 3) [7]. Median PFS was 66 months for patients
hypercalcaemia, renal insufficiency, anaemia or bone lesions)
with R-ISS stage I, 42 months for patients with R-ISS stage II and
that is felt to be related to the underlying plasma cell disorder.
29 months for patients with R-ISS stage III. The 5-year OS
Of note, renal insufficiency can be defined not only by creatinine
was 82% for R-ISS stage I, 62% for R-ISS stage II and 40% for
> 2 mg/dL but also by creatinine clearance < 40 mL/min [meas- R-ISS stage III. Median OS time was not reached for patients
ured by validated equations such as the Modification of Diet in with R-ISS stage I and was of 83 and 43 months for R-ISS stage II
Renal Disease (MDRD) or Chronic Kidney Disease Epidemiology and R-ISS stage III patients, respectively [7].
Collaboration (CKD-EPI)] [4]. Moreover, lytic lesions can also be Gene-expression profiling may segregate patients with stand-
defined by CT and not only by conventional X-ray. ard or high-risk disease, but this test is not yet established in rou-
• Any biomarkers of malignancy: tine practice.
- 60% clonal BM plasma cells Elderly patients with myeloma are heterogeneous and assess-
- Involved/uninvolved serum FLC ratio 100 ment strategies should be considered before starting therapy to
- > 1 focal lesion on MRI studies (each focal lesion must be define the frailty profile of the patient. The IMWG has proposed
5 mm in size).
Table 2. International staging system for MM

Stage Criteria
The course of MM is highly variable, and the clinical behaviour is
remarkably heterogeneous. Many studies have identified prog- I Serum b2M < 3.5 mg/mL and serum albumin 3.5 g/dL
nostic factors capable of predicting this heterogeneity in survival: II Not stage I or IIIa
serum b2-microglobulin, albumin, C-reactive protein and lactate III Serum b2M 5.5 mg//mL
dehydrogenase (LDH). a
There are two possibilities for stage II:
The International Staging System (ISS), a powerful and repro-
Serum b2M < 3.5 mg/L but serum albumin < 3.5 g/dL;
ducible three-stage classification (Table 2), relies on the combin- or
ation of serum levels of b2-microglobulin and albumin. ISS stage Serum b2M 3.5–5.5 mg/L irrespective of the serum albumin.
III is associated with the poorest outcome [5]. b2M, b2 microglobulin; MM, multiple myeloma.
Cytogenetics, evaluated by FISH, is a major prognostic Reprinted from [5] with permission. V C 2005 American Society of Clinical
factor. Three recurrent genetic abnormalities, t(4;14), deletion(17p) Oncology. All rights reserved.
and t(14;16), are mostly associated with a poorer outcome.
Chromosome 1 abnormalities are also adverse prognostic factors [6].
Table 1. Diagnostic criteria for plasma cell disorders

Plasma cell disorder Definition
Smouldering multiple Both criteria must be met:

myeloma • Serum M protein (IgG or IgA) 30 g/L or urinary M protein 500 mg per 24 h and/or clonal BM plasma cells 10%–60%
• Absence of myeloma-defining events or amyloidosis
Multiple myeloma Clonal BM plasma cells 10% or biopsy-proven bony or extramedullary plasmacytoma and any one or more of the following
myeloma-defining events:
• Evidence of end organ damage that can be attributed to the underlying plasma cell proliferative disorder, specifically:
- Hypercalcaemia: serum calcium > 0.25 mmol/L (> 1 mg/dL) higher than the upper limit of normal or > 2.75 mmol/L (> 11 mg/dL)
- Renal insufficiency: CrCl < 40 mL/min or serum creatinine > 177 lmol/L (> 2 mg/dL)
- Anaemia: haemoglobin value of > 20 g/L below the lower limit of normal or a haemoglobin value < 100 g/L
- Bone lesions: one or more osteolytic lesions on skeletal radiography, CT or PET-CT
• Any one or more of the following biomarkers of malignancy:
- 60% clonal BM plasma cells
- Involved/uninvolved serum-free light chain ratio 100
- > 1 focal lesion on MRI studies (each focal lesion must be 5 mm in size)
BM, bone marrow; CrCl, creatinine clearance; CT, computed tomography; M protein, monoclonal protein; MRI, magnetic resonance imaging; PET-CT, posi-
tron emission tomography-computed tomography.
Adapted from [2] with permission from Elsevier; permission conveyed through Copyright Clearance Center, Inc.

by guest
on 05 February 2018
Table 3. Standard risk factors for MM and the revised ISS Multiple myeloma
Prognostic Criteria Treatment should be initiated in all patients with MM according
factor to the updated definition proposed by the IMWG in 2014 [2].
Major treatment regimens in MM are shown in Table 6. Front-
ISS stage line treatment regimens are shown in Figure 1.
I Serum b2M < 3.5 mg/L, serum albumin 3.5 g/dL
II Not ISS stage I or III Elderly patients (non-transplant setting). The two following op-
III Serum b2M 5.5 mg/L
tions are recommended based on data from randomised phase III
CA by iFISH trials [I, A]: bortezomib (administered subcutaneously)/melpha-
High risk Presence of del(17p) and/or translocation t(4;14) and/or lan/prednisone (VMP) [11] or lenalidomide plus low-dose dexa-
translocation t(14;16) methasone (Rd) [12]; both VMP and Rd are approved in this
Standard risk No high-risk CA setting by the European Medicines Agency (EMA). Rd is
LDH approved until progression of the disease. Melphalan/prednis-
Normal Serum LDH < the upper limit of normal one/thalidomide (MPT) [13] is also approved by the EMA, but is
High Serum LDH > the upper limit of normal inferior to Rd in terms of PFS and OS [12]. Bortezomib-
cyclophosphamide and dexamethasone (VCD) is not EMA-
A new model for risk stratification for MM
approved (no controlled data), but is widely used and induces
R-ISS stage
I ISS stage I and standard-risk CA by iFISH and normal LDH
high response rates and prolonged PFS [III, A] [14]. Rd has re-
II Not R-ISS stage I or III cently been compared prospectively with Rd plus bortezomib
III ISS stage III and either high-risk CA by iFISH or high LDH (VRd), and the addition of bortezomib resulted in significantly
improved PFS and OS and had an acceptable risk–benefit pro-
b2M, b2 microglobulin; CA, chromosomal abnormalities; iFISH, inter- file [15]. Nevertheless, this triplet combination is not yet
phase fluorescent in situ hybridisation; ISS, International Staging approved by the EMA. Bendamustine plus prednisone is also
System; LDH, lactate dehydrogenase; MM, multiple myeloma; R-ISS, approved by the EMA in patients who have clinical neuropathy
revised International Staging System. at time of diagnosis, precluding the use of thalidomide accord-
Reprinted from [7] with permission. V
C 2015 American Society of Clinical
ing to the MPT regimen or bortezomib according to the VMP
Oncology. All rights reserved. regimen [II, C] [16].
Melphalan/prednisone/lenalidomide (MPR) has been eval-
uated in two prospective randomised studies versus melphalan
a frailty score (an additive scoring system based on age, comor- and prednisone (MP) [17] and versus MPT [18], but MPR was
bidities and cognitive and physical conditions) that predicts mor- not superior to the other combinations with a fixed number of
tality and the risk of toxicity in this group of patients [8]. cycles [II, C]. This triplet combination is approved by the EMA
but is not routinely used and cannot be considered as a standard
of care.
Response evaluation Cyclophosphamide/thalidomide/dexamethasone (CTD) has
also been compared with MP and is superior in terms of response
The definition of response established by the IMWG in 2006 has rates, but does not induce a clear survival advantage over MP [II,
been updated twice, in 2011 [9] and in 2016 [10] (Tables 4 and 5). C] [19].
The quality and the depth of response have improved over the last
5 years in the context of novel agent-based therapies, allowing for Younger patients (< 65 years or fit patients < 70 years in good
the introduction of new response grades, namely minimal residual clinical condition). For patients in good clinical condition (e.g. fit
disease (MRD) criteria including sequencing MRD negativity, flow patients), induction followed by high-dose therapy (HDT) with
MRD negativity, imaging plus negativity and sustained MRD nega- autologous stem cell transplantation (ASCT) is the standard
tivity. Nevertheless, MRD evaluation is not yet a reimbursed pro- treatment [II, B] [14]. Two recent phase III trials comparing
cedure, does not lead to treatment decisions, and is currently being front-line ASCT versus ASCT at the time of first relapse showed
evaluated in the context of clinical trials. that PFS was improved in the front-line ASCT arm (in the con-
There is a statistical relationship between the achievement of text of triplet novel agent-based induction) [20, 21]. Response
complete response (CR), MRD negativity and PFS or OS. rates to induction therapy have been significantly increased by
the use of novel agent-based combinations. Bortezomib-
dexamethasone, which is superior to the classical VAD regimen
(vincristine, doxorubicin and high-dose dexamethasone) [II, B],
Front-line treatment has become the backbone of induction therapy before ASCT
[14]. The addition of a third agent to bortezomib-
Smouldering myeloma dexamethasone, e.g. thalidomide (VTD), doxorubicin (PAD),
lenalidomide (RVD) or cyclophosphamide (VCD), has shown
Immediate treatment is not recommended at the present time for higher response rates in phase II trials [14]. Three prospective
patients with indolent myeloma. Clinical trials for high-risk studies have already shown that VTD is superior to thalidomide-
smouldering myeloma are strongly encouraged. dexamethasone (TD) or bortezomib-dexamethasone [I, A] [14].
iv54 | Moreau et al. Volume 28 | Supplement 4 | August 2017

by guest
on 05 February 2018
Table 4. 2011 response criteria
Response subcategory Response criteria
Molecular CR CR plus negative ASO-PCR, sensitivity 10-5

Immunophenotypic CR Stringent CR plus
Absence of phenotypically aberrant PCs (clonal) in BM with a minimum of 1 million total BM cells analysed by multiparametric
flow cytometry (with > 4 colours)
Stringent CR CR as defined below plus
Normal FLC ratio and
Absence of clonal PCs by immunohistochemistry or 2- to 4-colour flow cytometry
CR Negative immunofixation on the serum and urine and

Disappearance of any soft tissue plasmacytomas and
5% PCs in BM
VGPR Serum and urine M protein detectable by immunofixation but not on electrophoresis or 90% reduction in serum M protein
plus urine M protein level <100 mg per 24 h
PR 50% reduction of serum M protein and reduction in 24h urinary M protein by 90% or to < 200 mg per 24 h
If the serum and urine M protein are unmeasurable, a 50% decrease in the difference between involved and uninvolved FLC
levels is required in place of the M protein criteria
If serum and urine M protein are unmeasurable, and serum-free light assay is also unmeasurable, 50% reduction in PCs is
required in place of M protein, provided baseline BM PC percentage was 30%
In addition to the above listed criteria, if present at baseline, a 50% reduction in the size of soft tissue plasmacytomas is also
required
Progressive disease Increase of 25% from lowest confirmed response value in one of the following criteria:
Serum M protein (absolute increase must be 0.5 g/dL)
Serum M protein increase 1 g/dL, if the lowest M component was 5 g/dL
Urine M protein (absolute increase must be 200 mg/24 h)
ASO-PCR, allele-specific polymerase chain reaction; BM, bone marrow; CR, complete response; FLC, free light chain; M protein, monoclonal protein; PCs,
plasma cells; PR, partial response; VGPR, very good partial response.
Adapted from [9] with permission of the American Society of Hematology; permission conveyed through Copyright Clearance Center, Inc.
Table 5. 2016 response criteria

Response subcategory Response criteria
IMWG MRD Sustained MRD-negative MRD-negative in the marrow (next-generation flow and/or NGS) and by imaging as defined below,
negativity confirmed one year apart. Subsequent evaluations can be used to further specify the duration of
criteria negativity (e.g. MRD-negative at 5 years)
Flow MRD-negative Absence of phenotypically aberrant clonal plasma cells by next-generation flow cytometry on BM aspir-
ates using the EuroFlow standard operation procedure for MRD detection in MM (or validated
equivalent method) with a minimum sensitivity of 1 in 105 nucleated cells or higher
Sequencing MRD-negative Absence of clonal plasma cells by NGS on BM aspirates in which presence of a clone is defined as less
than two identical sequencing reads obtained after DNA sequencing of BM aspirates using the
LymphosightV platform (or validated equivalent method) with a minimum sensitivity of 1 in 105
R
nucleated cells or higher
Imaging þ MRD-negative MRD-negative as defined by next-generation flow cytometry or NGS plus

Disappearance of every area of increased tracer uptake found at baseline or a preceding PET-CT or de-
crease to < mediastinal blood pool SUV or decrease to less than that of surrounding normal tissue
BM, bone marrow; IMWG, International Myeloma Working Group; MM, multiple myeloma; MRD, minimal residual disease; NGS, next-generation sequenc-
ing; PET-CT, positron emission tomography-computed tomography; SUV, standardised uptake value.
Adapted from [10] with permission from Elsevier; permission conveyed through Copyright Clearance Center, Inc.

by guest
on 05 February 2018
Table 6. Major treatment regimens in multiple myeloma
Regimen Usual dosing schedule
Front-line:
Bortezomib/melphalan/prednisone Bortezomib 1.3 mg/m2 subcutaneously days 1, 8, 15, 22; melphalan 9 mg/m2 orally days 1–4; prednisone
(VMP) [11] 60 mg/m2 orally days 1–4; repeated every 35 days
Lenalidomide/low-dose dexa- Lenalidomide 25 mg orally days 1–21; dexamethasone 40 mg orally days 1, 8, 15, 22; repeated every
methasone (Rd) [12] 28 days
Melphalan/prednisone/thalidomide Melphalan 0.25 mg/kg orally days 1–4 (use 0.20 mg/kg/day orally days 1–4 in patients over the age of 75); pred-
(MPT) [13] nisone 2 mg/kg orally days 1–4; thalidomide 100–200 mg orally days 1–28 (use 100 mg dose in patients
>75); repeated every 6 weeks
Bortezomib/cyclophosphamide/ Cyclophosphamide 300 mg/m2 orally days 1, 8, 15 and 22; bortezomib 1.3 mg/m2 i.v. on days 1, 8, 15, 22; dexa-
dexamethasone (VCD) [14] methasone 40 mg orally on days 1, 8, 15, 22; repeated every 4 weeks
Bortezomib/thalidomide/dexa- Bortezomib 1.3 mg/m2 subcutaneously days 1, 8, 15, 22; thalidomide 100–200 mg orally days 1–21; dexametha-
methasone (VTD) [14] sone 20 mg on day of and day after bortezomib (or 40 mg days 1, 8, 15, 22); repeated every 4 weeks
4 cycles as pre-transplant induction therapy
Bortezomib/lenalidomide/dexa- Bortezomib 1.3 mg/m2 subcutaneously days 1, 8, 15; lenalidomide 25 mg orally days 1–14; dexamethasone
methasone (VRd) [14] 20 mg on day of and day after bortezomib (or 40 mg days 1, 8, 15, 22); repeated every 3 weeks
Relapse/refractory disease:
Carfilzomib/lenalidomide/dexa- Carfilzomib 20 mg/m2 (cycle 1) and 27 mg/m2 (subsequent cycles) i.v. on days 1, 2, 8, 9, 15, 16; lenalidomide
methasone (KRd) [24, 32] 25 mg orally days 1–21; dexamethasone 40 mg on days 1, 8, 15, 22; 28-day cycles
Bortezomib/dexamethasone/pano- Bortezomib 1.3 mg/m2 subcutaneously days 1, 8, 15, 22; dexamethasone 20 mg on day of and day after borte-
binostat (VD-Pano) [31] zomib; panobinostat 20 mg orally days 1, 3, 5 week 1 and 2; repeated every 3 weeks (cycles 1–8)
Carfilzomib/dexamethasone (Kd) Carfilzomib 56 mg/m2 i.v. days 1, 2, 8, 9, 15, 16 (20 mg/m2 days 1, 2, cycle 1 only); dexamethasone 20 mg days 1,
[33] 2, 8, 9, 15, 16, 22, 23; 28-day cycles
Lenalidomide/dexamethasone/ Lenalidomide 25 mg orally days 1–21; dexamethasone 40 mg weekly; elotuzumab 10 mg/kg i.v. weekly cycle 1
elotuzumab (Rd-Elo) [34] and 2, every other week cycles 3þ; repeated every 28 days
Lenalidomide/dexamethasone/ixa- Lenalidomide 25 mg orally days 1–21; dexamethasone orally 40 mg days 1, 8, 15, 22; ixazomib 4 mg orally days
zomib (IRd) [35] 1, 8, 15; repeated every 28 days
Bortezomib/dexamethasone/dara- Bortezomib 1.3 mg/m2 subcutaneously days 1, 4, 8, 11 (cycles 1–8); dexamethasone 20 mg orally days 1, 2, 4, 5,
tumumab (DVd) [38] 8, 9, 11, 12 (cycles 1–8); daratumumab 16 mg/kg i.v. every week (cycles 1–3), every 3 weeks (cycles 4–8), every
4 weeks (cycles 9þ); cycles 1–8: repeated every 21 days; cycles 9þ: repeated every 28 days
Lenalidomide/dexamethasone/dar- Lenalidomide 25 mg orally days 1–21; dexamethasone 40 mg orally weekly; daratumumab 16 mg/kg i.v. weekly
atumumab (DRd) [39] (cycles 1-2), every other week (cycles 3-6), every 4 weeks (cycles 7þ)
Two trials have prospectively compared VCD versus PAD [II, B] Tandem ASCT was evaluated before the era of novel agents.
[22], and VTD versus VCD [II, B] [23]. The first one showed that The benefit of tandem ASCT was observed in patients not achiev-
VCD and PAD were equally effective in terms of response, and ing very good partial response after the first ASCT [14]. In a re-
that VCD was less toxic. The second one showed that VTD is the cent study from The Netherlands and Germany (HOVON-65/
more effective regimen compared with VCD in terms of very GMMG-HD4 trial), in the context of bortezomib induction and
good partial response rates, but was associated with a higher rate maintenance treatment, OS was better in the GMMG group (tan-
of peripheral neuropathy. Based on response rates, depth of re- dem ASCT) in contrast to the HOVON group (single ASCT)
sponse and PFS as surrogate markers for outcome, three-drug [26]. Nevertheless, the trial was not powered to compare single
combinations including at least bortezomib and dexamethasone versus double ASCT. The recent EMN02/H095 trial compared
are currently the standard of care before ASCT [14]. In Europe, single versus tandem ASCT upfront; PFS was improved in the
VTD and VCD are the most preferred regimens [14]. RVD, when tandem ASCT arm of the study, hampered by a short follow-up
approved, will probably be widely used [20]. Carfilzomib- [21]. Additional data from a similar trial (BMT CTN 0702,
lenalidomide and dexamethasone (KRd) [24], currently being NCT01109004) being conducted in the USA will solve this im-
evaluated in ongoing phase III trials, is associated with high re- portant issue.
sponse rates, but is currently only approved for treatment of Allogeneic SCT is not indicated as part of front-line therapy
relapsed MM. and should only be carried out in the context of a clinical trial.
Four to six courses of induction are recommended before pro-
ceeding to stem cell collection.
Melphalan [200 mg/m2 intravenous (i.v.)] is the standard pre-
parative regimen before ASCT [II, B] [25]. Peripheral blood pro-
Consolidation
genitor cells are the preferred source of stem cells, rather than BM Several trials have shown that consolidation is improving the
[III, B] [14]. depth of response [14]. However, in the era of novel agent-based

by guest
on 05 February 2018
thalidomide or lenalidomide, prolongs PFS [I, A] [14]. A recent
meta-analysis based on individual patient data of more than 1200
cases demonstrated that lenalidomide maintenance following
ASCT is associated with an overall OS benefit of more than two
years [I, A] [29]. In February 2017, the EMA approved lenalido-
mide as monotherapy for the maintenance treatment of adult pa-
tients with newly diagnosed MM who have undergone ASCT.
Bortezomib maintenance was also evaluated during a two-year
study and was associated with a survival benefit over thalidomide
maintenance, but induction was not identical in the two arms of
this prospective trial [26]. Bortezomib and thalidomide are not
approved in this setting.
Treatment of relapsed/refractory disease

The choice of therapy in the relapse setting depends on several
parameters such as age, performance status, comorbidities, the
type, efficacy and tolerance of the previous treatment, the num-
ber of prior treatment lines, the available remaining treatment
options, the interval since the last therapy and the type of relapse
(i.e. clinical versus biochemical relapse; in the case of biochemical
relapse, treatment can be delayed) (Table 6) [30].
Until 2015, the EMA had approved, at the time of first relapse
and beyond, lenalidomide in combination with dexamethasone
[I, A] and bortezomib, either alone as single-agent or in combin-
Figure 1. Front-line treatment of symptomatic multiple myeloma ation with PEGylated doxorubicin [I, A]. Nevertheless, bortezo-
outside clinical trials. mib is mostly used in combination with dexamethasone in the
CTD, cyclophosphamide, thalidomide, dexamethasone; MP, melpha- relapse setting [30].
lan, prednisone; MPT, melphalan, prednisone, thalidomide; PAD, bor- In 2015 and 2016, based on the results of phase III prospective
tezomib, doxorubicin, dexamethasone; Rd, lenalidomide, low-dose randomised trials, new triplet combinations were approved by the
dexamethasone; RVD, lenalidomide, bortezomib, dexamethasone;
EMA. Panobinostat, a panHDAC inhibitor, in combination with
VCD, bortezomib, cyclophosphamide, dexamethasone; VMP, borte-
bortezomib and dexamethasone, is now indicated for the treat-
zomib, melphalan, prednisone; VRd, lenalidomide, low-dose dexa-
ment of patients with relapsed/refractory MM who have received
methasone, bortezomib; VTD, bortezomib, thalidomide,
dexamethasone. at least two prior regimens including bortezomib and an immu-
nomodulatory agent [II, C] [31]. Carfilzomib, the second-in-class
proteasome inhibitor, has also been approved at the dose of
27 mg/m2 in combination with lenalidomide and dexamethasone
for the treatment of patients with MM who have received at least
induction therapy, there is still not enough evidence that consoli-
one prior therapy [II, A] [32]. Carfilzomib has also been approved
dation therapy should be systematically applied. Ongoing trials
at the dose of 56 mg/m2 in combination with dexamethasone
will clarify the impact of consolidation, especially in the setting of
alone in patients with at least one line of prior therapy [II, A] [33].
front-line ASCT, such as the EMN02/H095 and BMT CTN 0702
Elotuzumab, a monoclonal antibody targeting SLAM-F7, has also
studies.
been approved in combination with lenalidomide and dexa-
methasone for the treatment of MM in patients who have received
at least one prior therapy [II, B] [34]. Ixazomib, the first oral pro-
Maintenance teasome inhibitor, in combination with lenalidomide and dexa-
In elderly patients following induction, several randomised trials methasone was also approved by the EMA in 2016 in patients who
have explored the benefit of maintenance therapy in terms of OS have received at least one prior line of therapy [II, A] [35].
using either immunomodulatory drugs (IMiDs) or bortezomib: In very advanced-stage disease, two other drugs are EMA-
MP or a reduced-dose regimen of CTD (CTDa) with or without approved for the treatment of relapsed MM. Pomalidomide, the
thalidomide maintenance [19], MP versus MPR versus MPR-R third-in-class IMiD, in combination with low-dose dexametha-
[17], VMPT-VT versus VMP [27], VMP versus VTP followed by sone, is approved in patients who have received at least two prior
either VP or VT maintenance [28]. These trials have not demon- therapies, including both lenalidomide and bortezomib, and
strated a clear benefit in OS, and the drugs are not yet approved whose disease progressed after treatment with these medicines
by the EMA; therefore, systematic maintenance therapy currently [II, A] [36]. Daratumumab, a monoclonal antibody targeting
cannot be recommended in elderly patients. CD38, was also recently approved for the treatment of adults with
In young patients following ASCT, phase III randomised trials relapsed/refractory MM whose previous treatment included a pro-
have demonstrated that maintenance therapy with IMiDs, either teasome inhibitor and an immunomodulatory agent and whose

by guest
on 05 February 2018
Figure 2. Treatment of relapse.

Bort, bortezomib; Cyclo, cyclophosphamide; Dara, daratumumab; DaraRd, daratumumab, lenalidomide, low dose dexamethasone; DaraVD, daratumu-
mab, bortezomib, dexamethasone; Dex, dexamethasone; Elo, elotuzumab; EloRd, elotuzumab, lenalidomide, low dose dexamethasone; EloVD, elotu-
zumab, bortezomib, dexamethasone; IMiD, immunomodulatory drug; Ixa, izaxomib; IxaRd, izaxomib, lenalidomide, low dose dexamethasone; Kd,
carfilzomib, low dose dexamethasone; KRd, carfilzomib, lenalidomide, low dose dexamethasone; PanoVD, panobinostat, bortezomib, dexamethasone;
Rd, lenalidomide, low-dose dexamethasone; VCD, bortezomib, cyclophosphamide, dexamethasone; Vd, bortezomib, low dose dexamethasone.
disease worsened after treatment [II, A] [37]. Daratumumab has maintenance until progression [44]. The role of novel agents
also shown significant efficacy at earlier stages of the disease, first re- such as monoclonal antibodies and immunotherapies, as well as
lapse and beyond in combination with bortezomib-dexamethasone metronomic approaches and allogeneic transplant should be for-
[II, A] [38] or lenalidomide-dexamethasone [II, A] [39] in two mally investigated in these patients.
randomised phase III clinical trials. These two new triplet combin-
ations may be considered in the near future as standards of care, in
the case of regulatory approval. Supportive care
In young patients, a second ASCT may be considered, provided
that the patient responded well to the previous ASCT and had a Bone disease and spinal cord compression
PFS of more than 24 months [40]. In the relapse setting, allogeneic
SCT should only be carried out in the context of a clinical trial. The i.v. agents pamidronate and zoledronic acid are of clinical
When possible, patients should be offered participation in clin- benefit in the treatment of bone disease in patients with MM
ical trials. [II, A] [4]. Pamidronate is administered at a monthly dose of
Treatment of relapse is shown in Figure 2. 90 mg via a 2 h i.v. infusion. Zoledronic acid is at least as effective
as pamidronate at a monthly dose of 4 mg and has the advantage to
Management of solitary plasmacytoma be administered via a 15 min infusion. In patients with moderate
renal function impairment (creatinine clearance 30–60 mL/min),
The diagnostic criteria require the existence of a histologically-
the dose of zoledronic acid must be reduced to a maximum of
confirmed solitary plasma cell tumour in the absence of BM
3 mg with no change to infusion time, while pamidronate should
infiltration and CRAB symptoms [41]. Local radiotherapy is the
be given via a 4 h infusion [4]. Patients with hypercalcaemia should
preferred treatment of choice, but about two-thirds of patients de-
also receive zoledronic acid. The most challenging complication is
velop MM at 10 years’ follow-up [42]. Moreover, following the use
osteonecrosis of the jaw. The current recommendations based on
of high sensitivity flow cytometry, half of the patients showed occult
consensus panels from both the IMWG and the American Society
BM infiltration, and half of these cases progressed at 2 years [43].
of Clinical Oncology do not recommend the initial use of
bisphosphonates for more than 2 years [4, 45]. In relapsed patients,
Management of plasma cell leukaemia treatment with bisphosphonates can be restarted and administered
The outcomes of patients with plasma cell leukaemia (PCL) re- concomitantly with active therapy. New molecules such as denosu-
mains uniformly poor, with a median OS of only around 1 year mab are under investigation. Orthopaedic surgery is required in
[44]. There are no specific treatment approaches for PCL. The patients with pathological fractures or at risk of long bones, and
use of multidrug combinations (including both a proteasome in- may need to be complemented with radiotherapy [4, 45]. Patients
hibitor and an IMiD) appears to be a logical choice, along with with severe back pain due to vertebral compression fractures can
the use of HDT in eligible patients, followed by prolonged benefit from vertebroplasty or kyphoplasty [4, 45].

by guest
on 05 February 2018
Spinal cord compression is an emergency that requires treat- Follow-up, long-term implications and
ment with high-dose dexamethasone and simultaneous local
survivorship
radiotherapy should be started as soon as possible; surgery should
be used in the case of bone fragments within the spinal route [4]. Full blood count, serum and urine electrophoresis and/or serum-
FLC determination, creatinine and calcium should be carried out
Anaemia, BM failure and infections every 2–3 months (outside the context of a clinical trial). Serum-
FLC may contribute to detect light chain escape.
Recombinant human erythropoietin and darbepoetin alfa can be In the case of bone pain, skeletal X-ray, MRI, WBLD-CT or
used for the treatment of myeloma-associated anaemia (haemo- PET-CT should be carried out to detect new bone lesions.
globin level < 10 g/dL), once other causes of anaemia have been MM has for a long time been considered as an incurable disease.
excluded [4]. The target is to maintain haemoglobin around Recent trials incorporating novel agents and ASCT identify CR as an
12 g/dL (below 14 g/dL to avoid thromboembolic complications important predictor of long-term survival for transplant eligible
and hypertension) [II, B] [46].
MM patients, with a statistical cure fraction of 15% [48]. These
Treatment with granulocyte colony-stimulating factor (G-CSF)
numbers will increase with the addition of quadruplet combinations
may be required to treat chemotherapy-induced severe granulocy-
including monoclonal antibodies as part of front-line therapies.
topaenia. Infectious episodes require immediate therapy with
broad spectrum antibiotics. Prophylaxis of infection remains con-
troversial but may be beneficial within the first 2–3 months of ini-
tiation of therapy, especially in patients receiving lenalidomide or Methodology
pomalidomide, or in patients at high risk of infection (previous These Clinical Practice Guidelines were developed in accordance
serious infections or neutropaenia) [4]. Influenza and pneumo- with the ESMO standard operating procedures for Clinical Practice
coccal vaccinations are recommended [4]. Acyclovir or valacyclo- Guidelines development http://www.esmo.org/Guidelines/ESMO-
vir for herpes-zoster virus prophylaxis is recommended for Guidelines-Methodology. The relevant literature has been selected
patients receiving proteasome inhibitor-based therapies [4]. i.v. by the expert authors. A summary of recommendations is shown in
immunoglobulin prophylaxis is not routinely recommended [4]. Table 7. Levels of evidence and grades of recommendation have
been applied using the system shown in Table 8. Statements without
Renal failure grading were considered justified standard clinical practice by the
Bortezomib-based therapies (in combination with dexametha- experts and the ESMO Faculty. This manuscript has been subjected
sone 6 thalidomide or doxorubicin or cyclophosphamide) is the to an anonymous peer review process.
treatment of choice in patients with renal failure [II, B] [4, 47].
Other proteasome inhibitors are under investigation. The use of
high cut-off dialysis filters to remove FLCs or to reverse renal fail- Disclosure
ure is under evaluation in randomised trials. PM has reported advisory boards and honoraria from Celgene,
Janssen, Takeda, Amgen and Novartis; JSM has reported advisory
Venous thromboembolism. Patients with MM have an increased boards for Janssen, Celgene, Takeda, Amgen, Merck Sharp and
risk of thrombosis, with a baseline risk of 3%–4% of venous Dohme, Bristol-Myers Squibb and Novartis; PS has reported ad-
thrombotic events, and this risk is significantly enhanced in the visory boards, honoraria and research support from Amgen,
face of therapy with use of specific agents. High-dose dexametha- Celgene, Janssen, Tanken, Karyopharm; MVM has reported hon-
sone, cytotoxic chemotherapy such as doxorubicin and IMiDs oraria from Janssen, Celgene, Amgen, Takeda; EZ has reported
(thalidomide and lenalidomide) increase this risk substantially. honoraria and advisory boards for Amgen, Bristol-Myers Squibb,
Other factors such as reduced mobility due to neurological com- Janssen and Celgene; RH has reported consultant and advisory
plications or bone pain, associated fractures, concurrent use of
erythropoiesis-stimulating agents and prior personal or family
history of thrombotic events increase the risk of thromboembolic Table 7. Summary of recommendations
events. The current recommendations for patients with MM who The diagnosis of MM must include the criteria updated in 2014 by the
are due to start IMiD therapy are to use aspirin (100 mg) in the International Myeloma Working Group.
absence of risk factors for thrombosis and to use full dose antico- Immediate treatment is not recommended for patients with indolent
agulants for those at higher risk (low molecular weight heparin or myeloma.
full-dose warfarin) [4]. Sub-therapeutic doses of anticoagulants For patients < 70 years in good clinical condition, induction followed by
such as small doses of warfarin are not recommended. high-dose therapy with ASCT is the standard treatment.
For relapsed/refractory MM, the most commonly used regimens are
proteasome inhibitor- or lenalidomide-containing regimens. New triplet
combinations are increasing PFS.
Personalised medicine In advanced cases, pomalidomide plus low-dose dexamethasone and
In 2016, no prognostic factor or staging system, including R-ISS daratumumab are approved.
or gene-expression profiling, is used routinely to define a risk-
adapted strategy. In this disease setting, more research is needed ASCT, autologous stem cell transplantation; PFS, progression-free sur-
to identify molecular markers which could lead to advances in vival; MM, multiple myeloma.
personalised medicine.

by guest
on 05 February 2018
Table 8. Levels of evidence and grades of recommendation (adapted from References
the Infectious Diseases Society of America-United States Public Health 1. Palumbo A, Bringhen S, Ludwig H et al. Personalized therapy in multiple
Service Grading Systema) myeloma according to patient age and vulnerability: a report of the
Levels of evidence European Myeloma Network (EMN). Blood 2011; 118: 4519–4529.
2. Rajkumar SV, Dimopoulos MA, Palumbo A et al. International
Myeloma Working Group updated criteria for the diagnosis of multiple
myeloma. Lancet Oncol 2014; 15: e538–e548.
analyses of well-conducted randomised trials without 3. Rajkumar SV. Updated diagnostic criteria and staging system for mul-
heterogeneity tiple myeloma. Am Soc Clin Oncol Educ Book 2016; 35: e418–e423.
II Small randomised trials or large randomised trials with a suspicion 4. Terpos E, Kleber M, Engelhardt M et al. European Myeloma Network
of bias (lower methodological quality) or meta-analyses of such guidelines for the management of multiple myeloma-related complica-
trials or of trials demonstrated heterogeneity tions. Haematologica 2015; 100: 1254–1266.
III Prospective cohort studies 5. Greipp PR, San Miguel J, Durie BG et al. International staging system for
IV Retrospective cohort studies or case–control studies multiple myeloma. J Clin Oncol 2005; 23: 3412–3420.
V Studies without control group, case reports, expert opinions 6. Sonneveld P, Avet-Loiseau H, Lonial S et al. Treatment of multiple mye-
loma with high-risk cytogenetics: a consensus of the International
Grades of recommendation Myeloma Working Group. Blood 2016; 127: 2955–2962.
A Strong evidence for efficacy with a substantial clinical benefit, 7. Palumbo A, Avet-Loiseau H, Oliva S et al. Revised international staging
strongly recommended system for multiple myeloma: a report from International Myeloma
B Strong or moderate evidence for efficacy but with a limited clin- Working Group. J Clin Oncol 2015; 33: 2863–2869.
ical benefit, generally recommended 8. Palumbo A, Bringhen S, Mateos MV et al. Geriatric assessment predicts
survival and toxicities in elderly myeloma patients: an International
C Insufficient evidence for efficacy or benefit does not outweigh
Myeloma Working Group report. Blood 2015; 125: 2068–2074.
the risk or the disadvantages (adverse events, costs,. . .),
9. Rajkumar SV, Harousseau JL, Durie B et al. Consensus recommendations
optional for the uniform reporting of clinical trials: report of the International
D Moderate evidence against efficacy or for adverse outcome, gen- Myeloma Workshop Consensus Panel 1. Blood 2011; 117: 4691–4695.
erally not recommended 10. Kumar S, Paiva B, Anderson KC et al. International Myeloma Working
E Strong evidence against efficacy or for adverse outcome, never Group consensus criteria for response and minimal residual disease as-
recommended sessment in multiple myeloma. Lancet Oncol 2016; 17: e328–e346.
11. San Miguel JF, Schlag R, Khuageva NK et al. Bortezomib plus melphalan
a
By permission of the Infectious Diseases Society of America [49]. and prednisone for initial treatment of multiple myeloma. N Engl J Med
2008; 359: 906–917.
12. Benboubker L, Dimopoulos MA, Dispenzieri A et al. Lenalidomide and
dexamethasone in transplant-ineligible patients with myeloma. N Engl J
boards from Amgen, Takeda, Bristol-Myers Squibb and Celgene, Med 2014; 371: 906–917.
13. Fayers PM, Palumbo A, Hulin C et al. Thalidomide for previously un-
honoraria for Amgen, Bristol-Myers Squibb, Takeda, Celgene
treated elderly patients with multiple myeloma: meta-analysis of 1685 in-
and Janssen and research funding from Takeda, Novartis, Amgen dividual patient data from 6 randomized clinical trials. Blood 2011; 118:
and Janssen; MAD has reported honoraria from Janssen, 1239–1247.
Celgene, Amgen and Takeda; HL has reported advisory boards 14. Moreau P, Attal M, Facon T. Frontline therapy of multiple myeloma.
and honoraria from Amgen, Celgene, Bristol-Myers Squibb, Blood 2015; 125: 3076–3084.
being a member of the speaker’s bureau of Janssen-Cilag, Amgen, 15. Durie BG, Hoering A, Abidi MH et al. Bortezomib with lenalidomide
and dexamethasone versus lenalidomide and dexamethasone alone in
Celgene, Bristol-Myers Squibb and research support from
patients with newly diagnosed myeloma without intent for immediate
Takeda and Amgen; HE has reported advisory boards, honoraria autologous stem-cell transplant (SWOG S0777): a randomised, open-
and research support from Janssen and Celgene and being a label, phase 3 trial. Lancet 2017; 389: 519–527.
member of their speaker’s bureau, advisory boards and honoraria 16. Pönisch W, Mitrou PS, Merkle K et al. Treatment of bendamustine and
from Amgen and Novartis and being a member of their speaker’s prednisone in patients with newly diagnosed multiple myeloma results in
bureau; SZ has reported research support from Takeda, Celgene, superior complete response rate, prolonged time to treatment failure and
improved quality of life compared to treatment with melphalan and
Janssen, advisory boards for Takeda, Celgene, Janssen and prednisone – a randomized phase III study of the East German Study
Novartis; TF has reported advisory boards for Celgene, Janssen, Group of Hematology and Oncology (OSHO). J Cancer Res Clin Oncol
Takeda and Amgen and being a member of the speaker’s bureau 2006; 132: 205–212.
of Celgene, Janssen, Takeda and Amgen; ET has reported honora- 17. Palumbo A, Hajek R, Delforge M et al. Continuous lenalidomide treat-
ria from Amgen, Bristol-Myers Squibb, Celgene, Janssen, ment for newly diagnosed multiple myeloma. N Engl J Med 2012; 366:
1759–1769.
Novartis, Takeda, research support by Amgen, Celgene and
18. Zweegman S, van der Holt B, Mellqvist UH et al. Melphalan, prednisone,
Janssen and participation in the steering committee of Amgen and lenalidomide versus melphalan, prednisone, and thalidomide in un-
and Takeda and in the data monitoring committee of Celgene; treated multiple myeloma. Blood 2016; 127: 1109–1116.
HG has reported research support and honoraria from Celgene, 19. Morgan GJ, Davies FE, Gregory WM et al. Cyclophosphamide, thalido-
Janssen, Chugai, Novartis and Bristol-Myers Squibb and advisory mide, and dexamethasone (CTD) as initial therapy for patients with mul-
boards for Janssen, Celgene, Novartis, Amgen Takeda and tiple myeloma unsuitable for autologous transplantation. Blood 2011;
118: 1231–1238.
Bristol-Myers Squibb; CB has reported research funds from
20. Attal M, Lauwers-Cances V, Hulin C et al. Autologous transplantation
Roche and Janssen and being a member of the speaker’s bureau for multiple myeloma in the era of new drugs: a phase III study of the
of Roche, Janssen and Pfizer; HAL, MC and MA have declared no Intergroupe Francophone du Myelome (IFM/DFCI 2009 trial). Blood
conflicts of interest. 2015; 126: abstr. 391.

by guest
on 05 February 2018
21. Cavo M, Palumbo A, Zweegman S et al. Upfront autologous stem cell relapsed or refractory multiple myeloma (ENDEAVOR): a randomised,
transplantation (ASCT) versus novel agent-based therapy for multiple phase 3, open-label, multicentre study. Lancet Oncol 2016; 17: 27–38.
myeloma (MM): a randomized phase 3 study of the European Myeloma 34. Lonial S, Dimopoulos M, Palumbo A et al. Elotuzumab therapy for relapsed
Network (EMN02/HO95 MM trial). J Clin Oncol 2016; 34 (Suppl): abstr. or refractory multiple myeloma. N Engl J Med 2015; 373: 621–631.
8000. 35. Moreau P, Masszi T, Grzasko N et al. Oral ixazomib, lenalidomide, and
22. Mai EK, Bertsch U, Dürig J et al. Phase III trial of bortezomib, cyclophos- dexamethasone for multiple myeloma. N Engl J Med 2016; 374:
phamide and dexamethasone (VCD) versus bortezomib, doxorubicin 1621–1634.
and dexamethasone (PAd) in newly diagnosed myeloma. Leukemia 36. San Miguel J, Weisel K, Moreau P et al. Pomalidomide plus low-dose
2015; 29: 1721–1729. dexamethasone versus high-dose dexamethasone alone for patients with
23. Moreau P, Hulin C, Macro M et al. VTD is superior to VCD prior to in- relapsed and refractory multiple myeloma (MM-003): a randomised,
tensive therapy in multiple myeloma: results of the prospective open-label, phase 3 trial. Lancet Oncol 2013; 14: 1055–1066.
IFM2013-04 trial. Blood 2016; 127: 2569–2574. 37. Usmani SZ, Weiss BM, Plesner T et al. Clinical efficacy of daratumumab
24. Korde N, Roschewski M, Zingone A et al. Treatment with carfilzomib- monotherapy in patients with heavily pretreated relapsed or refractory
lenalidomide-dexamethasone with lenalidomide extension in patients multiple myeloma. Blood 2016; 128: 37–44.
with smoldering or newly diagnosed multiple myeloma. JAMA Oncol 38. Palumbo A, Chanan-Khan A, Weisel K et al. Daratumumab, bortezomib,
2015; 1: 746–754. and dexamethasone for multiple myeloma. N Engl J Med 2016; 375:
25. Moreau P, Facon T, Attal M et al. Comparison of 200 mg/m2 melphalan 754–766.
and 8Gy total body irradiation plus 140 mg/m2 melphalan as condition- 39. Dimopoulos MA, Oriol A, Nahi H et al. Daratumumab, lenalidomide,
ing regimen for peripheral blood stem cell transplantation in patients and dexamethasone for multiple myeloma. N Engl J Med 2016; 375:
with newly diagnosed multiple myeloma: final analysis of the 1319–1331.
40. Giralt S, Garderet L, Durie B et al. American Society of Blood and
Intergroupe Francophone du Myelome 9502 randomized trial. Blood
Marrow Transplantation, European Society of Blood and Marrow
2002; 99: 731–735.
Transplantation, Blood and Marrow Transplant Clinical Trials
26. Sonneveld P, Schmidt-Wolf IG, van der Holt B et al. Bortezomib induc-
Network, and International Myeloma Working Group Consensus
tion and maintenance treatment in patients with newly diagnosed mul-
Conference on Salvage Hematopoietic Cell Transplantation in Patients
tiple myeloma: results of the randomized phase III HOVON-65/GMMG-
with Relapsed Multiple Myeloma. Biol Blood Marrow Transplant 2015;
HD4 trial. J Clin Oncol 2012; 30: 2946–2955.
21: 2039–2051.
27. Palumbo A, Bringhen S, Larocca A et al. Bortezomib-melphalan-prednis-
41. Rajkumar SV. Multiple myeloma: 2014 update on diagnosis, risk-
one-thalidomide followed by maintenance with bortezomib-thalidomide
stratification, and management. Am J Hematol 2014; 89: 999–1009.
compared with bortezomib-melphalan-prednisone for initial treatment
42. Reed V, Shah J, Medeiros LJ et al. Solitary plasmacytomas: outcome and prog-
of multiple myeloma: updated follow-up and improved survival. J Clin
nostic factors after definitive radiation therapy. Cancer 2011; 117: 4468–4474.
Oncol 2014; 32: 634–640. 43. Paiva B, Chandia M, Vidriales MB et al. Multiparameter flow cytometry
28. Mateos MV, Oriol A, Martinez-Lopez J et al. Maintenance therapy with for staging of solitary bone plasmacytoma: new criteria for risk of pro-
bortezomib plus thalidomide or bortezomib plus prednisone in elderly gression to myeloma. Blood 2014; 124: 1300–1303.
multiple myeloma patients included in the GEM2005MAS65 trial. Blood 44. Fernandez de Larrea C, Kyle RA, Durie BG et al. Plasma cell leukemia:
2012; 120: 2581–2588. consensus statement on diagnostic requirements, response criteria and
29. Attal M, Palumbo A, Holstein SA et al. Lenalidomide maintenance after treatment recommendations by the International Myeloma Working
high-dose melphalan and autologous stem-cell transplantation in mul- Group. Leukemia 2013; 27: 780–791.
tiple myeloma: a meta-analysis of overall survival. J Clin Oncol 2016; 45. Terpos E, Morgan G, Dimopoulos MA et al. International Myeloma
34(Suppl): abstr. 8001. Working Group recommendations for the treatment of multiple
30. Laubach J, Garderet L, Mahindra A et al. Management of relapsed mul- myeloma-related bone disease. J Clin Oncol 2013; 31: 2347–2357.
tiple myeloma: recommendations of the International Myeloma 46. Snowden JA, Ahmedzai SH, Ashcroft J et al. Guidelines for supportive
Working Group. Leukemia 2016; 30: 1005–1017. care in multiple myeloma 2011. Br J Haematol 2011; 154: 76–103.
31. San-Miguel JF, Hungria VT, Yoon SS et al. Panobinostat plus bortezo- 47. Dimopoulos MA, Terpos E, Chanan-Khan A et al. Renal impairment in
mib and dexamethasone versus placebo plus bortezomib and dexametha- patients with multiple myeloma: a consensus statement on behalf the
sone in patients with relapsed or relapsed and refractory multiple International Myeloma Working Group. J Clin Oncol 2010; 28:
myeloma: a multicentre, randomised, double-blind phase 3 trial. Lancet 4976–4984.
Oncol 2014; 15: 1195–1206. 48. Usmani SZ, Crowley J, Hoering A et al. Improvement in long-term out-
32. Stewart AK, Rajkumar SV, Dimopoulos MA et al. Carfilzomib, lenalido- comes with successive Total Therapy trials for multiple myeloma: are pa-
mide, and dexamethasone for relapsed multiple myeloma. N Engl J Med tients now being cured? Leukemia 2013; 27: 226–232.
2015; 372: 142–152. 49. Dykewicz CA. Summary of the guidelines for preventing opportunistic
33. Dimopoulos MA, Moreau P, Palumbo A et al. Carfilzomib and dexa- infections among hematopoietic stem cell transplant recipients. Clin
methasone versus bortezomib and dexamethasone for patients with Infect Dis 2001; 33: 139–144.

by guest
on 05 February 2018
Waldenström’s macroglobulinaemia: ESMO Clinical

and follow-up†
C. Buske1, V. Leblond2, M. Dimopoulos3, E. Kimby4, U. Jäger5 & M. Dreyling6, on behalf of the
1
Comprehensive Cancer Center Ulm, Institute of Experimental Cancer Research, University Hospital Ulm, Ulm, Germany; 2Departement of Hematology Pitié Salpêtrière
Hospital, Pierre et Marie Curie University, UPMC GRC11-GRECHY, 75013 Paris, France; 3Department of Clinical Therapeutics, University of Athens School of Medicine,
Alexandra Hospital, Athens, Greece; 4Department of Medicine at Huddinge, Division of Hematology, Karolinska Institutet and The Hematology Center, Karolinska University
Hospital, Stockholm, Sweden; 5Division of Haematology and Haemostaseology, Department of Medicine I, Medical University of Vienna (MUW), Vienna, Austria;
6
Department of Medicine III, University Hospital Grosshadern, LMU Munich, Germany
incidence assessments for individual patients are carried out with the same
methodology, optimally in the same laboratory because of known
Waldenström’s macroglobulinaemia (WM) is a rare disease. It intralaboratory as well as interlaboratory variations [4]. To test for
accounts for 1%–2% of haematological neoplasms with a IgM-associated coagulopathy and haemolysis, Coombs test and
reported age-adjusted incidence rate of 3.4 per million among testing for cold agglutinin disease and coagulation parameter are
clinical practice
males and 1.7 per million among females in the United States recommended. The presence of cold agglutinins or cryoglobulins
guidelines
and 7.3 and 4.2 per million European standard population may affect the determination of IgM levels; therefore, testing for
[1, 2]. WM is a disease of the elderly with a median age of cold agglutinins and cryoglobulins should be carried out at
63–68 years with a male predominance. diagnosis. Serum-free light chain testing is not advised in routine
practice as its relevance for the management of WM patients is
diagnosis currently under evaluation. Recent findings documented a strong
association between WM and the MYD88 L265P variant, which
To establish the diagnosis of WM, it is necessary to demonstrate might serve as an additional tool to diagnose WM and to separate
an IgM monoclonal protein, along with histological evidence of it from other entities such as multiple myeloma, monoclonal
infiltration of the bone marrow by lymphoplasmacytic cells in gammopathy of undetermined significance, splenic marginal zone
line with the diagnosis of lymphoplasmacytic lymphoma (LPL) lymphoma and MALT lymphoma [5].
[3]. Thus, detection of monoclonal IgM without the
histopathological diagnosis of LPL cannot be considered to be
WM. Conversely, the diagnosis of LPL without detection of staging and risk assessment
monoclonal IgM does not fulfil the criteria of WM. Besides documentation of monoclonal IgM gammopathy, as
Lymphoplasmacytic cell population in the bone marrow mentioned above, staging should include a complete blood count
should be documented by trephine biopsy and aspiration. The with differential and more detailed serum chemistry.
bone marrow infiltration should routinely be confirmed by Furthermore, the beta2 microglobulin and albumin levels should
immunophenotypic studies (flow cytometry and/or be determined, as these factors have prognostic impact. Serum
immunohistochemistry) showing expression of CD 19, CD 20, protein electrophoresis and quantification of immunoglobulin
CD 22 and CD 79a. levels (IgM, IgG, IgA) should be carried out. Some patients suffer
The presence of monoclonal IgM should be confirmed by from hyperviscosity caused by excessive levels of IgM. In this
immunofixation. Determination of IgM levels can be based on case, quantification of serum viscosity might be helpful.
either densitometry or total serum IgM quantitation by However, serum viscosity does not always correspond well to the
nephelometry. Because IgM values when assessed by clinical severity of hyperviscosity. More important are clinical
nephelometry are systematically higher than M protein values examinations such as fundoscopy, showing, e.g. venous
determined by densitometry, it is essential that sequential response engorgement (‘sausaging’) in the retinal veins, which is an
excellent indicator of clinically relevant hyperviscosity [6]. In the
*Correspondence to: ESMO Guidelines Working Group, ESMO Head Office, Via L. Taddei case of peripheral neuropathy, the evaluation of anti-myelin-
4, CH-6962 Viganello-Lugano, Switzerland. associated glycoprotein, antigangliosides M1 and anti-sulfatide
IgM antibodies may support the diagnosis of IgM-related
†
Approved by the ESMO Guidelines Working Group: July 2013. neuropathy. Also, the possibility of amyloid light-chain

by guest
on 05 February 2018
Table 1. International prognostic scoring system for Waldenström’s Rituximab can be combined with purine analogues such as
macroglobulinaemia (ISSWM) adapted from [7] cladribine or fludarabine [II, B]. Bendamustine, which carries
characteristics of both alkylating agents and purine
Risk Group Low Intermediate High analogues, in combination with rituximab is highly effective
Score 0–1 (except age) Age or 2 ≥3 in WM [II, B] [13].
5-year OSa (%) 87 68 36 Bortezomib has shown considerable activity in combination
Risk Factors Score with rituximab in the first-line treatment of WM with or
without dexamethasone [III, B] [14, 15].
Age ≥ 65 years 1
In medically non-fit patients (e.g. patients who do not tolerate
Other risk factorsb
chemotherapy because of non-lymphoma related co-
Hbc ≤11.5 g/dl 1
morbidities), single-agent rituximab is a treatment option,
Thrombod ≤100.000 × 109/l 1
Beta-2 Me >3 mg/l 1
which avoids chemotherapy-related toxic effects [III, B] [16].
IgMf >70 g/l 1 However, responses are delayed and, particularly in patients
with signs of hyperviscosity or patients with high IgM values,
a
OS, overall survival. there is the danger of so-called ‘IgM flare’, a transient increase of
b
Each of the risk factors is counted as one. serum IgM immediately following initiation of rituximab
c
Hb, haemoglobin. treatment [17, 18]. In these patients, plasmapheresis should
d
Thrombo, thrombocytes. precede rituximab application. Figure 1a summarises the
e
beta-2 M, beta-2 microglobulin. treatment algorithm for first-line therapy.
f
IgM, monoclonal protein concentration. Rituximab maintenance treatment outside of clinical trials is
not considered standard today. There are retrospective data
amyloidosis in association with peripheral neuropathy needs to suggesting clinical benefit for rituximab maintenance also in
be considered. At diagnosis, an ultrasound or computed WM, but prospectively randomised data are still missing [19].
tomography (CT) scan should be carried out to document
organomegaly/adenopathies. There is no routine role for
relapsed disease
positron emission tomography (PET) scanning unless a large-
cell lymphoma transformation is suspected. There is a consensus that an alternative rituximab/chemotherapy
With regard to risk assessment, all patients should be regimen should be used if the relapse occurs within the first year
categorised by the international prognostic scoring system for [8, 20]. The choice of the rituximab/chemotherapy depends on
WM (ISSWM), which divides patients into three risk groups the prior regimen. If the patient was treated initially with
with a 5-year survival rate ranging from 86% for the low-risk to rituximab plus alkylating agents, the salvage regimen could be
36% for the high-risk group. This is built on factors which are switched to rituximab in combination with nucleoside analogues,
easy to determine in clinical practice (Table 1) [7]. rituximab/bendamustine or bortezomib and vice versa.
If patients are chemosensitive and eligible for autologous
stem cell transplantation, myeloablative chemotherapy followed
treatment plan by reinfusion of autologous stem cells is a valid option in these
asymptomatic patients clinically aggressive cases [III, B] [21].
Allogeneic transplantation may be considered in young
Comparable with other indolent lymphomas, a watch and wait
relapsed patients with aggressive clinical course, but preferably
approach is standard for asymptomatic patients, meaning that
within clinical trials [22]. Figure 1b summarises the treatment
only patients suffering from lymphoma-related symptoms
algorithm for therapy in relapsed patients.
should start treatment [8]. In the case of WM, this includes
symptoms caused by circulating IgM such as hyperviscosity,
amyloidosis, symptomatic cryoglobulinaemia, cold agglutinin response evaluation
disease, neuropathy or disease-related haemoglobin level WM is distinct with regard to response criteria and differs in
<10 g/dl or platelet count <100 × 109/l. On the other hand, this respect from other lymphomas. This is in particular due to
monoclonal IgM per se is not a reason to initiate treatment [9]. the fact that the level of reduction of the monoclonal IgM affects
A close observation is appropriate for these patients. remission status, and that its disappearance is one of the
prerequisites for the definition of complete response in this
first line disease. There is an international consensus on how to define
Frontline treatment options include alkylating agents, the remission status in WM. These response criteria should be
nucleoside analogues, bortezomib and the monoclonal antibody used in and outside of clinical trials to be able to compare
rituximab. Fludarabine as a single agent is more effective than treatment results (Table 2) [4]. Because of the variability in
chlorambucil [I, B] [10]. The combination of rituximab with kinetics of IgM reduction with different treatment modalities
chemotherapy is among the most effective treatments and the and the apparent discrepancy between IgM and bone marrow/
first option to choose and should be considered in medically fit tissue response noted with many regimens, sequential bone
patients and, in particular, in patients who need rapid response. marrow trephine biopsies are strongly encouraged and
Options for rituximab in combination with alkylating agents mandatory inside clinical trials. Ultrasound or CT should only
are DCR (dexamethasone, cyclophosphamide and rituximab) be carried out in the case of initial splenomegaly/lymph node
[III, B] [11] or rituximab–CHOP [II, B] [12]. enlargements. A PET-CT scan is not indicated in WM.
vi | Buske et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
Figure 1. Treatment algorithms for patients with Waldenström’s macroglobulinaemia (WM, adapted from [23]). (a) Newly diagnosed WM. (b) relapsed WM.
personalised medicine follow-up

In this disease setting, more research is needed to identify Follow-up should include history, physical examination, blood
molecular markers which could lead to advances in count, routine chemistry and quantification of IgM every 3 months
personalised medicine. for 2 years, every 4–6 months for an additional 3 years, and
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt298 | vi

by guest
on 05 February 2018
Table 2. Response criteria in Waldenström’s macroglobulinaemia (WM) [4]
Response category Definition

Complete response (CR) (i) Absence of serum monoclonal IgM protein by immunofixation
(ii) Normal serum IgM level
(iii) Complete resolution of lymphadenopathy and splenomegaly if present at baseline
(iv) Morphologically normal bone marrow aspirate and trephine biopsy
Very good partial response (VGPR) (i) Monoclonal IgM protein is detectable
(ii) ≥90% reduction in serum IgM level from baselinea
(iii) Decreased lymphadenopathy/splenomegaly if present at baseline
(iv) No new signs or symptoms of active disease
Partial response (PR) (i) Monoclonal IgM protein is detectable

(ii) ≥50% but <90% reduction in serum IgM level from baseline
(iii) Decreased lymphadenopathy/splenomegaly if present at baseline
Minor response (MR) (i) Monoclonal IgM protein is detectable

(ii) ≥25% but <50% reduction in serum IgM level from baseline
(iii) No new signs or symptoms of active disease
Stable disease (SD) (i) Monoclonal IgM protein is detectable

(ii) <25% reduction and <25% increase in serum IgM level from baseline
(iii) No progression in lymphadenopathy/splenomegaly
Progressive disease (PD) (i) ≥25% increase in serum IgM levels from lowest nadir and/or
(ii) Progression in clinical features attributable to the disease
a
Sequential changes in IgM levels may be determined either by M protein quantitation by densitometry or total serum IgM quantitation by nephelometry ([4].
©2012 Blackwell Publishing Ltd. Reprinted with permission).
Grading Systema)
Levels of evidence
conducted, randomised trials without heterogeneity
V Studies without the control group, case reports, experts opinions
C Insufficient evidence for efficacy or benefit does not outweigh the risk or the disadvantages (adverse events, costs,...), optional
a
subsequently once a year with special attention to transformation

and secondary malignancies, including secondary leukaemia.
note
Minimal adequate radiological or ultrasound examinations every 6 Levels of evidence and grades of recommendation have
months for 2 years are recommended, and annually thereafter only been applied using the system shown in Table 3. Statements
in cases of initial splenomegaly or lymph node enlargement. without grading were considered justified standard
Regular CT scans are not necessary outside clinical trials. clinical practice by the experts and the ESMO faculty.
vi | Buske et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
conflict of interest macroglobulinemia, marginal zone lymphoma, or lymphoplasmacytic lymphoma.
J Clin Oncol 2013; 31: 301–307.
Prof. Buske has reported consultancy/honoraria from Celgene, 11. Dimopoulos MA, Anagnostopoulos A, Kyrtsonis MC et al. Primary treatment of
Pfizer and Roche. Prof. Leblond has reported speaker’s bureau Waldenström macroglobulinemia with dexamethasone, rituximab, and
for Roche, Mundipharma, GlaxoSmithKline; advisory board for cyclophosphamide. J Clin Oncol 2007; 25: 3344–3349.
Roche, Janssen. Dr Dimopoulos has reported honoraria from 12. Buske C, Hoster E, Dreyling M et al. The addition of rituximab to front-line therapy
Celgene, OrthoBiotech, Onyx. Prof. Kimby has reported with CHOP (R-CHOP) results in a higher response rate and longer time to
treatment failure in patients with lymphoplasmacytic lymphoma: results of a
research support from Roche; advisory board for Teva and randomized trial of the German Low-Grade Lymphoma Study Group (GLSG).
Janssen. Prof. Jäger has reported speaker’s honoraria and Leukemia 2009; 23: 153–161.
research support from Roche, Janssen-Cilag, Celgene. Prof. 13. Rummel MJ, Niederle N, Maschmeyer G et al. Bendamustine plus rituximab versus
Dreyling has reported: scientific advisory board for Celgene, CHOP plus rituximab as first-line treatment for patients with indolent and mantle-
Janssen, Pfizer, Roche; speaker’s honoraria for Celgene, Janssen, cell lymphomas: an open-label, multicentre, randomised, phase 3 non-inferiority
Pfizer, Roche; research funding to the institution from Celgene, trial. Lancet 2013; 381: 1203–1210.
Janssen, Mundipharma, Pfizer, Roche. 14. Ghobrial IM, Xie W, Padmanabhan S et al. Phase II trial of weekly bortezomib in
combination with rituximab in untreated patients with Waldenström
Macroglobulinemia. Am J Hematol 2010; 85: 670–674.
references 15. Treon SP, Ioakimidis L, Soumerai JD et al. Primary therapy of Waldenström
macroglobulinemia with bortezomib, dexamethasone, and rituximab: WMCTG
1. Groves FD, Travis LB, Devesa SS et al. Waldenström’s macroglobulinemia: incidence
clinical trial 05-180. J Clin Oncol 2009; 27: 3830–3835.
patterns in the United States, 1988–1994. Cancer 1998; 82: 1078–1081.
16. Dimopoulos MA, Zervas C, Zomas A et al. Treatment of Waldenström’s
2. Phekoo KJ, Jack RH, Davies E et al. The incidence and survival of Waldenström’s
macroglobulinemia with rituximab. J Clin Oncol 2002; 20: 2327–2333.
Macroglobulinaemia in South East England. Leuk Res 2008; 32: 55–59.
17. Ghobrial IM, Fonseca R, Greipp PR et al. Initial immunoglobulin M ‘flare’ after
3. Owen RG, Treon SP, Al-Katib A et al. Clinicopathological definition of
rituximab therapy in patients diagnosed with Waldenstrom macroglobulinemia: an
Waldenstrom’s macroglobulinemia: consensus panel recommendations from the
Eastern Cooperative Oncology Group Study. Cancer 2004; 101: 2593–2598.
Second International Workshop on Waldenstrom’s Macroglobulinemia. Semin
18. Treon SP, Branagan AR, Hunter Z et al. Paradoxical increases in serum IgM and
Oncol 2003; 30: 110–115.
viscosity levels following rituximab in Waldenstrom’s macroglobulinemia. Ann
4. Owen RG, Kyle RA, Stone MJ et al. Response assessment in Waldenström
Oncol 2004; 15: 1481–1483.
macroglobulinaemia: update from the VIth International Workshop. Br J Haematol
19. Treon SP, Hanzis C, Manning RJ et al. Maintenance Rituximab is associated with
2013; 160: 171–176.
improved clinical outcome in rituximab naive patients with Waldenstrom
5. Treon SP, Xu L, Yang G et al. MYD88 L265P somatic mutation in Waldenström’s
Macroglobulinaemia who respond to a rituximab-containing regimen. Br J
macroglobulinemia. N Engl J Med 2012; 367: 826–833.
Haematol 2011; 154: 357–362.
6. Menke MN, Feke GT, McMeel JW et al. Ophthalmologic techniques to assess the
20. Ansell SM, Kyle RA, Reeder CB et al. Diagnosis and management of Waldenström
severity of hyperviscosity syndrome and the effect of plasmapheresis in patients with
macroglobulinemia: Mayo stratification of macroglobulinemia and risk-adapted
Waldenström’s macroglobulinemia. Clin Lymphoma Myeloma 2009; 9: 100–103.
therapy (mSMART) guidelines. Mayo Clin Proc 2010; 85: 824–833.
7. Morel P, Duhamel A, Gobbi P et al. International prognostic scoring system for
21. Kyriakou C, Canals C, Sibon D et al. High-dose therapy and autologous stem-cell
Waldenstrom macroglobulinemia. Blood 2009; 113: 4163–4170.
transplantation in Waldenstrom macroglobulinemia: the Lymphoma Working Party
8. Dimopoulos MA, Gertz MA, Kastritis E et al. Update on treatment of the European Group for Blood and Marrow Transplantation. J Clin Oncol 2010;
recommendations from the Fourth International Workshop on Waldenstrom’s 28: 2227–2232.
Macroglobulinemia. J Clin Oncol 2009; 27: 120–126.
22. Kyriakou C, Canals C, Cornelissen JJ et al. Allogeneic stem-cell transplantation in
9. Kyle RA, Treon SP, Alexanian R et al. Prognostic markers and criteria to initiate patients with Waldenström macroglobulinemia: report from the Lymphoma
therapy in Waldenstrom’s macroglobulinemia: consensus panel recommendations Working Party of the European Group for Blood and Marrow Transplantation. J Clin
from the Second International Workshop on Waldenstrom’s Macroglobulinemia. Oncol 2010; 28: 4926–4934.
Semin Oncol 2003; 30: 116–120.
23. Buske C, Leblond V. How to manage Waldenstrom’s macroglobulinemia. Leukemia
10. Leblond V, Johnson S, Chevret S et al. Results of a randomized trial of 2013; 27: 762–772.
chlorambucil versus fludarabine for patients with untreated Waldenström
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt298 | vi

by guest
on 05 February 2018
Chronic myeloid leukaemia: ESMO Clinical Practice

A. Hochhaus1, S. Saussele2, G. Rosti3, F.-X. Mahon4, J. J. W. M. Janssen5, H. Hjorth-Hansen6, J. Richter7 &

C. Buske8, on behalf of the ESMO Guidelines Committee*
1
Klinik für Innere Medizin II, H€amatologie/Onkologie, Universit€atsklinikum Jena, Jena; 2III. Medizinische Klinik, Universit€atsmedizin Mannheim, Fakult€at für Klinische
Medizin Mannheim der Universit€at Heidelberg, Mannheim, Germany; 3Department of Experimental, Diagnostic and Specialty Medicine, Institute of Haematology
and Medical Oncology Lorenzo ed Ariosto Seragnoli, Bologna University School of Medicine, Bologna, Italy; 4CHU de Bordeaux, Bordeaux, France; 5Department of
Hematology, VU University Medical Center, Amsterdam, The Netherlands; 6Department of Hematology, St Olavs Hospital and Department of Cancer Research and
Molecular Medicine, NTNU, Trondheim, Norway; 7Department of Hematology, Oncology and Radiation Physics, Skane University Hospital, Lund, Sweden; 8CCC
Ulm, Institut für Experimentelle Tumorforschung, Universit€atsklinikum Ulm, Ulm, Germany
†
Approved by the ESMO Guidelines Committee: August 2003, last update May 2017. This publication supersedes the previously published version—Ann Oncol 2012;
23(Suppl 7): vi72–vi77.
Epidemiology BCR–ABL1 transcripts, or both, in peripheral blood or bone mar-

The incidence of chronic myeloid leukaemia (CML) ranges be- row (BM) cells. In 5% of cases the Philadelphia chromosome
tween 10 and 15 cases/106/year without any major geographic or cannot be detected and confirmation of diagnosis depends on the
ethnic differences [1]. The median age at diagnosis ranges be- confirmation of the BCR–ABL1 fusion by either fluorescent in
tween 60 and 65 years in Europe, but is considerably lower in situ hybridisation (FISH) or by reverse transcriptase polymerase
countries with a younger population. The prevalence of CML is chain reaction (RT-PCR). These patients should be treated the
steadily rising due to the substantial prolongation of survival that same way as Philadelphia-positive (Phþ) patients. Therapeutic
has been achieved with targeted therapy [2]. CML in children is response is comparable [4]. In some patients with features of
rare; biology and treatment strategies in paediatric patients reveal CML, no Philadelphia chromosome or BCR–ABL1 rearrange-
specific aspects [3]. Therefore, these recommendations are pri- ment can be detected [1]. These patients are referred to as
marily intended for use in adult patients. Philadelphia-negative (Ph) and BCR–ABL1 negative, or as
atypical CML, according to the World Health Organization
(WHO) classification, and represent a separate disease entity [5].
Treatment of atypical CML is beyond the scope of these
Pathophysiology and diagnostic evaluation guidelines.
CML is a disease of haemopoietic stem cells, arising from a trans- BCR–ABL1 positive cells are genetically unstable and are prone
location t(9;22)(q34;q11), with the shortened chromosome 22, to develop multiple and heterogeneous genomic abnormalities,
designated as Philadelphia chromosome, 22q-. The translocation resulting in the transformation of the leukaemic phenotype from
leads to a juxtaposition of the ABL1 gene from chromosome 9 chronic to acute, hence leading to the progression from chronic
and the BCR gene from chromosome 22, resulting in a BCR– (CP) to accelerated (AP) and blast (BP) phases (Table 1) [5, 6].
ABL1 fusion gene that codes for BCR–ABL1 transcripts and fu- Table 1 compares the WHO and the European LeukemiaNet
sion proteins with high tyrosine kinase activity. The molecular (ELN) definitions of CML phases. Of note, the ELN definition
pathogenesis of CML is well understood, but the mechanism that has been used in almost all clinical trials assessing the efficacy of
leads to the gene translocation is unknown [1]. tyrosine kinase inhibitors (TKIs) and is recommended as a basis
Diagnosis of CML is generally straightforward. In most cases, for treatment decisions.
the diagnosis can be made on the basis of a characteristic blood Progression may be based on BCR–ABL1-dependent factors,
count and differential (excessive granulocytosis with typical left e.g. point mutations of the kinase domain, associated with resist-
shift of granulopoiesis). Confirmation of diagnosis is obtained ance to TKIs, or BCR–ABL1-independent factors, e.g. additional
by the identification of the Philadelphia chromosome, 22q- or cytogenetic aberrations causing clonal evolution.
VC The Author 2017. Published by Oxford University Press on behalf of the European Society for Medical Oncology.

by guest
on 05 February 2018
Table 1. Clinical and haematological criteria for the definition of AP and BP according to WHO [5] and ELN [6]
Accelerated phase Blast phase
WHO ELN WHO ELN
Spleen Persisting or increasing splenomegaly unresponsive to therapy – – –

9
WBC count Persisting or increasing WBC count (> 10 x 10 /L) unresponsive – – –
to therapy
Blast cellsa 10%–19% 15%–29% 20% 30%
a
Basophils > 20% > 20% – –
9
Platelet count > 1000 x 10 /L uncontrolled by therapy – – –
< 100 x 109/L unrelated to therapy Yes – –
CCA/Phþ Any new clonal aberration during therapy Present – –
Additional clonal chromosomal abnormalities in Ph cells at
diagnosis that include ‘major route’ abnormalities (second
Ph, trisomy 8, isochromosome 17q, trisomy 19), complex
karyotype or abnormalities of 3q26.2
Extramedullary involvementb – – Present Present
‘Provisional’ response- Haematological resistance to the first TKI (or failure to achieve a
to-TKI criteria complete haematological responsec to the first TKI) or
Any haematological, cytogenetic or molecular indications of
resistance to 2 sequential TKIs or
Occurrence of 2 or more mutations in BCR–ABL1 during TKI
therapy
The criteria of AP are different, reflecting the difficulty of making the diagnosis of this transitory phase. The criteria of BP differ only for the percent of blast
cells. Only one of the listed criteria is sufficient for the diagnosis of AP or BP.
a
In peripheral blood or in BM.
b
Excluding liver and spleen, including lymph nodes, skin, CNS, bone and lung.
c
Complete haematological response: WBC < 10 109/L; platelet count < 450 109/L, no immature granulocytes in the differential and spleen non-
palpable.
AP, accelerated phase; BM, bone marrow; BP, blast phase; CCA/Phþ, clonal chromosome abnormalities in Phþ cells; CNS, central nervous system; ELN,
European LeukemiaNet; Ph, Philadelphia; TKI, tyrosine kinase inhibitor; WBC, white blood cell; WHO, World Health Organization.
BM biopsies taken at diagnosis show increased cellularity due tests are taken for some unrelated reason. At diagnosis, most
to proliferation of the myelopoiesis in all stages of maturation (90%–95%) CML patients present in CP; initial BP is rare [1, 7].
with predominance of mature forms. Basophilia is common, and The designation of an AP at diagnosis is conflicting but the term
eosinophils may be prominent. Proportion of blasts vary; but, ac- should be used during therapy. Common signs and symptoms
cording to ELN recommendations [6], CP disease is associated of CML CP, when present, result from anaemia and spleno-
with < 15% blasts in blood and BM. Megakaryocytes are smaller megaly. These include fatigue, weight loss, malaise and left
than normal with hypolobulated nuclei. Moderate to marked re- upper quadrant fullness or pain. Rare manifestations include
ticulin fibrosis is encountered in 30% of cases. Pseudo-Gaucher bleeding (associated with a low platelet count and/or platelet
cells and sea-blue histiocytes are usually observed. BM compos- dysfunction), thrombosis (associated with thrombocytosis and/
ition undergoes rapid changes during therapy. These consist of or marked leukocytosis), gouty arthritis (from elevated uric
reduction of the granulocytic cellularity, normalisation of mega- acid levels), retinal haemorrhages and upper gastrointestinal ul-
karyopoiesis, regression of fibrosis, lymphocytosis and normal- ceration (from elevated histamine levels due to basophilia).
isation of erythropoiesis. Leukostatic symptoms (priapism, dyspnoea, drowsiness, loss of
The recognition of disease progression from CP to BP is rele- coordination, confusion) due to leukaemic cells sludging in the
vant for prognosis and treatment. However, the clinical and blood vessels are uncommon in CP despite white blood cell
morphological boundaries between these stages are sometimes (WBC) count often exceeding 100 109/L. Splenomegaly is the
vague. Immunocytology by flow cytometry and histochemistry most consistent physical sign detected in 40%–50% of cases.
allow accurate assessment of immature cells and distinction between Hepatomegaly is less common. Extramedullary infiltration
myeloid (70%–80%) and lymphoid (20%–30%) blast crisis [1]. (apart from spleen and liver) is rare. Headaches, bone pain,
About 50% of patients with CML diagnosed in Europe are arthralgias, pain from splenic infarction and fever are more fre-
asymptomatic. The disease is frequently diagnosed after blood quent with CML transformation [7].
iv42 | Hochhaus et al. Volume 28 | Supplement 4 | August 2017

by guest
on 05 February 2018
Table 2. Recommendations for diagnostic work-up, assessment of response and monitoring
Baseline To assess the response To monitor the response and the treatment
(diagnostic work-up)
Blood counts and Yes Every 15 days until a CHR without Every 3 months
differential significant cytopaenias has been achieved
BM, cytology Yes No No
BM, karyotype Yes At 3 and 6 months Then every 6 months until CCyR has been achieved
Blood, iFISH No No Only if cytogenetics of BM metaphases cannot

be analysed or is normal and molecular response
cannot be assessed
Blood, RT-PCR (qualitative) Yes No No
Blood, qRT-PCR No Every 3 months Every 4–6 weeks in first year after
(quantitative, BCR–ABL %) treatment discontinuation
Mutational analysis Only in AP or BP No Only in the case of failure
AP, accelerated phase; BM, bone marrow; BP, blast phase; CCyR, complete cytogenetic response; CHR, complete haematological response; iFISH, inter-
phase fluorescent in situ hybridisation; qRT-PCR: quantitative RT-PCR; RT-PCR, reverse transcriptase polymerase chain reaction.
On TKI therapy, most patients restore normal haematopoiesis. achieving a complete cytogenetic response (CCyR) after 18 months
Transient cytopaenias occur due to delayed recovery of normal of TKI therapy can be estimated with the European Treatment and
haematopoiesis but good efficacy against leukaemia. CML AP Outcome Study (EUTOS) score. The EUTOS Long-Term Survival
might present with non-specific symptoms, worsening anaemia, (ELTS) score for patients on TKI therapy considers CML-related
splenomegaly and organ infiltration; CML BP presents as an deaths only (Table 3) [III, A] [8]. Despite randomised first-line tri-
acute leukaemia with worsening constitutional symptoms, bleed- als employing different scores, the intrinsic risk of early acceleration
ing, fever and infections. or blast crisis in low-risk patients is low with all available TKIs.
Diagnosis must be confirmed by cytogenetics showing Major route cytogenetic aberrations (þ8, iso(17-
t(9;22)(q34;q11) and by multiplex RT-PCR showing BCR–ABL1 q), þ19, þ22q-), chromosome 3 aberrations and BM fibrosis at
transcripts. In rare cases, BCR–ABL1 juxtaposition can be deter- diagnosis have been associated with an unfavourable outcome
mined by interphase FISH (iFISH) of blood cells, using dual after imatinib therapy [9] and are considered warning signs.
colour dual fusion probes that allow the detection of BCR–
ABL1þ nuclei. Cytogenetic assessment is required because it is
necessary to detect additional chromosome abnormalities.
Qualitative multiplex RT-PCR is carried out on blood or BM First-line management of chronic phase
RNA. It identifies the transcript type, either typical e14a2 or e13a2 CML
(also known as b3a2 and b2a2) or atypical variants. Determination
of the transcript type is crucial for later monitoring, in particular for The three commercially available TKIs for the front-line treatment
the accurate assessment of molecular response. of CML are imatinib, dasatinib and nilotinib (Figure 1); options for
Quantitative RT-PCR (qRT-PCR) measuring BCR–ABL1 tran- first-line therapy in CML CP are imatinib 400–800 mg/day, nilotinib
scripts level as BCR–ABL1 % on the International Scale (IS) and 300 mg twice daily or dasatinib 100 mg/day. TKI selection should be
BCR–ABL1 mutation analysis are not required at baseline. Baseline based on treatment goals, age and comorbidities and should take
mutational analysis in patients with newly diagnosed CML CP is into consideration the adverse event (AE) profile of the available
not advised, as this has not been proven to provide information on drugs. With all three TKIs, overall survival (OS) after 5 years is 85%–
optimal treatment selection and to predict therapeutic outcome. 95% [I, A]. So far, no significant survival difference between imati-
Recommendations for the baseline diagnostic work-up are nib and second generation inhibitors has been observed.
summarised in Table 2 [V, A]. Imatinib mesylate was the first TKI to receive approval for the
treatment of patients with CML CP. It acts via competitive inhibition
at the adenosine triphosphate (ATP)-binding site of the BCR–ABL1
oncoprotein, which results in the inhibition of phosphorylation of
Staging and risk assessment proteins involved in cell signal transduction. It efficiently inhibits the
The relative risk of a patient with CML can be calculated using sim- BCR–ABL1 kinase, but, among others, also blocks the platelet-
ple clinical and haematological data provided that they were col- derived growth factor (PDGF) receptors and the KIT tyrosine kinase.
lected before any treatment. The Sokal score has been developed in The International Randomised Study of Interferon and STI571
the chemotherapy era and the Euro score in the interferon alpha (IRIS) study is considered a landmark clinical trial for CML treatment
(IFNa) era, with survival as the endpoint for both. The chance of with TKIs. A total of 1106 patients in CML CP were randomised to

by guest
on 05 February 2018
Table 3. Calculation of the relative risk of a patient with CML using clinical and haematological data obtained before any treatment [8]
Sokal EURO EUTOS ELTS
Age (years) 0.116 (age - 43.4) 0.666 when age > 50 N/A 0.0025 (age/10)3
Spleen sizea (cm) 0.345 (spleen - 7.51) 0.042 spleen 4 spleen 0.0615 spleen
Platelet count (109/L) 0.188 [(platelets/700)2 - 0.563] 1.0956 when platelets 1500 N/A 0.4104 (platelets/1000)0.5
Blood blast cells (%) 0.887 (blast cells - 2.10) 0.0584 blast cells N/A 0.1052 blast cells
Blood basophils (%) N/A 0.20399 when basophils > 3% 7 basophils
Blood eosinophils (%) N/A 0.0413 eosinophils N/A
Relative risk Exponential of the total Total 1000 Total Total
Low < 0.8 780 87 1.5680
Intermediate 0.8–1.2 781–1480 N/A 1.5680–2.2185
High 1.2 1480 87 2.2185
Endpoint Survival Survival CCyR CML-specific survival
a
Spleen size is measured by manual palpation and expressed as maximum distance perpendicular from costal margin.
CCyR, complete cytogenetic response; CML, chronic myeloid leukaemia; ELTS, EUTOS Long-Term Survival; EUTOS, European Treatment and Outcome
Study; N/A, not applicable.
receive imatinib 400 mg/day or IFNa plus low-dose cytarabine. After was a phase III randomised study comparing dasatinib 100 mg/
a median follow-up of 19 months, outcomes for patients receiving day to imatinib 400 mg/day in 519 newly diagnosed patients with
imatinib were significantly better than in those treated with IFNa plus CML. Patients assigned to dasatinib achieved confirmed CCyR at
cytarabine, notably the rates of CCyR (74% versus 9%, P < 0.001), 12 months more often than those on imatinib (77% versus 66%,
and freedom from progression to AP or BP at 12 months (99% versus P < 0.007). A 5-year follow-up showed that dasatinib induced
93%, P < 0.001) [10]. Responses to imatinib were also durable: in a more rapid and deeper responses at early time points compared
10-year follow-up of the IRIS study, the estimated event-free survival with imatinib. At 3 months, a higher proportion of patients
rate was 79.6%, and the OS rate was 83.3% [11]. treated with dasatinib achieved BCR–ABL1 transcripts <10% on
the IS (84% versus 64%, P < 0.0001). Meeting this threshold in
High-dose imatinib and combination with IFNa either arm predicted for better progression-free survival (PFS)
and OS. Transformations to CML AP or CML BP were fewer in
Other strategies for front-line therapy include using higher doses patients treated with dasatinib versus imatinib at 5 years (4.6%
of imatinib or combining a TKI with an additional agent, such as versus 7.3%) [18, 19].
IFNa. In the German CML IV study, patients with tolerance- and Nilotinib is a structural analogue of imatinib. Compared with
response-adapted high-dose imatinib achieved deep molecular imatinib, the in vitro affinity for the BCR-ABL1 ATP-binding
remission (DMR) more quickly than patients on standard-dose
site is 30- to 50-fold higher. In the ENESTnd study, two doses of
imatinib [12, 13]. A recent meta-analysis revealed an advantage
nilotinib (300 or 400 mg twice daily) were compared with imati-
of high-dose imatinib with regard to achievement of major mo-
nib 400 mg/day. The primary endpoint, MMR rate at 12 months,
lecular response (MMR) at 12 months of therapy [14].
was achieved at higher rates for both doses of nilotinib compared
IFNa has re-emerged as an interesting therapeutic option in
with imatinib (44% and 43% versus 22%, P < 0.001). The cumu-
CML with the advent of PEGylated formulations (with polyethyl-
lative incidence of CCyR by 24 months was 87% and 85% with
ene glycol PEG) requiring less frequent administration, improved
nilotinib 300 mg twice daily, and 400 mg twice daily, respect-
efficacy and tolerability. In the French SPIRIT trial, patients were
randomised to receive imatinib 400 or 600 mg/day, imatinib ively, and 77% with imatinib 400 mg/day (P < 0.001). By 5 years,
400 mg/day plus PEG-IFNa-2a, or imatinib 400 mg/day plus the cumulative incidences of MMR by 60 months were 77%,
subcutaneous cytarabine. At 12 months, rates of CCyR were 77% and 60%, respectively (P < 0.0001). The incidences of DMR
similar among the 4 groups. The imatinib plus PEG-IFNa-2a- with BCR–ABL1 transcripts [IS] 0.0032% (equivalent to a 4.5
treated group obtained higher rates of MMR and DMR [15]. log reduction) by 72 months were 56%, 55% and 33%, respect-
PEGylated IFNa in combination with dasatinib appeared to ively (P < 0.0001). The incidences of transformation to AP or BP
improve molecular response rates in a single-armed phase II study were 3.9%, 2.1%, and 7.4%, respectively (P ¼ 0.06 and 0.003, re-
with historical controls [16]. IFNa maintenance after TKI therapy spectively). The estimated 5-year survival rates were 94%, 96%,
may help to bridge to treatment-free remission (TFR) [17]. In all, and 92%, respectively. While nilotinib was superior to imatinib
despite lack of registration, PEGylated IFNa is promising as an across all Sokal score categories in inducing higher rates of CCyR
agent to increase the proportion of patients that may discontinue and MMR, the advantage in reducing the rates of transformation
(see Table 4), but must still be considered investigational. was more pronounced in patients with intermediate- and high-
Dasatinib is an oral, second generation multikinase TKI that is Sokal-risk CML. The rates of transformations were 1%, 1% or
350 times more potent than imatinib in vitro and inhibits mul- 0% in Sokal low-risk patients treated with nilotinib 300 mg twice
tiple kinases including Src-family members. The DASISION trial daily, 400 mg twice daily or imatinib 400 mg/day. The rates were

by guest
on 05 February 2018
Table 4. Minimum requirements for treatment-free remission [40]
Institutional requirements for safe supervision of TFR:
1. Availability of high quality internationally standardised, accurate,
Any line, T315l mutation
sensitive qRT-PCR laboratory

2. Rapid turn-around of PCR test results—within 4 weeks
Ponatinib
3. Capacity to provide PCR tests every 4–6 weeks, when required
4. Structured follow-up established to enable rapid intervention
if BCR-ABL1 is rising
Green criteria, which support treatment-free remission:

1. Institutional criteria met
2. Sokal-Score at diagnosis non-high
intolerance to 2nd line TKIs
3. Typical b2a2- or b3a2-BCR–ABL1 transcripts, or atypical transcripts

which can be quantified over a 4.5 log dynamic range
Failure of and/or
remaining TKIs
Any of the
4. Chronic phase disease
3rd line
5. Optimal response to first-line therapy

6. Duration of TKI therapy > 5 years
SCT consideration In many cases 7. MR4.5 reached
8. Duration of deep molecular response (MR4 or MR4.5) (standardised
lab) > 2 years
MR, molecular response; PCR, polymerase chain reaction; qRT-PCR,

quantitative reverse transcriptase PCR; TFR, treatment-free remission;
nilotinib or dasatinib
Failure of 1st line
TKI, tyrosine kinase inhibitor.

dasatinib,
bosutinib,
ponatinib
Nilotinib,
Republished from [40] with permission of the American Society of

Hematology; permission conveyed through Copyright Clearance
Center, Inc.
2%, 1% or 10% among patients with intermediate Sokal risk and

9%, 5% or 11% among patients with high Sokal risk [20, 21].
CML, chronic myeloid leukaemia; SCT, stem cell transplantation; TKI, tyrosine kinase inhibitor.
Failure of 1st
line imatinib
Selecting first-line therapy

dasatinib,
bosutinib,
ponatinib
Nilotinib,
2nd line
Therapy goals should be discussed with the patient and defined be-
fore the selection of the first-line drug. With all three TKIs licensed
for first-line therapy, survival chances are similar [I, A]. However,
the chance to achieve DMR with an option to discontinue therapy
is higher with dasatinib and nilotinib as compared with imatinib
[V, C]. This may be particularly relevant for young female patients
with a wish to become pregnant and for all patients with a long life
Figure 1. Treatment options for chronic phase CML patients.
expectancy. Risk of transformation to AP and BP is lower in Sokal

Intolerance
dasatinib,
bosutinib
to 1st TKI
Imatinib,
nilotinib,
non-low-risk patients using dasatinib or nilotinib [I, A]. The use of

generic imatinib may be considered to reduce cost of therapy sub-
stantially, but also for its safety profile, particularly in elderly pa-
tients [22]. In some countries imatinib may be mandatory for first-
SCT investigation warranted
line use on cost-effectiveness/reimbursement grounds. However,

many different forms of generic imatinib are being commercialised
worldwide and precise information on the tolerance and efficacy of
each of these different compounds are rare. Recent results from a
randomised study (ClinicalTrials.gov Identifier: NCT02130557)
suggest the alternative use of bosutinib 400 mg/day as first-line ther-
dasatinib
Imatinib,
nilotinib,
1st line
apy for CP CML patients, but registration is pending.

Comorbidities are the major cause of death in CML patients
and may be aggravated by AEs [23]. Therefore, patient’s age,
comorbidities and the specific TKI toxicity profile should be con-
sidered [V, B]. For patients at risk of developing pleural effusions
(existing lung disorders or uncontrolled hypertension), dasatinib
should be avoided. Pulmonary arterial hypertension (PAH) is a
rare complication of dasatinib, and patients with pre-existing

by guest
on 05 February 2018
PAH may be considered for alternative TKIs in the front-line set- Table 5. Assessment of response
ting. Dasatinib also inhibits platelet function, and patients taking
CHR (Complete haematological response)
concomitant anticoagulants may be at an increased risk of haem-
WBC count < 10109/L
orrhagic complications [V, B].
No immature granulocytes
Nilotinib has been associated with hyperglycaemia; caution
Basophils < 5%
should be exercised in patients with uncontrolled diabetes melli- Platelet count < 450109/L
tus (DM) when initiating therapy. Patients should take nilotinib
Spleen non-palpable
on an empty stomach to avoid excess drug exposure with fat-
containing food. Nilotinib has also been associated with vaso- Cytogenetic response (CyR)
spastic and vaso-occlusive vascular events, such as ischaemic Complete CyR No Phþ metaphases by CBA, or < 1%
heart disease, ischaemic cerebrovascular events and peripheral ar- BCR–ABLþ nuclei by iFISH out of 200 cells
tery occlusive disease [I, C]. Nilotinib use should be prescribed Partial CyR 1%–35% Phþ metaphases by CBA
with caution in patients with risk factors such as DM or coronary, Minor CyR 36%–65% Phþ metaphases by CBA
cerebrovascular or peripheral arterial disease. A thorough inter- Minimal CyR 66%–95% Phþ metaphases by CBA
vention against cardiovascular risk factors such as smoking, No CyR > 95% Phþ metaphases by CBA
hyperlipidaemia, hypertension and DM is warranted [V, A].
Imatinib causes persisting but mostly mild to moderate side-ef- Molecular response (MR)
fects with significant impact on quality of life (QoL) including Major MR (MMR) BCR–ABL transcript level 0.1% on the
weight gain, fatigue, peripheral and periorbital oedema bone and International Scale
Deep MR:
muscle aches, nausea and others. All available TKIs may prolong the
MR4 BCR–ABL transcript level 0.01% on the
QT interval; thus, potassium and magnesium should be repleted to
International Scale or
appropriate serum levels before starting therapy [V, B].
BCR–ABL not detectable with at least
Hydroxyurea (40 mg/kg body weight/day) may be used as initial
10 000 ABL or 24 000 GUS transcripts
therapy before confirmation of the BCR–ABL1 fusion and imme- MR4.5 BCR–ABL transcript level 0.0032% on the
diate need for therapy because of high leukocyte counts or clinical International Scale or
symptoms. TKI therapy should be commenced immediately after BCR–ABL not detectable with at least
confirmation of BCR–ABL1 positivity. It is recommended to taper 32 000 ABL or 77 000 GUS transcripts
the hydroxyurea dose before its discontinuation.
To avoid tumour lysis syndrome, 2.5–3 L fluid intake is recom- CBA, chromosome banding analysis; iFISH, interphase fluorescent in
mended per day considering the individual cardiac and/or renal situ- situ hybridisation; Ph, Philadelphia; WBC, white blood cell.
ation. Sodium bicarbonate may be used to set the urine pH to 6.4–
6.8 for optimal uric acid clearance. Allopurinol may increase the risk
of xanthine accumulation with renal failure and should therefore be
restricted to patients with symptomatic hyperuricaemia.
The recommendations on follow-up therapy are based on the as-
Molecular monitoring
sessment of the response (Table 5) and on the definition of the re- A quantification of BCR–ABL1 mRNA, performing qRT-PCR
sponse (Table 6). The response to TKIs can be classified as optimal, from 10 to 20 mL ethylenediaminetetraacetic acid (EDTA)-anti-
meaning that continuing treatment survival is predicted to be nor- coagulated peripheral blood, is required every 3 months. This
mal or close to normal; and failure, meaning that treatment must method represents the most sensitive tool for the assessment of
be switched to an alternative TKI, or allogeneic stem cell trans- the disease status, particularly of measurable residual disease.
plantation (alloSCT) should be considered. Between optimal and BCR–ABL1 transcript levels should be expressed according to the
failure, there is a grey zone that is defined as ‘warning’, meaning IS (BCR–ABL1IS %) to guarantee comparability of results among
that the response must be monitored more carefully and that the laboratories. Therefore, local methods require thorough optimi-
patient may be eligible for potentially better treatments [V, A] [6]. sation and harmonisation with reference laboratories [24–26].
The choice of treatment, particularly the decision of moving from Early molecular response at 3 months (BCR–ABL1IS 10%) pre-
one treatment to another, strongly depends on the response to treat- dicts survival and chance of eventually achieving DMR [27].
ment, i.e. on the degree of the cytogenetic response, molecular re- Intervals can be prolonged from 3 to 6 months after repeated
sponse and on the detection of BCR–ABL1 kinase domain mutations. achievement of an MMR (BCR–ABL1IS 0.1%, 3 log reduction
from standardised baseline) or reduced to 4–6 weeks after treat-
ment discontinuation. Significant rises of BCR–ABL1 transcript
Cytogenetic monitoring levels (fivefold accompanied by loss of MMR) during long-term
Cytogenetic monitoring must be carried out by analysis of marrow therapy are early indicators for treatment failure or non-
cell metaphases, reporting the proportion of Phþ metaphases of at adherence. The achievement of DMR (MR4, MR4.5, MR5, i.e. 4–5
least 20 metaphases analysed. The cytogenetic response is defined log reduction) during TKI treatment is prerequisite for therapy
as complete (CCyR) with 0% Phþ metaphases, partial (PCyR) interruptions within controlled trials [III, A] [28].
with 1%–35% Phþ metaphases, minor with 36%–65% More than 100 different kinase domain mutations of BCR–
Phþ metaphases, minimal with 66%–95% Phþ metaphases and ABL1 that impair TKI binding have been reported in patients who
none if > 95% of metaphases are still Phþ. iFISH data cannot be develop TKI resistance. In the case of mutations, second-line ther-
used to calculate the cytogenetic response categories. apy should be selected according to the sensitivity of the individual

by guest
on 05 February 2018
Table 6. Definition of the response to TKI therapy (any line) Second and third generation TKIs
Milestones Failure Warning Optimal Before their approval to treat first-line CML CP, both nilotinib
response and dasatinib were approved for use in second-line CML CP fol-
lowing prior therapy including imatinib. In Europe, bosutinib is
Diagnosis Score: High-risk approved for CML patients previously treated with one or more
Additional TKIs and for whom imatinib, nilotinib and dasatinib are not con-
chromosomal sidered appropriate treatment options.
aberrations ‘major
Second-line treatment with nilotinib, dasatinib or bosutinib
route’ in
can yield high rates of response in patients who have inad-
Phþ metaphases
equate response to imatinib. Dose escalation of imatinib can
3 months No CHR Ph 36%–95% Ph 35% improve response rates in patients with inadequate response to
Ph > 95% BCR–ABL >10% BCR–ABL < 10% standard-dose imatinib, but switching to second-line TKIs can
6 months Ph > 35% Ph 1%–65% Ph 0% be more effective. Several studies have demonstrated signifi-
BCR–ABL > 10% BCR–ABL 1%–10% BCR–ABL < 1% cantly higher rates of complete haematological response
(CHR), CCyR and MMR with the newer TKIs than with high-
12 months Ph 1% BCR–ABL
dose imatinib. Moreover, PFS in these studies was better with
< 0.1%
the newer TKIs than with high-dose imatinib. An earlier switch
BCR–ABL > 1% BCR–ABL 0.1%–1%
to second-line TKI may be more effective than a later switch
> 18 months BCR–ABL 0.1%–1% BCR–ABL [30–32].
< 0.01%* Bosutinib was initially studied in patients who had resistance
Anytime Relapse, loss or intolerance to imatinib. After a dose escalation period, 500 mg
of MMR once daily was selected as the phase II dose. A total of 288 patients
were enrolled in the pivotal phase II trial; more than two-thirds
These definitions are a provisional adaptation of the original ELN defin- had imatinib-resistant disease. The primary endpoint of major
itions [6]. Operationally, ‘optimal’ means to continue the treatment, ‘fail- cytogenetic response (MCyR) at 6 months was achieved in 31%;
ure’ to change the treatment, and ‘warning’ to monitor more carefully 41% achieved a CCyR. The most common toxicities were diar-
and to be ready to consider a change of treatment. rhoea, nausea, vomiting and rash. Diarrhoea occurred in 84% of
*For patients with the aim to achieve treatment-free remission. patients, with 9% experiencing grade 3 diarrhoea. Other notable
CHR, complete haematological response; ELN, European LeukemiaNet; MMR, AEs included mild myelosuppression and liver function test
major molecular response; Ph, Philadelphia; TKI, tyrosine kinase inhibitor. abnormalities [32].
Modified from [6] with permission of the American Society of Hematology; Ponatinib is a third generation TKI, and the first TKI in class to
permission conveyed through Copyright Clearance Center, Inc. exhibit activity against CML with T315I mutation. It is 500 times
more potent than imatinib at inhibiting BCR–ABL1. After resist-
ance or intolerance to dasatinib or nilotinib, or in the case of the
mutation. In particular, V299L, T315A and F317L/V/I/C are resist- T315I mutation, 56% of CP CML patients achieved a MCyR by
ant to dasatinib. Y253H, E255K/V and F359V/C/I are resistant to 12 months on ponatinib 45 mg/day. Ponatinib should be con-
nilotinib, and V299L to bosutinib. T315I is resistant to all TKIs ex- sidered the agent of choice in patients with CML and T315I muta-
cept ponatinib. Recommendations on the use of mutational ana- tion, and in instances where other TKIs are not indicated. No
lysis have recently been presented by the ELN [V, A]: other commercially available TKIs have activity against this
• During first-line therapy, analysis is due in the case of failure
mutated BCR-ABL1. The risk of serious toxicities (vaso-occlusive
and in the case of an increase in BCR–ABL1 transcript levels disease, pancreatitis, hypertension, severe skin rashes) and of
leading to a loss of MMR. thrombotic events with ponatinib is significant, but the benefits
• During second-line therapy, analysis is due in the case of haem-
outweigh the risks for patients with a T315I mutation. The inci-
atological or cytogenetic failure or in the case of pre-existing dence of these side-effects is lower with ponatinib 15–30 mg daily
mutations. [33, 34].
• In the case of AP or BP, mutational analysis is always due.
In any case, the mutation result should be accompanied by an

estimation of the size of the mutated clone to confirm the associ- How to select a second- or third-line option
ation of refractoriness with this specific mutation [29]. At the time of treatment failure, patients should undergo BM
examination to allow proper determination of the CML phase
and documentation of any clonal evolution. All patients should
have CML cells tested for BCR–ABL1 mutational profile, as this
Management of resistant and refractory will help guide the selection of the TKI.
For second-line treatment, the choice can be guided by the type
disease of AEs which caused the switch, by the side-effect profiles of the dif-
Before defining a patient as having TKI resistance and modifying ferent TKIs, mutational profiles, drug interactions, compliance
therapy, treatment compliance and drug–drug interactions issues and the patient’s pre-existing medical conditions. Mutational
should be assessed. analysis is required in patients who are failing imatinib or second

by guest
on 05 February 2018
Diagnosis
• In most cases, the diagnosis can be made on the basis of a characteristic blood count and differential (excessive granulocytosis with typical left shift of
granulopoiesis). Confirmation of diagnosis is obtained by the identification of the Philadelphia chromosome, 22q-, or BCR–ABL1 transcripts, or both, in
peripheral blood or BM cells.
• The recognition of disease progression from CP to BP is relevant for prognosis and treatment. Immunocytology by flow cytometry and histochemistry
allow accurate assessment of immature cells and distinction between myeloid and lymphoid blast crisis.
• Diagnosis must be confirmed by cytogenetics showing t(9;22)(q34;q11), and by multiplex RT-PCR showing BCR–ABL1 transcripts. In rare cases, BCR–ABL1
juxtaposition can be determined by iFISH of blood cells, using dual colour dual fusion probes that allow the detection of BCR–ABL1þ nuclei. Cytogenetic
assessment is required because it is necessary to detect additional chromosome abnormalities.
• Qualitative multiplex RT-PCR is carried out on blood or BM RNA. It identifies the transcript type, either typical or atypical variants. Determination of the tran-
script type is crucial for later monitoring, in particular for the accurate assessment of molecular response.
• Baseline mutational analysis in patients with newly diagnosed CML CP is not advised.
• Major route cytogenetic aberrations (þ8, iso(17q), þ19, þ22q-), chromosome 3 aberrations and BM fibrosis at diagnosis are considered warning signs.
First-line management of chronic phase CML
• Options for first-line therapy in CML CP are imatinib 400–800 mg/day, nilotinib 300 mg twice daily or dasatinib 100 mg/day. TKI selection should be based
on treatment goals, age and comorbidities and should take into consideration the AE profile of the available drugs.
• Other strategies for front-line therapy include using higher doses of imatinib or combining a TKI with an additional agent, such as IFNa.
• In individual patients, therapy goals should be discussed with the patient and defined before the selection of the first-line drug. With all three TKIs licensed
for first-line therapy, survival chances are similar [I, A].
• Risk of transformation to AP and BP is lower in Sokal non-low risk patients using dasatinib or nilotinib [I, A].
• Patient’s age, comorbidities and the specific TKI toxicity profile should be considered [V, B].
• For patients at risk of developing pleural effusions (existing lung disorders or uncontrolled hypertension), dasatinib should be avoided. PAH is a rare com-
plication of dasatinib, and patients with pre-existing PAH may be considered for alternative TKIs in the front-line setting. Dasatinib also inhibits platelet
function, and patients taking concomitant anticoagulants may be at an increased risk of haemorrhagic complications [V, B].
• Patients should take nilotinib on an empty stomach to avoid excess drug exposure with fat-containing food. Nilotinib has also been associated with vaso-
spastic and vaso-occlusive vascular events, such as ischaemic heart disease, ischaemic cerebrovascular events and PAOD [I, C]. Nilotinib use should be pre-
scribed with caution in patients with risk factors such as DM or coronary, cerebrovascular or peripheral arterial disease. A thorough intervention against
cardiovascular risk factors such as smoking, hyperlipidaemia, hypertension and DM is warranted [V, A].
• All available TKIs may prolong the QT interval; thus, potassium and magnesium should be repleted to appropriate serum levels before starting therapy [V, B].
• TKI therapy should be commenced immediately after confirmation of BCR–ABL1 positivity. It is recommended to taper the hydroxyurea dose before its
discontinuation.
• To avoid tumour lysis syndrome, 2.5–3 L fluid intake is recommended per day considering the individual cardiac and/or renal situation. Sodium bicarbonate
may be used to set the urine pH to 6.4–6.8 for optimal uric acid clearance. Allopurinol may increase the risk of xanthine accumulation with renal failure
and should therefore be restricted to patients with symptomatic hyperuricemia.
• If the response to TKIs is a failure, the treatment must be switched to an alternative TKI, or alloSCT should be considered. Between optimal and failure, there
is a grey zone that is defined as ‘warning’, meaning that the response must be monitored more carefully and that the patient may be eligible for poten-
tially better treatments [V, A].
• Cytogenetic monitoring must be carried out by analysis of marrow cell metaphases, reporting the proportion of Phþ metaphases of at least 20 meta-
phases analysed. iFISH data cannot be used to calculate the cytogenetic response categories.
• A quantification of BCR–ABL1 mRNA, performing qRT-PCR from 10 to 20 mL EDTA-anticoagulated peripheral blood, is required every 3 months.
• The achievement of DMR (MR4, MR4.5, MR5, i.e. 4–5 log reduction) during TKI treatment is prerequisite for therapy interruptions within controlled trials [III, A].
• During first-line therapy, mutation analysis is due in the case of failure and in the case of an increase in BCR–ABL1 transcript levels leading to a loss of MMR
[V, A]. During second-line therapy, analysis is due in the case of haematological or cytogenetic failure or in the case of pre-existing mutations [V, A]. In the
case of AP or BP, mutational analysis is always due [V, A].
Management of resistant and refractory disease
• At the time of treatment failure, patients should undergo BM examination to allow proper determination of the CML phase and documentation of any clo-
nal evolution. All patients should have CML cells tested for BCR–ABL1 mutational profile, as this will help guide the selection of the TKI.
• Mutational analysis is required in patients who are failing imatinib or second generation TKIs, or those who progress to AP/BP [V, A].
• Options are imatinib, nilotinib, dasatinib, bosutinib or ponatinib [V, A]. Ponatinib should be considered the agent of choice in patients with CML and T315I
mutation, and in instances where other TKIs are not indicated.
• AlloSCT remains an important therapeutic option for patients in CML CP who fail at least 2 TKIs or are potentially harbouring the T315I mutation (after a trial
of ponatinib therapy) [V, B]. Patients with a high risk for transformation should be considered for alloSCT, since outcome of alloSCT after transformation is
unfavourable.
• The only curative option for patients in BP disease is alloSCT. AlloSCT should also be considered early in patients developing AP during TKI treatment or
high-risk patients with insufficient treatment response [V, B].
• AlloSCT for advanced disease with a high transplant risk should not be advocated; ongoing drug treatment or best supportive care might be the better option.

by guest
on 05 February 2018
Table 7. Continued
Management of AEs, quality of life
• Treatment discontinuation may be considered in individual patients, if proper, high-quality and certified monitoring can be ensured. Prerequisites for safe
stopping are institutional requirements for safe supervision, identification of typical BCR–ABL1 transcripts at diagnosis, at least 5 years of TKI therapy,
achievement of MR4.5 and a stability of DMR (at least MR4) for at least 2 years [III, B].
AE, adverse event; alloSCT, allogeneic stem cell transplantation; AP, accelerated phase; BM, bone marrow; BP, blast phase; CML, chronic myeloid leukae-
mia; CP, chronic phase; DM, diabetes mellitus; DMR, deep molecular remission; EDTA, ethylenediaminetetraacetic acid; iFISH, interphase fluorescent in
situ hybridisation, IFNa, interferon alpha; MMR, major molecular response; MR4.5, 4.5-log reduction; PAH, pulmonary arterial hypertension; PAOD, periph-
eral artery occlusive disease; Ph, Philadelphia; qRT-PCR, quantitative RT-PCR; RT-PCR, reverse transcriptase polymerase chain reaction; TKI, tyrosine kinase
inhibitor.
generation TKIs, or those who progress to AP/BP [V, A]. Options tolerable, but affect the QoL and are a cause of decreased compli-
are imatinib, nilotinib, dasatinib, bosutinib or ponatinib [V, A]. ance. The third category includes late, so-called ‘off-target’ com-
The optimisation of dosages of bosutinib (400 mg/day) and ponati- plications, which can affect the cardiovascular system, the
nib (15–30 mg/day; 45 mg/day in advanced disease or in the case of respiratory system, liver, pancreas, the immune system, second
problematic mutations) is still under investigation. malignancies, glucose and lipid metabolism, etc.
All TKIs can be cardiotoxic and should be used with caution in
Allogeneic stem cell transplantation patients with heart failure. Nilotinib has been reported to be asso-
ciated in particular with arterial pathology, both peripheral and
The number of patients undergoing alloSCT for CML CP has coronary. Dasatinib has been reported to be associated in particu-
decreased significantly since TKIs were introduced. AlloSCT re- lar with pleura and lung complications. Because these complica-
mains an important therapeutic option for patients in CML CP tions are a potential cause of morbidity and mortality, continued
who fail at least 2 TKIs or are potentially harbouring the T315I mu- clinical monitoring of all patients is required [37].
tation (after a trial of ponatinib therapy) [V, B]. Patients at high risk
for transformation should be considered for alloSCT, since out-
come of alloSCT after transformation is unfavourable. Patients
referred to transplant may have a better outcome if entering the Treatment-free remission
transplant with a better response (lower CML burden). Assessment
of donor availability will be prerequisite to achieve this goal. TKI discontinuation studies in patients with durable DMR dem-
onstrate that stopping TKI therapy is feasible. The Stop Imatinib
(STIM) trial investigated the risk of relapse in patients on imati-
AlloSCT in advanced stage CML nib with ongoing complete molecular response for longer than
The only curative option for patients in BP disease is alloSCT. 2 years who then stopped treatment. In the most recent update,
AlloSCT should also be considered early in patients developing 100 patients had a median follow-up of 50 months and were
AP during TKI treatment or high-risk patients with insufficient monitored closely for evidence of molecular relapse. Overall,
treatment response [V, B]. TKI monotherapy or in combination 61% experienced a molecular relapse, with 95% of the events
with chemotherapy may serve as a good option for those who are occurring within 7 months of stopping imatinib [38]. With nilo-
not candidates for transplant, or as a bridge and debulking option tinib, at 48 weeks after stopping, 98 patients (51.6%) remained in
before alloSCT [35]. AlloSCT for advanced disease with a high MMR or better (primary endpoint) [39].
transplant risk should not be advocated; ongoing drug treatment The economic impact of long-term discontinuation of TKI
or best supportive care might be the better option and will also may be substantial. Ongoing studies will help to guide physicians
save costs [36]. in determining when it is safe and most promising to stop TKI
therapy in CML patients.
Treatment discontinuation may be considered in individual pa-
tients, if proper, high-quality and certified monitoring can be
Management of AEs, QoL ensured. Prerequisites for safe stopping are institutional require-
The TKIs are associated with different patterns of side-effects, ments for safe supervision, identification of typical BCR–ABL1 tran-
and this should be considered for treatment decisions. Side- scripts at diagnosis, at least 5 years of TKI therapy, achievement of
effects can be divided into three general categories. The first cat- MR4.5 (4.5-log reduction) and a stability of DMR (at least MR4) for
egory includes major, grade 3 to 4 side-effects that typically occur at least 2 years [III, B]. Less stringent criteria do not exclude success-
during the initial phase of treatment, are manageable, but require ful TFR, but stability of TFR is improved with longer TKI therapy
temporary treatment discontinuation and dose reduction, and and longer DMR. Informed consent should include the information
may lead to treatment discontinuation in 10% of patients. The on the estimated risk of recurrence of the disease and the need for
second category includes minor, grade 1/2, side-effects that start frequent molecular monitoring, monthly during the first half-year,
early during treatment but persist. They are also manageable and 6-weekly during the second half-year and 3-monthly later on [40,

by guest
on 05 February 2018
Grading Systema)
Levels of evidence
II Small randomised trials or large randomised trials with a suspicion of bias (lower methodological quality) or meta-analyses of such trials or
of trials demonstrated heterogeneity
C Insufficient evidence for efficacy or benefit does not outweigh the risk or the disadvantages (adverse events, costs, ... ), optional
a
41]. It should be noted that a TKI withdrawal syndrome consisting References

of musculoskeletal pain, in most cases transient, has been observed
1. Hehlmann R, Hochhaus A, Baccarani M; European LeukemiaNet.
in several cessation studies in up to 30% of the patients [42]. Chronic myeloid leukaemia. Lancet 2007; 370: 342–350.
2. Bower H, Björkholm M, Dickman PW et al. Life expectancy of patients
with chronic myeloid leukemia approaches the life expectancy of the gen-
eral population. J Clin Oncol 2016; 34: 2851–2857.
3. Hijiya N, Schultz KR, Metzler M et al. Pediatric chronic myeloid leuke-
Methodology mia is a unique disease that requires a different approach. Blood 2016;
These Clinical Practice Guidelines were developed in accordance 127: 392–399.
with the ESMO standard operating procedures for Clinical Practice 4. Hochhaus A, Mahon FX, le Coutre P et al. Nilotinib first-line therapy in
patients with Philadelphia chromosome-negative/BCR-ABL-positive
Guidelines development, http://www.esmo.org/Guidelines/ESMO-
chronic myeloid leukemia in chronic phase: ENEST1st sub-analysis.
Guidelines-Methodology. The relevant literature has been selected J Cancer Res Clin Oncol 2017; 143: 1225–1233.
by the expert authors. A summary of recommendations is shown in 5. Arber DA, Orazi A, Hasserjian R et al. The 2016 revision to the World
Table 7. Levels of evidence and grades of recommendation have Health Organization classification of myeloid neoplasms and acute leu-
been applied using the system shown in Table 8. Statements without kemia. Blood 2016; 127: 2391–2405.
grading were considered justified standard clinical practice by the 6. Baccarani M, Deininger MW, Rosti G et al. European LeukemiaNet rec-
experts and the ESMO Faculty. This manuscript has been subjected ommendations for the management of chronic myeloid leukemia: 2013.
Blood 2013; 122: 872–884.
to an anonymous peer review process.
7. Hoffmann VS, Baccarani M, Hasford J et al. Treatment and outcome of
2904 CML patients from the EUTOS population-based registry.
Leukemia 2017; 31: 593–601.
8. Pfirrmann M, Baccarani M, Saussele S et al. Prognosis of long-term sur-
vival considering disease-specific death in patients with chronic myeloid
Disclosure leukemia. Leukemia 2016; 30: 48–56.
AH has reported research support and honoraria from Novartis, 9. Fabarius A, Leitner A, Hochhaus A et al. Impact of additional cytogenetic
Bristol-Myers Squibb, Pfizer and Ariad; SS received honoraria aberrations at diagnosis on prognosis of CML: long-term observation of
from Novartis, Bristol-Myers Squibb, Pfizer and Ariad and re- 1151 patients from the randomized CML Study IV. Blood 2011; 118:
6760–6768.
search support from Novartis and Bristol-Myers Squibb; GR, F-
10. O’Brien SG, Guilhot F, Larson RA et al. Imatinib compared with inter-
XM, and JR have received research support and/or honoraria feron and low-dose cytarabine for newly diagnosed chronic-phase
from Novartis, Bristol-Myers Squibb, Pfizer and Ariad; JJWMJ chronic myeloid leukemia. N Engl J Med 2003; 348: 994–1004.
has received research support from Novartis and Bristol-Myers 11. Hochhaus A, Larson RA, Guilhot F et al. Long-term outcomes of imati-
Squibb and honoraria from Novartis, Bristol-Myers Squibb, nib treatment for chronic myeloid leukemia. N Engl J Med 2017; 376:
Pfizer and Incyte; HHH has received honoraria from Ariad, 917–927.
Janssen, Bristol-Myers Squibb and Pfizer, and he is chairman of 12. Hehlmann R, Müller MC, Lauseker M et al. Deep molecular response is
reached by the majority of patients treated with imatinib, predicts survival,
Nordic CML Study Group which has received research funding
and is achieved more quickly by optimized high-dose imatinib: results from
from Bristol-Myers Squibb, Novartis, Merck, Pfizer and the the randomized CML-Study IV. J Clin Oncol 2014; 32: 415–423.
Nordic Cancer Union; CB has reported research funds from 13. Kalmanti L, Saussele S, Lauseker M et al. Safety and efficacy of imatinib
Roche and Janssen and is a member of the speaker’s bureau of in CML over a period of 10 years: data from the randomized CML-study
Roche, Janssen and Pfizer. IV. Leukemia 2015; 29: 1123–1132.

by guest
on 05 February 2018
14. Hoffmann VS, Hasford J, Deininger M et al. Systematic review and meta- 29. Soverini S, Hochhaus A, Nicolini FE et al. BCR-ABL kinase domain mu-
analysis of standard dose imatinib vs. high dose imatinib and 2nd gener- tation analysis in chronic myeloid leukemia patients treated with tyrosine
ation tyrosine kinase inhibitors for chronic myeloid leukemia. J Cancer kinase inhibitors: recommendations from an expert panel on behalf of
Res Clin Oncol 2017; 143: 1311–1318. European LeukemiaNet. Blood 2011; 118: 1208–1215.
15. Preudhomme C, Guilhot J, Nicolini FE et al. Imatinib plus peginterferon 30. Shah NP, Rousselot P, Schiffer C et al. Dasatinib in imatinib-resistant or
alfa-2a in chronic myeloid leukemia. N Engl J Med 2010; 363: -intolerant chronic-phase, chronic myeloid leukemia patients: 7-year
2511–2521. follow-up of study CA180-034. Am J Hematol 2016; 91: 869–874.
16. Hjorth-Hansen H, Stentoft J, Richter J et al. Safety and efficacy of the 31. Giles FJ, le Coutre PD, Pinilla-Ibarz J et al. Nilotinib in imatinib-resistant
combination of pegylated interferon-a2b and dasatinib in newly diag- or imatinib-intolerant patients with chronic myeloid leukemia in chronic
nosed chronic-phase chronic myeloid leukemia patients. Leukemia 2016; phase: 48-month follow-up results of a phase II study. Leukemia 2013;
30: 1853–1860. 27: 107–112.
17. Burchert A, Saussele S, Eigendorff E et al. Interferon alpha 2 maintenance 32. Cortes JE, Khoury HJ, Kantarjian HM et al. Long-term bosutinib for
therapy may enable high rates of treatment discontinuation in chronic chronic phase chronic myeloid leukemia after failure of imatinib plus
myeloid leukemia. Leukemia 2015; 29: 1331–1335. dasatinib and/or nilotinib. Am J Hematol 2016; 91: 1206–1214.
18. Kantarjian H, Shah NP, Hochhaus A et al. Dasatinib versus imatinib in 33. Cortes JE, Kim DW, Pinilla-Ibarz J et al. A phase 2 trial of ponatinib in
newly diagnosed chronic-phase chronic myeloid leukemia. N Engl J Med Philadelphia chromosome-positive leukemias. N Engl J Med 2013; 369:
2010; 362: 2260–2270. 1783–1796.
19. Cortes JE, Saglio G, Kantarjian HM et al. Final 5-year study results of 34. Lipton JH, Chuah C, Guerci-Bresler A et al. Ponatinib versus imatinib
DASISION: the dasatinib versus imatinib study in treatment-naı̈ve chronic for newly diagnosed chronic myeloid leukaemia: an international, rando-
myeloid leukemia patients trial. J Clin Oncol 2016; 34: 2333–2340. mised, open-label, phase 3 trial. Lancet Oncol 2016; 17: 612–621.
20. Saglio G, Kim DW, Issaragrisil S et al. Nilotinib versus imatinib for newly 35. Hehlmann R, Saußele S, Voskanyan A, Silver RT. Management of CML-
diagnosed chronic myeloid leukemia. N Engl J Med 2010; 362: 2251–2259. blast crisis. Best Pract Res Clin Haematol 2016; 29: 295–307.
21. Hochhaus A, Saglio G, Hughes TP et al. Long-term benefits and risks of 36. Gratwohl A, Pfirrmann M, Zander A et al. Long-term outcome of pa-
frontline nilotinib vs imatinib for chronic myeloid leukemia in chronic tients with newly diagnosed chronic myeloid leukemia: a randomized
phase: 5-year update of the randomized ENESTnd trial. Leukemia 2016; comparison of stem cell transplantation with drug treatment. Leukemia
30: 1044–1054. 2016; 30: 562–569.
22. Padula WV, Larson RA, Dusetzina SB et al. Cost-effectiveness of tyrosine 37. Steegmann JL, Baccarani M, Breccia M et al. European LeukemiaNet rec-
kinase inhibitor treatment strategies for chronic myeloid leukemia in ommendations for the management and avoidance of adverse events of
chronic phase after generic entry of imatinib in the United States. J Natl treatment in chronic myeloid leukaemia. Leukemia 2016; 30: 1648–1671.
Cancer Inst 2016; 108: pii: djw003. 38. Mahon FX, Réa D, Guilhot J et al. Discontinuation of imatinib in pa-
23. Saussele S, Krauss MP, Hehlmann R et al. Impact of comorbidities on tients with chronic myeloid leukaemia who have maintained complete
overall survival in patients with chronic myeloid leukemia: results of the molecular remission for at least 2 years: the prospective, multicentre Stop
randomized CML-Study IV. Blood 2015; 126: 42–49. Imatinib (STIM) trial. Lancet Oncol 2010; 11: 1029–1035.
24. Branford S, Fletcher L, Cross NC et al. Desirable performance characteristics 39. Hochhaus A, Masszi T, Giles FJ et al. Treatment-free remission following
for BCR-ABL measurement on an international reporting scale to allow con- frontline nilotinib in patients with chronic myeloid leukemia in chronic
sistent interpretation of individual patient response and comparison of re- phase: results from the ENESTfreedom study. Leukemia 2017; 31:
sponse rates between clinical trials. Blood 2008; 112: 3330–3338. 1525–1531.
25. Branford S, Cross NC, Hochhaus A et al. Rationale for the recommenda- 40. Hughes TP, Ross DM. Moving treatment-free remission into mainstream
tions for harmonizing current methodology for detecting BCR-ABL clinical practice in CML. Blood 2016; 128: 17–23.
transcripts in patients with chronic myeloid leukaemia. Leukemia 2006; 41. Saußele S, Richter J, Hochhaus A, Mahon FX. The concept of treatment-
20: 1925–1930. free remission in chronic myeloid leukemia. Leukemia.2016; 30:
26. Cross NC, White HE, Müller MC et al. Standardized definitions of molecu- 1638–1647.
lar response in chronic myeloid leukemia. Leukemia 2012; 26: 2172–2175. 42. Richter J, Söderlund S, Lübking A et al. Musculoskeletal pain in patients
27. Hanfstein B, Müller MC, Hehlmann R et al. Early molecular and cytogen- with chronic myeloid leukemia after discontinuation of imatinib: a tyro-
etic response is predictive for long-term progression-free and overall sur- sine kinase inhibitor withdrawal syndrome? J Clin Oncol 2014; 32:
vival in chronic myeloid leukemia (CML). Leukemia 2012; 26: 2096–2102. 2821–2823.
28. Cross NC, White HE, Colomer D et al. Laboratory recommendations for 43. Dykewicz CA. Summary of the guidelines for preventing opportunistic
scoring deep molecular response following treatment for chronic mye- infections among hematopoietic stem cell transplant recipients. Clin
loid leukemia. Leukemia 2015; 29: 999–1003. Infect Dis 2001; 33: 139–144.

by guest
on 05 February 2018
Chronic lymphocytic leukaemia: ESMO Clinical Practice

B. Eichhorst1, T. Robak2, E. Montserrat3, P. Ghia4, P. Hillmen5, M. Hallek6 & C. Buske7, on behalf of
the ESMO Guidelines Committee*
1
Klinik I für Innere Medizin, Universität zu Köln, Köln, Germany; 2Department of Hematology, Medical University of Lodz, Copernicus Memorial Hospital, Lodz, Poland;
3
Institute of Hematology and Oncology, Department of Hematology, Hospital Clinic, University of Barcelona, Barcelona, Spain; 4Department of Onco-Hematology and
Division of Molecular Oncology, Università Vita-Salute San Raffaele, Fondazione Centro San Raffaele, IRCCS Istituto Scientifico San Raffaele, Milan, Italy; 5Institute of
Oncology, St James’s University Hospital, Leeds, UK; 6Department I Internal Medicine, Center for Integrated Oncology Köln-Bonn, Center of Excellence on Cellular Stress
Responses in Aging-Associated Diseases, University of Cologne, Köln; 7Comprehensive Cancer Center Ulm, Institute of Experimental Cancer Research, University Hospital,
Ulm, Germany
incidence and epidemiology and mantle cell lymphoma (MCL). These tumour cells express B-
cell surface antigens and MCL also expresses CD5, but usually not
Chronic lymphocytic leukaemia (CLL) is the most common CD23. For cases that express CD23, staining for cyclin D1 or
leukaemia in the Western world with an incidence of SOX11 and fluorescence in situ hybridisation (FISH) for detecting
4.2:100 000/year. The incidence increases to >30:100 000/year a translocation (11;14) are useful for establishing the diagnosis of
at an age of >80 years. The median age at diagnosis is 72 years. MCL. FMC7 may also help differentiating CLL from MCL, but
About 10% of the CLL patients are reported to be younger there are also FMC7 positive (atypical) CLL cases. Marginal zone
than 55 years. There is an inherited genetic susceptibility for lymphoma or lymphoplasmacytic lymphoma may also be differ-
clinical practice
CLL, with a 6- to 9-fold increased risk for family members of entiated by a negative or lower CD43 expression in comparison
guidelines
CLL patients. to CLL.

In the World Health Organization classification, small
lymphocytic lymphoma (SLL) and CLL are considered to be a
diagnosis and molecular biology single entity. The diagnosis of SLL requires the presence of
lymphadenopathy and/or splenomegaly with a number of B
The diagnosis of CLL is established by the following criteria [1]:
lymphocytes in the peripheral blood not exceeding 5 × 109/l.
– Presence in the peripheral blood of ≥5000 monoclonal B SLL cells show the same immunophenotype as CLL. The diag-
lymphocytes/µl. The clonality of the circulating B lympho- nosis of SLL should be confirmed by histopathological evalu-
cytes needs to be confirmed by flow cytometry. ation of a lymph node biopsy, whenever possible.
– The leukaemia cells found in the blood smear are character- In absence of lymphadenopathy, organomegaly, cytopaenia
istically small, mature-appearing lymphocytes with a narrow and clinical symptoms, the presence of fewer than 5000 mono-
border of cytoplasm and a dense nucleus lacking discernible clonal B lymphocytes/µl defines ‘monoclonal B-lymphocytosis’
nucleoli, and having partially aggregated chromatin. Larger, (MBL) [1], which can be detected in 5% of subjects with normal
atypical lymphocytes or prolymphocytes may be seen but blood count [2]. Progression to CLL occurs in 1%–2% of MBL
must not exceed 55%. cases per year [2].
CLL cells co-express the CD5 antigen and B-cell surface antigens
CD19, CD20 and CD23. The levels of surface immunoglobulin, staging and risk assessment
CD20 and CD79b are characteristically low compared with The following examinations are recommended before any treat-
those found on normal B cells. Each clone of leukaemia cells ment (Table 1) [III, B] [1]:
is restricted to expression of either kappa or lambda immuno-
globulin light chains. – History and physical examination including a careful palpa-
Other lymphoma entities to be separated from CLL are leu- tion of all lymph node areas, spleen and liver
kaemic marginal zone lymphoma, lymphoplasmacytic lymphoma – Complete blood cell count and differential count
– Serum chemistry including lactate dehydrogenase, bilirubin,
serum immunoglobulins, direct antiglobulin test
Viganello-Lugano CH-6962, Switzerland. – The history and status of relevant infections [i.e. hepatitis B
E-mail: clinicalguidelines@esmo.org and C, cytomegalovirus, human immunodeficiency virus]
† should be evaluated before chemoimmunotherapy or allo-
This publication supersedes the previously published version—Ann Oncol 2011; 22 geneic stem-cell transplantation (alloSCT), to avoid virus re-
(Suppl. 6): vi50–vi54. activation

by guest
on 05 February 2018
Table 1. Diagnostic and staging work-up Table 2. Staging systems for chronic lymphocytic leukaemia (CLL)
Pretreatment Response Stage Definition Median
evaluation evaluation survival
History, physical examination and + + Binet system

performance status
Binet A Hb ≥ 10.0 g/dl, thrombocytes >10 years
Complete blood count and differential + + ≥100 × 109/l,
Serum chemistry including serum + + <3 lymph node regions
immunoglobulin and direct Binet B Hb ≥ 10.0 g/dl, thrombocytes >8 years
antiglobulin test ≥100 × 109/l,
Cytogenetics (FISH) for del (17p)/ + − ≥3 lymph node regions
molecular genetics for TP53 mutation Binet C Hb < 10.0 g/dl, thrombocytes 6.5 years
a b <100 × 109/l
Marrow aspirate and biopsy + +
Rai system
Hepatitis B and C, CMV and HIV + −
serology Low risk
Rai 0 Lymphocytosis >15 × 109/l >10 years
a Intermediate risk
Only if clinically indicated.
b Rai I Lymphocytosis and lymphadenopathy >8 years
Only for confirmation of CR within clinical studies.
FISH, fluorescence in situ hybridisation; CMV, cytomegalovirus; HIV, Rai II Lymphocytosis and hepatomegaly and/
human immunodeficiency virus; CR, complete remission. or splenomegaly with/without
lymphadenopathy
High risk
Rai III Lymphocytosis and Hb < 11.0 g/dl 6.5 years
– FISH for detection of deletion of the chromosome 17 [del with/without lymphadenopathy/
(17p)] affecting the tumour protein p53 expression and in organomegaly
the absence of del(17p) molecular genetics for detection of Rai IV Lymphocytosis and thrombocytes
TP53 gene mutation (at least exons 4–10, eventually exons <100 × 109/l
2–11) [III, A] [3]. with/without lymphadenopathy/
organomegaly
The following additional examinations before treatment are de-
sirable [III, B] [1]: The overall survival times included in this table were adapted and have
changed during the past 30 years [7].
– Although a bone marrow biopsy is not required for diagno- Binet’s lymphoid areas consist in: lymphadenopathy either uni- or bi-
sis, it is recommended for the diagnostic evaluation of lateral in (1) cervical, (2) axillary, (3) inguinal areas, (4) spleen, (5) liver.
unclear cytopaenias, or FISH or molecular genetics if per- Hb, haemoglobin.
ipheral blood cell lymphocytosis does not allow adequate
immunophenotyping
– An extended FISH analysis is recommended before the start
of therapy because the detection of additional cytogenetic Additional markers are available to predict the prognosis of
abnormalities [del(11q) or trisomy 12] may have therapeutic patients with CLL, in particular at early stages [7, 8]. Patients
consequences with a detectable del(17p) or a mutation of TP53 (∼5% at diag-
– Molecular analysis for detecting immunoglobulin heavy nosis and up to 10% at treatment initiation) have the poorest
chain variable (IGHV) mutation status and better estimation prognosis, with a median OS of 2–5 years. The formerly poor
of duration of response prognosis of patients with a del(11q) (∼20%) has been improved
– Imaging studies by computed tomography (CT) scans may by chemoimmunotherapy with FCR (fludarabine, cyclophos-
be helpful to assess the tumour load or to determine the phamide and rituximab) [9]. More recently described gene
cause of unclear symptoms in individual patients, but they mutations such as NOTCH1, SF3B1, MYD88 or BIRC3 [10]
should not generally be used in asymptomatic patients or for may also predict an unfavourable prognosis in the absence of
clinical staging. In addition, CT scans may be useful for base- TP53 deletion/mutation [11, 12], but their clinical impact needs
line and final assessment in clinical trials [III, C]. In elderly further investigation [III, C]. Because leukaemic clones may
patients, abdominal ultrasound might be considered instead. evolve, FISH and TP53 mutation analyses should be repeated
before relapse treatment is administered [III, B] [13].
Two clinical staging systems are used to predict median survival About 50% of CLL patients present with an unmutated IGHV
(Table 2) [4, 5]. In Europe, the Binet staging system is more status [14, 15]. CLL cells with unmutated IGVH status have a
widely used, whereas in the United States, the Rai system is higher genetic instability with a higher risk of gaining unfavour-
more commonly applied. Both Binet and Rai staging systems able genetic mutations. OS and time to treatment intervention
separate three groups of patients with different prognoses are significantly shorter in this patient group. The expression of
(Table 2) [4, 5]. With the new treatment options available, the CD38 and ZAP70 correlates to some extent with the IGHV mu-
overall survival (OS) of patients with advanced disease stages tational status, but has no therapeutic impact and is therefore
has improved [6]. not required [III, C].
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv303 | v

by guest
on 05 February 2018
management of early disease stage without the above-mentioned conditions is not an indication
for treatment.
Binet stage A and B without active disease; Rai 0,
I and II without active disease front-line treatment. In physically fit patients ( physically
Previous studies have shown that early treatment with che- active, with no major health problems, normal renal function)
motherapeutic agents does not translate into a survival advan- without TP53 deletion/mutation, FCR is the standard first-line
tage in patients with early-stage CLL [16]. The standard therapy: improvement of OS has been demonstrated with this
treatment of patients with early disease is a watch-and-wait first-line chemoimmunotherapy (Figure 1) [I, A] [9]. Combinations
strategy [I, A]. Blood cell counts and clinical examinations based on other purine analogues such as cladribine [17] or
should be carried out every 3–12 months. pentostatin [18] have shown similar activity, but it is uncertain
Due to the lack of clinical trials, no evidence-based treatment whether they can replace fludarabine in the FCR regimen [II, B]. In
recommendation can be given for localised, early-stage SLL fit but elderly patients, FCR was shown to be associated with a
[I, A]. higher rate of severe infections when compared with bendamustine
plus rituximab (BR) [19]. Therefore, in this group of patients,
therapy with BR may be considered, although it produces fewer
treatment of advanced disease stage complete remissions than FCR [I, B]. Further studies evaluating
BR as front-line therapy in fit but elderly patients are therefore
Binet stage A and B with active disease or Binet required.
stage C; Rai 0–II with active disease or Rai III–IV In patients with relevant co-morbidity, who are usually older,
treatment indication. Treatment should only be initiated in but without TP53 deletion/mutation, the combination of chlor-
patients with symptomatic, active disease. The following conditions ambucil plus an anti-CD20 antibody (rituximab, ofatumumab
define active disease: significant B symptoms, cytopaenias not or obinutuzumab) prolongs progression-free survival (PFS)
caused by autoimmune phenomena and symptoms or compli- when compared with monotherapy and is therefore the standard
cations from lymphadenopathy, splenomegaly or hepatomegaly, approach [I, A] [20, 21]. In a head-to-head comparison of chlor-
lymphocyte doubling time of <6 months (only in patients with ambucil-based combinations, the type II antibody obinutuzu-
more than 30G lymphocytes/l), as well as autoimmune anaemia mab was superior to the type I antibody rituximab with regard
and/or thrombocytopaenia poorly responsive to conventional to PFS, complete remission (CR) and minimal residual disease
therapy [I, A]. The presence of del(17p) or TP53 mutation (MRD)-negative remissions.
Figure 1. Front-line treatment. CLL, chronic lymphocytic leukaemia; SLL, small lymphocytic leukaemia; BCR, B-cell receptor; R, rituximab; alloHSCT,
allogeneic haematopoietic stem cell transplantation; FCR, fludarabine, cyclophosphamide and rituximab; BR, bendamustine plus rituximab; Clb, chlorambucil.
v | Eichhorst et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Patients with TP53 deletion/mutation have a poor prognosis Treatment options include [III, B]:
even after FCR therapy [9]. Therefore, it is recommended that
patients with TP53 deletion/mutation are treated with novel – BCL2 antagonists alone or in combination within a clinical
inhibitors (ibrutinib; idelalisib and rituximab) in front-line and study
relapse settings [V, A]. For fit patients responding to inhibitor – Bruton’s tyrosine kinase inhibitor ibrutinib [23]
treatment, an allogeneic haematopoietic stem-cell transplant- – PI3K inhibitor idelalisib in combination with rituximab [24]
ation (HSCT) may be discussed, using individual and trans- – Other chemoimmunotherapy combinations should only
plant-related risk factors [III, B] [22]. be administered if TP53 deletion/mutation was excluded
Maintenance therapy in CLL patients with higher risk of (Figure 2).
relapse may have some benefit, but cannot be generally recom- Patients not responding nor progressing upon therapy with
mended. kinase inhibitors might be switched to a different kinase inhibi-
tor or to BCL2 antagonists when available (according to clinical
treatment of relapse and refractory disease. As for the first-line trials). Fit patients achieving second remission following the
therapy, treatment at relapse should only be started in second application of an inhibitor should proceed to allogeneic
symptomatic patients. Many patients with relapsed but HSCT [V, B] [22].
asymptomatic CLL can be followed with no therapy for a long
period of time. role of haematopoietic stem-cell transplantation. Autologous
First-line treatment may be repeated if the relapse or progres- stem-cell transplantation is not useful in CLL [I, A] [25]. An
sion occurs at least 24–36 months after chemoimmunotherapy alloSCT should be considered in patients achieving remission
and if TP53 deletion/mutation was excluded [III, B]. with kinase inhibitors or BCL2 antagonists after early relapse
If relapse occurs within 24–36 months after chemoimmu- from chemoimmunotherapy and/or with del(17p) or TP53
notherapy, or if the disease does not respond to any first-line mutation. In this situation, long-term treatment with inhibitors
therapy, the therapeutic regimen should be changed. is an alternative option. The decision should be based on
Figure 2. Relapse treatment. CLL, chronic lymphocytic leukaemia; SLL, small lymphocytic leukaemia; BCR, B-cell receptor; R, rituximab; BR, bendamustine
plus rituximab; FCR, fludarabine, cyclophosphamide and rituximab; alloHSCT, allogeneic haematopoietic stem cell transplantation.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv303 | v

by guest
on 05 February 2018
transplant- and disease-risk (e.g. availability of matched donor, morbidity of the patient (as assessed e.g. by using the cumulative
patient age and comorbidities and response to treatment) and illness rating scale [31]) may influence the choice of therapy.
the patient’s preferences, following a careful discussion of the
risks and benefits of an allogeneic transplant [22]. In patients
failing to several lines of therapy, allogeneic bone marrow
transplantation should be considered [III, B]. CLL is an incurable disease. Therefore, life-long observation and
follow-up is recommended for all patients. Follow-up of asymp-
treatment of CLL complications. Treatment of patients with tomatic patients should include a blood cell count and the pal-
autoimmune cytopaenia should be carried out according to the pation of lymph nodes, liver and spleen every 3–12 months
statement from the ‘ESMO guidelines consensus conference on depending on the dynamics of the leukaemic development.
malignant lymphoma: CLL’ [26]. Most patients with autoimmune Special attention should be paid to the appearance of auto-
cytopaenia respond to corticosteroids [III, B]. For patients not immune cytopaenias. Moreover, CLL patients have a two- to
responding to corticosteroids, rituximab administration might be sevenfold increased risk of developing secondary malignancies
a reasonable treatment option before splenectomy [III, B]. In (mostly solid cancers, but also secondary myelodysplastic syn-
patients with resistant autoimmune cytopaenia, treatment of the dromes or acute myeloblastic leukaemia).
underlying CLL is recommended. The transformation into a diffuse large B-cell lymphoma
Infections are a common complication in CLL patients; there- (DLBCL) or Hodgkin’s lymphoma (HL) occurs in 2%–15% of
fore, use of immunosuppressive agents, as for example corticos- CLL patients during the course of their disease. The diagnosis
teroids, should be restricted to a possible minimum. The use of has to be confirmed by histopathology exam of a lymph node
prophylactic systemic immunoglobulin does not have an impact (biopsy or excision). A positron emission tomography–CT
on OS [27, 28], and is only recommended in patients with might be useful to guide biopsy [IV, C]. The transformation of
severe hypogammaglobulinaemia and repeated infections [I, A]. CLL into Hodgkin’s disease represents a separate entity, where
Antibiotic and antiviral prophylaxis should be used in patients conventional chemotherapy against HL often achieves long-
with recurrent infections and/or very high risk of developing lasting remissions. The transformation into DLBCL is called
infections (e.g. pneumocystis prophylaxis with co-trimoxazole Richter’s transformation (RT) and usually has a very poor prog-
during treatment with chemoimmunotherapies based on purine nosis, with the exception of clonally-unrelated de novo DLBCL.
analogues or bendamustine) [IV, B]. Pneumococcal vaccination Treatment regimens for RT include therapies used in DLBCL
as well as seasonal influenza vaccination is recommended in such as rituximab plus CHOP (cyclophosphamide, vincristine,
early-stage CLL [IV, B]. doxorubicin and dexamethasone). More intense treatment regi-
mens such as rituximab plus hyper CVAD (cyclophosphamide,
vincristine, doxorubicin and dexamethasone alternating with
response evaluation. Response evaluation includes a careful methotrexate and cytarabine) or OFAR (oxaliplatin, fludarabine,
physical examination and a blood cell count. A bone marrow cytarabine and rituximab) have not been proven to induce
biopsy may be carried out to define CR [III, B] [1]. Chest X-ray better outcomes than R-CHOP (rituximab, cyclophosphamide,
and an abdominal ultrasound or CT for response evaluation doxorubicin, vincristine and prednisolone) [IV, B]. Response
may be carried out, if abnormal before therapy [IV, C] [1]. duration of RT is typically short, and an allogeneic HSCT
Detection of MRD by four-colour flow cytometry has a strong should be recommended to all patients with clonally related
prognostic impact [29, 30]. Patients who are MRD-negative after DLBCL with an available donor and sufficient fitness [IV, B].
therapy show a longer response duration and survival. Additional
clinical consequences of MRD positivity post-therapy remain
unclear except for patients after an allogeneic transplantation, where methodology
a positive MRD signal may trigger the reduction of immunosup- These clinical practice guidelines were developed in accordance
pressive therapies or the start of anti-leukaemic maintenance with the ESMO standard operating procedures for clinical practice
therapy. Therefore, MRD assessment is not generally recommended guidelines development. The relevant literature has been selected
for monitoring post-therapy outside clinical studies. by the expert authors. Levels of evidence and grades of recommen-
dation have been applied using the system shown in Table 3.
Statements without grading were considered justified standard
personalised medicine clinical practice by the experts and the ESMO faculty. This manu-
Principles of personalised or precision medicine in CLL have script has been subjected to an anonymous peer review process.
remained somewhat elusive. There is ample evidence that the
comprehensive evaluation of prognostic factors in early-stage
CLL improves the prediction of the individual prognosis [7].
However, a complete prognostic work-up often has no clinical BE has reported honoraria from Celgene, Gilead,
consequences. Therefore, the initial assessment of a CLL patient GlaxoSmithKline, Janssen, Mundipharma and Roche and has
can be carried out with a limited set of parameters (see above). received research grants from Mundipharma and Roche. TR has
In contrast, a complete genetic work-up should be carried out reported honoraria from Roche and Janssen and research grants
when CLL becomes symptomatic and requires treatment. In this from Roche, Janssen, GlaxoSmithKline, Pharmacyclics and
situation, genetic aberrations such as TP53 mutations alter the Gilead. EM has reported honoraria from Janssen, Pharmacyclics
choice of therapy (see above). Moreover, fitness and co- and Pfizer. PG has reported honoraria from Abbvie, Gilead,
v | Eichhorst et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Table 3. Levels of evidence and grades of recommendation 7. Pflug N, Bahlo J, Shanafelt TD et al. Development of a comprehensive prognostic
(adapted from the Infectious Diseases Society of America-United index for patients with chronic lymphocytic leukemia. Blood 2014; 124: 49–62.
States Public Health Service Grading Systema) 8. Döhner H, Stilgenbauer S, Benner A et al. Genomic aberrations and survival in
chronic lymphocytic leukemia. N Engl J Med 2000; 343: 1910–1916.
Levels of evidence 9. Hallek M, Fischer K, Fingerle-Rowson G et al. Addition of rituximab to fludarabine
and cyclophosphamide in patients with chronic lymphocytic leukemia: a
I Evidence from at least one large randomised, controlled trial of randomised, open-label, phase III trial. Lancet 2010; 376: 1164–1174.
good methodological quality (low potential for bias) or meta- 10. Wang L, Lawrence MS, Wan Y et al. SF3B1 and other novel cancer genes in
analyses of well-conducted randomised trials without chronic lymphocytic leukemia. N Engl J Med 2011; 365: 2497–2506.
heterogeneity 11. Stilgenbauer S, Schnaiter A, Paschka P et al. Gene mutations and treatment
II Small randomised trials or large randomised trials with a outcome in chronic lymphocytic leukemia: results from the CLL8 trial. Blood 2014;
suspicion of bias (lower methodological quality) or meta- 123: 3247–3254.
analyses of such trials or of trials with demonstrated 12. Villamor N, Conde L, Martinez-Trillos A et al. NOTCH1 mutations identify a genetic
heterogeneity subgroup of chronic lymphocytic leukemia patients with high risk of transformation
III Prospective cohort studies and poor outcome. Leukemia 2013; 27: 1100–1106.
IV Retrospective cohort studies or case–control studies 13. Landau DA, Carter SL, Stojanov P et al. Evolution and impact of subclonal
V Studies without control group, case reports, experts opinions mutations in chronic lymphocytic leukemia. Cell 2013; 152: 714–726.
14. Damle RN, Wasil T, Fais F et al. Ig V gene mutation status and CD38 expression as
Grades of recommendation novel prognostic indicators in chronic lymphocytic leukemia. Blood 1999; 94:
A Strong evidence for efficacy with a substantial clinical benefit, 1840–1847.
strongly recommended 15. Hamblin TJ, Orchard JA, Ibbotson RE et al. CD38 expression and immunoglobulin
B Strong or moderate evidence for efficacy but with a limited variable region mutations are independent prognostic variables in chronic
lymphocytic leukemia, but CD38 expression may vary during the course of the
disease. Blood 2002; 99: 1023–1029.
C Insufficient evidence for efficacy or benefit does not outweigh
16. Dighiero G, Maloum K, Desablens B et al. Chlorambucil in indolent chronic
lymphocytic leukemia. French Cooperative Group on Chronic Lymphocytic
optional Leukemia. N Engl J Med 1998; 338: 1506–1514.
D Moderate evidence against efficacy or for adverse outcome, 17. Robak T, Jamroziak K, Gora-Tybor J et al. Comparison of cladribine plus
generally not recommended cyclophosphamide with fludarabine plus cyclophosphamide as first-line therapy for
E Strong evidence against efficacy or for adverse outcome, never chronic lymphocytic leukemia: a phase III randomized study by the Polish Adult
recommended Leukemia Group (PALG-CLL3 Study). J Clin Oncol 2010; 28: 1863–1869.
18. Kay NE, Geyer SM, Call TG et al. Combination chemoimmunotherapy with
a
By permission of the Infectious Diseases Society of America [32]. pentostatin, cyclophosphamide, and rituximab shows significant clinical activity
with low accompanying toxicity in previously untreated B chronic lymphocytic
leukemia. Blood 2007; 109: 405–411.
Pharmacyclics, Janssen, Boehringer Ingelheim, Roche Italia, 19. Eichhorst B, Fink AM, Busch R et al. Frontline chemoimmunotherapy with
fludarabine (F), cyclophosphamide (C), and rituximab (R) (FCR) shows superior
Celgene, Merck and research grants from GlaxoSmithKline,
efficacy in comparison to bendamustine (B) and rituximab (BR) in previously
Gilead and Roche. MH has reported honoraria for consulting and untreated and physically fit patients ( pts) with advanced chronic lymphocytic
speaker activities from Abbvie, Celgene, Gilead, GlaxoSmithKline, leukemia (CLL): final analysis of an international, randomized study of the German
Janssen, Mundipharma and Roche and has received research CLL Study Group (GCLLSG) (CLL10 Study). Blood 2014; 124: Abstract 19.
grants from Mundipharma, Janssen, Abbvie and Roche. CB has 20. Goede V, Fischer K, Busch R et al. Obinutuzumab plus chlorambucil in patients
reported honoraria from Roche, Pfizer, Celgene, Pharmacyclics with CLL and coexisting conditions. N Engl J Med 2014; 370: 1101–1110.
and Janssen and research grants from Roche and Janssen. PH has 21. Hillmen P, Robak T, Janssens A et al. Ofatumumab + chlorambucil versus
not reported any potential conflicts of interest. chlorambucil alone in patients with untreated chronic lymphocytic leukemia (CLL):
results of the phase III study complement 1 (OMB110911). Blood 2013; 122:
Abstract 528.
22. Dreger P, Schetelig J, Andersen N et al. Managing high-risk CLL during transition
references to a new treatment era: stem cell transplantation or novel agents? Blood 2014;
1. Hallek M, Cheson BD, Catovsky D et al. Guidelines for the diagnosis and treatment 124: 3841–3849.
of chronic lymphocytic leukemia: a report from the International Workshop on 23. Byrd JC, Brown JR, O’Brien S et al. Ibrutinib versus ofatumumab in previously
Chronic Lymphocytic Leukemia updating the National Cancer Institute-Working treated chronic lymphoid leukemia. N Engl J Med 2014; 371: 213–223.
Group 1996 guidelines. Blood 2008; 111: 5446–5456. 24. Furman RR, Sharman JP, Coutre SE et al. Idelalisib and rituximab in relapsed
2. Rawstron AC, Bennett FL, O’Connor SJ et al. Monoclonal B-cell lymphocytosis and chronic lymphocytic leukemia. N Engl J Med 2014; 370: 997–1007.
chronic lymphocytic leukemia. N Engl J Med 2008; 359: 575–583. 25. Brion A, Mahé B, Kolb B et al. Autologous transplantation in CLL patients with B
3. Baliakas P, Hadzidimitriou A, Sutton LA et al. Recurrent mutations refine prognosis and C Binet stages: final results of the prospective randomized GOELAMS LLC 98
in chronic lymphocytic leukemia. Leukemia 2015; 29: 329–336. trial. Bone Marrow Transplant 2012; 47: 542–548.
4. Binet JL, Auquier A, Dighiero G et al. A new prognostic classification of chronic 26. Ghielmini M, Vitolo U, Kimby E et al. ESMO Guidelines consensus conference on
lymphocytic leukemia derived from a multivariate survival analysis. Cancer 1981; malignant lymphoma 2011 part 1: diffuse large B-cell lymphoma (DLBCL),
48: 198–206. follicular lymphoma (FL) and chronic lymphocytic leukemia (CLL). Ann Oncol
5. Rai KR, Sawitsky A, Cronkite EP et al. Clinical staging of chronic lymphocytic 2013; 24: 561–576.
leukemia. Blood 1975; 46: 219–234. 27. Intravenous immunoglobulin for the prevention of infection in chronic lymphocytic
6. Abrisqueta P, Pereira A, Rozman C et al. Improving survival in patients with chronic leukemia. A randomized, controlled clinical trial. Cooperative Group for the Study of
lymphocytic leukemia (1980–2008): the Hospital Clinic of Barcelona experience. Immunoglobulin in Chronic Lymphocytic Leukemia. N Engl J Med 1988; 319:
Blood 2009; 114: 2044–2050. 902–907.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv303 | v

by guest
on 05 February 2018
28. Raanani P, Gafter-Gvili A, Paul M et al. Immunoglobulin prophylaxis in chronic lymphocytic leukemia: a multivariate analysis from the randomized GCLLSG CLL8
lymphocytic leukemia and multiple myeloma: systematic review and meta-analysis. trial. J Clin Oncol 2012; 30: 980–988.
Leuk Lymphoma 2009; 50: 764–772. 31. Extermann M, Overcash J, Lyman GH et al. Comorbidity and functional
29. Moreton P, Kennedy B, Lucas G et al. Eradication of minimal residual disease in B- status are independent in older cancer patients. J Clin Oncol 1998; 16:
cell chronic lymphocytic leukemia after alemtuzumab therapy is associated with 1582–1587.
prolonged survival. J Clin Oncol 2005; 23: 2971–2979. 32. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
30. Bottcher S, Ritgen M, Fischer K et al. Minimal residual disease quantification is among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
an independent predictor of progression-free and overall survival in chronic 139–144.
v | Eichhorst et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Diffuse large B-cell lymphoma (DLBCL): ESMO Clinical

follow-up†
H. Tilly1, M. Gomes da Silva2, U. Vitolo3, A. Jack4, M. Meignan5, A. Lopez-Guillermo6, J. Walewski7,
M. André8, P. W. Johnson9, M. Pfreundschuh10 & M. Ladetto11, on behalf of the ESMO
1
Centre Henri-Becquerel, Université de Rouen, Rouen, France; 2Portuguese Institute of Oncology, Lisbon, Portugal; 3A.O. Città della Salute e della Scienza di Torino, Turin,
Italy; 4St James’s University Hospital, Leeds, UK; 5Henri Mondor University Hospital, Créteil, France; 6Hospital Clinic, Barcelona, Spain; 7Maria Sklodowska-Curie Memorial
Institute and Oncology Centre, Warsaw, Poland; 8CHU Dinant-Godinne, UCL Namur, Yvoir, Belgium; 9Cancer Research UK, University of Southampton, Southampton, UK;
10
Innere Medizin I, Universität des Saarlandes, Hamburg, Germany; 11Divisione di Ematologia, Azienda Ospedaliera Santi Antonio e Biagio e Cesare Arrigo, Alessandria, Italy
incidence and epidemiology immunohistochemistry (IHC) or flow cytometry or a combin-

ation of both techniques [V, A]. Panels used must be designed
Diffuse large B-cell lymphoma (DLBCL) constitutes 30%–58% of to confirm B-cell lineage, and must be comprehensive enough to
non-Hodgkin’s lymphoma series. The crude incidence in Europe is highlight possible variant forms such as immunoblastic lymph-
3.8/100 000/year [1]. The incidence increases with age and varies oma [4], primary mediastinal B cell lymphoma (PMBCL), T-cell/
considerably across Europe. A family history of lymphoma, auto- histiocyte rich large B-cell lymphoma, primary cutaneous
clinical practice
immune disease, human immunodeficiency virus (HIV) infection, DLBCL leg-type or EBV-positive DLBCL of the elderly. They
guidelines
hepatitis C virus (HCV) seropositivity, a high body mass as a young must also distinguish alternative diagnoses that may be difficult
adult and some occupational exposures have been identified as risk to make on the basis of morphology alone, and which have im-
factors of DLBCL [2]. In recent years, there have been important portant clinical consequences as plasmablastic lymphoma or
survival improvements for DLBCL in all European regions [3]. soft tissue involvement by myeloma, Burkitt lymphoma, unclas-
sifiable B-cell lymphoma with features intermediate between
diagnosis and pathology/molecular diffuse large cell lymphoma and Burkitt lymphoma, blastic
biology mantle cell lymphoma and some cases of Hodgkin’s lymphoma.
A suggested immunohistochemical panel would include CD20,
The diagnosis of DLBCL should be carried out in a reference CD79a, BCL6, CD10, MYC, BCL2, Ki67, IRF4, CyclinD1, CD5
haematopathology laboratory with expertise in morphological and CD23. EBER-1 staining may be used to identify the
interpretation and the facilities to carry out the full range of Epstein–Barr virus-positive DLBCL subtype of the elderly popu-
phenotypic and molecular investigations [V, A]. lation. The histological report should give the diagnosis accord-
A surgical excision biopsy remains the optimal method of diag- ing to the current World Health Organization classification [5].
nosis [V, A]. This allows assessment of nodal architecture and pro- Where the level of confidence in the diagnosis is reduced, for
vides adequate material for phenotypic and molecular studies. example, because only a small biopsy specimen is available or
Ideally, the biopsy should be sent unfixed to the laboratory to allow where the putatively neoplastic population has a normal pheno-
flow cytometric studies to be carried out and high-quality DNA and type by IHC, demonstration of B-cell monoclonality by a poly-
RNA to be extracted. Needle-core and endoscopic biopsies should merase chain reaction-based method should be considered
be reserved for patients for whom a surgical approach is impractical [IV, C] [6].
or would entail excessive risk [IV, B]. A fine-needle aspirate should The cell of origin phenotype determined by gene expression
not be used as the sole basis for a diagnosis of DLBCL [V, E]. profiling is also a major prognostic factor in DLBCL [7–9].
A morphological diagnosis of DLBCL should be confirmed in Tumours with a germinal centre phenotype have a significantly
all cases by immunophenotypic investigations, either better clinical outcome that those with an activated B-cell pheno-
type. The nature of type 3 or unclassified subgroups requires
further clarification. Newer methods, based on evaluation of a
limited set of genes, have been validated in comparison with
E-mail: clinicalguidelines@esmo.org standard gene expression, and are now used in the setting of clin-
ical trials [9, 10]. Cell of origin can also be determined by IHC
†
Approved by the ESMO Guidelines Committee: February 2002, last update August
but published data on the prognostic effect of immunohistochem-
2015.
This publication supersedes the previously published version—Ann Oncol 2012; 23 ical techniques are contradictory, and it is not recommended to
(Suppl. 7): vii78–vii82. routinely base clinical decisions on these results [11, 12]. General

by guest
on 05 February 2018
issues of reproducibility may also limit the value of IHC as a prog- A diagnostic lumbar puncture should be considered in high-
nostic biomarker [13]. risk patients as described above [V, A]. Flow cytometry, when
The presence of an MYC rearrangement in combination available, enhances the detection of lymphoma cells in the cere-
with BCL2 rearrangement, and possibly other genetic abnor- brospinal fluid [24].
malities, has been described as a special entity (‘double-hit’ or Cardiac function (left ventricular ejection fraction) should be
‘triple-hit’ lymphoma). However, the prognostic significance assessed before treatment [V, A]. The risks of infertility and pos-
of these rearrangements remains controversial and optimal sibilities of fertility preservation should be discussed, depending
clinical management is not established [14–16]. This assess- on the type of treatment being proposed.
ment is recommended, wherever technically possible, in newly The staging is established according to the Ann Arbor classifi-
diagnosed and relapsed patients being treated with curative cation system [I, A] (Table 1). A new staging system that has not
intent, using interphase fluorescence in situ hybridisation [IV, been prospectively validated has been recently proposed [20].
B]. The immunohistochemical expression of MYC and/or For prognostic purposes, the International Prognostic Index
BCL2 or both (double expressors) is only partly correlated (IPI) and age-adjusted IPI (aaIPI) should be calculated [I, A]
with genetic abnormalities, but the concurrent expression of [25] (Table 2). Other factors that may affect prognosis and treat-
MYC and BCL2 is usually associated with a poor outcome ment strategies, including the maximum bulk of the disease,
[17–19]. should be assessed [30].
Table 1. Ann Arbor staging classification

Stage
Physical examination, performance status (PS) and assessment
I Involvement of a single lymphatic region (I) or localized
of B symptoms are necessary [V, A]. A complete blood count, involvement of single extralymphatic organ or site (IE)
routine blood chemistry including lactate dehydrogenase II Involvement of two or more lymphatic regions on the same side
(LDH) and uric acid, as well as screening tests for HIV, hepa- of the diaphragm (II) or localized involvement of a single
titis B virus (HBV) (HBs antigen, anti-HBs and anti-HBc anti- extralymphatic organ or site and of one or more lymphatic
bodies) and HCV are required [V, A]. Protein electrophoresis regions on the same side of the diaphragm (IIE)
is recommended [IV, B]. III Involvement of lymphatic regions on both sides of the diaphragm
Based on recent consensus recommendations for staging and IV Diffuse or disseminated involvement of one or more
restaging of lymphoma developed by the clinical and imaging extralymphatic organs with or without lymphatic involvement
working groups of the international conference of malignant
lymphomas (Lugano classification), fluorodeoxyglucose positron
emission tomography (FDG-PET)/computed tomography (CT)
scan is now recommended as the gold standard for staging Table 2. International prognostic index (IPI)
DLBCL patients [V, A] [20, 21]. PET/CT is more accurate than International prognostic index (IPI) Estimated 3-year
contrast-enhanced CT (CeCT), with increased sensitivity for overall survival
nodal and extranodal sites; in practice, CeCT is often carried out [26–29] (95% CI)
before PET/CT. Should this not be the case, a full diagnostic Risk factors Age >60 years
high-dose CeCT should be carried out when necessary, in com- Serum LDH > normal
bination with PET/CT and after the PET scan [V, B]. Indeed, Stage III–IV
CeCT may be necessary for a better delineation of lymphadenop- Performance status 2–4
athy from the bowel; the detection of compression/thrombosis of Extranodal sites >1
central/mediastinal vessels, radiation planning or a more accurate
Risk categories Low 0–1 91 (89–94)
measurement of nodal sites in the context of a trial. The findings
Low intermediate 2 81 (73–86)
of CeCT when carried out at baseline determine whether the low-
High intermediate 3 65 (58–73)
dose non-enhanced CT part of the PET/CT scan will be sufficient
High 4–5 59 (49–69)
for restaging.
Focal bone marrow FDG uptake with or without increased Age-adjusted international prognostic index
diffuse uptake is more sensitive than bone marrow biopsy (BMB) (aaIPI) in patients ≤60 years
for infiltration in DLBCL and is highly specific [22]. Low-volume Risk factors Serum LDH > normal
involvement (<10%–20%) and discordant lymphoma may be Stage III–IV
missed by PET/CT imaging but these positive BMB/negative Performance status 2–4
PET/CT findings are <10% [23]. Thus, biopsy is no longer
required when a PET/CT scan demonstrates bone or marrow in- Risk categories Low 0 98 (96–100)
volvement indicating advanced-stage disease but is appropriate in Low intermediate 1 92 (87–95)
case of negative PET, when its results would change prognosis High intermediate 2
and treatment, especially when a shortened number of immuno- High 3 }75 (66–82)
chemotherapy cycles is proposed [V, C].
For suspected central nervous system (CNS) lymphoma, mag- LDH, lactate dehydrogenase; CI, confidence interval.
netic resonance imaging is the modality of choice [III, A].
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv304 | v

by guest
on 05 February 2018
Table 3. Recommended treatment strategies in diffuse large B-cell lymphoma

Patients ≤60 years
IPI low risk (aaIPI = 0) and no bulk IPI low risk (aaIPI = 0) with bulk or IPI IPI intermediate-high risk or IPI high risk
low-intermediate risk (aaIPI = 1) (aaIPI = 2, 3)
R-CHOP21 × 6 R-ACVBP and sequential consolidation R-CHOP21 × 6–8

or or
R-CHOP21 × 6 + IF-RT on bulk R-CHOP14 × 6 with 8 R
Consider more intensive regimens in
selected patients:
R-CHOEP14 × 6
or
R-CHOP or R-ACVBP plus HDCT with
ASCT
Consider CNS prophylaxis in patients at risk for CNS progression
Elderly >60 years
Fit, 60–80 years >80 years without cardiac dysfunction Unfit or frail or >60 years with cardiac
dysfunction
R-CHOP21 × 6–8 Attenuated regimens: Doxorubicin substitution with

(R-CHOP21 × 6 for IPI low risk) R-miniCHOP21 × 6 gemcitabine, etoposide or liposomal
or doxorubicin or others:
R-CHOP14 × 6 with 8 R R-C(X)OP21 × 6
or
palliative care
Consider CNS prophylaxis in patients at risk
First relapse/progress
Eligible for transplant Not eligible for transplant
Platinum-based chemotherapy regimens (i.e. R-DHAP, R-ICE, R- Platinum- and/or gemcitabine-based

GDP) as salvage treatment regimens
For chemosensitive patients: R-HDCT with ASCT as Clinical trials with novel drugs
remission consolidation
Consider allogeneic transplantation in patients relapsed
after R-HDCT with ASCT or in patients with
poor-risk factors at relapse
>2 relapse/progress
Eligible for transplant Not eligible for transplant
Allogeneic transplantation Clinical trials with novel drugs

Clinical trials with novel drugs Palliative care
IPI, International Prognostic Index; aaIPI, age-adjusted IPI; R, rituximab; CHOP, cyclophosphamide, doxorubicin, vincristine, prednisone; ACVBP,
doxorubicin, vindesine, cyclophosphamide, bleomycin and prednisolone; IF-RT, involved-field radiotherapy; HDCT, high-dose chemotherapy; ASCT,
autologous stem-cell transplantation; DHAP, cisplatin, cytarabine, dexamethasone; ICE, ifosfamide, carboplatin, etoposide; GDP, cisplatin, gemcitabine,
dexamethasone; CNS, central nervous system; CHOEP, cyclophosphamide, doxorubicin, vincristine, etoposide, prednisolone; R-C(X)OP, R-CHOP with
substitution of doxorubicin.
haematopoietic growth factors in patients treated with curative

treatment
intent and in patients older than 60 years of age [I, A].
Treatment strategies should be stratified according to age,
IPI and feasibility of dose-intensified approaches (Table 3). young low-risk patients (aa-IPI = 0) without
Whenever available, the inclusion in a clinical trial is bulky disease
recommended. Six cycles of combination chemotherapy with cyclophospha-
In cases with high tumour load, precautions such as the ad- mide, doxorubicin, vincristine and prednisone (CHOP) treat-
ministration of prednisone ( p.o.) several days as ‘prephase’ ment combined with six doses of rituximab given every 21 days
treatment are advised to avoid tumour lysis syndrome [I, A]. is the current standard [I, A] [31]. Consolidation by radiother-
Dose reductions due to haematological toxicity should be apy to initial non-bulky sites has no proven benefit in patients
avoided [I, A]. Febrile neutropaenia justifies prophylactic use of treated with rituximab or not [I, A] [32, 33].
v | Tilly et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
young low-intermediate-risk patients (aa-IPI = 1) exposure has been shown to improve outcome of elderly poor-
or IPI low risk (aa-IPI = 0) with bulky disease prognosis patients without increasing toxicity [III, C] [46]. A
Rituximab (R)-CHOP 21 × 6 with radiotherapy to the sites of comprehensive geriatric assessment in order to ascertain co-
previous bulky disease was shown to be effective in this group of morbidities and functional decline is recommended to guide the
patients, based on the results of the MINT study [II, B] [31]. choice of treatment in these patients [III, A] [47, 48]. R-CHOP
Alternatively, an intensification of chemotherapy with R-ACVBP treatment can usually be used up to 80 years of age in fit patients
(rituximab, doxorubicin, vindesine, cyclophosphamide, bleo- [I, A] but modulation of treatment according to geriatric assess-
mycin and prednisolone), given every 2 weeks followed by ment is recommended [III, C] [49].
sequential consolidation, has been shown to improve survival
compared with eight cycles of R-CHOP in this category, but patients aged >80 years
radiotherapy was omitted in both arms of this trial [I, A] [26]. The combination of rituximab with attenuated chemotherapy,
In this group of patients, either R-CHOP21 × 6 with radiother- such as R-miniCHOP, can induce complete remission and long
apy to the sites of previous bulky disease or the intensified survival in fit patients older than 80 years [III, B] [50].
regimen R-ACVBP is recommended [II, B]. Substitution of doxorubicin by gemcitabine, etoposide or liposo-
mal doxorubicin, or even its omission, can be considered from
young high- and high-intermediate-risk patients the beginning or after a few cycles in patients with cardiac dys-
(aa-IPI ≥ 2) function or who are frail or unfit [III, C] [51].
There is no current standard in this subgroup, and in this group
especially, enrolment in clinical trials should be a priority. Six to central nervous system (CNS) prophylaxis
eight cycles of chemotherapy with CHOP combined with eight Patients with high-intermediate and high-risk IPI, especially
doses of rituximab given every 21 days are most frequently those with more than one extranodal site or elevated LDH, are
applied [III, B]. Dose dense treatment with R-CHOP given every at higher risk of CNS relapse [52]. Testicular, renal and adrenal
14 days has not demonstrated a survival advantage over standard involvements have been validated as additional risk factors [53].
R-CHOP given every 21 days [I, C] [34]. In this trial, R-CHOP 14 CNS prophylaxis should be recommended in these populations
failed to show a better outcome in any DLBCL subset, including [II, A]. MYC gene rearrangement is associated with a high risk of
young poor-risk patients, although the trial was not powered to CNS relapse [43]. Although widely used, intrathecal injections of
compare multiple clinical subgroups. Intensive treatment with methotrexate may not be an optimal method. Intravenous high-
R-ACVBP or R-CHOEP (rituximab, cyclophosphamide, doxo- dose methotrexate has been shown to be associated with efficient
rubicin, vincristine, etoposide and prednisolone) is frequently disease control [IV, C] [54–56]. Prospective trials are ongoing to
used but these regimens have not been directly compared with evaluate this alternative approach.
R-CHOP in this category [II, B] [27, 35].
Four randomised trials comparing rituximab chemotherapy some DLBCLs require special consideration
(R-chemotherapy) followed by high-dose chemotherapy • Extranodal DLBCLs and PMBCLs are considered in other
(HDC) and autologous stem-cell transplantation (ASCT) guidelines.
versus R-chemotherapy alone have been presented. Two trials • Patients with HIV infection should usually receive the same
showed a progression-free survival (PFS) benefit for HDC treatment as HIV-negative patients in association with anti-
with ASCT but no impact, at present, on overall survival (OS) viral therapy [II, A] [57].
[36, 37], while two trials failed to demonstrate an improve- • Patients previously exposed to HBV (HBs antigen-negative,
ment for the HDC arm [35, 38]. Therefore, HDC with ASCT anti-HBc-positive) are at risk of reactivation during treatment
in first line remains experimental or may be proposed for with R-CHOP. Antiviral prophylaxis or periodic HBV DNA
selected high-risk patients [II, C]. The role of consolidation by monitoring and antiviral treatment in the case of reactivation
radiotherapy to initial sites of bulky disease is unknown. The are recommended [III, A] [58].
role of interim PET to select patients who could benefit from
consolidative ASCT [39] or from radiotherapy [40] is under
evaluation [I, C]. response evaluation
patients aged 60–80 years post-treatment evaluation
Six to eight cycles of combination chemotherapy with CHOP FDG-PET/CT is now the recommended standard for post-treat-
plus eight doses of rituximab given every 21 days is the current ment assessment in DLBCL [I, A] [59]. The recent Lugano classifi-
standard [I, A] [41]. R-CHOP given every 14 days did not dem- cation based on the visual Deauville criteria (5-point scale, Table 4)
onstrate a survival advantage over R-CHOP 21 [I, C] [34, 42]. If
R-CHOP is given every 14 days, six cycles of CHOP with eight Table 4. PET 5-point scale (Deauville criteria)
cycles of rituximab are sufficient [I, A] [43]. In patients with
1 No uptake
localised disease (IPI = 0), no benefit of consolidation by radio-
2 Uptake ≤mediastinum
therapy was shown in patients treated before the introduction of
3 Uptake >mediastinum but ≤liver
rituximab [I, A] [44], but a recent study indicated that irradi-
4 Moderately increased uptake compared with liver
ation could improve the outcome of elderly patients with bulky 5 Markedly increased uptake to liver and/or new lesions
disease [II, C] [45]. In a phase II study, extended rituximab
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv304 | v

by guest
on 05 February 2018
has proposed different response categories, termed ‘metabolic to be around 1/100 000/year. In addition to initial prognostic
response categories’ [20, 21]: factors, the nature of previous treatments and time from initial
treatment are of utmost importance [63].
• Complete metabolic response (CMR) is defined when no
residual uptake exists or if the residual uptake is lower to or diagnosis. In patients who are suspected of having relapsed on
equal to the liver activity (Deauville score 1–3), with or the basis of imaging studies, the diagnosis should be confirmed
without evidence of residual mass on the CT part of the exam- by biopsy before proceeding to second-line therapy. In these
ination, and without FDG-avid lesions in the bone marrow. circumstances, a needle-core biopsy is acceptable as primary
Since most patients with score 3 (uptake greater than medias- investigation.
tinal activity) have a good prognosis with standard treatment,
score 3 has been included in the CMR category but a careful staging and risk assessment. Patients still amenable to curative
evaluation of these patients is recommended. therapy should have the same examinations as at first diagnosis.
• Deauville scores 4 and 5 indicate residual disease in most
cases. Three categories of response are defined by comparing treatment. The following recommendations apply to patients
the residual uptake with the tumour uptake in baseline scan: who received adequate rituximab and anthracycline-containing
partial metabolic response when the uptake has decreased, no first-line therapy.
metabolic response when it has not changed or progressive In patients aged <65–70 years with good PS and no major
metabolic disease (PMD) when it has increased. A new site of organ dysfunction, salvage regimens with rituximab and chemo-
FDG uptake consistent with lymphoma is graded score 5 and therapy followed, in responsive patients, by HDC and ASCT,
indicates a PMD but should be biopsied or followed by inter- are recommended [II, A] [63–65] Salvage regimens such as
val scans in case of aetiological uncertainties. In the presence R-DHAP (rituximab, cisplatin, cytarabine, dexamethasone) or
of residual metabolically active tissue, where salvage treatment R-ICE (rituximab, ifosfamide, carboplatin, etoposide) appear to
is being considered, a biopsy is recommended [III, A]. have similar outcomes [I, A] [63]. However, R-GDP (rituximab,
cisplatin, gemcitabine, dexamethasone) has been shown to have
interim evaluation similar efficacy but less toxicity than R-DHAP [I, A] [66]. One
study suggested a possible advantage of R-DHAP in the germi-
Mid-treatment imaging after three to four cycles may be used to
nal centre B-cell subtype, but this needs confirmation [IV, C]
rule out progression in clinical practice [V, B]. It is usually carried
[67]. BEAM (carmustine, etoposide, cytarabine and melphalan)
out with CT but PET/CT can also be used when available [20].
is the most commonly used high-dose regimen [III, B].
Changing treatment solely on the basis of interim PET/CT is dis-
Additional involved-field radiation or iceberg radiation may
couraged [II, E], unless there is clear evidence of progression.
be used, especially in the few cases with limited stage disease,
Early PET evaluation carried out after one to two cycles of
but this has never been evaluated in controlled trials [IV, C].
treatment has been shown to be predictive of outcome, but
Maintenance with rituximab is not recommended [I, E] [68].
should be reserved for clinical trials at the present time [II, D].
Allogeneic transplantation with a sibling or matched unrelated
donor may be considered in patients with refractory disease,
follow-up early relapse or relapse after ASCT [III, B] [69].
Patients not suitable for high-dose therapy may be treated with
Patients with DLBCL who are event-free at 2 years have an iden- the same or other salvage regimens as R-GEMOX (rituximab,
tical OS to that of the general population, emphasising the need gemcitabine, oxaliplatin) [III, B] [70]. Pixantrone, a new anthra-
to only specifically monitor the disease in this early period [60]. cycline-like drug with reduced cardiotoxicity, demonstrated some
Careful history and physical examination every 3 months for efficacy in heavily treated patients [II, C] [71]. However, these
1 year, every 6 months for 2 more years and then once a year patients should be preferably enrolled in clinical trials testing the
with attention to development of secondary tumours or other activity of other novel drugs.
long-term side-effects of chemotherapy is recommended [V, D].
Blood count should be carried out at 3, 6, 12 and 24 months, response evaluation. Response criteria are identical to those of
then only as needed for evaluation of suspicious symptoms or first-line treatment evaluation. An evaluation should be carried out
clinical findings in those patients suitable for further therapy [V, C]. after three to four cycles of the salvage regimen (before high-dose
Minimal radiological examinations at 6, 12 and 24 months treatment) and after the end of all therapy. Results of PET before
after end of treatment by CT scan are common practice, but high-dose treatment are correlated to clinical outcome [72].
there is no definitive evidence that routine imaging in patients in
complete remission provides any outcome advantage, and it may follow-up. Follow-up of patients in second response is the same
increase the incidence of secondary malignancies [V, D] [61, 62]. as for first response.
Routine surveillance with PET scan is not recommended [V, E].
High-risk patients with curative options may potentially mandate
more frequent evaluation. personalised medicine
Progress in the knowledge of pathological and molecular hetero-
relapsed and refractory DLBCL geneity of DLBCL has led to the study of new agents that have
incidence. Overall, more than 30% of DLBCL will ultimately distinct activity in molecular subtypes, or have specific efficacy
relapse. The incidence in the European Union is therefore estimated on molecular targets involved in disease pathogenesis. At the
v | Tilly et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
• Diagnosis should be carried out in a reference haematopathology laboratory with expertise in morphological interpretation and the facilities to carry out
the full range of phenotypic and molecular investigations [V, A].
• Surgical biopsy is the optimal method of diagnosis. [V, A].
• Needle-core and endoscopic biopsies should be reserved for patients for whom a surgical approach is impractical or would entail excessive risk [IV, B].
• A fine-needle aspirate should not be used as the sole basis for a diagnosis of DLBCL [V, E].
• A morphological diagnosis of DLBCL should be confirmed in all cases by immunophenotypic investigations [V, A].
• If there is doubt in the diagnosis, demonstration of B-cell monoclonality by a PCR-based method should be considered [IV, C].
• Assessment of MYC and BCL2 rearrangement is recommended (whenever technically possible) in newly diagnosed and relapsed patients treated with
curative intent, using interphase FISH [IV, B].
• Physical exam, performance status and assessment of B symptoms are necessary [V, A].
• A complete blood count, routine blood chemistry including LDH and uric acid, as well as screening tests for HIV, HBV and HCV are required [V, A].
• Protein electrophoresis is recommended [IV, B].
• FDG-PET/CT scan is recommended as the gold standard for staging DLBCL patients [V, A].
• If CeCT is not carried out before PET/CT, a full diagnostic high-dose CeCT should be carried out when necessary, in combination with PET/CT [V, B].
Biopsy may be avoided when PET/CT scans demonstrate bone or marrow involvement indicating advanced-stage disease but is appropriate in the case of
negative PET, when its results would change prognosis and treatment, especially when a shortened number of immunochemotherapy cycles is proposed
[V, C].
• For suspected CNS lymphoma, MRI is the modality of choice [III, A].
• A diagnostic lumbar puncture should be considered in high-risk patients [V, A].
• Cardiac function (LVEF) should be assessed before treatment [V, A].
• The staging is established according to the Ann Arbor classification system [I, A].
• For prognostic purposes, the IPI and aa-IPI should be calculated [I, A].
Treatment
• Treatment strategies should be stratified according to age, IPI and feasibility of dose-intensified approaches.
• Whenever available, inclusion in a clinical trial is recommended.
• In cases with high tumour load, precautions are advised to avoid tumour lysis syndrome [I, A].
• Dose reductions due to haematological toxicity should be avoided whenever possible [I, A].
• The risk of febrile neutropenia justifies prophylactic use of haematopoietic growth factors in patients treatment with curative intent and in
patients >60 years of age [I, A].
• For young, low-risk patients (aa-IPI = 0) without bulky disease:
○ six cycles of combination chemotherapy with CHOP treatment combined with six doses of rituximab given every 21 days is the current standard [I, A];
○ consolidation by radiotherapy to initial non-bulky sites has no proven benefit in patients treated with rituximab or not [I, A].
• For young low-intermediate-risk patients (aa-IPI = 1) or IPI low risk (aa-IPI = 0) with bulky disease:
○ either R-CHOP21 × 6 with radiotherapy to the sites of previous bulky disease or the intensified regimen R-ACVBP is recommended [II, B].
• For young high- and high-intermediate-risk patients (aa-IPI ≥ 2):

○ enrolment in clinical trials should be a priority;
○ six to eight cycles of chemotherapy with CHOP combined with eight doses of rituximab given every 21 days are most frequently applied [III, B];
○ dose dense treatment with R-CHOP given every 14 days has not demonstrated a survival advantage over standard R-CHOP given every 21 days [I, C];
○ intensive treatment with R-ACVBP or R-CHOEP is frequently used but these regimens have not been directly compared with R-CHOP in this category
[II, B];
○ HDC with ASCT in first line remains experimental or may be proposed for selected high-risk patients [II, C];
○ the role of interim PET to select patients who could benefit from consolidative ASCT or from radiotherapy is under evaluation [I, C].
• For patients aged 60–80 years:

○ six to eight cycles of combination chemotherapy with CHOP plus eight doses of rituximab given every 21 days is the current standard [I, A];
○ if R-CHOP is given every 14 days, six cycles of CHOP with eight cycles of rituximab are sufficient [I, A];
○ a comprehensive geriatric assessment in order to ascertain comorbidities and functional decline is recommended to guide the choice of treatment in
elderly poor-prognosis patients [III, A];

○ R-CHOP treatment can usually be used up to 80 years of age in fit patients [I, A], but modulation of treatment according to geriatric assessment is
recommended [III, C].

• For patients aged >80 years:
○ the combination of rituximab with attenuated chemotherapy, such as R-miniCHOP, can induce complete remission and long survival in fit patients
older than 80 years [III, B];

○ substitution of doxorubicin by gemcitabine, etoposide or liposomal doxorubicin, or even its omission, can be considered from the beginning or after a
few cycles in patients with cardiac dysfunction or who are frail or unfit [III, C].
Continued
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv304 | v

by guest
on 05 February 2018
Table 5. Continued
• CNS prophylaxis:
○ should be recommended for patients with high-intermediate-risk and high-risk IPI, especially those with more than one extranodal site or elevated
LDH or for patients with testicular, renal or adrenal involvement [II, A];
○ intravenous high-dose methotrexate has been shown to be associated with efficient disease control [IV, C].
• Patients with human immunodeficiency virus (HIV) infection should usually receive the same treatment as HIV-negative patients in association with
antiviral therapy [II, A].
• Antiviral prophylaxis or periodic HBV DNA monitoring and antiviral treatment are recommended for patients previously exposed to HBV who
experience reactivation of the virus during treatment [III, A].
Response evaluation
• FDG-PET/CT is the recommended standard for post-treatment assessment in DLBCL [I, A].
• In the presence of residual metabolically active tissue, where salvage treatment is being considered, a biopsy is recommended [III, A].
• Interim evaluation:
○ mid-treatment imaging after three to four cycles may be used to rule out progression in clinical practice [V, B];
○ changing treatment solely on the basis of interim PET/CT is discouraged [II, E], unless there is clear evidence of progression;
○ early PET evaluation carried out after one to two cycles of treatment has been shown to be predictive of outcome, but should be reserved for clinical
trials at the present time [II, D].
Follow-up
• Careful history and physical examination every 3 months for 1 year, every 6 months for 2 further years and then once a year with attention to
development of secondary tumours or other long-term side-effects of chemotherapy is recommended [V, D].
• Blood count should be carried out at 3, 6, 12 and 24 months, then only as needed for evaluation of suspicious symptoms or clinical findings in those
patients suitable for further therapy [V, C].
• Minimal radiological examinations at 6, 12 and 24 months after end of treatment, by CT scan, is common practice, but there is no definitive evidence
that routine imaging in patients in complete remission provides any outcome advantage and it may increase the incidence of secondary malignancies
[V, D]. Routine surveillance with PET scan is not recommended [V, E].
DLBCL, diffuse large B-cell lymphoma; PCR, polymerase chain reaction; FISH, fluorescence in situ hybridisation; LDH, lactate dehydrogenase; HIV,
human immunodeficiency virus; HBV, hepatitis B virus; HCV, hepatitis C virus; FDG, fluorodeoxyglucose; PET, positron emission tomography; CT,
computed tomography; CeCT, contrast-enhanced CT; CNS, central nervous system; MRI, magnetic resonance imaging; LVEF, left ventricular ejection
fraction; IPI, International Prognostic Index; aa-IPI, age-adjusted IPI; CHOP, cyclophosphamide, doxorubicin, vincristine and prednisone; R, rituximab; R-
AVCBP, rituximab, doxorubicin, vindesine, cyclophosphamide, bleomycin and prednisolone; R-CHOEP, rituximab, cyclophosphamide, doxorubicin,
vincristine, etoposide and prednisolone; HDC, high-dose chemotherapy; ASCT, autologous stem-cell transplantation.
Levels of evidence
I Evidence from at least one large randomised, controlled trial of good methodological quality (low potential for bias) or meta-
analyses of well-conducted randomised trials without heterogeneity
II Small randomised trials or large randomised trials with a suspicion of bias (lower methodological quality) or meta-analyses of
such trials or of trials with demonstrated heterogeneity
C Insufficient evidence for efficacy or benefit does not outweigh the risk or the disadvantages (adverse events, costs, … ),
optional
a
v | Tilly et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
present time, pending results of large comparative studies, none conflict of interest
of these agents is appropriate for routine therapy in practice.
The activated B-cell (ABC) subtype has been shown to have a HT has received honoraria from Celgene, Roche, Janssen-Cilag
worse prognosis when compared with germinal centre B-cell and Takeda; research contracts from Celgene, Roche/Genentech
(GCB) in patients treated by R-CHOP [8]. A subgroup analysis and Janssen-Cilag. MG has reported advisory boards for Roche,
suggested that R-ACVBP could have a survival benefit over R- Celgene, Janssen-Cilag, Pfizer and Ferrer. UV has reported ad-
CHOP in the non-GCB population [III, C] [73]. The ABC visory boards for Roche; lectures sponsored by Roche, Janssen,
subtype is characterised by a constitutive activation of the Celgene and Mundipharma. AL-G has reported advisory boards
NF-κB pathway, which could be targeted by different agents for Roche, Celgene, Novartis, Mundipharma, Infinity, Bayer and
as bortezomib and lenalidomide. Bortezomib combined with Gilead. JW has declared advisory boards for Roche, Celgene,
dose-adjusted-EPOCH (etoposide, vincristine, doxorubicin, Janssen-Cilag and Takeda; research support from Roche,
cyclophosphamide and prednisone) (DA-EPOCH) has shown Celgene, Janssen-Cilag, Takeda, GlaxoSmithKline, Gilead and
selective activity in a small study of relapsed/refractory ABC- Seattle Genetics. PJ has reported research grants partially
DLBCL [74]. A UK/Swiss phase III trial (REMoDL-B) compar- funded by Janssen-Cilag and Epizyme; member of data moni-
ing R-CHOP with R-CHOP-bortezomib in gene expression toring and safety committee for Boehringer Ingelheim; advisory
prolife defined cell of origin subgroups of DLBCL is nearly com- board member for Roche, Bristol-Myers Squibb, Janssen-Cilag
pleted. Lenalidomide, as a single agent, demonstrated selective and Takeda. MP has declared advisory boards for Boehringer-
efficacy in the non-GCB subtype [75, 76]. In two phase II Ingelheim, Celgene and Roche; research support from Roche,
studies, the combination of lenalidomide and R-CHOP showed Amgen and Spectrum. ML has reported honoraria from
acceptable toxicity [77, 78]. In one of these studies, the PFS and Celgene, Janssen-Cilag, Roche, Amgen Mundipharma and Teva;
OS of the patients treated with the combination were identical research contracts from Celgene, Pfizer, Mundipharma and
in non-GCB and GCB subtypes [78], leading to the initiation of Roche; funds received from Amgen, Roche and Takeda. MM
a randomised study in the ABC subtype. and MA have reported no potential conflicts of interest. UV has
Ibrutinib, a novel oral Bruton’s tyrosine kinase inhibitor, has not reported any potential conflicts of interest.
shown selective activity in ABC-DLBCL. The combination of
ibrutinib with R-CHOP has demonstrated promising responses,
leading to the initiation of a phase III trial in the non-GCB references
population [79].
DLBCL with MYC rearrangement and/or MYC overexpres- 1. Sant M, Allemani C, Tereanu C et al. Incidence of hematologic malignancies in
Europe by morphologic subtype: results of the HAEMACARE project. Blood 2010;
sion is usually considered a subgroup with aggressive behaviour.
116: 3724–3734.
However, many uncertainties remain about the extent of this
2. Morton LM, Slager SL, Cerhan JR et al. Etiologic heterogeneity among non-
subgroup concerning translocation partners, additional defects Hodgkin lymphoma subtypes: the InterLymph Non-Hodgkin Lymphoma Subtypes
(double or triple hit), combination of genetic abnormalities and Project. J Natl Cancer Inst Monogr 2014: 130–144.
MYC protein overexpression and dual overexpression with 3. Sant M, Minicozzi P, Mounier M et al. Survival for haematological malignancies in
MYC and BCL2 [80]. Although R-CHOP gives poor outcomes Europe between 1997 and 2008 by region and age: results of EUROCARE-5, a
for double-hit lymphomas, only preliminary results have sug- population-based study. Lancet Oncol 2014; 15: 931–942.
gested better results with more intensive regimens, and clinical 4. Ott G, Ziepert M, Klapper W et al. Immunoblastic morphology but not the
trials are required in this subtype [81, 82]. immunohistochemical GCB/non-GCB classifier predicts outcome in diffuse large B-cell
lymphoma in the RICOVER-60 trial of the DSHNHL. Blood 2010; 116: 4916–4925.
Whole-genome next-generation sequencing studies have
5. Swerdlow SH, Campo E, Harris NL et al. World Health Organization Classification of
identified frequent and recurrent mutations which may play a Tumours of Haematopoietic and Lymphoid Tissues, 4th edition. Lyon, France: IARC
crucial role in lymphoma development [83–85]. These molecu- 2008.
lar defects may prove useful targets in the future treatment and 6. Langerak AW, Groenen PJ, Brüggemann M et al. EuroClonality/BIOMED-2
management of DLBCL. guidelines for interpretation and reporting of Ig/TCR clonality testing in suspected
lymphoproliferations. Leukemia 2012; 26: 2159–2171.
7. Wright G, Tan B, Rosenwald A et al. A gene expression-based method to diagnose
clinically distinct subgroups of diffuse large B cell lymphoma. Proc Natl Acad Sci
USA 2003; 100: 9991–9996.
methodology 8. Lenz G, Wright GW, Emre NC et al. Molecular subtypes of diffuse large B-cell
lymphoma arise by distinct genetic pathways. Proc Natl Acad Sci USA 2008; 105:
These clinical practice guidelines were developed in accordance 13520–13525.
with the ESMO standard operating procedures for clinical prac- 9. Scott DW, Wright GW, Williams PM et al. Determining cell-of-origin subtypes of
tice guidelines development. The relevant literature has been diffuse large B-cell lymphoma using gene expression in formalin-fixed paraffin-
selected by the expert authors. A summary of recommendations embedded tissue. Blood 2014; 123: 1214–1217.
is shown in Table 5. Levels of evidence and grades of recom- 10. Mareschal S, Ruminy P, Bagacean C et al. Accurate classification of germinal
center b-cell-like/activated b-cell-like diffuse large b-cell lymphoma using a simple
mendation have been applied using the system shown in
and rapid reverse transcriptase-multiplex ligation-dependent probe amplification
Table 6. Statements without grading were considered justified assay: a CALYM study. Mol Diagn 2015 [Epub ahead of print].
standard clinical practice by the experts and the ESMO faculty. 11. Hans CP, Weisenburger DD, Greiner TC et al. Confirmation of the molecular
This manuscript has been subjected to an anonymous peer classification of diffuse large B-cell lymphoma by immunohistochemistry using a
review process. tissue microarray. Blood 2004; 103: 275–282.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv304 | v

by guest
on 05 February 2018
12. Choi WW, Weisenburger DD, Greiner TC et al. A new immunostain algorithm 32. Reyes F, Lepage E, Ganem G et al. ACVBP versus CHOP plus radiotherapy for
classifies diffuse large B-cell lymphoma into molecular subtypes with high localized aggressive lymphoma. N Engl J Med 2005; 352: 1197–1205.
accuracy. Clin Cancer Res 2009; 15: 5494–5502. 33. Lamy T, Damaj G, Gyan E et al. R-CHOP with or without radiotherapy in non-bulky
13. de Jong D, Rosenwald A, Chhanabhai M et al. Immunohistochemical prognostic limited-stage diffuse large B cell lymphoma (DLBCL): preliminary results of the
markers in diffuse large B-cell lymphoma: validation of tissue microarray as a prospective randomized phase III 02-03 trial from the Lysa/Goelams Group. In:
prerequisite for broad clinical applications—a study from the Lunenburg 2014 ASH Annual Meeting, Orlando, Florida. Abstract 393.
Lymphoma Biomarker Consortium. J Clin Oncol 2007; 25: 805–812. 34. Cunningham D, Hawkes EA, Jack A et al. Rituximab plus cyclophosphamide,
14. Savage KJ, Johnson NA, Ben-Neriah S et al. MYC gene rearrangements are doxorubicin, vincristine, and prednisolone in patients with newly diagnosed diffuse
associated with a poor prognosis in diffuse large B-cell lymphoma patients treated large B-cell non-Hodgkin lymphoma: a phase 3 comparison of dose intensification
with R-CHOP chemotherapy. Blood 2009; 114: 3533–3537. with 14-day versus 21-day cycles. Lancet 2013; 381: 1817–1826.
15. Barrans S, Crouch S, Smith A et al. Rearrangement of MYC is associated with poor 35. Schmitz N, Nickelsen M, Ziepert M et al. Conventional chemotherapy (CHOEP-14)
prognosis in patients with diffuse large B-cell lymphoma treated in the era of with rituximab or high-dose chemotherapy (MegaCHOEP) with rituximab for young,
rituximab. J Clin Oncol 2010; 28: 3360–3365. high-risk patients with aggressive B-cell lymphoma: an open-label, randomised,
16. Lin P, Dickason TJ, Fayad LE et al. Prognostic value of MYC rearrangement in cases phase 3 trial (DSHNHL 2002–1). Lancet Oncol 2012; 13: 1250–1259.
of B-cell lymphoma, unclassifiable, with features intermediate between diffuse large 36. Vitolo U, Chiappella A, Brusamolino E et al. Rituximab dose-dense
B-cell lymphoma and Burkitt lymphoma. Cancer 2012; 118: 1566–1573. chemotherapy followed by intensified high-dose chemotherapy and autologous
17. Green TM, Young KH, Visco C et al. Immunohistochemical double-hit score is a stem cell transplantation (HDC+ASCT) significantly reduces the risk of
strong predictor of outcome in patients with diffuse large B-cell lymphoma treated progression compared to standard rituximab dose-dense chemotherapy as first
with rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone. J line treatment in young patients with high-risk (aa-IPI 2-3) diffuse large B-cell
Clin Oncol 2012; 30: 3460–3467. lymphoma (DLBCL): final results of phase III randomized trial DLCL04 of the
18. Johnson NA, Slack GW, Savage KJ et al. Concurrent expression of MYC and BCL2 Fondazione Italiana Linfomi (FIL). Blood (ASH Annual Meeting Abstracts), 2012;
in diffuse large B-cell lymphoma treated with rituximab plus cyclophosphamide, 120: 688.
doxorubicin, vincristine, and prednisone. J Clin Oncol 2012; 30: 3452–3459. 37. Stiff PJ, Unger JM, Cook JR et al. Autologous transplantation as consolidation for
19. Horn H, Ziepert M, Becher C et al. MYC status in concert with BCL2 and BCL6 aggressive non-Hodgkin’s lymphoma. N Engl J Med 2013; 369: 1681–1690.
expression predicts outcome in diffuse large B-cell lymphoma. Blood 2013; 121: 38. Le Gouill S, Milpied NJ, Lamy T et al. First-line rituximab (R) high-dose therapy (R-
2253–2263. HDT) versus R-CHOP14 for young adults with diffuse large B-cell lymphoma:
20. Cheson BD, Fisher RI, Barrington SF et al. Recommendations for initial evaluation, preliminary results of the GOELAMS 075 prospective multicenter randomized trial.
staging, and response assessment of Hodgkin and non-Hodgkin lymphoma: the J Clin Oncol (ASCO Annual Meeting Abstracts Part 1) 2011; 29: 8003.
Lugano classification. J Clin Oncol 2014; 32: 3059–3068. 39. Casasnovas RO, Ysebaert L, Thieblemont C et al. Final results of a randomized
21. Barrington SF, Mikhaeel NG, Kostakoglu L et al. Role of imaging in the staging and phase II GELA/LYSA study of rituximab plus ACVBP or CHOP, using a PET-driven
response assessment of lymphoma: consensus of the International Conference on consolidation strategy, in patients with high-risk diffuse large B-cell lymphoma
Malignant Lymphomas Imaging Working Group. J Clin Oncol 2014; 32: 3048–3058. (DLBCL). J Clin Oncol (ASCO Annual Meeting Abstracts) 2014; 32: 8503.
22. Khan AB, Barrington SF, Mikhaeel NG et al. PET-CT staging of DLBCL accurately 40. Sehn LH, Hardy ELG, Gill KK et al. Phase 2 trial of interim PET scan-tailored
identifies and provides new insight into the clinical significance of bone marrow therapy in patients with advanced stage diffuse large B-cell lymphoma (DLBCL) in
involvement. Blood 2013; 122: 61–67. British Columbia (BC). Blood (ASH Annual Meeting Abstracts) 2014; 124: 392.
23. Pelosi E, Penna D, Douroukas A et al. Bone marrow disease detection with FDG- 41. Coiffier B, Thieblemont C, Van Den Neste E et al. Long-term outcome of patients in
PET/CT and bone marrow biopsy during the staging of malignant lymphoma: results the LNH-98.5 trial, the first randomized study comparing rituximab-CHOP to
from a large multicentre study. Q J Nucl Med Mol Imaging 2011; 55: 469–475. standard CHOP chemotherapy in DLBCL patients: a study by the Groupe d’Etudes
24. Benevolo G, Stacchini A, Spina M et al. Final results of a multicenter trial des Lymphomes de l’Adulte. Blood 2010; 116: 2040–2045.
addressing role of CSF flow cytometric analysis in NHL patients at high risk for 42. Delarue R, Tilly H, Mounier N et al. Dose-dense rituximab-CHOP compared with
CNS dissemination. Blood 2012; 120: 3222–3228. standard rituximab-CHOP in elderly patients with diffuse large B-cell lymphoma
25. A predictive model for aggressive non-Hodgkin’s lymphoma. The International Non- (the LNH03-6B study): a randomised phase 3 trial. Lancet Oncol 2013; 14:
Hodgkin’s Lymphoma Prognostic Factors Project. N Engl J Med 1993; 329: 987–994. 525–533.
26. Récher C, Coiffier B, Haioun C et al. Intensified chemotherapy with ACVBP plus 43. Pfreundschuh M, Schubert J, Ziepert M et al. Six versus eight cycles of bi-weekly
rituximab versus standard CHOP plus rituximab for the treatment of diffuse large CHOP-14 with or without rituximab in elderly patients with aggressive CD20+ B-
B-cell lymphoma (LNH03-2B): an open-label randomised phase 3 trial. Lancet cell lymphomas: a randomised controlled trial (RICOVER-60). Lancet Oncol 2008;
2011; 378: 1858–1867. 9: 105–116.
27. Fitoussi O, Belhadj K, Mounier N et al. Survival impact of rituximab combined with 44. Bonnet C, Fillet G, Mounier N et al. CHOP alone compared with CHOP plus
ACVBP and upfront consolidation autotransplantation in high-risk diffuse large B- radiotherapy for localized aggressive lymphoma in elderly patients: a study by the
cell lymphoma for GELA. Haematologica 2011; 96: 1136–1143. Groupe d’Etude des Lymphomes de l’Adulte. J Clin Oncol 2007; 25: 787–792.
28. Ziepert M, Hasenclever D, Kuhnt E et al. Standard international prognostic index 45. Held G, Murawski N, Ziepert M et al. Role of radiotherapy to bulky disease in
remains a valid predictor of outcome for patients with aggressive CD20+ B-cell elderly patients with aggressive B-cell lymphoma. J Clin Oncol 2014; 32:
lymphoma in the rituximab era. J Clin Oncol 2010; 28: 2373–2380. 1112–1118.
29. Ketterer N, Coiffier B, Thieblemont C et al. Phase III study of ACVBP versus ACVBP 46. Pfreundschuh M, Poeschel V, Zeynalova S et al. Optimization of rituximab for the
plus rituximab for patients with localized low-risk diffuse large B-cell lymphoma treatment of diffuse large B-cell lymphoma (II): extended rituximab exposure time
(LNH03–1B). Ann Oncol 2013; 24: 1032–1037. in the SMARTE-R-CHOP-14 trial of the German high-grade non-Hodgkin
lymphoma study group. J Clin Oncol 2014; 32: 4127–4133.
30. Pfreundschuh M, Ho AD, Cavallin-Stahl E et al. Prognostic significance of
maximum tumour (bulk) diameter in young patients with good-prognosis diffuse 47. Morrison VA, Hamlin P, Soubeyran P et al. Diffuse large B-cell lymphoma in the
large-B-cell lymphoma treated with CHOP-like chemotherapy with or without elderly: impact of prognosis, comorbidities, geriatric assessment, and
rituximab: an exploratory analysis of the MabThera International Trial Group (MInT) supportive care on clinical practice. An International Society of Geriatric
study. Lancet Oncol 2008; 9: 435–444. Oncology (SIOG) Expert Position Paper. J Geriatr Oncol 2015; 6: 141–152.
31. Pfreundschuh M, Kuhnt E, Trumper L et al. CHOP-like chemotherapy with or 48. Tucci A, Martelli M, Rigacci L et al. Comprehensive geriatric assessment is an
without rituximab in young patients with good-prognosis diffuse large-B-cell essential tool to support treatment decisions in elderly patients with diffuse large
lymphoma: 6-year results of an open-label randomised study of the MabThera B-cell lymphoma: a prospective multicenter evaluation in 173 patients by the
International Trial (MInT) Group. Lancet Oncol 2011; 12: 1013–1022. Lymphoma Italian Foundation (FIL). Leuk Lymphoma 2015; 56: 921–926.
v | Tilly et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
49. Spina M, Balzarotti M, Uziel L et al. Modulated chemotherapy according to 68. Gisselbrecht C, Schmitz N, Mounier N et al. Rituximab maintenance therapy after
modified comprehensive geriatric assessment in 100 consecutive elderly patients autologous stem-cell transplantation in patients with relapsed CD20(+) diffuse large
with diffuse large B-cell lymphoma. Oncologist 2012; 17: 838–846. B-cell lymphoma: final analysis of the collaborative trial in relapsed aggressive
50. Peyrade F, Jardin F, Thieblemont C et al. Attenuated immunochemotherapy lymphoma. J Clin Oncol 2012; 30: 4462–4469.
regimen (R-miniCHOP) in elderly patients older than 80 years with diffuse large B- 69. Glass B, Hasenkamp J, Wulf G et al. Rituximab after lymphoma-directed conditioning
cell lymphoma: a multicentre, single-arm, phase 2 trial. Lancet Oncol 2011; 12: and allogeneic stem-cell transplantation for relapsed and refractory aggressive non-
460–468. Hodgkin lymphoma (DSHNHL R3): an open-label, randomised, phase 2 trial. Lancet
51. Fields PA, Townsend W, Webb A et al. De novo treatment of diffuse large B-cell Oncol 2014; 15: 757–766.
lymphoma with rituximab, cyclophosphamide, vincristine, gemcitabine, and 70. Mounier N, El Gnaoui T, Tilly H et al. Rituximab plus gemcitabine and oxaliplatin in
prednisolone in patients with cardiac comorbidity: a United Kingdom National patients with refractory/relapsed diffuse large B-cell lymphoma who are not
Cancer Research Institute trial. J Clin Oncol 2014; 32: 282–287. candidates for high-dose therapy. A phase II Lymphoma Study Association trial.
52. Kridel R, Dietrich PY. Prevention of CNS relapse in diffuse large B-cell lymphoma. Haematologica 2013; 98: 1726–1731.
Lancet Oncol 2011; 12: 1258–1266. 71. Pettengell R, Coiffier B, Narayanan G et al. Pixantrone dimaleate versus other
53. Savage KJ, Zeynalova S, Kansara RR et al. Validation of a prognostic model to chemotherapeutic agents as a single-agent salvage treatment in patients with
assess the risk of CNS disease in patients with aggressive B-cell lymphoma. Blood relapsed or refractory aggressive non-Hodgkin lymphoma: a phase 3, multicentre,
(ASH Annual Meeting Abstracts) 2014; 124: 394. open-label, randomised trial. Lancet Oncol 2012; 13: 696–706.
54. Tilly H, Lepage E, Coiffier B et al. Intensive conventional chemotherapy (ACVBP 72. Dickinson M, Hoyt R, Roberts AW et al. Improved survival for relapsed diffuse large
regimen) compared with standard CHOP for poor-prognosis aggressive non- B cell lymphoma is predicted by a negative pre-transplant FDG-PET scan following
Hodgkin lymphoma. Blood 2003; 102: 4284–4289. salvage chemotherapy. Br J Haematol 2010; 150: 39–45.
55. Abramson JS, Hellmann M, Barnes JA et al. Intravenous methotrexate as central 73. Molina TJ, Canioni D, Copie-Bergman C et al. Young patients with non-germinal
nervous system (CNS) prophylaxis is associated with a low risk of CNS recurrence center B-cell-like diffuse large B-cell lymphoma benefit from intensified
in high-risk patients with diffuse large B-cell lymphoma. Cancer 2010; 116: chemotherapy with ACVBP plus rituximab compared with CHOP plus rituximab:
4283–4290. analysis of data from the Groupe d’Etudes des Lymphomes de l’Adulte/lymphoma
56. Cheah CY, Herbert KE, O’Rourke K et al. A multicentre retrospective comparison of study association phase III trial LNH 03–2B. J Clin Oncol 2014; 32: 3996–4003.
central nervous system prophylaxis strategies among patients with high-risk diffuse 74. Dunleavy K, Pittaluga S, Czuczman MS et al. Differential efficacy of bortezomib
large B-cell lymphoma. Br J Cancer 2014; 111: 1072–1079. plus chemotherapy within molecular subtypes of diffuse large B-cell lymphoma.
57. Coutinho R, Pria AD, Gandhi S et al. HIV status does not impair the outcome of Blood 2009; 113: 6069–6076.
patients diagnosed with diffuse large B-cell lymphoma treated with R-CHOP in the 75. Hernandez-Ilizaliturri FJ, Deeb G, Zinzani PL et al. Higher response to lenalidomide in
cART era. AIDS 2014; 28: 689–697. relapsed/refractory diffuse large B-cell lymphoma in nongerminal center B-cell-like
58. Seto WK, Chan TS, Hwang YY et al. Hepatitis B reactivation in patients with than in germinal center B-cell-like phenotype. Cancer 2011; 117: 5058–5066.
previous hepatitis B virus exposure undergoing rituximab-containing chemotherapy 76. Czuczman MS, Davies A, Linton KM et al. A phase 2/3 multicenter, randomized study
for lymphoma: a prospective study. J Clin Oncol 2014; 32: 3736–3743. comparing the efficacy and safety of lenalidomide versus investigator’s choice in
59. Meignan M, Itti E, Gallamini A, Haioun C. Interim 18F-fluorodeoxyglucose relapsed/refractory DLBCL. Blood (ASH Annual Meeting Abstracts) 2014; 124: 628.
positron emission tomography in diffuse large B-cell lymphoma: qualitative or 77. Vitolo U, Chiappella A, Franceschetti S et al. Lenalidomide plus R-CHOP21 in
quantitative interpretation—where do we stand? Leuk Lymphoma 2009; 50: elderly patients with untreated diffuse large B-cell lymphoma: results of the
1753–1756. REAL07 open-label, multicentre, phase 2 trial. Lancet Oncol 2014; 15: 730–737.
60. Maurer MJ, Ghesquieres H, Jais JP et al. Event-free survival at 24 months is a 78. Nowakowski GS, LaPlant B, Macon WR et al. Lenalidomide combined with R-
robust end point for disease-related outcome in diffuse large B-cell lymphoma CHOP overcomes negative prognostic impact of non-germinal center B-cell
treated with immunochemotherapy. J Clin Oncol 2014; 32: 1066–1073. phenotype in newly diagnosed diffuse large B-cell lymphoma: a phase II study. J
61. Thompson CA, Ghesquieres H, Maurer MJ et al. Utility of routine post-therapy Clin Oncol 2015; 33: 251–257.
surveillance imaging in diffuse large B-cell lymphoma. J Clin Oncol 2014; 32: 79. Younes A, Thieblemont C, Morschhauser F et al. Combination of ibrutinib with
3506–3512. rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP)
62. Chien SH, Liu CJ, Hu YW et al. Frequency of surveillance computed tomography in for treatment-naive patients with CD20-positive B-cell non-Hodgkin lymphoma: a
non-Hodgkin lymphoma and the risk of secondary primary malignancies: a non-randomised, phase 1b study. Lancet Oncol 2014; 15: 1019–1026.
nationwide population-based study. Int J Cancer 2015; 137: 658–665. 80. Swerdlow SH. Diagnosis of "double hit" diffuse large B-cell lymphoma and B-cell
63. Gisselbrecht C, Glass B, Mounier N et al. Salvage regimens with autologous lymphoma, unclassifiable, with features intermediate Burkitt lymphoma: when and how,
transplantation for relapsed large B-cell lymphoma in the rituximab era. J Clin FISH versus IHC. Hematology Am Soc Hematol Educ Program 2014; 2014: 90–99.
Oncol 2010; 28: 4184–4190. 81. Oki Y, Noorani M, Lin P et al. Double hit lymphoma: the MD Anderson Cancer
64. Horwitz SM, Negrin RS, Blume KG et al. Rituximab as adjuvant to high-dose Center clinical experience. Br J Haematol 2014; 166: 891–901.
therapy and autologous hematopoietic cell transplantation for aggressive non- 82. Petrich AM, Gandhi M, Jovanovic B et al. Impact of induction regimen and stem
Hodgkin lymphoma. Blood 2004; 103: 777–783. cell transplantation on outcomes in double-hit lymphoma: a multicenter
65. Kewalramani T, Zelenetz AD, Nimer SD et al. Rituximab and ICE as second-line retrospective analysis. Blood 2014; 124: 2354–2361.
therapy before autologous stem cell transplantation for relapsed or primary 83. Intlekofer AM, Younes A. Precision therapy for lymphoma—current state and
refractory diffuse large B-cell lymphoma. Blood 2004; 103: 3684–3688. future directions. Nat Rev Clin Oncol 2014; 11: 585–596.
66. Crump M, Kuruvilla J, Couban S et al. Randomized comparison of gemcitabine, 84. Dunleavy K, Roschewski M, Wilson WH. Precision treatment of distinct molecular
dexamethasone, and cisplatin versus dexamethasone, cytarabine, and cisplatin subtypes of diffuse large B-cell lymphoma: ascribing treatment based on the
chemotherapy before autologous stem-cell transplantation for relapsed and refractory molecular phenotype. Clin Cancer Res 2014; 20: 5182–5193.
aggressive lymphomas: NCIC-CTG LY.12. J Clin Oncol 2014; 32: 3490–3496. 85. Jardin F. Next generation sequencing and the management of diffuse large B-cell
67. Thieblemont C, Briere J, Mounier N et al. The germinal center/activated B-cell lymphoma: from whole exome analysis to targeted therapy. Discov Med 2014; 18:
subclassification has a prognostic impact for response to salvage therapy in 51–65.
relapsed/refractory diffuse large B-cell lymphoma: a bio-CORAL study. J Clin Oncol 86. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections among
2011; 29: 4079–4087. hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33: 139–144.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv304 | v

by guest
on 05 February 2018
Extranodal diffuse large B-cell lymphoma (DLBCL)

and primary mediastinal B-cell lymphoma: ESMO
Clinical Practice Guidelines for diagnosis,
treatment and follow-up†
U. Vitolo1, J. F. Seymour2, M. Martelli3, G. Illerhaus4, T. Illidge5, E. Zucca6, E. Campo7 & M. Ladetto8
on behalf of the ESMO Guidelines Committee*
1
Hematology, University-Hospital Città della Salute e della Scienza, Torino, Italy; 2Peter MacCallum Cancer Centre, and University of Melbourne, Melbourne, Australia;
3
Department of Cellular Biotechnologies and Hematology, University ‘Sapienza’, Rome, Italy; 4Clinic of Hematology, Oncology and Palliative Care, Klinikum Stuttgart,
Stuttgart, Germany; 5Institute of Cancer Sciences, Manchester Academic Health Sciences Centre, Christie Hospital, Manchester, UK; 6Oncology Institute of Southern
Switzerland, Bellinzona, Switzerland; 7Department of Anatomic Pathology, Hospital Clinic, Institut d’Investigacions Biomèdiques August Pi i Sunver (IDIBAPS), University of
Barcelona, Barcelona, Spain; 8Divisione di Ematologia, Azienda Ospedaliera Santi Antonio e Biagio e Cesare Arrigo, Alessandria, Italy
aim of the guidelines guidelines for the clinical management of DLBCLs arising as
primary tumours of the mediastinum, testis, CNS, breast and
clinical practice
The aim of these guidelines is to provide practical clinical bone, whereas nodal DLBCLs have been considered separately [1].
guidelines
guidance and recommendations to clinicians to manage diffuse These subtypes of extranodal DLBCLs represent ∼1%–5% of all
large B-cell lymphomas (DLBCLs) that arise in extranodal NHLs, and this low frequency makes the analysis of disease-specif-
sites. DLBCL may arise in virtually any extranodal sites, but the ic epidemiological factors difficult [3].
majority of these are treated as the nodal counterpart, i.e. gastric
DLBCL. For the recommended treatment, the reader can refer
to the 2015 ESMO DLBCL guidelines [1]. The choice of the diagnosis and pathology/molecular
entities included in these guidelines is based on the clinical need biology
to give recommendations for those entities which require a
The diagnosis of DLBCL should be carried out in a reference
specific therapeutic approach. Recommendations for primary
haematopathology laboratory with expertise in morphological
cutaneous DLBCL have been already reported by ESMO.
interpretation and the facilities to carry out the full range of
phenotypic and molecular investigations. Details on methods
incidence and epidemiology for pathological diagnosis, immunohistochemistry and molecu-
lar biology have been fully reported in ESMO guidelines for
DLBCL is one of the most common lymphoid neoplasms, repre-
nodal DLBCLs [1].
senting 30%–58% of non-Hodgkin’s lymphoma (NHL) and with a
crude incidence in Europe of 3.8/100 000/year. Despite a common
morphology characterised by the diffuse proliferation of mature primary mediastinal large B-cell lymphoma
large B cells, these tumours are clinically and biologically very het- Primary mediastinal large B-cell lymphoma (PMBCL) represents
erogeneous. Most DLBCLs originate in lymph nodes, but ≤40% ∼10% of all DLBCLs and it is more commonly seen in women in
initially present in extranodal sites [2]. Differences in molecular their third to fourth decades of life [3]. These tumours derive from
pathogenesis, clinical presentation and natural history indicate that a medullary thymic B cell and are composed of large cells expres-
extranodal DLBCLs are distinct entities [2]. The most common sing pan B-cell markers, but are negative for surface Ig. The
site of origin is the gastrointestinal tract, but many other organs tumour cells are usually positive for CD23, weakly positive for
and tissues may be involved such as the mediastinum, testis, CD30 and negative for CD15. BCL6, CD10 and IRF4 are expressed
central nervous system (CNS), breast and bone. Here, we provide in a variable number of cases. Some tumours have morphological,
phenotypic and molecular features close to nodular sclerosing
Hodgkin’s lymphoma (HL). These tumours have been included in
*Correspondence to: ESMO Guidelines Committee, ESMO Head Office, Via L. Taddei 4, the World Health Organization (WHO) category of B-cell lymph-
CH-6962 Viganello-Lugano, Switzerland. oma as unclassifiable and with features intermediate between
DLBCL and classical HL [3]. The molecular alterations detected in
†
Approved by the ESMO Guidelines Committee: June 2016. PMBCL target three major mechanisms: NFkB activation (REL

by guest
on 05 February 2018
amplification, inactivating mutations of TNFAIP3), JAK-STAT primary diffuse large B-cell lymphoma of the bone
pathway activation (inactivating mutations of SOCS1, JAK2 ampli- Primary bone lymphoma (PBoL) is a rare category accounting
fication and IL13 overexpression) and modulation of tumour cell for <1% of NHLs and ∼5% of either all primary extranodal
interactions with the microenvironment and immune system NHLs or all bone tumours. It is defined as a lymphoma confined
(down-regulation of HLA class II, PDL-1 and 2 amplifications or to bones and is usually radiologically evident. These tumours
translocations) [4]. may be clinically distinct from primary bone marrow lymph-
omas, in which the tumour cells are restricted to the marrow
primary testicular lymphomas cavity without radiological or histological evidence of cortical
bone destruction [10]. However, this distinction is challenged in
Primary testicular lymphomas (PTLs) are mainly DLBCLs
the present positron emission tomography (PET) era by the in-
(80%–90%) and a minority of other histological subtypes such
creasing recognition of asymptomatic lesions with focal bone
as plasmablastic, Burkitt’s or mantle cell lymphoma and, rarely,
marrow uptake, sometimes but not necessarily accompanied by
low-grade follicular lymphomas (FLs) or T-cell lymphomas [5].
alterations of the bone tissue. They are more common in mid-
Some features of DLBCL (e.g. plasmacytoid differentiation,
elderly men and the vast majority are DLBCLs, although other
somatic hypermutation of immunoglobulin heavy-chain genes
subtypes may also rarely occur. Non-GC phenotype is slightly
and higher frequency of the loss of HLA-DR and DQ expres-
more common than GC. BCL2, BCL6 and MYC rearrangements
sion) suggest the possibility of antigen-driven stimulation. In
may be found in 19%, 14% and 9% of the cases, respectively.
addition, an altered expression of cell surface adhesion mole-
There is insufficient literature and too few reported patient out-
cules may explain the propensity to disseminate to extranodal
comes to enable firm conclusions regarding the prognostic
sites. Cell of origin (COO) studies show an activated B-cell
impact of the COO.
(ABC) pattern in 60%–96% of cases. Mutations of MYD88 were
found in ∼70% of PTLs, compared with <20% in patients with
nodal DLBCLs. Interestingly, 19% of these MYD88-mutated
cases also carry CD79a mutations [6]. Inactivating mutations of
B2M and rearrangements of PDL, CIITA and FOXP1 have also Baseline assessments and procedures do not generally differ
been found in 2%–10% of the cases, emphasising the role of from those required for patients presenting with nodal DLBCLs
tumour microenvironment interactions in the pathogenesis of and include: physical exam, determination of performance
these tumours. status (PS), B symptom assessment, complete blood count,
routine blood chemistry with lactate dehydrogenase (LDH) and
screening tests for human immunodeficiency virus and hepatitis
primary central nervous system lymphomas
B and C (Table 1).
Primary CNS lymphomas (PCNSLs) in immunocompetent Based on the new consensus recommendations for staging
patients represent 2%–3% of all brain tumours. Most of these and restaging of lymphoma developed by the clinical and
cases are DLBCLs and 10%–20% develop intraocular lesions. imaging working groups of the International Conference on
Extraneural dissemination is rare, but clonally related cells have Malignant Lymphomas, fluorodeoxyglucose positron emission
been detected in the blood of some of these patients [7]. tomography/computed tomography (FDG-PET/CT) scan is
Different studies to determine the COO have yielded inconclu- now recommended as standard practice both for staging
sive results with many tumours expressing both germinal centre patients with DLBCL and for response assessment based on the
(GC) markers and the ABC marker IRF4 [7]. Molecular studies visual Deauville criteria (five-point scale, Table 2) [11].
have shown activation of NFkB and BCR/MYD88 pathways and Additional and specific staging procedures are reported in
deletions of the HLA gene locus at 6p21.32, suggesting that, each section for different extranodal sites and are summarised in
similarly to PTLs and PMBCLs, interactions with the micro- Table 1.
environment are important pathogenic mechanisms [7]. The disease staging is established as nodal DLBCL according to
MYD88 mutations have been detected in 30%–75% of cases and the Ann Arbor classification system (Table 3). In extranodal
CD79A/B mutations in up to 45%, with substantial overlap in DLBCL, the discriminating utility of the International Prognostic
the positive cases, so that many carry both mutations [8]. Index (IPI) and age-adapted IPI (aaIPI) is not fully evaluated;
however, these tools should be calculated for prognostic purposes
primary diffuse large B-cell lymphoma of the breast at least to differentiate between localised versus advanced stage
Primary breast lymphoma (PBL) is an uncommon tumour disease [II, A] (Table 4). Specific risk score models have been devel-
representing <1% of NHLs and <0.5% of all breast tumours. oped by the International Extranodal Study Group (IELSG) for
Histologically, these lymphomas are heterogeneous and include PCNSL and PBoL; they are described in each related paragraph
DLBCLs, which are the most common form, mucosa-associated below and are summarised in Table 5 [12] and Table 6 [13].
lymphoid tissue (MALT) lymphomas, FL, Burkitt’s lymphomas
(mainly occurring during pregnancy or lactation) and anaplastic
lymphoma kinase-negative T-cell lymphomas, which are asso-
treatment
ciated with breast implants. Molecularly, primary breast DLBCLs The treatment and recommendations for each entity are detailed
are mainly of ABC-subtype and cytogenetic studies have observed separately in each section. The International Lymphoma Radiation
trisomy 3 and 18 and chromosome 18 translocations involving Oncology Group have recently published comprehensive guide-
IGH/MALT1 as in low-grade MALT lymphomas [9]. lines on the use of radiotherapy (RT) in extranodal lymphomas,
v | Vitolo et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Table 1. Summary of diagnostic work-up
Diagnostic work-up for all entities

Physical examination Peripheral lymph nodes, liver, spleen
Laboratory Blood cell and differential count, renal and liver function, LDH, uric acid
Serology Hepatitis B (including HBsAg, anti-HBs and anti-HBc antibodies) and C, HIV
serology
Imaging CT neck, chest, abdomen and pelvis, PET/CT
Bone marrow Histology
Additional diagnostic work-up for specific entity PTL PCNSL PBL PBoL
Brain MRI Yes Yes Yes Noa

Cytology and flow cytometry of cerebral spinal fluid Yes Yes Yes Noa
Testicular ultrasound Yes Elderly males yes No No
Ocular slit-lamp examination No Yes No No
Involved MRI of bone lesion No No No Yes
LDH, lactate dehydrogenase; HIV, human immunodeficiency virus; CT, computed tomography; PET/CT, positron emission tomography/computed
tomography; PTL, primary testicular lymphoma; PCNSL, primary central nervous system lymphoma; PBL, primary breast lymphoma; PBoL, primary bone
lymphoma; MRI, magnetic resonance imaging.
a
Only if involvement of skull and/or spine.
Table 2. Positron emission tomography (PET) five-point scale Table 4. International prognostic index (IPI)
(Deauville criteria) International prognostic index—IPI
1 No uptake Risk factors Age >60 years
2 Uptake ≤mediastinum Serum LDH > normal
3 Uptake >mediastinum but ≤liver Stage III–IV
4 Moderately increased uptake compared with liver PS 2–4
5 Markedly increased uptake to liver and/or new lesions Extranodal sites >1
Risk categories Low 0–1

Low intermediate 2
Table 3. Ann Arbor staging classification
High intermediate 3
Stage High 4–5
I Involvement of a single lymphatic region (I) or localised Age-adjusted international prognostic index—aaIPI—in patients ≤60 years
involvement of single extralymphatic organ or site (IE)
II Involvement of two or more lymphatic regions on the same side Risk factors Serum LDH > normal
of the diaphragm (II) or localised involvement of a single Stage III–IV
extralymphatic organ or site and of one or more lymphatic PS 2–4
regions on the same side of the diaphragm (IIE) Risk categories Low 0
III Involvement of lymphatic regions on both sides of the Low intermediate 1
diaphragm High intermediate 2
IV Diffuse or disseminated involvement of one or more High 3
extralymphatic organs with or without lymphatic involvement
LDH, lactate dehydrogenase; PS, performance status.
and these guidelines should be referenced in every section regard-

ing the detailed techniques and dose requirements [14].
PMBCL is limited by the age distribution of the disease and its
usual confinement to the mediastinum.
primary mediastinal large B-cell lymphomas FDG-PET/CT scan is mandatory to assess disease extent and
PMBCLs usually present with a bulky tumour in the anterior to obtain better definition of the residual mediastinal masses at
mediastinum, with local compressive symptoms including dys- the completion of treatment, because PMBCL shows universal
pnoea, cough, dysphagia and a superior vena cava syndrome in avidity for [18F]-2-fluoro-2-deoxyglucose.
∼50% of cases. Pleural or pericardial effusions are often present.
Particular extranodal sites such as kidneys, adrenal glands, liver treatment. The combination of rituximab with cyclophosphamide,
and ovaries may be involved, particularly in the setting of doxorubicin, vincristine and prednisolone (R-CHOP) or with
disease recurrence. For prognostic characterisation, the IPI VACOP-B (etoposide, doxorubicin, cyclophosphamide, vincristine,
remains the standard score, but its discriminatory utility in prednisolone and bleomycin)/MACOP-B (methotrexate,
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw175 | v

by guest
on 05 February 2018
The role of consolidative mediastinal RT in the 50% of patients

Table 5. The International Extranodal Lymphoma Study Group with CMR (PET/CT negative) post-chemoimmunotherapy is cur-
(IELSG) prognostic score for PCNSL rently being investigated. The results of the ongoing comparative
Variable Favourable feature Unfavourable feature randomised trial (IELSG-37) (NCT01599559) will determine
(value ‘0’) (value ‘1’) whether consolidative RT could be omitted in patients with a
Age (years) <60 >60 CMR at the post-chemoimmunotherapy PET/CT scan following
R-CHOP-like regimens. Outside of a clinical trial, there are in-
ECOG PS score 0–1 >1
sufficient data to support omission, especially as relapses are dif-
Lactate dehydrogenase Normal Elevated ficult to salvage with poor outcomes.
serum level Based on the good results achieved with rituximab chemo-
Cerebrospinal fluid Normal Elevated therapy combination regimens ± RT, high-dose chemotherapy
protein level followed by autologous stem cell transplantation (HDCT/
Involvement of deep No Yes ASCT) is not recommended in patients who achieved complete
regions of the CNSa remission (CR), even in initially poor-risk patients who attain
an adequate response to initial therapy [III, A].
IELSG score 0–1, low risk; 2–3, intermediate risk; 4–5, high risk. In young patients who do not obtain an adequate response
PCNSL, primary central nervous system lymphoma; ECOG PS, Eastern (i.e. less than partial response) with an elevated FDG uptake at
Cooperative Oncology Group performance status; CNS, central nervous the post-chemoimmunotherapy PET/CT scan, if residual
system. disease is confirmed by biopsy when feasible, an intensification
Reprinted from [12] with permission. @2003 American Society of therapy with HDCT/ASCT is recommended [III, B].
Clinical Oncology. All rights reserved.
a
Deep regions include the corpus callosum, basal ganglia, brain stem and
post-treatment evaluation. A PET/CT scan in PMBCL has an
cerebellum.
excellent negative predictive value, but it also has a low positive
predictive value due to the high frequency of false-positive
scans. Thus, positive PET scans require further investigation
Table 6. The International Extranodal Lymphoma Study Group before modifying planned therapy. Several patients with
(IELSG) staging system for DLBCL of the bone persistent metabolically active masses underwent biopsies,
Lymphoma extension IELSG which showed necrosis but no lymphoma. False increased FDG
Stage uptake may be due to an inflammatory response produced
by the addition of rituximab to chemotherapy or a thymic
Single bony lesion IE
rebound, which is particularly relevant, given the location of the
Single bony lesion with involvement of regional lymph IIE
disease [17]. Thus, positive PET scans require further investigation
nodes
Multifocal disease in a single bone or lesions in multiple IVE
before modifying planned therapy. A post-treatment PET/CT
bones in a disease exclusively limited to the skeleton
response evaluation should not be carried out until at least 5–6
(without lymph nodal or visceral disease)a weeks from the last infusion of chemotherapy to minimise the
Disseminated lymphoma with at least one bony lesion IV incidence of these false-positive scans. Serial scanning may be
required to fully evaluate areas of residual post-treatment PET
Reprinted from [13], with permission from Elsevier. tracer uptake, with many patients manifesting gradual resolution.
a
Also called multifocal osteolymphoma or polyostotic lymphoma.
treatment of relapsed/refractory PMBCL. The probability of
recurrence after successful initial therapy for PMBCL appears to
be lower than for other DLBCLs, and those patients who achieve
doxorubicin, cyclophosphamide, vincristine, prednisolone and a response lasting longer than 18 months are likely to be cured.
bleomycin) (R-V/MACOP-B), dose-dense CHOP (R-CHOP14) or Salvage treatment strategies of relapsed/resistant patients are
more intensive chemotherapy regimens such as DA-EPOCH-R similar to nodal DLBCLs, and include attempting reinduction
(dose-adjusted etoposide, prednisone, vincristine with non-cross-resistant agents followed by consolidation with
cyclophosphamide, doxorubicin and rituximab) are the current HDCT/ASCT in patients with chemosensitive disease [II, A].
standard treatments for PMBCLs and any of them can be used If radiation has not been a component of initial therapy, RT
according to the centre’s experience [III, A] [15, 16]. should be incorporated into the salvage treatment programme
Consolidative mediastinal RT is recommended in responding for patients with relapsed disease, ideally post-transplant if sig-
patients treated with standard-dose chemoimmunotherapy (R- nificant mediastinal or lung volumes are involved.
CHOP/R-V/MACOP-B) [II, A] [16].
Based on the results of a small phase II study, consolidative medi-
astinal RT could be omitted in patients with a complete metabolic primary testicular lymphomas
response (CMR) only after DA-EPOCH-R [III, C] [15]. However, PTLs usually present as a limited stage disease with most patients
these results need to be validated in a larger, multicentre series. having a unilateral testicular mass, although bilateral testicular in-
Where used, consolidative mediastinal RT should be adminis- volvement occurs in ∼10%. There is a propensity for an aggres-
tered with doses in the range 30–36 Gy, in 1.5–2.0 Gy fractions sive clinical behaviour with dissemination to multiple extranodal
[II, A]. sites, such as lung, pleura, skin (up to 35%), soft tissues and
v | Vitolo et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Waldeyer’s ring (5%). CNS relapses are frequent and occur in up treatment of relapsed/refractory PTL. Standard therapeutic
to 30% of PTL patients within 1–2 years of diagnosis, but can option for these patients has not yet been defined in prospective
also be delayed in presentation [18]. For staging, in addition to trials. HDCT/ASCT, if feasible, should be the preferred treatment
the standard tests, ultrasound of the contralateral testis, brain strategy in chemosensitive relapse as in nodal DLBCL [I, A] [22].
nuclear magnetic resonance imaging (MRI) and diagnostic However, many PTL patients are elderly, with poor PS and multi-
lumbar puncture with cytological and flow cytometric analysis on organ dysfunction; the inclusion of these patients in clinical trials
cerebrospinal fluid (CSF) are mandatory. Orchiectomy remains a testing novel drugs is advisable.
mandatory step in the management of PTL, both for diagnostic
and for therapeutic aims [III, A] [19].
primary CNS lymphomas
treatment. In the IELSG retrospective series, chemotherapy PCNSLs are rare diseases defined by involvement of the cerebral
with anthracycline-containing regimens significantly improved parenchyma, leptomeninges, eyes or spinal cord without evi-
the outcome in all patients [19]. Despite the lack of randomised dence of systemic disease.
studies, some retrospective studies and a prospective one Patients with PCNSL usually present with focal neurological
(IELSG10) provided evidence of the benefit of the addition of deficits and/or neuropsychiatric symptoms. Symptoms of
rituximab to chemotherapy [20]. Six to eight cycles of R-CHOP increased intracranial pressure and seizures are less frequent
given every 21 days is the current standard [II, A]. [12]. Approximately 10%–20% of patients have ocular involve-
If orchiectomy is not carried out, RT is given to the involved ment, with floaters, blurred vision, diminished visual acuity and
testis. The addition of RT to the contralateral testis significantly painful red eyes. About one-third of patients present with multi-
reduces or abrogates testicular relapses in retrospective and pro- focal neurological disease, with highly diverse deficits.
spective studies [19–21]. Hence, prophylactic RT to the contra- Stereotactic biopsy is recommended as a minimal invasive
lateral testis and scrotum is strongly recommended [III, A]. The technique to obtain histopathological diagnosis [III, A]; therefore,
dose to the testis should be 25–30 Gy in 1.5–2 Gy fractions. resection of PCNSL cannot be recommended, because of the
Such treatment to the remaining testis is associated with a sub- associated morbidity and lack of evidence for benefit [23].
stantial risk of hypogonadism; therefore, sequential monitoring Glucocorticoid therapy can interfere with an accurate histopatho-
of testosterone levels and replacement therapy are clinically im- logical diagnosis; therefore, it should be avoided before biopsy if
portant and should be included in the follow-up programme for possible without clinically compromising the patient [24].
these patients. RT was given in the past to involved abdomino- At diagnosis, patients should undergo contrast-enhanced brain
pelvic nodes in stage IIE disease; however, the lack of well- MRI as the standard for baseline and response assessment, as
conducted trials makes it difficult to assess the role of nodal RT well as cytological evaluation and flow cytometry of the CSF. In
in this setting. Indeed, also in nodal DLBCL, with the introduc- addition, slit-lamp examination should be carried out to investi-
tion of PET reassessment after R-CHOP, consolidation RT is gate possible ocular involvement. CT scans of the chest, abdomen
less frequently used in patients who achieved a PET-negative and pelvis or a PET/CT scan is carried out to exclude systemic
CR. Based on these indirect data and considering the increased disease. Additionally, testicular ultrasound examination in elderly
toxicity, RT to involved nodes could be safely omitted if PET male patients is recommended [V, B] [23]. In PCNSL, the value
negative after R-CHOP chemotherapy [III, C], as in the ongoing of a bone marrow examination is controversial; however, it is still
IELSG30 phase II study (NCT00945724). part of baseline investigations in current PCNSL trials.
Although some conflicting data exist, a prophylactic therapy to The IELSG has proposed a prognostic system for PCNSL
reduce the risk of CNS relapse should be added to the treatment based on the following independent prognostic factors: age >60
[III, B]. Two options can be considered: intrathecal chemotherapy years, Eastern Cooperative Oncology Group (ECOG) PS >1,
with methotrexate (MTX) is the most common one; alternatively, elevated serum LDH, elevated CSF protein concentration
intravenous systemic MTX can be employed as part of overall treat- and tumour localisation within the deep regions of the brain.
ment [III, B] [5]. The best strategy to further reduce CNS recurrence Three risk classes are defined for patients with 0–1 (good-risk),
remains an open issue. The ongoing IELSG30 phase II study that 2–3 (intermediate-risk) or 4–5 ( poor-risk) prognostic factors
includes both intrathecal chemotherapy with liposomal cytarabine (Table 5) [12]. Another validated risk score developed by Abrey
and systemic prophylaxis with a lower dose of MTX (1.5 g/m2) to et al. [25] only requires age and clinical PS to build three differ-
spare toxicity in elderly PTL patients aims at addressing this issue. ent risk groups good risk ( patients <50 years), intermediate risk
In summary, the recommended treatment for stage I–II PTL ( patients ≥50 years and Karnofsky performance score [KPS]
consists of R-CHOP21 × 6–8 courses, with the addition of CNS ≥70) and high risk ( patients ≥50; KPS <70).
prophylaxis and prophylactic RT to the contralateral testis.
The therapy of advanced-stage PTLs is not different from the first-line treatment. Historically, whole-brain radiotherapy
standard treatment of advanced-stage nodal DLBCLs. Unlike (WBRT) has been the standard treatment for PCNSL. Despite
nodal DLBCLs, two treatment issues should be incorporated in the high CR rate, the disease relapses in almost all patients after
the therapeutic strategy: prophylactic testicular RT and CNS a few months, with a median survival of 12 months and few
prophylaxis [18]. long-term survivors. The addition of chemotherapy to WBRT
has been recommended to improve survival of PCNSL patients.
post-treatment evaluation. As in nodal DLBCLs, FDG-PET/CT High-dose MTX (HD-MTX; at least 3 g/m² over 2–4 hours
can be considered the recommended standard for post-treatment rapid infusion time) is considered the most effective single agent
assessment in PTL. for treating PCNSL [25]; however, based on prospective studies,
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw175 | v

by guest
on 05 February 2018
CR rates are usually <30% with single-agent, high-dose MTX. routine component of front-line therapy in elderly patients able
The addition of high-dose cytarabine (HD-ara-C) has been to tolerate other therapies, due to its high rate of neurotoxicity,
shown to significantly improve response rate (RR) and failure- and is reserved for frail patients or patients with refractory
free survival, compared with high-dose MTX alone in a phase II disease [14].
randomised trial [26]. Other CHOP-like protocols were asso- In summary, HD-MTX-based regimens, in combination with
ciated with unsatisfactory results in PCNSL [26]. Several single- HD-ara-C, are recommended for induction treatment if possible
arm studies have investigated other drugs in combination with based on patient age, PS and organ function [I, A]. HD-MTX-
HD-MTX without clear additional benefit, including ifosfamide, based treatment should be considered whenever possible in
thiotepa, procarbazine, vincristine and temozolomide [27]. elderly patients with PCNSL [II, A].
The role of prophylactic intrathecal chemotherapy is unclear. The use of WBRT as consolidation after HD-MTX-based
Retrospective studies do not suggest a benefit from adding intra- chemotherapy is still commonly applied in young patients and
thecal chemotherapy combined to systemic high-dose MTX. On generally avoided in those >60 years, as the risk of neurotoxicity
the other hand, other non-randomised studies suggested that (including cognitive deterioration) is particularly high in elderly
intraventricular treatment combined with an MTX-based chemo- patients [III, B].
therapy was effective in reducing relapse rate [28]. Thus, the ef- Thus, we recommend that consolidation RT should be
fectiveness of intrathecal therapy is contradictory and may avoided in elderly patients who achieved CR [III, B], while in
different according to the route of administration. In summary, those not reaching CR, it is preferable to use lower doses of
intrathecal chemotherapy cannot be routinely recommended WBRT (36 or 23.4 Gy) based on response [III, B].
if appropriate systemic chemotherapy is applied, but could be Whether RT can safely be omitted without compromising
considered in the case of severe meningeal involvement [IV, C] long-term outcome also in younger patients in CR remains con-
[24, 29]. troversial [III, C]. HDCT/ASCT as consolidation treatment
RT with doses between 40 and 45 Gy in 20–25 fractions is pri- could be considered as an alternative to WBRT in eligible
marily used as consolidative therapy following HD-MTX-based patients [III, B].
chemotherapy. Higher RT doses following chemotherapy are
associated with an increased risk of neurotoxicity, which is par- post-treatment evaluation. Brain MRI is the standard for
ticularly pronounced in the older age group (30% of all patients response evaluation. For patients with previous CSF involvement,
and 40%–50% of those >60 years of age). This neurotoxicity led CSF analysis (cytology and flow cytometry) should be added, as
to testing lower RT doses, as low as 30–36 Gy for WBRT or well as spinal imaging as clinically indicated. In the follow-up,
23.4 Gy after chemotherapy-induced CR [27]. These regimens brain MRI monitoring may be helpful to rule out progression,
appeared to be effective and without prohibitive RT-related tox- but the overall clinical utility of such a surveillance strategy is
icity, even in elderly patients [27, 29–31]. unproven. For patients with prior CSF or ocular involvement,
HDCT/ASCT represents an alternative strategy for consolida- specific assessments (i.e. CSF evaluation or ophthalmologic
tion treatment aimed at sparing neurotoxicity, avoiding RT and evaluation) are needed as clinically indicated. Based on the rarity
improving outcome. HDCT/ASCT has been used as part of of systemic progressions, additional systemic evaluation (i.e. CT
front-line treatment in PCNSL patients with encouraging or CT/PET scans) are indicated only in the case of clinical signs
results, mostly using thiotepa-based conditioning regimens with or symptoms of systemic progression.
overall remission rates of up to 91% and a 5-year overall survival
(OS) of 87% for those who completed HDCT [32]. Another ap- treatment of relapsed/refractory PCNSL. HDCT/ASCT has
proach to avoid WBRT is consolidation with dose-intensive been prospectively investigated in patients with disease relapse
conventional chemotherapy, such as etoposide and ara-C after after HD-MTX with a median survival of 18.3 months (58.6
HD-MTX-based induction [24]. months for those who received HDCT/ASCT) [36]. Only a few
Recently, the preliminary results of IELSG32 randomised agents have been investigated prospectively in the relapse/
study (NCT01011920, Eudra CT number 2009-014722-42) sug- refractory setting, namely temozolomide [36], temozolomide
gested a benefit of the addition of rituximab and thiotepa in plus rituximab [37], topotecan and pemetrexed. However, none
newly diagnosed PCNSL with regard to RR, progression-free of these showed convincing activity sufficient to be established
survival (PFS) and OS [33]. This is also supported by retrospective as standard in this situation. Re-exposure to the same HD-
series investigating combination of rituximab with HD-MTX. MTX-based protocol can be considered; however, there are no
data on the relationship between efficacy and the duration of
elderly patients (age >70 years or patients with comorbidities). prior response. When RT is used as a single modality therapy, i.e.
Selected elderly patients are often able to tolerate HD-MTX- WBRT as primary treatment for non-candidates for chemotherapy,
based treatment, with dosage guided by renal function [24]. higher doses (40–50 Gy, 1.5–1.8 Gy/fraction) are required. Doses of
Based on a recent systematic review and individual patient data 36 Gy have been shown to be beneficial in a salvage setting, but the
meta-analysis, HD-MTX-based treatment was associated with optimal dose and the role of boost doses remain uncertain [38].
an improved survival outcome [34]. Further combinations with In summary, choice of salvage treatments depends on the
oral alkylating agents such as lomustine (CCNU) and clinical status, toxicities from previous treatments and duration
procarbazine ± vincristine are feasible, active and are warranted of remission. Based on the limited evidence available so far, no
to be compared with HD MTX alone [31, 35]. Overall, HD standard protocol can be recommended. Fit patients should
MTX-based treatment should be considered whenever possible be considered for HDCT/ASCT [III, B]. WBRT remains an im-
in elderly patients with PCNSL [II, A]. WBRT should not be a portant palliative treatment in PCNSL lymphoma for those
v | Vitolo et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
unable to tolerate or relapsing after high-dose, CNS-penetrating stratification by risk, abbreviating the number of chemotherapy
chemotherapy and not fit for further chemotherapy [III, B]. cycles to fewer than 4 has resulted in inferior outcomes [39, 41].
Although there are no randomised studies addressing the ques-
tion, most historical comparisons suggest some favourable
primary breast lymphoma
impact from the incorporation of rituximab, and given its estab-
PBL is typically a disease of older women (median age 62–64 lished role in DLBCL treatment, it should be routinely included
years), but with lower median ages in specific geographic areas in front-line chemoimmunotherapy as part of R-CHOP q 21d
(i.e. in East Asian countries) [39, 40]. PBL almost exclusively [III, B]. There are no data supporting the use of more aggressive
affects women, with very rare male cases reported, but with or dose-escalated regimens as part of primary therapy [III, B].
apparently similar outcomes [41]. The typical presentation is a Whole-breast RT is generally recommended after chemother-
painless breast mass (median 4 cm diameter), slightly more fre- apy based on the benefit in OS observed in the large IESLG
quently on the right side and with systemic symptoms present study [41], but some experts consider partial breast RT in cases
in <5% of patients, almost always those with disseminated where the resulting morbidity reduction is likely to offset the po-
disease [42]. tential increase in marginal relapse rate, especially where PET
Currently, up to 20% of patients are diagnosed incidentally scanning has been used in initial staging to allow accurate delin-
[42, 43], but there are no definitive radiologic features of PBL; eation of the involved area and nodal status [14]. The unin-
therefore, a diagnostic biopsy is mandatory [IV, A]. volved lymph nodes need not be included in the RT volume,
The specific additional staging procedures recommended for provided initial staging has used PET [14].
PBL include contralateral breast examination for potential bilat-
eral involvement, which would usually be dealt with adequately
treatment recommendations. The recommended treatment for
by whole-body PET/CT. CNS imaging with MRI of the brain
PBL is as follows: 6 cycles of R-CHOP, if therapy is well
and CSF analysis by cytology and flow cytometry should also
tolerated [III, A], followed by consolidative whole ipsilateral
be carried out even in the absence of CNS symptoms [IV, B].
breast RT (30–36 Gy) [III, B]. CNS-directed prophylaxis
Approximately 70% of patients have stage IE disease, with the
should be considered for individual patients [III, B], and is
other 30% having regional nodal involvement (stage IIE) [41].
recommended in high-risk patients, i.e. bilateral involvement [III,
From 4% to 13% of patients have bilateral breast involvement at
A]. No specific CNS prophylaxis treatment can be recommended
presentation [41].
due to the lack of data; in nodal DLBCL, either intrathecal or i.v.
The standard IPI is predictive of outcome, but this has low
MTX could be appropriate [48]. Patients with bilateral breast
discriminatory ability, while the stage-modified IPI has a better
involvement represent a particularly high-risk group, and
prognostic discrimination, mainly through the adverse impact
investigation of more intensive chemotherapy regimens in the
of stage IIE disease [41, 44]. Other reproducible adverse prog-
context of clinical trials is justified in such cases [IV, C].
nostic factors among patients with localised disease include
tumour size >4–5 cm [44, 45] and bilateral involvement that
was found to be a significant adverse predictor of both PFS and post-treatment evaluation. As in nodal DLBCL, FDG PET/CT
OS, with a high risk of CNS relapse [41, 43]. can be considered the recommended standard for post-treatment
Reminiscent of PTL, PBL displays extranodal tropism at assessment in PBL (Table 2). The Lugano 2014 criteria are also
relapse [46]. The ipsilateral and contralateral breasts were applied to PBL.
involved at relapse in 12%–44% of series with available data
[40], with ipsilateral breast relapse occurring in the first few treatment of relapsed/refractory PBL. The outcome of patients
years following treatment, while in contrast, contralateral breast after relapse is poor; in the IELSG-15 series, the median survival
relapse occurred up to 13.3 years later [41]. Other extranodal after relapse was 1.0 year, with a 5-year OS of 20% [41]. The
sites including the bone marrow, lung or pleura, skin, gastro- management strategy should be as per other instances of
intestinal tract and CNS have all been reported to occur more relapsed DLBCL, with reinduction chemoimmunotherapy
commonly in PBL than among nodal DLBCLs [39, 44]. aiming to proceed to HDCT/ASCT, where feasible, in patients
From larger series, CNS relapse developed in 5%–16% of with responsive disease [IV, B].
patients, with a weighted average of ∼12%. Other than poten-
tially bilateral disease, there are no reproducible predictors of primary bone lymphoma
CNS relapse risk. PBoL may present as a single bone lesion with or without an
associated soft tissue mass arising from local extension and with
treatment. Surgical resection as a sole modality results in or without regional lymphadenopathy or as a multifocal polyos-
inadequate local control, and the application of mastectomy is totic disease exclusively involving the skeleton [13].
associated with inferior outcomes [41, 44]. In the rare instance The median age at diagnosis varies from 45 to 60 years.
of patients who have undergone surgical excision, they should Patients with PBoL typically present with bone pain (80%–
still be managed identically as recommended below [V, B]. 95%), a tumour mass in 30%–40% of the cases and pathological
The retrospective IELSG-15 study reinforced the importance fracture, most frequently of the humerus, in 10%–15% of
of an anthracycline-containing chemotherapy regimen with a patients. Approximately 15% of patients show spinal cord com-
strongly favourable impact on both PFS and OS. Most recent pression and 10% have hypercalcaemia, usually in the presence
contemporary series have used CHOP ± rituximab and RT with of rapidly progressive disease. The femur, the pelvic bones and
5-year PFS and OS of 50%–70% [41, 42, 47]. In cohorts without the spine are the most common sites of involvement [13, 49].
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw175 | v

by guest
on 05 February 2018
Most patients have early-stage disease at diagnosis [13], and response assessment and post-treatment evaluation. Post-
PET/CT has increased the detection of asymptomatic bone lesions treatment evaluation is difficult because bone lesions remain
as part of the initial presentation of disseminated DLBCL [49]. In detectable upon CT scan. PET scan is mandatory in response
comparison with CT scan, MRI better defines the local extent of assessment as in nodal DLBCL; however, residual PET uptake
the disease and cortical changes. An adapted staging system for can persist and may resolve slowly, representing bone healing
PBoL has been proposed by the IELSG (Table 6) [13, 50]. rather than active disease. New treatments such as HDCT/ASCT
The prognosis of primary bone DLBCLs mainly depends on or alternative chemoimmunotherapy are only recommended in
the disease extent: 5-year OS rates vary from >80% for stage IE the case of biopsy-proven persisting disease or clear clinical or
to <40% for disseminated DLBCL with bone localisation [51]. radiological progression [III, A].
The role of the IPI in predicting prognosis of primary DLBCL of Finally, long-term bone health preventive measures should
the bone seems limited [13, 49, 51]. also be taken into account in patients with PBoL, including
evaluation and treatment of any underlying osteoporosis, and/
treatment. Primary bone DLBCL should be treated with or vitamin D deficiency.
anthracycline-containing chemotherapy regimens together with
rituximab, although the benefit of the addition of rituximab has
not formally studied specifically in the subset of patients with follow-up
PBoL. The role of consolidation RT is not well defined because
the available data are very controversial and mainly come from Follow-up monitoring is not different from that of nodal
retrospective analyses that suggest benefit [51, 52]. Indeed, R- DLBCL [1].
CHOP ± consolidation RT remains the standard approach for
the patients with any stage of DLBCL with bone involvement
[III, B]. Whether RT to sites of bone involvement is truly personalised medicine
needed or can be spared (at least in cases with a negative PET There are still many open issues in the treatment of extranodal
scan after chemoimmunotherapy) should be addressed in DLBCL. New agents targeting distinct molecular pathways
appropriately designed prospective randomised trials [51]. involved in disease pathogenesis are in progress and are being
When consolidation RT is given to localised lesions, there is tested in ongoing trials. So far, none of these agents is appropri-
limited additional benefit from the use of extensive radiation ate for standard clinical practice.
volumes and RT and dose range is 30–40 Gy, depending on the For instance, in PMBCLs based on some molecular alterations
certainty that a CR has been obtained with chemotherapy [13]. detected such as JAK-STAT pathway activation and modulation
The risk of CNS recurrence associated with skeletal involve- of tumour cell interactions with the microenvironment and
ment is also a matter of debate [13]. In the IELSG-14 study, con- immune system, the role of biological drugs targeting selective
ducted in the pre-rituximab era, CNS involvement (mainly pathways as JAK-STAT and PDL1, PDL2 should be evaluated.
meningeal) occurred in 2.5% of patients with localised primary PTLs are predominantly of ABC subtype, and lenalidomide
bone DLBCL [51]. Hence, CNS prophylaxis is not routinely and ibrutinib, which preferentially target this DLBCL subtype
required in these patients [III, B]. However, an accurate assess- and have been proved to be effective in nodal ABC DLBCL, may
ment (CSF flow cytometry and brain MRI) and prophylaxis can be tested in PTL patients in an effort to improve their outcome
be recommended in patients with involvement of anatomic [53, 54]. Moreover, both PTLs and PCNSL show a high fre-
areas in close apposition to the CNS (skull and/or spine) [III, quency of MYD88 and CD79a mutations [6] and ibrutinib that
B]. CNS recurrence occurs in 7% of these patients, particularly targets this pathway may be active in these patients. Recently,
those with disseminated DLBCL and other high-risk features, ibrutinib was reported to cross the blood–brain barrier and have
and decisions regarding CNS prophylaxis should be made on potential activity in PCNSL.
the basis of established risk profiling [48].
The presence of pathological fracture at presentation is asso-
ciated with an inferior outcome in PBoL in the IELSG14 study
[49]. The initial surgical stabilisation of the pathological fracture
methodology
should be directed at achieving a better quality of life (e.g. These clinical practice guidelines were developed in accordance
control pain, prevent bone displacement, allow weight-bearing); with the ESMO standard operating procedures for clinical
however, there is no evidence that initial surgical stabilisation practice guidelines development, www.esmo.org/Guidelines/
improves the lymphoma outcome, with a 5-year OS of 45% and ESMO-Guidelines-Methodology. The relevant literature has
54%, respectively, for patients who either did or did not receive been selected by the expert authors. A summary of common
surgery. These data suggest that initial surgery should be consid- and specific diagnostic procedures for each extranodal DLBCL
ered only if chemotherapy delays can be avoided [IV, C] [49]. type is shown in Table 1. Table 7 summarises the recommenda-
Patients with a pathological fracture should be initially tions for first-line treatment. A summary of recommendations is
managed similarly to standard PBoL. Consolidation RT (30–40 shown in Table 8. Levels of evidence and grades of recommen-
Gy) to the fractured bone may be given [IV, C]. Indeed, the dation have been applied using the system shown in Table 9.
initial RT of a fractured bone before chemotherapy does not Statements without grading were considered justified standard
seem to improve disease control or OS when compared with the clinical practice by the experts and the ESMO faculty. This
chemoradiotherapy sequence but may improve pain and poten- manuscript has been subjected to an anonymous peer review
tially hasten healing [13]. process.
v | Vitolo et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Table 7. Recommended first-line treatment strategies in extranodal diffuse large B-cell lymphoma (DLBCL)
Primary sites Treatment Consolidation CNS prophylaxis
Primary testicular R-CHOP21X6-8 RT to contralateral testis (25–30 Gy) IT MTX or i.v. systemic
lymphoma MTX
Primary central HD-MTX (MTX ≥ 3 g/m2) plus HD-ara-C WBRT is not routinely recommended N/A
nervous lymphoma HD-MTX-based (adjusted dose on ECOG HDCT/ASCT suggested in young patients
PS, renal function, etc.) in elderly patients (clinical trial)
Primary mediastinal R-CHOP or R-V/MACOP-B or R-CHOP14 Mediastinal RT (30 Gy) in responding patients; RT Not recommended
lymphoma or DA-EPOCH-R could be omitted in CMR only after DA-EPOCH-R
HDCT/ASCT is not recommended in CR1
Primary breast R-CHOP21X6 Whole ipsilateral breast RT (30–36 Gy). Partial To be considered in all
lymphoma breast RT in selected cases (see text) patients
Mandatory in high risk
Primary bone R-CHOP21X6-8 RT (30–40 Gy) to involved bone Only if involvement of the
lymphoma RT before chemotherapy is not skull and/or spine
recommended
CNS, central nervous system; R-CHOP21, cyclophosphamide, doxorubicin, vincristine and prednisone treatment combined with rituximab given every 21
days; RT, radiotherapy; IT, intrathecal; MTX, methotrexate; i.v., intravenous; HD-MTX, high-dose methotrexate; HD-ara-C, high-dose cytarabine; ECOG
PS, Eastern Cooperative Oncology Group performance status; WBRT, whole-brain radiotherapy; HDCT/ASCT, high-dose chemotherapy followed by
autologous stem cell transplantation; R-CHOP, rituximab, cyclophosphamide, doxorubicin, vincristine, prednisolone; R-V/MACOP-B, rituximab,
etoposide, doxorubicin, cyclophosphamide, vincristine, prednisolone, bleomycin)/rituximab, methotrexate, doxorubicin, cyclophosphamide, vincristine,
prednisolone, bleomycin; R-CHOP14, cyclophosphamide, doxorubicin, vincristine and prednisone treatment combined with rituximab given every 14
days; DA-EPOCH-R, dose-adjusted etoposide, prednisone, vincristine cyclophosphamide, doxorubicin and rituximab; CMR, complete metabolic response;
CR1, first complete remission.
The disease staging is established as nodal DLBCL according to the Ann Arbor system. In extranodal DLBCL, the discriminating utility of the IPI and aaIPI
is not fully evaluated; however, they should be calculated for prognostic purposes at least to differentiate between localised versus advanced stage disease
[II, A]
PMBCLs
• R-CHOP, R-VACOP-B, V-MACOP-B, R-CHOP14 or more intensive chemotherapy regimens such as DA-EPOCH-R should be considered the current
standard treatment for PMBCLs; any of them can be used according to the centre’s experience [III, A]
• Consolidative mediastinal RT is recommended in responding patients treated with standard dose chemoimmunotherapy (R-CHOP/R-V/MACOP-B)
[II, A]
• Consolidative mediastinal RT could be omitted in patients with a CMR only after DA-EPOCH-R [III, C]
• Where used, consolidative mediastinal RT should be administered with doses in the range 30–36 Gy, in 1.5–2.0 Gy fractions [II, A]
• HDCT/ASCT is not recommended in patients who achieved CR, even in initially poor-risk patients who attain an adequate response to initial therapy
[III, A]
• In young patients who do not obtain an adequate response (i.e. less than partial response) with an elevated FDG uptake at the post-
chemoimmunotherapy PET/CT scan, if residual disease is confirmed by biopsy when feasible, an intensification therapy with HDCT/ASCT is
recommended [III, B]
• Salvage treatment strategies of relapsed/resistant patients include attempting reinduction with non-cross-resistant agents followed by consolidation with
HDCT/ASCT in patients with chemosensitive disease [II, A]
PTLs
• Orchiectomy remains a mandatory step for diagnostic and therapeutic aims [III, A]
• Six to eight cycles of R-CHOP given every 21 days is the current standard [II, A]
• Prophylactic RT to the contralateral testis and scrotum is strongly recommended [III, A]
• RT to involved nodes could be safely omitted if PET is negative after R-CHOP chemotherapy [III, C]
• A prophylactic therapy to reduce the risk of CNS relapse such as intrathecal chemotherapy with (MTX or intravenous systemic MTX should be added to
the treatment) [III, B]
• HDCT/ASCT, if feasible, should be the preferred treatment strategy in chemosensitive relapse [I, A]
Continued
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw175 | v

by guest
on 05 February 2018
Table 8. Continued
PCNSLs
• Stereotactic biopsy is recommended as a minimal invasive technique to obtain a histopathological diagnosis [III, A]; therefore, resection of PCNSL
cannot be recommended
• Additionally, testicular ultrasound examination in elderly male patients is recommended [V, B]
• HD-MTX-based regimens, in combination with HD-ara-C, are recommended for induction treatment if possible based on patient age, PS and organ
function [I, A]
• HD-MTX-based treatment should be considered whenever possible in elderly patients [II, A]
• Intrathecal chemotherapy cannot be routinely recommended if appropriate systemic chemotherapy is applied and could be considered in the case of
severe meningeal involvement [IV, C]
• WBRT as consolidation after HD-MTX-based chemotherapy is still commonly applied in young patients and generally avoided in those >60 years, as the
risk of neurotoxicity (including cognitive deterioration) is particularly high in elderly patients [III, B]
• Consolidation RT should be avoided in elderly patients who achieved CR [III, B], while in those not reaching a CR, it is preferable to use lower doses of
WBRT (36 or 23.4 Gy) based on response [III, B]
• Whether RT can safely be omitted without compromising long-term outcome also in younger patients in CR remains controversial [III, C]
• HDCT/ASCT as consolidation treatment could be considered as an alternative in eligible patients [III, B]
• For salvage treatments, no standard protocol can be recommended. Fit patients should be considered for HDCT/ASCT [III, B]
• WBRT remains an important palliative treatment in PCNSL lymphoma for those unable to tolerate or relapsing after high-dose, CNS-penetrating
chemotherapy and not fit for further chemotherapy [III, B]
PBL
• Diagnostic biopsy is mandatory [IV, A]

• In the rare instance of patients who have undergone surgical excision, they should still be managed identically as recommended below [V, B]
• Incorporation of rituximab should be routinely included in front-line chemoimmunotherapy as part of R-CHOP q 21d [III, B]
• The recommended treatment is 6 cycles of R-CHOP [III, A], followed by consolidative whole ipsilateral breast RT (30–36 Gy) [III, B]
• More aggressive or dose-escalated regimens as part of primary therapy are not indicated [III, B]
• CNS-directed prophylaxis should be considered for individual patients [III, B], it is recommended in high-risk patients, i.e. bilateral involvement [III, A]
• Patients with bilateral breast involvement represent a particularly high-risk group, and investigation of more intensive chemotherapy regimens in the
context of clinical trials is justified in such cases [IV, C]
• After relapse, the management strategy should be as for other instances of relapsed DLBCL, with reinduction chemoimmunotherapy aiming to proceed
to HDCT/ASCT, where feasible, in patients with responsive disease [IV, B]
PBoL
• R-CHOP ± consolidation RT (30–40 Gy) remains the standard approach for the patients with any stage of DLBCL with bone involvement [III, B]
• CNS prophylaxis is not routinely required in these patients [III, B]; however, an accurate assessment (CSF flow cytometry and brain MRI) and
prophylaxis can be recommended in patients with involvement of anatomic areas in close apposition to the CNS (skull and/or spine) [III, B]
• Patients with a pathological fracture should be initially managed similarly to standard PBoL. Consolidation RT (30–40 Gy) to the fractured bone may be
given [IV, C]
• Initial surgery should be considered only if chemotherapy delays can be avoided [IV, C]
New treatments such as HDCT/ASCT or alternative chemoimmunotherapy are only recommended in the case of biopsy-proven persisting disease or clear
clinical or radiological progression [III, A]
DLBCL, diffuse large B-cell lymphoma; IPI, international prognostic index; aaIPI, age-adapted IPI; PMBCLs, primary mediastinal large B-cell lymphomas;
R-CHOP, rituximab, cyclophosphamide, doxorubicin, vincristine and prednisolone; R-VACOP-B; rituximab, etoposide, doxorubicin, cyclophosphamide,
vincristine, prednisolone, bleomycin; R-MACOP-B, rituximab, methotrexate, doxorubicin, cyclophosphamide, vincristine, prednisolone, bleomycin; R-
CHOP14, rituximab, cyclophosphamide, doxorubicin, vincristine, prednisolone every 14 days; DA-EPOCH-R, dose-adjusted etoposide, prednisone,
vincristine cyclophosphamide, doxorubicin and rituximab; RT, radiotherapy; CMR, complete metabolic response; HDCT/ASCT, high-dose chemotherapy
followed by autologous stem cell transplantation; CR, complete remission; FDG, fluorodeoxyglucose; PET/CT, positron emission tomography/computed
tomography; PTLs, primary testicular lymphomas; CNS, central nervous system; MTX, methotrexate; PCNSLs; primary central nervous system
lymphomas; HD-MTX, high-dose methotrexate; HD-ara-C, high-dose cytarabine; PS, performance status; WBRT, whole-brain radiotherapy; PBL, primary
diffuse large B-cell breast lymphoma of the breast; PBoL, primary diffuse large B-cell bone lymphoma of the bone; CSF, cerebral spinal fluid; MRI, magnetic
resonance imaging.
v | Vitolo et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Table 9. Levels of evidence and grades of recommendation and Takeda. GI, EC and MM have declared no potential con-
(adapted from the Infectious Diseases Society of America-United flicts of interest.
States Public Health Service Grading System)a
Levels of evidence references
I Evidence from at least one large randomised, controlled trial of 1. Tilly H, Gomes da Silva M, Vitolo U et al. Diffuse large B-cell lymphoma (DLBCL):
good methodological quality (low potential for bias) or meta- ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann
analyses of well-conducted randomised trials without Oncol 2015; 26(Suppl. 5): v116–v125.
heterogeneity 2. Møller MB, Pedersen NT, Christensen BE. Diffuse large B-cell lymphoma: clinical
II Small randomised trials or large randomised trials with a implications of extranodal versus nodal presentation—a population-based study of
suspicion of bias (lower methodological quality) or meta- 1575 cases. Br J Haematol 2004; 124: 151–159.
analyses of such trials or of trials with demonstrated 3. Swerdlow SH. Keynote lecture: Kn06 update on the WHO classification of
heterogeneity malignant lymphomas. Pathology 2014; 46(Suppl. 2): S2.
III Prospective cohort studies 4. Twa DD, Chan FC, Ben-Neriah S et al. Genomic rearrangements involving
IV Retrospective cohort studies or case–control studies programmed death ligands are recurrent in primary mediastinal large B-cell
lymphoma. Blood 2014; 123: 2062–2065.
5. Kuper-Hommel MJJ, Janssen-Heijnen MLG, Vreugdenhil G et al. Clinical and
Grades of recommendation pathological features of testicular diffuse large B-cell lymphoma: a heterogeneous
disease. Leuk Lymphoma 2012; 53: 242–246.
6. Kraan W, Horlings HM, van Keimpema M et al. High prevalence of oncogenic
MYD88 and CD79B mutations in diffuse large B-cell lymphomas presenting at
immune-privileged sites. Blood Cancer J 2013; 3: e139.
7. Deckert M, Engert A, Brück W et al. Modern concepts in the biology, diagnosis,
C Insufficient evidence for efficacy or benefit does not outweigh differential diagnosis and treatment of primary central nervous system lymphoma.
the risk or the disadvantages (adverse events, costs, …), Leukemia 2011; 25:1797–1807.
optional 8. Poulain S, Boyle EM, Tricot S et al. Absence of CXCR4 mutations but high
D Moderate evidence against efficacy or for adverse outcome, incidence of double mutant in CD79A/B and MYD88 in primary central nervous
generally not recommended system lymphoma. Br J Haematol 2015; 170: 285–287.
E Strong evidence against efficacy or for adverse outcome, never 9. Cheah CY, Campbell BA, Seymour JF. Primary breast lymphoma. Cancer Treat Rev
recommended 2014; 40: 900–908.
10. Martinez A, Ponzoni M, Agostinelli C et al. Primary bone marrow lymphoma: an
a
By permission of the Infectious Diseases Society of America [55]. uncommon extranodal presentation of aggressive non-Hodgkin lymphomas. Am J
Surg Pathol 2012; 36: 296–304.
11. Cheson BD, Fisher RI, Barrington SF et al. Recommendations for initial evaluation,
staging, and response assessment of Hodgkin and non-Hodgkin lymphoma: the
Lugano classification. J Clin Oncol 2014; 32: 3059–3068.
12. Ferreri AJ, Blay JY, Reni M et al. Prognostic scoring system for primary CNS
acknowledgements lymphomas: the International Extranodal Lymphoma Study Group experience. J
We would like to thank Maura Nicolosi, MD, for her assistance Clin Oncol 2003; 21: 266–272.
with manuscript preparation. 13. Messina C, Christie D, Zucca E et al. Primary and secondary bone lymphomas.
Cancer Treat Rev 2015; 41: 235–246.
14. Yahalom J, Illidge T, Specht L et al. Modern radiation therapy for extranodal
lymphomas: field and dose guidelines from the International Lymphoma Radiation
Oncology Group. Int J Radiat Oncol Biol Phys 2015; 92: 11–31.
15. Dunleavy K, Pittaluga S, Maeda LS et al. Dose-adjusted EPOCH-rituximab therapy
UV has reported global advisory boards for lymphoma for in primary mediastinal B-cell lymphoma. N Engl J Med 2013; 368: 1408–1416.
Roche and Janssen; lectures and educational activities for Roche, 16. De Sanctis V, Finolezzi E, Osti MF et al. MACOP-B and involved-field radiotherapy
Janssen, Gilead, Celgene and Takeda; research sponsored by is an effective and safe therapy for primary mediastinal large B cell lymphoma. Int
Celgene. JFS has reported research funding from Abbvie and J Radiat Oncol Biol Phys 2008; 72: 1154–1160.
17. Martelli M, Ceriani L, Zucca E et al. 18-fluorodeoxyglucose positron-emission
Janssen; advisory boards for Abbvie, Celgene, Genentech,
tomography predicts survival following chemoimmunotherapy for primary
Gilead, Infinity, Janssen, Roche and Takeda; speakers’ bureau mediastinal large B cell lymphoma: results of the International Extranodal
for Abbvie, Celgene, Gilead, Janssen and Roche; honoraria from Lymphoma Study Group IELSG-26 Study. J Clin Oncol 2014; 32: 1769–1775.
Abbvie, Celgene, Gilead, Janssen, Roche and Takeda; travel 18. Vitolo U, Ferreri AJ, Zucca E. Primary testicular lymphoma. Crit Rev Oncol Hematol
support from Celgene and Roche; and consultancy fees from 2008; 65: 183–189.
Celgene, Roche and Takeda. TI has reported speakers’ bureau 19. Zucca E, Conconi A, Mughal TI et al. Patterns of outcome and prognostic factors in
for Takeda and Roche; research sponsored by Merck, Roche and primary large-cell lymphoma of the testis in a survey by the International
Millennium. EZ has reported advisory honoraria and/or Extranodal Lymphoma Study Group. J Clin Oncol 2003; 21: 20–27.
support of investigator-initiated studies (for the institution) 20. Vitolo U, Chiappella A, Ferreri AJ et al. First-line treatment for primary testicular
diffuse large B-cell lymphoma with rituximab-CHOP, CNS prophylaxis, and
from Celgene, Johnson and Johnson/Janssen, Gilead,
contralateral testis irradiation: final results of an international phase II trial. J Clin
Mundipharma, Roche and Bayer. ML has reported honoraria Oncol 2011; 29: 2766–2772.
from Celgene, Janssen-Cilag, Roche, Amgen, Mundipharma 21. Brouwer CL, Wiesendanger EM, van der Hulst PC et al. Scrotal irradiation in
and Teva; research contracts from Celgene, Pfizer, primary testicular lymphoma: review of the literature and in silico planning
Mundipharma and Roche; funds received from Amgen, Roche comparative study. Int J Radiat Oncol Biol Phys 2013; 85: 298–308.
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw175 | v

by guest
on 05 February 2018
22. Gisselbrecht C, Glass B, Mounier N et al. Salvage regimens with autologous 38. Hottinger AF, DeAngelis LM, Yahalom J, Abrey LE. Salvage whole brain
transplantation for relapsed large B-cell lymphoma in the rituximab era. J Clin radiotherapy for recurrent or refractory primary CNS lymphoma. Neurology 2007;
Oncol 2010; 28: 4184–4190. 69: 1178–1782.
23. Abrey LE, Batchelor TT, Ferreri AJ et al. Report of an international workshop to 39. Yhim HY, Kang HJ, Choi YH et al. Clinical outcomes and prognostic factors in
standardize baseline evaluation and response criteria for primary CNS lymphoma. patients with breast diffuse large B cell lymphoma; Consortium for Improving
J Clin Oncol 2005; 23: 5034–5043. Survival of Lymphoma (CISL) study. BMC Cancer 2010; 10: 321.
24. Ferreri AJ. How I treat primary CNS lymphoma. Blood 2011; 118: 510–522. 40. Jung SP, Kim M, Han KM et al. Primary breast lymphoma: a single institution’s
25. Abrey LE, Ben-Porat L, Panageas KS et al. Primary central nervous system experience. J Korean Surg Soc 2013; 84: 267–272.
lymphoma: the Memorial Sloan-Kettering Cancer Center prognostic model. J Clin 41. Ryan G, Martinelli G, Kuper-Hommel M et al. Primary diffuse large B-cell
Oncol 2006; 24: 5711–5715. lymphoma of the breast: prognostic factors and outcomes of a study by the
26. Ferreri AJ, Reni M, Foppoli M et al. High-dose cytarabine plus high-dose International Extranodal Lymphoma Study Group. Ann Oncol 2008; 19: 233–241.
methotrexate versus high-dose methotrexate alone in patients with primary CNS 42. Hosein PJ, Maragulia JC, Salzberg MP et al. A multicentre study of primary breast
lymphoma: a randomised phase 2 trial. Lancet 2009; 374: 1512–1520. diffuse large B-cell lymphoma in the rituximab era. Br J Haematol 2014; 165:
27. Shah GD, Yahalom J, Correa DD et al. Combined immunochemotherapy with 358–363.
reduced whole-brain radiotherapy for newly diagnosed primary CNS lymphoma. J 43. Ganjoo K, Advani R, Mariappan MR et al. Non-Hodgkin lymphoma of the breast.
Clin Oncol 2007; 25: 4730–4735. Cancer 2007; 110: 25–30.
28. Pels H, Juergens A, Glasmacher A et al. Early relapses in primary CNS lymphoma 44. Jeanneret-Sozzi W, Taghian A, Epelbaum R et al. Primary breast lymphoma:
after response to polychemotherapy without intraventricular treatment: results of a patient profile, outcome and prognostic factors. A multicentre Rare Cancer
phase II study. J Neurooncol 2009; 91: 299–305. Network study. BMC Cancer 2008; 8: 86.
29. Hoang-Xuan K, Bessell E, Bromberg J et al. Diagnosis and treatment of primary 45. Fukuhara S, Watanabe T, Munakata W et al. Bulky disease has an impact on
CNS lymphoma in immunocompetent patients: guidelines from the European outcomes in primary diffuse large B-cell lymphoma of the breast: a retrospective
Association for Neuro-Oncology. Lancet Oncol 2015;16: e322–e332. analysis at a single institution. Eur J Haematol 2011; 87: 434–440.
30. Correa DD, Ahles TA. Neurocognitive changes in cancer survivors. Cancer J 2008; 46. Cheah CY, Wirth A, Seymour JF. Primary testicular lymphoma. Blood 2014; 123:
14: 396–400. 486–493.
31. Morris PG, Correa DD, Yahalom J et al. Rituximab, methotrexate, procarbazine, 47. Avilés A, Castañeda C, Neri N et al. Rituximab and dose dense chemotherapy in
and vincristine followed by consolidation reduced-dose whole-brain radiotherapy primary breast lymphoma. Haematologica 2007; 92: 1147–1148.
and cytarabine in newly diagnosed primary CNS lymphoma: final results and long- 48. Cheah CY, Seymour JF. Central nervous system prophylaxis in non-Hodgkin
term outcome. J Clin Oncol 2013; 31: 3971–3979. lymphoma: who, what, and when? Curr Oncol Rep 2015; 17: 25.
32. Illerhaus G, Fritsch K, Egerer G et al. Sequential high dose immuno-chemotherapy 49. Govi S, Christie D, Messina C et al. The clinical features, management and
followed by autologous peripheral blood stem cell transplantation for patients with prognostic effects of pathological fractures in a multicenter series of 373 patients
untreated primary central nervous system lymphoma—a multicentre study by the with diffuse large B-cell lymphoma of the bone. Ann Oncol 2014; 25: 176–181.
collaborative PCNSL Study Group Freiburg. Blood 2012; 120: abstr. 302. 50. Messina C, Ferreri AJ, Govi S et al. Clinical features, management and prognosis
33. Ferreri AJ, Cwynarski K, Pulczynski E et al. Addition of thiotepa and rituximab to of multifocal primary bone lymphoma: a retrospective study of the international
antimetabolites significantly improves outcome in primary CNS lymphoma: first extranodal lymphoma study group (the IELSG 14 study). Br J Haematol 2014; 164:
randomization of the IELSG32 trial. Clin Adv Hematol Oncol 2015; 33(Suppl. 9): 10–11. 834–840.
34. Kasenda B, Ferreri AJ, Marturano E et al. First-line treatment and outcome of 51. Bruno Ventre M, Ferreri AJ, Gospodarowicz M et al. Clinical features,
elderly patients with primary central nervous system lymphoma (PCNSL)—a management, and prognosis of an international series of 161 patients with
systematic review and individual patient data meta-analysis. Ann Oncol 2015; 26: limited-stage diffuse large B-cell lymphoma of the bone (the IELSG-14 study).
1305–1313. Oncologist 2014; 19: 291–298.
35. Illerhaus G, Marks R, Müller F et al. High-dose methotrexate combined with 52. Held G, Zeynalova S, Murawski N et al. Impact of rituximab and radiotherapy on
procarbazine and CCNU for primary CNS lymphoma in the elderly: results of a outcome of patients with aggressive B-cell lymphoma and skeletal involvement. J
prospective pilot and phase II study. Ann Oncol 2009; 20: 319–325. Clin Oncol 2013; 31: 4115–4122.
36. Soussain C, Hoang-Xuan K, Taillandier L et al. Intensive chemotherapy followed by 53. Wilson WH, Young RM, Schmitz R et al. Targeting B cell receptor signaling with
hematopoietic stem-cell rescue for refractory and recurrent primary CNS and ibrutinib in diffuse large B cell lymphoma. Nat Med 2015; 21: 922–926.
intraocular lymphoma: Société Française de Greffe de Moëlle Osseuse-Thérapie 54. Vitolo U, Chiappella A, Franceschetti S et al. Lenalidomide plus R-CHOP21 in
Cellulaire. J Clin Oncol 2008; 26: 2512–2518. elderly patients with untreated diffuse large B-cell lymphoma: results of the
37. Nayak L, Abrey LE, Drappatz J et al. Multicenter phase II study of rituximab and REAL07 open-label, multicentre, phase 2 trial. Lancet Oncol 2014; 15: 730–737.
temozolomide in recurrent primary central nervous system lymphoma. Leuk 55. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections among
Lymphoma 2013; 54: 58–61. hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33: 139–144.
v | Vitolo et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Newly diagnosed and relapsed mantle cell

lymphoma: ESMO Clinical Practice Guidelines for
M. Dreyling1, E. Campo2, O. Hermine3, M. Jerkeman4, S. Le Gouill5, S. Rule6, O. Shpilberg7, J. Walewski8 &

M. Ladetto9, on behalf of the ESMO Guidelines Committee*
1
Department of Medicine III, University Hospital – LMU Munich, Munich, Germany; 2Hematopathology Section, Hospital Clinic of Barcelona, University of Barcelona,
Barcelona, Spain; 3Department of Hematology, Imagine Institute and Descartes University, INSERM U1163 and CNRS ERL 8564, Necker Hospital, Paris, France;
4
Department of Hematology, University Lund, Lund, Sweden; 5CHU de Nantes, Service d’Hématologie Clinique, Université de Nantes, Nantes, France; 6Peninsula
School of Medicine and Dentistry, University of Plymouth, Plymouth, UK; 7Institute of Hematology, Assuta Medical Center, Tel-Aviv, Israel; 8Department of
Lymphoid Malignancy, Maria Sklodowska-Curie Institute and Oncology Centre, Warsaw, Poland; 9Divisione di Ematologia, Azienda Ospedaliera Santi Antonio e
Biagio e Cesare Arrigo, Alessandria, Italy
*Correspondence to: ESMO Guidelines Committee, ESMO Head Office, Via, L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland. E-mail: clinicalguidelines@esmo.org
†
Approved by the ESMO Guidelines Committee: August 2014, last update May 2017. This publication supersedes the previously published version—Ann Oncol 2014; 25
(Suppl. 3): iii83–iii92.
Incidence and epidemiology marginal zone-like, pleomorphic and blastoid cells. In the
updated WHO classification, a leukaemic non-nodal subtype has
Mantle cell lymphoma (MCL) is a relatively uncommon subtype been characterised based on the clinical presentation usually with
of lymphoid malignancy and represents 5%–7% of malignant a more indolent clinical course [1]. As only the minority of these
lymphoma in Western Europe. The annual incidence of this dis- cases is correctly diagnosed based on classical histology only, re-
ease has increased during recent decades to 1–2/100 000 recently. view by an expert haematopathologist is advised. In particular,
MCL is more common in males than in women with a 3 : 1 ratio. additional immunohistochemistry for detection of cyclin D1
overexpression is mandatory.
In the rare cyclin D1-negative cases, detection of SOX11 may
Diagnosis and pathology/molecular biology help to establish the diagnosis [2].
If possible, additional biopsy material should be stored freshly
Diagnosis should be based on a surgical specimen, preferably a
frozen to allow additional molecular analyses (currently still
lymph node biopsy. Core biopsies should only be carried out in
investigational).
patients without easily accessible lymph nodes (e.g. retroperiton-
eal bulk), keeping in mind the heterogeneity of MCL. In the rare
cases with leukaemic manifestation only, a bone marrow (BM)
biopsy may be sufficient if additional diagnostic measures are
applied, immunophenotype (CD5þ, CD19/20þ), detection of Staging and risk assessment
t(11;14)(q13;q32) and overexpression of cyclin D1. Fine needle Since treatment may differ depending on the stage of the disease,
aspirations are inappropriate for a reliable evaluation of add- initial staging should be thorough, particularly in the rare cases
itional risk factors (cytology, cell proliferation). with non-bulky stages I and II (Table 1). Initial work-up should
The histological report should give the diagnosis according to include a computed tomography (CT) scan of the neck, thorax,
the World Health Organization (WHO) classification and Ki-67 abdomen and pelvis, and a BM aspirate and biopsy (Table 2).
as the most established histomorphological risk factor [I, A] [1]. Positron emission tomography (PET)-CT scan is especially rec-
Most tumours have a classic morphology of small-medium sized ommended in the rare limited stages I/II, before localised radio-
cells with irregular nuclei. However, the malignant lymphocytes therapy (RT) [IV, C]. Gastrointestinal endoscopy is also
may present with a spectrum of morphological variants, includ- recommended in these rare cases to detect asymptomatic involve-
ing small round (resembling chronic lymphocytic leukaemia), ment but otherwise only in symptomatic patients. Of note, when
V

by guest
on 05 February 2018
Table 1. Lugano classification [4] Table 3. MIPI and MIPI-c risk stratification [6, 7]
Stage Area of involvement Points Age (years) ECOG LDH (ULN) WBC (109/L)
I (IE) One lymph node region or extranodal site (IE) 0 < 50 0–1 < 0.67 < 6700
II (IIE) Two or more lymph node regions or localised extranodal sites 1 50–59 – 0.67–0.99 6700–9999
(IIE) on the same side of the diaphragm 2 60–69 2–4 1.00–1.49 10 000–14 999
III Lymph node regions or lymphoid structures (e.g. thymus, 3 70 – 1.50 15 000
Waldeyer’s ring) on both sides of the diaphragm
MIPI risk group Ki-67 index MIPI-c risk group
IV Diffuse or disseminated extralymphatic organ involvement
(weight in MIPI-c) (weight in MIPI-c) (sum of weights)
Low (0) < 30% (0) Low (0)

Low (0) 30% (1) Low-intermediate (1)
Table 2. Diagnostic work-up Intermediate (1) < 30% (0) Low-intermediate (1)
Intermediate (1) 30% (1) High-intermediate (2)
History B symptoms
High (2) < 30% (0) High-intermediate (2)
High (2) 30% (1) High (3)
Physical examination Waldeyer’s ring, peripheral lymph nodes,
liver, spleen
For each prognostic factor, 0–3 points were given to each patient and
Laboratory work-up Blood and differential count points were summed up to a maximum of 11. Patients with 0–3 points
In leukaemic cases: in summary were classified as low-risk, patients with 4 to 5 points as
FACS (CD5/CD19/20þ) intermediate-risk, and patients with 6–11 points as high-risk. ECOG
FISH for t(11;14) recommended performance status was weighted with 2 points if patients were unable
LDH, uric acid, liver and renal function to work or bedridden (ECOG 2–4). LDH was weighted according to the
Electrophoresis (optional: immune fixation) ratio to the ULN. Thus, for an ULN of 240 U/L, the limits were 180, 240
Serology Hepatitis B, C and HIV serology and 360 U/L for low-, intermediate- and high-risk groups, respectively.
ECOG, Eastern Cooperative Oncology Group; LDH, lactate dehydrogenase;
Imaging Abdominal ultrasound MIPI, mantle cell lymphoma international prognostic index; MIPI-c, com-
CT neck, chest, abdomen, pelvis bined MIPI; ULN, upper limit of normal range; WBC, white blood count.
or PET-CT
MRI only in selected locations (CNS)
Bone marrow Histology (cyclin D1 immunohistochemistry)
Cytology
Recommended: FACS, FISH for t(11;14) carried out according to the Lugano classification system
Optional: PCR for IGH gene rearrangement (Table 1), with mention of bulky disease > 5 cm when appro-
Toxicity Electrocardiogram, cardiac ultrasound
priate [4].
(before anthracyclines, ASCT) The evaluation of the cell proliferation antigen Ki-67 is the
Pulmonary function (before ASCT) most applicable method to evaluate cell proliferation, and is con-
Creatinine clearance sidered the most established biological risk factor in MCL. As the
Optional: reproductive counselling in young patients reproducibility of quantitative scores among pathologists may
vary, a standardised method has been suggested [5].
ASCT, autologous stem cell transplantation; CNS, central nervous sys- For prognostic purposes, a combined MCL International
tem; CT, computed tomography; FACS, fluorescence-activated cell sort- Prognostic Index (MIPI-c) (Table 3; web-based calculator: www.
ing; FISH, fluorescent in situ hybridisation; HIV, human european-mcl.net/de/clinical_mipi.php) has been established
immunodeficiency virus; LDH, lactate dehydrogenase; MRI, magnetic [I, A] [6, 7].
resonance imaging; PCR, polymerase chain reaction; PET, positron
emission tomography.
Leukaemic non-nodal subtype of MCL
Most patients with MCL follow an aggressive clinical course.
However, a subset of patients may exhibit a more indolent evolu-
analysed, the majority of MCL patients will have gastrointestinal tion. Most of these cases are commonly characterised by a leu-
involvement. kaemic non-nodal presentation with BM involvement only and
Central nervous system (CNS) involvement is rare in splenomegaly [1, 8]. SOX11 negativity may help to identify these
asymptomatic patients at diagnosis, but a lumbar puncture may cases. In addition, conventional MCL (SOX11-positive) with low
be considered in high-risk cases [at least two of the following Ki-67 ( 10%) tend to have a more indolent course. However,
risk factors: blastoid variant, elevated lactate dehydrogenase additional TP53 mutations may cause an aggressive clinical evo-
(LDH), impaired performance status or neurological symptoms] lution (Figure 1) [9].
[III, C] [3]. Unfortunately, there are no markers that definitely predict in-
A full blood count, blood chemistry including LDH and uric dolent behaviour, but a short course of ‘watch and wait’ period
acid, as well as screening tests for human immunodeficiency under close observation seems to be appropriate in suspected in-
virus (HIV) and hepatitis B and C, are required. Staging is dolent cases with low tumour burden [III, B] [10].

by guest
on 05 February 2018
Progression
(TP53 and other
oncogenic
abnormalities)
Classical MCL Blastoid MCL
Unmutated/minimally
Lymph
mutated IG
node,
SOX11 + B cell
extranodal
In situ MC Pleomorphic
CCND1-R neoplasia MCL
Mantle
zone
Germinal
centre
Pre-
B cell
NaÏve
B cell Hypermutated IG
SOX11–B cell PB, BM,
Cyclin spleen
D1 neg
Leukaemic
non-nodal MCL
Figure 1. Molecular pathogenesis of MCL.

BM, bone marrow; IG, immunoglobulin; MC, mantle cell; MCL, mantle cell lymphoma; neg, negative; PB, peripheral blood.
Republished from [1] with permission. Copyright 2016 American Society of Hematology. All rights reserved.
If treatment is required, recommendations for classical MCL regimens. Three prospective first-line trials, a salvage trial and a
apply. systematic meta-analysis support an improved overall response,
progression-free survival (PFS) and overall survival (OS) if ritux-
imab was added to ChT (Table 4) [I, A] [13].
Treatment Rituximab in combination with ChT such as CHOP (cyclo-
phosphamide, doxorubicin, vincristine and prednisone) or
First-line bendamustine should be used [I, B] [14–16]. Recently, a com-
bination with bortezomib achieved almost doubled median PFS
Stages I–II. In the small proportion of patients with limited non- but resulted in significant thrombocytopaenia [17]. Rituximab
bulky stages I–II, RT (involved field, 30–36 Gy) has been sug- in combination with cyclophosphamide, vincristine and pred-
gested to achieve long-term remissions [11]. In contrast, in a nisone (R-CVP) resulted in inferior response rates and PFS
randomised study, all patients with early-stage MCL relapsed [18]. Purine analogue-based schemes, rituximab with fludara-
within 1 year [12]. Thus, a shortened conventional chemotherapy bine and cyclophosphamide (R-FC) or with fludarabine and
(ChT) induction followed by consolidation RT (similar to diffuse mitoxantrone (R-FM), are also discouraged due to early failures
large cell lymphoma) may be the most appropriate treatment in and long-lasting myelosuppression [I, D] [15]. The addition of
these cases [IV, B]. high-dose cytarabine (HD-AraC) to CHOP is currently being
In stage I–II patients with large tumour burden or adverse tested in elderly patients. Recently, rituximab, bendamustine
prognostic features, systemic therapy as indicated for advanced and cytarabine (R-BAC) has been explored also in first-line
stages should be applied; consolidation RT may be considered de- therapy [19].
pending on tumour location and expected side-effects [IV, B]. In frail patients, a less intense immunochemotherapy,
chlorambucil or vincristine, doxorubicin, oral dexametha-
Stages III–IV sone and chlorambucil (VADC) or prednisone, etoposide,
Induction: In all symptomatic patients and generally in procarbazine and cyclophosphamide (PEP-C) may be con-
cases with high tumour burden, therapy should be initiated at sidered, aiming primarily at palliation [II, B]. However, tar-
diagnosis [I, A]. The current therapeutic approach is based geted therapy exhibiting a low toxicity profile might be used
on clinical risk factors, symptoms and patient characteristics in this population [20].
(Figure 2). Antibody monotherapy [rituximab, radioimmunotherapy
Elderly patients: Based on a median age of 65 years at first diag- (RIT)] achieves only moderate response rates and is therefore not
nosis, the majority of patients do not qualify for dose-intensified recommended [III, B] [21].
iv64 | Dreyling et al. Volume 28 | Supplement 4 | August 2017

by guest
on 05 February 2018
Young patient (≤65 years) Elderly patient (>65 years) Compromised patient
First-line treatment
Best supportive care?

Dose-intensified
Conventional immunochemotherapy R-chlorambucil
immunochemotherapy
(e.g. R-CHOP, VR-CAP, BR, R-BAC) BR (dose-reduced)
(e.g. R-CHOP, high dose Ara-C)
R-CVP
ASCT Rituximab maintenance

Rituximab maintenance
Relapse
Immunochemotherapy Immunochemotherapy Immunochemotherapy

(e.g. R-BAC, BR) (e.g. BR, R-BAC) (e.g. BR dose-reduced)
or targeted approaches or targeted approaches or targeted approaches
Discuss: Discuss:
AlloSCT Rituximab
Radioimmunotherapy
maintenance
Higher relapse
Targeted approaches: ibrutinib, lenalidomide

Temsirolimus, bortezomib (preferable in combination with chemotherapy)
Alternatively: repeat previous therapy (long remissions)
Figure 2. Therapeutic recommendations.

AlloSCT, allogeneic stem cell transplantation; ASCT, autologous stem cell transplantation; BAC, bendamustine and cytarabine; BR, bendamus-
tine and rituximab; Ara-C, cytarabine; CHOP, cyclophosphamide, doxorubicin, vincristine and prednisone; CVP, cyclophosphamide, vincristine
and prednisone; R, rituximab; VR-CAP, rituximab, cyclophosphamide, doxorubicin and prednisone with bortezomib.
In patients with positive hepatitis B serology, prophylactic In addition, a randomised trial confirmed that a cytarabine-
antiviral medication and/or virus load monitoring is strongly rec- containing induction achieves a significantly improved median
ommended [I, A] [22]. time to treatment failure (P ¼ 0.038) [I, B] [28]. In contrast, an
induction-based on HD-AraC alone achieves only insufficient re-
Consolidation/maintenance. Rituximab maintenance significantly sponse rates [III, D] [29].
improves PFS and OS after rituximab and CHOP (R-CHOP) In a retrospective study comparison of the Nordic, HOVON
[I, A] and PFS in a systematic meta-analysis [15, 23]. and MCL Younger protocols, total body irradiation (TBI) before
RIT consolidation also prolongs PFS after ChT, but its benefit ASCT was confirmed to be beneficial only in partial response
seems to be inferior in comparison to rituximab maintenance (PR) patients [II, B] [30]. In contrast, the benefit of RIT has not
[II, B] [24]. been demonstrated in inter-study comparisons.
An upfront dose-intensified approach (R-hyper-CVAD, rit-
Younger patients: Although no curative treatment is available uximab in combination with hyperfractionated cyclophospha-
for MCL so far, an intensive approach, e.g. by autologous stem mide, vincristine, doxorubicin and dexamethasone alternating
cell transplantation (ASCT), has been demonstrated to induce with high-dose methotrexate/cytarabine cycles) also achieved
higher response and survival rates in fit patients, independently very high response and survival rates in phase II studies, but
of the addition of rituximab [I, B] (Table 5) [25–27]. its feasibility is hampered by a significant therapy-associated

by guest
on 05 February 2018
Table 4. Conventional first-line immunochemotherapy in MCL (phase III studies)
Author Study features Assessable Therapeutic ORR% (CR%) Median PFS 2-Year OS
patients regimen (months)
Lenz et al. [14] Phase III, 112 CHOP versus R-CHOP 75 (7) versus 94 (34) 21 versus 14 (TTF) 76% versus 76%
randomised
Herold et al. [54] Phase III, 90 MCP versus R-MCP 63 (15) versus 71 (32) 18 versus 20 52% versus 55%
randomised (4-year OS)
Kluin-Nelemans Phase III, 485 R-CHOP versus R-FC 86 (34) versus 78 (40) 58% versus 29% 62% versus 47%
et al. [15] randomised (4-year DOR) (4-year OS)
R versus IFNa 79% versus 67%
(4-year OS)
Rummel et al. Phase III, 514 (94 MCL) R-CHOP versus BR 91 (30) versus 93 (40) 21 versus 35 No differences
[16] randomised
Robak et al. [17] Phase III, 487 R-CHOP versus 89 (42) versus 92 (53) 16 versus 31 54% versus 64%
randomised VR-CAP (4 years)
BR, bendamustine and rituximab; CHOP, cyclophosphamide, doxorubicin, vincristine and prednisone; CR, complete response; DOR, duration of response;
FC, fludarabine and cyclophosphamide; IFNa, interferon alpha; MCL, mantle cell lymphoma; MCP, melphalan, chlorambucil and prednisone; ORR, object-
ive response rate; OS, overall survival; PFS, progression-free survival; R, rituximab; TTF, time to failure; VR-CAP, bortezomib in combination with rituximab,
cyclophosphamide, doxorubicin and prednisone.
toxicity [II, C] [31–33] and low success rate of stem cell applied in combination with ChT based on phase II/III studies
mobilisation. [17, 42–45].
Rituximab maintenance following a rituximab with dexa- Rituximab maintenance has a favourable safety profile and
methasone, cytarabine and cisplatin (R-DHAP)-based induction prolongs PFS and OS in relapsed disease [I, A] [46]. However,
and ASCT improves PFS and OS and represents the current second-line maintenance approaches have not been investigated
standard of care [I, A] [27]. in patients relapsing after front-line maintenance [IV, D].
So far, there are no data to support the application of allogeneic RIT consolidation seems to result in extended remission dur-
stem cell transplantation (alloSCT) as part of front-line treatment ations [47, 48] especially in elderly patients with comorbidities
[II, D] [34]. not eligible for dose intensification [IV, B].
High-dose ChT with ASCT may be considered in patients
relapsed after conventional first-line therapy. However, the bene-
Relapsed disease fit seems to be marginal in this setting [49], and there is no role
for a second autograft at relapse.
A repeated biopsy is recommended to identify important prog- In younger patients, alloSCT is potentially curative and
nostic features of MCL. has achieved long-term remissions even in patients following
Selection of salvage treatment depends on efficacy of early relapse and with refractory disease [III, B]. Based on
prior regimens. In early relapses (< 12–24 months), a non- the advanced age of most patients, dose-reduced conditioning
cross-resistant scheme should be preferred [bendamustine or is appropriate [IV, B] [50]. Haploidentical BM transplant-
HD-AraC-containing regimens, e.g. rituximab, bendamustine ation achieves high response rates but is still experimental in
and cytarabine (R-BAC)] after CHOP or vice versa [19]. MCL.
Rituximab should be added if the previous antibody-
containing scheme achieved > 6 months duration of remission
[IV, B].
Response evaluation
In cases of early relapses or in refractory cases, newer tar- PET-CT according to the Lugano classification for response
geted approaches should be strongly considered (Figure 2; evaluation is optional [4].
Table 6). Among the registered compounds, ibrutinib achieves Radiological tests should be carried out mid- and post-
the highest response rates and, in some cases, long-term remis- completion of ChT. Patients who achieve less than a PR should
sions [35–37], but early relapses display very aggressive fea- be considered for early salvage regimens. Patients achieving a PR
tures. When there are contraindications to ibrutinib therapy, may convert to a complete response after post-induction
particularly a high risk of bleeding, lenalidomide (preferable treatment.
in combination with rituximab) may also achieve ongoing re- The independent prognostic role of minimal residual disease
missions in some cases [38–41]. Temsirolimus and bortezo- (MRD) applying patient-specific primers has been confirmed in
mib have been shown to be effective but should preferably be numerous studies [51, 52]. However, because of the current

by guest
on 05 February 2018
by guest
on 05 February 2018
Table 5. Dose-intensified first-line therapy in MCL (phase II/III trials)
Study features Assessable Therapeutic regimen ORR% (CR%) Median PFS Median OS (years) Dropout TRM Secondary
patients (years) rate tumour rate
Annals of Oncology
ASCT-based regimens
Dreyling et al. [25, 26] Phase III, randomised 122 CHOP þ TBI þ ASCT 98 (81) versus 3.3 versus 1.4 NR (83% 3-year OS) 13% versus 5% versus 5%
versus 99 (37) versus N/A 0%
CHOP þ TBI þ IFNa NR (77%
3-year OS)

Hermine et al. [28] Phase III, randomised 455 R-CHOP þ TBI þ ASCT 98 (63) versus 3.8 versus 7.3 6.8 versus NR N/A 4% N/A
versus 99 (61)
R-CHOP/R-DHAP þ
HD-AraC þ ASCT
Eskelund et al. [55] Phase II 160 R-Maxi-CHOP þ HD-AraC 96 (54) 7.4 NR (64% 10-year OS) 9% 5% 4%
þ ASCT
Delarue et al. [56] Phase II 60 R-CHOP/R-DHAP þ 100 (96) 6.9 NR (75% 5-year OS) 18% 1.5% 18%
HD-AraC þ ASCT

Le Gouill et al. [27] Phase III, randomised 299 R-DHAP þ ASCT 83 (77) NR (73% 3-year NR (84% 3-year OS) N/A N/A N/A
versus PFS) versus versus
R-DHAP þ ASCT þ NR (89% 3-year NR (93%
R maintenance PFS) 3-year OS)
Non-ASCT-based regimens
Romaguera et al. [31] Phase II, monocentric 97 R-hyper-CVAD N/A 4.6 NR (64% 10-year OS) 29% 8% 5%
Merli et al. [32] Phase II, multicentric 60 R-hyper-CVAD 83 (72) NR (73% 5-year NR (61% 5-year OS) 63% 6.5% 1.5%
PFS)
Bernstein et al. [33] Phase II, multicentric 49 R-hyper-CVAD 86 (55) 4.8 6.8 39% 2% 4%
ASCT, autologous stem cell transplantation; CHOP, cyclophosphamide, doxorubicin, vincristine, prednisone; DHAP, dexamethasone, cytarabine and cisplatin; CR, complete response; HD-AraC, high dose cytara-
bine; hyper-CVAD, hyperfractionated cyclophosphamide, vincristine, doxorubicin, dexamethasone alternating with high-dose methotrexate and cytarabine; IFNa, interferon alpha; Maxi-CHOP, maximum-
strength CHOP; MCL, mantle cell lymphoma; N/A, not applicable; NR, not reached; ORR, objective response rate; OS, overall survival; PFS, progression-free survival; R, rituximab; TBI, total body irradiation; TRM,
transplant-related mortality.
doi:10.1093/annonc/mdx223 | iv67
Table 6. Published clinical studies investigating molecular targeted approaches in relapsed MCL
Author Study Assessable Therapeutic regimen ORR% (CR%) Median PFS Median OS
features patients (months) (months)
Proteasome inhibitors
Goy et al. [45] Phase II 141 Bortezomib 33 (8) 6.7 (TTP) 23.5
mTOR inhibitors
Hess et al. [42] Phase III, 54 Temsirolimus 175 mg/75 mg 22 (2) 4.8 12.8
randomised 54 Temsirolimus 175 mg/25 mg 6 (0) 3.4 10.0
53 Investigator’s choice 2 (2) 1.9 9.7
Ansell et al. [43] Phase II 69 Temsirolimus, rituximab 59 (19) 9.7 29.5
Hess et al. [44] Phase II 32 Temsirolimus, BR 87 (8) 18 36.0
Immunomodulatory drugs
Zinzani et al. [38] Phase II 57 Lenalidomide 35 (12) 8.8 NR
Goy et al. [39] Phase II 134 Lenalidomide 28 (8) 4 19.0
Trneny et al. [40] Phase III 254 Lenalidomide 46 (11) 8,7 27.9
Investigator’s choice 23 (8) 5.2 21.2
Wang et al. [41] Phase II 44 Lenalidomide, rituximab 57 (36) 11.1 24.3
Antibody-based approaches
90
Wang et al. [47] Phase II 32 Y-ibritumumab tiuxetan 31 (16) 6 (EFS) 21
Ferrero et al. [48] Phase II 15þa 90
Y-ibritumumab tiuxetan 40 (20) 3.7 13.8
BCR signalling inhibitors
Wang et al. [35] Phase II 111 Ibrutinib 68 (21) 13.9 NR (1.5-year OS 58%)
Dreyling et al. [36] Phase III 280 Ibrutinib versus temsirolimus 72 (26) 14.6 NR (68% at 1 year)
Wang et al. [37] Phase II 50 Ibrutinib, rituximab 88 (44) NR NR
Kahl et al. [57] Phase I 16 Idelalisib 62 (N/A) 3 (DOR) N/A
BCL2 inhibitors
Davids et al. [53] Phase I 32 (8 MCL) Venetoclaxb 100 (0) N/A N/A
a
Fifteen patients received the antibody as relapse monotherapy, 30 patients as consolidation after salvage treatment.
b
Not registered in MCL.
BCL2, B-cell lymphoma 2; BCR, B-cell receptor; BR, bendamustine and rituximab; CR, complete response; DOR, duration of response; EFS, event-free sur-
vival; MCL, mantle cell lymphoma; mTOR, mechanistic target of rapamycin; N/A, not applicable; NR, not reached; ORR, overall response rate; OS, overall
survival; PFS, progression-free survival; TTP, time to progression.
Table 7. Recommended follow-up

Examination Details Year 1–2 Year 3–5 Year > 5
History B symptoms Every 3 months Twice annually Annually

Physical examination Peripheral lymph nodes, liver, spleen Every 3 months Twice annually Annually
Laboratory work-up Blood and differential count Every 3 months Twice annually Annually
LDH
Imaging Abdominal ultrasound Optional: every 3–6 months Optional: every 6–12 months If progress suspected
CT neck, chest, abdomen, pelvis
Toxicity TSH if irradiated Annually
CT, computed tomography; LDH, lactate dehydrogenase; TSH, thyroid-stimulating hormone.
limitations in knowing how to react, especially in MRD-positive Personalised medicine

patients, its use is advised in clinical trials but not recommended
In this disease setting, more research is needed to identify mo-
in clinical routine, except for the setting of donor lymphocyte in-
lecular markers, which could translate into a more personalised
fusion post-allograft.
approach.

by guest
on 05 February 2018
Table 8. Summary of recommendations The selection of the optimal treatment of any given patient is
based mainly on clinical and biological risk factors, symptoms
Diagnostic procedures include histomorphology by an expert haematopa-
and tumour load (Figure 2). PET and MRD-based tailored treat-
thologist, immunophenotype (CD5þ, CD19/20þ), Ki-67 staining and man-
ments are currently being evaluated in studies but are not yet part
datory detection of cyclin D1 overexpression or t(11;14)(q13;q32)
of routine clinical practice.
Clinical and biological prognosticators (MIPI-c) should be applied in clinical
routine to estimate the clinical behaviour
New agents, especially other inhibitors targeting the B-cell re-
In localised stages: discuss conventional ChT followed by RT (30–36 Gy)
ceptor (BCR) pathway, B-cell lymphoma 2 (BCL2) or cyclin-
In advanced stages: dependent kinases (CDK), are currently being investigated [53].
– Younger patients: HD-AraC containing regimens plus rituximab
followed by ASCT and rituximab maintenance
– Elderly patients: conventional immunochemotherapy (e.g. R-CHOP, Follow-up, long-term implications and
VR-CAP, BR, R-BAC) followed by rituximab maintenance.
In relapse:
survivorship
– Targeted approaches (ibrutinib, lenalidomide) should be considered, The following recommendations are based on consensus rather
temsirolimus and bortezomib should be preferably combined with ChT than on evidence (see Table 7):
– In younger patients, an alloSCT should be discussed
– Consider participation in trials
• History and physical examination, blood counts and routine
chemistry every 3 months for 2 years, every 6 months for 3
alloSCT, allogeneic stem cell transplantation; ASCT, autologous stem additional years and, subsequently, once a year [V, D].
cell transplantation; BR, bendamustine and rituximab; ChT, chemother- • Annual evaluation of thyroid function in patients after irradi-
apy; HD-AraC, high dose cytarabine; MIPI-c, combined mantle cell ation of the neck.
lymphoma international prognostic index; R-BAC, rituximab, benda- • Optional CT scan (or ultrasound examinations to reduce ra-
mustine and cytarabine; R-CHOP, rituximab plus cyclophosphamide, diation exposure) every 3–6 months for 2 years and every 6–
doxorubicin, vincristine and prednisone; VR-CAP, rituximab, cyclophos- 12 months up to 5 years. However, there is no strong evidence
phamide, doxorubicin, and prednisone with bortezomib; RT, to support a regular radiological follow-up. These recommen-
radiotherapy. dations are driven by the concern to minimise radiation ex-
posure and lack of evidence for survival advantage conferred
with routine surveillance imaging. PET-CT should not be
used for surveillance.
• Some studies suggest that pre-emptive treatment may be effi-
Table 9. Levels of evidence and grades of recommendation (adapted from cient. However, MRD screening may be carried out but
the Infectious Diseases Society of America-United States Public Health should not guide therapeutic strategies outside clinical
Service Grading Systema) studies.
Levels of evidence
Methodology
heterogeneity These Clinical Practice Guidelines were developed in accord-
II Small randomised trials or large randomised trials with a suspicion ance with the ESMO standard operating procedures for Clinical
of bias (lower methodological quality) or meta-analyses of such Practice Guidelines development http://www.esmo.org/
trials or of trials with demonstrated heterogeneity Guidelines/ESMO-Guidelines-Methodology. The relevant lit-
III Prospective cohort studies erature has been selected by the expert authors. A summary of
IV Retrospective cohort studies or case–control studies recommended treatment strategies outside of clinical studies is
V Studies without control group, case reports, experts opinions provided in Figure 2, and a summary of recommendations is
Grades of recommendation provided in Table 8. Levels of evidence and grades of recom-
A Strong evidence for efficacy with a substantial clinical benefit, mendation have been applied using the system shown in
strongly recommended Table 9. Statements without grading were considered justified
B Strong or moderate evidence for efficacy but with a limited clin- standard clinical practice by the experts and the ESMO Faculty.
ical benefit, generally recommended This manuscript has been subjected to an anonymous peer re-
C Insufficient evidence for efficacy or benefit does not outweigh view process.
the risk or the disadvantages (adverse events, costs, . . .),
optional
D Moderate evidence against efficacy or for adverse outcome, gen-
erally not recommended
Disclosure
E Strong evidence against efficacy or for adverse outcome, never MD has reported being a member of the advisory board of Bayer,
recommended Celgene, Gilead, Janssen, Mundipharma, Sandoz and Roche,
received honoraria from Bayer, Celgene, Gilead, Janssen and
a
By permission of the Infectious Diseases Society of America (58). Roche and research grants from Celgene, Janssen, Mundipharma,
Pfizer and Roche; MJ received research support from Janssen,

by guest
on 05 February 2018
Celgene, Giliead and Abbvie; SLG has reported research grants 16. Rummel MJ, Niederle N, Maschmeyer G et al. Bendamustine plus rituxi-
and honoraria from Roche Genentech and Jenssen-Cilag and mab versus CHOP plus rituximab as first-line treatment for pa-
tients with indolent and mantle-cell lymphomas: an open-label,
honoraria from Celgene; OS has reported to be a member of the
multicentre, randomised, phase 3 non-inferiority trial. Lancet 2013; 381:
scientific advisory board for Roche, AbbVie, Gilead sciences and 1203–1210.
Takeda; JW has reported advisory board for Roche, Janssen- 17. Robak T, Huang H, Jin J et al. Bortezomib-based therapy for newly diag-
Cilag, Celgene, Amgen, Bristol-Myers Squibb and Incyte, nosed mantle-cell lymphoma. N Engl J Med 2015; 372: 944–953.
received lecture honoraria from Roche, Takeda, Celgene, Servier, 18. Flinn IW, van der Jagt R, Kahl BS et al. Randomized trial of
Gilead and Janssen-Cilag, research funding from Roche, bendamustine-rituximab or R-CHOP/R-CVP in first-line treatment
Mundipharma, Celgene and GlaxoSmithKline/Novartis, and of indolent NHL or MCL: the BRIGHT study. Blood 2014; 123:
2944–2952.
travel grants from Roche, Celgene, Sanofi and Servier; EC, OH,
19. Visco C, Chiappella A, Nassi L et al. Rituximab, bendamustine, and low-
SR and ML have reported no conflict of interest. dose cytarabine as induction therapy in elderly patients with mantle cell
lymphoma: a multicentre, phase 2 trial from Fondazione Italiana
Linfomi. Lancet Haematol 2017; 4: e15–e23.
References 20. Ruan J, Martin P, Shah B et al. Lenalidomide plus rituximab as initial treat-
1. Swerdlow SH, Campo E, Pileri SA et al. The 2016 revision of the World ment for mantle-cell lymphoma. N Engl J Med 2015; 373: 1835–1844.
Health Organization classification of lymphoid neoplasms. Blood 2016; 21. Ghielmini M, Schmitz SF, Cogliatti S et al. Effect of single-agent rituxi-
127: 2375–2390. mab given at the standard schedule or as prolonged treatment in patients
2. Mozos A, Royo C, Hartmann E et al. SOX11 expression is highly specific with mantle cell lymphoma: a study of the Swiss Group for Clinical
for mantle cell lymphoma and identifies the cyclin D1-negative subtype. Cancer Research (SAKK). J Clin Oncol 2005; 23: 705–711.
Haematologica 2009; 94: 1555–1562. 22. Mallet V, van Bömmel F, Doerig C et al. Management of viral hepatitis in
3. Cheah CY, George A, Giné E et al. European Mantle Cell Lymphoma patients with haematological malignancy and in patients undergoing
Network. Central nervous system involvement in mantle cell lymphoma: haemopoietic stem cell transplantation: recommendations of the 5th
clinical features, prognostic factors and outcomes from the European European Conference on Infections in Leukaemia (ECIL-5). Lancet
Mantle Cell Lymphoma Network. Ann Oncol 2013; 24: 2119–2123. Infect Dis 2016; 16: 606–617.
4. Cheson BD, Fisher RI, Barrington SF et al. Recommendations for initial 23. Vidal L, Gafter-Gvili A, Dreyling MH et al. Maintenance therapy for pa-
evaluation, staging and response assessment of Hodgkin and non- tients with mantle cell lymphoma (MCL) - a systematic review and meta-
Hodgkin lymphoma – the Lugano Classification. J Clin Oncol 2014; 32: analysis of randomized controlled trials (RCTs). Blood 2016; (ASH
3059–3068. Annual Meeting Abstracts); 128: 1802.
5. Klapper W, Hoster E, Determann O et al. Ki-67 as a prognostic marker 24. Smith MR, Li H, Gordon L et al. Phase II study of rituximab plus cyclo-
in mantle cell lymphoma-consensus guidelines of the pathology panel of phosphamide, doxorubicin, vincristine, and prednisone immunochemo-
the European MCL Network. J Hematop 2009; 2: 103–111. therapy followed by yttrium-90-ibritumomab tiuxetan in untreated
6. Hoster E, Dreyling M, Klapper W et al. A new prognostic index (MIPI) mantle-cell lymphoma: Eastern Cooperative Oncology Group Study
for patients with advanced-stage mantle cell lymphoma. Blood 2008; E1499. J Clin Oncol 2012; 30: 3119–3126.
111: 558–565. 25. Dreyling M, Lenz G, Hoster E et al. Early consolidation by myeloablative
7. Hoster E, Rosenwald A, Berger F et al. Prognostic value of Ki-67 index, radiochemotherapy followed by autologous stem cell transplantation in
cytology, and growth pattern in mantle-cell lymphoma: results from first remission significantly prolongs progression-free survival in mantle-
randomized trials of the European Mantle Cell Lymphoma Network. cell lymphoma: results of a prospective randomized trial of the European
J Clin Oncol 2016; 34: 1386–1394. MCL Network. Blood 2005; 105: 2677–2684.
8. Fernandez V, Salamero O, Espinet B et al. Genomic and gene expression 26. Hoster E, Metzner B, Forstpointner R et al. Autologous Stem Cell
profiling defines indolent forms of mantle cell lymphoma. Cancer Res Transplantation and Addition of Rituximab Independently Prolong
2010; 70: 1408–1418. Response Duration in Advanced Stage Mantle Cell Lymphoma. Blood
9. Nygren L, Baumgartner Wennerholm S, Klimkowska M et al. Prognostic 2009; (ASH Annual Meeting Abstracts); 114: 880.
role of SOX11 in a population-based cohort of mantle cell lymphoma. 27. Le Gouill S, Thieblemont C, Oberic L et al. Rituximab maintenance after
Blood 2012; 119: 4215–4223. autologous stem cell transplantation prolongs survival in younger pa-
10. Martin P, Chadburn A, Christos P et al. Outcome of deferred initial ther- tients with mantle cell lymphoma. Blood 2016; (ASH Annual Meeting
apy in mantle-cell lymphoma. J Clin Oncol 2009; 27: 1209–1213. Abstracts); 128: 145.
11. Leitch HA, Gascoyne RD, Chhanabhai M et al. Limited-stage mantle-cell 28. Hermine O, Hoster E, Walewski J et al. Addition of high-dose cytarabine
lymphoma. Ann Oncol 2003; 14: 1555–1561. to immunochemotherapy before autologous stem-cell transplantation in
12. Engelhard M, Unterhalt M, Hansmann M et al. Follicular lymphoma, patients aged 65 years or younger with mantle cell lymphoma (MCL
immunocytoma, and mantle cell lymphoma: randomized evaluation of Younger): a randomised, open-label, phase 3 trial of the European
curative radiotherapy in limited stage nodal disease. Ann Oncol 2008; 19 Mantle Cell Lymphoma Network. Lancet 2016; 388: 565–575.
(suppl. 4): 418. 29. Laurell A, Kolstad A, Jerkeman M et al. High dose cytarabine with rituxi-
13. Schulz H, Bohlius JF, Trelle S et al. Immunochemotherapy with rituximab is not enough in first-line treatment of mantle cell lymphoma with
mab and overall survival in patients with indolent or mantle cell lymph- high proliferation: early closure of the Nordic Lymphoma Group Mantle
oma: a systematic review and meta-analysis. J Natl Cancer Inst 2007; 99: Cell Lymphoma 5 trial. Leuk Lymphoma 2014; 55: 1206–1208.
706–714. 30. Hoster E, Geisler CH, Doorduijn J et al. Total body irradiation after
14. Lenz G, Dreyling M, Hoster E et al. Immunochemotherapy with rituxi- high-dose cytarabine in mantle cell lymphoma: A comparison of Nordic
mab and cyclophosphamide, doxorubicin, vincristine, and prednisone MCL2, HOVON-45, and European MCL Younger trials. Leukemia 2016;
significantly improves response and time to treatment failure, but not 30: 1428–1430.
long-term outcome in patients with previously untreated mantle cell 31. Romaguera JE, Fayad L, Rodriguez MA et al. High rate of durable remis-
lymphoma: results of a prospective randomized trial of the German Low sions after treatment of newly diagnosed aggressive mantle-cell lymphoma
Grade Lymphoma Study Group (GLSG). J Clin Oncol 2005; 23: with rituximab plus hyper-CVAD alternating with rituximab plus high-
1984–1992. dose methotrexate and cytarabine. J Clin Oncol 2005; 23: 7013–7023.
15. Kluin-Nelemans HC, Hoster E, Hermine O et al. Treatment of older pa- 32. Merli F, Luminari S, Ilariucci F et al. Rituximab plus HyperCVAD alter-
tients with mantle-cell lymphoma. N Engl J Med 2012; 367: 520–531. nating with high dose cytarabine and methotrexate for the initial

by guest
on 05 February 2018
treatment of patients with mantle cell lymphoma, a multicentre trial 46. Forstpointner R, Unterhalt M, Dreyling M et al. Maintenance therapy
from Gruppo Italiano Studio Linfomi. Br J Haematol 2012; 156: with rituximab leads to a significant prolongation of response duration
346–353. after salvage therapy with a combination of rituximab, fludarabine,
33. Bernstein SH, Epner E, Unger JM et al. A phase II multicenter trial of cyclophosphamide, and mitoxantrone (R-FCM) in patients with recur-
hyperCVAD MTX/Ara-C and rituximab in patients with previously untreated ring and refractory follicular and mantle cell lymphomas: results of a
mantle cell lymphoma; SWOG 0213. Ann Oncol 2013; 24: 1587–1593. prospective randomized study of the German Low Grade Lymphoma
34. Krüger WH, Hirt C, Basara N et al. Allogeneic stem cell transplantation Study Group (GLSG). Blood 2006; 108: 4003–4008.
for mantle cell lymphoma-final report from the prospective trials of the 47. Wang M, Oki Y, Pro B et al. Phase II study of yttrium-90-ibritumomab
East German Study Group Haematology/Oncology (OSHO). Ann tiuxetan in patients with relapsed or refractory mantle cell lymphoma.
Hematol 2014; 93: 1587–1597. J Clin Oncol 2009; 27: 5213–5218.
35. Wang ML, Rule S, Martin P et al. Targeting BTK with ibrutinib in relapsed 48. Ferrero S, Pastore A, Scholz CW et al. Radioimmunotherapy in relapsed/
or refractory mantle-cell lymphoma. N Engl J Med 2013; 369: 507–516. refractory mantle cell lymphoma patients: final results of a European
36. Dreyling M, Jurczak W, Jerkeman M et al. Ibrutinib versus temsirolimus in MCL network phase II trial. Leukemia 2016; 30: 984–987.
patients with relapsed or refractory mantle-cell lymphoma: an interna- 49. Cassaday RD, Guthrie KA, Budde EL et al. Specific features identify pa-
tional, randomised, open-label, phase 3 study. Lancet 2016; 387: 770–778. tients with relapsed or refractory mantle cell lymphoma benefitting from
37. Wang ML, Lee H, Chuang H et al. Ibrutinib in combination with rituxi- autologous hematopoietic cell transplantation. Biol Blood Marrow
mab in relapsed or refractory mantle cell lymphoma: a single-centre, Transplant 2013; 19: 1403–1406.
open-label, phase 2 trial. Lancet Oncol 2016; 17: 48–56. 50. Robinson S, Dreger P, Caballero D et al. The EBMT/EMCL consensus
38. Zinzani PL, Vose JM, Czuczman MS et al. Long-term follow-up of lenali- project on the role of autologous and allogeneic stem cell transplantation
domide in relapsed/refractory mantle cell lymphoma: subset analysis of in mantle cell lymphoma. Leukemia 2015; 29: 464–473.
the NHL-003 study. Ann Oncol 2013; 24: 2892–2897. 51. Pott C, Hoster E, Delfau-Larue MH et al. Molecular remission is an inde-
39. Goy A, Sinha R, Williams ME et al. Single-agent lenalidomide in patients pendent predictor of clinical outcome in patients with mantle cell
with mantle-cell lymphoma who relapsed or progressed after or were re- lymphoma after combined immunochemotherapy: a European MCL
fractory to bortezomib: phase II MCL-001 (EMERGE) study. J Clin intergroup study. Blood 2010; 115: 3215–3223.
Oncol 2013; 31: 3688–3695. 52. Ladetto M, Magni M, Pagliano G et al. Rituximab induces effective clear-
40. Trneny M, Lamy T, Walewski J et al. Lenalidomide versus investigator’s ance of minimal residual disease in molecular relapses of mantle cell
choice in relapsed or refractory mantle cell lymphoma (MCL-002; lymphoma. Biol Blood Marrow Transplant 2006; 12: 1270–1276.
SPRINT): a phase 2, randomised, multicentre trial. Lancet Oncol 2016; 53. Davids MS, Roberts AW, Seymour JF et al. Phase I first-in-human study
17: 319–331. of venetoclax in patients with relapsed or refractory non-Hodgkin
41. Wang M, Fayad L, Wagner-Bartak N et al. Lenalidomide in combination lymphoma. J Clin Oncol 2017; 35: 826–833.
with rituximab for patients with relapsed or refractory mantle-cell 54. Herold M, Haas A et al. Immunochemotherapy (R-MCP) in advanced
lymphoma: a phase 1/2 clinical trial. Lancet Oncol 2012; 13: 716–723. mantle cell lymphoma is not superior to chemotherapy (MCP) alone -
42. Hess G, Herbrecht R, Romaguera J et al. Phase III study to evaluate tem- 50 months update of the OSHO phase III study (OSHO#39). Ann Oncol
sirolimus compared with investigator’s choice therapy for the treatment 2008; 19 (Suppl. 4): iv85–iv86.
of relapsed or refractory mantle cell lymphoma. J Clin Oncol 2009; 27: 55. Eskelund CW, Kolstad A, Jerkeman M et al. 15-year follow-up of the
3822–3829. Second Nordic Mantle Cell Lymphoma trial (MCL2): prolonged remis-
43. Ansell SM, Tang H, Kurtin PJ et al. Temsirolimus and rituximab in pa- sions without survival plateau. Br J Haematol 2016; 175: 410–418.
tients with relapsed or refractory mantle cell lymphoma: a phase 2 study. 56. Delarue R, Haioun C, Ribrag V et al. CHOP and DHAP plus rituximab
Lancet Oncol 2011; 12: 361–368. followed by autologous stem cell transplantation in mantle cell lymph-
44. Hess G, Karola W, LaRosee P et al. Temsirolimus in combination with oma: a phase 2 study from the Groupe d’Etude des Lymphomes de
bendamustine and rituximab (BeRT) for the treatment of relapsed man- l’Adulte. Blood 2013; 121: 48–53.
tle cell and follicular lymphoma: final phase I/II results. Blood 2016; 57. Kahl BS, Spurgeon SE, Furman RR et al. A phase 1 study of the PI3Kd in-
(ASH Annual Meeting Abstracts) 128: 2977. hibitor idelalisib in patients with relapsed/refractory mantle cell lymph-
45. Goy A, Bernstein SH, Kahl BS et al. Bortezomib in patients with relapsed oma (MCL). Blood 2014; 123: 3398–3405.
or refractory mantle cell lymphoma: updated time-to-event analyses of 58. Dykewicz CA. Summary of the guidelines for preventing opportunistic
the multicenter phase 2 PINNACLE study. Ann Oncol 2009; 20: infections among hematopoietic stem cell transplant recipients. Clin
520–525. Infect Dis 2001; 33: 139–144.

by guest
on 05 February 2018
Myelodysplastic syndromes: ESMO Clinical Practice

P. Fenaux1, D. Haase2, G. F. Sanz3, V. Santini4 & C. Buske5 on behalf of the ESMO Guidelines
Working Group*
1
Service d’Hématologie Clinique, Groupe Francophone des Myélodysplasies (GFM), Hôpital St Louis (Assistance Publique, Hôpitaux de Paris) and Paris 7 University, Paris,
France; 2Clinics of Hematology and Medical Oncology, University Medicine, Goettingen, Germany; 3Department of Haematology, Hospital Universitario y Politécnico La Fe,
Valencia, Spain; 4Functional Unit of Haematology, AOU Careggi, University of Florence, Firenze, Italy; 5Comprehensive Cancer Center Ulm, Institute of Experimental Cancer
Research, University Hospital, Ulm, Germany
introduction predisposition should also be assessed in MDS occurring in

young adults or in families with other cases of MDS, AML or aplas-
Myelodysplastic syndromes (MDS) are clonal haematopoietic tic anaemia. Environmental factors include previous use of chemo-
stem cell (HSC) disorders predominating in the elderly, charac- therapy, especially alkylating agents and more recently of purine
terised by ineffective haematopoiesis leading to blood cytopaenias analogues [12] radiotherapy or ionising radiation [13, 14], and
and by progression to acute myeloid leukaemia (AML) in one- tobacco smoking [15]. Recognised occupational factors include
third of cases [1]. Their pathophysiology is a multistep process benzene and its derivatives [16], while an excess of MDS is
involving cytogenetic changes and/or gene mutations [2] and reported in agricultural and industrial workers [15, 17]. Those
widespread gene hypermethylation at advanced stages [3–5]. ‘secondary MDS’, particularly cases occurring after chemother-
clinical practice
Diagnosis of MDS is based on the blood and marrow examin- apy, generally have poor prognostic factors, including complex
guidelines
ation, showing blood cytopaenias, hypercellular marrow with cytogenetic findings involving chromosomes 5 and/or 7 and/or
dysplasia, with or without an excess of immature marrow cells 17p, constituting the so-called alkylator type, therapy-associated
(blasts) [6]. Prognosis is largely based on the marrow blast haematological malignancies.
percentage, number and extent of cytopaenias and cytogenetic ab-
normalities, which are grouped in a recently Revised International
Prognostic Scoring system (IPSS/IPSS-R) [7, 8]. Treatment diagnosis
varying from symptomatic therapy of cytopaenias, especially by
Well-established and necessary diagnostic tools for MDS with
transfusions, to allogeneic stem cell transplantation (alloSCT)
widespread availability are peripheral and differential blood
has improved in the last few years.
counts, cytomorphology of peripheral blood and bone marrow
smears and cytogenetics of bone marrow cells. At initial diagno-
incidence and aetiology sis, histology of bone marrow trephine biopsies is strongly
MDS are diseases of the elderly, with a median age at diagnosis recommended, especially in the case of difficult diagnosis and
of ∼70 years and with >10% of the patients being younger than because of its potential prognostic information.
50 years of age [9]. The incidence of MDS in Europe is about 4 The medical history of the patient can provide important
cases/100 000 inhabitants/year (reaching 40–50/100 000 in information relating to differential diagnoses such as history of
patients aged ≥70 years) [9]. There are no known ethnic differ- medication or ingestion of alcohol or other drugs, as well as an ex-
ences in the incidence of MDS, but MDS in Asian populations clusion of other diseases including autoimmune disorders, renal
tend to occur at an earlier age, more often have a hypocellular failure, malignancies, chronic infections or inflammations, aplastic
marrow and present less often with isolated 5q deletion (5q- anaemia and paroxysmal nocturnal haemoglobinuria [18]. Beyond
syndrome), while trisomy 8 seems to be more frequent than in the mere diagnosis of MDS, one should classify every case accord-
Western populations [10, 11]. ing to the World Health Organisation (WHO) criteria [19] and
The aetiology of MDS is known in only 15% of cases. Inherited should establish the prognosis by IPSS [7] and IPSS-R [8].
predisposition to MDS is seen in one-third of paediatric MDS
cases, including in Down’s syndrome, Fanconi anaemia and neu-
rofibromatosis. It is less frequent in adults, where an inherited
peripheral blood counts and differential
blood counts
*Correspondence to: ESMO Guidelines Working Group, ESMO Head Office, Via Almost all patients with MDS have peripheral blood cytopaenias,
L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland. mostly anaemia with or without other cytopaenias. If blood
counts only modestly deviate from normal values, repeated
†
Approved by the ESMO Guidelines Working Group: June 2014. controls are recommended.

by guest
on 05 February 2018
laboratory parameters and must be evaluated morphologically in marrow aspirates (less

so with other methods) according to WHO and International
Important laboratory values supporting or excluding the diagno- Working Group (IWG) criteria. ‘Blasts’ should include agranular
sis of MDS are lactate dehydrogenase (LDH), ferritin, transferrin blasts and myeloblasts, but not promyelocytes.
and transferrin saturation, reticulocyte counts, vitamin B12 and
folate concentrations, haptoglobin, endogenous erythropoietin
(EPO) and creatinine levels. They can serve to exclude the differ- histopathology
ential diagnoses of iron deficiency anaemia, haemolytic anaemia,
In Europe, contrary to in the United States, MDS (like acute leu-
vitamin B12 or folate deficiency and renal anaemia. If MDS is
kaemias) are mainly diagnosed by bone marrow aspirate rather
diagnosed, ferritin and LDH also have a certain prognostic value,
than with a biopsy. Bone marrow trephine biopsy, however, is
and the EPO level can support a decision for or against treatment
very useful in the case of hypocellular aspirates or dry puncture,
with erythropoiesis-stimulating agents (ESAs).
where hypoplastic MDS or fibrotic MDS may be diagnosed. It
may also be important for other differential diagnoses. In expert
cytomorphology hands, histomorphology can provide additional information
on dysplastic features and prognostic information, especially
The hallmarks of cytomorphology in MDS are the determin-
by showing fibrosis (see below). It is therefore strongly recom-
ation of dysplastic signs in erythropoiesis, granulopoiesis and
mended in addition to bone marrow aspiration.
megakaryopoiesis in the bone marrow and/or peripheral blood
and the enumeration of blast cells again in the bone marrow
and/or peripheral blood. In the bone marrow, histology of cytogenetics
trephine biopsies is of great additional value.
In early MDS with only mild morphological abnormalities, In MDS, clonal chromosome abnormalities can be observed in
certain cases with persistent, unexplained cytopaenias are called 30% to >80% of patients depending on the MDS subtype and
idiopathic cytopaenias of uncertain significance (ICUS). In whether the disease is de novo or chemo- or radiotherapy-
patients with dysplastic features in the bone marrow but no or
only very mild peripheral blood cytopaenias and a normal kar-
yotype, idiopathic dysplasia of unknown significance (IDUS) Table 2. Signs of dysplasia in myelodysplastic syndromes
[20] can be diagnosed if no other cause of dysplasia is apparent
(see Table 1). Of note, however, the terms ICUS and IDUS Peripheral blood
(although they are consensus statements from a MDS working Granulocytes Pseudo-Pelger cells, abnormal chromatin
conference) are not universally accepted and are not included in clumping, hypo-/degranulation, left shift
the current 2008 WHO classification. Platelets Giant platelets, anisometry of platelets
When evaluating MDS peripheral blood films and bone Red cells Anisocytosis, poikilocytosis, dimorphic
marrow slides, a number of cytological abnormalities should be erythrocytes, polychromasia,
taken into account; a list of those items is given in Table 2 [21]. hypochromasia, megalocytes, basophilic
For the diagnosis of MDS, the recommended number of cells that stippling, presence of nucleated
should be reviewed per slide is 200 for the peripheral blood film erythroid precursors, teardrop cells,
and up to 500 for the bone marrow [19]. The bone marrow blast ovalocytes, fragmentocytes
count is crucial in MDS, given its paramount prognostic value Bone marrow
Cellularity of the Typically hypercellularity, rarely
marrow hypocellularity
Erythropoiesis Megaloblastoid changes, multi-nuclearity,
Table 1. Definition of ICUS and IDUS [18]
nuclear budding, nuclear bridges,
Characteristics atypical mitoses, sideroblastosis, ring
sideroblasts, periodic Acid–Schiff-
ICUS Mild cytopaenias (haemoglobin <11.0 g/dl, neutropaenia
positive red cell precursors
<1500/μl and/or thrombocytopaenia <100.000/μl and
Megakaryopoiesis Micro-megakaryocytes, mononuclear
lack of significant dysplasia in the bone marrow but
megakaryocytes, dumbbell-shaped
exclusion of other diseases and/or
nuclei, hypersegmentation, multi-
no clonal cytogenetic/molecular markers
nuclearity with multiple isolated nuclei
IDUS Mild cytopaenias for >6 months (hb ≥11/dl, neutrophils
Granulocytopoiesis Left shift, increased medullary blast count,
≥1500/μl, platelets ≥100 000/μl, all below lower limit of
Auer rods or Auer bodies, hypo-/
normal) or
degranulation, Pseudo-Pelger cells,
no cytopaenias but marked dysplasia in >10% of cell
nuclear anomalies (e.g.
lineages and
hypersegmentation, abnormal
no clonal cytogenetic/molecular markers
chromatin clumping), deficiency of
myeloperoxidase, increase and
ICUS, idiopathic cytopaenias of uncertain significance; IDUS,
morphological abnormality of
idiopathic dysplasia of unknown significance.
monocytes
iii | Fenaux et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
induced [22]. In the remaining 20%–70% of patients with a In an international database of 2124 MDS patients, 52% of
normal karyotype, there is growing evidence that sub-microscopic patients had one or more clonal cytogenetic abnormality by
alterations such as point mutations, micro-deletions, micro- chromosome banding. Abnormal karyotypes showed a clear as-
amplifications, epigenetic changes or copy-number neutral loss of sociation with the severity of MDS, increasing with the medul-
genetic information as by uniparental disomy provide the genetic lary blast count and the intensity of cellular dysplasias [22].
basis for the disease [2, 23, 24]. Karyotype also has the highest Several independent studies have proven the dismal outcome
prognostic weight of all parameters in the IPSS-R [8]. related to complex abnormalities, especially in the presence of
Chromosome banding analysis is carried out on dividing monosomies (monosomal karyotype); however, there is a growing
metaphase cells. Generally, whenever possible, 20–25 meta- body of evidence that it is not the monosomies per se but rather
phases should be structurally analysed not to miss smaller cell the complexity of chromosomal changes that determines the
clones which are quite frequent, especially in low-risk MDS. dismal outcome [22, 27–30].
According to the International System for Human Cytogenetic
Nomenclature (ISCN), an abnormal clone is defined by at least
two metaphases with the same supernumerary chromosome or
additional diagnostic tests
structural change, or at least three metaphases with loss of the When MDS is uncertain, especially in cases of ICUS or IDUS
same chromosome. Complex abnormalities are defined as three with normal karyotype, analysis of somatic mutations and flow
or more independent abnormalities in at least two metaphases. cytometry analysis of marrow cells could be useful to ascertain
An adequate cytogenetic report should contain a correct diagnosis. Acquired mutations, especially in genes involved
formula describing the karyotype according to the most recent in epigenetic regulation and chromatin remodelling (TET2,
ISCN criteria [25]. Cytogenetic analysis should follow minimal DNMT3A, ASXL1, IDH1/2, EZH2), pre-mRNA splicing factors
standards fixed by the ‘Workpackage Cytogenetics’ of the (SF3B1, SRSF2, U2AF1) transcription [TP53, RUNX1, (II) (III)]
European Leukaemia Net (Figure 1). These relate to cell cultur- and signalling transduction (e.g. NRAS, CBL) are seen in most
ing, analytical expenditure and the use of additional fluores- MDS, and can demonstrate clonal disease [31, 32] (Table 3),
cence in situ hybridisation analyses [26]. while some flow cytometry abnormalities of myeloid precursors
Cytogenetics: At least 2 different cell cultures (24 h and 48 h)

(if 4 cultures were established additional 24 h and 72 h cultures are
recommendable)
Analysis of at least 10 (–20) metaphases (if abnormal),

or 20 (–25) if a normal karyotype is diagnosed
Normal karyotype or
insufficient number of Abnormal karyotype
metaphases
FISH
Optional:
for all subtypes:
Probes for 5q31, cen7, 7q31, cen8,
TP53, 20q, cenY
Documentation and reporting

according to ISCN
Figure 1. Recommended standard algorithm for cytogenetic analysis in myelodysplastic syndromes [26]. FISH, fluorescence in situ hybridisation.
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu180 | iii

by guest
on 05 February 2018
Table 3. Most frequent somatic mutations observed in MDS (other

recommended for predicting outcome and planning treatment
mutations are seen in <5% of the cases) [38]. The IPSS-R is used to stratify patients into five risk groups
(very low, low, intermediate, high and very high risk), with clear
Gene function Gene Mutation differences in OS and risk of progression to AML [8]. Of note,
frequency currently conventional IPSS remains the most widely used
Epigenetic regulators TET 2 15%–25% system, and the system used by health agencies for drug approval
and chromatin- ASXL1 10%–20% in MDS. For therapeutic purposes, IPSS low and intermediate-1
remodelling factors DNMT3a 10% patients are generally grouped in ‘lower risk’ MDS, and inter-
IDH1/2 5%–10% mediate-2 and high-risk patients in ‘higher risk’ MDS.
Pre-mRNA splicing SF3B1 15%–30% Patient-related characteristics such as age [8], Eastern
factors SRSF2 10%–15% Cooperative Oncology Group (ECOG) performance status
U2AF1 5%–10% [8] and comorbidities [39] are also relevant for establishing
Transcription factors RUNX 1 10%–15% prognosis and treatment choice, particularly in lower risk MDS.
TP 53 5%–10% Other disease-related factors include multi-lineage dysplasia
Signaling moleculesa N RAS/K RAS 10% [40], red blood cell (RBC) transfusion dependence [40], serum
LDH, ferritin and β2-microglobulin [7] and bone marrow fibrosis
a
In this group, NPM1 mutations and FLT3 duplications are rare in [39]. Finally, flow cytometry immunophenotyping [33] and gene
MDS, and suggest imminent progression to AML; JAK 2 mutations somatic mutation profiling (especially the most frequent, i.e. TET 2,
are also rare, contrary to myeloproliferative neoplasms. SF3B1, SRSF2, ASXL1, RUNX1, DNMT3a, EZH2, TP53 and
MDS, myelodysplastic syndrome; AML, acute myeloid leukaemia.
RAS, mutations) may also improve risk stratification [2, 31, 32],
but data on their independent prognostic impact are still lacking
to recommend use in routine practice. The prognostic impact of
combined mutations as well as combinations of gene mutations
are highly suggestive of MDS, although both procedures are still with distinct karyotype abnormalities is also unclear at this time.
used only in a minority of centres [33]. Finally, most prognostic factors in MDS have been established
independently of treatment, particularly in cohorts receiving
mostly supportive care. With the availability of treatments having
classification
an impact on disease evolution, including alloSCT and hypo-
A WHO classification of MDS, published in 2001 [19] and methylating agents (HMAs), factors that may be prognostic of
updated in 2008 (Table 4) [34], divides MDS with <5% blasts outcome in patients treated with these treatments are starting to be
into those with either unilineage or multi-lineage dysplasia. defined.
Within the subgroup of MDS with unilineage dysplasia, patients
with ring sideroblasts ( pure sideroblastic anaemia) have a very
low AML progression rate and an excellent overall survival response criteria in MDS
(OS). Refractory anaemia with excess blasts (RAEB) patients
Response criteria to treatment, in MDS, are based on recom-
were sub-divided in RAEB-1 and RAEB-2, i.e. patients with
mendations of an IWG (most recently updated in 2006), that
5%–9% blasts and patients with 10%–19% blasts (20% and more
define two types of responses. The first type considers responses
blasts are now considered to be AML patients). While all of
to treatments aimed at modifying the disease course (mainly
these subgroups can be defined solely on a morphological basis,
alloSCT, intensive chemotherapy and HMAs), and includes
the entity of del(5q) MDS is not defined by morphological
complete remission (CR), partial remission (PR), stable disease
criteria, but by the occurrence of isolated del(5q), making cyto-
and progression. The second type evaluates improvement of
genetic analysis mandatory (see above). Finally, chronic myelo-
cytopaenias (‘haematological improvement’ or HI) in one, two
monocytic leukaemia was excluded and was moved to the
or three lineages (erythroid, platelet and neutrophil responses),
subgroup of myelodysplastic/myeloproliferative neoplasms.
and is particularly adapted to treatments which, like growth
factors, can improve these cytopaenias, but with no obvious
prognosis and risk assessment effect on the disease course. While CR and PR are generally
associated with improvement in cytopaenias, the second type of
The natural course of MDS is highly variable, with survival response is often designed as ‘stable disease with HI (on the
ranging from a few weeks to several years [35]. The median OS erythroid and/or platelet and/or neutrophil) lineage.’
is 15–30 months and the risk of progression to AML is 25%–
35% at 5 years [36]. Bone marrow failure (infection and haem-
orrhage) is the leading cause of death, with more patients dying treatment of IPSS INT-2 and high-risk
before overt AML has occurred [36].
An individual risk-adapted treatment strategy is essential in
(higher risk) MDS patients
MDS. Specific cytogenetic abnormalities, percentage of marrow Although the division is schematic, it is customary since publi-
blasts and number and severity of cytopaenias are the main prog- cation of the classical IPSS to separate MDS into ‘higher risk’
nostic factors in MDS. The IPSS [7] and its recent revision (IPSS- MDS (corresponding to IPSS high or intermediate-2) and ‘lower
R, Table 5) [8] are based on these three variables. They have been risk’ (corresponding to IPSS low or intermediate-1). Higher risk
validated in external series [37], and their use is strongly MDS carry a major risk of progression to AML and short
iii | Fenaux et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
Table 4. The WHO classification of myelodysplastic syndromes [34]
Disease Blood findings Bone marrow findings
Refractory cytopaenias with Unicytopaenia or Unilineage dysplasia: ≥10% of the cells in one myeloid lineage
unilineage dysplasia (RCUD) bicytopaeniaa <5% blasts
Refractory anaemia (RA) No or rare blasts (<1%)b <15% of erythroid precursors are ring sideroblasts
Refractory neutropaenia (RN)
Refractory thrombocytopaenia (RT)
Refractory anaemia with ring Anaemia ≥15% of erythroid precursors are ring sideroblasts
sideroblasts (RARS) No blasts Erythroid dysplasia only
<5% blasts
Refractory cytopaenia with multi- Cytopaenia(s) Dysplasia in ≥10% of cells in two or more myeloid lineages
lineage dysplasia (RCMD) No or rare blasts (<1%)b <5% blasts in marrow
No Auer rods No Auer rods
<1 × 109/l monocytes ±15% ring sideroblasts
Refractory anaemia with excess blasts- Cytopaenia(s) Unilineage or multi-lineage dysplasia
1 (RAEB-1) <5% blastsb 5%–9% blastsb
No Auer rods No Auer rods
<1 × 109/l monocytes
Refractory anaemia with excess blasts- Cytopaenia(s) Unilineage or multi-lineage dysplasia
2 (RAEB-2) 5%–19% blasts 10%–19% blasts
Auer rods±c Auer rods±c
<1 × 109/l monocytes
Myelodysplastic syndrome: Cytopaenias Unequivocal dysplasia in <10% of cells in one or more myeloid cell lines when
unclassified (MDS-U) ≤1% blastsb accompanied by a cytogenetic abnormality considered as presumptive evidence
for a diagnosis of MDSd
<5% blasts
MDS associated with isolated del(5q) Anaemia Normal to increased megakaryocytes with hypolobulated nuclei <5% blasts
Usually normal or No Auer rods
increased platelet count Isolated del(5q) cytogenetic abnormality
No or rare blasts (<1%)
a
Bicytopaenia may occasionally be observed. Cases with pancytopaenia should be classified as MDS-U.
b
If the marrow myeloblast percentage is <5% but there are 2%–4% myeloblasts in the blood, the diagnostic classification is RAEB-1. Cases of RCUD
and RCMD with 1% myeloblasts in the blood should be classified as MDS-U.
c
Cases with Auer rods and <5% myeloblasts in the blood and <10% in the marrow should be classified as RAEB-2.
d
Unbalanced abnormalities: −7 or del(7q), −5 or del(5q), i(17q) or t(17p), −13 or del(13q), del(11q), del(12p) or t(12p), del(9q), idic(X)(q13).
Balanced abnormalities: t(11;16)(q23;p13.3), t(3;21)(q26.2;q22.1), t(1;3)(p36.3;q21.2), inv(3)(q21q26.2), t(2;11)(p21;q23), t(6;9)(p23;q34).
Not considered as definitive evidence for MDS: +8, del(20q), –Y.
survival, and treatment in those patients should aim, whenever Due to the fact that most patients respond only after several
possible, to modify the natural disease course including alloSCT, courses, at least six courses of azacitidine are recommended,
HMAs and, although now less often, chemotherapy (mainly in- with the following schedule: azacitidine 75 mg/m²/day s.c. for
tensive anthracycline–AraC combinations) [41]. In most higher 7 days every 28 days [II, B] in order to properly evaluate its
risk MDS, HMAs are the first-line reference treatment. efficacy. Alternative schedules (such as 5-day regimens), which
appear to give similar response rates as the classical 7-day
hypomethylating agents regimen in lower risk MDS, have not demonstrated their efficacy
In patients with MDS IPSS INT2 high risk, without major coin terms of survival advantage in higher risk MDS.
morbidities and not eligible for alloSCT, azacitidine is recom- Besides induction of CR and PR, achievement of HI according
mended [I, A]. The use of azacitidine may be recommended to IWG 2006 criteria, i.e. an improvement in cytopaenias
compared with the other HMA decitabine, because in a rando- (mainly anaemia and/or thrombocytopaenia), should be consid-
mised trial, azacitidine has been shown to be superior to con- ered indicative of response, because it has been shown to be
ventional care regimens (i.e. supportive care, low-dose AraC associated with a prolongation of survival [III, B] [44].
and AML-like chemotherapy) [42, 43], whereas there was no The use of azacitidine before HSC transplantation (HSCT)
clear survival advantage with decitabine over conventional appears promising and is currently being evaluated in clinical
treatment in two phase III trials. trials.
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu180 | iii

by guest
on 05 February 2018
Table 5. Revised International Prognostic Scoring System (IPSS-R) for myelodysplastic syndromes [8]
Prognostic characteristics Points

0 0.5 1 1.5 2 3 4
a
Cytogenetic risk category Very good Good Intermediate Poor Very poor
Blasts in bone marrow, % ≤2 >2%–5% 5%–10% >10%
Haemoglobin, g/dl ≥10 8–<10 <8
Platelet count, ×109/l ≥100 50–<100 <50
Absolute neutrophil count, ×109/l ≥0.8 <0.8
IPSS-R risk group Score Median overall survival, years Median time to 25% AML evolution, years
Very low ≤1.5 8.8 NR
Low >1.5–3 5.3 9.4
Intermediate >3–4.5 3.0 2.5
High >4.5–6 1.6 1.7
Very high >6 0.8 0.7
a
Very good: –Y and del(11q) as single abnormalities; good: normal, del(5q), del(12p) and del(20q) as single abnormalities, double abnormalities
including del(5q); intermediate: del(7q), +8, +19, i(17q) and any other single abnormalities, any other double abnormalities; poor: −7 and inv(3)/t(3q)/
del(3q) as single abnormalities, double abnormalities including −7/del(7q), complex (3 abnormalities); very poor; >3 abnormalities.
AML-like chemotherapy are reached in 15%–20% of cases with significant myelosuppres-

AML-like intensive chemotherapy has limited indication in sive effects [48, 49].
higher risk MDS patients. In particular, MDS patients with un-
favourable karyotype show few CRs and shorter CR duration second-line treatment
[45]. This treatment can be envisaged for younger patients (gen- IPSS higher risk MDS patients who fail to respond to azacitidine
erally <60–65 years of age) with favourable cytogenetics accord- or are primary refractory to HMAs have an extremely poor sur-
ing to IPSS categories and marrow blasts >10%, preferably as a vival (median <6 months) except for patients potentially eligible
bridge to alloSCT [I, B]. for alloSCT [50]. Retreatment with AML-like chemotherapy or
Suggested regimens with equivalent efficacy deduced from low-dose AraC yields dismal results. The recommended ap-
retrospective analyses are combinations of cytarabine with idar- proach is to enrol such patients in a clinical trial with investiga-
ubicin, fludarabine or topotecan [IV, B] [41]. No improvement tional agents and, if the patient has become eligible for alloSCT,
of outcome was reported by the addition of granulocyte–macro- proceed to transplant [IV, B] [50].
phage colony-stimulating factor (GM-CSF) or granulocyte
colony-stimulating factor (G-CSF) [IV, C]. allogeneic stem cell transplantation
A direct comparison of the efficacy of AML-like chemother- At present, alloSCT is the only potentially curative treatment of
apy versus azacitidine has been carried out in a small number of higher risk MDS patients [I, A]. The major obstacle to alloSCT is
MDS patients and only in one randomised phase III trial: it sug- the fact that most MDS patients are above the age of 70 years.
gested a superiority of the HMA without reaching statistical Co-morbidity, age, IPSS and IPSS-R score, cytogenetics, condi-
significance, but the number of patients was too small for any tioning regimen and donor selection are predictors of post-trans-
conclusion. A retrospective comparison of AML-like chemo- plant outcome [51, 52] and should be taken into account carefully
therapy versus decitabine was carried out [46] in two groups of during the decision process. All patients diagnosed with higher
matched MDS patients and, while CR rates were equivalent, risk MDS aged <65–70 years (although particularly ‘fit’ patients
survival advantage was obtained only with the use of an HMA. aged >70 years may sometimes be considered) should be evaluated
for alloSCT eligibility and human leukocyte antigen (HLA)-identi-
cal (or single antigen mismatched) siblings or matched unrelated
low-dose chemotherapy individuals should be considered as suitable donors [I, A] [52].
Low-dose cytarabine at the most prevalent schedules and dose Regarding conditioning regimens, there is no randomised
(AraC 20 mg/m2/day, 14–21 days/every 4 weeks) was found to trial comparing reduced-intensity conditioning (RIC) to mye-
be significantly inferior to azacitidine (in terms of response and loablative approaches. At present, relapse seems to be higher in
survival) in a randomised phase III study [42], especially in patients receiving RIC; therefore, patients aged <55 years and
patients with unfavourable cytogenetics. However, low-dose without co-morbidities should probably be offered myeloabla-
cytarabine may still be a treatment option in higher risk MDS tive HSCT [IV, C].
patients with normal karyotype [47] who are not candidates for Alternative sources such as cord blood should be further eval-
any intensive chemotherapy or alloSCT, in particular when ad- uated in clinical trials [II, C] [52]. It is debated whether treat-
ministration of azacitidine or decitabine is not possible (includ- ment aimed at reducing the blast count should be carried out
ing for economic reasons) [IV, C]. In these patients, CR and PR before alloSCT with AML-like chemotherapy or HMAs. This is
iii | Fenaux et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
generally considered when marrow blasts are >10%, especially transfusion independence (RBC-TI) of 2–2.5 years [I, A] [70, 71].
for non-myeloablative alloSCT. The recommended initial dose is 10 mg/day, 3 weeks out of every 4
weeks [71]. Cytogenetic response (CyR) is achieved in 50%–75% of
treatment of lower risk MDS subjects (including 30%–45% complete CyR). TP53 gene
mutations, found in ∼20% of lower risk MDS with del 5q, seem to
In lower risk MDS, the risk of AML progression is smaller and confer resistance to LEN and a higher risk of AML progression
survival longer than in higher risk MDS, with about one-half of [72], and their presence may require more aggressive treatment.
elderly patients dying from causes other than the consequences of Grade 3 or 4 neutropaenia and thrombocytopaenia, seen in ∼60%
MDS or AML [53]. In lower risk MDS, the main priority is gener- of patients during the first weeks of treatment, constitute the most
ally the treatment of cytopaenias, mainly of anaemia (usually the common adverse events associated with LEN [70, 71]. Close
predominant cytopaenia), and the improvement in quality of life monitoring of blood counts is therefore required during this
(QoL). Still, some of these patients may be identified as carrying period, with dose reduction and/or addition of G-CSF if required.
poorer prognosis, either rapidly by their revised IPSS score [8] or
by other biological characteristics [54, 55], or subsequently by
their resistance to first-line treatment [56], and may benefit from second-line treatments for anaemia of lower risk
treatments generally applied to higher risk MDS. MDS
Anaemia, because of failure of specific treatments, often even- *patients without del 5q. Treatment after ESA failure ( primary
tually requires repeated RBC transfusions, leading to potential resistance or relapse after a response) in patients who remain
iron overload [42]. with IPSS low or intermediate-1 MDS is still disappointing
overall, with most patients eventually requiring long-term RBC
treatment of anaemia transfusions. Second-line treatments currently used include
anti-thymocyte globulin (ATG), HMAs and LEN.
RBC transfusions or drugs? Immunosuppressive drugs, including anti-lymphocyte or ATG,
Chronic RBC transfusions could be considered as the sole with or without ciclosporin, can yield an erythroid response (asso-
approved treatment of anaemia of lower risk MDS, as very few ciated with response of other cytopaenias, especially thrombocyto-
drugs are approved in this situation and none has been demon- paenia), in 25%–40% of the patients treated [73–77]. ATG results
strated to improve survival. However, chronic RBC transfusions are better in relatively young (<65 years) low-risk MDS patients
are associated with chronic anaemia, leading to excess morbid- with a RBC transfusion history of <2 years, with normal karyotype
ity, and they cannot completely correct impaired QoL [57, 58]. (or possibly trisomy 8), with no excess blasts, HLA DR15 geno-
Although this remains disputed (see below), iron overload due type, and possibly in patients with thrombocytopaenia in addition
to RBC transfusions may also be deleterious to various organs to anaemia, a small paroxysmal nocturnal haemoglobinuria clone
[57, 59]. Finally, researchers have found recently that, in lower or with marrow hypocellularity [III, B] [73]. Therefore, this treat-
risk MDS with anaemia, receiving ESAs had no impact on pro- ment is generally proposed to a relatively small minority of
gression to AML but was an independent, favourable prognostic patients. HMAs have been reported to yield RBC-TI in 30%–40%
factor for survival [IV, B] [60–63]. of the patients [78, 79], and may also be effective on other cytopae-
nias in lower risk MDS [III, B]. They are approved in this situation
first-line treatment of anaemia in lower risk MDS in several countries, including the United States.
LEN yields RBC-TI in 25%–30% of lower risk MDS without
*patients without del(5q): ESAs. ESAs, i.e. recombinant EPO or
del 5q resistant to ESA [80, 81], and the combination of LEN
darbepoetin (DAR), remain the first choice treatment of
and ESA may yield higher RBC-TI rates than LEN alone in
anaemia in most lower risk MDS without del(5q) [60]. Weekly
patients resistant to an ESA alone [I, B] [82].
doses of 30 000–80 000 units of EPO or 150–300 μg of DAR alfa
injection yield ∼60% of erythroid responses, according to IWG
2006 response criteria [64], when the baseline EPO level is low *patients with del 5q. Resistance to LEN in lower risk MDS
and transfusion requirement is absent or limited, which is now with del 5q is associated with poor prognosis, even if no
the case in most patients where this treatment is considered immediate progression to high-risk MDS is observed. Patients
[I, A] [60–62, 65, 66]. The efficacy of ESAs can be further with TP53 gene mutation may have a particularly poor outcome
improved by the addition of G-CSF [67, 68], and there are no [72]. Although no prospective data exist, these patients are likely
data showing that one ESA is superior to another. to be suitable candidates for approaches having demonstrated
Responses to ESA occur within 8–12 weeks of treatment. a survival benefit in MDS, including HMAs and, whenever
Median duration of response to ESA is ∼2 years, with longer possible, alloSCT [IV, B] [82].
responses in patients with major response according to IWG
2000 criteria [64], IPSS low or intermediate-1, marrow blasts treatment of neutropaenia
<5% and no multi-lineage dysplasia [60–62].
and thrombocytopaenia
*lower risk MDS with del 5q: lenalidomide. Anaemia of lower In lower risk MDS, neutropaenia and thrombocytopaenia are
risk MDS with del 5q, compared with that of other lower risk MDS, less frequent than anaemia, and are infrequently isolated or
shows lower response rates and significantly shorter responses to profound.
ESA [69]. However after ESA failure, it responds to lenalidomide White blood cells are <1.500 mm3 in only 7% of lower risk
(LEN) in 60%–65% of the subjects, with a median duration of RBC MDS [83], and neutropaenia is rarely associated with life
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu180 | iii

by guest
on 05 February 2018
Higher-risk MDS
(IPSS high and int 2)
Patients aged Patients >65 or 70 years Frail patients

<65 to 70 years or without donor for
(or sometimes allogeneic SCT
slightly older if
“fit”) with a
donor Age <60–65 Other patients
years and no
unfavourable
karyotype
Allogeneic SCT AML-like Azacitidine (at least 6 Supportive care

(preceded or not by CT or chemotherapy cycles) (RBC transfusions,
HMA to reduce blast %) or In case of failure or antibiotics...)
azacitadine relapse, consider clinical
trials or symptomatic
treatment
Figure 2. Treatment algorithm for higher risk myelodysplastic syndromes.
threatening infection if no drugs worsening neutropaenia are dependent anaemia not responding to any of the agents
used. G-CSF and GM-CSF can improve neutropaenia in 60%– described above [90]. In patients requiring repeated RBC
75% of these cases, and can be considered in the treatment of transfusions, it is recommended to administer transfusions at
neutropaenic fever in addition to anti-infective drugs, but their sufficiently high haemoglobin threshold, i.e. at least 8 g/dl, and
prolonged use has not had a demonstrated impact on survival. 9 g/dl or even 10 g/dl in case of co-morbidities worsened by
Platelets <50 000/mm3 are seen in ∼30% of low-risk MDS [83]. anaemia or in case of poor functional tolerance and/or poor
High-dose androgens can improve thrombocytopaenia in about QoL or in elderly persons who are still very active. A sufficient
one-third of thrombocytopaenic lower risk MDS, but response is number of RBC concentrates should be transfused each time,
generally transient [III, C] [84–86]. The thrombopoietic (TPO) if necessary over 2 or 3 days, to increase the haemoglobin
receptor-agonist romiplostim at high dose (500–1.500 μg/week) level >10 g/dl, and thereby limit the effects of chronic anaemia,
yielded 55% platelet responses in a phase II trial in lower risk especially on QoL [IV, A].
MDS with thrombocytopaenia. However, in ∼15% of the patients, Except in patients receiving myelosuppressive drugs, prophy-
a transient rise in marrow blasts was seen, which was reversible lactic platelet transfusions are less used than RBC transfusions
after drug discontinuation. In a randomised phase II study versus in MDS, especially in the long term. Likewise, prophylactic anti-
placebo in lower risk MDS with thrombocytopaenia, romiplostim biotics and/or G-CSF are not recommended in case of neutro-
significantly reduced the incidence of severe bleeding and platelet paenia, as they have not shown any impact on survival, but
transfusions [87]. While there was a suspected increase in the rapid onset of broad spectrum antibiotics is mandatory in these
AML risk upon first analysis, this was not confirmed by later patients in case of fever or symptoms of infection. Short-term
follow-up [87]. Eltrombopag, the other available TPO receptor use of G-CSF during severe infections could be useful in neu-
agonist, is also currently being tested in both lower risk tropaenic patients, but this indication has not been validated.
and higher risk MDS. Lower risk MDS patients seem to be par- Psychosocial support and contacts with patient support groups
ticularly responsive to treatment with eltrombopag [88]. (when they exist) should be systematically offered.
ATG and HMAs appear to give platelet response in 35%–40% A large debate exists about the deleterious effect of iron over-
of the cases of lower risk MDS, in addition to erythroid load in MDS patients and whether iron chelation may be useful
responses [III, C] [74, 77, 89]. See Figures 2 and 3. in patients with iron overload. In particular, while heart iron
overload is a well-documented cause of heart failure in children
supportive care and chelation therapy with thalassaemia [91, 92], its incidence and clinical conse-
quences are less certain in MDS patients receiving transfusions,
in MDS particularly as many already have other causes of cardiac morbid-
Supportive care is required in all patients with MDS at some ity [40, 93]. However, heart MRI studies show that heart iron
point of the disease, and may be the only treatment of some overload (reflected by a decrease in MRI heart T2*) is frequent in
patients in the long term, especially those with transfusion- patients having received at least 70–80 RBC concentrates or more,
iii | Fenaux et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
Lower-risk MDS
(IPSS low and int 1)
Symptomatic Symptomatic anaemia Moderate and

thrombocytopaenia (generally if Hb below asymptomatic cytopaenias
10g/dl)
No treatment
- TPO agonist RBC transfusions ≥ 2 RBC transfusions < 2
receptor in a clinical concentrates/month concentrates/month
trial and serum EPO >500 u/l or serum EPO <500 u/l
- ATG if favourable
feature (see text)
- azacitidine if
del 5q No del 5 q EPO +/– G-CSF
approved or in
EPO +/– G-CSF (low
clinical trial
success rate)
lenalidomide If failure or relapse
or second-line treatment
Second-line
Symptomatic No del 5q del 5q
treatment
neutropaenia:
- broad spectrum
antibiotics if fever lenalidomide
Age <60–65 Others
-short term use of and favourable features azacitidine (if approved)
G-CSF ? for response to ATG or clinical trial (with
- ATG if favourable (see text) azacitidine, lenalidomide
feature (see text) +/– EPO or experimental
- azacitidine if drug
approved or in ATG
clinical trial
Figure 3. Treatment algorithm for lower risk myelodysplastic syndromes.
a frequent situation in low-risk MDS, and that a heart T2* value personalised medicine
<20 ms is associated with decreased left ventricular ejection frac-
tion and a risk of heart failure [94]. It has been suggested in retro- While many prognostic factors have been established in MDS,
spective studies that adequate chelation in highly transfused as seen above, most of them have been defined irrespective of
patients may improve their survival [IV, C] [95–97]. Prospective treatment, in patient cohorts that mainly received supportive
randomised studies are underway to confirm those results. care, and it is often unclear if they are predictive of the efficacy
In the absence of prospective studies, published recommenda- of a given treatment. Furthermore, in spite of recent improve-
tions for iron chelation therapy so far only result from expert ments, there are still too few effective treatment options in MDS,
opinions [V] [98], which generally advocate starting chelation and there is limited choice for most patients.
in patients with relatively favourable prognosis (i.e. low or in- The classical IPSS offers a valuable patient stratification, and
termediate-1-risk MDS), who have received 20–60 RBC con- this is why it served as a basis in Figures 2 and 3 summarising
centrates, or if serum ferritin raises above 1000–2500 U/l, or if treatment indications: e.g. anaemia of IPSS low and intermedi-
cardiac T2* is significantly reduced. Future candidates to alloSCT ate-1 MDS often responds to ESAs, except in case of del 5q,
should be chelated early. Indeed, although the underlying where LEN is very efficacious. On the other hand, in IPSS inter-
mechanisms are unclear, it appears that even relatively moderate mediate-2 and high-risk patients, while azacitidine has shown it
iron overload before alloSCT is associated with increased trans- could improve survival, there are currently limited other options
plant-related mortality [II, B] [99–101]. (except alloSCT, possible in a minority of patients).
Iron chelation is now made easier by the availability of oral Given the usual age of MDS patients, considering the patient’s
iron chelators (especially deferasirox), in addition to the classical age, general condition and co-morbidities is also crucial before
parenteral deferoxamine. Deferasirox is however frequently making any treatment decisions.
associated with gastrointestinal side-effects, and cannot be used
in patients with renal failure [102]. Deferiprone, another oral
iron chelator, is currently not approved for MDS in most coun-
patient follow-up
tries, and can cause neutropaenia in a small percentage of Except for follow-up of specific treatments, follow-up of MDS is
patients, a side-effect that is problematic in MDS [103]. largely based on regular blood counts, to detect anaemia that
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu180 | iii

by guest
on 05 February 2018
Levels of evidence
II Small randomised trials or large randomised trials with a suspicion of bias (lower methodological quality) or meta-analyses of such trials or
of trials with demonstrated heterogeneity
C Insufficient evidence for efficacy or benefit does not outweigh the risk or the disadvantages (adverse events, costs, … ), optional
a
will require RBC transfusions, while severe thrombocytopaenia 3. Shen J, Wang S, Zhang YJ et al. Genome-wide aberrant DNA methylation of
may require platelet transfusions and severe neutropaenia man- microRNA host genes in hepatocellular carcinoma. Epigenetics 2012; 7:
1230–1237.
dates preventive measures against infection (e.g. during invasive
4. Will B, Zhou L, Vogler TO et al. Stem and progenitor cells in myelodysplastic
procedures), or, more importantly, rapid onset of broad spec-
syndromes show aberrant stage-specific expansion and harbor genetic and
trum antibiotics in case of fever or symptoms of infection. epigenetic alterations. Blood 2012; 120: 2076–2086.
Bone marrow examination, with or without karyotype, is 5. Jiang Y, Dunbar A, Gondek LP et al. Aberrant DNA methylation is a dominant
generally triggered by worsening of cytopaenias or appearance of mechanism in MDS progression to AML. Blood 2009; 113: 1315–1325.
circulating blasts rather than systematically carried out at regular 6. Vardiman JW, Thiele J, Arber DA et al. The 2008 revision of the World Health
intervals. Organization (WHO) classification of myeloid neoplasms and acute leukemia:
rationale and important changes. Blood 2009; 114: 937–951.
7. Greenberg P, Cox C, LeBeau MM et al. International scoring system for evaluating
note prognosis in myelodysplastic syndromes. Blood 1997; 89: 2079–2088.
Levels of evidence and grades of recommendation have been 8. Greenberg PL, Tuechler H, Schanz J et al. Revised international prognostic
applied using the system shown in Table 6. Statements without scoring system for myelodysplastic syndromes. Blood 2012; 120: 2454–
grading were considered justified standard clinical practice by 2465.
the experts and the ESMO faculty. 9. Neukirchen J, Schoonen WM, Strupp C et al. Incidence and prevalence of
myelodysplastic syndromes: data from the Dusseldorf MDS-registry. Leuk Res
2011; 35: 1591–1596.
conflict of interest 10. Matsuda A, Germing U, Jinnai I et al. Difference in clinical features between
Japanese and German patients with refractory anaemia in myelodysplastic
VS has reported honoraria received for lecturing from: Janssen- syndromes. Blood 2005; 106: 2633–2640.
Cilag, GlaxoSmithKline, Novartis and Celgene. GS has reported 11. Qu S, Xu Z, Zhang Y et al. Impacts of cytogenetic categories in the Revised
Amgen (honoraria, advisory board, research funding); Boehringer- International Prognostic Scoring System on the prognosis of primary
Ingelheim (advisory board); Celgene (honoraria, advisory board, myelodysplastic syndromes: results of a single-center study. Leuk Lymphoma
2012; 53: 940–946.
research funding); Jansen-Cilag (honoraria, advisory board);
12. Leone G, Fianchi L, Voso MT. Therapy-related myeloid neoplasms. Curr Opin
Merck Sharp & Dohme (advisory board); Mirati Therapeutics, (ad- Oncol 2011; 23: 672–680.
visory board); and Novartis (honoraria, advisory board, research 13. Cardis E, Vrijheid M, Blettner M et al. Risk of cancer after low doses of ionising
funding). CB has reported consultancy/honoraria from Celgene, radiation: retrospective cohort study in 15 countries. BMJ 2005; 331: 77.
Pfizer and Roche. The other authors have reported no potential 14. Iwanaga M, Hsu WL, Soda M et al. Risk of myelodysplastic syndromes in people
conflicts of interest. exposed to ionizing radiation: a retrospective cohort study of Nagasaki atomic
bomb survivors. J Clin Oncol 2011; 29: 428–434.
15. Nisse C, Haguenoer JM, Grandbastien B et al. Occupational and environmental
references risk factors of the myelodysplastic syndromes in the north of France. Br J
1. Tefferi A, Lim KH, Levine R. Mutation in TET2 in myeloid cancers. N Engl J Med Haematol 2001; 112: 927–935.
2009; 361: 1117–1118. 16. Aksoy M, Ozeris S, Sabuncu H et al. Exposure to benzene in Turkey between
2. Bejar R, Stevenson K, Abdel-Wahab O et al. Clinical effect of point mutations in 1983 and 1985: a haematological study on 231 workers. Br J Ind Med 1987;
myelodysplastic syndromes. N Engl J Med 2011; 364: 2496–2506. 44: 785–787.
iii | Fenaux et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
17. Rigolin GM, Cuneo A, Roberti MG et al. Exposure to myelotoxic agents and 38. Malcovati L, Hellström-Lindberg E, Bowen D et al. Diagnosis and treatment of
myelodysplasia: case-control study and correlation with clinicobiological findings. primary myelodysplastic syndromes in adults: recommendations from the
Br J Haematol 1998; 103: 189–197. European LeukemiaNet. Blood 2013; 122: 2943–2964.
18. Valent P, Horny HP, Bennet JM et al. Definitions and standards in the diagnosis 39. Della Porta MG, Malcovati L, Boveri E et al. Clinical relevance of bone marrow
and treatment of the myelodysplastic syndromes: consensus statements and fibrosis and CD34-positive cell clusters in primary myelodysplastic syndromes. J
report from a working conference. Leuk Res 2007; 31: 727–736. Clin Oncol 2009; 27: 754–762.
19. Jaffe ES, Harris NL, Stein H et al. (eds). World Health Organization Classification 40. Malcovati L, Germing U, Kuendgen A et al. Time-dependent prognostic scoring
of Tumours, Pathology and Genetics of Haematopoietic and Lymphoid Tissues. system for predicting survival and leukemic evolution in myelodysplastic
Lyon: IARC Press, 2008, pp. 75–104. syndromes. J Clin Oncol 2007; 25: 3503–3510.
20. Valent P, Horny HP. Minimal diagnostic criteria for myelodysplastic syndromes 41. Estey EH, Thall PF, Cortes JE et al. Comparison of idarubicin + ara-C-,
and separation from ICUS and IDUS: update and open questions. Eur J Clin Invest fludarabine + ara-C-, and topotecan + ara-C-based regimens in treatment of
2009; 39: 548–553. newly diagnosed acute myeloid leukemia, refractory anaemia with excess blasts
21. Germing U, Hildebrandt B, Pfeilstöcker M et al. Refinement of the international in transformation, or refractory anaemia with excess blasts. Blood 2001; 98:
prognostic scoring system (IPSS) by including LDH as an additional prognostic 3575–3583.
variable to improve risk assessment in patients with primary myelodysplastic 42. Fenaux P, Mufti GJ, Hellström-Lindberg E et al. Efficacy of azacitidine compared
syndromes (MDS). Leukemia 2005; 19: 2223–2231. with that of conventional care regimens in the treatment of higher-risk
22. Haase D, Germing U, Schanz J et al. New insights into the prognostic impact of myelodysplastic syndromes: a randomised, open-label, phase III study. Lancet
the karyotype in MDS and correlation with subtypes: evidence from a core Oncol 2009; 10: 223–232.
dataset of 2124 patients. Blood 2007; 110: 4385–4395. 43. Gurion R, Vidal L, Gafter-Gvili A et al. 5-azacitidine prolongs overall survival in
23. Mohamedali A, Gäken J, Twine NA et al. Prevalence and prognostic significance patients with myelodysplastic syndrome—a systematic review and meta-analysis.
of allelic imbalance by single-nucleotide polymorphism analysis in low-risk Haematologica 2010; 95: 303–310.
myelodysplastic syndromes. Blood 2007; 110: 3365–3373. 44. Gore SD, Fenaux P, Santini V et al. A multivariate analysis of the relationship
24. Boultwood J, Wainscoat JS. Gene silencing by DNA methylation in between response and survival among patients with higher-risk myelodysplastic
haematological malignancies. Br J Haematol 2007; 138: 3–11. syndromes treated within azacitidine or conventional care regimens in the
25. Shaffer LG, McGowan-Jordan J, Schmid M (eds). ISCN: an International System randomized AZA-001 trial. Haematologica 2013; 98: 1067–1072.
for Human Cytogenetic Nomenclature. Basel: S. Karger, 2013. 45. Knipp S, Hildebrand B, Kündgen A et al. Intensive chemotherapy is not
26. Haferlach C, Rieder H, Lillington DM et al. Proposals for standardized protocols recommended for patients aged >60 years who have myelodysplastic syndromes or
for cytogenetic analyses of acute leukemias, chronic lymphocytic leukemia, acute myeloid leukemia with high-risk karyotypes. Cancer 2007; 110: 345–352.
chronic myeloid leukemia, chronic myeloproliferative disorders, and 46. Kantarijan HM, O’Brien S, Huang X et al. Survival advantage with decitabine versus
myelodysplastic syndromes. Genes Chromosomes Cancer 2007; 46: intensive chemotherapy in patients with higher risk myelodysplastic syndrome:
494–499. comparison with historical experience. Cancer 2007; 109: 1133–1137.
27. Solé F, Luño E, Sanzo C et al. Identification of novel cytogenetic markers with 47. Zwierzinq H, Suciu S, Loeffler-Ragg J et al. Low-dose cytosine arabinoside
prognostic significance in a series of 968 patients with primary myelodysplastic (LD-AraC) vs LD-AraC plus granulocyte/macrophage colony stimulating factor vs
syndromes. Haematologica 2005; 90: 1168–1178. LD-AraC plus Interleukin-3 for myelodysplastic syndrome patients with a high
28. Schanz J, Steidl C, Fonatsch C et al. Coalesced multicentric analysis of 2,351 risk of developing acute leukemia: final results of a randomized phase III study
patients with myelodysplastic syndromes indicates an underestimation of poor- (06903) of the EORTC Leukemia Cooperative Group. Leukemia 2005; 19:
risk cytogenetics of myelodysplastic syndromes in the international prognostic 1929–1933.
scoring system. J Clin Oncol 2011; 29: 1963–1970. 48. Cheson BD, Simon R. Low-dose ara-C in acute nonlymphocytic leukemia and
29. Valcárcel D, Ademà V, Solé F et al. Complex, not monosomal, karyotype is the myelodysplastic syndromes: a review of 20 years’ experience. Semin Oncol
cytogenetic marker of poorest prognosis in patients with primary myelodysplastic 1987; 14(2 Suppl 1):126–133.
syndrome. J Clin Oncol 2013; 31: 916–922. 49. Fenaux P, Lai JL, Gardin C et al. Cytogenetics are a predictive factor of response
30. Schanz J, Tüchler H, Solé F et al. Monosomal karyotype in MDS: explaining the to low dose Ara-C in acute myelogenous leukemia (AML) in the elderly. Leukemia
poor prognosis? Leukemia 2013; 27: 1988–1995. 1990; 4: 312.
31. Yoshida K, Sanada M, Shiraishi Y et al. Frequent pathway mutations of splicing 50. Prebet T, Gore SD, Esterni B et al. Outcome of high-risk myelodysplastic
machinery in myelodysplasia. Nature 2011; 478: 64–69. syndrome after azacitidine treatment failure. J Clin Oncol 2011; 29: 3322–3327.
32. Papaemmanuil E, Gerstung M, Malcovati L et al. Clinical and biological 51. Brand R, Putter H, van Biezen A et al. Comparison of allogeneic stem cell
implications of driver mutations in myelodysplastic syndromes. Blood 2013; 122: transplantation and non-transplant approaches in elderly patients with advanced
3616–3627. myelodysplastic syndrome: optimal statistical approaches and a critical appraisal
33. Westers TM, Ireland R, Kern W et al. Standardization of flow cytometry in of clinical results using non-randomized data. PLoS One 2013; 8: e74368.
myelodysplastic syndromes: a report from an international consortium 52. Vaughn JE, Scott BL, Deeg HJ. Transplantation for myelodysplastic syndromes
and the European LeukemiaNet Working Group. Leukemia 2012; 26: 1730– 2013. Curr Opin Hematol 2013; 20: 494–500.
1741. 53. Guardiola P, Runde V, Bacigalupo A et al. Retrospective comparison of bone
34. Swerdlow SH, Campo E, Harris NL et al. (eds). WHO Classification of Tumours of marrow and granulocyte colony-stimulating factor-mobilized peripheral blood
Haematopoietic and Lymphoid Tissues, 4th edition. Lyon: IARC Press, 2008. progenitor cells for allogeneic stem cell transplantation using HLA identical sibling
35. Tefferi A, Vardiman JW. Myelodysplastic syndromes. N Engl J Med 2009; 361: donors in myelodysplastic syndromes. Blood 2002; 99: 4370–4378.
1872–1885. 54. Mittelman M, Oster HS, Hoffman M, Neumann D. The lower risk MDS patient at
36. Sanz GF, Sanz MA, Vallespi T et al. Two regression models and a scoring system for risk of rapid progression. Leuk Res 2010; 34: 1551–1555.
predicting survival and planning treatment in myelodysplastic syndromes: a 55. Bejar R, Stevenson KE, Caughey BA et al. Validation of a prognostic model and
multivariate analysis of prognostic factors in 370 patients. Blood 1989; 74: the impact of mutations in patients with lower-risk myelodysplastic syndromes. J
395–408. Clin Oncol 2012; 30: 3376–3382.
37. Voso MT, Fenu S, Latagliata R et al. Revised International Prognostic Scoring 56. Kelaidi C, Park S, Sapena R et al. Long-term outcome of anemic lower-risk
System (IPSS) predicts survival and leukemic evolution of myelodysplastic myelodysplastic syndromes without 5q deletion refractory to or relapsing after
syndromes significantly better than IPSS and WHO Prognostic Scoring System: erythropoiesis-stimulating agents. Leukemia 2013; 27: 1283–1290.
validation by the Gruppo Romano Mielodisplasie Italian Regional Database. J Clin 57. Fenaux P, Rose C. Impact of iron overload in myelodysplastic syndromes. Blood
Oncol 2013; 31: 2671–2677. Rev 2009; 23 (Suppl 1): S15–S19.
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu180 | iii

by guest
on 05 February 2018
58. Crawford J, Cella D, Cleeland CS et al. Relationship between changes in 9221 by the Cancer and Leukemia Group B. J Clin Oncol 2006; 24:
hemoglobin level and quality of life during chemotherapy in anemic cancer 3895–3903.
patients receiving epoetin alfa therapy. Cancer 2002; 95: 888–895. 79. Silverman LR, Demakos EP, Peterson BL et al. Randomized controlled trial of
59. Roy NB, Myerson S, Schuh AH et al. Cardiac iron overload in transfusion- azacitidine in patients with the myelodysplastic syndrome: a study of the cancer
dependent patients with myelodysplastic syndromes. Br J Haematol 2011; 154: and leukemia group B. J Clin Oncol 2002; 20: 2429–2440.
521–524. 80. Raza A, Reeves JA, Feldman EJ et al. Phase 2 study of lenalidomide in
60. Park S, Grabar S, Kelaidi C et al. Predictive factors of response and survival in transfusion-dependent, low-risk, and intermediate-1 risk myelodysplastic
myelodysplastic syndrome treated with erythropoietin and G-CSF: the GFM syndromes with karyotypes other than deletion 5q. Blood 2008; 111: 86–93.
experience. Blood 2008; 111: 574–582. 81. Sibon D, Cannas G, Baracco F et al. Lenalidomide in lower-risk myelodysplastic
61. Jadersten M, Malcovati L, Dybedal I et al. Erythropoietin and granulocyte-colony syndromes with karyotypes other than deletion 5q and refractory to
stimulating factor treatment associated with improved survival in myelodysplastic erythropoiesis-stimulating agents. Br J Haematol 2012; 156: 619–625.
syndrome. J Clin Oncol 2008; 26: 3607–3613. 82. Komrokji RS, List AF. Lenalidomide for treatment of myelodysplastic syndromes.
62. Greenberg PL, Sun Z, Miller KB et al. Treatment of myelodysplastic Curr Pharm Des 2012; 18: 3198–3203.
syndrome patients with erythropoietin with or without granulocyte colony- 83. Garcia-Manero G, Shan J, Faderl S et al. A prognostic score for patients with
stimulating factor: results of a prospective randomized phase 3 trial by the lower risk myelodysplastic syndrome. Leukemia 2008; 22: 538–543.
Eastern Cooperative Oncology Group (E1996). Blood 2009; 114: 2393– 84. Montero AJ, Estrov Z, Freireich EJ et al. Phase II study of low-dose interleukin-11
2400. in patients with myelodysplastic syndrome. Leuk Lymphoma 2006; 47:
63. Golshayan AR, Jin T, Maciejewski J et al. Efficacy of growth factors compared to 2049–2054.
other therapies for low-risk myelodysplastic syndromes. Br J Haematol 2007; 85. Wattel E, Cambier N, Caulier MT et al. Androgen therapy in myelodysplastic
137: 125–132. syndromes with thrombocytopenia: a report on 20 cases. Br J Haematol 1994;
64. Cheson BD, Greenberg PL, Bennett JM et al. Clinical application and proposal for 87: 205–208.
modification of the International Working Group (IWG) response criteria in 86. Bourgeois E, Caulier MT, Rose C et al. Role of splenectomy in the treatment of
myelodysplasia. Blood 2006; 108: 419–425. myelodysplastic syndromes with peripheral thrombocytopenia: a report on six
65. Mannone L, Gardin C, Quarre MC et al. High-dose darbepoetin alpha in the cases. Leukemia 2001; 15: 950–953.
treatment of anaemia of lower risk myelodysplastic syndrome results of a phase II 87. Giagounidis A, Mufti GJ, Fenaux P et al. Results of a randomized, double-blind
study. Br J Haematol 2006; 133: 513–519. study of romiplostim versus placebo in patients with low/intermediate-1-risk
66. Gabrilove J, Paguette R, Lyons RM et al. Phase 2, single-arm trial to evaluate the myelodysplastic syndrome and thrombocytopenia. Cancer 2014; 120:
effectiveness of darbepoetin alfa for correcting anaemia in patients with 1838–1846.
myelodysplastic syndromes. Br J Haematol 2008; 142: 379–393. 88. Oliva EN, Santini V, Zini G et al. Eltrombopag for the treatment of
67. Hellstrom-Lindberg E, Birgegård G, Carlsson M et al. A combination of thrombocytopenia of low and intermediate-1 IPSS risk myelodysplastic
granulocyte colony-stimulating factor and erythropoietin may synergistically syndromes: results of a prospective, randomized, trial. Haematologica 2013; 98
improve the anaemia in patients with myelodysplastic syndromes. Leuk (Suppl 1): s1110.
Lymphoma 1993; 11: 221–228. 89. Stadler M, Germing U, Kliche KO et al. A prospective, randomised, phase II study
68. Hellstrom-Lindberg E, Gulbrandsen N, Lindberg G et al. A validated decision of horse antithymocyte globulin vs rabbit antithymocyte globulin as immune-
model for treating the anaemia of myelodysplastic syndromes with modulating therapy in patients with low-risk myelodysplastic syndromes.
erythropoietin + granulocyte colony-stimulating factor: significant effects on Leukemia 2004; 18: 460–465.
quality of life. Br J Haematol 2003; 120: 1037–1046. 90. Kelaidi C, Beyne-Rauzy O, Braun T et al. High response rate and improved
69. Kelaidi C, Park S, Brechignac S et al. Treatment of myelodysplastic syndromes exercise capacity and quality of life with a new regimen of darbepoetin alfa with
with 5q deletion before the lenalidomide era; the GFM experience with EPO and or without filgrastim in lower-risk myelodysplastic syndromes: a phase II study by
thalidomide. Leuk Res 2008; 32: 1049–1053. the GFM. Ann Hematol 2013; 92: 621–631.
70. List A, Dewald G, Bennett J et al. Lenalidomide in the myelodysplastic syndrome 91. Davis BA, O’Sullivan C, Jarritt PH et al. Value of sequential monitoring of left
with chromosome 5q deletion. N Engl J Med 2006; 355: 1456–1465. ventricular ejection fraction in the management of thalassemia major. Blood
71. Fenaux P, Giagounidis A, Selleslag D et al. A randomized phase 3 study of 2004; 104: 263–269.
lenalidomide versus placebo in RBC transfusion-dependent patients with low-/ 92. Kirk P, Roughton M, Porter JB et al. Cardiac T2* magnetic resonance for
intermediate-1-risk myelodysplastic syndromes with del5q. Blood 2011; 118: prediction of cardiac complications in thalassemia major. Circulation 2009; 120:
3765–3776. 1961–1968.
72. Jadersten M, Saft L, Smith A et al. TP53 mutations in low-risk myelodysplastic 93. Malcovati L, Porta MG, Pascutto C et al. Prognostic factors and life expectancy in
syndromes with del(5q) predict disease progression. J Clin Oncol 2011; 29: myelodysplastic syndromes classified according to WHO criteria: a basis for
1971–1979. clinical decision making. J Clin Oncol 2005; 23: 7594–7603.
73. Sauntharajah Y, Nakamura R, Wesley R et al. A simple method to predict 94. Jensen PD, Jensen FT, Chistensen T et al. Evaluation of myocardial iron by
response to immunosuppressive therapy in patients with myelodysplastic magnetic resonance imaging during iron chelation therapy with deferrioxamine:
syndrome. Blood 2003; 102: 3025–3027. indication of close relation between myocardial iron content and chelatable iron
74. Yazji S, Giles FJ, Tsimberidou AM et al. Antithymocyte globulin (ATG)-based pool. Blood 2003; 101: 4632–4639.
therapy in patients with myelodysplastic syndromes. Leukemia 2003; 17: 95. Rose C, Brechignac S, Vassilief D et al. Does iron chelation therapy improve
2101–2106. survival in regularly transfused lower risk MDS patients? A multicenter study by
75. Molldrem JJ, Caples M, Mavroudis D et al. Antithymocyte globulin for patients the GFM (Groupe Francophone des Myelodysplasies). Leuk Res 2010; 34:
with myelodysplastic syndrome. Br J Haematol 1997; 99: 699–705. 864–870.
76. Sloand EM, Wu CO, Greenberg P et al. Factors affecting response and survival in 96. Leitch HA. Improving clinical outcome in patients with myelodysplastic syndrome
patients with myelodysplasia treated with immunosuppressive therapy. J Clin and iron overload using iron chelation therapy. Leuk Res 2007; 31 (Suppl 3):
Oncol 2008; 26: 2505–2511. S7–S9.
77. Lim ZY, Killick S, Germing U et al. Low IPSS score and bone marrow 97. Neukirchen J, Fox F, Kündgen A et al. Improved survival in MDS patients
hypocellularity in MDS patients predict hematological responses to antithymocyte receiving iron chelation therapy—a matched pair analysis of 188 patients from
globulin. Leukemia 2007; 21: 1436–1441. the Dusseldorf MDS registry. Leuk Res 2012; 36: 1067–1070.
78. Silverman LR, McKenzie DR, Peterson BL et al. Further analysis of trials with 98. Greenberg PL. Myelodysplastic syndromes: iron overload consequences and
azacitidine in patients with myelodysplastic syndrome: studies 8421, 8921, and current chelating therapies. J Natl Compr Canc Netw 2006; 4: 91–96.
iii | Fenaux et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
99. Armand P, Kim HT, Cutler CS et al. Prognostic impact of elevated 102. Gattermann N, Finelli C, Porta MD et al. Deferasirox in iron-overloaded
pretransplantation serum ferritin in patients undergoing myeloablative stem cell patients with transfusion-dependent myelodysplastic syndromes: results
transplantation. Blood 2007; 109: 4586–4588. from the large 1-year EPIC study. Leuk Res 2010; 34: 1143–1150.
100. Pullarkat V. Objectives of iron chelation therapy in myelodysplastic syndromes: 103. Cermak J, Jonasova A, Vondrakova J et al. Efficacy and safety of administration
more than meets the eye? Blood 2009; 114: 5251–5255. of oral iron chelator deferiprone in patients with early myelodysplastic syndrome.
101. Platzbecker U, Bornhäuser M, Germing U et al. Red blood cell transfusion Hemoglobin 2011; 35: 217–227.
dependence and outcome after allogeneic peripheral blood stem cell 104. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
transplantation in patients with de novo myelodysplastic syndrome (MDS). Biol among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
Blood Marrow Transplant 2008; 14: 1217–1225. 139–144.
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu180 | iii

by guest
on 05 February 2018
Newly diagnosed and relapsed follicular lymphoma:

ESMO Clinical Practice Guidelines for diagnosis,
M. Dreyling1, M. Ghielmini2, S. Rule3, G. Salles4, U. Vitolo5 & M. Ladetto6, on behalf of the ESMO
1
Department of Medicine III, University of Munich, Germany; 2Oncology Institute of Southern Switzerland, Ospedale San Giovanni, Bellinzona, Switzerland; 3Haematology,
Peninsula School of Medicine, Plymouth, UK; 4Hospices Civils de Lyon, Centre Hospitalier Lyon-Sud, Service d’Hématologie & Université Claude Bernard Lyon-1, Pierre-
Benite, France; 5Haematology, University-Hospital Città della Salute e della Scienza, Torino; 6Divisione di Ematologia, Azienda Ospedaliera Santi Antonio e Biagio e Cesare
Arrigo, Alessandria, Italy
incidence and epidemiology staging and risk assessment

Follicular lymphomas (FLs) are the second most frequent Since treatment largely depends on the stage of the disease, initial
subtype of nodal lymphoid malignancies in Western Europe. staging should be thorough, particularly in the small proportion
clinical practice
The annual incidence of this disease has rapidly increased of patients with early stages I and II (10%–15%) (Table 2). Initial
guidelines
during recent decades and has risen from 2–3/100 000 during work-up should include a computed tomography (CT) scan of
the 1950s to 5/100 000 recently [1]. the neck, thorax, abdomen and pelvis, and a bone marrow aspir-
ate and biopsy (Table 3). Positron emission tomography (PET)–
CT improves the accuracy of staging for nodal and extranodal
sites and thus should be recommended for routine staging in FL
biology [IV, C] [4]. This is particularly important to confirm localised
Diagnosis should be based on a surgical specimen/excisional stage I/II before involved-field radiotherapy.
lymph node biopsy. Core biopsies should only be carried A complete blood count, routine blood chemistry including
out in patients without easily accessible lymph nodes (e.g. lactate dehydrogenase (LDH), β2 microglobulin and uric acid as
retroperitoneal bulk), keeping in mind the possible hetero- well as screening tests for human immunodeficiency virus
geneity of FL grading can be difficult to appreciate on core (HIV), hepatitis B virus (HBV) and hepatitis C are required.
biopsies and re-biopsy may be required if the material is not The staging is carried out according to the Ann Arbor classifica-
adequate. Fine needle aspirations are inappropriate for a reli- tion system (Table 2), with mention of bulky disease (>7 cm)
able diagnosis. when appropriate.
The histological report should give the diagnosis according For prognostic purposes, a ‘Follicular Lymphoma-specific
to the World Health Organization (WHO) classification. International Prognostic Index’ (FLIPI, Table 4) has been
Grading of lymph node biopsies is carried out according to established [I, A] [6]. A revised FLIPI 2 (incorporating β2
the number of blasts/high-power field (Table 1). FL grade 3A microglobulin, diameter of largest lymph node, bone marrow
(with sheets of blasts) is considered an aggressive lymphoma involvement and haemoglobin level) has been suggested for
and treated as such [2], whereas grade 1, 2 and 3A should be patients requiring treatment which may be more informative on
treated as indolent disease [3]. Review, especially of grade 3A progression-free survival (PFS) [7].
or 3B, by an expert haematopathologist is advised if the infil- Extended gene-expression profiling of tumour biopsy suggests
tration pattern is atypical (diffuse areas, even with small a more favourable clinical course in cases with infiltrating
cells). T cells, in comparison with cases with non-specific macrophage
bystander cells [5]. Recently, a clinicogenetic risk score (m7-
FLIPI) has been proposed based on mutation status of seven
candidate genes (EZH2, ARID1A, MEF2B, EP300, FOXO1,
E-mail: clinicalguidelines@esmo.org CREBBP and CARD11) [8]; however, none of the techniques are
yet established in clinical routine practice. In addition, several
†
recent immunohistochemistry studies have reported conflicting
This publication supersedes the previously published version—Ann Oncol 2014; 25
(Suppl. 3): iii76–82. data; hence, biological parameters are still investigational for

by guest
on 05 February 2018
Table 1. Grading of follicular lymphoma (FL) Table 4. ‘Follicular Lymphoma-specific International Prognostic
Grade Description Index’ (FLIPI) risk factors
Parameter Definition of risk factors
1 ≤5 blasts/high-power field
FLIPI 1 FLIPI 2
2 6–15 blasts/high-power field
3A >15 blasts/high-power field, centroblasts with intermingled Nodal sites >4 lymph node regions Long diameter of largest
centrocytes (definition in [5]) lymph node >6 cm
3B >15 blasts/high-power field, pure sheets of blasts
Age >60 years >60 years
Serum marker Elevated LDH Elevated β2
microglobulin
Table 2. Ann Arbor classification
Stage Advanced (III–IV according Bone marrow
Stage Area of involvement to Ann Arbor involvement
I (IE) One lymph node region or extralymphatic site (IE) classification)
II (IIE) Two or more lymph node regions or at least one lymph Haemoglobin <12 g/dl <12 g/dl
node region plus a localised extralymphatic site(IIE) on
the same side of the diaphragm 0–1 risk factors, low risk; 2 risk factors, intermediate risk; 3–5 risk
III (IIIE, IIIS) Lymph node regions or lymphoid structures (e.g. thymus, factors, high risk.
Waldeyer’s ring) on both sides of the diaphragm with LDH, lactate dehydrogenase.
optional localised extranodal site (IIIE) or spleen (IIIS)
IV Diffuse or disseminated extralymphatic organ
involvement Table 5. High tumour burden criteria in FL [Groupe d’Etude des
Lymphomes Folliculaires (GELF)
A: no symptoms.
Parameter High tumour burden criteria
B: unexplained fever of >38°C, drenching night sweats; or loss of >10%
body weight within 6 months. Lymph nodes Bulk (>7 cm) or 3 lymph nodes in distinct areas
>3 cm
Spleen Symptomatic splenic enlargement
(Potential) Organ compression by tumour, pleural or
complication peritoneal effusion
Table 3. Diagnostic work-up
Serum markers Elevated LDH or elevated β2-microglobuline
History B symptoms Clinical presentation B symptoms (see Table 2)
Physical examination Peripheral lymph nodes, liver, spleen
LDH, lactate dehydrogenase.
Laboratory work-up Blood and differential count
Optional: FACS on peripheral blood, PCR for
BCL2 rearrangement
prognostic assessment and are not yet suitable for clinical deci-
LDH, uric acid
sion-making [9]. If possible, additional biopsy material should
Electrophoresis (optional: immune fixation)
β2 microglobulin (FLIPI 2)
be stored fresh frozen to allow for the possible future application
of additional molecular analyses.
Serology Hepatitis B, C and HIV serology
Imaging CT neck, chest, abdomen, pelvis
treatment
Recommended: PET–CTa
Optional: abdominal ultrasound first line
Bone marrowb Histology stage I–II. In the small proportion of patients with limited
Cytology non-bulky stages I–II, radiotherapy (involved field, 24 Gy) is the
Optional: FACS, PCR for BCL2 rearrangement preferred treatment with a potentially curative potential, whereas
Toxicity Electrocardiogram, cardiac ultrasound (before the 2 × 2 Gy schedule is inferior and is merely palliative [II, B]
anthracyclines, ASCT) [10]. In selected cases, watchful waiting or rituximab monotherapy
Creatinine clearance may be considered to avoid the side-effects of radiation (e.g.
Reproductive counselling in young patients cervical: sicca syndrome, hypothyroidism; abdominal: mucositis,
myeloablative suppression) [11, 12].
a
To confirm localised disease or in the case of suspected transformation. In stage I–II patients with large tumour burden, adverse clin-
b
If clinically indicated. ical or biological prognostic features or when local radiotherapy
FACS, fluorescence-activated cell sorting; PCR, polymerase chain is not applicable (e.g. lung, liver), systemic therapy as indicated
reaction; LDH, lactate dehydrogenase; FLIPI 2, Follicular Lymphoma for advanced stages should be applied [IV, B] [12].
International Prognostic Index 2; HIV, human immunodeficiency virus;
CT, computed tomography; PET–CT, positron emission tomography– stage III–IV
computed tomography; ASCT, autologous stem cell transplantation. induction: In the majority of patients with advanced stage III
and IV disease, no curative therapy is yet established. Since the
v | Dreyling et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
natural course of the disease is characterised by spontaneous symptoms, including B symptoms, haematopoietic impairment,
regressions in 10%–20% of cases and varies significantly from case bulky disease, vital organ compression, ascites, pleural effusion or
to case, therapy should be initiated only upon the occurrence of rapid lymphoma progression (Table 5) [I, A].
In three randomised trials before the rituximab era, an early
initiation of therapy in asymptomatic patients did not result in
Evaluate
any improvement of disease-specific survival or overall survival
(OS) [13]. In a more recent study, early initiation of rituximab
resulted in improved PFS (80% versus 48%, P <0.001), but no
survival benefit has been determined so far [14], and the benefit
Prognosis Patient priority
Symptoms
Longer survival
of rituximab maintenance in this setting appears doubtful [15].
Stage Not mild
FLIPI 1/2 Long remission Thus, the current therapeutic approach is based on clinical risk
Life/organ threatening
Grade Better quality of life factors, symptoms and patient perspective (Figure 1).
Four prospective first-line trials, two salvage trials and a sys-
tematic meta-analysis confirmed an improved overall response,
PFS and OS if rituximab was added to chemotherapy (Table 6)
Choose among [16–20]. If complete remission and long PFS is to be achieved,
rituximab in combination with chemotherapy such as CHOP
(cyclophosphamide, doxorubicin, vincristine and prednisone) or
bendamustine should be used [I, B] [17, 21]. CVP (cyclophos-
Asymptomatic cases: Mild symptoms: High tumour burden: phamide, vincristine and prednisone) is not as effective as these
Watch and wait Non-chemotherapy Chemoimmunotherapy
treatment R-bendamustine two regimens with respect to PFS but not OS [22]. Full courses of
Rituximab R-CHOP purine analogue-based schemes [FC (fludarabine and cyclophos-
Radioimmunotherapy R-CVP
phamide) or FM (fludarabine and mitoxantrone)] are not recom-
Consider
rituximab maintenance
mended due to higher haematological toxicities, but a brief
(or radioimmunotherapy) course of chemoimmunotherapy with full rituximab course is an
alternative in elderly patients, with good efficacy and low toxicity
Figure 1. Therapeutic algorithm. R-CHOP, rituximab, cyclophosphamide, [II, B] [22, 23]. If there is evidence (histological grade 3B or clin-
doxorubicin, vincristine and prednisolone; R-CVP, rituximab, cyclophos- ical signs of transformation) of more aggressive lymphoma, an
phamide, vincristine and prednisolone. anthracycline-based regimen [rituximab, cyclophosphamide,
Table 6. Combined chemoimmunotherapy in FL (first line)

Study Total no. of patients Median follow-up Overall response Time to treatment failure Overall survival
(months)
Marcus et al. [16] 159 53 months 81% 27 83% (4 years)

R-CVP (P < 0.0001) (P < 0.0001) (P = 0.029)
Hiddemann et al. [17] 223 58 months 96% NR 90% (2 years)
R-CHOP (P < 0.001) (P = 0.0493)
Herold et al. [18] 105 48 months 92% NR 87% (4 years)
R-MCP (P = 0.0009) (P < 0.0001) (P = 0.0096)
Bachy et al. [19] 175 99 months 81% 66 79% (8 years)
R-CHVP-IFN (P = 0.035) (P = 0.0004) (P = 0.076)
Rummel et al. [21] 139 34 months 93% NR 84% (4 years)
BR
Federico et al. [22] 34 months 95% (3 years)
R-CVP 178 88% 46% (3 years)
R-CHOP 178 93% 62% (3 years)
R-FM 178 91% 59% (3 years)
+ R maintenance
Vitolo et al. [23] 234 42 months 86% NR 89% (3 years)
4× R-FND + 4× R
± R maintenance
P, significance levels in comparison with chemotherapy only.

FL, follicular lymphoma; R-CVP, cyclophosphamide, vincristine and prednisolone; R-CHOP, rituximab, cyclophosphamide, doxorubicin, vincristine,
prednisolone; R-MCP: mitoxantrone, chlorambucil, prednisone; R-CHVP-IFN, rituximab, cyclophosphamide, doxorubicin, etoposide, prednisone,
interferon; BR, bendamustine–rituximab; R-FM, rituximab, fludarabine and mitoxantrone; R-FND, cyclophosphamide, vincristine and prednisolone;
NR, not reached.
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw400 | v

by guest
on 05 February 2018
Table 7. Recommended follow-up after end of therapy

Examination Details Year 1–2 Year 3–5 Year >5
History B symptoms Every 3–4 months Twice annually Annually

Physical examination Particular: peripheral lymph nodes, liver, spleen Every 3–4 months Twice annually Annually
Laboratory work-up Blood and differential count Every 3–4 months Twice annually Annually
LDH Every 3–4 months Twice annually If progress suspected
Imaging Abdominal ultrasound Twice annually Every 12 months If progress suspected
CT neck, chest, abdomen, pelvis Optional: 6–12 months Optional: 12–24 months If progress suspected
LDH, lactate dehydrogenase; CT, computed tomography.
doxorubicin, vincristine and prednisolone (R-CHOP)] should be

Low tumour burden
applied.
Antibody monotherapy (rituximab, radioimmunotherapy) or
chlorambucil plus rituximab remain alternatives for patients
with a low-risk profile or when conventional chemotherapy is
contraindicated [III, B] [24, 25]. Stage I/II Stage III/IV
In patients with positive hepatitis B serology including occult

carrier (HBS Ag negative and anti-core positive), prophylactic
antiviral medication and regular monitoring of HBV DNA are Radiotherapy Watch and wait
strongly recommended [I, A] [26]. (involved field) 24 Gy
Front line
In selected cases, In selected cases,

consider watchful consider rituximab
monotherapy
consolidation/maintenance waiting or rituximab
monotherapy
Rituximab maintenance for 2 years improves PFS (59% versus
43% after 6 years, P < 0.0001) [I, B] [27], whereas a shorter
maintenance period results in inferior benefit [28]. Watch and wait
Relapse/progression
Radioimmunotherapy consolidation also prolongs PFS after Watch and wait

Rituximab
chemotherapy, but its benefit seems to be inferior in comparison monotherapy Chemoimmunotherapy
with rituximab maintenance for 2 years [II, B] [29, 30]. In selected cases,
In selected cases,
Myeloablative consolidation followed by autologous stem cell palliative radiation
rituximab monotherapy
(e.g. 2 x 2 Gy)
transplantation (ASCT) prolongs PFS after chemotherapy, but
its benefit after a rituximab-containing induction is minor and no
OS has been observed [31]. Therefore, such an approach is not
Later relapse/progression
Watch and wait

recommended in first-line therapy of responding patients [I, D]. Watch and wait
Rituximab monotherapy
Rituximab monotherapy
Chemoimmunotherapy
relapsed disease In selected cases, (long prior remission)
palliative radiation
At relapse, it is strongly recommended to obtain a new biopsy in (e.g. 2 x 2 Gy)
ldelalisib
(double refractory cases)
order to exclude transformation into an aggressive lymphoma.
It may be useful to target the biopsy based on PET scanning.
Figure 2. Consensus-driven recommendations outside of clinical studies—
As at first presentation, observation is an accepted approach low tumour burden.
in asymptomatic patients with low tumour burden.
Selection of salvage treatment depends on efficacy of prior
regimens. In early relapses (<12–24 months), a non-cross-resistant Radioimmunotherapy (90yttrium–ibritumomab–tiuxetan) may
scheme should be preferred (e.g. bendamustine after CHOP or vice represent an effective therapeutic approach in elderly patients
versa). Other options, including fludarabine-based, platinum salts- with comorbidities not appropriate for chemotherapy [IV, B].
based or alkylating agents-based regimens, could also be useful. Rituximab maintenance for up to 2 years has a favourable
Rituximab should be added if the previous antibody-containing side-effect profile and, based on a systematic meta-analysis, sub-
scheme achieved >6- to 12-month duration of remission [IV, B]. stantially prolongs PFS and OS in relapsed disease, even after
On the other hand, obinutuzumab has recently received a positive antibody-containing induction in patients who have not
recommendation for approval by the European Medicines Agency received antibody as first-line therapy [I, A] [33]. A second-line
for rituximab-refractory cases based on an improved PFS in com- maintenance treatment has not been investigated in the setting
parison with bendamustine only [I, B] [32]. of maintenance use in first line and probably should not be used
In symptomatic cases with low tumour burden, rituximab for those patients who had relapsed during their first mainten-
monotherapy may be applied. ance period [IV, D].
v | Dreyling et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
High-dose chemotherapy with ASCT prolongs PFS and OS acyclovir) is strongly recommended. Cytomegalovirus monitor-
and should be considered, especially in patients who experience ing may be also advised.
short-lived first remissions (<2–3 years) after rituximab-con- In selected younger patients with later relapses of high-risk
taining regimens, which usually have a much worse long-term profile or relapse after ASCT, a potentially curative allogeneic
outcome, but its general role in the rituximab era has to be rede- stem cell transplantation ( preferably with dose-reduced condi-
fined [I, B] [34–37]. A subsequent rituximab maintenance may tioning) may be considered, especially in patients with early
achieve some improvement in PFS [II, B] [38]. relapse and refractory disease [IV, B] [36].
In later relapses, monotherapy is an established option with
palliative intent [II, B]. The PI3K inhibitor idelalisib has been
registered in double-refractory FL, based on a phase II study innovative approaches
[39]. Recent analyses suggest an increased mortality risk as a In recent years, new approaches, including lenalidomide–rituxi-
consequence of pulmonary morbidity (atypical pneumonias/ mab and additional inhibitors of the B-cell signalling pathway,
pneumonitis), so appropriate prophylaxis (cotrimoxazole/ have proved active in phase II studies, but to date their benefit
High tumour burden
Stage Ill/IV Stage Ill/IV

(<65 yearsa) (>65 yearsa)
Chemoimmunotherapy (e.g. BRc, R-CHOPb, R-CVP) or brief

Front line
Chemoimmunotherapy (e.g BR, R-CHOPb, R-CVP)

chemoimmunotherapy
In selected cases, rituximab monotherapy In selected cases,
rituximab-chlorambucil or rituximab monotherapy
CR/PR: CR/PR:
Recommend rituximab maintenance Recommend rituximab maintenance
(every 2 months, up to 2 years) (every 2 months, up to 2 years)
Dependent on first-line regimen and remission duration

Relapse/progression
Dependent on first-line regimen and remission duration

• Chemoimmunotherapy + rituximab
• Chemoimmunotherapy (e.g BR, R-CHOP, R-CVP)
maintenance (every 3 months, up to 2 years)
+/- rituximab maintenance
• Alternatively, radioimmunotherapy
(every 3 months, up to 2 years)
• In early relapses, discuss high-dose
• Alternatively, radioimmunotherapy
consolidation with ASCTb
Dependent on prior regimens and remission duration

Later relapse/progression
• Chemoimmunotherapy (long prior remission) Dependent on prior regimens and remission duration
+ rituximab maintenance (if not previously applied)
• In early relapses, discuss high-dose • Chemoimmunotherapy (long prior remission)
consolidation with ASCTb + rituximab maintenance (if not previously applied)
• Radioimmunotherapy or rituximab monotherapy • Radioimmunotherapy or rituximab monotherapy
• ldelalisib (double refractory cases) • ldelalisib (double refractory cases)
• In selected cases, discuss allogeneic transplantation
Figure 3. Consensus-driven recommendations outside of clinical studies—high tumour burden. R-CHOP, rituximab, cyclophosphamide, doxorubicin, vin-
cristine, prednisolone; R-CVP, cyclophosphamide, vincristine and prednisolone; BR, bendamustine–rituximab; CR, complete response; PR, partial response;
ASCT, autologous stem cell transplantation. aAccording to biological age; bespecially if transformation is suspected; c70–90 mg/m2, 4–6 cycles [42].
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw400 | v

by guest
on 05 February 2018
has yet to be confirmed in randomised phase III studies. The • After (during continuous) systemic treatment: history and phys-
combination of bortezomib–rituximab has shown only a minor ical examination every 3–4 months for 2 years, every 6 months
benefit compared with antibody monotherapy [I, D]. for 3 additional years, and subsequently once a year [V, D].
• Blood count and routine chemistry every 6 months for 2 years,
response evaluation then only as needed for evaluation of suspicious symptoms.
Appropriate imaging evaluation should be carried out midterm • Evaluation of thyroid function in patients with irradiation of
and after completion of chemotherapy. Patients with an inad- the neck at 1, 2 and 5 years.
equate response [less than partial response (PR)] should be eval- • Minimal adequate radiological or ultrasound examinations
uated for early salvage regimens. PR patients may convert to every 6 months for 2 years and optionally annually up to 5
complete response (CR) under rituximab maintenance. years. Regular CT scans are not mandatory outside of clinical
PET–CT after completion of chemotherapy induction has trials, especially if abdominal ultrasound is applicable. PET–
been recommended for prognostic reasons as persistent PET- CT should be not used for surveillance.
positivity (using appropriate Deauville scales) identifies a small • MRD screening may be carried out in clinical studies but
group (20%–25%) of patients with a worse prognosis [40], but should not guide therapeutic strategies.
therapeutic consequences remain undefined [II, B].
Minimal residual disease (MRD) analysis by polymerase
methodology
chain reaction at the end of the treatment is an independent pre-
dictor of long-term outcome, but should not guide therapeutic These clinical practice guidelines were developed in accordance
strategies outside of clinical studies. with the ESMO standard operating procedures for clinical prac-
personalised medicine been selected by the expert authors. A summary of recom-
As various therapeutic approaches may achieve durable mended treatment strategies outside of clinical studies is pro-
responses in the vast majority of patients, the selection of vided in Figures 2 and 3, and a summary of recommendations is
optimal treatment is mainly based on clinical risk factors, symp- provided in Table 8. Levels of evidence and grades of recom-
toms and patient perspective (Figure 1). PET- and MRD-based mendation have been applied using the system shown in
tailored treatments are currently evaluated in studies but are not Table 9. Statements without grading were considered justified
yet routine clinical practice.
Paediatric FL is an FL variant originally described in children, Table 9. Levels of evidence and grades of recommendation
but occurs in adults as well. It is characterised by a localised (adapted from the Infectious Diseases Society of America-United
disease, the absence of bcl-2 aberrations, lack of t(14;18), grade States Public Health Service Grading Systema)
III and a high proliferation rate. It shows a much more indolent
Levels of evidence
course and should be managed with local therapy only, despite
displaying histologically more aggressive features [41]. I Evidence from at least one large randomised, controlled trial
meta-analyses of well-conducted randomised trials without
follow-up and long-term implications and heterogeneity
survivorship II Small randomised trials or large randomised trials with a
The following minimal recommendations are based on consen- suspicion of bias (lower methodological quality) or meta-
sus rather than on evidence (Table 7):
heterogeneity
• After local radiotherapy: history and physical examination III Prospective cohort studies
every 6 months for 2 years, subsequently once a year if clinic- IV Retrospective cohort studies or case–control studies
ally indicated. V Studies without control group, case reports, experts opinions
In localised stages: radiation (24 Gy) strongly recommended
In advanced stages: treatment depends on clinical risk factors, symptoms B Strong or moderate evidence for efficacy but with a limited
and patient perspective clinical benefit, generally recommended
Standard approach in asymptomatic advanced cases: watch and wait C Insufficient evidence for efficacy or benefit does not outweigh
In advanced symptomatic cases the risk or the disadvantages (adverse events, costs, …),
Combined chemoimmunotherapy for long-term remissions optional
Recommend rituximab maintenance for consolidation D Moderate evidence against efficacy or for adverse outcome,
Relapse is frequently sensitive to conventional approaches generally not recommended
Autologous (and allogeneic) transplantation should be only discussed in E Strong evidence against efficacy or for adverse outcome, never
relapse recommended
Monotherapy (antibodies, idelalisib) is appropriate, especially in later
a
relapses By permission of the Infectious Diseases Society of America [43].
v | Dreyling et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
standard clinical practice by the experts and the ESMO faculty. 12. Friedberg JW, Byrtek M, Link BK et al. Effectiveness of first-line management
This manuscript has been subjected to an anonymous peer strategies for stage I follicular lymphoma: analysis of the National LymphoCare
Study. J Clin Oncol 2012; 30: 3368–3375.
review process.
13. Ardeshna KM, Smith P, Norton A et al. Long-term effect of a watch and wait policy
versus immediate systemic treatment for asymptomatic advanced-stage non-
conflict of interest Hodgkin lymphoma: a randomised controlled trial. Lancet 2003; 362: 516–522.
MD has reported institutional research support from Celgene, 14. Ardeshna KM, Qian W, Smith P et al. Rituximab versus a watch-and-wait approach
Janssen, Mundipharma, Pfizer and Roche; speaker’s honoraria in patients with advanced-stage, asymptomatic, non-bulky follicular lymphoma: an
open-label randomised phase 3 trial. Lancet Oncol 2014; 15: 424–435.
from Bayer, Celgene, Gilead, Janssen and Roche; advisory
15. Kahl BS, Hong F, Williams ME et al. Rituximab extended schedule or re-treatment
boards for Bayer, Celgene, Gilead, Janssen, Pfizer and Roche.
trial for low-tumor burden follicular lymphoma: Eastern Cooperative Oncology
MG has reported honoraria from Roche, Mundipharma, Gilead Group protocol e4402. J Clin Oncol 2014; 32: 3096–3102.
and Janssen. SR has reported research support from Janssen, 16. Marcus R, Imrie K, Solal-Céligny P et al. Phase III study of R-CVP compared with
Roche and Celgene; advisory boards for Janssen, Gilead, Roche, cyclophosphamide, vincristine, and prednisolone alone in patients with previously
AstraZeneca and Celgene; speaker's honoraria from Janssen, untreated advanced follicular lymphoma. J Clin Oncol 2008; 26: 4579–4586.
Celgene and Roche. GS has reported personal fees for advisory 17. Hiddemann W, Kneba M, Dreyling M et al. Frontline therapy with rituximab added
boards or participation in meetings from Amgen, Celgene, to the combination of cyclophosphamide, doxorubicin, vincristine, and prednisone
Gilead, Janssen, Mundipharma, Novartis and Roche; grant (CHOP) significantly improves the outcome of patients with advanced stage
follicular lymphoma compared with therapy with CHOP alone: results of a
support from Roche. UV has reported advisory boards for
prospective randomized study of the German Low-Grade Lymphoma Study Group.
Roche and Janssen; honoraria for lectures from Roche, Janssen, Blood 2005; 106: 3725–3732.
Celgene, Takeda; conducting research as a global PI in multicen- 18. Herold M, Haas A, Srock S et al. Rituximab added to first-line mitoxantrone,
tre studies sponsored by Roche and Celgene. ML has reported chlorambucil, and prednisolone chemotherapy followed by interferon maintenance
institutional research support from Celgene, Janssen, prolongs survival in patients with advanced follicular lymphoma: an East German
Mundipharma, Pfizer and Roche; speaker’s honoraria from Study Group Hematology and Oncology Study. J Clin Oncol 2007; 25:
Bayer, Celgene, Gilead, Janssen and Roche; advisory boards for 1986–1992.
Bayer, Celgene, Gilead, Janssen, Pfizer and Roche. 19. Bachy E, Houot R, Morschhauser F et al. Long-term follow up of the FL2000 study
comparing CHVP-interferon to CHVP-interferon plus rituximab in follicular
lymphoma. Haematologica 2013; 98: 1107–1114.
references 20. Schulz H, Bohlius JF, Trelle S et al. Immunochemotherapy with rituximab and
1. Mounier M, Bossard N, Remontet L et al. Changes in dynamics of excess mortality overall survival in patients with indolent or mantle cell lymphoma: a systematic
rates and net survival after diagnosis of follicular lymphoma or diffuse large B-cell review and meta-analysis. J Natl Cancer Inst 2007; 99: 706–714.
lymphoma: comparison between European population-based data (EUROCARE-5). 21. Rummel M, Niederle N, Maschmeyer G et al. Bendamustine plus rituximab versus
Lancet Haematol 2015; 2: e481–e491. CHOP plus rituximab as first-line treatment for patients with indolent and mantle-
2. Tilly H, Gomes da Silva M, Vitolo U et al. Diffuse large B-cell lymphoma (DLBCL): cell lymphomas: an open-label, multicentre, randomised, phase 3 non-inferiority
ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann trial. Lancet 2013; 381: 1203–1210.
Oncol 2015; 26(Suppl. 5): v116–v125. 22. Federico M, Luminari S, Dondi A et al. R-CVP versus R-CHOP versus R-FM for the
3. Ott G, Katzenberger T, Lohr A et al. Cytomorphologic, immunohistochemical, and initial treatment of patients with advanced-stage follicular lymphoma: results of the
cytogenetic profiles of follicular lymphoma: 2 types of follicular lymphoma grade 3. FOLL05 trial conducted by the Fondazione Italiana Linfomi. J Clin Oncol 2013; 31:
Blood 2002; 99: 3806–3812. 1506–1513.
4. Cheson BD, Fisher RI, Barrington SF et al. Recommendations for initial evaluation, 23. Vitolo U, Ladetto M, Boccomini C et al. Rituximab maintenance compared with
staging, and response assessment of Hodgkin and non-Hodgkin lymphoma—the observation after brief first-line R-FND chemoimmunotherapy with rituximab
Lugano Classification. J Clin Oncol 2014; 32: 3059–3068. consolidation in patients age older than 60 years with advanced follicular
5. Dave SS, Wright G, Tan B et al. Prediction of survival in follicular lymphoma based lymphoma: a phase III randomized study by the Fondazione Italiana Linfomi. J Clin
on molecular features of tumor-infiltrating immune cells. N Engl J Med 2004; 351: Oncol 2013; 31: 3351–3359.
2159–2169. 24. Martinelli G, Schmitz SF, Utiger U et al. Long-term follow-up of patients with
6. Solal-Céligny P, Roy P, Colombat P et al. Follicular lymphoma international follicular lymphoma receiving single-agent rituximab at two different schedules in
prognostic index. Blood 2004; 104: 1258–1265. trial SAKK 35/98. J Clin Oncol 2010; 28: 4480–4484.
7. Federico M, Bellei M, Marcheselli L et al. Follicular lymphoma international 25. Scholz CW, Pinto A, Linkesch W et al. (90)Yttrium-ibritumomab-tiuxetan as first-
prognostic index 2: a new prognostic index for follicular lymphoma developed by line treatment for follicular lymphoma: 30 months of follow-up data from an
the international follicular lymphoma prognostic factor project. J Clin Oncol 2009; international multicenter phase II clinical trial. J Clin Oncol 2013; 31: 308–313.
27: 4555–4562. 26. Huang YH, Hsiao LT, Hong YC et al. Randomized controlled trial of entecavir
8. Pastore A, Jurinovic V, Kridel R et al. Integration of gene mutations in risk prophylaxis for rituximab-associated hepatitis B virus reactivation in patients with
prognostication for patients receiving first-line immunochemotherapy for follicular lymphoma and resolved hepatitis B. J Clin Oncol 2013; 31: 2765–2772.
lymphoma: a retrospective analysis of a prospective clinical trial and validation in a 27. Salles G, Seymour JF, Feugier P et al. Updated 6 year follow-up of the PRIMA
population-based registry. Lancet Oncol 2015; 16: 1111–1122. study confirms the benefit of 2-year rituximab maintenance in follicular lymphoma
9. Sander B, de Jong D, Rosenwald A et al. The reliability of immunohistochemical patients responding to frontline immunochemotherapy. Blood 2013; 122: abstr.
analysis of the tumor microenvironment in follicular lymphoma: a validation study 509.
from the Lunenburg Lymphoma Biomarker Consortium. Haematologica 2014; 99: 28. Taverna CJ, Martinell G, Hitz F et al. Rituximab maintenance treatment for a
715–725. maximum of 5 years in follicular lymphoma: results of the randomized phase III
10. Hoskin PJ, Kirkwood AA, Popova B et al. 4 Gy versus 24 Gy radiotherapy for trial SAKK 35/03. Blood 2013; 122: abstr. 508.
patients with indolent lymphoma (FORT): a randomised phase 3 non-inferiority 29. Morschhauser F, Radford J, Van Hoof A et al. 90Yttrium-ibritumomab-tiuxetan
trial. Lancet Oncol 2014; 15: 457–463. consolidation of first remission in advanced-stage follicular non-Hodgkin
11. Solal-Céligny P, Bellei M, Marcheselli L et al. Watchful waiting in low-tumor burden lymphoma: updated results after a median follow-up of 7.3 years from the
follicular lymphoma in the rituximab era: results of an F2-study database. J Clin international, randomized, phase III first-line indolent trial. J Clin Oncol 2013; 31:
Oncol 2012; 30: 3848–3853. 1977–1983.
Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw400 | v

by guest
on 05 February 2018
30. Lopez-Guillermo A, Canales MA, Dlouhy I et al. A randomized phase II study 36. Montoto S, Corradini P, Dreyling M et al. Indications for hematopoietic stem cell
comparing consolidation with a single dose of 90y ibritumomab tiuxetan (Zevalin®) transplantation in patients with follicular lymphoma: a consensus project of the
(Z) vs. maintenance with rituximab (R) for two years in patients with newly EBMT-Lymphoma Working Party. Haematologica 2013; 98: 1014–1021.
diagnosed follicular lymphoma (FL) responding to R-CHOP. Preliminary results at 37. Casulo C, Byrtek M, Dawson KL et al. Early relapse of follicular lymphoma after
36 months from randomization. Blood 2013; 122: abstr. 369. rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone defines
31. Hiddemann W, Dreyling MH, Metzner B et al. Evaluation of myeloablative therapy patients at high risk for death: an analysis from the National LymphoCare Study.
followed by autologous stem cell transplantation in first remission in patients with J Clin Oncol 2015; 33: 2516–2522.
advanced stage follicular lymphoma after initial immuno-chemotherapy (R-CHOP) 38. Pettengell R, Schmitz N, Gisselbrecht C et al. Rituximab purging and/or maintenance
or chemotherapy alone: analysis of 940 patients treated in prospective randomized in patients undergoing autologous transplantation for relapsed follicular lymphoma: a
trials of the German Low Grade Lymphoma Study Group (GLSG). Blood 2013; prospective randomized trial from the Lymphoma Working Party of the European
122: abstr. 419. Group for Blood and Marrow Transplantation. J Clin Oncol 2013; 31: 1624–1630.
32. Sehn LH, Chua N, Mayer J et al. Obinutuzumab plus bendamustine versus 39. Gopal AK, Kahl BS, de Vos S et al. PI3Kδ inhibition by idelalisib in patients with
bendamustine monotherapy in patients with rituximab-refractory indolent non- relapsed indolent lymphoma. N Engl J Med 2014; 370: 1008–1018.
Hodgkin lymphoma (GADOLIN): a randomised, controlled, open-label, multicentre, 40. Trotman J, Luminari S, Boussetta S et al. Prognostic value of PET-CT after first-
phase 3 trial. Lancet Oncol 2016; 17: 1081–1093. line therapy in patients with follicular lymphoma: a pooled analysis of central scan
33. Vidal L, Gafter-Gvili A, Salles G et al. Rituximab maintenance for the treatment of review in three multicentre studies. Lancet Haematol 2014; 1: e17–e27.
patients with follicular lymphoma: an updated systematic review and meta-analysis 41. Louissaint A, Jr, Ackerman AM, Dias-Santagata D et al. Pediatric-type nodal
of randomized trials. J Natl Cancer Inst 2011; 103: 1799–1806. follicular lymphoma: an indolent clonal proliferation in children and adults with high
34. Schouten HC, Qian W, Kvaloy S et al. High-dose therapy improves progression-free proliferation index and no BCL2 rearrangement. Blood 2012; 120: 2395–2404.
survival and survival in relapsed follicular non-Hodgkin’s lymphoma: results from 42. Cheson BD, Brugger W, Damaj G et al. Optimal use of bendamustine in
the randomized European CUP trial. J Clin Oncol 2003; 21: 3918–3927. hematologic disorders: treatment recommendations from an international
35. Le Gouill S, De Guibert S, Planche L et al. Impact of the use of autologous stem consensus panel—an update. Leuk Lymphoma 2016; 57: 766–782.
cell transplantation at first relapse both in naive and previously rituximab exposed 43. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
follicular lymphoma patients treated in the GELA/GOELAMS FL2000 study. among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
Haematologica 2011; 96: 1128–1135. 139–144.
v | Dreyling et al. Volume 27 | Supplement 5 | September 2016

by guest
on 05 February 2018
Gastric marginal zone lymphoma of MALT type:

E. Zucca1, C. Copie-Bergman2, U. Ricardi3, C. Thieblemont4, M. Raderer5 & M. Ladetto6, on behalf
of the ESMO Guidelines Working Group*
1
Lymphoma Unit, Oncology Institute of Southern Switzerland, Bellinzona, Switzerland; 2APHP, Groupe Henri Mondor-Albert Chenevier, Département de Pathologie, Créteil,
France; 3Department of Oncology, Radiation Oncology Unit, University of Turin, Turin, Italy; 4Department of Hematology, APHP-Saint-Louis Hospital, University Paris-Diderot,
Paris, France; 5Department of Internal Medicine I and Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria; 6Division of Hematology, Department of
Experimental Medicine and Oncology, University of Turin, San Giovanni Battista Hospital, Turin, Italy
incidence diagnosis nor pathognomonic, as they can be seen under

some reactive conditions as well as in other lymphoma
Extranodal marginal zone lymphomas of mucosa-associated subtypes.
lymphoid tissue (MALT) type represent ∼7% of all non- If the presence of active H. pylori infection is not
Hodgkin’s lymphomas in the western world and can arise at any demonstrated by histochemistry, it must be ruled out by
clinical practice
extranodal site. At least one-third of them present as a primary serology, urea breath test and/or stool antigen test [5, 6].
guidelines
gastric lymphoma, which in approximately two-thirds of cases In addition to routine histology and immunohistochemistry,
is associated with a chronic Helicobacter pylori infection [1]. fluorescence in situ hybridisation studies for detection of
t(11;18) ( p21;p21) may be useful for identifying patients who
diagnosis are unlikely to respond to antibiotic therapy [III, B] [5, 6].
The most common presenting symptoms of a gastric MALT
lymphoma are non-specific upper gastrointestinal complaints
that often lead to an endoscopy usually revealing non-specific staging and risk assessment
gastritis or peptic ulcer with mass lesions being unusual [2, 3]. The question of which is the best system for the staging of
Diagnosis is based on the histopathological evaluation of the gastric MALT lymphoma is controversial [2, 6]. The ‘Lugano
gastric biopsies [III, A]. The diagnosis should be in accordance staging system’ has been widely used in the past two decades,
with the current World Health Organisation (WHO) but more modern systems have been proposed, such as the
classification and accurate assessment of a potential associated ‘Paris staging system’, which describes more accurately the
large B-cell lymphoma is essential [4]. The diagnosis should, depth of gastric wall involvement, a parameter that may predict
therefore, be confirmed by an expert haematopathologist [5]. It the lymphoma response to H. pylori eradication (Table 1) [7, 8].
should be noted that the term ‘high grade MALT lymphomas’ is The initial staging procedures must include an
no longer accepted in the current WHO classification, hence esophagogastroduodenoscopy with multiple biopsies taken from
cases with solid or sheet-like proliferation of transformed large each region of the stomach, duodenum and gastroesophageal
cells have to be diagnosed as diffuse large B-cell lymphoma [4]. junction and from any site with an abnormal appearance.
Differentiation from other indolent lymphomas is not always Endoscopic ultrasound is recommended to evaluate the regional
straightforward and a minimum immunohistochemistry panel lymph nodes and gastric wall infiltration [III, A] [2, 3, 5, 6, 9].
should include CD20, CD10, CD5 and cyclin D1 [IV, B]. It is Work-up studies should include history and physical
noteworthy that lymphoepithelial lesions, despite being very examination, including lymph node regions, eye and ear, nose
typical of MALT lymphoma, are neither essential for the and throat areas, liver and spleen evaluation; complete blood
counts, basic biochemical studies, which may include evaluation
of renal and liver function, lactate dehydrogenase and
L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland. β2-microglobulin, serum protein immunofixation, human
E-mail: clinicalguidelines@esmo.org immunodeficiency virus, hepatitis C virus and hepatitis B virus
†
serology, computed tomography scan of the chest, abdomen
Approved by the ESMO Guidelines Working Group: December 2006, last update July
2013. This publication supersedes the previously published version—Ann Oncol 2010; and pelvis [IV, B] [2, 3, 5, 6, 9]. A bone marrow aspirate and
21 (Suppl. 5): v175–v176. biopsy is recommended [IV, B] [5]. The value of a positron
© The Author 2013. Published by Oxford University Press on behalf of the Infectious Diseases Society of America.
All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com.

by guest
on 05 February 2018
Table 1. Comparison of the Lugano and Paris staging systems for gastrointestinal tract lymphoma
Lugano staging system [7] Paris staging system [8] Tumour extension
Stage I = confined to the GI tract T1m N0 M0 Mucosa
(single primary or multiple, non-contiguous) T1sm N0 M0 Submucosa
T2 N0 M0 Muscularis propria
T3 N0 M0 Serosa
Stage II = extending into abdomen T1-3 N1 M0 Perigastric lymph nodes

II1 = local nodal involvement T1-3 N2 M0 More distant regional nodes
II2 = distant nodal involvement
Stage IIE = penetration of serosa to involve T4 N0-2 M0 Invasion of adjacent structures with or without
adjacent organs or tissues abdominal lymph nodes
Stage IV = disseminated extranodal involvement T1-4 N3 M0 Extra-abdominal lymph nodes

or concomitant supra-diaphragmatic nodal involvement T1-4 N0-3 M1 And/or additional distant (non-continuous)
gastrointestinal sites
T1-4 N0-3 M2 Or non-gastrointestinal sites
T1-4 N0-3 M0-2 BX Bone marrow not assessed
T1-4 N0-3 M0-2 B0 Bone marrow not involved
T1-4 N0-3 M2 B1 Bone marrow involvement
GI, gastrointestinal; T describes the gastric wall infiltration; N describes the regional lymph node involvement; M describes distant dissemination; B describes
the bone marrow assessment. Reproduced from: Rohatiner A et al. [7] With permission of Oxford University Press. Ruskone-Fourmestrauxet al. [8]. With
permission from BMJ Publishing Group Ltd.
emission tomography scan is controversial and has little clinical worthwhile since occasional lymphoma responses have been
utility [IV, D] [5]. reported (possibly due to a false-negative test or to infection by
other Helicobacter species) [6]. In these H. pylori-negative
patients, an oncological treatment (usually radiotherapy as
treatment plan described below) should, however, be considered if no signs of
lymphoma regression are seen at a repeat endoscopy assessment
Helicobacter pylori eradication therapy must be given to all 2 to 3 months after antibiotics administration [6].
gastric MALT lymphomas, independently of stage [5]. Anti- In patients who do not achieve a lymphoma regression
helicobacter regimens combining proton-pump inhibitor (PPI) following antibiotic therapy, irradiation and systemic
plus clarithromycin-based triple therapy with either amoxicillin oncological therapies should be applied depending on the stage
or metronidazole for 10–14 days are usually highly effective of disease. Radiotherapy might be the preferred option for
[10]. The outcome of the eradication therapy should be checked localised stage. Excellent disease control using radiation therapy
by a urea breath test (or by a monoclonal stool antigen test) at alone has been reported by several institutions supporting the
least 6 weeks after eradication therapy and at least 2 weeks after use of moderate-dose involved-field radiotherapy (24–30 Gy
PPI withdrawal. In case of unsuccessful H. pylori eradication, a radiation to the stomach and perigastric nodes given in 3 to
second-line therapy should be attempted with alternative triple- 4 weeks) [III, B] [11, 12].
or quadruple-therapy regimens of PPI plus antibiotics [1, 6]. Chemotherapy and/or immunotherapy are effective in
Eradication of H. pylori with antibiotics should be the sole patients with MALT lymphoma of all stages.
initial therapy for a localised H. pylori-positive gastric MALT Chemoimmunotherapy should be preferred in case of
lymphoma, where this treatment can induce lymphoma histological transformation, contraindications to radiotherapy,
regression and long-term clinical disease control in most and vice versa. However, there is no definitive evidence to guide
patients [II, A]. The length of time necessary to obtain a the choice between radiotherapy and systemic treatment in
remission can span from very few months to >12 months. It is localised gastric MALT lymphoma, which depends very much
reasonable to wait for at least 12 months before starting another on the local expertise of the attending physicians [1, 5]. Patients
treatment in patients who achieve a clinical and endoscopic with t(11;18) will most probably be unresponsive to alkylating
remission together with eradication of H. pylori, albeit having agents as a sole treatment [1, 6]. Surgery has not been shown to
persistent (residual) lymphoma at the histological level [III, B] achieve superior results in comparison with more conservative
[6]. Several studies of post-antibiotic molecular follow-up have approaches in various trials. It may impair the quality of life and
shown the frequent persistence of monoclonal B-cells after no longer has a role in the initial treatment [13].
histological regression of the lymphoma [1, 6]. Patients with symptomatic systemic disease should be
In H. pylori-negative cases, a regression of the lymphoma considered for systemic treatment [III, A]. As in other
after antibiotic treatment is unlikely and the immediate start of disseminated low-grade lymphomas, rituximab plus
oncological treatments (see below) should be considered, but chemotherapy would then be the most appropriate choice when
the administration of an anti-helicobacter regimen may be treatment is needed.
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt343 | vi

by guest
on 05 February 2018
Figure 1. Treatment algorithms for either localised or advanced gastric MALT lymphoma (stage is defined according to the Lugano system described in Table 1).
vi | Zucca et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
Only a few compounds and regimens have been tested treatment gastric biopsies can be very difficult and there are no
specifically in MALT lymphomas. Oral alkylating agents (either uniform criteria for the definition of histological remission.
cyclophosphamide or chlorambucil) or purine nucleoside Comparison with previous biopsies should be carried out to
analogues (fludarabine, cladribine) and the combination of assess response, and we recommend the GELA (Group d’Etude
rituximab and bendamustine have shown a high rate of disease des Lymphomes de l’Adult) scoring system (Table 2) as a
control in non-randomised studies. The activity of rituximab reproducible method [IV, B] [19].
has also been demonstrated in phase II studies [14] and its Following the documentation of the achieved H. pylori
efficacy in combination with chlorambucil has been proven in a eradication, a strict endoscopic follow-up is recommended, with
randomised study [II, A] [15]. This combination was very well- multiple biopsies taken 2 to 3 months after treatment to rule out
tolerated but no overall survival benefit has been shown [15] tumour progression, and subsequently (twice per year for 2
and there is not yet an accepted standard chemotherapy to be years) to monitor the histological regression of the lymphoma.
recommended. It should, however, be mentioned that treatment Gastric MALT lymphomas have a limited tendency to distant
with purine analogues might be associated with an increased spreading and to histological transformation. Transient,
risk of secondary myelodysplasia. There are no data supporting apparent histological relapses are occasionally observed in
a rituximab maintenance strategy. Aggressive anthracycline- endoscopic follow-up biopsies, but they have to be sustained
containing regimens are not usually necessary and should be and progressive in order to be considered a relapse, as they tend
reserved for the few patients with a very aggressive clinical to be self-limiting, especially in the absence of H. pylori
course or histological transformation [3]. Treatment algorithms reinfection. Hence, in the case of persistent but stable residual
summarising the above discussed management strategies for disease or histological relapse (without distant dissemination
either localised or advanced gastric MALT lymphoma are and/or gross endoscopic tumour), a watch-and-wait policy
shown in Figure 1. appears to be safe [IV, C] [16, 20–22]. Nevertheless, a long-term
careful endoscopic and systemic follow-up (clinical
examination, blood counts and minimal adequate radiological
personalised medicine or ultrasound examinations every 12–18 months) is
Absence of H. pylori, deep invasion of the gastric wall (beyond recommended for all patients. Indeed, the risk of gastric
the sub-mucosa), regional lymph node involvement, the adenocarcinoma among patients diagnosed with gastric MALT
presence of chromosomal translocations that result in lymphomas has been reported to be sixfold higher than in the
deregulation of MALT1 or Bcl-10 and other genetic features general population [23] and also the risk of other non-
(such as overexpression of miR-142-5p and miR-155) can be Hodgkin’s lymphomas may be increased [24].
associated with a reduced probability of lymphoma regression
after antibiotics [1–3, 16–18]. As discussed above, detection of t
(11;18) may also help to distinguish patients who may not
note
respond to alkylating agents alone. Nevertheless, in this disease Levels of evidence and grades of recommendation have been
setting, more research is needed to identify molecular markers applied using the system shown in Table 3. Statements without
which could lead to advances in personalised medicine. grading were considered justified standard clinical practice by
the experts and the ESMO faculty.
response evaluation and follow-up

Histological evaluation of repeat biopsies remains an essential
follow-up procedure to exclude either the possibility of Dr Zucca has reported research support from Roche,
persistent significant disease or, particularly in patients with Mundipharma, Janssen, Novartis, GSK and Celgene.
persistent H. pylori infection, the appearance of early epithelial Dr Ladetto has reported speaker’s bureau from Celgene,
changes, which may be related to gastric carcinoma. Janssen-Cilag, Roche, Bayer, Amgen, Mundipharma; research
Unfortunately, the interpretation of lymphoid infiltrate in post- contracts from Celgene, Pfizer, Mundipharma, Roche; funds
Table 2. GELA grading system proposed to define the histological response of gastric MALT lymphoma after H. pylori eradication [17]
Response (score) Description Histological Characteristics

CR Complete histological remission Normal or empty LP and/or fibrosis with absent or scattered
plasma cells and small lymphoid cells in the LP, no LEL
pMRD Probable minimal residual disease Empty LP and/or fibrosis with aggregates of lymphoid cells or lymphoid
nodules in the LP/MM and/or SM, no LEL
rRD Responding residual disease Focal empty LP and/or fibrosis with dense, diffuse or nodular
lymphoid infiltrate, extending around glands in the LP, focal LEL or absent
NC No change Dense, diffuse or nodular lymphoid infiltrate, LEL usually present
LEL, lymphoepithelial lesions; LP, lamina propria; MM, muscularis mucosa; SM, submucosa.
Copie-Bergman C et al. [18]. Reprinted with permission. ©2012 Blackwell Publishing Ltd.
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt343 | vi

by guest
on 05 February 2018
Grading Systema)
Levels of evidence
I Evidence from at least one large randomised, controlled trial of good methodological quality (low potential for bias) or
meta-analyses of well-conducted randomised trials without heterogeneity
II Small randomised trials or large randomised trials with a suspicion of bias (lower methodological quality) or meta-analyses of
such trials or of trials with demonstrated heterogeneity
a
received from Amgen, Roche, Italfarmaco. The other authors 13. Koch P, Probst A, Berdel WE et al. Treatment results in localized primary gastric
have declared no potential conflicts of interest. lymphoma: data of patients registered within the German multicenter study
(GIT `NHL 02/96). J Clin Oncol 2005; 23: 7050–7059.
14. Martinelli G, Laszlo D, Ferreri AJ et al. Clinical activity of rituximab in gastric
references marginal zone non-Hodgkin’s lymphoma resistant to or not eligible for anti-
Helicobacter pylori therapy. J Clin Oncol 2005; 23: 1979–1983.
1. Bertoni F, Coiffier B, Salles G et al. MALT Lymphomas: pathogenesis can drive 15. Zucca E, Conconi A, Laszlo D et al. Addition of rituximab to chlorambucil produces
treatment. Oncology 2011; 25: 1134–1142, 1147. superior event-free survival in the treatment of patients with extranodal marginal-
2. Bertoni F, Zucca E. State-of-the-art therapeutics: marginal-zone lymphoma. J Clin zone B-cell lymphoma: 5-year analysis of the IELSG-19 randomized study. J Clin
Oncol 2005; 23: 6415–6420. Oncol 2013; 31: 565–572.
3. Thieblemont C. Clinical presentation and management of marginal zone 16. Nakamura S, Sugiyama T, Matsumoto T et al. Long-term clinical outcome of
lymphomas. Hematol Am Soc Hematol Educ Program 2005; 307–313. gastric MALT lymphoma after eradication of Helicobacter pylori: a multicentre
4. Isaacson PG, Chott A, Nakamura S et al. Extranodal marginal zone B-cell cohort follow-up study of 420 patients in Japan. Gut 2012; 61: 507–513.
lymphoma of mucosa-associated lymphoid tissue (MALT lymphoma). In Swerdlow 17. Zullo A, Hassan C, Cristofari F et al. Effects of Helicobacter pylori eradication on
SH, Campo E, Harris NL et al. (eds), WHO Classification of Tumours of early stage gastric mucosa-associated lymphoid tissue lymphoma. Clin
Haematopoietic and Lymphoid Tissues, 4th edition. Lyon, France: IARC Press Gastroenterol Hepatol 2010; 8: 105–110.
2008; 214–217. 18. Saito Y, Suzuki H, Tsugawa H et al. Overexpression of miR-142-5p and miR-155
5. Dreyling M, Thieblemont C, Gallamini A et al. ESMO Consensus conferences: in gastric mucosa-associated lymphoid tissue (MALT) lymphoma resistant to
guidelines on malignant lymphoma. part 2: marginal zone lymphoma, mantle cell Helicobacter pylori eradication. PLoS One 2012; 7: e47396.
lymphoma, peripheral T-cell lymphoma. Ann Oncol 2013; 24: 857–877. 19. Copie-Bergman C, Wotherspoon AC, Capella C et al. Gela histological scoring
6. Ruskone-Fourmestraux A, Fischbach W, Aleman BM et al. EGILS Consensus system for post-treatment biopsies of patients with gastric MALT lymphoma is
report. Gastric extranodal marginal zone B-cell lymphoma of MALT. Gut 2011; 60: feasible and reliable in routine practice. Br J Haematol 2013; 160: 47–52.
747–758. 20. Stathis A, Chini C, Bertoni F et al. Long-term outcome following Helicobacter
7. Rohatiner A, d’Amore F, Coiffier B et al. Report on a workshop convened to discuss pylori eradication in a retrospective study of 105 patients with localized gastric
the pathological and staging classifications of gastrointestinal tract lymphoma. Ann marginal zone B-cell lymphoma of MALT type. Ann Oncol 2009; 20:
Oncol 1994; 5: 397–400. 1086–1093.
8. Ruskone-Fourmestraux A, Dragosics B, Morgner A et al. Paris staging system for 21. Fischbach W, Goebeler ME, Ruskone-Fourmestraux A et al. Most patients with
primary gastrointestinal lymphomas. Gut 2003; 52: 912–913. minimal histological residuals of gastric MALT lymphoma after successful
9. Raderer M, Wöhrer S, Streubel B et al. Assessment of disease dissemination in eradication of Helicobacter pylori can be managed safely by a watch and wait
gastric compared with extragastric mucosa-associated lymphoid tissue lymphoma strategy: experience from a large international series. Gut 2007; 56: 1685–1687.
using extensive staging: a single-center experience. J Clin Oncol 2006; 24: 22. Wündisch T, Thiede C, Morgner A et al. Long-term follow-up of gastric MALT
3136–3141. lymphoma after Helicobacter pylori eradication. J Clin Oncol 2005; 23:
10. Fuccio L, Laterza L, Zagari RM et al. Treatment of Helicobacter pylori infection. 8018–8024.
BMJ 2008; 337: a1454. 23. Capelle LG, de Vries AC, Looman CW et al. Gastric MALT lymphoma: epidemiology
11. Tsang RW, Gospodarowicz MK. Radiation therapy for localized low-grade non- and high adenocarcinoma risk in a nation-wide study. Eur J Cancer 2008; 44:
Hodgkin’s lymphomas. Hematol Oncol 2005; 23: 10–17. 2470–2476.
12. Wirth A, Gospodarowicz M, Aleman BM et al. Long-term outcome for gastric 24. Wündisch T, Dieckhoff P, Greene B et al. Second cancers and residual disease
marginal zone lymphoma treated with radiotherapy: a retrospective, multi-centre, in patients treated for gastric mucosa-associated lymphoid tissue lymphoma by
International Extranodal Lymphoma Study Group Study. Ann Oncol 2013; 24: Helicobacter pylori eradication and followed for 10 years. Gastroenterology 2012;
1344–1351. 143: 936–942.
vi | Zucca et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
Hairy cell leukaemia: ESMO Clinical Practice Guidelines

T. Robak1, E. Matutes2, D. Catovsky3, P. L. Zinzani4 & C. Buske5, on behalf of the ESMO Guidelines
Committee*
1
Department of Hematology, Medical University of Lodz, Lodz, Poland; 2Haematopathology Unit, Hospital Clinic, Barcelona University, Barcelona, Spain;
3
Haemato-Oncology Research Unit, The Institute of Cancer Research, Sutton, Surrey, UK; 4Seràgnoli Institute of Hematology, University of Bologna, Bologna, Italy;
5
Comprehensive Cancer Center Ulm, Institute of Experimental Cancer Research, University Hospital, Ulm, Germany
incidence and epidemiology combination will allow for the differentiation of HCL from
other B-cell leukaemias and lymphomas with circulating villous
Classical hairy cell leukaemia (HCL) is a B-cell chronic lympho- cells [5, 6]. In addition, strong expression of CD200 is character-
proliferative disorder characterised by splenomegaly, pancytopae- istic of HCL and may be useful for the diagnosis in difficult
nia and bone marrow involvement with fibrosis. HCL represents cases [7].
2% of adult leukaemia. Approximately 1600 new cases per year HCL-V typically presents with high lymphocyte counts, with
are diagnosed in Europe [1], with a median age of 52 years at the the cells being nucleolated and lacking monocytopaenia. Flow
time of diagnosis. The disease occurs more often in men than in cytometry will indicate the cells to be CD11c+ and often CD103
women, with a ratio of approximately 4:1 [2]. In the USA, a
clinical practice
+ but very rarely CD25+ and CD123+. There is an overlap with

higher frequency of HCL is observed among white Americans
guidelines
splenic diffuse red pulp lymphoma, and distinguishing between

than among African-Americans or Asians, as well as in patients these two diseases may be difficult. A bone marrow examination
following exposure to the herbicide ‘Agent Orange’, used during is required for the diagnosis, particularly after treatment, to
the Vietnam War [3]. HCL variant (HCL-V) is classified among assess response [I, C].
the unclassifiable splenic B-cell leukaemia/lymphoma that is no As the bone marrow can rarely be aspirated (‘dry-tap’) in clas-
longer biologically related to classical HCL. It is included in the sical HCL, diagnosis is typically performed by a bone marrow
World Health Organization (WHO) classification as a provisional trephine biopsy. The degree and pattern of infiltration varies
entity [4]. HCL-V is an uncommon disorder, accounting for ap- from mild interstitial to diffuse, and the lymphoid cells are sur-
proximately 0.4% of chronic lymphoid malignancies and 10% of rounded by a clear halo due to the abundant cytoplasm, giving
all HCL cases, without sexual predominance. The median age of the characteristic ‘fried egg’ pattern. In HCL-V, the infiltration is
the patients is 71 years. either intrasinusoidal or interstitial.
Immunohistochemistry with the monoclonal antibodies CD20,
diagnosis and pathology/molecular CD72 (DBA44), CD11c, CD25, CD103, annexin A1 and tartrate-
biology resistant acid phosphatase stain will highlight the lymphoid infil-
trates and support the diagnosis of classical HCL [6]. However,
The examination of peripheral blood films and immunopheno- annexin A1 is not suitable for detecting residual disease after
typing allows for a diagnosis to be established in most cases treatment as it stains myeloid cells also. Cyclin D1 may be weakly
(Table 1) [I, C] [5, 6]. The neoplastic cells are twice the size of a positive but differential diagnosis with mantle cell lymphoma
lymphocyte and have a round or kidney-shaped nucleus with rarely arises. Recently, monoclonal antibodies that detect the
loose chromatin and abundant pale cytoplasm with projections. mutated BRAF protein have been developed and shown to be
Monocytopaenia and macrocytosis are very common; other useful for the diagnosis and detection of minimal residual
cytopaenias may be present. A diagnosis of HCL based on cy- disease (MRD) [8]. Although this requires validation, prelimin-
tology can be effectively confirmed by flow cytometry studies ary data suggest that this marker has high specificity and sensi-
using anti-B-cell monoclonal antibodies such as CD19, CD20 or tivity for HCL [II, C].
CD22, together with a panel of antibodies such as CD11c, The main distinguishing features between HCL and HCL-V
CD25, CD103 and CD123, which are more specific to HCL; this are outlined in Table 2. HCL does not have a distinct chromo-
somal abnormality. The majority of the cases have mutations
*Correspondence to: ESMO Guidelines Committee, ESMO Head Office, via L. Taddei 4, of the immunoglobulin heavy chain (IGHV) gene, suggesting
CH-6962 Viganello-Lugano, Switzerland. that the disease arises on a memory B cell. Unlike other spleno-
megalic disorders such as splenic marginal zone lymphoma or
†
Approved by the ESMO Guidelines Committee, July 2015. HCL-V, there is no evidence of specific IGHV, IGHD, IGHJ

by guest
on 05 February 2018
Table 1. Diagnostic work-up of classical HCL as well as the presence of lymphadenopathy, which predicts a
poor response to purine analogues [15]. Patients who achieve a
Peripheral blood film morphology [I, C]
complete response (CR) have a significantly longer disease-free
Flow cytometry in peripheral blood and bone marrow aspirate [I, C]
survival than those who achieve a partial response (PR) [II, B]
Bone marrow trephine biopsy with immunohistochemistry [I, C]
[16]. Biological factors that have been associated with a poor
BRAF mutation of exon 15 in difficult cases [II, C]
outcome are the presence of TP53 mutations and the lack of
HCL, hairy cell leukaemia.
somatic mutations in the IGVH genes, which occur in a minor
proportion of cases [IV, C] [10] and the VH4-34 family usage, a
feature more frequent in HCL-V [11]. The presence of TP53
mutations (but not VH4-34) usage appears to be an adverse
Table 2. Diagnostic criteria for HCL and HCL-V
prognostic factor in HCL-V [IV, C] [13].
Characteristics Classical HCL HCL-V
Bone marrow aspiration Difficult (often dry tap) Easy

Lymphocytosis − +
treatment plan
Monocytopaenia + − Treatment is not indicated in asymptomatic patients [V, B].
Prominent nucleoli − + However, untreated patients should be closely monitored with
Cytoplasmic projections + + a complete history, physical examination, and complete blood
CD25 + − cell count with a differential test every 3–6 months. In contrast
FMC7, CD20, CD22, CD11c + + to chronic lymphocytic leukaemia, asymptomatic patients, who
CD103, CD123 + Variable
may be diagnosed by chance, are rare and in practice most
Annexin + −
patients need treatment shortly after diagnosis, either because of
BRAF V600E mutation + −
symptoms or to correct cytopaenias including monocytopaenia.
Splenomegaly + +
Treatment should be initiated in patients with symptomatic
Response to purine analogues Good Poor
disease manifested by bulky or progressive, symptomatic spleno-
HCL, hairy cell leukaemia; HCL-V, HCL variant. megaly cytopaenias (haemoglobin <10 g/dl and/or platelets
<100 × 109/l and/or neutrophils <1 × 109/l), recurrent or severe
infections and/or systemic symptoms [II, A] [17, 18].
repertoires or stereotypes in HCL [9–11]. In 2011, Tiacci et al.
[12] reported the presence of the V600E mutation of the BRAF
gene in exon 15 in all 47 investigated HCL cases. The mutation first line
rarely occurs in exon 11. The BRAF mutation leads to the activa- purine analogues
tion of the RAF/MEK-ERK pathway, resulting in enhanced cell
Purine analogues, cladribine (2-CldA) or pentostatin (DCF), are
proliferation and survival. These results have been validated by
recommended as initial treatment of symptomatic HCL patients
other groups and new simpler and more sensitive methods to
who are young and fit (Figure 1) [II, A]. 2-CldA induces durable
detect the mutation, such as quantitative real-time polymerase
and unmaintained response in 87%–100% patients, including CR
chain reaction (q-PCR) have been developed. This finding is rele-
in 85%–91%, after a single course of therapy [19]. 2-CldA is
vant to the diagnosis and pathogenesis of HCL with potential
administered either as a continuous intravenous (i.v.) infusion at
therapeutic implications. In the near future, testing for this muta-
a dose of 0.09 mg/kg over a 5–7 day period, or as a 2-h i.v. infu-
tion may be incorporated on a routine basis into the diagnostic
sion at a dose of 0.12–0.14 mg/kg for 5–7 days [20, 21]. 2-CldA is
work-up of HCL [II, C]. Although HCL-V patients lack the BRAF
also an effective drug when administered at a dose of 0.12–0.15
mutation, TP53 mutations are present in one-third of cases [13].
mg/kg in 2-h infusion once a week over 6 courses [I, B] [22, 23].
Both CR and overall response (OR) rates are similar in weekly
and daily administration. Neither less frequent infections nor
staging and risk stratification haematological toxicity have been confirmed in the weekly sched-
There is no worldwide accepted staging system for HCL. In add- ule by randomised trials [I, B] [23, 24]. Similar results were
ition to the diagnostic tests on the blood and bone marrow tre- achieved when the drug was given as a subcutaneous injection
phine biopsy, a staging work-up should include full blood cell [II, B] [24, 25]. A subcutaneous 2-CldA is given at a dose of 0.1
counts with differential and reticulocytes, renal and liver bio- mg/kg/day for 5–7 days or 0.14 mg/kg/day for 5 days as a single
chemistry, serum immunoglobulins, β2 microglobulin, direct course [24, 25]. Grade 3–4 toxicity, largely represented by neutro-
antiglobulin test (DAT), Coombs test and hepatitis virus B and paenic fevers and infections, were less frequent when lower total
C and human immunodeficiency virus screening. Computed doses were used (0.5 mg/kg) than higher doses (0.7 mg/kg) with
tomography (CT) imaging is desirable at the time of diagnosis similar OR [25]. Subcutaneous administration does not usually
(as around 10% of HCL patients have abdominal lymphadenop- require hospitalisation, and seems to be an easier way of deliver-
athy) and should be performed at relapse [14]. ing the drug than i.v. administration. Although no randomised
There is no international prognostic system for risk stratifica- trial comparing i.v. versus subcutaneous 2-CldA administration
tion of HCL. Clinical variables that have been considered to has been performed, the efficacy of both methods is similar [II,
have an adverse prognosis include the degree of cytopaenias B]. A CR following 2-CldA administration is durable even
(Hb < 10 g/dl, platelets < 100 × 109/l) and neutrophils (<1 × 109/l), without maintenance therapy [16]. In patients demonstrating a
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv200 | v

by guest
on 05 February 2018
Figure 1. Therapeutic algorithm for newly diagnosed classical hairy cell leukaemia (HCL). CR, complete response; PR, partial response.
PR after the first course of 2-CldA, a second course should be the treatment of HCL is limited. However, IFN-α may still have
repeated to achieve a CR at least 6 months after the end of the a place in the treatment of HCL in pregnancy [V, B]. It can also
first course, with or without rituximab [IV, B] [26]. be used in patients presenting with very severe neutropaenia
Similarly to 2-CldA, DCF induces a high rate of long-lasting (neutrophil count <0.2 × 109/l) to increase the neutrophil count
CR. In patients with a normal creatinine clearance (>60 ml/min), prior to nucleoside analogue therapy [V, C] [29].
DCF is usually given at a dose of 4 mg/m2 i.v. every second week
until CR, plus one or two consolidating injections [27]. After 8–9
courses, the full blood count usually normalises, and the bone response evaluation
marrow biopsy should be performed to confirm a CR [III, B]. If a Responses are defined according to the ‘Consensus Resolution’
CR is documented, one or two further DCF injections are indi- criteria (Table 3) [30]. Response evaluation includes careful
cated [16]. DCF and 2-CldA appear to induce similar high re- physical examination and a blood cell count. A marrow biopsy
sponse rates, duration of response, recurrence rates and adverse is recommended to establish a CR. A chest X-ray and an abdom-
events [III, B] [16]. However, no randomised, direct comparison inal ultrasound or CT for response evaluation should be
between the two drugs has been performed. The advantage of performed. A CR requires the morphological absence of hairy
DCF over interferon-α (IFN-α) in HCL patients has been con- cells in peripheral blood and bone marrow aspiration or biopsy
firmed in a multicentre, randomised trial [I, A] [28]. 2-CldA ad- specimens, and normalisation of any organomegaly and cytopae-
ministration is more convenient than DCF and is used more nia. Immunophenotypic analysis of peripheral blood or bone
frequently. marrow is not required but is useful to detect MRD. A PR is
defined as normalisation of peripheral counts, associated with at
least a 50% reduction in organomegaly and bone marrow hairy
interferon-α cells, and <5% circulating hairy cells. All other outcomes are con-
As purine analogues produce higher and more durable remis- sidered as nonresponse. The eradication of MRD is generally not
sions, and are more convenient to patients, the use of IFN-α in recommended in routine clinical practice. Assessment of response
v | Robak et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Table 3. Response criteria for HCL Other promising drugs active in purine analogue refractory
HCL patients include moxetumomab pasudotox, an anti-CD22
Definition of response categories
recombinant immunotoxin, and vemurafenib, a BRAF V600E
Complete No hairy cells on peripheral blood and bone marrow inhibitor [44–46]. A phase I trial of moxetumomab pasudotox
response aspiration or biopsy specimens, normalisation of in relapsed/refractory HCL induced an 86% OR rate, and a 46%
organomegaly and peripheral blood counts CR [44]. However, moxetumomab pasudotox is not licensed in
Partial response Normalisation of peripheral blood counts, at least Europe yet.
50% reduction in organomegaly and bone marrow Vemurafenib also showed remarkable activity in multiply
hairy cells, and <5% circulating hairy cells relapsed and refractory HCL patients with rapidly decreased
splenomegaly, increased platelet counts and normalisation
Relapse Any deterioration in blood counts related to the
of haemoglobin and granulocyte counts [45, 46]. The optimal
detection of hairy cells in peripheral blood and/or
bone marrow and/or increasing splenomegaly
dosing and duration of treatment using vemurafenib is un-
known. A multicentre, phase II study of vemurafenib in the
HCL, hairy cell leukaemia. Data from [30]. treatment of refractory HCL is ongoing (NCT01711632). The
therapeutic potential of the Bruton’s tyrosine kinase (BTK) in-
hibitor ibrutinib also needs to be taken into consideration in a
relapse setting, given the proven in vitro activity of this drug in
should be performed 4–6 months after treatment with 2-CldA hairy cells [47]. A phase II study with ibrutinib in patients with
and after 8–9 courses of DCF [31]. Relapse is defined as any de- relapsed HCL is ongoing (NCT01841723).
terioration in blood counts related to the detection of hairy cells
in peripheral blood and/or bone marrow.
splenectomy
Splenectomy may be indicated in patients with resistant massive
treatment of relapsed patients and symptomatic splenomegaly (>10 cm below the costal margin)
with accompanied low-level bone marrow infiltration [IV, B]
patients refractory to purine analogues [48]. OR rates of 60%–100% were documented in eight major
Relapsed patients can be successfully retreated with 2-CldA or reports [29]. Splenectomised patients respond better and faster
DCF if relapse occurs after 12–18 months [IV, B] [32] (Figure 2). to the subsequent chemotherapy [V, B]. Splenectomy can also
The alternative nucleoside analogue can be used in early relapse be considered when progressive HCL develops during preg-
within 2 years after the first-line treatment [31]. The ability to nancy and in patients refractory to nucleoside analogues and
attain CR decreases with each course of therapy, but CR dura- IFN-α [V, B] [29]. Systemic therapy should not be performed
tion appears to be similar after first-, second- or third-line therapy earlier than 6 months after splenectomy to reach full benefits
[16, 33]. of this treatment [31]. Vaccination against Haemophilus influ-
Rituximab at a dose of 375 mg/m2 for 4–8 doses given weekly enza and pneumococcus before splenectomy is recommended
as i.v. infusions can be used in early relapsed patients [III, B] [IV, B] [49].
[34–36]. However, rituximab alone is inferior to purine analo-
gues and is not the treatment of choice as a single agent in
relapsed patients. Outcomes for patients with recurrent HCL
allogeneic stem cell transplantation
appear to be better when a combination of rituximab and 2- Allogeneic stem cell transplantation has a potential role in
CldA or DCF is used rather than the purine analogue alone [III, younger, heavily pretreated HCL patients who have had multiple
B] [37, 38]. Concurrent therapy of rituximab and a purine ana- relapses and are refractory to purine analogues and rituximab
logue induces higher response rates, and higher rates of toxic [V, C] [50, 51].
events than in the sequential administration [III, B] [39].
IFN-α is also a possible option for selected patients relapsing
after purine analogue therapy [IV, B] [40, 41]. Patients refrac-
treatment during pregnancy
tory to purine analogue therapy should be enrolled on clinical Only a few cases of HCL during pregnancy have been reported
trials that use new agents, whenever possible. Fludarabine at a [52, 53]. Treatment is indicated only when truly warranted. The
dose of 40 mg/m2 oral ( p.o.) for five consecutive days in com- appropriate therapy depends on the stage of pregnancy, rate of
bination with an i.v. injection of 375 mg/m2 rituximab on day 1, disease progression and response to previous therapies. If treat-
every 28 days for four cycles, can be a therapeutic option in ment is indicated, the lower risks associated with IFN-α for both
relapsed or refractory patients previously treated with 2-CldA the mother and the fetus make it the treatment of choice [V, B].
[IV, B]. After a median follow-up of 35 months with 14 progres- IFN-α shows good tolerance, uncomplicated pregnancy and
sion-free patients, a 5-year progression-free survival (PFS) of delivery, and normal child development [54]. Splenectomy is
89%, an overall survival (OS) of 83% and a recurrence rate of another option to consider, should INF-α fail, especially in early
7% were observed in 15 treated patients [42]. Bendamustine at gestation when the risks are lower [V, B] [52, 55]. Administration
70–90 mg/m2 combined with rituximab is another therapeutic of 2-CldA and rituximab are not indicated in pregnancy because
option in multiply relapsed/refractory HCL, and could be con- of the risk of teratogenic effects. However, successful pregnancy
sidered in HCL patients after the failure of standard therapies after the administration of 2-CldA and rituximab has also been
[IV, B] [43]. reported [56].
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv200 | v

by guest
on 05 February 2018
Figure 2. Therapeutic algorithm for relapsed and refractory classical hairy cell leukaemia (HCL).
supportive treatment treatment of HCL-V

Patients treated with nucleoside analogues and lymphopaenia The results of the treatment of HCL-V with purine analogues
should receive prophylactic co-trimoxazole 960 mg three times are poor (Figure 3). The analysis of 19 cases treated with 2-CldA
per week and aciclovir (200 mg three times daily) until the showed a response rate of 55%, including only 2 with CR [59].
lymphocyte count increases to >1 × 109/l to prevent pneumocystis In addition, the majority of HCL-V patients required more than
infections and herpes reactivation [IV, B]. Co-trimoxazole and 1 cycle to maintain a response. DCF induced a PR in only 8 out
aciclovir prophylaxis should be started one week after purine ana- of 15 (54%) patients and no CR was achieved [60]. Rituximab
logue administration to reduce allergic skin reaction [18]. The alone, or splenectomy followed by rituximab therapy, can also
routine adjunctive use of filgrastim with 2-CldA in the treatment induce CR in HCL-V [61, 62]. However, administration of
of patients with HCL and severe neutropaenia is not recom- 2-CldA immediately followed by rituximab, typically involving
mended [IV, C] [57]. However, granulocyte colony-stimulating 2-CldA being given at a dose of 0.15 mg/kg on days 1–5, with
factor use should be considered in individual cases with severe 8 weekly doses of 375 mg/m2 rituximab beginning day 1,
neutropaenia and life-threatening infection. Patients treated with appears to be more effective than 2-CldA alone or rituximab
2-CldA or DCF should receive irradiated blood products to alone and should be considered as the initial treatment of HCL-
prevent transfusion-associated graft-versus-host disease [58]. V patients [IV, B] [38, 63]. In a recent study based on 8 patients,
v | Robak et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Figure 3. Therapeutic algorithm for hairy cell leukaemia variant (HCL-V).
all achieved a CR, including 7 (88%) with MRD negativity at in 17 (77.3%) of 22 HCL cases by PCR, and was demonstrated
12–38 (median 24) months of follow-up. Alternatively, indi- in 20 (90.9%) cases using VE1 immunohistochemistry staining
vidual case reports suggest that alemtuzumab is an active agent [68]. Immunohistochemical detection of BRAF V600E mutant
in treating HCL-V, even in patients who have relapsed after protein is highly sensitive and specific for the diagnosis of HCL.
rituximab [64]. Splenectomy induces clinical responses in some
patients with HCL-V, and is recommended because it corrects
cytopaenias, removes the bulk of the tumour and may improve
response to purine nucleoside analogues [V, B] [65]. Splenic A follow-up of asymptomatic patients should include a com-
irradiation could be performed in elderly patients with a high plete history, physical examination, a blood cell count and
surgical risk of splenectomy [V, B]. Clinical case reports support routine chemistry every 3–12 months [II, B]. Regular CT scans
the use of moxetumomab pasudotox in patients with HCL-V are not necessary outside clinical trials. The frequency of second
[V, B] [44]. Some case reports indicate that autologous or malignancies is increased in HCL patients whether treated
allogeneic hematopoietic cell transplantation can be effective in or untreated. Solid and haematological malignancies develop
refractory patients [V, B] [66]. However, the data to support in 10% of patients after HCL diagnosis, particularly chronic
these approaches are limited. lymphoproliferative diseases (myeloma, Hodgkin’s and non-
Hodgkin’s lymphoma), melanoma and thyroid cancer [70, 71].
In HCL and HCL-V, more research is needed to identify bio-
methodology
markers which could be used for medical treatment decisions. These clinical practice guidelines were developed in accordance
Recently, BRAF V600E mutation has been identified as a charac- with the ESMO standard operating procedures for clinical prac-
teristic biomarker of HCL [III, B] [67–69]. It is present in nearly tice guidelines development. The relevant literature has been
all cases of HCL but virtually absent in diseases that mimic selected by the expert authors. Levels of evidence and grades of
HCL. In a recent study, a BRAF V600E mutation was detected recommendation have been applied using the system shown in
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv200 | v

by guest
on 05 February 2018
Service Grading System)a
Levels of evidence
I Evidence from at least one large randomised, controlled trial of good methodological quality (low potential for bias) or meta-analyses
of well-conducted randomised trials without heterogeneity
II Small randomised trials or large randomised trials with a suspicion of bias (lower methodological quality) or meta-analyses of such
C Insufficient evidence for efficacy or benefit does not outweigh the risk or the disadvantages (adverse events, costs, ....), optional
a
Table 4. Statements without grading were considered justified lymphoma provided by the analysis of their IGH rearrangements and somatic
standard clinical practice by the experts and the ESMO faculty. hypermutation patterns. Br J Haematol 2010; 148: 666–669.
This manuscript has been subjected to an anonymous peer 10. Forconi F, Sozzi E, Cencini E et al. Hairy cell leukemias with unmutated IGHV genes
define the minor subset refractory to single-agent cladribine and with more
review process.
aggressive behaviour. Blood 2009; 114: 4696–4702.
11. Arons E, Suntum T, Stetler-Stevenson M, Kreitman RJ. VH4–34+ hairy cell
conflict of interest leukemia, a new variant with poor prognosis despite standard therapy. Blood
2009; 114: 4687–4695.
TR has reported honoraria from Roche and research grants 12. Tiacci E, Trifonov V, Schiavoni G et al. BRAF mutations in hairy cell leukemia. N
from Roche, Janssen, GlaxoSmithKline and Medimmune. PLZ Engl J Med 2011; 364: 2305–2315.
has reported honoraria from Takeda Millennium, Roche, 13. Hockley SL, Else M, Morilla A et al. The prognostic impact of clinical and molecular
Celgene, Pfizer, Gilead and Sandoz. CB has reported honoraria features in hairy cell leukaemia variant and splenic marginal zone lymphoma . Br J
from Roche, Pfizer, Celgene, Pharmacyclics and Janssen and re- Haematol 2012; 158: 347–354.
14. Mercieca J, Puga M, Matutes E et al. Incidence and significance of abdominal
search grants from Roche and Janssen. The other authors have
lymphadenopathy in hairy cell leukaemia. Leuk Lymphoma 1994; 14(Suppl. 1):
reported no potential conflicts of interest. 79–83.
15. Mercieca J, Matutes E, Emmett E et al. 2-chlordeoxyadenosine in the treatment of
hairy cell leukaemia: differences in response in patients with and without
references abdominal lymphadenopathy. Br J Haematol 1996; 93: 409–411.
1. Ferlay J, Steliarova-Foucher E, Lortet-Tieulent J et al. Cancer incidence and 16. Else M, Dearden CE, Matutes E et al. Long-term follow-up of 233 patients with
mortality patterns in Europe: estimates for 40 countries in 2012. Eur J Cancer hairy cell leukaemia, treated initially with pentostatin or cladribine, at a median of
2013; 49: 1374–1403. 16 years from diagnosis. Br J Haematol 2009; 145: 733–740.
2. Morton LM, Wang SS, Devesa SS et al. Lymphoma incidence patterns by WHO 17. Grever MR. How I treat hairy cell leukemia. Blood 2010; 115: 21–28.
subtype in the United States, 1992–2001. Blood 2006; 107: 265–276. 18. Cornet E, Delmer A, Feugier P et al. Recommendations of the SFH (French Society
3. Grever MR, Blachly JS, Andritsos LA. Hairy cell leukemia: update on molecular of Haematology) for the diagnosis, treatment and follow-up of hairy cell leukaemia.
profiling and therapeutic advances. Blood Rev 2014; 28: 197–203. Ann Hematol 2014; 93: 1977–1983.
4. Swerdlow SH, Campo E, Harris NL et al. World Health Organization (WHO) 19. Saven A, Burian C, Koziol JA, Piro LD. Long-term follow-up of patients with hairy
Classification of Tumours: Pathology and Genetics of Haematopoietic and cell leukemia after cladribine treatment. Blood 1998; 92: 1918–1926.
Lymphatic Tissues. 4th edition. Lyon, France: IARC Press, 2008. 20. Cheson BD, Sorensen JM, Vena DA et al. Treatment of hairy cell leukemia with
5. Del Giudice I, Matutes E, Morilla R et al. The diagnostic value of CD123 in B-cell 2-chlorodeoxyadenosine via the Group C protocol mechanism of the National
disorders with hairy or villous lymphocytes. Haematologica 2004; 89: 303–308. Cancer Institute: a report of 979 patients. J Clin Oncol 1998; 16: 3007–3015.
6. Matutes E. Immunophenotyping and differential diagnosis of hairy cell leukemia. 21. Robak T, Blasińska-Morawiec M, Krykowski E et al. 2-chlorodeoxyadenosine (2-
Hematol Oncol Clin North Am 2006; 20: 1051–1063. CdA) in 2-hour versus 24-hour intravenous infusion in the treatment of patients
7. Pillai V, Pozdnyakova O, Charest K et al. CD200 flow cytometric assessment and with hairy cell leukemia. Leuk Lymphoma 1996; 22: 107–111.
semiquantitative immunohistochemical staining distinguishes hairy cell leukemia 22. Robak T, Jamroziak K, Gora-Tybor J et al. Cladribine in a weekly versus daily
from hairy cell leukemia-variant and other B-cell lymphoproliferative disorders. Am schedule for untreated active hairy cell leukemia: final report from the Polish Adult
J Clin Pathol 2013; 140: 536–543. Leukemia Group (PALG) of a prospective, randomized, multicenter trial. Blood
8. Andrulis M, Penzel R, Weichert W et al. Application of a BRAF V600E mutation- 2007; 109: 3672–3675.
specific antibody for the diagnosis of hairy cell leukemia. Am J Surg Pathol 2012; 23. Zenhäusern R, Schmitz SF, Solenthaler M et al. Randomized trial of daily versus
36: 1796–1800. weekly administration of 2-chlorodeoxyadenosine in patients with hairy cell
9. Hockley SL, Giannouli S, Morilla A et al. Insight into the molecular pathogenesis of leukemia: a multicenter phase III trial (SAKK 32/98). Leuk Lymphoma 2009; 50:
hairy cell leukaemia, hairy cell leukaemia variant and splenic marginal zone 1501–1511.
v | Robak et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
24. von Rohr A, Schmitz SF, Tichelli A et al. Treatment of hairy cell leukemia with 49. Davies JM, Barnes R, Milligan D; British Committee for Standards in Haematology.
cladribine (2-chlorodeoxyadenosine) by subcutaneous bolus injection: a phase II Working Party of the Haematology/Oncology Task Force. Update of guidelines for
study. Ann Oncol 2002; 13: 1641–1649. the prevention and treatment of infection in patients with an absent or
25. Lauria F, Cencini E, Forconi F. Alternative methods of cladribine administration. dysfunctional spleen. Clin Med 2002; 2: 440–443.
Leuk Lymphoma 2011; 52(Suppl. 2): 34–37. 50. Zinzani PL, Bonifazi F, Pellegrini C et al. Hairy cell leukemia: allogeneic
26. Dearden CE, Else M, Catovsky D. Long-term results for pentostatin and cladribine transplantation could be an optimal option in selected patients. Clin Lymphoma
treatment of hairy cell leukemia. Leuk Lymphoma 2011; 52(Suppl. 2): 21–24. Myeloma Leuk 2012; 12: 287–289.
27. Flinn IW, Kopecky KJ, Foucar MK et al. Long-term follow-up of remission duration, 51. Kiyasu J, Shiratsuchi M, Ohtsuka R et al. Achievement of complete remission of
mortality and second malignancies in hairy cell leukemia patients treated with refractory hairy cell leukemia by rituximab progressing after allogeneic
pentostatin. Blood 2000; 96: 2981–2986. hematopoietic stem cell transplantation. Int J Hematol 2009; 89: 403–405.
28. Grever M, Kopecky K, Foucar MK et al. Randomized comparison of pentostatin 52. Stiles GM, Stanco LM, Saven A, Hoffmann KD. Splenectomy for hairy cell leukemia
versus interferon alfa-2a in previously untreated patients with hairy cell leukemia: in pregnancy. J Perinatol 1998; 18: 200–201.
an intergroup study. J Clin Oncol 1995; 13: 974–982. 53. Orlowski RZ. Successful pregnancy after cladribine therapy for hairy cell leukemia.
29. Habermann TM, Rai K. Historical treatments of in hairy cell leukemia, splenectomy Leuk Lymphoma 2004; 45: 187–188.
and interferon: past and current uses. Leuk Lymphoma 2011; 52(Suppl. 2): 54. Baer MR, Ozer H, Foon KA. Interferon-alpha therapy during pregnancy in chronic
18–20. myelogenous leukaemia and hairy cell leukaemia. Br J Haematol 1992; 81:
30. Anonymous. Consensus resolution: proposed criteria for evaluation of response to 167–169.
treatment in hairy cell leukemia. Leukemia 1987; 1: 405. 55. Adeniji BA, Fallas M, Incerpi M et al. Laparoscopic splenectomy for hairy cell
31. Jones G, Parry-Jones N, Wilkins B et al. Revised guidelines for the diagnosis and leukemia in pregnancy. Case Rep Med 2010; 2010: 136823.
management of hairy cell leukaemia and hairy cell leukaemia variant. Br J 56. Daver N, Nazha A, Kantarjian HM et al. Treatment of hairy cell leukemia during
Haematol 2012; 156: 186–195. pregnancy: are purine analogues and rituximab viable therapeutic options. Clin
32. Naik RR, Saven A. My treatment approach to hairy cell leukemia. Mayo Clin Proc Lymphoma Myeloma Leuk 2013; 13: 86–89.
2012; 87: 67–76. 57. Saven A, Burian C, Adusumalli J, Koziol JA. Filgrastim for cladribine-induced
33. Zinzani PL, Pellegrini C, Stefoni V et al. Hairy cell leukemia: evaluation of the long- neutropenic fever in patients with hairy cell leukemia. Blood 1999; 93:
term outcome in 121 patients. Cancer 2010; 116: 4788–4792. 2471–2477.
34. Hagberg H, Lundholm L. Rituximab, a chimaeric anti-CD20 monoclonal antibody in 58. Treleaven J, Gennery A, Marsh J et al. Guidelines on the use of irradiated blood
the treatment of hairy cell leukaemia. Br J Haematol 2001; 115: 609–611. components prepared by the British Committee for Standards in Haematology
35. Nieva J, Bethel K, Saven A. Phase 2 study of rituximab in the treatment of blood transfusion task force. Br J Haematol 2011; 152: 35–51.
cladribine-failed patients with hairy cell leukemia. Blood 2003; 102: 810–813. 59. Robak T. Hairy-cell leukemia variant: recent view on diagnosis, biology and
36. Forconi F, Toraldo F, Sozzi E et al. Complete molecular remission induced by treatment. Cancer Treat Rev 2011; 37: 3–10.
concomitant cladribine--rituximab treatment in a case of multi-resistant hairy cell 60. Matutes E, Wotherspoon A, Catovsky D. The variant form of hairy-cell leukaemia.
leukemia. Leuk Lymphoma 2007; 48: 2441–2443. Best Pract Res Clin Haematol 2003; 16: 41–56.
37. Else M, Dearden CE, Matutes E et al. Rituximab with pentostatin or cladribine: an 61. Narat S, Gandla J, Dogan A, Mehta A. Successful treatment of hairy cell leukemia
effective combination treatment for hairy cell leukemia after disease recurrence. variant with rituximab. Leuk Lymphoma 2005; 46: 1229–1232.
Leuk Lymphoma 2011; 52(Suppl. 2): 75–78. 62. Yoshida T, Mihara K, Sugihara S et al. Splenectomy followed by administration of
38. Ravandi F, O’Brien S, Jorgensen J et al. Phase 2 study of cladribine followed by rituximab is useful to treat a patient with hairy cell leukemia-variant. Ann Hematol
rituximab in patients with hairy cell leukemia. Blood 2011; 118: 3818–3823. 2013; 92: 711–713.
39. Else M, Osuji N, Forconi F et al. The role of rituximab in combination with 63. Kreitman RJ, Wilson W, Calvo KR et al. Cladribine with immediate rituximab for the
pentostatin or cladribine for the treatment of recurrent/refractory hairy cell treatment of patients with variant hairy cell leukemia. Clin Cancer Res 2013; 19:
leukemia. Cancer 2007; 110: 2240–2247. 6873–6881.
40. Seymour JF, Estey EH, Keating MJ, Kurzrock R. Response to interferon-alpha in 64. Telek B, Batár P, Udvardy M. Successful alemtuzumab treatment of a patient with
patients with hairy cell leukemia relapsing after treatment with 2- atypical hairy cell leukaemia variant. Orv Hetil 2007; 148: 1805–1807.
chlorodeoxyadenosine. Leukemia 1995; 9: 929–932. 65. Zinzani PL, Lauria F, Buzzi M et al. Hairy cell leukemia variant: A morphologic,
41. Hoffman MA. Interferon-alpha is a very effective salvage therapy for patients with immunologic and clinical study of 7 cases. Haematologica 1990; 75: 54–57.
hairy cell leukemia relapsing after cladribine: a report of three cases. Med Oncol 66. Busemann C, Schüler F, Krüger W et al. Late extramedullary relapse after
2011; 28: 1537–1541. allogeneic transplantation in a case of variant hairy cell leukaemia. Bone Marrow
42. Gerrie AS, Zypchen LN, Connors JM. Fludarabine and rituximab for relapsed or Transplant 2010; 45: 1117–1118.
refractory hairy cell leukemia. Blood 2012; 119: 1988–1991. 67. Uppal G, Ly V, Wang ZX et al. The utility of BRAF V600E mutation-specific
43. Burotto M, Stetler-Stevenson M, Arons E et al. Bendamustine and rituximab in antibody VE1 for the diagnosis of hairy cell leukemia. Am J Clin Pathol 2015; 143:
relapsed and refractory hairy cell leukemia. Clin Cancer Res 2013; 19: 120–125.
6313–6321. 68. Brown NA, Betz BL, Weigelin HC et al. Evaluation of allele-specific PCR and
44. Kreitman RJ, Tallman MS, Robak T et al. Phase I trial of anti-CD22 recombinant immunohistochemistry for the detection of BRAF V600E mutations in hairy cell
immunotoxin moxetumomab pasudotox (CAT-8015 or HA22) in patients with hairy leukemia. Am J Clin Pathol 2015; 143: 89–99.
cell leukemia. J Clin Oncol 2012; 30: 1822–1828. 69. Wang XJ, Kim A, Li S. Immunohistochemical analysis using a BRAF V600E
45. Tiacci E, De Carolis L, Zinzani PL et al. Vemurafenib is safe and highly active in mutation specific antibody is highly sensitive and specific for the diagnosis of hairy
hairy cell leukemia patients refractory to or relapsed after purine analogs: a phase- cell leukemia. Int J Clin Exp Pathol 2014; 7: 4323–4328.
2 Italian clinical trial. Haematologica 2014; 99(s1): Abstract S696. 70. Cornet E, Tomowiak C, Tanguy-Schmidt A et al. Long-term follow-up and second
46. Samuel J, Macip S, Dyer MJ. Efficacy of vemurafenib in hairy-cell leukemia. N Engl malignancies in 487 patients with hairy cell leukaemia. Br J Haematol 2014; 166:
J Med 2014; 370: 286–288. 390–400.
47. Sivina M, Kreitman RJ, Arons E et al. The bruton tyrosine kinase inhibitor 71. Hisada M, Chen BE, Jaffe ES, Travis LB. Second cancer incidence and cause-
ibrutinib (PCI-32765) blocks hairy cell leukaemia survival, proliferation and B cell specific mortality among 3104 patients with hairy cell leukemia: a population-
receptor signalling: a new therapeutic approach. Br J Haematol 2014; 166: based study. J Natl Cancer Inst 2007; 99: 215–222.
177–188. 72. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
48. Jansen J, Hermans J. Splenectomy in hairy cell leukemia: a retrospective among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
multicenter analysis. Cancer 1981; 47: 2066–2076. 139–144.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv200 | v

by guest
on 05 February 2018
Hodgkin’s lymphoma: ESMO Clinical Practice

D. A. Eichenauer1, A. Engert1, M. André2, M. Federico3, T. Illidge4, M. Hutchings5, & M. Ladetto6
on behalf of the ESMO Guidelines Working Group*
1
First Department of Internal Medicine, University Hospital Cologne and German Hodgkin Study Group (GHSG), Cologne, Germany; 2CHU Dinant-Godinne, UCL Namur,
Yvoir, Belgium; 3Department of Oncology, Hematology and Respiratory Diseases, University of Modena and Reggio Emilia, Modena, Italy; 4Institute of Cancer Sciences,
University of Manchester, Manchester, UK; 5Department of Hematology, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark; 6Divisione di Ematologia,
Azienda Ospedaliera Santi Antonio e Biagio e Cesare Arrigo, Alessandria, Italy
incidence fatigue, pruritus and alcohol-induced pain, as well as the results

of a physical examination, should be recorded [1].
The crude incidence of Hodgkin’s lymphoma (HL) in the Chest X-ray and a contrast-enhanced computed tomography
European Union is 2.3 and the mortality is 0.4 cases/100 000/ (CT) scan of neck, chest and abdomen are mandatory. In add-
year. Young adults aged 20–40 years are most often affected; ition, a baseline positron emission tomography (PET) should be
however, a second incidence peak is seen in individuals aged 55 carried out according to the recommendations for staging and
and older. Slightly more men than women are diagnosed with response assessment in lymphoma whenever this diagnostic tool
clinical practice
HL. Histologically, classical HL (cHL) accounting for ∼95% is available [1, 2].
guidelines
of all HL cases is distinguished from nodular lymphocyte- Given the high sensitivity of PET/CT for bone marrow involve-
predominant HL (NLPHL) representing ∼5% of all HL cases. ment, a bone marrow biopsy is no longer indicated in patients
undergoing PET/CT evaluation [III, B] [1–3]. However, bone
diagnosis marrow biopsy must be carried out if PET/CT is not available.
Full blood cell count, erythrocyte sedimentation rate (ESR)
Pathological diagnosis should be made according to the World and blood chemistry including C-reactive protein, alkaline
Health Organization (WHO) classification from a sufficiently large phosphatase, lactate dehydrogenase, liver enzymes and albumin
surgical specimen or excisional lymph node biopsy to provide are obligatory. Screening for hepatitis B, hepatitis C and human
enough material for fresh frozen and formalin-fixed samples. immunodeficiency virus (HIV) is compulsory [II–III, A].
In cHL, the presence of Hodgkin and Reed–Sternberg (HRS) Staging is carried out according to the Ann Arbor classification
cells is disease-defining while the detection of lymphocyte in consideration of defined clinical risk factors. After completion
predominant (LP) cells is required for the diagnosis of NLPHL. of staging, patients are allocated to three categories (limited, inter-
The immunophenotype of the malignant cells in cHL and mediate and advanced stages). Table 2 illustrates the European
NLPHL differs significantly. In contrast to HRS cells that stain Organisation for Research and Treatment of Cancer/Lymphoma
consistently positive for CD30 and CD15, occasionally positive Study Association and the German Hodgkin Study Group defini-
for CD20 and negative for CD45, LP cells are characterised by the tions of limited, intermediate and advanced stages [II–III, A].
expression of CD20 and CD45 but they lack CD15 and CD30. To identify patients at increased risk for acute and/or long-
term complications, cardiac and pulmonary function tests
staging and risk assessment should be carried out before the start of treatment.
Since chemotherapy and radiotherapy (RT) can potentially
The diagnostic work-up is shown in Table 1. The medical
cause permanent fertility damage, reproductive counselling
history including the presence of B symptoms (fever, drenching
must be offered to young patients of both genders before
night sweats, unexplained weight loss >10% of total body weight
treatment.
over 6 months) and other disease-related symptoms such as
*Correspondence to: ESMO Guidelines Working Group, ESMO Head Office, Via treatment of cHL
L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland.
E-mail: clinicalguidelines@esmo.org limited-stage patients
†
Combined modality treatment consisting of a brief chemother-
2014. This publication supersedes the previously published version—Ann Oncol 2011; apy followed by RT was shown to result in superior tumour
22 (Suppl. 6): vi55–vi58. control compared with RT alone [I, A] [4, 5] (Figure 1).

by guest
on 05 February 2018
Currently, two or three cycles of adriamycin/bleomycin/vin- The question of whether RT can be omitted in patients with
blastine/dacarbazine (ABVD) (Table 3) followed by involved- complete metabolic response at interim PET is currently a
field RT (IFRT) is considered standard of care for limited-stage matter of debate and cannot be fully answered to date. Several
HL. A large multicentre trial in which patients were randomly randomised trials addressing this issue have been initiated in
assigned to either two or four cycles of ABVD followed by either recent years. Emerging data consistently demonstrate a progres-
20 or 30 Gy IFRT showed similar freedom from treatment sion-free survival advantage also for patients with a complete
failure (FFTF) and overall survival (OS) rates for all treatment metabolic response at interim PET when treatment with com-
groups. Thus, the least toxic approach consisting of two cycles bined modality approaches is applied. A population that can be
of ABVD followed by 20 Gy IFRT appears to be sufficient for safely treated with chemotherapy alone could not yet be defined
limited-stage HL [I, A] [6]. However, the current RT guidelines [8, 9]. Therefore, interim PET-guided treatment in limited-stage
of the International Lymphoma Radiation Oncology Group HL is not recommended outside clinical studies.
(ILROG) recommend involved-site RT (ISRT) after chemother-
apy in limited stages although this recent strategy has not yet
been validated in a prospective study [7]. intermediate-stage patients
Intermediate-stage HL is usually treated with combined modal-
ity approaches.
Four cycles of ABVD followed by 30 Gy IFRT is widely consid-
Table 1. Diagnostic work-up in Hodgkin’s lymphoma
ered standard for intermediate-stage HL [I, A] [5]. In patients
Diagnosis ≤60 years who are eligible for a more intensive treatment, this
Lymph node biopsy (or a biopsy from another organ with standard is challenged by a protocol consisting of two cycles of
suspected affection) bleomycin/etoposide/adriamycin/cyclophosphamide/vincristine/
Staging and risk stratification procarbazine/prednisone in escalated dose (BEACOPPescalated)
Medical history and physical examination (Table 4) followed by two cycles of ABVD and 30 Gy IFRT.
X-ray of the chest After a median follow-up of 43 months, FFTF with this protocol
Contrast-enhanced CT scan of neck, chest and abdomen was superior in comparison with four cycles of ABVD followed
PET by 30 Gy IFRT. An advantage in OS could not be shown [I, B–C]
Full blood cell count and blood chemistry [10]. Although no results of a prospective study addressing this
HBV, HCV and HIV screening issue are available to date, the ILROG guidelines recommend
Pre-treatment examinations ISRT instead of IFRT after chemotherapy in intermediate
ECG stages [7].
Echocardiography The question of whether RT is dispensable in intermediate-
Pulmonary function test stage patients with complete metabolic response at interim PET
Reproductive counselling (in younger patients) is unanswered. Trials addressing this issue are ongoing.
Serum pregnancy test (in younger female patients)
CT, computed tomography; PET, positron emission tomography; advanced-stage patients

HBV, hepatitis B; HCV, hepatitis C; HIV, human immunodeficiency
Advanced-stage HL is usually treated with chemotherapy alone.
virus; ECG, electrocardiography.
Additional RT is confined to patients with residual disease after
chemotherapy.
Table 2. Definition of Hodgkin’s lymphoma risk groups according to the European Organisation for Research and Treatment of Cancer /Lymphoma
Study Association and the German Hodgkin Study Group
Treatment group EORTC/LYSA GHSG
Limited stages CS I–II without risk factors (supra-diaphragmatic) CS I–II without risk factors
Intermediate stages CS I–II with ≥1 risk factors (supra-diaphragmatic) CS I, CS IIA with ≥1 risk factors;
CS IIB with risk factors C/D, but not A/B
Advanced stages CS III–IV CS IIB with risk factors A/B, CS III/IV
Risk factors (A) Large mediastinal mass (A) Large mediastinal mass
(B) Age ≥50 years (B) Extranodal disease
(C) Elevated ESR (C) Elevated ESR
(D) ≥4 nodal areas (D) ≥3 nodal areas
Elevated ESR: >50 mm/h without B symptoms, >30 mm/h with B symptoms.
Large mediastinal mass: more than one-third of the maximum horizontal chest diameter.
B symptoms: fever, night sweat, unexplained weight loss >10% over 6 months.
EORTC: European Organisation for Research and Treatment of Cancer; LYSA: Lymphoma Study Association; GHSG: German Hodgkin Study Group;
CS: clinical stage; ESR: erythrocyte sedimentation rate.
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu181 | iii

by guest
on 05 February 2018
Diagnosis of HL
Limited stages Intermediate stages Advanced stages
4 cycles of ABVD
6 to 8 cycles of ABVD
or
2 or 3 cycles of or
2 cycles of BEACOPPesc +
ABVD 6 cycles of BEACOPPesc
2 cycles of ABVD
(£60 years)
(£60 years)
Localised RT to residual
lymphoma >1.5 cm
20 Gy IFRT 30 Gy IFRT (after ABVD)
or or or
ISRT ISRT Localised RT to PET-positive
residual lymphoma >2.5 cm
(after BEACOPPesc)
Follow-up
Figure 1. Therapeutic algorithm for newly diagnosed Hodgkin’s lymphoma. HL, Hodgkin’s lymphoma; RT, radiotherapy; ABVD, adriamycin, bleomycin, vin-
blastine, dacarbazine; BEACOPPesc, bleomycin, etoposide, adriamycin, cyclophosphamide, vincristine, procarbazine, prednisone escalated dose regimen; ISRT,
involved-site radiotherapy; PET, positron emission tomography; NLPHL, nodular lymphocyte-predominant Hodgkin’s lymphoma; IFRT, involved-field RT.
Table 3. The adriamycin, bleomycin, vinblastine, dacarbazine Table 4. The bleomycin/etoposide/adriamycin/cyclophosphamide/

(ABVD) regimen vincristine/ procarbazine/ prednisone in escalated dose
(BEACOPPescalated) regimen
Adriamycin 25 mg/m2 i.v. Days 1 + 15
Bleomycin 10 mg/m2 i.v. Days 1 + 15 Bleomycin 10 mg/m2 i.v. Day 8
Vinblastine 6 mg/m2 i.v. Days 1 + 15 Etoposide 200 mg/m2 i.v. Days 1–3
Dacarbazine 375 mg/m2 i.v. Days 1 + 15 Adriamycin 35 mg/m2 i.v. Day 1
Cyclophosphamide 1250 mg/m2 i.v. Day 1
Recycle: day 29. Vincristine 1.4 mg/m2 i.v. Day 8
(maximum: 2 mg)
Procarbazine 100 mg/m2 p.o. Days 1–7
Prednisone 40 mg/m2 p.o. Days 1–14
Patients ≤60 years are treated with either six to eight cycles of G-CSF s.c. From day 8
ABVD followed by localised RT of residual lymphoma larger
than 1.5 cm or six cycles of BEACOPPescalated followed by loca- Recycle: day 22.
lised RT of PET-positive residual lymphoma larger than 2.5 cm G-CSF, granulocyte colony-stimulating factor.
[I, A] [11, 12]. Several trials randomly comparing ABVD and
BEACOPPescalated have shown a superior tumour control with
BEACOPPescalated [13–15]. A recent network meta-analysis in-
cluding 9993 patients also indicated a significantly better OS with Retrospective analyses have indicated that early interim
BEACOPPescalated when compared with ABVD. The survival PET might be a good predictor for treatment failure in patients
advantage was 10% at 5 years [16]. However, given the relevant with advanced HL receiving ABVD chemotherapy [18, 19].
acute toxicity of BEACOPPescalated, appropriate surveillance and Therefore, ongoing trials aim at guiding treatment on the basis of
supportive care must be available when this protocol is used. In early interim PET which is used to distinguish between patients
patients >60 years, the BEACOPP regimen should not be given, who can potentially be cured with reduced therapy and patients
as an increased rate of treatment-related mortality has been who require standard or even more intensive treatment. However,
observed in this age group [II, A] [17]. Thus, ABVD represents given a lack of mature prospective data, treatment stratification
the standard regimen for older HL patients who are fit enough on the basis of early interim PET cannot be considered standard
for treatment with multi-agent chemotherapy. as yet and further evidence from randomised trials is necessary.
iii | Eichenauer et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
relapsed disease non-Hodgkin’s lymphoma must be excluded. According to
newer analyses, transformation rates appear to be substantially
For most patients with refractory or relapsed HL, the treatment higher than previously reported [IV, A] [34, 35].
of choice consists of high-dose chemotherapy followed by au- Localised NLPHL relapses can be effectively treated with
tologous stem cell transplantation (ASCT) [II, A] [20, 21]. rituximab alone [III, B] [36].
High-risk patients may benefit from tandem ASCT [III, B] [22]. Patients with more advanced disease at relapse often require a
Salvage regimens such as dexamethasone/high-dose Ara-C/ more aggressive salvage therapy possibly combined with an
cisplatin (DHAP), ifosfamide/gemcitabine/vinorelbine (IGEV) anti-CD20 antibody. However, prospective data on the use of
or ifosfamide/carboplatin/etoposide (ICE) are given to reduce high-dose chemotherapy followed by ASCT are not available
the tumour burden and mobilise stem cells before high-dose yet.
chemotherapy and ASCT [II–III, A] [23–25]. Given the lack of CD30 on the malignant LP cells in NLPHL,
A subset of low-risk patients relapsing after primary treat- brentuximab vedotin does not represent a treatment option in
ment with two cycles of chemotherapy followed by RT can be this entity.
successfully salvaged with a second, more intensive convention-
al chemotherapy such as BEACOPPescalated [IV, B–C] [26].
In some patients with localised late relapse, salvage RT alone response evaluation
appears to be sufficient [IV, B–C] [27]. Interim response evaluation by contrast-enhanced CT should be
The use of the antibody-drug conjugate brentuximab vedotin carried out after completion of chemotherapy/before RT in limited
represents an option in patients failing ASCT. After a pivotal and intermediate stages and after four cycles of chemotherapy as
phase II study including 102 HL patients with relapse after ASCT well as before RT in advanced stages. Retrospective studies includ-
had revealed an overall response rate (ORR) of 75% with single- ing advanced-stage and relapsed patients, respectively, have shown
agent brentuximab vedotin, the drug was recently approved for that interim PET appears to be a useful tool to identify poor-risk
the treatment of such patients [III, B] [28]. Alternatively, patients individuals [18, 19, 37]. However, interim PET-guided treatment
can be enrolled in clinical trials evaluating novel agents. cannot be considered standard and should be restricted to clinical
Reduced-intensity conditioning allogeneic stem cell trans- trials except for the decision of whether patients with advanced
plantation (RIC-aSCT) can be considered in young, chemo- HL receiving BEACOPPescalated require RT [12].
sensitive patients in good general condition who relapse after Final staging should be carried out after completion of treat-
high-dose chemotherapy and ASCT [III, C] [29]. However, ment. Physical examination, laboratory analyses and contrast-
RIC-aSCT is not a standard approach in HL and should be con- enhanced CT are mandatory. In addition, PET should be
ducted within clinical trials whenever possible. carried out at final staging according to the guidelines for
In patients with multiple relapses who have no other treat- staging and response assessment in lymphoma whenever this
ment options, acceptable remission rates, satisfying quality of diagnostic tool is available [1, 2].
life and prolonged survival can be achieved by palliative single-
agent chemotherapy with gemcitabine or bendamustine and/or
regional RT [30, 31]. As brentuximab vedotin has also been prognosis
approved for the treatment of HL patients with disease recur- With modern treatment strategies, 80%–90% of HL patients
rence after at least two lines of treatment who are not candidates achieve permanent remission and can be considered cured.
for high-dose chemotherapy followed by ASCT, its use can also
be considered in this patient group.
treatment of NLPHL In HL, personalised treatment based on certain genetic features
as known for some malignancies is not established.
stage IA without risk factors Treatment intensity is chosen according to the clinical stage
30 Gy IFRT alone is the standard treatment for stage IA NLPHL and the presence or absence of clinical risk factors (as described
patients without risk factors [III, A] [32]. in the staging and risk assessment section). The use of risk-
adapted therapy has led to excellent cure rates in HL patients ir-
respective of the stage at diagnosis.
other stages Prospective studies evaluating interim PET-guided strategies
Usually, NLPHL is treated identically to cHL in all stages except have been initiated, with the aim to discriminate between low-
for stage IA without risk factors [33]. As the malignant LP cells risk patients who may be sufficiently treated with reduced-inten-
of NLPHL consistently express CD20, addition of an anti-CD20 sity approaches and high-risk patients who require standard or
antibody may improve treatment efficacy [V, C]. However, even intensified treatment. In patients with limited and inter-
prospective data on this issue are not yet available. mediate stages, the goal is to define a group of patients with
complete metabolic response after chemotherapy not requiring
consolidating RT. In advanced HL, it has been shown that RT is
relapsed NLPHL patients dispensable in patients without PET-positive residual lymph-
Even more importantly than in cHL, a renewed biopsy should oma larger than 2.5 cm after BEACOPPescalated chemotherapy.
be obtained in patients with suspected NLPHL relapse before Ongoing trials in advanced HL evaluate whether it is possible to
salvage therapy is initiated, since transformation into aggressive modify the intensity of chemotherapy based on the result of
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu181 | iii

by guest
on 05 February 2018

out every three months for the first half year, every 6 months
until the fourth year and once per year thereafter [V, B].
• After staging is completed, Hodgkin’s lymphoma (HL) patients Additional evaluation of thyroid function (thyroid-stimulat-
are allocated to distinct risk groups depending on their clinical ing hormone) after irradiation of the neck at one, two and at
stage and the presence of clinical risk factors. least five years is recommended. Furthermore, testosterone and
• First-line treatment of HL patients usually consists of combined oestrogen levels should be monitored, particularly in younger
modality approaches or chemotherapy alone. Intensity of patients who had intensive chemotherapy.
treatment depends on the patient’s risk profile. CT scans and previously pathologic radiographic tests must
• The standard of care for most patients with disease recurrence be carried out once to confirm the remission status. Thereafter,
after first-line treatment consists of high-dose chemotherapy surveillance scans are not indicated unless clinical symptoms
followed by autologous stem cell transplantation (ASCT). occur [IV, B] [1, 2].
• Brentuximab vedotin is approved for the treatment of patients Patients should be asked for symptoms indicating the exist-
failing ASCT and those with multiple relapses. ence of long-term toxicity, particularly of heart and lung.
• HL patients should be treated within clinical trial protocols Cancer screening should be conducted regularly due to the
whenever possible.
increased risk of haematological and solid secondary malignan-
• Follow-up should be conducted regularly to detect disease
cies after HL treatment.
recurrence and therapy-related late effects.
note
A summary of recommendations is provided in Table 5. Levels
Table 6. Levels of evidence and grades of recommendation of evidence and grades of recommendation have been applied
(adapted from the Infectious Diseases Society of America-United using the system shown in Table 6. Statements without grading
States Public Health Service Grading Systema) were considered justified standard clinical practice by the
experts and the ESMO faculty.
Levels of evidence

acknowledgements
meta-analyses of well-conducted randomised trials We gratefully thank Dr Sally F. Barrington and Dr Bruce
without heterogeneity D. Cheson for providing their manuscripts on the role of
II Small randomised trials or large randomised trials with a imaging in staging and response assessment.
heterogeneity conflict of interest
IV Retrospective cohort studies or case–control studies AE has reported research funding and speakers’ bureau for
V Studies without control group, case reports, experts Takeda. ML has reported speakers’ bureau from Celgene,
opinions Janssen-Cilag, Roche, Bayer, Amgen, Mundipharma; research
contracts from Celgene, Pfizer, Mundipharma, Roche; funds
received from Amgen, Roche, Italfarmaco. The other authors
A Strong evidence for efficacy with a substantial clinical have reported no potential conflicts of interest.
benefit, strongly recommended
B Strong or moderate evidence for efficacy but with a limited
clinical benefit, generally recommended references
C Insufficient evidence for efficacy or benefit does not 1. Cheson BD, Fisher RI, Barrington SF et al. Recommendations for initial evaluation,
outweigh the risk or the disadvantages (adverse events, staging and response assessment of Hodgkin and non-Hodgkin lymphoma—the
costs, … ), optional Lugano classification. J Clin Oncol 2014; in press.
D Moderate evidence against efficacy or for adverse outcome, 2. Barrington SF, Mikhaeel NG, Kostakoglu L et al. The role of imaging in the staging
generally not recommended and response assessment of lymphoma: consensus of the ICML Imaging Working
E Strong evidence against efficacy or for adverse outcome, Group. J Clin Oncol 2014; in press.
never recommended 3. El-Galaly TC, d’Amore F, Mylam KJ et al. Routine bone marrow biopsy has little or
no therapeutic consequence for positron emission tomography/computed
a
By permission of the Infectious Diseases Society of America [38]. tomography-staged treatment-naive patients with Hodgkin lymphoma. J Clin Oncol
2012; 30: 4508–4514.
4. Engert A, Franklin J, Eich HT et al. Two cycles of doxorubicin, bleomycin,
vinblastine, and dacarbazine plus extended-field radiotherapy is superior to
early interim PET. However, no mature data addressing this
radiotherapy alone in early favorable Hodgkin’s lymphoma: final results of the
issue are available to date. GHSG HD7 trial. J Clin Oncol 2007; 25: 3495–3502.
5. Fermé C, Eghbali H, Meerwaldt JH et al. Chemotherapy plus involved-field
follow-up radiation in early-stage Hodgkin’s disease. N Engl J Med 2007; 357:
1916–1927.
History, physical examination and laboratory analysis including 6. Engert A, Plutschow A, Eich HT et al. Reduced treatment intensity in patients with
full blood cell count, ESR and blood chemistry should be carried early-stage Hodgkin’s lymphoma. N Engl J Med 2010; 363: 640–652.
iii | Eichenauer et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
7. Specht L, Yahalom J, Illidge T et al. Modern radiation therapy for Hodgkin Hodgkin’s lymphoma: results of the prospective multicenter H96 trial by the GELA/
lymphoma: field and dose guidelines from the International Lymphoma Radiation SFGM study group. J Clin Oncol 2008; 26: 5980–5987.
Oncology Group (ILROG). Int J Radiat Oncol Biol Phys 2014; 89: 854–862. 23. Josting A, Rudolph C, Reiser M et al. Time-intensified dexamethasone/cisplatin/
8. Raemaekers JM, André MP, Federico M et al. Omitting radiotherapy in early cytarabine: an effective salvage therapy with low toxicity in patients with relapsed
positron emission tomography-negative stage I/II Hodgkin lymphoma is associated and refractory Hodgkin’s disease. Ann Oncol 2002; 13: 1628–1635.
with an increased risk of early relapse: clinical results of the preplanned interim 24. Santoro A, Magagnoli M, Spina M et al. Ifosfamide, gemcitabine, and vinorelbine:
analysis of the randomized EORTC/LYSA/FIL H10 Trial. J Clin Oncol 2014; 32: a new induction regimen for refractory and relapsed Hodgkin’s lymphoma.
1188–1194. Haematologica 2007; 92: 35–41.
9. Radford J, Barrington S, Counsell N et al. Involved field radiotherapy versus no 25. Moskowitz CH, Nimer SD, Zelenetz AD et al. A 2-step comprehensive high-dose
further treatment in patients with clinical stages IA and IIA Hodgkin lymphoma and chemoradiotherapy second-line program for relapsed and refractory Hodgkin
a ‘negative’ PET scan after 3 cycles ABVD. Results of the UK NCRI RAPID trial. disease: analysis by intent to treat and development of a prognostic model. Blood
Blood (ASH Annual Meeting Abstracts) 2012; 120: 547. 2001; 97: 616–623.
10. von Tresckow B, Plütschow A, Fuchs M et al. Dose-intensification in early 26. Sieniawski M, Franklin J, Nogová L et al. Outcome of patients experiencing
unfavorable Hodgkin’s lymphoma: final analysis of the German Hodgkin Study progression or relapse after primary treatment with two cycles of chemotherapy
Group HD14 trial. J Clin Oncol 2012; 30: 907–913. and radiotherapy for early-stage favorable Hodgkin’s lymphoma. J Clin Oncol
11. Canellos GP, Niedzwiecki D, Johnson JL. Long-term follow-up of survival in 2007; 25: 2000–2005.
Hodgkin’s lymphoma. N Engl J Med 2009; 361: 2390–2391. 27. Josting A, Nogová L, Franklin J et al. Salvage radiotherapy in patients with relapsed
12. Engert A, Haverkamp H, Kobe C et al. Reduced-intensity chemotherapy and PET- and refractory Hodgkin’s lymphoma: a retrospective analysis from the German
guided radiotherapy in patients with advanced stage Hodgkin’s lymphoma (HD15 Hodgkin Lymphoma Study Group. J Clin Oncol 2005; 23: 1522–1529.
trial): a randomised, open-label, phase 3 non-inferiority trial. Lancet 2012; 379: 28. Younes A, Gopal AK, Smith SE et al. Results of a pivotal phase II study of
1791–1799. brentuximab vedotin for patients with relapsed or refractory Hodgkin’s lymphoma.
13. Federico M, Luminari S, Iannitto E et al. ABVD compared with BEACOPP compared J Clin Oncol 2012; 30: 2183–2189.
with CEC for the initial treatment of patients with advanced Hodgkin’s lymphoma: 29. Sureda A, Canals C, Arranz R et al. Allogeneic stem cell transplantation after
results from the HD2000 Gruppo Italiano per lo Studio dei Linfomi Trial. J Clin reduced intensity conditioning in patients with relapsed or refractory Hodgkin’s
Oncol 2009; 27: 805–811. lymphoma. Results of the HDR-ALLO study—a prospective clinical trial by the
14. Viviani S, Zinzani PL, Rambaldi A et al. ABVD versus BEACOPP for Hodgkin’s Grupo Español de Linfomas/Trasplante de Médula Osea (GEL/TAMO) and the
lymphoma when high-dose salvage is planned. N Engl J Med 2011; 365: Lymphoma Working Party of the European Group for Blood and Marrow
203–212. Transplantation. Haematologica 2012; 97: 310–317.
15. Carde PP, Karrasch M, Fortpied C et al. ABVD (8 cycles) versus BEACOPP 30. Santoro A, Bredenfeld H, Devizzi L et al. Gemcitabine in the treatment of refractory
(4 escalated cycles => 4 baseline) in stage III-IV high-risk Hodgkin lymphoma Hodgkin’s disease: results of a multicenter phase II study. J Clin Oncol 2000; 18:
(HL): first results of EORTC 20012 Intergroup randomized phase III clinical trial. J 2615–2619.
Clin Oncol (ASCO Meeting Abstracts) 2012; 30: 8002. 31. Moskowitz AJ, Hamlin PA, Jr, Perales MA et al. Phase II study of bendamustine in
16. Skoetz N, Trelle S, Rancea M et al. Effect of initial treatment strategy on survival of relapsed and refractory Hodgkin lymphoma. J Clin Oncol 2013; 31: 456–460.
patients with advanced-stage Hodgkin’s lymphoma: a systematic review and 32. Nogová L, Reineke T, Eich HT et al. Extended field radiotherapy, combined modality
network meta-analysis. Lancet Oncol 2013; 14: 943–952. treatment or involved field radiotherapy for patients with stage IA lymphocyte-
17. Ballova V, Rüffer JU, Haverkamp H et al. A prospectively randomized trial carried predominant Hodgkin’s lymphoma: a retrospective analysis from the German
out by the German Hodgkin Study Group (GHSG) for elderly patients with advanced Hodgkin Study Group (GHSG). Ann Oncol 2005; 16: 1683–1687.
Hodgkin’s disease comparing BEACOPP baseline and COPP-ABVD (study 33. Nogová L, Reineke T, Brillant C et al. Lymphocyte-predominant and classical
HD9elderly). Ann Oncol 2005; 16: 124–131. Hodgkin’s lymphoma: a comprehensive analysis from the German Hodgkin Study
18. Hutchings M, Loft A, Hansen M et al. FDG-PET after two cycles of chemotherapy Group. J Clin Oncol 2008; 26: 434–439.
predicts treatment failure and progression-free survival in Hodgkin lymphoma. 34. Biasoli I, Stamatoullas A, Meignin V et al. Nodular, lymphocyte-predominant
Blood 2006; 107: 52–59. Hodgkin lymphoma: a long-term study and analysis of transformation to diffuse
19. Gallamini A, Hutchings M, Rigacci L et al. Early interim 2-[18F]fluoro-2-deoxy-D- large B-cell lymphoma in a cohort of 164 patients from the Adult Lymphoma Study
glucose positron emission tomography is prognostically superior to international Group. Cancer 2010; 116: 631–639.
prognostic score in advanced-stage Hodgkin’s lymphoma: a report from a joint 35. Al-Mansour M, Connors JM, Gascoyne RD et al. Transformation to aggressive
Italian-Danish study. J Clin Oncol 2007; 25: 3746–3752. lymphoma in nodular lymphocyte-predominant Hodgkin’s lymphoma. J Clin Oncol
20. Linch DC, Winfield D, Goldstone AH et al. Dose intensification with autologous 2010; 28: 793–799.
bone-marrow transplantation in relapsed and resistant Hodgkin’s disease: results 36. Schulz H, Rehwald U, Morschhauser F et al. Rituximab in relapsed lymphocyte-
of a BNLI randomised trial. Lancet 1993; 341: 1051–1054. predominant Hodgkin lymphoma: long-term results of a phase 2 trial by the
21. Schmitz N, Pfistner B, Sextro M et al. Aggressive conventional chemotherapy German Hodgkin Lymphoma Study Group (GHSG). Blood 2008; 111: 109–111.
compared with high-dose chemotherapy with autologous haemopoietic stem-cell 37. Moskowitz AJ, Yahalom J, Kewalramani T et al. Pretransplantation functional
transplantation for relapsed chemosensitive Hodgkin’s disease: a randomised trial. imaging predicts outcome following autologous stem cell transplantation for
Lancet 2002; 359: 2065–2071. relapsed and refractory Hodgkin lymphoma. Blood 2010; 116: 4934–4937.
22. Morschhauser F, Brice P, Fermé C et al. Risk-adapted salvage treatment with 38. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections among
single or tandem autologous stem-cell transplantation for first relapse/refractory hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33: 139–144.
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu181 | iii

by guest
on 05 February 2018
Peripheral T-cell lymphomas: ESMO Clinical Practice

F. d’Amore1, P. Gaulard2, L. Trümper3, P. Corradini4, W.-S. Kim5, L. Specht6, M. Bjerregaard
Pedersen1 & M. Ladetto7, on behalf of the ESMO Guidelines Committee*
1
Department of Hematology, Aarhus University Hospital, Aarhus, Denmark; 2Department of Pathology, Hôpital Henri Mondor, Créteil, France; 3Department of Hematology
and Oncology, Georg August University, Göttingen, Germany; 4Department of Hematology, Fondazione IRCCS Istituto Nazionale dei Tumori, University of Milan, Milan,
Italy; 5Department of Medicine, Division of Hematology-Oncology, Samsung Medical Center, Seoul, Korea; 6Department of Oncology and Hematology, Rigshospitalet,
University of Copenhagen, Copenhagen, Denmark; 7Divisione di Ematologia, Azienda Ospedaliera Santi Antonio e Biagio e Cesare Arrigo, Alessandria, Italy
introduction autoimmune disorders, such as HSTCL to Crohn’s disease [3]. In

PTCL, the male/female ratio is 2:1 and the median age at diagnosis
The present guidelines cover the systemic subtypes of primary is between the sixth and seventh decades of life, but both sex and
nodal and primary extranodal peripheral T-cell lymphomas age patterns vary according to different subtypes [2, 4, 5]. The
(PTCLs). ESMO guidelines for primary cutaneous T-cell lymph- ALCL ALK+ subtype has a better prognosis than the other PTCL
omas are published separately [1]. Primary leukaemic PTCL sub- entities, including its ALK−counterpart [6]. Recent reports have
types (i.e. T-cell prolymphocytic leukaemia, T-cell large granular suggested that the prognostic difference between ALCL ALK+ and
lymphocytic leukaemia, adult T-cell leukaemia/lymphoma and ag- ALCL ALK− may at least be partly due to age-related differences
clinical practice
gressive NK-cell leukaemia) are not covered by the present guide- (ALK+ patients are generally younger than the other PTCL
guidelines
lines. Primary nodal PTCLs include PTCL-not otherwise specified patients) [7]. HSTCL occurs most frequently in younger to middle-
(PTCL-NOS), anaplastic large-cell lymphoma (ALCL), both aged males in the setting of immunosuppressive treatment [8].
fusion protein ALCL anaplastic lymphoma kinase positive (ALCL
ALK+) and ALCL anaplastic lymphoma kinase negative (ALCL
ALK−), and angioimmunoblastic T-cell lymphoma (AITL). The diagnosis
primary extranodal PTCL subtypes covered by the present guide- A PTCL diagnosis should be made by an expert haematopatholo-
lines comprise enteropathy-associated T-cell lymphoma (EATL), gist and should, whenever possible, rely on an excisional tumour
extranodal natural killer/T-cell lymphoma (ENKTCL), and hepa- tissue biopsy that provides enough material for formalin-fixed
tosplenic T-cell lymphoma (HSTCL). samples. According to the WHO classification (2008), the distinc-
tion among different PTCL entities requires the integration of
epidemiology the clinical picture, morphology, immunohistochemistry, flow
cytometry, cytogenetics, and molecular biology [3]. In PTCL, the
PTCLs are uncommon and heterogeneous malignant lymphopro- indication of the neoplastic nature of a given T-cell population is
liferative disorders that originate from post-thymic (peripheral) based on (i) morphology, (ii) aberrant T-cell phenotype, and (iii)
T cells or mature natural killer (NK) cells. They represent 10%– clonally rearranged T-cell receptor (TCR) genes (αβ versus γδ
15% of all non-Hodgkin’s lymphomas. Nodal subtypes are the genotypes) [9].
most frequent in Caucasian patients (>80% of PTCL in Europe, Table 1 summarises the immunophenotype of the PTCL en-
PTCL-NOS 34%, AITL 28%, ALCL ALK+ 6%, ALCL ALK− 9%) tities covered by the present guidelines along with their TCR re-
[2]. In Asia, the PTCL incidence is higher due to the endemic oc- arrangement features and putative cell of origin. Accumulating
currence of the Epstein-Barr virus (EBV)-associated ENKTCL evidence indicates that information on TCR and the cell of origin
(44% of PTCL in Asia excluding Japan, where there is a relatively plays an important role in both tumour biology and clinical be-
higher frequency of adult T-cell leukaemia/lymphoma) [2]. EATL haviour, underscoring the clinical relevance of this information in
is more frequent in Northern Europe (9%–10% as compared with the light of an increasing number of targeted therapeutic options.
1%–2% in Asia) [2], where there is a higher occurrence of human TIA1, granzyme B and perforin suggest a cytotoxic profile, which
leukocyte antigen (HLA) haplotypes associated with coeliac may imply a more aggressive clinical behaviour in PTCL-NOS
disease. Other PTCL subtypes have been associated with chronic [10]. At least three of the following markers: CD10, Bcl6,
CXCL13, PD1, SAP, ICOS, and CCR5 are suggestive of a follicular
*Correspondence to: ESMO Guidelines Committee, ESMO Head Office, via L. Taddei 4, T-helper (FTH) cell origin [9, 11, 12]. Although FTH cells are
CH-6962 Viganello-Lugano, Switzerland. considered to be the cell of origin in AITL, this diagnosis should
also be based on morphological parameters such as hyperplasia
†
Approved by the ESMO Guidelines Committee: June 2015. of follicular dendritic cells (FDCs), arborising high endothelial

by guest
on 05 February 2018
Table 1. Nodal and extranodal PTCL subtypes—cell of origin and related phenotypes (adapted from [9])
PTCL entity Immunophenotypic features TCR Presumed cell of origin
Nodal PTCL-NOS CD4>CD8, frequent antigen loss (CD5, CD7), CD30+/−, αβ, rarely γδ Variable, mostly T-helper cell
CD56+/−, subset FTH features, cytotoxic granules+/−
AITL CD4+, CD10+/−, BCL+/−, CXCL13+, PD1+, ICOS+/−, αβ FTH
SAP+/, CCR5+/−, hyperplasia of FDC, EBV+ B blasts
ALCL ALK+ ALK+, CD30+, EMA+, CD25+, cytotoxic granules+, CD4+/−, αβ Cytotoxic T-cell
CD3+/−
ALCL ALK- ALK−, CD30+, EMA+, CD25+, cytotoxic granules+, CD4+/−, αβ Cytotoxic T-cell
CD3+/−
Extranodal EATL, type 1 CD8(+)/−, CD56−, HLA-DQ2/-DQ8 αβ Intra-epithelial T cells (αβ), pre-
existing enteropathy
EATL, type 2 CD8+, CD56+, HLA-DQ2/-DQ8 γδ or αβ Intra-epithelial T cells or NK,
no pre-existing enteropathy
NKTCL CD2+, CD56+, surface CD3−, cytoplasmic CD3ε+, gr B+, TCR in germline NK, rarely cytotoxic T cells
TIA-1+, perforin+, EBV+, LMP1 configuration, rarely
αβ or γδ
HSTCL CD3+, CD56+/−, CD4−, CD8+/−, CD5−, TIA1+, gr M+, γδ, rarely αβ Cytotoxic T cell of the innate
gr B−, perforin− immune system
PTCL, peripheral T-cell lymphomas; PTCL-NOS, PTCL-not otherwise specified; AITL, angioimmunoblastic T-cell lymphoma; ALCL ALK+, anaplastic
large-cell lymphoma anaplastic lymphoma kinase positive; ALCL ALK-, ALCL anaplastic lymphoma kinase negative; EATL, enteropathy-associated T-cell
lymphoma; NKTCL, natural killer/T-cell lymphoma; HSTCL, hepatosplenic T-cell lymphoma; FTH, follicular T helper; FDC, follicular dendritic cell;
EMA, epithelial membrane antigen; HLA, human leukocyte antigen; EBV, Epstein-Barr virus; TCR, T-cell receptor; NK, natural killer.
venules and a substantial B-cell component, including EBV- (CT) scan of the chest and abdomen, as well as a bone marrow
infected B-cell blasts [13]. ENKTCL cases show intra-cytoplasmic aspirate and biopsy. 18Fluorodeoxyglucose positron emission
CD3 (ɛ-chain), in contrast to other PTCL subtypes that only tomography combined with computed tomography (18F-FDG
express CD3 on the cell surface [3]. CD56 is helpful in differenti- PET/CT) is increasingly used in nodal PTCL at baseline and re-
ating between EATL type I (CD8+/CD56−) and type II (CD8 staging, but its role at the subtype-specific level still needs
−/CD56+), which is also more often γδ+ and is not associated further elucidation. PET may be useful for detecting residual
with coeliac disease [3]. CD30 plays a central role in the recogni- disease at the end of treatment, although residual FDG-avid
tion of ALCL. ALCL is systematically PAX5-negative, frequently lesions lack specificity and biopsy confirmation is recommended.
epithelial membrane antigen (EMA)-positive and, in one-third of The use of PET/CT is recommended in ENKTCL, where it is
the cases, CD45-negative. It is further categorised as ALK+ or documented to be a valuable modality for staging and treatment
ALK− depending on the occurrence or lack of occurrence of the planning [14–18].
classical t(2;5) translocation (or one of its variants) [3, 9]. CD20
and PAX5 allow for the identification of B-cell components and prognostic indices
can help in distinguishing ALCL ALK− from morphologically ag-
The International Prognostic Index (IPI) [19] is the most com-
gressive classical Hodgkin’s lymphoma (PAX5+) with anaplastic
monly used prognostic tool in nodal PTCL. A prognostic index
features. CD21 is useful in revealing the content of FDCs in AITL;
for PTCL-NOS [20] with later modification [21] has been pro-
CD68 visualises the histiocytic component that can occasionally
posed, but does not univocally appear to be more useful than
outnumber the neoplastic cell population (e.g. lymphoepithelioid
the original IPI [4, 22]. For clinical practice purposes, the IPI is
PTCL-NOS, Lennert’s variant and the lymphohistiocytic variant
therefore still the recommended tool. Male sex has been
of ALCL). The assessment of EBV (Epstein-Barr encoding region
reported as an adverse prognostic factor [5, 23]. In ENKTCL,
[EBER] in situ hybridisation) is important in T-cell malignancies,
high EBV-DNA copy number is correlated with tumour load
as some of the entities (e.g. ENKTCL and a subset of PTCL-NOS)
and is an adverse outcome predictor [24].
show EBER positivity in the neoplastic cells.
staging and risk assessment treatment

A complete blood count, routine blood chemistry including nodal PTCL (PTCL-NOS, AITL, ALCL ALK+,
lactate dehydrogenase (LDH), and uric acid as well as screening ALCL ALK−)
tests for HIV, HTLV-1, and hepatitis B and C are required. At first-line treatment. Treatment strategies should be adapted
baseline, patients should have at least a computed tomography according to factors such as age, IPI, and co-morbidity that
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv201 | v

by guest
on 05 February 2018
define a patient’s eligibility for dose-intensified approaches. treatment may also be useful to bridge eligible patients towards
Whenever possible, inclusion in a clinical trial is recommended. allogeneic stem-cell transplantation (alloSCT). A proposed
A treatment algorithm for newly diagnosed PTCL is shown in treatment algorithm is summarised in Figure 1B. For relapsed/
Figure 1A. Cyclophosphamide, hydroxydaunorubicin, vincristine refractory nodal PTCL other than ALCL, inclusion into clinical
and prednisone (CHOP), or variants of it, has been the most trials is highly encouraged. Outside clinical trials, in fit patients,
commonly used regimen in nodal PTCL. In patients less than 60 combination chemotherapy regimens such as DHAP
years of age with ALCL ALK+ histology, CHOP with the addition (dexamethasone, high-dose cytarabine, cisplatin) or ICE (ifosphamide,
of etoposide (CHOEP) has shown some outcome benefits in terms etoposide, carboplatin) can be attempted in chemosensitive
of event-free but not overall survival (OS). CHOEP was mostly patients with an available donor, aiming at alloSCT as a
feasible in younger patients (≤60 years), toxicity being a limiting potentially curative modality. In unfit patients, monotherapy with
factor in older patients [25]. In a large cohort of treatment-naïve gemcitabine or bendamustine are generally well-tolerated, with
PTCL patients, a schedule of 6 courses of bi-weekly CHOEP an ORR of approximately 50% but with modest durations of
followed by autologous stem-cell transplantation (autoSCT) response [30, 31]. Promising new drugs are under current
demonstrated an overall response rate (ORR) of 82%, with 51% evaluation in clinical trials. Of these new compounds, the anti-
achieving a complete response (CR) [23]. At a median follow-up folate pralatrexate and the histone deacetylase inhibitors
of 4.5 years, the three included nodal PTCL entities had an romidepsin and belinostat, have recently been conditionally
estimated 5-year OS and a progression-free survival (PFS) of approved in the US, based on phase II trial results [33, 34]. The
70% and 61% (ALCL ALK−), 52% and 49% (AITL) and 47% same is the case for the anti-CCR4 antibody mogamulizumab,
and 38% (PTCL-NOS). Recent population-based data also whose label in Japan has recently been extended from adult T-cell
indicate that upfront autoSCT in chemosensitive patients is leukaemia/lymphoma to cover also relapsed/refractory PTCL and
associated with improved OS [5]. On the basis of these data, a transformed mycosis fungoides [35]. Phase III studies are
dose-dense CHOEP schedule followed by autoSCT in ongoing in the upfront setting for all of these new compounds.
chemosensitive and transplant-eligible patients represents an
evidence-based approach adoptable outside of a clinical trial [III, EATL
B] (Figure 1A).
first-line treatment. In EATL, outcome after standard CHOP
Other induction regimens have been tried, e.g. platinum and
chemotherapy is generally poor. Recent reports indicate that, for
gemcitabine combinations. In newly diagnosed patients, a recent
patients sufficiently fit to tolerate more aggressive chemotherapy
phase II trial testing the PEGS regimen (cisplatin, etoposide, gem-
regimens, outcome can be significantly improved. A regimen
citabine, and methylprednisolone) revealed a disappointing ORR
with ifosphamide, vincristine, etoposide, and methotrexate (IVE/
of 39% and a 2-year PFS of only 14% [26]. Therefore, although
MTX) followed by autoSCT has shown promising results, with
the role of anthracyclines in PTCL is still debated, anthracycline-
5-year OS and PFS of 60% and 52%, respectively [36]. In
void regimens have, so far, failed to demonstrate their superiority
addition, CHOEP-14 consolidated with autoSCT has shown
to CHOP/CHOEP as the standard chemotherapy regimen
improved outcomes compared with standard CHOP [III, B] [23].
outside clinical trials. In low-risk (low-/low-intermediate IPI)
In a European Society for Blood and Marrow Transplantation-
ALCL ALK+ patients, consolidation with autoSCT is not recom-
based registry study, 4-year OS and PFS for EATL patients
mended, since these patients seem to have a more favourable
receiving intensive induction regimens followed by autoSCT in first
outcome compared with other PTCL subtypes, with a 5-year
CR/partial response (PR) were 59% and 54%, respectively [37]. It is
failure-free survival (FFS) of 60%–80% [6]. The few patients with
difficult to estimate the proportion of all EATL patients amenable
truly localised (stage I) disease should receive a shortened chemo-
to intensive therapies, but at least one-half of the patients aged <70
therapy schedule (e.g. 3 courses), followed by local radiotherapy,
years may be considered for standard-dose chemotherapy. If these
since retrospective analyses seem to indicate a survival advantage
patients respond to therapy, their performance status may improve,
for a combined modality approach in early stage disease [27–29]
thus allowing for subsequent autoSCT.
(see ‘Radiotherapy’ section). Frail patients not eligible for inten-
sive chemotherapy schedules may be considered for less toxic
relapse. No evidence-based specific relapse regimen can be
approaches such as monotherapy schedules, e.g. with gemcitabine
recommended in relapsed/refractory EATL. Therefore,
[30] or bendamustine [31].
considerations similar to those described for ‘nodal entities’ (see
above) are also applicable to relapsed/refractory EATL. In
relapse. Although a fair number of patients with nodal PTCL are
transplant-eligible patients who retain chemosensitivity at relapse
chemosensitive, their response duration is often short and relapses
and have a suitable donor, alloSCT should be attempted.
are frequent. Except for CD30+ ALCL, there is no standard of care
for relapsed/refractory nodal PTCL. The only globally approved
salvage treatment in PTCL is the anti-CD30 antibody conjugate ENKTCL
brentuximab vedotin (BV) administered in the setting of relapsed first-line treatment. The treatment strategy for this subtype of
systemic ALCL (regardless of the ALK status). In a pivotal phase II lymphoma is unique in the context of PTCL. L-asparaginase-
study, BV monotherapy in heavily pre-treated, noncutaneous containing regimens such as SMILE (dexamethasone, methotrexate,
ALCL patients yielded an ORR of 86% and a CR rate of 57%, ifosphamide, L-asparaginase, etoposide) and AspaMetDex
with a median response duration of 12.6 months [32]. Anti- (L-asparaginase, methotrexate, dexamethasone) have produced
CD30-directed BV monotherapy in relapsed/refractory ALCL is, promising results [38, 39]. Anthracycline-based regimens (CHOP
therefore, evidence-supported and recommended [III, A]. This or CHOP-like) are not effective [40]. Addition of radiation to
v | d’Amore et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
A B
Annals of Oncology

doi:10.1093/annonc/mdv201 | v
Figure 1. Integrated management algorithm (according to, e.g. risk factors, stage and histological subtype) in the (A) front-line and (B) relapsed/refractory setting. (A) § Stage I: shortened chemotherapy schedule
(e.g. 3 courses) followed by curatively intended RT (see ‘Radiotherapy’ section). *ALCL ALK+ with a high-risk profile (e.g. IPI >2) should be considered for autoSCT consolidation, while autoSCT in low risk profile
patients is not recommended. # if donor available. ¤ SMILE or AspaMetDex. (B) $: Pralatrexate and romidepsin: FDA but not EMA approved. PTCL, peripheral T-cell lymphomas; PTCL-NOS, PTCL-not otherwise
specified; AITL, angioimmunoblastic T-cell lymphoma; ALCL ALK+, anaplastic large-cell lymphoma anaplastic lymphoma kinase positive; ALCL ALK-, ALCL anaplastic lymphoma kinase negative;
EATL, enteropathy-associated T-cell lymphoma; HSTCL, hepatosplenic T-cell lymphoma; ENKTCL, extranodal natural killer/T-cell lymphoma; CHOEP, cyclophosphamide, hydroxydaunorubicin, vincristine,
etoposide, prednisone; CHOP, cyclophosphamide, hydroxydaunorubicin, vincristine, prednisone; IVE/MTX, ifosfamide, vincristine, etoposide/methotrexate; ICE, ifosphamide, etoposide, and carboplatin; IVAC,
ifosphamide, cytarabine, etoposide; PR, partial response; CR, complete response; alloSCT, allogeneic stem-cell transplantation; autoSCT, autologous stem-cell transplantation; rel/ref, relapsed/refractory;
BV, brentuximab vedotin; DHAP, dexamethasone, high-dose cytarabine, cisplatin; SMILE, dexamethasone, methotrexate, ifosphamide, L-asparaginase, etoposide; CS, clinical stage; RT, radiotherapy.

by guest
on 05 February 2018
chemotherapy is the preferred treatment of localised disease [40]. intense regimens such as ICE, IVAC (ifosphamide, cytarabine,
EBV DNA copy number from plasma or whole blood can be etoposide), or dose-dense CHOEP/EPOCH (etoposide, vincristine,
used as a biomarker for response; therefore, serial monitoring of doxorubicin, cyclophosphamide and prednisone) have been
EBV DNA copy number is recommended [15]. ENKTCL is proposed and auto- or alloSCT consolidation in fit patients is
FDG-avid, and although the role of PET/CT for response recommended [IV, B] [23, 44].
evaluation is not yet fully clarified, PET/CT is the recommended
imaging modality in ENKTCL.
relapse. No evidence-based specific relapse regimen can be
recommended in relapsed/refractory HSTCL. If chemosensitivity
stages I–II. Most patients present with stage I–II nasal disease. is achieved by current relapse regimens, alloSCT should be
In these cases, radiation combined with chemotherapy is the attempted in eligible patients, since a graft-versus-host effect has
preferred treatment [40, 41]. Concurrent chemoradiotherapy also been described in relapsed/refractory disease.
and sequential chemotherapy with L-asparaginase-containing
regimens followed by radiation appear to have comparable efficacy
[III, A]. A radiation dose >50 Gy is recommended when treating
radiotherapy
with radiation alone. However, with radiosensitisers such as
cisplatin, a weekly dose of ∼40 Gy can give a comparable outcome. PTCL seem to be somewhat less radiosensitive than the aggres-
Central nervous system prophylaxis is not recommended, although sive B-cell lymphomas [45], and higher radiation doses may be
the disease involves nasal and/or paranasal areas. Instances of needed, although still lower than for most solid tumours.
localised disease outside the nasal region are rare. If feasible,
radiation with or without chemotherapy seems to be a more
effective treatment compared with chemotherapy alone. The role of front-line treatment
high-dose chemotherapy followed by haematopoietic stem-cell Most types of PTCLs usually present with advanced disease.
transplantation (HSCT) is still controversial. For elderly and/or frail The few patients with localised disease may be treated with local
patients, radiation alone is recommended. radiotherapy after chemotherapy [27–29], although no rando-
mised evidence regarding this approach exists. Recommended
stages III–IV. L-asparaginase-containing chemotherapy regi- doses are 30–40 Gy [46]. Because of the somewhat lower radio-
mens should be preferred as front-line treatment [III, A]. If complete sensitivity of PTCLs, doses of 40 Gy should be preferred, in par-
remission is achieved, high-dose chemotherapy with HSCT is recom- ticular if residual lymphoma is present after chemotherapy. The
mended. AutoSCT is preferable due to higher treatment-related mor- treated volume should include only the initially involved volume
tality after alloSCT. For elderly and frail patients, L-asparaginase with appropriate margins for uncertainty, according to the prin-
single agent or mild chemotherapy regimens (AspaMetDex or dose- ciples of involved site radiotherapy (ISRT) [47]. High-quality
modified SMILE) can be recommended [42]. imaging before chemotherapy should be obtained in order to
allow for optimal planning of subsequent radiotherapy. Modern
advanced radiation treatment techniques should be used to min-
relapse. A repeated pre-therapeutic biopsy is strongly recom-
imise long-term toxicity [47].
mended, since some of the PET-positive lesions can represent
ENKTCLs have special features. They most commonly
inflammatory changes secondary to ulceration. Selection of a
involve the nasal cavity, paranasal sinuses, and Waldeyer’s ring.
salvage regimen depends on the type of primary treatment and
Radiotherapy is an important part of the treatment and should
response duration. In early relapse (<12 months) after anthracy-
be administered early [III, A] [15, 48]. Patients with stage I
cline-based prior treatment, L-asparaginase-containing regimens
disease may be treated with radiotherapy alone to a dose of
should be recommended. For patients who received L-asparagi-
≥50 Gy [49]. Alternatively, and for patients with risk factors or
nase regimens upfront, a gemcitabine-based (e.g. GELOX, gem-
stage II disease, concomitant chemoradiotherapy with a plat-
citabine, L-asparaginase, oxaliplatin) regimen can be considered
inum-containing regimen and radiotherapy dose of ≥50 Gy is
for salvage treatment [41]. Although transplant-specific data are
an option [41]. With more effective chemotherapy regimens, se-
very limited, either auto- or alloSCT should be considered in
quential chemoradiation with radiation doses of 45–50 Gy may
transplant-eligible patients. Both modalities should preferably
be used. ENKTCL is often locally destructive and may infiltrate
be tested in clinical trials.
extensively in the submucosa of the upper aerodigestive tract.
A generous ISRT volume is recommended, covering the entire
HSTCL organ(s) involved with lymphoma before chemotherapy plus ad-
first-line treatment. HSTCL has one of the worst prognoses jacent structures with concern for subclinical disease. Advanced
among PTCLs, with 5-year FFS and OS rates of less than 10%. All imaging including PET/CT and magnetic resonance imaging and
cases should be treated with chemotherapy at diagnosis. Although conformal radiotherapy techniques should be used.
most patients have only brief responses to anthracycline-based
therapy, limited evidence suggests that they may respond to a
platinum/cytarabine-based induction regimen. In the case of relapse/refractory disease
chemosensitivity to induction therapy, upfront consolidation with Palliative radiotherapy may be used to treat locally symptomatic
auto- or alloSCT should be offered to all eligible patients, since it disease. Usual palliative doses of 30 Gy in 10 fractions may
may offer the only chance for durable remission [8, 43]. Recently, be used.
v | d’Amore et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
autologous stem-cell transplantation The literature remains controversial with regard to the long-term
(autoSCT) outcomes with autoSCT in this setting [60].
front-line treatment personalised medicine

There are no randomised prospective studies to guide clinicians Two personalised approaches are currently widely accepted
in the decision on whether to perform autoSCT consolidation in and formally approved in the treatment of PTCL. One is the use
first remission versus expectant observation. A body of proof L-asparaginase in the treatment of ENKTCL and the other
spective literature has accumulated that evaluates an up-front is the use of the anti-CD30 antibody conjugate BV for the treat-
autoSCT in PTCL [23, 50–52]. Recent population-based data ment of relapsed/refractory ALCL also with the purpose of
also indicate that upfront autoSCT in chemosensitive patients is bridging eligible patients to alloSCT. BV is also currently being
associated with improved OS [5]. The nodal entities (PTCL- tested as part of the upfront treatment in CD30-positive PTCL
NOS, AITL, and ALCL mostly restricted to ALK− cases) re- other than ALCL. Increasing knowledge of the molecular
present the majority of patients enrolled in clinical trials. ALCL mechanisms involved in PTCL pathogenesis is accumulating.
ALK+ patients usually have been excluded from upfront This may lead to new and more targeted treatment approaches
autoSCT trials, given their superior outcome following CHOP in the near future. Currently, no evidence-based personalised
or CHOP-like regimens. The largest prospective study evaluat- medicine approaches are available for patients with PTCL-NOS,
ing autoSCT consolidation in de novo PTCL was carried out by AITL, EATL, and HSTCL. In these settings, more research is
the Nordic group, where patients achieving CR or PR after a needed to identify molecular markers which could lead to
CHOEP-based dose-dense induction schedule received BEAM advances in personalised medicine.
(carmustine, etoposide, cytarabine, melphalan) conditioning
and autoSCT. With the caveat of phase II data, autoSCT in first
remission seems quite feasible and possibly beneficial in the response evaluation and follow-up
selected subset of transplant-eligible PTCL [III, B]. In systemic PTCL, a midway interim evaluation should be
carried out in order to assess chemosensitivity. Increasing evi-
dence points at PTCL as consistently FDG-avid tumours [61, 62]
allogeneic stem-cell transplantation providing the rationale for the use of PET/CT, particularly in the
(alloSCT) context of residual disease evaluation. Diagnostic imaging (CT or
PET/CT) should be repeated at the end of treatment along with a
AlloSCT is a potentially curative option for patients affected by bone marrow biopsy (only if initially involved). Presently, no evi-
PTCL. The first prospective phase II results demonstrated sus- dence-based recommendation is possible with regard to the
tained responses in relapsed/refractory PTCL patients, suggest- length of follow-up. However, in the Nordic NLG-T-01 trial,
ing the existence of a possible ‘graft-versus-T-cell lymphoma’ where a cohort of 160 evaluable systemic PTCL patients were fol-
effect [III, B] [53, 54]. Non-relapse mortality (NRM) was low, lowed over a median period of 5 years (range: 2–8 years), 7% of
supporting the feasibility of a reduced-intensity conditioning all relapses occurred after 2 years [23]. On this basis, a follow-up
alloSCT (RIC-alloSCT) even in heavily pre-treated patients. schedule consisting of history and physical examination every 3
Retrospective and registry-based analyses also confirmed that months for 1 year, every 6 months for 2 more years, and then
alloSCT can yield long-term responses in relapsed/refractory once a year for detection of secondary tumours or other long-
PTCL [55–58]. In extranodal subtypes data are anecdotal, but term side-effects [V, C]. CT examinations at 6, 12, and 24
generally supportive of the feasibility and efficacy of alloSCT. In months after the end of treatment are usual practice, but there is
a recently reported prospective trial carried out in treatment- no definitive evidence that routine imaging in patients in com-
naïve PTCL patients, after an induction phase with intensive plete remission provides any outcome advantage [V, C]. Routine
chemoimmunotherapy, responding patients were randomised to surveillance with PET scan is not recommended.
auto- or alloSCT based on the availability of a HLA identical
sibling or a matched unrelated donor [59]. The sample size did
not allow identification of a preferred approach among the two; methodology
however, allo- and autografted patients had a 4-year PFS of 69% These clinical practice guidelines were developed in accordance
and 70%, respectively. In conclusion, alloSCT is a valid treatment with the ESMO standard operating procedures for clinical prac-
option in transplant-eligible relapsed PTCL patients, also after a tice guidelines development. The relevant literature has been
failed prior autograft. The benefit is most evident in chemosensi- selected by the expert authors. Levels of evidence and grades of
tive patients. A RIC-alloSCT should be preferred to a myeloabla- recommendation have been applied using the system shown in
tive approach in order to reduce NRM. In the upfront setting, Table 2. Statements without grading were considered justified
alloSCT should be carried out primarily within clinical trials. standard clinical practice by the experts and the ESMO faculty.
review process.
relapsed/refractory disease
For patients with relapsed/refractory PTCLs, potential curative
options include consolidation with either an autoSCT or alloSCT.
Although often appropriately offered, not all patients are able to FA has reported Scientific advisory boards and speaker’s honor-
have a transplant due to highly refractory and progressive disease. aria: Takeda, Norpharma, Kyowa Kirin, CTI Life Sciences,
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv201 | v

by guest
on 05 February 2018
Levels of evidence
a
Seattle Genetics, Infinity. Research support: Sanofi, Amgen. PG 9. Gaulard P, de Leval L. Pathology of peripheral T-cell lymphomas: where do we
has reported Scientific advisory boards and speaker’s honoraria: stand? Semin Hematol 2014; 51: 5–16.
Takeda. LT has reported Research grants from Sanofi, Amgen, 10. Asano N, Suzuki R, Kagami Y et al. Clinicopathologic and prognostic significance
of cytotoxic molecule expression in nodal peripheral T-cell lymphoma, unspecified.
Roche and Mundipharma. WSK: Research supported by grants
Am J Surg Pathol 2005; 29: 1284–1293.
from Takeda, Novartis, Celgene, Roche. LS has reported
11. de Leval L, Rickman DS, Thielen C et al. The gene expression profile of nodal
Scientific advisory board and speaker’s honoraria: Takeda. PC peripheral T-cell lymphoma demonstrates a molecular link between
has reported Scientific advisory boards and speaker’s honoraria: angioimmunoblastic T-cell lymphoma (AITL) and follicular helper T (TFH) cells.
Takeda, Celgene, Roche, Novartis, Sanofi, Gilead, Janssen. ML Blood 2007; 109: 4952–4963.
has reported honoraria from Celgene, Janssen-Cilag, Roche, 12. Iqbal J, Weisenburger DD, Greiner TC et al. Molecular signatures to improve
Amgen, Mundipharma and Teva; research contracts from diagnosis in peripheral T-cell lymphoma and prognostication in angioimmunoblastic
Celgene, Pfizer, Mundipharma and Roche; funds received from T-cell lymphoma. Blood 2010; 115: 1026–1036.
Amgen, Roche and Takeda. MBP has reported no conflicts of 13. de Leval L, Gaulard P. Pathology and biology of peripheral T-cell lymphomas.
Histopathology 2011; 58: 49–68.
interests.
14. Chan WK, Au WY, Wong CY et al. Metabolic activity measured by F-18 FDG PET in
natural killer-cell lymphoma compared to aggressive B- and T-cell lymphomas.
references Clin Nucl Med 2010; 35: 571–575.
1. Willemze R, Hodak E, Zinzani PL et al. Primary cutaneous lymphomas: ESMO 15. Kwong YL, Anderson BO, Advani R et al. Management of T-cell and natural-killer-
Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol cell neoplasms in Asia: consensus statement from the Asian Oncology Summit
2013; 24(Suppl 6): vi149–vi154. 2009. Lancet Oncol 2009; 10: 1093–1101.
2. Vose J, Armitage J, Weisenburger D. International peripheral T-cell and natural 16. Zhou X, Lu K, Geng L et al. Utility of PET/CT in the diagnosis and staging of
killer/T-cell lymphoma study: pathology findings and clinical outcomes. J Clin extranodal natural killer/T-cell lymphoma: a systematic review and meta-analysis.
Oncol 2008; 26: 4124–4130. Medicine (Baltimore) 2014; 93: e258.
3. Swerdlow S, Campo E, Harris NL et al. Eds. WHO Classification of Tumours of 17. Tse E, Kwong YL. How I treat NK/T-cell lymphomas. Blood 2013; 121: 4997–5005.
Haematopoietic and Lymphoid Tissue. Lyon: International Agency for Research on 18. Moon SH, Cho SK, Kim WS et al. The role of 18F-FDG PET/CT for initial staging of
Cancer, 2008. nasal type natural killer/T-cell lymphoma: a comparison with conventional staging
4. Pedersen MB, Hamilton-Dutoit SJ, Bendix K et al. Evaluation of clinical trial eligibility methods. J Nucl Med 2013; 54: 1039–1044.
and prognostic indices in a population-based cohort of systemic peripheral T-cell 19. A predictive model for aggressive non-Hodgkin’s lymphoma. The International
lymphomas from the Danish Lymphoma Registry. Hematol Oncol 2014 [Epub ahead Non-Hodgkin’s Lymphoma Prognostic Factors Project. N Engl J Med 1993; 329:
of print]. 987–994.
5. Ellin F, Landstrom J, Jerkeman M, Relander T. Real-world data on prognostic 20. Gallamini A, Stelitano C, Calvi R et al. Peripheral T-cell lymphoma unspecified
factors and treatment in peripheral T-cell lymphomas: a study from the Swedish (PTCL-U): a new prognostic model from a retrospective multicentric clinical study.
Lymphoma Registry. Blood 2014; 124: 1570–1577. Blood 2004; 103: 2474–2479.
6. Savage KJ, Harris NL, Vose JM et al. ALK- anaplastic large-cell lymphoma is 21. Went P, Agostinelli C, Gallamini A et al. Marker expression in peripheral T-cell
clinically and immunophenotypically different from both ALK+ ALCL and peripheral lymphoma: a proposed clinical-pathologic prognostic score. J Clin Oncol 2006;
T-cell lymphoma, not otherwise specified: report from the International Peripheral 24: 2472–2479.
T-Cell Lymphoma Project. Blood 2008; 111: 5496–5504. 22. Gutierrez-Garcia G, Garcia-Herrera A, Cardesa T et al. Comparison of
7. Sibon D, Fournier M, Briere J et al. Long-term outcome of adults with systemic four prognostic scores in peripheral T-cell lymphoma. Ann Oncol 2011; 22:
anaplastic large-cell lymphoma treated within the Groupe d’Etude des Lymphomes 397–404.
de l’Adulte trials. J Clin Oncol 2012; 30: 3939–3946. 23. d’Amore F, Relander T, Lauritzsen GF et al. Up-front autologous stem-cell
8. Ferreri AJ, Govi S, Pileri SA. Hepatosplenic gamma-delta T-cell lymphoma. Crit transplantation in peripheral T-cell lymphoma: NLG-T-01. J Clin Oncol 2012; 30:
Rev Oncol Hematol 2012; 83: 283–292. 3093–3099.
v | d’Amore et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
24. Suzuki R, Yamaguchi M, Izutsu K et al. Prospective measurement of Epstein-Barr results in improved outcome for patients with hepatosplenic T-cell lymphoma: a
virus-DNA in plasma and peripheral blood mononuclear cells of extranodal NK/T- single institution experience. Clin Lymphoma Myeloma Leuk 2013; 13: 8–14.
cell lymphoma, nasal type. Blood 2011; 118: 6018–6022. 45. Sakata K, Fuwa N, Kodaira T et al. Analyses of dose-response in radiotherapy for
25. Schmitz N, Trumper L, Ziepert M et al. Treatment and prognosis of mature T-cell patients with mature T/NK-cell lymphomas according to the WHO classification.
and NK-cell lymphoma: an analysis of patients with T-cell lymphoma treated in Radiother Oncol 2006; 79: 179–184.
studies of the German High-Grade Non-Hodgkin Lymphoma Study Group. Blood 46. Lowry L, Smith P, Qian W et al. Reduced dose radiotherapy for local control in non-
2010; 116: 3418–3425. Hodgkin lymphoma: a randomised phase III trial. Radiother Oncol 2011; 100:
26. Mahadevan D, Unger JM, Spier CM et al. Phase 2 trial of combined cisplatin, 86–92.
etoposide, gemcitabine, and methylprednisolone (PEGS) in peripheral T-cell non- 47. Illidge T, Specht L, Yahalom J et al. Modern radiation therapy for nodal non-Hodgkin
Hodgkin lymphoma: Southwest Oncology Group Study S0350. Cancer 2013; 119: lymphoma-target definition and dose guidelines from the International Lymphoma
371–379. Radiation Oncology Group. Int J Radiat Oncol Biol Phys 2014; 89: 49–58.
27. Weisenburger DD, Savage KJ, Harris NL et al. Peripheral T-cell lymphoma, not 48. Huang MJ, Jiang Y, Liu WP et al. Early or up-front radiotherapy improved survival
otherwise specified: a report of 340 cases from the International Peripheral T-cell of localized extranodal NK/T-cell lymphoma, nasal-type in the upper aerodigestive
Lymphoma Project. Blood 2011; 117: 3402–3408. tract. Int J Radiat Oncol Biol Phys 2008; 70: 166–174.
28. Zhang XM, Li YX, Wang WH et al. Survival advantage with the addition of radiation 49. Bi XW, Li YX, Fang H et al. High-dose and extended-field intensity modulated
therapy to chemotherapy in early stage peripheral T-cell lymphoma, not otherwise radiation therapy for early-stage NK/T-cell lymphoma of Waldeyer’s ring:
specified. Int J Radiat Oncol Biol Phys 2013; 85: 1051–1056. dosimetric analysis and clinical outcome. Int J Radiat Oncol Biol Phys 2013; 87:
29. Zhang XM, Li YX, Wang WH et al. Favorable outcome with doxorubicin-based 1086–1093.
chemotherapy and radiotherapy for adult patients with early stage primary systemic 50. Corradini P, Dodero A, Farina L et al. Allogeneic stem cell transplantation following
anaplastic large-cell lymphoma. Eur J Haematol 2013; 90: 195–201. reduced-intensity conditioning can induce durable clinical and molecular
30. Zinzani PL, Venturini F, Stefoni V et al. Gemcitabine as single agent in pretreated remissions in relapsed lymphomas: pre-transplant disease status and histotype
T-cell lymphoma patients: evaluation of the long-term outcome. Ann Oncol 2010; heavily influence outcome. Leukemia 2007; 21: 2316–2323.
21: 860–863. 51. Rodriguez J, Conde E, Gutierrez A et al. Frontline autologous stem cell
31. Damaj G, Gressin R, Bouabdallah K et al. Results from a prospective, open-label, transplantation in high-risk peripheral T-cell lymphoma: a prospective study from
phase II trial of bendamustine in refractory or relapsed T-cell lymphomas: the The Gel-Tamo Study Group. Eur J Haematol 2007; 79: 32–38.
BENTLY trial. J Clin Oncol 2013; 31: 104–110. 52. Reimer P, Rudiger T, Geissinger E et al. Autologous stem-cell transplantation as
32. Pro B, Advani R, Brice P et al. Brentuximab vedotin (SGN-35) in patients with first-line therapy in peripheral T-cell lymphomas: results of a prospective
relapsed or refractory systemic anaplastic large-cell lymphoma: results of a phase multicenter study. J Clin Oncol 2009; 27: 106–113.
II study. J Clin Oncol 2012; 30: 2190–2196. 53. Corradini P, Dodero A, Zallio F et al. Graft-versus-lymphoma effect in relapsed
33. O’Connor OA, Pro B, Pinter-Brown L et al. Pralatrexate in patients with relapsed or peripheral T-cell non-Hodgkin’s lymphomas after reduced-intensity conditioning
refractory peripheral T-cell lymphoma: results from the pivotal PROPEL study. J followed by allogeneic transplantation of hematopoietic cells. J Clin Oncol 2004;
Clin Oncol 2011; 29: 1182–1189. 22: 2172–2176.
34. Piekarz RL, Frye R, Prince HM et al. Phase 2 trial of romidepsin in patients with 54. Dodero A, Spina F, Narni F et al. Allogeneic transplantation following a reduced-
peripheral T-cell lymphoma. Blood 2011; 117: 5827–5834. intensity conditioning regimen in relapsed/refractory peripheral T-cell lymphomas:
35. Ogura M, Ishida T, Hatake K et al. Multicenter phase II study of mogamulizumab long-term remissions and response to donor lymphocyte infusions support the role
(KW-0761), a defucosylated anti-cc chemokine receptor 4 antibody, in patients of a graft-versus-lymphoma effect. Leukemia 2012; 26: 520–526.
with relapsed peripheral T-cell lymphoma and cutaneous T-cell lymphoma. J Clin 55. Kim SW, Tanimoto TE, Hirabayashi N et al. Myeloablative allogeneic hematopoietic
Oncol 2014; 32: 1157–1163. stem cell transplantation for non-Hodgkin lymphoma: a nationwide survey in
36. Sieniawski M, Angamuthu N, Boyd K et al. Evaluation of enteropathy-associated T- Japan. Blood 2006; 108: 382–389.
cell lymphoma comparing standard therapies with a novel regimen including 56. Le Gouill S, Milpied N, Buzyn A et al. Graft-versus-lymphoma effect for aggressive
autologous stem cell transplantation. Blood 2010; 115: 3664–3670. T-cell lymphomas in adults: a study by the Societe Francaise de Greffe de Moelle
37. Jantunen E, Boumendil A, Finel H et al. Autologous stem cell transplantation for et de Therapie Cellulaire. J Clin Oncol 2008; 26: 2264–2271.
enteropathy-associated T-cell lymphoma: a retrospective study by the EBMT. 57. Smith SM, Burns LJ, van Besien K et al. Hematopoietic cell transplantation for
Blood 2013; 121: 2529–2532. systemic mature T-Cell non-Hodgkin lymphoma. J Clin Oncol 2013; 31: 3100–3109.
38. Kwong YL, Kim WS, Lim ST et al. SMILE for natural killer/T-cell lymphoma: 58. Kyriakou C, Canals C, Goldstone A et al. High-dose therapy and autologous stem-
analysis of safety and efficacy from the Asia Lymphoma Study Group. Blood 2012; cell transplantation in angioimmunoblastic lymphoma: complete remission at
120: 2973–2980. transplantation is the major determinant of outcome – Lymphoma Working Party of
39. Jaccard A, Gachard N, Marin B et al. Efficacy of L-asparaginase with the European Group for Blood and Marrow Transplantation. J Clin Oncol 2008; 26:
methotrexate and dexamethasone (AspaMetDex regimen) in patients with 218–224.
refractory or relapsing extranodal NK/T-cell lymphoma, a phase 2 study. Blood 59. Corradini P, Vitolo U, Rambaldi A et al. Intensified chemo-immunotherapy with or
2011; 117: 1834–1839. without stem cell transplantation in newly diagnosed patients with peripheral T-cell
40. Kim WS, Song SY, Ahn YC et al. CHOP followed by involved field radiation: is it lymphoma. Leukemia 2014; 28: 1885–1891.
optimal for localized nasal natural killer/T-cell lymphoma? Ann Oncol 2001; 12: 60. Zinzani PL. High-dose therapy and stem cell transplantation. Semin Hematol
349–352. 2010; 47(Suppl 1): S15–S17.
41. Yamaguchi M, Tobinai K, Oguchi M et al. Concurrent chemoradiotherapy 61. Tsukamoto N, Kojima M, Hasegawa M et al. The usefulness of (18)F-
for localized nasal natural killer/T-cell lymphoma: an updated analysis of the Japan fluorodeoxyglucose positron emission tomography ((18)F-FDG-PET) and a
clinical oncology group study JCOG0211. J Clin Oncol 2012; 30: 4044–4046. comparison of (18)F-FDG-pet with (67)gallium scintigraphy in the evaluation of
42. Yang L, Liu H, Xu XH et al. Retrospective study of modified SMILE chemotherapy lymphoma: relation to histologic subtypes based on the World Health Organization
for advanced-stage, relapsed, or refractory extranodal natural killer (NK)/T cell classification. Cancer 2007; 110: 652–659.
lymphoma, nasal type. Med Oncol 2013; 30: 720. 62. Khong PL, Pang CB, Liang R et al. Fluorine-18 fluorodeoxyglucose positron
43. Tanase A, Schmitz N, Stein H et al. Allogeneic and autologous stem cell emission tomography in mature T-cell and natural killer cell malignancies. Ann
transplantation for hepatosplenic T-cell lymphoma: a retrospective study of the Hematol 2008; 87: 613–621.
EBMT Lymphoma Working Party. Leukemia 2015; 29: 686–688. 63. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
44. Voss MH, Lunning MA, Maragulia JC et al. Intensive induction chemotherapy among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
followed by early high-dose therapy and hematopoietic stem cell transplantation 139–144.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv201 | v

by guest
on 05 February 2018
Philadelphia chromosome-negative chronic

myeloproliferative neoplasms: ESMO Clinical Practice
A. M. Vannucchi1, T. Barbui2, F. Cervantes3, C. Harrison4, J.-J. Kiladjian5, N. Kröger6, J. Thiele7
& C. Buske8, on behalf of the ESMO Guidelines Committee*
1
Department of Experimental and Clinical Medicine, University of Florence, Florence; 2Hematology and Foundation for Research, Ospedale Papa Giovanni XXIII, Bergamo,
Italy; 3Department of Hematology, Hospital Clínic, IDIBAPS, University of Barcelona, Barcelona, Spain; 4Department of Haematology, Guy’s and St Thomas’ NHS
Foundation Trust, London, UK; 5Centre d’Investigations Cliniques (INSERM CIC1427), Hôpital Saint-Louis and Paris Diderot University, Paris, France; 6Department of Stem
Cell Transplantation, University Hospital Hamburg, Hamburg; 7University of Cologne, Cologne; 8Comprehensive Cancer Center Ulm, Institute of Experimental Cancer
Research, University Hospital Ulm, Ulm, Germany
incidence In PV, classical BM features are a moderate to overt increase

in age-matched cellularity, due to a trilineage proliferation
According to the 2008 World Health Organization (WHO) clas- ( panmyelosis) of erythroid and granulocytic precursors and
sification, classical Philadelphia chromosome/BCR-ABL nega- megakaryocytes in variable proportions. Megakaryocytes are
tive chronic myeloproliferative neoplasms (MPNs) include
clinical practice
characterised by a pleomorphic appearance due to the variability
polycythaemia vera (PV), essential thrombocythaemia (ET) and
guidelines
in sizes, from small to giant cells, without gross abnormalities of
primary myelofibrosis (PMF) [1]. The reported worldwide maturation. There may be minimal (grade 1) reticulin fibrosis,
annual incidence rate of MPNs ranges from 0.44 to 5.87/105, which is very rarely grade 2.
with the lowest incidence being reported in Japan and Israel [2]. In ET, age-adjusted cellularity is normal or sometimes slightly
These great ranges may reflect racial/geographic differences as increased; there is no left-shifted neutrophil granulopoiesis. Any
well as differences in study design, diagnostic criteria and case with a mild to moderate panmyelosis is suspicious for early
methods of reporting, among others. The estimated incidence PV rather than ET. Megakaryocytes have increased in number
rate in Europe is 0.4–2.8 × 105/year for PV, 0.38–1.7 × 105/year and are randomly distributed within the BM, with scattered
for ET and 0.1–1 × 105/year for PMF [2]. There are few reliable forms or a few loose clusters. Large to giant mature megakaryo-
estimates of the prevalence [2, 3]. The last is likely to be rising cytes with extensively folded (staghorn-like) nuclei and mature
due to earlier diagnosis and trends towards prolonged survival cytoplasm are in the majority. Gross disturbances of the histo-
[4]. The reported median age at diagnosis ranges from 65–74 logic topography or extensive dense clustering of megakaryo-
years for PV, 64–73 years for ET, and 69–76 years for PMF [2]. cytes should not be detectable. There is no substantial increase
of reticulin fibres. The WHO classification suggests that these
diagnosis and pathology/molecular features clearly distinguish ET from pre-fibrotic/early PMF;
biology however, minor diagnostic criteria should also be present in
order to assign this diagnosis (Table 1).
To achieve the most accurate diagnosis possible, the 2008 WHO In the initial phases of PMF, the BM is often hypercellular
classification is recommended. It is based upon standardised with prominent granulocytic and megakaryocytic proliferation,
morphological features, ideally using specimens obtained before frequently with a reduction of erythroid precursors. If reticulin
treatment, and is integrated with haematological, molecular and fibrosis is present, grade 1 is allocated. Megakaryopoiesis is
clinical diagnostic criteria (Table 1). For full details and images characterised by the extensive formation of loose to dense clus-
of classical morphological features, the reader is referred to the ters of megakaryocytes, with abnormal localisation toward the
WHO publication [1]. European consensus-based criteria for endosteal borders. Megakaryocyte anomalies include a high
grading of cellularity and bone marrow (BM) fibrosis should be degree of cellular atypia (from small to giant forms), abnormal
followed (Table 2) [6]. nuclear folding and an aberrant nuclear cytoplasmic ratio
created by large, bulbous and hyperchromatic cloud-shaped
*Correspondence to: ESMO Guidelines Committee, ESMO Head Office, Via L. Taddei 4, nuclei. Naked (bare) megakaryocytic nuclei are often visible.
CH-6962 Viganello-Lugano, Switzerland. Overall, the megakaryocytes in PMF show a more pronounced
degree of atypia than in other MPN subtypes. The more advanced
†
Approved by the ESMO Guidelines Committee: July 2015. fibro-osteosclerotic phases of PMF are characterised by grade ≥2

by guest
on 05 February 2018
Table 1. The WHO diagnostic criteria for Philadelphia chromosome-negative chronic myeloproliferative neoplasms [1]
Polycythaemia vera (PV) Essential thrombocythaemia (ET) Primary myelofibrosis (PMF)
Major criteria 1. Haemoglobin 1. Platelet count ≥450 × 10 /l

9
1. Megakaryocyte proliferation and atypiab
>18.5 g/dl (men) 2. Megakaryocyte proliferation with large accompanied by either reticulin and/or collagen
>16.5 g/dl (women) and mature morphology fibrosis, or c
or a any other evidence of increased
red cell volume
2. Presence of JAK2V617F or JAK2 3. Not meeting WHO criteria for CML, 2. Not meeting WHO criteria for CML, PV, MDS
exon 12 mutation PV, PMF, MDS or other myeloid or other myeloid neoplasm
neoplasm
4. Demonstration of JAK2V617F or 3. Demonstration of JAK2V617F or
other clonal marker or other clonal marker or
no evidence of reactive thrombocytosis no evidence of reactive BM fibrosis
Minor criteria 1. BM trilineage myeloproliferation 1. Leukoerythroblastosis

2. Sub-normal sEPO level 2. Increased serum LDH level
3. Endogenous erythroid colony 3. Anaemia
growth 4. Palpable splenomegaly
a
Hb or HCT >99th percentile of reference range for age, sex or altitude of residence or red cell mass >25% above mean normal predicted or Hb >17 g/dl
(men)/>15 g/dl (women) if associated with a sustained increase of ≥2 g/dl from baseline that cannot be attributed to correction of iron deficiency.
b
Small to large megakaryocytes with aberrant nuclear/cytoplasmic ratio and hyperchromatic and irregularly folded nuclei and dense clustering.
c
In the absence of reticulin fibrosis, the megakaryocyte changes must be accompanied by increased marrow cellularity, granulocytic proliferation and often
decreased erythropoiesis (i.e. pre-fibrotic PMF).
PV diagnosis requires meeting either both major criteria and one minor criterion or the first major criterion and two minor criteria. ET diagnosis requires
meeting all four major criteria. PMF diagnosis requires meeting all three major criteria and two minor criteria.
Note: mutations in calreticulin (CALR) will be included as major diagnostic criteria for ET and PMF in the upcoming (2015) revised WHO classification [5].
BM, bone marrow; WHO, World Health Organization; CML, chronic myelogenous leukaemia; MDS, myelodysplastic syndromes; LDH, lactate
dehydrogenase; sEPO, serum erythropoietin; Hb, haemoglobin; HCT, haematocrit.
Table 2. European consensus on the grading of myelofibrosis (MF) [6] Clusters of megakaryocytes (often with hyperchromatic and
abnormal nuclei) are prominent, while erythropoiesis and
MF—0 Scattered linear reticulin with no intersection (cross-overs) granulopoiesis are decreased. Osteosclerosis may occur. However,
corresponding to normal bone marrow these findings must be integrated with other features to achieve
MF—1 Loose network of reticulin with many intersections, especially in a diagnosis (Table 3).
perivascular areas Although a number of clinicopathological studies have
MF—2 Diffuse and dense increase in reticulin with extensive demonstrated that a reliable morphological differentiation can
intersections, occasionally with only focal bundles of collagen and/or be achieved with high consensus rates [8–10], both the reprodu-
focal osteosclerosis cibility and the clinical usefulness of the WHO classification of
MF—3 Diffuse and dense increase in reticulin with extensive MPNs remain controversial issues, especially concerning the
intersections with coarse bundles of collagen, often associated with distinction between ET and pre-fibrotic/early myelofibrosis
significant osteosclerosis (MF) as well as initial cases of PV from ET or even PMF [11,
12]. To avoid incorrect classification, comprehensive evaluation
Fibre density should be assessed in haematopoietic (cellular) areas. and, if necessary, re-evaluation of patients are suggested. This
Republished with permission. Obtained from the Haematologica Journal also ensures that the patient is not unnecessarily diagnosed as
website http://www.haematologica.org. ‘MPN-unclassified’.
MPNs are characterised by somatic recurrent mutations and
are included as the main criteria in the 2008 WHO classification
reticulin deposition, and the appearance of coarse bundles of col- (Table 1), for which a further revision is expected in the future
lagen fibres. Additional features include endophytic bone forma- due to the newly discovered calreticulin (CALR) mutations [5].
tion (osteosclerosis) associated with extension of adipose tissue. These mutations include the Janus kinase (JAK) 2V617F muta-
Dilated marrow sinuses with intraluminal haematopoiesis, espe- tion, found in ≥95% of PV and ∼60% of ET and PMF patients.
cially made up of megakaryocytes, are often seen. 3%–5% of ET and 5%–8% of PMF patients have point mutations
The morphological hallmark of post-polycythaemia vera mye- at codon 515 of the gene encoding the thrombopoietin receptor
lofibrosis (PPV-MF) and post-essential thrombocythaemia MPL (W>L, K or A).
myelofibrosis (PET-MF) is overt reticulin and collagen fibrosis Abnormalities (deletions/duplications/substitutions) located
of the BM. Cellularity varies, but hypocellularity is common. in exon 12 of JAK2 are detected exclusively in 2%–4% of PV.
v | Vannucchi et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Table 3. Diagnostic criteria for post-polycythaemia vera myelofibrosis (PPV-MF) and post-essential thrombocythaemia myelofibrosis (PET-MF)
according to the International Working Group for Myeloproliferative Neoplasm Research and Treatment (IWG-MRT) [7]
PPV-MF PET-MF
Required criteria (both required):
1. Documentation of a previous diagnosis of PV as defined by the 2008 1. Documentation of a previous diagnosis of ET as defined by the 2008
WHO criteria WHO criteria
2. Bone marrow fibrosis grade 2–3 (on a 0–3 scale) or grade 3–4 (on a 0–4 2. Bone marrow fibrosis grade 2–3 (on a 0–3 scale) or grade 3–4 (on a
scale) 0–4 scale)
Additional criteria (≥2 required):
Anaemia or sustained loss of requirement for phlebotomy in the absence of Anaemia and a Hb ≥2 g/dl decrease from baseline Hb level
cytoreductive therapy
Leukoerythroblastic peripheral blood picture Leukoerythroblastic peripheral blood picture
Increasing splenomegaly defined as either an increase in palpable Increasing splenomegaly defined as either an increase in palpable
splenomegaly of ≥5 cm from the LCM, or the appearance of a newly splenomegaly of ≥5 cm from the LCM, or the appearance of a newly
palpable splenomegaly palpable splenomegaly
Development of ≥1 of the constitutional symptoms (>10% weight loss in 6 Development of >1 of the constitutional symptoms (>10% weight loss in 6
months, night sweats, unexplained fever >37.5°C) months, night sweats, unexplained fever >37.5°C)
Increased lactate dehydrogenase
WHO, World Health Organization; LCM, left costal margin; Hb, haemoglobin; PV, polycythaemia vera; ET, essential thrombocythaemia.
Reprinted by permission from Macmillan Publishers Ltd.: Leukemia [7], copyright 2008.
About 60%–80% of JAK2 and MPL un-mutated patients with or PET-MF is made using the criteria described by the
ET and PMF have mutations in the exon 9 of CALR. Therefore, International Working Group for Myeloproliferative Neoplasm
virtually all patients with PV have a mutation in JAK2. Research and Treatment (IWG-MRT) (Table 3).
Conversely, 10%–15% of ET and PMF patients remain molecu-
larly uncharacterised, and are operationally defined as ‘triple
negative’ for the three phenotypic driver mutations. The pres- assessment of prognosis
ence of any of these mutations excludes reactive forms of ery- Any patient newly diagnosed with MPN should be categorised
throcytosis, thrombocytosis and MF, but does not indicate a at baseline according to the risks associated with the disease
specific MPN subtype. These mutations can be found using [I, B]. It must be realised that the prognostic scoring systems
methods such as conventional sequencing, qualitative and quan- used for risk-adapted therapy in PV and ET are based on the
titative polymerase chain reaction (PCR) and high-resolution likelihood of patients developing thrombotic complications. These
melting analysis (may have a sensitivity of 1% and higher). complications are the leading cause of morbidity and mortality in
Whole blood or purified granulocytes are harvested and tested; PV and ET patients. Scores predicting for overall survival in PV
the latter is preferred in cases with low mutation burden, as is and ET are also available, yet considering the long survival of these
often the case with JAK2 exon12 mutations. Genotyping should disorders, they do not currently impact treatment decisions [14,
be obtained at diagnosis. It is not recommended to measure the 15]. The recommended prognostic scoring system for PV and ET
mutation burden serially during follow-up or to assess response is based upon two variables: age >60 years and previous history of
to treatment, except following allogeneic stem-cell transplant- thrombosis. These variables separate patients into low- or high-
ation (alloSCT) and, possibly, interferon (IFN) treatment. In risk categories (Table 4). In ET, an intermediate-risk group is
such instances, a detection limit of JAK2V617F allele burden of sometimes advocated. However, this group is variably defined and
≤0.1% is recommended [13]. there is no clear evidence of how to manage patients in this cat-
egory. Thrombocytosis (>1000 × 109/l) is a risk factor for haemor-
rhage, and advocates caution for the use of aspirin. Extreme
achieving an accurate diagnosis thrombocytosis (>1500 × 109/l) is regarded as an indication for
Accurate differentiation among the three unique MPN subtypes therapy in ET, and less frequently in PV. Improved risk stratifica-
as well as the exclusion of reactive conditions (in mutation- tion is desirable, but any new risk stratification should be robust,
negative patients only) and disorders such as myelodysplasia easily measurable and ideally validated in a prospective manner.
and chronic myeloid leukaemia (CML) are critical for appropri- Since the median survival in PMF is ∼6 years, ranging from
ate prognosis and therapy decision making. It is not acceptable <2 to >10 years, the relevant end point for current prognostic
to use the generic diagnostic label ‘MPN’ alone [I, A]. The 2008 scoring systems in PMF is represented by survival (Table 5). The
WHO diagnostic criteria for MPN outlined in Table 1 should be International Prognostic Scoring System (IPSS) [18] is used at
followed strictly [1]. Any patient with suspected MPN should be the time of diagnosis to outline four risk categories (low, inter-
tested for the three driver mutations [I, B]. A simplified diagnos- mediate-1, intermediate-2 and high risk), with median survival
tic algorithm is presented in Figure 1. A diagnosis of PPV-MF of 135, 95, 48 and 27 months, respectively. The ‘dynamic’ IPSS
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv203 | v

by guest
on 05 February 2018
v | Vannucchi et al.
Figure 1. Diagnostic algorithm for MPN, based on the 2008 WHO criteria. PV, polycythaemia vera; ET, essential thrombocythaemia; PMF, primary myelofibrosis; SN, sub-normal EPO level; N, EPO levels in the
normal range; IN, increased EPO level above normal range; BM biopsy, bone marrow biopsy; WHO, World Health Organization.
Annals of Oncology
by guest
on 05 February 2018
(DIPSS) [19], which utilises the same variables as the IPSS, is incorporated in an updated prognostic score. Appropriate risk
employed during follow-up. A refinement is represented by stratification in PMF has relevance for risk-adapted therapy,
DIPSS-plus score [20] that incorporates thrombocytopaenia, particularly for alloSCT. These scores are commonly used for
transfusion requirements and abnormal cytogenetics (Table 5). PPV-MF and PET-MF. It should be underlined that they have
The role of CALR [21] and other mutations (i.e. EZH2, ASXL1, not been validated in such a context, and their appropriateness
SRSF2, IDH1/2 mutations), comprising a high-molecular risk has been questioned [23].
category in PMF [22], has been underscored but has yet to be
Table 4. Risk stratification and risk-adapted therapy in

management
polycythaemia vera (PV) and essential thrombocythaemia (ET) [16, 17] Treatment aims are to reduce the risk of thrombosis and haem-
Risk Risk variables Therapy orrhage, control symptoms and perhaps reduce the risk of pro-
category PV ET gression (Figure 2). Cure is not presently possible, except in
selected MF patients who are successfully receiving alloSCT. All
Low • Age <60 • Phlebotomy, • Correction of
patients should be informed regarding the disease course, and
years and CV risk factors,
vascular risk factors such as smoking should be aggressively
• No • Correction of and
managed.
thrombosis CV risk factors, • Low-dose
history and aspirina
• Low-dose polycythaemia vera
aspirina PV therapy should address both short- and long-term objectives
High • Age ≥60 • Cytoreduction, • Cytoreduction, (Figure 2A). In the short-term, therapeutic aims are to reduce
years and and the risk of occurrence and recurrence of thrombosis. The long-
and /or • Correction of • Correction of term objective is to reduce the risk of evolution to MF, myelo-
• Thrombosis CV risk factors, CV risk factors,
dysplastic syndrome (MDS) and/or acute myeloid leukaemia
history and and
(AML) [16]. Risk stratification (Table 4) aims at selecting the
• Low-dose • Low-dose
patients with a low risk of vascular events. PV is associated with
aspirina aspirina
elevated haematocrit (HCT); with these patients, phlebotomy is
• Phlebotomy if
required
carried out to control the HCT and low-dose aspirin is used, as
it may delay the need for cytoreductive therapy.
a
Depending on the thrombosis type, oral anti-coagulation instead of
low-dose aspirin. first-line therapy. Phlebotomy can be an emergency therapy at
CV, cardiovascular. diagnosis, in patients presenting with very high HCT and clinical
signs of hyperviscosity, as well as a long-term maintenance
Table 5. Risk stratification in primary myelofibrosis (MF)

Variable IPSS [18] DIPSS [19] DIPSS-plus [20]
Age >65 years ✓ ✓ ✓

Constitutional symptoms ✓ ✓ ✓
Haemoglobin (Hb) <10 g/dl ✓ ✓ ✓
Leukocyte count >25 × 109/l ✓ ✓ ✓
Circulating blasts >1% ✓ ✓ ✓
Platelet count <100 × 109/l ✓
RBC transfusion need ✓
Unfavourable karyotypea ✓
1 point each 1 point each but Hb = 2 Calculated by the DIPSS score (Int
1 = 1, Int 2 = 2, High = 3) plus one
additional point for each of the
three additional variables
Risk group Points Median survival (years) Points Median survival (years) Points Median survival (years)
Low 0 11.3 0 n.r. 0 15.4

Intermediate-1 1 7.9 1–2 14.2 1 6.5
Intermediate-2 2 4.0 3–4 4 2–3 2.9
High ≥3 2.3 5–6 1.5 ≥4 1.3
a
Unfavourable karyotype includes +8, –7/7q–, i(17q), inv(3), –5/5q–, 12p–, 11q23 rearrangements.
IPSS, International Prognostic Scoring System; DIPSS, dynamic International Prognostic Scoring System; RBC, red blood cell; Int, intermediate; n.r., not
reached.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv203 | v

by guest
on 05 February 2018
v | Vannucchi et al.
Figure 2. Therapeutic algorithms for polycythaemia vera, essential thrombocythaemia, and myelofibrosis. * Dynamic IPSS and Dynamic IPSS-plus after diagnosis. **Hydroxyurea for symptomatic splenomegaly in
countries where ruxolitinib is not approved for low-risk patients. If anaemia is the problem, erythropoietin, corticosteroids, danazol, immunomodulators or splenectomy. ***For patients presenting with symptomat-
ic splenomegaly and/or constitutional symptoms if allowed by the label. §For patients presenting with symptomatic splenomegaly and/or constitutional symptoms. PV, polycythaemia vera; ET, essential thrombo-
cythaemia; LD-Asa, low-dose aspirin; HU, hydroxyurea; INF-α, interferon-α; IPSS, International Prognostic Score System; Int, intermediate; AlloSCT, allogeneic stem cell transplantation.
Annals of Oncology
by guest
on 05 February 2018
therapy to control the HCT [I, A]. The optimal target of HCT HU. These results suggest that ruxolitinib could be a new option
levels for reducing vascular events was a matter of debate, as a second-line therapy in PV [I, A] [34].
but, a recent multicentre, randomised clinical trial (CYTO-
PV) showed that the HCT should be maintained strictly essential thrombocythaemia
below 45% to efficiently reduce the risk of thrombotic events
Aspirin in ET has never been evaluated in a randomised, con-
[I, A] [24]. Low-dose aspirin is the second cornerstone of PV
trolled trial, and there is a concern that bleeding is a particular
therapy [I, A]. It has been shown in the European
risk for ET patients with extreme thrombocytosis (>1000 × 109/l)
Collaboration on Low-dose Aspirin in Polycythaemia Vera
[35]. Low-dose aspirin is recommended in the setting of high-
(ECLAP) study, a large European double-blind, placebo-
risk ET patients without a clear contraindication to this therapy
controlled, randomised trial, to significantly reduce a primary
[III, B]. For low-risk ET patients, a retrospective analysis sug-
combined end point, including: cardiovascular death, non-fatal
gested that only those who are either JAK2V617F-positive or
myocardial infarction, non-fatal stroke and major venous
have cardiovascular risk factors may benefit from anti-platelet
thromboembolism [25].
therapy [36]. Pending further data, we recommend low-dose
A cytoreductive drug should be prescribed in high-risk PV
aspirin for low-risk ET patients, since thrombosis remains the
patients, i.e. >60 years and/or with a history of a vascular event.
major clinical hazard [III, B] (Figure 2B).
In 2011, the European LeukemiaNet (ELN) published recom-
The choice of the most appropriate first-line cytoreductive
mendations for the management of PV, concluding that hydro-
therapy for high-risk ET is based on three randomised trials
xyurea (HU) [II, A] and IFN-α [III, B] were recommended as
[37–39]. There is a debate regarding whether HU, anagrelide or
first-line treatments for high-risk patients [16]. HU is a well-
IFN-α should be the treatment of choice. In addition, there is
known cytoreductive agent, with good efficacy and tolerance in
an increased interest in ‘pegylated’ forms of IFN. HU versus
the majority of patients. In the unique randomised trial compar-
no myelosuppressive therapy significantly reduced the rate of
ing HU with another cytoreductive drug ( pipobroman) in PV,
thrombosis in high-risk ET patients, most of whom received
the cumulative incidence of AML/MDS at 10, 15 and 20 years
anti-platelet prophylaxis with aspirin or ticlopidine [37]. The
was 6.6%, 16.5% and 24% in the HU arm and 13%, 34% and
use of anagrelide versus HU has been evaluated in the PT-1
52% in the pipobroman arm, respectively (P = 0.004) [26].
study and the non-inferiority ANAHYDRET study. Anagrelide
Other studies from registry data and prospective analysis with
was equivalent to HU in reducing platelet counts in both
shorter follow-up failed to attribute a clear leukaemogenic risk
studies. In the PT-1 study, an excess of arterial thrombosis was
to HU [17, 27]. Overall, there is no definitive evidence for (or
seen in the anagrelide arm compared with HU [38]. However, in
against) a leukaemogenic risk of HU, but it should be empha-
the ANAHYDRET, study equivalence was reported [39]. The
sised that this risk may appear after prolonged exposure to this
use of HU and low-dose aspirin as first-line therapy for high-
drug. Thus, it seems reasonable to adopt a conservative ap-
risk ET is recommended, but in specific groups of patients, IFN
proach and to consider alternative treatments in young subjects,
or sometimes anagrelide may be appropriate [I, B]. Whenever
and in those previously treated with other myelosuppressive
possible, patients with ET should be enrolled in randomised
agents. IFN-α has been shown to induce a high rate of haemato-
studies evaluating HU compared with IFN. According to current
logical response and to significantly reduce the malignant clone,
label approvals in Europe, anagrelide may be used as second-line
as shown by the percentage of mutated allele JAK2V617F in two
therapy for patients who are resistant or intolerant to HU. IFN or
phase II studies [28, 29]. However, this drug (in any of its
busulfan are also options available in this setting. The use of
various presentations) is not approved for the treatment of PV.
cytotoxic agents, in the youngest patients and/or especially in
Two phase III studies comparing HU to pegylated forms of
combination, should be avoided where possible [III, B]. In the
IFN-α are ongoing in the United States and Europe. These
presence of extensive thrombocytosis (>1.500 × 109/l), the bene-
studies should help to better define the efficacy and tolerance of
fits of anti-platelet agents should be balanced against the risks of
each drug in the short term [30]. Whenever possible, patients
haemorrhages due the occurrence of acquired von Willebrand
should be recruited into these studies [II, B]. Aquagenic pruritus
disease [35].
is a disabling symptom in some PV patients: IFN-α or JAK2
inhibitors can be used to treat this symptom. Other options
include antihistamines, selective serotonin reuptake inhibitors myelofibrosis
and PUVA ( psolarens + ultraviolet A) therapy [IV, B] [31, 32]. Since there is no curative therapy other than alloSCT for PMF
and PPV-/PET-MF, treatment is essentially palliative and is gen-
erally guided by the predominant symptoms, anaemia and
second-line therapy. The choice of second-line myelosuppres- splenomegaly (Figure 2C).
sive drugs for PV should be carefully evaluated because some
drugs administered after HU may enhance the risk of acute anaemia. A haemoglobin <10 g/dl usually triggers consider-
leukaemia [II, B] [33]. Finally, in selected patients, alkylating ation of treatment, but there are individual variations depending
agents like busulfan may be useful when other drugs have failed upon age and comorbidities. One of the first options is
or are contraindicated, although they are characterised as erythropoiesis-stimulating agents, which produce improvements
inducing an increased leukaemogenic risk [III, C] [16]. In PV, a in 23%–60% of patients [40, 41]. Response is often restricted to
phase III study showed the use of ruxolitinib was superior, patients with inadequate erythropoietin levels (<125 mU/ml),
compared with the best available therapy, to control HCT and and less frequently when there is significant splenomegaly or
splenomegaly in PV patients who are resistant or intolerant to transfusion dependence. If no response is obtained at three
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv203 | v

by guest
on 05 February 2018
months, treatment should be stopped [III, B]. Androgens such It has been reported that sudden ruxolitinib withdrawal can
as nandrolone, fluoxymesterone, methandrostenolone and oxy- provoke a shock-like syndrome, due to the re-emergence of the
metholone improve anaemia in 30%–60% of patients [IV, B]. suppressed inflammatory cytokines [51]. Though this side-effect
Similar results with less toxicity are obtained with danazol, with is rare, abrupt interruption should be avoided and withdrawal of
the overall response rate being 35% [III, B] [42]. The recommen- the drug should be tapered.
ded dose is 400–600 mg daily maintained for at least 6 months, Two phase III studies compared ruxolitinib with placebo
then progressively reduced to the minimum necessary for main- (COMFORT-I) [52] or best available therapy (COMFORT-II)
tenance. Immunomodulating drugs may also be useful in man- [53]; both attained the primary end point of ≥35% reduction in
aging anaemia but are frequently withdrawn early, due to spleen volume by imaging techniques at 24 or 48 weeks of treat-
toxicity. Low-dose thalidomide combined with oral prednisone ment, respectively [I, A]. The effect on JAK2V617F allele burden
provides a 23%–29% response [III, C] [43]. Lenalidomide, com- was modest [54]. A survival advantage for patients treated with
bined with low-dose prednisone taper, produces a 19% response ruxolitinib was first shown from historical comparison with
[III, C] [44]. Lenalidomide as a single agent is the treatment of matched MF populations [55, 56]. Extended follow-up of the
choice for MF patients with 5q deletion [V, C] [45]. Splenectomy phase III studies indicated a survival advantage for patients on
can be useful in patients with transfusion-dependent anaemia re- ruxolitinib [54]. However, there is little evidence of a disease-
fractory to drug therapy [IV, B] [46], but needs careful consider- modifying effect. Other JAK inhibitors are currently being
ation due to complication rates (see below). Corticosteroids alone studied in clinical trials, although several trials have been with-
may also be used for the management of refractory anaemia in drawn due to emergent neurological toxicity. The precise mech-
patients unresponsive to the above drugs, and who are not eligible anism for this toxicity is unclear but merits close monitoring.
for alloSCT or splenectomy. Corticosteroids used alone are often
observed to result in modest haemoglobin increases and improve- stem-cell transplantation. AlloSCT is currently the only
ments in patient well-being [V, C]. curative treatment approach for MF, resulting in resolution of
BM fibrosis, molecular remission and restoration of normal
haematopoiesis [57, 58]. Depending on the disease status, patient’s
splenomegaly and extra-medullary haematopoiesis. Traditionally,
performance status, comorbidities and donor availability, ∼40%–
treatment of splenomegaly was not instituted before the
70% of patients can be cured [57, 58]. Careful patient selection is
appearance of associated symptoms, particularly due to the
mandatory, due to the inherent risks of alloSCT. Patients with PV
inherent risk of worsening cytopaenias. HU was previously the
or ET are not candidates for alloSCT, unless their disease has
first-line therapy for symptomatic splenomegaly, with an overall
transformed into MF or secondary acute leukaemia [59]. Results
response of 40% [IV, B] [47]. However, published experience
with completely matched, related and unrelated donors are superior
suggests that after 1 year of treatment, ∼80% of patients require
to those with human leukocyte antigen-mismatched donors [57,
an alternative therapy. The use of HU is now largely superseded
60]. Reduced-intensity conditioning regimens resulting in lower
by JAK inhibitors (see below). Splenectomy is indicated in patients
therapy-related complications have broadened the availability of
with large and painful splenomegaly where JAK inhibitors are
alloSCT to older patients, but a direct comparison to standard
not available or prove ineffective [46]. Splenectomy requires an
myeloablative conditioning is lacking [57, 59]. According to ELN
experienced surgical team and critical care support to minimise
recommendations, it is justified to offer alloSCT to eligible patients
the risks associated with the procedure; a perioperative
with MF whose median survival is expected to be <5 years. This
mortality rate of 5%–10% and a morbidity rate up to 25% can be
includes patients with intermediate-2 and high risk according to
expected [IV, D] [46]. Splenic irradiation can also be applied in
IPSS [III, A] [16]. Splenectomy is generally not recommended
patients who do not tolerate JAK inhibitors and are poor
in preparation for alloSCT [IV, D]. Pre-transplant JAK inhibitor
candidates for surgery [IV, D] [48]. However, benefit is transient
treatment can reduce spleen size and improve constitutional
and involves the risk of severe cytopaenias; therefore, its routine
symptoms, but is currently being tested in clinical studies and
use is not recommended. Low-dose radiation is the therapy of
should be regarded as experimental [IV, D].
choice for symptomatic extra-medullary haematopoiesis in
places other than the spleen and liver, as well as for MF-
associated pulmonary hypertension, due to extra-medullary personalised medicine
haematopoiesis [IV, B] [49].
MPNs are diseases that typically affect people within a mid-
advanced age group. Therefore, any treatment decisions, espe-
JAK inhibitors. The JAK inhibitors act mainly by inhibiting cially regarding thrombosis prevention and selection of patients
dysregulated JAK-STAT signalling, present in all MF patients. for alloSCT, should consider the patient’s general condition and
They are not selective of the mutated JAK2; therefore, they are comorbidities. The diagnostic approach has improved remark-
indicated in both JAK2-mutated and JAK2-unmutated MF. ably with the discovery of recurrent phenotypic driver muta-
Ruxolitinib, an oral JAK1/JAK2 inhibitor, is the first in-class tions, but there is still a need for standardised and validated
drug approved for MF treatment [50]. Thrombocytopaenia is a mutational tests. The interpretation of BM histopathology fea-
main adverse event observed with JAK inhibitors, and tures, as required by the current WHO classification, requires
worsening anaemia is often seen, especially at the beginning of experienced pathologists, particularly for the differential diagno-
therapy. These drugs are also associated with an increased risk sis of early and/or for mutation-negative cases.
of infection. Spleen reduction and symptom control are usually There are a number of prognostic scores that are useful for
dramatic but are also drug- and dose-dependent. guiding treatment, yet they have been built on retrospective
v | Vannucchi et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
series and are not validated prospectively, nor in the settings of HU dosage and/or IFN with very infrequent/absent phlebotomy
conventional or new therapies. This is a challenge of particular needs, a 4- to 6-month interval may be appropriate. In low-risk
relevance today, when novel targeted drugs are being used in ET patients, a visit at the 6- to 12-month interval is sufficient,
clinical trials, and the first in-class JAK inhibitor has been while for those with high-risk ET receiving cytoreduction, a 3-
approved. However, in spite of these recent advancements, we to 4-month interval is recommended. Clinical and laboratory
have to acknowledge that there are still too few effective treat- evaluation should be more frequent in the initial phases of
ment options for highly heterogeneous diseases such as MPNs, disease management until a stable phlebotomy rate and/or drug
which additionally show high degrees of variability from patient dose is achieved.
to patient. In this disease setting, more research is needed to Most patients maintain their HU dose steadily for several years,
identify molecular markers which could lead to advances in although in a minority of patients, excessive myelosuppression
personalised medicine. may develop that requires dose adjustment. Routine chemistry as-
sessment (including at least lipid panel, glucose, renal and hepatic
function tests) is recommend at 1-year intervals. While an ultra-
long-term follow-up and response sound scan of the abdomen is suggested at first visits to exclude
evaluation subclinical splanchnic vein thrombosis or spleen infarcts, assess-
In patients with asymptomatic and well-controlled PV, follow- ment of splenomegaly can be conveniently carried out by palpa-
up visits can be scheduled at 2–4 month intervals to determine tion at each follow-up visit. Unless there is a suspicion of disease
phlebotomy needs; in those with high-risk PV receiving stable progression to post-PV/post-ET MF, there is no indication to
Table 6. Criteria for assessing response to treatment in polycythaemia vera (PV) and essential thrombocythaemia (ET) according to the ELN criteria
[61]
Criteria
PV ET
Complete remission
A Durable resolution of disease-related signs including palpable Durable resolution of disease-related signs including palpable
haepatosplenomegaly, large symptom improvement AND haepatosplenomegaly, large symptom improvement AND
B Durable peripheral blood count remission, defined as HCT lower than 45% Durable peripheral blood count remission, defined as platelet count
without phlebotomies; platelet count ≤400 × 109/l, WBC count ≤400 × 109/l, WBC count <10 × 109/l, absence of
<10 × 109/l, AND leukoerythroblastosis, AND
C Without progressive disease, and absence of any haemorrhagic or Without progressive disease, and absence of any haemorrhagic or
thrombotic events, AND thrombotic events, AND
D Bone marrow histological remission defined as the presence of age-adjusted Bone marrow histological remission defined as disappearance of
normal cellularity and disappearance of trilinear hyperplasia, and megakaryocyte hyperplasia and absence of >grade 1 reticulin fibrosis
absence of >grade 1 reticulin fibrosis
Partial remission
A Durable resolution of disease-related signs including palpable Durable resolution of disease-related signs including palpable
haepatosplenomegaly, large symptom improvement AND haepatosplenomegaly, large symptom improvement AND
B Durable peripheral blood count remission, defined as HCT lower than 45% Durable peripheral blood count remission, defined as platelet count
without phlebotomies; platelet count ≤400 × 109/l, WBC count ≤400 × 109/l, WBC count <10 × 109/l, absence of
<10 × 109/l, AND leukoerythroblastosis, AND
C Without progressive disease, and absence of any haemorrhagic or Without progressive disease, and absence of any haemorrhagic or
thrombotic events, AND thrombotic events, AND
D Without bone marrow histological remission defined as persistence of Without bone marrow histological remission defined as persistence of
trilinear hyperplasia megakaryocyte hyperplasia
No response
Any response that does not satisfy partial remission Any response that does not satisfy partial remission
Progressive disease
Transformation into post-PV myelofibrosis, myelodysplastic syndrome Transformation into post-PV myelofibrosis, myelodysplastic syndrome
or acute leukaemia or acute leukaemia
Durable = lasting at least 12 weeks.

Large symptom improvement = a ≥10-point decrease in the Myeloproliferative Neoplasm-Symptom Assessment Form Total Symptom Score (MPN-SAF
TSS) [62].
Molecular response is not required for assignment as complete or partial response.
WBC, white blood cells; HCT, haematocrit; ELN, European LeukemiaNet.
Republished with permission of the American Society of Hematology, from [61]; permission conveyed through Copyright Clearance Center, Inc.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv203 | v

by guest
on 05 February 2018
Table 7. Criteria for assessing response to treatment in PMF according to the IWG-MRT and ELN criteria [63]
Response categories Required criteria (benefit must last >12 weeks to qualify as a response)
Complete remission (CR) Bone marrow: age-adjusted normal cellularity; <5% blasts; ≤grade 1 fibrosis, and
Peripheral blood: Hb ≥100 g/l, and <UNL; neutrophil count ≥1 × 109/l and <UNL;
Platelets ≥100 × 109/l and <UNL; <2% immature myeloid cells, and
Clinical: resolution of disease symptoms; spleen and liver not palpable; no evidence of EMH
Partial remission (PR) Peripheral blood: Hb ≥100 g/l, and <UNL; neutrophil count ≥1 × 109/l and <UNL;
platelets ≥100 × 109/l and <UNL; <2% immature myeloid cells, and
Clinical: resolution of disease symptoms; spleen and liver not palpable; no evidence of EMH, or
Bone marrow: age-adjusted normal cellularity; <5% blasts; <grade 1 fibrosis, and
Peripheral blood: Hb ≥85 g/l but <100 g/l and <UNL; neutrophil count ≥1 × 109/l and <UNL; platelets ≥50 × 109/l but
<100 × 109/l and <UNL; <2% immature myeloid cells, and
Clinical: resolution of disease symptoms; spleen and liver not palpable; no evidence of EMH
Clinical improvement (CI) The achievement of anaemia, spleen or symptom response without progressive disease or increase in severity of anaemia,
thrombocytopaenia or neutropaenia
Anaemia response Transfusion-independent patients: a >20 g/l increase in haemoglobin level
Transfusion-dependent patients: becoming transfusion-independent
Spleen response A baseline splenomegaly that is palpable at 5–10 cm below the LCM becomes not palpable, or
A baseline splenomegaly that is palpable at >10 cm below the LCM decreases by ≥50%
A baseline splenomegaly that is palpable at <5 cm below the LCM is not eligible for spleen response
A spleen response requires confirmation by MRI or CT showing ≥35% spleen volume reduction
Symptom response A ≥50% reduction in the MPN-SAF TSS
Progressive disease Appearance of a new splenomegaly that is palpable at least 5 cm below the LCM, or
A ≥100% increase in palpable distance below LCM for baseline splenomegaly 5–10 cm, or
A 50% increase in palpable distance below LCM for baseline splenomegaly of >10 cm, or
Leukaemic transformation confirmed by a bone marrow blast count of ≥20%, or
A peripheral blood blast content of ≥20% associated with an absolute blast count of ≥1 × 109/l that last for at least 2 weeks
Stable disease Belonging to none of the above response categories
Relapse No longer meeting criteria for at least a CI after achieving CR, PR or CI, or
Loss of anaemia response persisting for at least 1 month, or
Loss of spleen response persisting for at least 1 month
Cytogenetic remission At least 10 metaphases must be analysed for cytogenetic response evaluation and requires confirmation by repeat testing
within a 6-month window
CR: eradication of a pre-existing abnormality
PR: ≥50% reduction in abnormal metaphases
(PR applies only to patients with at least 10 abnormal metaphases at baseline)
Molecular remission Molecular response evaluation must be analysed in peripheral blood granulocytes and requires confirmation by repeat
testing within a 6-month window
CR: eradication of a pre-existing abnormality
PR: >50% decrease in allele burden
(partial response applies only to patients with at least 20% mutant allele burden at baseline)
Cytogenetic/molecular relapse Re-emergence of a pre-existing cytogenetic or molecular abnormality that is confirmed by repeat testing
Grading of fibrosis is according to European scale (see Table 2).

Transfusion dependency before study enrolment is defined as transfusions of at least 6 units of packed red cells (PRBC), in the 12 weeks before study
enrolment, for a haemoglobin level of <85 g/l, in the absence of bleeding or treatment-induced anaemia. In addition, the most recent transfusion episode
must have occurred in the 28 days before study enrolment. Response in transfusion-dependent patients requires absence of any PRBC transfusions during
any consecutive ‘rolling’ 12-week interval during the treatment phase, capped by a haemoglobin level of ≥85 g/l.
Spleen or liver responses must be confirmed by imaging studies where a ≥35% reduction in spleen volume, as assessed by MRI or CT, is required. Furthermore,
a ≥35% volume reduction in the spleen or liver, by MRI or CT, constitutes a response regardless of what is reported with physical examination.
MPN-SAF TSS, MPN symptom self-assessment form total symptom score. This is assessed by the patients themselves and includes fatigue, concentration, early
satiety, inactivity, night sweats, itching, bone pain, abdominal discomfort, weight loss and fevers, which are ranked each from 0 (absent/as good as it can be) to
10 (worst imaginable/as worst as it can be); the TSS is the summation of all individual scores (0–100 scale).
PMF, primary myelofibrosis; IWG-MRT, International Working Group for Myeloproliferative Neoplasm Research and Treatment; ELN, European Leukemia
Net; UNL, upper normal limit; EMH, extra-medullary haematopoiesis; LCM, left costal margin; MRI, magnetic resonance imaging; CT, computed tomography;
Hb, haemoglobin.
Republished with permission of the American Society of Hematology, from [63]; permission conveyed through Copyright Clearance Center, Inc.
v | Vannucchi et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
serially repeat BM biopsy. Also, outside a clinical study and with end points. Cytogenetic and molecular remissions, although
the exception of patients IFNs, repeated assessment of mutation considered, are not formally required for complete remission to
burden is not recommended. For patients with MF, visits must be defined (Table 7) [63]. However, these criteria have been
be more frequent, ranging from every week to 3 months depend- developed mainly for use in the clinical trials setting, and there-
ing on the general status, the presence of blood cell abnormalities fore should not be used as a tool to assess response to conven-
and the type of therapy instituted. There is no firm evidence to tional therapies. In daily practice, the optimal but unvalidated
suggest a specific type and/or frequency of follow-up; therefore, the target in ET include platelets <400 × 109/l [V]. On the other
above indications are mainly derived from experts’ experience. hand, the randomised, controlled CYTO-PV trial established
Criteria for assessing the response to treatment have been the superiority of an HCT <45% as the goal of treatment of
developed by the ELN and the IWG-MRT (Table 6). The cri- patients with PV [I, A] [24].
teria for PV and ET include definitions of complete and partial
remission, incorporating clinical, haematological and histological re- special situations
sponse assessments, including a standardised symptom assessment antithrombotic therapy. The prophylactic use of low-dose aspirin
form [61]. In the case of PMF, the revised criteria include six re- in PV and ET has been already discussed. In patients with MF and
sponse categories encompassing clinical and haematological normal/increased platelet counts, the use of low-dose aspirin is not
Table 8. European LeukemiaNet criteria for definition of resistance/intolerance to hydroxyurea in patients with polycythaemia vera, essential
thrombocythaemia and primary myelofibrosis
(A) Polycythaemia vera [66]

1. Need for phlebotomy to keep HCT <45% after 3 months of at least 2 g/day of HU, OR
2. Uncontrolled myeloproliferation, i.e. platelet count >400 × 109/l AND white blood cell count >10 × 109/l after 3 months of at least 2 g/day of HU,
OR
3. Failure to reduce massivea splenomegaly by more than 50% as measured by palpation, OR failure to completely relieve symptoms related to
splenomegaly, after 3 months of at least 2 g/day of HU, OR
4. Absolute neutrophil count <1.0 × 109/l OR platelet count <100 × 109/l OR Hb <100 g/l at the lowest dose of HU required to achieve a complete or
partial clinicohaematological response, OR
5. Presence of leg ulcers or other unacceptable HU-related non-haematological toxicities, such as mucocutaneous manifestations, gastrointestinal
symptoms, pneumonitis or fever at any dose of HU
(B) Essential thrombocythaemia [67]
• Platelet count <600 × 109/l after 3 months of at least 2 g/day of HU (2.5 g/day in patients with a body weight >80 kg)
• Platelet count <400 × 109/l and white blood cells <2500/µl at any dose of HU
• Platelet count <400 × 109/l and Hb <10 g/dl at any dose of HU
• Presence of leg ulcers or other unacceptable mucocutaneous manifestations at any dose of HU
• HU-related fever
(C) Primary myelofibrosis [66]
1. Failure to: (i) reduce massivea or progressiveb splenomegaly or haepatomegaly in splenectomised patients, by more than 50% as measured by
palpation, OR (ii) completely relieve symptoms of splenomegaly or haepatomegaly in splenectomised patients, after 3 months of at least 2 g/day of
HU
2. Uncontrolled myeloproliferation, i.e. platelet count >400 × 109/l AND white blood cell count >10 × 109/l after 3 months of at least 2 g/day of HU,
OR
3. Failure to reduce massivea splenomegaly by more than 50% as measured by palpation, OR failure to completely relieve symptoms related to
splenomegaly, after 3 months of at least 2 g/day of HU, OR
4. Absolute neutrophil count <1.0 × 109/l OR platelet count <50 × 109/l at the lowest dose of HU required to achieve a complete or major
clinicohaematological response, OR
5. Presence of leg ulcers or other unacceptable HU-related non-haematological toxicities, such as mucocutaneous manifestations, gastrointestinal
symptoms, pneumonitis or fever at any dose of HU
a
Organ extending by more than 10 cm from the costal margin.
b
Organ increasing by more than 3 cm in the last 3 months.
For (A), complete response was defined as: HCT <45% without phlebotomy, platelet count <400 × 109/l, white blood cell count <10 × 109/l and no disease-
related symptoms. Partial response was defined as: HCT <45% without phlebotomy, or response in three or more of the other criteria [68].
For (C), complete response was defined as a complete response in anaemia, splenomegaly and constitutional symptoms; major response was defined as any
response in anaemia and splenomegaly without progression in constitutional symptoms, OR complete response in anaemia (or partial response in anaemia
that was transfusion-dependent), and response in constitutional symptoms without progression in splenomegaly, OR any response in splenomegaly and
response in constitutional symptoms without progression in anaemia [69].
HCT, haematocrit; Hb, haemoglobin; HU, hydroxyurea.
Reprinted from [66]. Copyright © 2010 by John Wiley & Sons, Inc. Reprinted by permission of John Wiley & Sons, Inc.
Reprinted by permission from Macmillan Publishers Ltd.: Leukemia [67], copyright © 2007.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv203 | v

by guest
on 05 February 2018
supported by clinical trials but may be appropriate considering Table 9. Criteria for defining ‘high risk’ a pregnancy in the course
a rate of thrombosis (fatal and non-fatal) of 1.75%, comparable of a myeloproliferative neoplasm (MPN) (adapted from [76])
with ET [64]. In patients who have already experienced major
cardiovascular events, prevention of recurrence should be carried • Sustained rise in platelet count rising to above 1500 × 109/la
• Previous venous or arterial thrombosis
out according to general lines of management, depending on the
• Previous haemorrhage attributed to MPNa
type and site of previous thrombosis. In patients with splanchnic
• Previous pregnancy complication
vein thrombosis or recurrent venous thrombosis and pulmonary
a. ≥1 unexplained deaths of a morphologically normal foetus ≥10
embolism, lifelong oral anti-coagulation is usually suggested, weeks of gestation
although there is debate among experts in the absence of controlled b. ≥1 premature delivery of a morphologically normal foetus <34
studies [65]. The use of new anti-coagulants in these specific weeks gestation because of:
settings has not been evaluated yet. Indications for special situations (i) Severe pre-eclampsia or eclampsia defined according to
are described below. standard definitions
(ii) Recognised features of placental insufficiency
resistance or intolerance to first-line cytoreductive agents: The c. ≥3 unexplained consecutive miscarriages <10 weeks gestation,
phenomenon of HU resistance or intolerance, as defined by the with maternal and paternal factors (anatomic, hormonal or
ELN (Table 8), is important. It identifies a group of ET or PV chromosomal abnormalities) excluded
patients with a poor prognosis [70, 71], who require a change of d. Otherwise unexplained intra-uterine growth restriction
treatment, and for whom novel therapies such as JAK inhibitors e. Significant antepartum or postpartum haemorrhage requiring
may be attractive. We recommend options for management that transfusion
include (in the face of HU resistance or modest intolerance): • Abnormal uterine artery Doppler at 20 weeks (mean pulsatility index
reducing the dose by adjusting therapeutic targets (e.g. raising the >1.4)
platelet count target to 600 × 109/l) or switch therapy, usually to
IFN (PV, ET) or anagrelide (ET). Busulfan may also be employed, a
Represents indication for IFN only rather than IFN plus low molecular
preferably in older patients. It is important to consider that when
weight heparin.
HU is used with (or succeeded by) other agents, including
busulfan, it will significantly increase the long-term risk of
leukaemia [III, B]. Ruxolitinib has been approved for patients
with PV who are refactory or resistant to HU [I, A] [34]. Doppler should be carried out at 20 weeks. In the presence of a
mean pulsatility index >1.4, the pregnancy may be considered
leukaemic transformation: Treatment of blast-phase MPN is high risk and treatment can be escalated, along with additional
usually disappointing [72]. Acute leukaemia-like regimens can growth scans, as appropriate.
be used in patients who are potential candidates for alloSCT Local protocols with regard to interruption of LMWH should
[IV, B] [72, 73]. Benefits have been reported with azacitidine be adhered to during labour, and dehydration should be
[IV, B] [74]. avoided. Postpartum LMWH thromboprophylaxis for 6 weeks
should be considered. Thrombosis has been documented in the
pregnancy: Current literature for pregnancy outcomes in postpartum period, and blood counts should be monitored at
MPN is sparse and likely to be subject to reporting bias. ET is this time.
the most common MPN in women of childbearing age. Breastfeeding is safe with low-dose aspirin, heparin and war-
Maternal morbidity is uncommon in ET, but has been reported. farin ( providing the new-born receives adequate vitamin K), but
Successful pregnancy occurs in up to 70% of patients. The it is traditionally contraindicated with cytoreductive therapy.
literature regarding PV and MF, while very limited, is concordant Decisions about breastfeeding while taking IFN should be made
with pregnancy outcome in ET. The suggested management of on an individual basis, after discussion regarding possible risks
pregnancy in MPN derives mainly from a few single-centre and benefits [III, B].
studies [III, B] [75]. Disease management should be optimised
before conception. HU and anagrelide should be stopped with an hormonal therapy: An individualised risk benefit assessment
adequate wash-out period. If cytoreductive therapy is needed, IFN should be undertaken in considering hormonal therapy. The
should be considered. Depending on the risk assessment results, overall evidence grade is poor here [V]. For testosterone therapy,
women should be managed according to a standard-risk or high- the major MPN-specific risk is of provoking an erythrocytosis
risk pregnancy protocol, with multidisciplinary review. Unless which should be monitored. Concerning oestrogen and
contraindicated, all women should receive low-dose aspirin progesterone, the major risk is of thrombosis. In this case, topical
throughout pregnancy. For women with PV, venesection can be therapies including coated intra-uterine devices are likely to be
continued to maintain HCT levels within a gestation-appropriate extremely low-risk and acceptable. For oral contraception,
target range, and is sufficient for standard-risk pregnancy. For progesterone-only preparations are acceptable, but the combined
high-risk pregnancy, i.e. if one (or more) of the factors in Table 9 oral contraceptive (i.e. both progesterone and oestrogen) is not
is present or is likely to occur during pregnancy, additional recommended [V]. Where short-term hormonal manipulation is
treatment including cytoreductive therapy with IFN and low required, for example in fertility treatment, thromboprophylaxis
molecular weight heparin (LMWH) may be considered. should be considered and the patient should be counselled about
Foetal monitoring is suggested during pregnancy; scans thrombosis risk. Lastly, menopausal hormone replacement has
should be carried out at 20, 26 and 34 weeks. Uterine artery recently been suggested to have minimal associated thrombosis
v | Vannucchi et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Table 10. Levels of evidence and grades of recommendation practice guidelines development. The relevant literature has
(adapted from the Infectious Diseases Society of America-United been selected by the expert authors. Levels of evidence and
States Public Health Service Grading Systema) grades of recommendation have been applied using the system
Levels of evidence shown in Table 10. Statements without grading were considered
justified standard clinical practice by the experts and the ESMO
I Evidence from at least one large randomised, controlled trial of faculty. This manuscript has been subjected to an anonymous
good methodological quality (low potential for bias) or meta- peer review process.
heterogeneity
II Small randomised trials or large randomised trials with a suspicion conflict of interest
of bias (lower methodological quality) or meta-analyses of such
AMV has reported advisory board or lecture honoraria from
Novartis, lecture honoraria from Shire and institutional research
support from Novartis. TB has reported advisory board and
lecture honoraria from Novartis. FC has reported advisory
board and honoraria from Novartis, Gilead, CTI/Baxter and
Grades of recommendation AOP Orphan Pharmaceuticals, and lecture honoraria from
A Strong evidence for efficacy with a substantial clinical benefit, Novartis and CTI/Baxter. CH has received advisory board or
strongly recommended lecture honoraria from Novartis, CTI, YM Bioscience, Sanofi,
B Strong or moderate evidence for efficacy but with a limited clinical Celgene, Gilead and Shire and institutional research support
benefit, generally recommended from Shire and Novartis. JJK has received advisory board or
C Insufficient evidence for efficacy or benefit does not outweigh the lecture honoraria from Novartis and Shire, and institutional re-
risk or the disadvantages (adverse events, costs, …), optional search support from AOP Orphan Pharmaceuticals and
D Moderate evidence against efficacy or for adverse outcome, Novartis. NK has reported advisory board and honoraria from
generally not recommended Novartis and Sanofi, lecture honoraria from Novartis and Sanofi
E Strong evidence against efficacy or for adverse outcome, never and institutional research support from Novartis and Sanofi. JT
recommended has reported consultant or advisory roles and honoraria from
Novartis, Incyte Corporation, AOP Orphan Pharmaceuticals
a
By permission of the Infectious Diseases Society of America [78]. and Sanofi. CB has reported honoraria from Roche, Pfizer,
Celgene, Pharmacyclics and Janssen and research grants from
Roche and Janssen.
risk. An individual assessment weighing up the risks and need for
treatment should be carried out.
references
surgery: Disease phenotype, individual patient variables and 1. Swerdlow SH, Campo E, Harris NL et al. WHO Classification of Tumors of
surgery-specific factors including choice of anaesthesia contribute Haematopoietic and Lymphoid Tissues. Lyon: International Agency for Research on
Cancer, 2008.
to the personal risk of postoperative thrombosis and bleeding in
2. Moulard O, Mehta J, Fryzek J et al. Epidemiology of myelofibrosis, essential
MPN. Most of the data exist for ET patients, and arterial events
thrombocythemia, and polycythemia vera in the European Union. Eur J Haematol
are increased postoperatively at ∼3.8%, but there is also a 10.5% 2014; 92: 289–297.
bleeding risk with surgery [77]. This side-effect is thought to be 3. Titmarsh GJ, Duncombe AS, McMullin MF et al. How common are
caused by a combination of disease-related primary platelet myeloproliferative neoplasms? A systematic review and meta-analysis. Am J
abnormalities, anti-platelet agents and anti-coagulant therapy. For Hematol 2014; 89: 581–587.
this reason, careful preoperative review is required, with consideration 4. Cervantes F, Dupriez B, Passamonti F et al. Improving survival trends in primary
given to temporary control of platelet counts (especially for myelofibrosis: an international study. J Clin Oncol 2012; 30: 2981–2987.
procedures where there is a significant risk if bleeding occurs or 5. Tefferi A, Thiele J, Vannucchi AM, Barbui T. An overview on CALR and CSF3R
when patients are at increased risk of bleeding or thrombosis, as mutations and a proposal for revision of WHO diagnostic criteria for
myeloproliferative neoplasms. Leukemia 2014; 28: 1407–1413.
a consequence of the procedure). The use of anti-platelet and
6. Thiele J, Kvasnicka HM, Facchetti F et al. European consensus on grading bone
anti-coagulants should be adjusted according to local policy. marrow fibrosis and assessment of cellularity. Haematologica 2005; 90:
Postoperative thromboprophylaxis with LMWH is recommended. 1128–1132.
It is not necessary to extend thromboprophylaxis beyond the 7. Barosi G, Mesa RA, Thiele J et al. Proposed criteria for the diagnosis of post-
normal period of time postoperatively simply because of an MPN polycythemia vera and post-essential thrombocythemia myelofibrosis: a consensus
diagnosis. Controlling blood counts preoperatively to standard statement from the International Working Group for Myelofibrosis Research and
targets for high-risk patients should be considered in MPN Treatment. Leukemia 2008; 22: 437–438.
patients undergoing surgery, when bleeding is a risk or when 8. Thiele J, Kvasnicka HM, Müllauer L et al. Essential thrombocythemia versus early
primary myelofibrosis: a multicenter study to validate the WHO classification. Blood
thromboprophylaxis would normally be prescribed [IV, C].
2011; 117: 5710–5718.
9. Gianelli U, Bossi A, Cortinovis I et al. Reproducibility of the WHO histological
methodology criteria for the diagnosis of Philadelphia chromosome-negative myeloproliferative
neoplasms. Mod Pathol 2014; 27: 814–822.
These clinical practice guidelines were developed in accordance 10. Madelung AB, Bondo H, Stamp I et al. World Health Organization-defined
with the ESMO standard operating procedures for clinical classification of myeloproliferative neoplasms: morphological reproducibility and
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv203 | v

by guest
on 05 February 2018
clinical correlations—the Danish experience. Am J Hematol 2013; 88: 32. Tefferi A. JAK inhibitors for myeloproliferative neoplasms: clarifying facts from
1012–1016. myths. Blood 2012; 119: 2721–2730.
11. Thiele J, Kvasnicka HM. The 2008 WHO diagnostic criteria for polycythemia vera, 33. Finazzi G, Caruso V, Marchioli R et al. Acute leukemia in polycythemia vera: an
essential thrombocythemia, and primary myelofibrosis. Curr Hematol Malig Rep analysis of 1638 patients enrolled in a prospective observational study. Blood
2009; 4: 33–40. 2005; 105: 2664–2670.
12. Barbui T, Thiele J, Vannucchi AM, Tefferi A. Problems and pitfalls regarding WHO- 34. Vannucchi AM, Kiladjian JJ, Griesshammer M et al. Ruxolitinib proves superior to
defined diagnosis of early/prefibrotic primary myelofibrosis versus essential best available therapy in a prospective, randomized, phase 3 study (RESPONSE) in
thrombocythemia. Leukemia 2013; 27: 1953–1958. patients with polycythemia vera resistant to or intolerant of hydroxyurea. EHA
13. Jovanovic JV, Ivey A, Vannucchi AM et al. Establishing optimal quantitative- Annual Meeting 2014; LB2436.
polymerase chain reaction assays for routine diagnosis and tracking of minimal 35. Tefferi A. Polycythemia vera and essential thrombocythemia: 2013 update on
residual disease in JAK2-V617F-associated myeloproliferative neoplasms: a joint diagnosis, risk-stratification, and management. Am J Hematol 2013; 88: 507–516.
European LeukemiaNet/MPN&MPNr-EuroNet (COST action BM0902) study. 36. Alvarez-Larrán A, Cervantes F, Pereira A et al. Observation versus antiplatelet
Leukemia 2013; 27: 2032–2039. therapy as primary prophylaxis for thrombosis in low-risk essential
14. Passamonti F, Thiele J, Girodon F et al. A prognostic model to predict survival in thrombocythemia. Blood 2010; 116: 1205–1210.
867 World Health Organization-defined essential thrombocythemia at diagnosis: a 37. Cortelazzo S, Finazzi G, Ruggeri M et al. Hydroxyurea for patients with essential throm-
study by the International Working Group on Myelofibrosis Research and bocythemia and a high risk of thrombosis. N Engl J Med 1995; 332: 1132–1136.
Treatment. Blood 2012; 120: 1197–1201. 38. Harrison CN, Campbell PJ, Buck G et al. Hydroxyurea compared with anagrelide in
15. Tefferi A, Rumi E, Finazzi G et al. Survival and prognosis among 1545 patients high-risk essential thrombocythemia. N Engl J Med 2005; 353: 33–45.
with contemporary polycythemia vera: an international study. Leukemia 2013; 27: 39. Gisslinger H, Gotic M, Holowiecki J et al. Anagrelide compared to hydroxyurea in
1874–1881. WHO-classified essential thrombocythemia: the ANAHYDRET Study, a randomized
16. Barbui T, Barosi G, Birgegard G et al. Philadelphia-negative classical controlled trial. Blood 2013; 121: 1720–1728.
myeloproliferative neoplasms: critical concepts and management recommendations 40. Cervantes F, Alvarez-Larrán A, Hernández-Boluda JC et al. Erythropoietin
from European LeukemiaNet. J Clin Oncol 2011; 29: 761–770. treatment of the anaemia of myelofibrosis with myeloid metaplasia: results in 20
17. Marchioli R, Finazzi G, Landolfi R et al. Vascular and neoplastic risk in a large patients and review of the literature. Br J Haematol 2004; 127: 399–403.
cohort of patients with polycythemia vera. J Clin Oncol 2005; 23: 2224–2232. 41. Cervantes F, Alvarez-Larrán A, Hernández-Boluda JC et al. Darbepoetin-alpha for
18. Cervantes F, Dupriez B, Pereira A et al. New prognostic scoring system for primary the anaemia of myelofibrosis with myeloid metaplasia. Br J Haematol 2006; 134:
myelofibrosis based on a study of the International Working Group for Myelofibrosis 184–186.
Research and Treatment. Blood 2009; 113: 2895–2901. 42. Cervantes F, Alvarez-Larrán A, Domingo A et al. Efficacy and tolerability of danazol
19. Passamonti F, Cervantes F, Vannucchi AM et al. A dynamic prognostic model to as a treatment for the anaemia of myelofibrosis with myeloid metaplasia: long-
predict survival in primary myelofibrosis: a study by the IWG-MRT (International term results in 30 patients. Br J Haematol 2005; 129: 771–775.
Working Group for Myeloproliferative Neoplasms Research and Treatment). Blood 43. Mesa RA, Steensma DP, Pardanani A et al. A phase 2 trial of combination low-
2010; 115: 1703–1708. dose thalidomide and prednisone for the treatment of myelofibrosis with myeloid
20. Gangat N, Caramazza D, Vaidya R et al. DIPSS Plus: a refined Dynamic metaplasia. Blood 2003; 101: 2534–2541.
International Prognostic Scoring System for primary myelofibrosis that incorporates 44. Mesa RA, Yao X, Cripe LD et al. Lenalidomide and prednisone for myelofibrosis:
prognostic information from karyotype, platelet count, and transfusion status. J Eastern Cooperative Oncology Group (ECOG) phase 2 trial E4903. Blood 2010;
Clin Oncol 2011; 29: 392–397. 116: 4436–4438.
21. Rumi E, Pietra D, Pascutto C et al. Clinical effect of driver mutations of JAK2, 45. Tefferi A, Lasho TL, Mesa RA et al. Lenalidomide therapy in del(5)(q31)-associated
CALR or MPL in primary myelofibrosis. Blood 2014; 124: 1062–1069. myelofibrosis: cytogenetic and JAK2V617F molecular remissions. Leukemia 2007;
22. Vannucchi AM, Lasho TL, Guglielmelli P et al. Mutations and prognosis in primary 21: 1827–1828.
myelofibrosis. Leukemia 2013; 27: 1861–1869. 46. Tefferi A, Mesa RA, Nagorney DM et al. Splenectomy in myelofibrosis with myeloid
23. Hernández-Boluda JC, Pereira A, Gómez M et al. The International Prognostic metaplasia: a single-institution experience with 223 patients. Blood 2000; 95:
Scoring System does not accurately discriminate different risk categories in 2226–2233.
patients with post-essential thrombocythemia and post-polycythemia vera 47. Martinez-Trillos A, Gaya A, Maffioli M et al. Efficacy and tolerability of hydroxyurea
myelofibrosis. Haematologica 2014; 99: e55–e57. in the treatment of the hyperproliferative manifestations of myelofibrosis: results in
24. Marchioli R, Finazzi G, Specchia G et al. Cardiovascular events and intensity of 40 patients. Ann Hematol 2010; 89: 1233–1237.
treatment in polycythemia vera. N Engl J Med 2013; 368: 22–33. 48. Bouabdallah R, Coso D, Gonzague-Casabianca L et al. Safety and efficacy of
25. Landolfi R, Marchioli R, Kutti J et al. Efficacy and safety of low-dose aspirin in splenic irradiation in the treatment of patients with idiopathic myelofibrosis: a
polycythemia vera. N Engl J Med 2004; 350: 114–124. report on 15 patients. Leuk Res 2000; 24: 491–495.
26. Kiladjian JJ, Chevret S, Dosquet C et al. Treatment of polycythemia vera with 49. Steensma DP, Hook CC, Stafford SL, Tefferi A. Low-dose, single-fraction, whole-
hydroxyurea and pipobroman: final results of a randomized trial initiated in 1980. J lung radiotherapy for pulmonary hypertension associated with myelofibrosis with
Clin Oncol 2011; 29: 3907–3913. myeloid metaplasia. Br J Haematol 2002; 118: 813–816.
27. Björkholm M, Derolf AR, Hultcrantz M et al. Treatment-related risk factors for 50. Verstovsek S, Kantarjian H, Mesa RA et al. Safety and efficacy of INCB018424,
transformation to acute myeloid leukemia and myelodysplastic syndromes in a JAK1 and JAK2 inhibitor, in myelofibrosis. N Engl J Med 2010; 363: 1117–1127.
myeloproliferative neoplasms. J Clin Oncol 2011; 29: 2410–2415. 51. Tefferi A, Pardanani A. Serious adverse events during ruxolitinib treatment
28. Kiladjian JJ, Cassinat B, Chevret S et al. Pegylated interferon-alfa-2a induces discontinuation in patients with myelofibrosis. Mayo Clin Proc 2011; 86: 1188–1191.
complete hematologic and molecular responses with low toxicity in polycythemia 52. Verstovsek S, Mesa RA, Gotlib J et al. A double-blind, placebo-controlled trial of
vera. Blood 2008; 112: 3065–3072. ruxolitinib for myelofibrosis. N Engl J Med 2012; 366: 799–807.
29. Quintás-Cardama A, Kantarjian H, Manshouri T et al. Pegylated interferon alfa-2a yields 53. Harrison C, Kiladjian JJ, Al-Ali HK et al. JAK inhibition with ruxolitinib versus best
high rates of hematologic and molecular response in patients with advanced essential available therapy for myelofibrosis. N Engl J Med 2012; 366: 787–798.
thrombocythemia and polycythemia vera. J Clin Oncol 2009; 27: 5418–5424. 54. Cervantes F, Vannucchi AM, Kiladjian JJ et al. Three-year efficacy, safety, and
30. Kiladjian JJ, Mesa RA, Hoffman R. The renaissance of interferon therapy for the survival findings from COMFORT-II, a phase 3 study comparing ruxolitinib with
treatment of myeloid malignancies. Blood 2011; 117: 4706–4715. best available therapy for myelofibrosis. Blood 2013; 122: 4047–4053.
31. Mesa RA, Niblack J, Wadleigh M et al. The burden of fatigue and quality of life in 55. Verstovsek S, Kantarjian HM, Estrov Z et al. Long term outcomes of 107 patients
myeloproliferative disorders (MPDs): an international internet-based survey of with myelofibrosis receiving JAK1/JAK2 inhibitor ruxolitinib: survival advantage in
1179 MPD patients. Cancer 2007; 109: 68–76. comparison to matched historical controls. Blood 2012; 120: 1202–1209.
v | Vannucchi et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
56. Passamonti F, Maffioli M, Cervantes F et al. Impact of ruxolitinib on the natural 67. Barosi G, Besses C, Birgegard G et al. A unified definition of clinical resistance/
history of primary myelofibrosis: a comparison of the DIPSS and the COMFORT-2 intolerance to hydroxyurea in essential thrombocythemia: results of a consensus
cohorts. Blood 2014; 123: 1833–1835. process by an international working group. Leukemia 2007; 21: 277–280.
57. Kröger N, Holler E, Kobbe G et al. Allogeneic stem cell transplantation after 68. Barosi G, Birgegard G, Finazzi G et al. Response criteria for essential
reduced-intensity conditioning in patients with myelofibrosis: a prospective, thrombocythemia and polycythemia vera: result of a European LeukemiaNet
multicenter study of the Chronic Leukemia Working Party of the European Group consensus conference. Blood 2009; 113: 4829–4833.
for Blood and Marrow Transplantation. Blood 2009; 114: 5264–5270. 69. Barosi G, Bordessoule D, Briere J et al. Response criteria for myelofibrosis with
58. Ballen KK, Shrestha S, Sobocinski KA et al. Outcome of transplantation for myeloid metaplasia: results of an initiative of the European Myelofibrosis Network
myelofibrosis. Biol Blood Marrow Transplant 2010; 16: 358–367. (EUMNET). Blood 2005; 106: 2849–2853.
59. Lussana F, Rambaldi A, Finazzi MC et al. Allogeneic hematopoietic stem cell 70. Alvarez-Larrán A, Pereira A, Cervantes F et al. Assessment and prognostic value of
transplantation in patients with polycythemia vera or essential thrombocythemia the European LeukemiaNet criteria for clinicohematologic response, resistance, and
transformed to myelofibrosis or acute myeloid leukemia: a report from the MPN intolerance to hydroxyurea in polycythemia vera. Blood 2012; 119: 1363–1369.
Subcommittee of the Chronic Malignancies Working Party of the European Group 71. Hernández-Boluda JC, Alvarez-Larrán A, Gómez M et al. Clinical evaluation of the
for Blood and Marrow Transplantation. Haematologica 2014; 99: 916–921. European LeukaemiaNet criteria for clinicohaematological response and resistance/
60. Rondelli D, Goldberg JD, Isola L et al. MPD-RC 101 prospective study of reduced intolerance to hydroxycarbamide in essential thrombocythaemia. Br J Haematol
intensity allogeneic hematopoietic stem cell transplantation in patients with 2011; 152: 81–88.
myelofibrosis. Blood 2014; 124: 1183–1191. 72. Kennedy JA, Atenafu EG, Messner HA et al. Treatment outcomes following
61. Barosi G, Mesa R, Finazzi G et al. Revised response criteria for polycythemia vera leukemic transformation in Philadelphia-negative myeloproliferative neoplasms.
and essential thrombocythemia: an ELN and IWG-MRT consensus project. Blood Blood 2013; 121: 2725–2733.
2013; 121: 4778–4781. 73. Alchalby H, Zabelina T, Stubig T et al. Allogeneic stem cell transplantation for
62. Emanuel RM, Dueck AC, Geyer HL et al. Myeloproliferative neoplasm (MPN) myelofibrosis with leukemic transformation: a study from the Myeloproliferative
symptom assessment form total symptom score: prospective international Neoplasm Subcommittee of the CMWP of the European Group for Blood and Marrow
assessment of an abbreviated symptom burden scoring system among patients Transplantation. Biol Blood Marrow Transplant 2014; 20: 279–281.
with MPNs. J Clin Oncol 2012; 30: 4098–4103. 74. Thepot S, Itzykson R, Seegers V et al. Treatment of progression of Philadelphia-
63. Tefferi A, Cervantes F, Mesa R et al. Revised response criteria for myelofibrosis: negative myeloproliferative neoplasms to myelodysplastic syndrome or acute
International Working Group-Myeloproliferative Neoplasms Research and Treatment myeloid leukemia by azacitidine: a report on 54 cases on the behalf of the Groupe
(IWG-MRT) and European LeukemiaNet (ELN) consensus report. Blood 2013; 122: Francophone des Myelodysplasies (GFM). Blood 2010; 116: 3735–3742.
1395–1398. 75. Harrison CN, Robinson SE. Myeloproliferative disorders in pregnancy. Hematol
64. Barbui T, Carobbio A, Cervantes F et al. Thrombosis in primary myelofibrosis: Oncol Clin North Am 2011; 25: 261–275.
incidence and risk factors. Blood 2010; 115: 778–782. 76. Harrison C. Pregnancy and its management in the Philadelphia negative
65. Ellis MH, Lavi N, Vannucchi A, Harrison C. Treatment of thromboembolic events myeloproliferative diseases. Br J Haematol 2005; 129: 293–306.
coincident with the diagnosis of myeloproliferative neoplasms: a physician survey. 77. Ruggeri M, Rodeghiero F, Tosetto A et al. Postsurgery outcomes in patients with
Thromb Res 2014; 134: 251–254. polycythemia vera and essential thrombocythemia: a retrospective survey. Blood
66. Barosi G, Birgegard G, Finazzi G et al. A unified definition of clinical resistance and 2008; 111: 666–671.
intolerance to hydroxycarbamide in polycythaemia vera and primary myelofibrosis: 78. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
results of a European LeukemiaNet (ELN) consensus process. Br J Haematol among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
2010; 148: 961–963. 139–144.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv203 | v

by guest
on 05 February 2018
Primary cutaneous lymphomas: ESMO Clinical Practice

R. Willemze1, E. Hodak2, P. L. Zinzani3, L. Specht4 & M. Ladetto5, on behalf of the ESMO Guidelines
Working Group*
1
Department of Dermatology, Leiden University Medical Centre, Leiden, The Netherlands; 2Department of Dermatology, Rabin Medical Centre, Beilinson Hospital, Petach
Tikva, Israel; 3Institute of Haematology and Medical Oncology, University of Bologna, Bologna, Italy; 4Department of Oncology and Haematology, Rigshospitalet, University
of Copenhagen, Copenhagen, Denmark; 5Department of Molecular Biotechnology and Health Sciences, University of Torino, Torino, Italy,
epidemiology skin or peripheral blood may be a valuable adjunct in selected

cases. However, clinical and histopathological features are, in
Primary cutaneous lymphomas (PCLs) are defined as non- most cases, the most important deciding factors for therapeutic
Hodgkin’s lymphomas that present in the skin with no planning. PCLs should be classified according to the criteria of
evidence of extracutaneous disease at the time of diagnosis. the World Health Organisation–European Organisation for
After the gastrointestinal lymphomas, PCLs are the second Research and Treatment of Cancer (WHO–EORTC)
most common group of extranodal non-Hodgkin’s classification (Table 1) [1].
lymphomas with an estimated annual incidence of
clinical practice
1/100 000 in Western countries. PCLs must be distinguished
guidelines
from nodal or systemic malignant lymphomas involving the staging
skin secondarily, which often have another clinical In all cases, adequate staging should be carried out to exclude
behaviour, have a different prognosis and require a different the presence of extracutaneous disease. Staging includes
therapeutic approach. In recent lymphoma classifications, complete physical examination, complete and differential
PCLs are therefore included as separate entities. Within the blood cell count and serum biochemistry and appropriate
group of PCLs, distinct types of cutaneous T-cell lymphoma imaging studies (computed tomography scans ± [18F]2-fluoro-
(CTCL) and cutaneous B-cell lymphoma (CBCL) can be 2-deoxy-D-glucose–positron emission tomography scans in all
distinguished [1, 2]. In the western world, CTCL constitutes but stage IA), although they are not required in patients with
∼75%–80% of all PCLs, with mycosis fungoides (MF) as the lymphomatoid papulosis (LyP) [5, 6]. Flow cytometry of the
most common type of CTCL, and CBCL ∼20%–25% [1]. peripheral blood should only be carried out in selected cases,
However, different distributions have been observed in other but is mandatory in patients with (suspected) Sézary syndrome
parts of the world. In southeast Asian countries, CTCLs (SS). Bone marrow biopsy and aspiration should be carried out
other than MF, in particular Epstein-Barr virus-associated in cutaneous lymphomas with an intermediate or aggressive
natural killer/T-cell lymphomas, are much more common clinical behaviour, but is not required in cutaneous
than in Western countries, while CBCLs are much more lymphomas with an indolent clinical behaviour (MF,
uncommon [3, 4]. cutaneous anaplastic large-cell lymphoma and cutaneous
marginal zone lymphoma), unless indicated by other staging
diagnosis assessments [5, 6]. The significance of bone marrow
examination in primary cutaneous follicle centre lymphoma
The diagnosis and classification of PCLs should always be based (PCFCL) is controversial [6, 7]. Prognosis is extremely variable
on a combination of clinical, histological and depending on the type of PCL and the stage of disease. For
immunophenotypical data. Demonstration of clonal T-cell clinical staging of MF and SS, the revised International Society
receptor or immunoglobulin gene rearrangements in lesional for Cutaneous Lymphomas/European Organization of
Research and Treatment of Cancer (ISCL/EORTC) TNMB
(tumour–node–metastasis-blood) staging system should be
L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland; used (Tables 2 and 3) [5]. For PCL other than MF/SS, a
E-mail: clinicalguidelines@esmo.org separate ISCL/EORTC TNM classification system has been
†
published [6]. This staging system is primarily intended to
Approved by the ESMO Guidelines Working Group: December 2006, last update June
2013. This publication supersedes the previously published version. Ann Oncol 2010; 21
document the extent of disease and cannot be used as a
(Suppl. 5): v177–v180. prognostic guide.

by guest
on 05 February 2018
Table 1. World Health Organisation–European Organisation for Table 2. Revised TNMB classification of mycosis fungoides (MF) and
Research and Treatment of Cancer (WHO–EORTC) classification Sézary syndrome (SS)
Cutaneous T-cell lymphoma (CTCL) T (skin)

Mycosis fungoides (MF) T1 Limited patch/plaque (involving <10% of total skin surface)
Variants of MF T2 Generalised patch/plaque (involving ≥10% of total skin surface)
Folliculotropic MF T3 Tumour(s)
Pagetoid reticulosis T4 Erythroderma
Granulomatous slack skin
N (lymph node)
Sézary syndrome (SS)
N0 No clinically abnormal peripheral lymph nodes
Primary cutaneous CD30-positive lymphoproliferative disorders
N1 Clinically abnormal peripheral lymph nodes; histologically
Primary cutaneous anaplastic large-cell lymphoma
uninvolved
Lymphomatoid papulosis
N2 Clinically abnormal peripheral lymph nodes; histologically involved
Subcutaneous panniculitis-like T-cell lymphoma
(nodal architecture uneffaced)
Extranodal natural killer (NK)/T-cell lymphoma, nasal-type
N3 Clinically abnormal peripheral lymph nodes; histologically involved
Primary cutaneous peripheral T-cell lymphoma-not otherwise specified
(nodal architecture (partially) effaced)
Aggressive epidermotropic CD8+ CTCLa
Nx Clinically abnormal peripheral lymph nodes; no histological
Cutaneous γ/δ T-cell lymphomaa
confirmation
CD4+ small/medium-sized pleomorphic CTCLa
Cutaneous B-cell lymphoma (CBCL) M (viscera)
Primary cutaneous marginal zone B-cell lymphoma M0 No visceral involvement
Primary cutaneous follicle centre lymphoma M1 Visceral involvement
Primary cutaneous diffuse large B-cell lymphoma, leg type
B (blood)
a
Provisional entities. B0 No circulating atypical (Sézary) cells (or <5% of lymphocytes)
Republished with permission of the American Society of Hematology, B1 Low blood tumour burden (≥5% of lymphocytes are Sézary cells,
from Willemze R, Jaffe ES, Burg G et al., WHO/EORTC classification but not B2)
for cutaneous lymphomas Blood 2005; 105: 3768–3785; permission B2 High blood tumour burden (≥1000/µl Sézary cells and positive
conveyed through Copyright Clearance Center, Inc. clone)
Republished with permission of American Society of Hematology, from

Olsen E, Vonderheid E, Pimpinelli N et al., Revisions to the staging and
therapy classification of mycosis fungoides and Sézary syndrome: a proposal of the
International Society for Cutaneous Lymphomas (ISCL) and the Cutaneous
The choice of treatment depends on the type of PCL and the Lymphoma Task Force of the European Organization of Research and
stage of disease. Due to their heterogeneity and rarity, controlled Treatment of Cancer (EORTC), Blood 2007; 110: 1713–1722; permission
clinical trials in PCLs are almost non-existent, with a few conveyed through Copyright Clearance Center, Inc.
exceptions mainly concerning recently marketed drugs.
Recommendations are therefore largely based on (retrospective)
cohort studies and expert opinions discussed during consensus
meetings of the EORTC Cutaneous Lymphoma Group, the
International Society for Cutaneous Lymphomas (ISCL), the
United States Cutaneous Lymphoma Consortium (USCLC) and Table 3. Revised clinical staging system for mycosis fungoides (MF) and
the International Lymphoma Radiation Oncology Group. Sézary syndrome (SS)
Consensus recommendations for clinical end points and
response criteria in MF/SS and in PCLs other than MF/SS have Clinical stage
recently been published (Olsen EA, Willemze R, Wood GS et al. IA T1 N0 M0 B0-1
Clinical endpoints and response criteria in primary cutaneous IB T2 N0 M0 B0-1
lymphomas other than mycosis fungoides and Sézary IIA T1–2 N1-2 M0 B0-1
syndrome: a Consensus Statement of the International Society IIB T3 N0-2 M0 B0-1
for Cutaneous Lymphomas (ISCL), the United States III T4 N0-2 M0 B0-1
Cutaneous Lymphoma Consortium (USCLC) and the IVA1 T1-4 N0-2 M0 B2
Cutaneous Lymphoma Task Force of the European IVA2 T1-4 N3 M0 B0-2
Organization for the Research and Treatment of Cancer IVB T1-4 N0-3 M1 B0-2
(EORTC). Blood. In preparation.) [8].
Republished with permission of American Society of Hematology, from
Olsen E, Vonderheid E, Pimpinelli N et al., Revisions to the staging and
classification of mycosis fungoides and Sézary syndrome: a proposal of the
mycosis fungoides and variants
International Society for Cutaneous Lymphomas (ISCL) and the Cutaneous
Since early aggressive chemotherapy is associated with Lymphoma Task Force of the European Organization of Research and
considerable side-effects but does not improve survival, a stage- Treatment of Cancer (EORTC), Blood 2007; 110: 1713–1722; permission
adapted conservative therapeutic approach is recommended for conveyed through Copyright Clearance Center, Inc.
vi | Willemze et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
MF and its variants [9–12]. Patients with only patches and/or suggested as the treatment of choice in SS and erythrodermic
plaques covering <10% (stage IA) or ≥10% of the skin surface MF [10–12]. Overall response rates range from 30% to 80% with
(stage IB) should be treated with skin-directed therapies, complete response rates ranging from 14% to 25%, depending
including topical steroids, psoralens + ultraviolet A (PUVA), on the ECP regimen and the type of combination used.
narrow-band ultraviolet B (UVB) and topical cytostatic agents, However, the suggested superiority of ECP over the traditional
such as mechlorethamine or carmustine (BCNU). Narrow-band low-dose chemotherapy regimens has not yet been substantiated
UVB should only be used in patients with patches or very thin by randomised, controlled trials [21]. Prolonged treatment with
plaques. Topical steroids can be recommended as monotherapy a combination of low-dose chlorambucil and prednisone is
for patches/flat plaques stage IA disease. In stage IB, topical often effective in controlling the disease, but is unlikely to yield
steroids can be used as adjuvant therapy for selected skin complete responses. Low-dose methotrexate, bexarotene,
lesions. In patients developing one or few infiltrated plaques or denileukin diftitox, alemtuzumab (low-dose) and multiagent
tumours (stage IIB), additional local radiotherapy may suffice. chemotherapy have been recommended as second-line
Local radiotherapy can be curative in patients with early treatment of SS [10–12, 22]. It should be emphasised that
localised disease, particularly in patients with unilesional MF comparison of treatment results in the different studies is almost
and pagetoid reticulosis. Local radiotherapy is most commonly impossible due to differences in diagnostic criteria used for SS.
administered with electrons (energy dependent on the thickness
of the lesion), with bolus to achieve full skin dose, a margin of
≥2 cm and a total dose of 24–36 Gy. For patients with more primary cutaneous CD30-positive
extensive infiltrated plaques and tumours or patients refractory lymphoproliferative disorders (LPDs)
to skin-directed therapies, a combination of PUVA and
The group of primary cutaneous CD30-positive LPDs includes
interferon alpha or PUVA and retinoids (including bexarotene),
primary cutaneous anaplastic large lymphoma (C-ALCL) and
a combination of interferon alpha and retinoids or total skin
LyP, which form a spectrum of disease. Both C-ALCL and LyP
electron beam irradiation can be considered. However, these
have an excellent prognosis, with a 10-year survival of 90% and
treatments are not curative, although remission may last for
almost 100%, respectively [23, 24]. Patients with C-ALCL
several years. Total skin electron beam irradiation was often
generally present with solitary or localised (ulcerating) tumours
given to total doses of 30–36 Gy, but recently lower doses
or nodules and should be treated with radiotherapy or surgical
(10–12 Gy) have been employed with the advantages of shorter
excision. Patients with C-ALCL presenting with multifocal skin
duration of the treatment period, fewer side-effects and
lesions can be best treated with low-dose methotrexate, as in
opportunity for re-treatment [13]. In patients with advanced
LyP, or radiotherapy in case of only a few lesions. Radiotherapy
and refractory disease, gemcitabine or liposomal doxorubicin
is commonly administered with electrons, with bolus, a margin
may be considered [14, 15]. Other agents like the fusion toxin
of ≥2 cm and a total dose of 40 Gy [25]. This dose is effective
denileukin diftitox and histone deacetylase (HDAC) inhibitors,
and well-tolerated. Lower doses may achieve the same result, but
such as vorinostat and romidepsin, have been approved in the
data have not been published. In patients with multiple lesions,
United States by the Food and Drug Administration (FDA) for
lower doses of radiation may be used for palliation. In cases not
patients with relapsed and refractory CTCL, but have not yet
responsive to these treatments, systemic retinoids including
been registered for CTCL in Europe [16–18]. Multiagent
bexarotene or interferon alpha can be used [24]. Recent
chemotherapy is only indicated in patients with effaced lymph
preliminary studies report high response rates of brentuximab
nodes or visceral involvement (stage IV), or in patients with
vedotin (anti-CD30 monoclonal antibody coupled to the anti-
widespread tumour stage MF which cannot be controlled with
tubulin agent monomethyl auristatin E) in patients with C-
skin-targeted and immunomodulating therapies. Local
ALCL as well as patients with MF expressing CD30, but
palliation of cutaneous as well as extracutaneous lesions may be
controlled clinical trials have just started [26, 27]. Multiagent
achieved with local radiotherapy to doses ≥8 Gy [19]. In
chemotherapy is only indicated in patients presenting with or
relatively young patients with refractory, progressive MF or with
developing extracutaneous disease and in rare patients with
SS, allogeneic stem cell transplantation may be considered.
rapidly progressive skin disease.
Durable responses have been reported, but experience is still
limited, and the optimal conditioning regimen and the optimal
timing for an allogeneic transplant are currently unknown [20]. subcutaneous panniculitis-like T-cell
Results with autologous stem cell transplantation in MF and SS
have been disappointing.
lymphoma (SPTCL)
The term SPTCL is only used for cases with an α/β T-cell
phenotype, which have a favourable prognosis, particularly if
not associated with a haemophagocytic syndrome (HPS), which
Sézary syndrome is frequently an extremely aggressive clinical syndrome
Being a systemic disease (i.e. leukaemia) by definition, systemic requiring immediate intervention. One study reported 5-year
treatment is required. Skin-directed therapies like PUVA or overall survival rates of 91% and 46% in SPTCL patients without
potent topical steroids may be used as adjuvant therapy. and with an HPS, respectively [28]. In SPTCL without
Extracorporeal photopheresis (ECP), either alone or in associated HPS, systemic steroids or other immunosuppressive
combination with other treatment modalities such as interferon agents should be considered first, whereas in cases of solitary or
alpha, retinoids, total skin electron beam and PUVA, has been localised skin lesions, radiotherapy with electrons is advised.
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt242 | vi

by guest
on 05 February 2018
Table 4. Recommendations for the initial management of CBCL
Extent First-line therapy Alternative therapies

PCMZL Solitary/localised Local radiotherapy IFN alpha i.l.
Excision Rituximab i.l.
(Antibiotics)a Intralesional steroids
Multifocal Wait and see IFN alpha i.l.
Local radiotherapy Rituximab i.l.
Chlorambucilb Topical or intralesional steroids
Rituximab i.v.
(Antibiotics)a
PCFCL Solitary/localised Local radiotherapy IFN alpha i.l.

Excision Rituximab i.l.
Multifocal Wait and see R-CVP/CHOPc

Local radiotherapy
Rituximab i.v.
PCLBCL, LT Solitary/localised R-CHOP ± IFRT Local radiotherapy

Rituximab i.v.
Multifocal R-CHOP Rituximab i.v.
a
In case of evidence for Borrelia burgdorferi infection,
b
or other single or combination regimens appropriate for low-grade B-cell lymphomas,
c
in exceptional cases or for patients developing extracutaneous disease i.l. intralesional; i.v.: intravenous; IFRT: involved field radiotherapy.
Republished with permission of American Society of Hematology, from Senff NJ, Noordijk EM, Kim YH et al., European Organization for Research and
Treatment of Cancer and International Society for Cutaneous Lymphoma consensus recommendations for the management of cutaneous B-cell lymphomas,
Blood 2008; 112: 1600–1609; permission conveyed through Copyright Clearance Center, Inc.
Little information on radiation dose is available, but a dose of more advanced disease, these lymphomas show an aggressive
40 Gy has been used. Bexarotene may be also effective in SPTCL clinical behaviour and are often resistant to chemotherapy.
[29]. Multiagent chemotherapy is required only in cases with Recently, an intensive chemotherapy regimen including
progressive disease not responding to immunosuppressive L-asparaginase (the SMILE regimen) was shown to be
therapy or in cases with HPS. effective [31].
extranodal natural killer (NK)/T-cell primary cutaneous peripheral T-cell

lymphoma, nasal type lymphoma-not otherwise specified
(PTCL-NOS)
Extranodal NK/T-cell lymphoma, nasal type is a rare, nearly
always Epstein-Barr virus-positive lymphoma, which is more Within the group of primary cutaneous PTCL-NOS, three
common in Asia and Central and South America. The skin is somewhat better defined subgroups have been included as
the second most common site of involvement after the nasal provisional entities (see Table 1). However, most of the cases
cavity/nasopharynx. Patients generally present with multiple have in common a generally aggressive clinical course and poor
(ulcerating) plaques and tumours, or in the case of nasal survival, and should therefore be treated as systemic PTCL-NOS
NK/T-cell lymphoma with a midfacial destructive tumour. Skin with multiagent chemotherapy. Since the results are often
involvement may be a primary or secondary manifestation of disappointing, early allogeneic stem cell transplantation may be
the disease. One study reported a median survival of 27 months considered. The only exception is the group of CD4-positive
for patients presenting with only skin lesions, compared with small-medium pleomorphic CTCL. These patients often present
5 months for patients presenting with cutaneous and with a solitary tumour, most commonly on the head, should be
extracutaneous diseases [30]. Since both groups have an treated with local radiotherapy or excision, and have an
aggressive clinical behaviour, distinction between primary and excellent prognosis.
secondary cutaneous involvement seems not to be useful for
this category. In patients with stage I disease, radiotherapy is the
first choice of treatment. Recommended radiation doses are
cutaneous B-cell lymphoma
higher than for other lymphomas, with 50 Gy to the initial In the WHO–EORTC classification, three main types of CBCL
lesion and a boost of 5–10 Gy to residual disease. In the case of are distinguished: primary cutaneous marginal zone lymphoma
vi | Willemze et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
Table 5. Levels of evidence and grades of recommendation (adapted from (histology, blood examination, imaging, etc.) should only be
the Infectious Diseases Society of America-United States Public Health carried out if required.
Levels of evidence
note
I Evidence from at least one large randomised, controlled trial of good According to the levels of evidence and grades of
methodological quality (low potential for bias) or meta-analyses of recommendation shown in Table 5, the levels of evidence in
well-conducted, randomised trials without heterogeneity these guidelines are mostly level IV and the recommendations
II Small randomised trials or large randomised trials with a suspicion are grade B. This is due to the heterogeneity and rarity of the
of bias (lower methodological quality) or meta-analyses of such diseases.
IV Retrospective cohort studies or case–control studies conflict of interest
Dr Ladetto has reported speaker’s bureau from Celgene,
Grades of recommendation Janssen-Cilag, Roche, Bayer, Amgen, Mundipharma; research
A Strong evidence for efficacy with a substantial clinical benefit, contracts from Celgene, Pfizer, Mundipharma, Roche; funds
strongly recommended received from Amgen, Roche, Italfarmaco. The other authors
B Strong or moderate evidence for efficacy but with a limited clinical have declared no potential conflicts of interest.
benefit, generally recommended
C Insufficient evidence for efficacy or benefit does not outweigh the
risk or the disadvantages (adverse events, costs, ...), optional references
D Moderate evidence against efficacy or for adverse outcome, generally
not recommended 1. Willemze R, Jaffe ES, Burg G et al. WHO–EORTC classification for cutaneous
lymphomas. Blood 2005; 105: 3768–3785.
2. Swerdlow SH, Campo E, Harris NL et al. WHO Classification of Tumours
recommended
Haematopoietic and Lymphoid Tissues, 4th Edition, Geneva: WHO Press,
a
Dykewicz CA. Summary of the guidelines for preventing opportunistic 2008.
infections among hematopoietic stem cell transplant recipients. Clin Infect 3. Tan SH, Sim CS, Ong BH. Cutaneous lymphomas other than mycosis fungoides in
Singapore: a clinicopathological analysis using recent classification systems. Br J
Dis 2001; 33: 139–144. By permission of the Infectious Diseases Society of
Dermatol 2003; 149: 542–553.
America.
4. Park JH, Shin HT, Lee DY et al. World Health Organization–European Organization
for Research and Treatment of Cancer classification of cutaneous lymphoma in
Korea: a retrospective study at a single tertiary institution. J Am Acad Dermatol
2012; 67: 1200–1209.
(PCMZL), PCFCL and primary cutaneous diffuse large B-cell 5. Olsen EA, Vonderheid E, Pimpinelli N et al. Revisions to the staging and
lymphoma, leg type (PCLBCL-LT). PCMZL and PCFCL are classification of mycosis fungoides and Sézary syndrome: a proposal of the
indolent types of CBCL with a disease-related 10-year survival International Society for Cutaneous Lymphomas (ISCL) and the Cutaneous
rate of 90%, while PCLBCL-LT has a more unfavourable Lymphoma Task Force of the European Organization of Research and Treatment of
prognosis (disease-related 5-year survival, ∼50%). EORTC/ISCL Cancer (EORTC). Blood 2007; 110: 1713–1722.
consensus recommendations for the management of these three 6. Kim YH, Willemze R, Pimpinelli N et al. TNM classification system for primary
cutaneous lymphomas other than mycosis fungoides and Sézary syndrome: a
types of CBCL have been formulated and are summarised in
proposal of the International Society for Cutaneous Lymphomas (ISCL) and the
Table 4 [32]. Recommended radiation doses for localised Cutaneous Lymphoma Task Force of the European Organization of Research and
PCMZL and PCFCL are 24–36 Gy, whereas for palliative Treatment of Cancer (EORTC). Blood 2007; 110: 479–484.
treatment of multifocal disease, low-dose radiation (4 Gy) is 7. Senff NJ, Kluin-Nelemans HC, Willemze R. Results of bone marrow examination in
often sufficient [19]. For the more aggressive PCLBCL-LT, a 275 patients with histological features that suggest an indolent type of cutaneous
radiation dose of 40 Gy is recommended. B-cell lymphoma. Br J Haematol 2008; 142: 52–56.
8. Olsen EA, Whittaker S, Kim YH et al. Clinical endpoints and response criteria in
mycosis fungoides and Sézary syndrome: a consensus statement of the
personalised medicine International Society for Cutaneous Lymphomas (ISCL), the United States
Cutaneous Lymphoma Consortium (USCLC) and the Cutaneous Lymphoma Task
In this disease setting, more research is needed to identify Force of the European Organization for Research and Treatment of Cancer
molecular markers which could lead to advances in (EORTC). J Clin Oncol 2011; 29: 2598–2607.
personalised medicine. 9. Kaye FJ, Bunn PA Jr., Steinberg SM et al. A randomized trial comparing
combination electron-beam radiation and chemotherapy with topical therapy
in the initial treatment of mycosis fungoides. N Eng J Med 1989; 321:
follow-up 1784–1790.
The frequency of follow-up visits depends on the type of PCL 10. Whittaker SJ, Marsden JR, Spittle M et al. Joint British Association of
Dermatologists and UK Cutaneous Lymphoma Group guidelines for the
and the stage of disease. It may vary from every 6 or 12 months
management of primary cutaneous T-cell lymphomas. Br J Dermatol 2003;
in patients with indolent types of PCL and stable disease or 149: 1095–1107.
patients in complete remission to every 4–6 weeks in patients 11. Trautinger F, Knobler R, Willemze R et al. EORTC consensus recommendations for
with active or progressive disease. Follow-up visits should focus the treatment of mycosis fungoides/Sézary syndrome. Eur J Cancer 2006; 42:
on history and physical examination, and additional testing 1014–1030.
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt242 | vi

by guest
on 05 February 2018
12. Prince HM, Whittaker S, Hoppe RT. How I treat mycosis fungoides and Sézary Cutaneous Lymphoma Group on the long-term follow-up data of 219 patients and
syndrome. Blood 2009; 114: 4337–4353. guidelines for diagnosis and treatment. Blood 2000; 95: 3653–3661.
13. Kamstrup MR, Lindahl LM, Gniadecki R et al. Low-dose total skin electron beam 24. Kempf W, Pfaltz K, Vermeer MH et al. EORTC, ISCL, USCLC consensus
therapy as a debulking agent for cutaneous T-cell lymphoma: an open-label recommendations for the treatment of primary cutaneous CD30-positive
prospective phase II study. Br J Dermatol 2012; 166: 399–404. lymphoproliferative disorders: lymphomatoid papulosis and primary cutaneous
14. Marchi E, Alinari L, Tani M et al. Gemcitabine as frontline treatment for cutaneous anaplastic large-cell lymphoma. Blood 2011; 118: 4024–4035.
T-cell lymphoma: phase II study of 32 patients. Cancer 2005; 104: 2437–2441. 25. Yu JB, McNiff JM, Lund MW, Wilson LD. Treatment of primary cutaneous CD30+
15. Dummer R, Quaglino P, Becker JC et al. Prospective international multicenter anaplastic large-cell lymphoma with radiation therapy. Int J Radiat Oncol Biol Phys
phase II trial of intravenous pegylated liposomal doxorubicin monochemotherapy in 2008; 70: 1542–1545.
patients with stage IIB, IVA, or IVB advanced mycosis fungoides: final results from 26. Pro B, Advani R, Brice P et al. Brentuximab vedotin (SGN-35) in patients with
EORTC 21012. J Clin Oncol 2012; 30: 4091–4097. relapsed or refractory systemic anaplastic large-cell lymphoma: results of a phase
16. Prince HM, Duvic M, Martin A et al. Phase III placebo-controlled trial of denileukin II study. J Clin Oncol 2012; 30: 2190–2196.
diftitox for patients with cutaneous T-cell lymphoma. J Clin Oncol 2010; 28: 27. Krathen M, Sundram U, Bashey S et al. Brentuximab vedotin demonstrates
1870–1877. significant clinical activity in relapsed or refractory mycosis fungoides with
17. Olsen EA, Kim YH, Kuzel TM et al. Phase IIb multicenter trial of vorinostat in variable CD30 expression. Blood ASH Annual Meeting Abstracts 2012;
patients with persistent, progressive, or treatment refractory cutaneous T-cell 120: 797.
lymphoma. J Clin Oncol 2007; 25: 3109–3115. 28. Willemze R, Jansen PM, Cerroni L et al. Subcutaneous panniculitis-like
18. Whittaker SJ, Demierre MF, Kim EJ et al. Final results from a multicenter, T-cell lymphoma: definition, classification and prognostic factors. An
international, pivotal study of romidepsin in refractory cutaneous T-cell lymphoma. EORTC Cutaneous Lymphoma Group study of 83 cases. Blood 2008; 111:
J Clin Oncol 2010; 28: 4485–4491. 838–845.
19. Neelis KJ, Schimmel EC, Vermeer MH et al. Low-dose palliative radiotherapy for 29. Mehta N, Wayne AS, Kim YH et al. Bexarotene is active against subcutaneous
cutaneous B- and T-cell lymphomas. Int J Radiat Oncol Biol Phys 2009; 74: panniculitis-like T-cell lymphoma in adult and pediatric populations. Clin
154–158. Lymphoma Myeloma Leuk 2012; 12: 20–25.
20. Duarte RF, Schmitz N, Servitje O, Sureda A. Haematopoietic stem cell 30. Bekkenk MW, Jansen PM, Meijer CJ et al. CD56+ hematological neoplasms
transplantation for patients with primary cutaneous T-cell lymphoma. Bone Marrow presenting in the skin: a retrospective analysis of 23 new cases and 130 cases
Transplant 2008; 41: 597–604. from the literature. Ann Oncol 2004; 15: 1097–1108.
21. Russell-Jones R. Extracorporeal photopheresis in cutaneous T-cell lymphoma. 31. Yamaguchi M, Kwong YL, Kim WS et al. Phase II study of SMILE chemotherapy
Inconsistent data underline the need for randomized studies. Br J Dermatol 2000; for newly diagnosed stage IV, relapsed, or refractory extranodal natural killer
142: 16–21. (NK)/T-cell lymphoma, nasal type: the NK-cell Tumor Study Group study. J Clin
22. Bernengo MG, Quaglino P, Comessatti A et al. Low-dose intermittent alemtuzumab Oncol 2011; 29: 4410–4416.
in the treatment of Sézary syndrome: clinical and immunologic findings in 14 32. Senff NJ, Noordijk EM, Kim YH et al. European Organization for Research
patients. Haematologica 2007; 92: 784–794. and Treatment of Cancer and International Society for Cutaneous
23. Bekkenk MW, Geelen FA, van Voorst Vader PC et al. Primary and secondary Lymphoma consensus recommendations for the management of cutaneous B-cell
cutaneous CD30(+) lymphoproliferative disorders: a report from the Dutch lymphomas. Blood 2008; 112: 1600–1609.
vi | Willemze et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
SPECIAL ARTICLE
ESMO Consensus Conference on malignant

lymphoma: general perspectives and
recommendations for the clinical management
of the elderly patient with malignant lymphoma
C. Buske1*, M. Hutchings2, M. Ladetto3, V. Goede4, U. Mey5, P. Soubeyran6, M. Spina7, R. Stauder8,

M. Trneny9, U. Wedding10, P. Fields11 & The ESMO Lymphoma Consensus Conference Panel Members†
1
Comprehensive Cancer Center Ulm and Department of Internal Medicine III, Institute of Experimental Cancer Research, University Hospital, Ulm, Germany;
2
Department of Haematology, Rigshospitalet, Copenhagen, Denmark; 3Hematology Division, Azienda Ospedaliera Santi Antonio e Biagio e Cesare Arrigo,
Alessandria, Italy; 4Department of Internal Medicine, University Hospital Cologne, Cologne, Germany; 5Department of Oncology and Haematology, Kantonsspital
Graubünden, Chur, Switzerland; 6Department of Medical Oncology, Institut Bergonié, Comprehensive Cancer Centre, Bordeaux, France; 7Division of Medical
Oncology A, National Cancer Institute, Aviano, Italy; 8Haematology and Oncology Department, Innsbruck Medical University, Innsbruck, Austria; 9Institute of
Hematology and Blood Transfusion, Ist Department of Medicine, 1st Faculty of Medicine, Charles University General Hospital, Prague, Czech Republic;
10
Department of Palliative Care, University Hospital, Jena, Germany; 11Department of Haematology, Guys and St Thomas’ and King’s College Hospitals, London, UK
*Correspondence to: Prof. Christian Buske, ESMO Guidelines Committee, ESMO Head Office, Via L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland.
†
See Appendix for members of the ESMO Lymphoma Consensus Conference.
The European Society for Medical Oncology (ESMO) consensus conference on mature B cell lymphomas and chronic lymphocytic
leukaemia (CLL) was held on 20 June 2015 in Lugano, Switzerland, and included a multidisciplinary panel of 25 leading experts. The
aim of the conference was to develop recommendations on critical subjects difficult to consider in detail in the ESMO Clinical
Practice Guidelines. The following areas were identified: (1) the elderly patient, (2) prognostic factors suitable for clinical use, and (3)
the ‘ultra-high-risk’ group. Before the conference, the expert panel was divided into three working groups; each group focused on
one of these areas in order to address clinically-relevant questions relating to that topic. All relevant scientific literature, as identified
by the experts, was reviewed in advance. During the consensus conference, each working group developed recommendations to
address each of the four questions assigned to their group. These recommendations were presented to the entire panel and a
consensus was reached. This consensus, which was further developed in continuous post-meeting discussions, formed the basis of
three manuscripts, each covering one of the three key areas identified. This manuscript presents the consensus recommendations
regarding the clinical management of elderly patients diagnosed with malignant lymphoma. Four clinically-relevant topics
identified by the panel were: 1) how to define patient fitness, 2) assessing quality of life, 3) diagnostic work-up and 4) clinical
management of elderly patients with lymphoma. Each of these key topics is addressed in the context of five different lymphoma
entities, namely: CLL, follicular lymphoma, mantle cell lymphoma, peripheral T-cell lymphoma and diffuse large B-cell lymphoma.
Results, including a summary of evidence supporting each recommendation, are detailed in this manuscript.
Key words: lymphoma, consensus, elderly patient, quality of life, diagnosis, treatment
Introduction
Western industrial countries, as well as developing countries, are old; specifically, 4.4% were 70–74, 3.8% were 75–79 and 5.3%
facing dramatic demographic changes in the near future, with an were 80 years old [1]. Moreover, by 2060, it is estimated that the
increasing proportion of elderly persons in these societies. In 2015, proportion of people 65 years old will rise to 28% and the propor-
18.9% of the population in the European Union were 65 years tion who are 80 years old will rise to 12% [2]. Consequently, the
V
Downloaded from https://academic.oup.com/annonc/advance-article-abstract/doi/10.1093/annonc/mdx413/4675064

by guest
on 05 February 2018
number of elderly patients with haematological malignancies A literature review was conducted by each working group before
will rise continuously and dramatically since they are typical the consensus conference, with each group responsible for compil-
diseases of the elderly, with a median age at initial diagnosis ing a summary of relevant information required to develop recom-
of >70 years for the most common lymphoma subtypes, such as mendations relating to each of their questions at the conference.
diffuse large B-cell lymphoma (DLBCL) or chronic lymphocytic No systematic literature search was undertaken. During the confer-
leukaemia (CLL) [3, 4]. On the other hand, the number of ence, in parallel sessions, the three working groups discussed and
treatment options has increased dramatically over the past agreed on recommendations relating to each of their assigned
years, ranging from best supportive care to haematopoietic questions. The level of evidence and strength of each recommenda-
stem cell transplantation. In more recent years, well-tolerated tion were also noted, which were defined based on the ‘Infectious
and effective cytostatic drugs as well as immuno-modulatory Diseases Society of America-United States Public Health Service
drugs, monoclonal antibodies and small molecules with favour- Grading System’, as shown in Table 1 [8]. Recommendations from
able efficacy profiles have been developed. Based on the com- each group were then presented to the entire panel of experts,
plexity of elderly patients with malignant lymphoma and the where they were discussed and modified as required. Finally, a vote
lack of evidence-based recommendations regarding treatment was conducted to determine the level of agreement amongst the ex-
algorithms for these patients, the rational integration of this pert panel for each of the recommendations. When necessary,
armamentarium of new drugs remains a challenge. This is par- more recent developments emerging after the consensus confer-
ticularly true given that decision making for elderly patients ence in Lugano in June 2015 were taken into account when finalis-
with lymphoma depends on aspects of the ageing process including this consensus manuscript.
ing comorbidities, malnutrition and/or impairments in functional For working group 1, four topics were identified for discussion
capacities. As such, at this present time, evidence-based treatment in terms of their potential suitability to guide physicians in the
algorithms and recommendations for elderly patients with lymph- management of elderly patients with lymphoma. As such, the
oma are rare. This is also due to the lack of integration of elderly pa- following topics were considered:
tients into clinical trials. Indeed, clinical trials are typically limited
1. Assessing fitness in elderly patients with lymphoma.
to younger patients and those with an excellent performance status
2. Assessing ‘quality of life’ in elderly patients with lymphoma.
(PS) and a lack of, or only minor, comorbidities. A systematic re-
3. Diagnostic work-up in elderly patients with lymphoma.
view by Hamaker et al. [5] showed that only 5% of trials in haem-
4. Treatment of elderly patients with lymphoma.
atological malignancies focus exclusively on elderly patients, and
69% of trials excluded older patients based on their chronolo-
gical age alone. Furthermore, end points that are particularly
relevant for the elderly, such as improvement or maintenance
of quality of life and functional capacities, are rarely considered Table 1. Levels of evidence and grades of recommendation (adapted from
within clinical trials [5]. the Infectious Diseases Society of America-United States Public Health
Treatment algorithms in a given elderly patient with lymph- Service Grading System [8])
oma should be based on the individual risk and prognosis, with
Levels of evidence
personalised treatment algorithms based on the integration of
age-adjusted models including age-adjusted life expectancy and of good methodological quality (low potential for bias) or
the evaluation of patient status by geriatric assessment [6, 7]. meta-analyses of well-conducted randomised trials without
Based on the huge and continuously increasing numbers of eld- heterogeneity
erly patients with malignant lymphoma, and the lack of recom- II Small randomised trials or large randomised trials with a suspicion
mendations on their treatment so far, the European Society for of bias (lower methodological quality) or meta-analyses of such
Medical Oncology (ESMO) convened a panel of experts in order to trials or of trials with demonstrated heterogeneity
develop consensus-based recommendations regarding the clinical III Prospective cohort studies
management of elderly patients with malignant lymphoma. IV Retrospective cohort studies or case–control studies
Methods Grades of recommendation

A Strong evidence for efficacy with a substantial clinical benefit,
A consensus panel, comprising a multidisciplinary panel of 25 strongly recommended
experts in the management of lymphoma, was convened by ESMO. B Strong or moderate evidence for efficacy but with a limited clin-
Three consensus conference chairs (C. Buske, M. Ladetto, M. ical benefit, generally recommended
Hutchings) were also appointed. The consensus panel was divided C Insufficient evidence for efficacy or benefit does not outweigh
into three working groups, each of which was assigned a specific the risk or the disadvantages (adverse events, costs,. . .),
subject area and a working group chair as follows: optional
D Moderate evidence against efficacy or for adverse outcome,
1. The elderly patient (Chair: C. Buske).
2. Prognostic factors suitable for clinical use (Chair: M. Ladetto).
3. The ‘ultra-high-risk’ group (Chair: M. Hutchings). recommended
The consensus conference was held on 20 June 2015 in Lugano,
Switzerland. Before this consensus conference, clinically relevant By permission of the Infectious Diseases Society of America [8].
questions were identified for each subject area.
2 | Buske et al. Volume 0 | Issue 0 | 2017

by guest
on 05 February 2018
For topics 3 and 4, the following lymphoma entities were life expectancy. In addition, the patient’s aims in life and treat-
discussed: CLL, DLBCL, follicular lymphoma (FL), mantle cell ment preferences are essential. However, a cut-off at which the
lymphoma (MCL) and peripheral T-cell lymphoma (PTCL). likelihood of harm means that a patient should be defined as
Results from the section of the consensus conference dedicated to unfit does not exist. Ziepert et al. [16] identified the following
the above topics related to elderly patients with lymphoma, together factors as contributing to haematological toxicity of CHOP-like
with a summary of evidence supporting each recommendation, protocols: low weight, female gender, poor Eastern Cooperative
are detailed in this article. Topics 3 and 4 are presented together Oncology Group (ECOG) PS, high lactate dehydrogenase (LDH),
and subdivided according to the different lymphoma entities. initial cytopenia and grade 4 haematological toxicity during the
A summary of these recommendations is included in Table 2. first cycle. A geriatric assessment was not included in this trial.
Importantly, these additional recommendations should be read in In addition, the association between objective toxicity [defined
conjunction with the already-published ESMO Clinical Practice according to the National Cancer Institute Common Terminology
Guidelines (CPGs) for the diagnosis, treatment and follow-up of Criteria (NCI-CTC)] and subjective toxicity (defined as toxicity
the aforementioned malignant lymphoma entities [9–13]. compromising quality of life) was not well studied. Current
projects address the topic of patient-reported toxicity [17].
In the field of geriatric oncology, a number of trials have
been conducted to evaluate the association between the results of
Results geriatric assessment and patient outcomes. Some of them
included lymphoma patients, but only a few trials were set up
1. Assessing fitness in elderly patients with especially for elderly patients with malignant lymphoma. The
malignant lymphoma only patient characteristics included in the analysis for prognostic
variables in patients with DLBCL are therefore age and ECOG PS.
The term ‘elderly’ patients is used frequently in the literature Both are included in the new National Comprehensive Cancer
without clearly defining the criteria for ‘elderly’. Although Network International Prognostic Index (NCCN-IPI), with
numerical or chronological age itself is not a good tool to define age of 75 years considered as the most important prognostic
fitness in patients, it is well accepted that age-related changes factor [18]. As patient characteristics determined via geriatric
result in decreased fitness. Increasing age is associated with assessment are not included, it might be that numerical age is just
a higher prevalence of comorbidities, functional decline, a surrogate for other age-related changes, such as comorbidities,
cognitive impairment, depressive mood and dependence in functional decline, cognitive impairment, depressive mood and
activities of daily living (ADL). In addition, increasing age is dependence in ADL. Poor performance according to geriatric
associated with increasing rates of toxicity from treatment and assessment, such as limitations in ADL, has demonstrated prog-
treatment-related mortality. According to data of the DSHNHL nostic importance for survival and toxicity, and might change the
(Deutsche Studiengruppe Hochmaligne Non-Hodgkin-Lymphome) treatment decision. It is also associated with poor health-related
in patients with DLBCL within clinical trials treated with quality of life [19]. However, data to show that assessment-based
cyclophosphamide/doxorubicin/vincristine/prednisone (CHOP) care improves outcome are still missing.
or CHOP-like protocols, the treatment-related mortality In a systematic review, Wildiers et al. [20] showed that the use
increased from 4% in those aged 50–64 years to 20% in those aged of a geriatric assessment results in detection of impairment not
75–79 years when no antibacterial or antiviral prophylaxis was identified in routine assessments (patient history or physical
given [14]. examination). Geriatric assessment was also able to predict severe
Age-related changes occur as a gradual process, but increase treatment-related toxicity and overall survival (OS) in a variety of
substantially in patients of 70 years and older. Therefore, the tumours and treatment settings, and so could help tailor the
authors recommend using the term ‘elderly’ for all patients of treatment choice and intensity in each individual patient [20].
70 years and older, in line with recommendations of the Currently, judgement of fitness of elderly patients with malignant
International Society of Geriatric Oncology (SIOG) and the lymphoma should include some form of a geriatric assessment
European Organisation for Research and Treatment of Cancer [20]. The Task Force for Cancer in the Elderly of the EORTC
(EORTC) [7, 15]. recommends the use of the G8 questionnaire. It is a very simple
Less aggressive treatment may reduce the chances of cure, screening tool, which includes seven mini nutritional assessment
prolongation of life and/or symptom control, whereas too aggres- items and age (<80, 80–85, >85) for a total score ranging from 0
sive treatment may result in treatment-related morbidity and (poor score) to 17 (good score). The task force recommended
mortality, and/or compromised quality of life. Thus, a cornerstone a cut-off value of 14 for an ‘impaired’ reference test score [21].
in the decision-making process is to judge fitness for treatment. In an exploratory analysis of a prospective cohort study of 1435
However, this is not a well-defined term in haematology. In addition, assessable cancer patients accrued before treatment in 23 health-care
willingness to accept impairment by toxicity will depend on the aim facilities, an abnormal G8 score was an independent prognostic
of treatment and the likelihood of reaching this aim. Thus, if the factor of 1-year survival. Importantly, the mean time to complete
chance for cure, or more generally speaking the likelihood of benefit, G8 questionnaire was about 5 min [22]. In a busy clinic, such a
is high, as is the case in DLBCL, the readiness to accept toxicity is screening approach might be preferred [23, 24]. To quantify
high and vice versa. Consideration of fitness has to include: (i) the comorbidity, the Cumulative Illness Rating Scale for Geriatrics
lymphoma, as it defines the potential aim of treatment; (ii) the (CIRS-G) is one of the most widely available scores and has
treatment, as it defines the risk of toxicity; and (iii) the patient, already been used to allocate patients with CLL to less intensive
as the individual characteristics also contribute to toxicity and treatment protocols in a large prospective clinical trial [15, 21,
Volume 0 | Issue 0 | 2017 doi:10.1093/annonc/mdx413 | 3

by guest
on 05 February 2018
Guidelines statement LoE GoR Consensus
1. Assessing fitness in elderly patients with malignant lymphoma

Recommendations:
1.1 The panel suggests that geriatric assessment should be included in the diagnostic process of II B 100% yes (23 voters)
clinical trials in order to assess patient fitness
1.2 The panel suggests that a geriatric assessment is included in the diagnostic process to assess II B 100% yes (23 voters)
patient fitness in routine clinical practise. In cases when geriatric assessment is not possible,
geriatric screening (e.g. G-8) can be carried out
2. Assessing quality of life in elderly patients with malignant lymphoma
Recommendations:
2.1 Quality of life should be considered as a prognostic indicator of survival I A 100% yes (23 voters)
2.2 Quality of life should be included as a major end point in clinical trials in the elderly, either alone II B 100% yes (23 voters)
(e.g. in the palliative setting) or in combination with a survival end point (as a co-primary or
composite end point)
2.3 Other PROs can be considered, including maintenance of functional capacity/dependence, III C 100% yes (23 voters)
either alone or in combination with survival end points. This type of end point is encouraged in
clinical trials but methodological questions remain to be solved. Standardised instruments such
as ADL and IADL are available and their use during treatment in clinical practice is encouraged
3. Diagnostic work-up and treatment of elderly patients with CLL
Recommendations:
3.1 Similar methodology should be used in older compared with younger patients to diagnose V B 100% yes (23 voters)
and to stage CLL (i.e. history taking, physical examination, differential blood count, blood smear
microscopy, flow cytometry of blood)
3.2 Older patients with CLL should to be screened for Del(17p) and/or TP53mut whenever III B 100% yes (23 voters)
treatment is planned
3.3 Clinical judgement supported by geriatric assessments should be used to stratify older patients III B 100% yes (23 voters)
with CLL for fitness and treatment goals (fit versus unfit versus terminally ill)
3.4 Only a small minority of carefully selected and very fit older patients with untreated CLL not I B 100% yes (23 voters)
harbouring Del(17p) or TP53mut might be treated with full-dose FCR
3.5 Fit older patients with untreated CLL not harbouring Del(17p) and TP53mut should be II B 100% yes (23 voters)
evaluated for alternative treatments such as BR or dose-attenuated FCR
3.6 Vulnerable older patients with untreated CLL not harbouring Del(17p) or TP53mut should be I A 100% yes (23 voters)
treated with G-CLB, O-CLB or R-CLB
3.7 Vulnerable older patients with untreated CLL not harbouring Del(17p) or TP53mut may be II B 100% yes (23 voters)
considered for alternative treatments such as BR or dose-attenuated FCR or ibrutinib
3.8 Older patients with untreated CLL harbouring Del(17p) and/or TP53mut should be considered III B 100% yes (23 voters)
for treatment with ibrutinib
3.9 Older patients with relapsed or refractory CLL should be considered for treatment with I A 100% yes (23 voters)
ibrutinib or idelalisib plus rituximab (irrespective of Del[17p] and TP53mut status)
3.10 Older patients with relapsed or refractory CLL not harbouring Del(17p) or TP53mut could be III C 100% yes (23 voters)
evaluated for alternative treatments (e.g. CD20 antibody)
3.11 Older patients with late relapse CLL not harbouring Del(17p) or TP53mut may be evaluated for V B 100% yes (23 voters)
re-administration of chemo-immunotherapy
4. Diagnostic work-up and treatment of elderly patients with FL
Recommendations:
4.1 Elderly patients should be diagnosed based on lymph node histology whenever possible. The V A 100% yes (23 voters)
aim of staging is to discriminate between patients with limited disease and those with advanced
stage disease. Any diagnostics that do not impact on treatment decisions should be avoided,
particularly in terminally ill patients
4.2 Asymptomatic elderly patients should undergo a watch and wait strategy I A 100% yes (23 voters)
4.3 Symptomatic patients with mild symptoms should be offered a chemotherapy-free approach III B 100% yes (23 voters)
such as rituximab single agent, if possible
4.4 For patients with high tumour burden tolerating chemotherapy, a rituximab/chemotherapy regimen I A 100% yes (23 voters)
such as BR is recommended (with bendamustine dose reduction or fewer treatment cycles, if neces-
sary. Be aware of bendamustine-associated infections; consider antibacterial/antiviral prophylaxis)
4.5 For relapsed patients, rituximab/chemotherapy adjusted to the fitness of the patient is also III B 100% yes (23 voters)
standard in the elderly. Rituximab maintenance is optional both first line and in relapse.
Continued

by guest
on 05 February 2018
Table 2. Continued
Idelalisib should be used with caution in relapsed patients not responding to rituximab/chemo-
therapy because of its toxicity profile
5. Diagnostic work-up and treatment of elderly patients with MCL
Recommendations:
5.1 Diagnostic work-up in elderly patients should generally not differ from patients of a younger V A 100% yes (23 voters)
age. Histological confirmation by excisional lymph node biopsy or at least core biopsy is manda-
tory. Detection of cyclin D1 overexpression or chromosomal translocation t(11;14) is essential.
Imaging should include at least a CT scan with iodine contrast of the neck, chest, abdomen and
pelvis. The use of PET-CT imaging is considered optional for fit elderly patients. BM aspirate and
biopsy should be carried out in elderly fit patients, whereas BM examination is not required in
vulnerable or terminally ill patients.
5.2 For elderly fit patients, the following chemo-immunotherapeutic regimens are recommended 100% yes (23 voters)
as the preferred first-line treatment options in routine clinical practice:
1. R-CHOP followed by rituximab maintenance I A
2. BR I A
3. VR-CAP I A
4. R-BAC V B
5.3 For vulnerable elderly patients, dose-adapted chemo-immunotherapeutic regimens are con- V B 100% yes (23 voters)
sidered appropriate. Options include dose-reduced BR, R-CVP or R-CLB
5.4 For vulnerable patients with severe comorbidities, mild chemo-immunotherapeutic regimens V B 100% yes (23 voters)
like R-CLB, (dose-reduced) BR or PEP-C are considered appropriate
5.5 While newer targeted drugs such as ibrutinib and lenalidomide might offer benefits, particularly 100% yes (23 voters)
in vulnerable patients, no data are available from clinical trials for this subset of patients and
therefore clear recommendations cannot be given. For relapsed or refractory disease, treatment
should be adapted to the age and PS of the patient. Besides non-cross-resistant combination
regimens, treatment options include:
1. Ibrutinib II A
2. Lenalidomide 6 rituximab II B
3. Temsirolimus 6 rituximab II B
4. Bortezomib V B
6. Diagnostic work-up and treatment of elderly patients with PTCL
Recommendations:
6.1 The final histological diagnosis requires full analysis and integration of the clinical context and IV A 100% yes (23 voters)
expert haematopathological review
6.2 Whenever possible, patients should be entered into clinical trials. First-line regimes for elderly III B 100% yes (23 voters)
patients should be based on a CHOP induction backbone
6.3 Whenever possible, patients should be entered into clinical trials testing novel agents. However, 100% yes (23 voters)
for elderly relapsed patients considered unsuitable for clinical trials, treatment options include:
1. Salvage chemotherapy with gemcitabine or platinum-containing agents IV C
2. Novel agents such as brentuximab vedotin monotherapy for patients with CD30þ T-cell III B
lymphoma
7. Diagnostic work-up and treatment of elderly patients with DLBCL
Recommendations:
7.1 For patients treated with curative intent, diagnosis should be carried out in an expert haemato- V A 100% yes (23 voters)
pathology laboratory with full diagnostic capabilities (immunophenotypic and molecular) and
staging should be with PET-CT
7.2 Cardiac assessment (LVEF) is required for patients treated with curative intent V A 100% yes (23 voters)
7.3 The IPI score should be calculated I A 100% yes (23 voters)
7.4 A CGA is recommended to guide treatment choice III A 100% yes (23 voters)
7.5 The aim of treatment in fully fit patients who are <80 years old should be curative, with a full- I A 100% yes (23 voters)
dose anthracycline-based regimen preferred. R-CHOP is the recommended first-line treatment
choice
7.6 For fully fit patients who are >80 years old without comorbidities, dose-attenuated R-CHOP may III B 100% yes (23 voters)
be appropriate
7.7 For relapsed fit (no organ dysfunction, PS 0–1, no comorbidities), transplant-eligible patients, II A 100% yes (23 voters)
appropriate salvage treatment with R–DHAP, R-ESHAP, R-ICE or R-GDP is indicated. In the event
of an adequate response, ASCT is recommended
Continued

by guest
on 05 February 2018
Table 2. Continued
7.8 For transplant-ineligible patients, dose attenuated R-DHAP, R-ESHAP, R-ICE, or less intense regi- III B 100% yes (23 voters)
mens such as R-Gem-Ox, are appropriate
7.9 For transplant-ineligible patients, single-agent chemotherapies such as bendamustine or pixan- I (pixantrone) C 100% yes (23 voters)
trone may be considered II (bendamustine)
ADL, activities of daily living; ASCT, autologous stem cell transplantation; BAC, bendamustine/cytarabine; BM, bone marrow; BR, bendamustine/rituximab;
CGA, comprehensive geriatric assessment; CHOP, cyclophosphamide/doxorubicin/vincristine/prednisone; CLB, chlorambucil; CLL, chronic lymphocytic
leukaemia; CT, computed tomography; CVP, cyclophosphamide/vincristine/prednisone; DHAP, dexamethasone/high-dose cytarabine/cisplatin; DLBCL,
diffuse large B-cell lymphoma; ESHAP, etoposide/methylprednisolone/cytarabine/cisplatin; FCR, fludarabine/cyclophosphamide/rituximab; FL, follicular
lymphoma; G, obinutuzumab; GDP, gemcitabine/dexamethasone/cisplatin; Gem-Ox, gemcitabine/oxaliplatin; GoR, grade of recommendation; IADL,
instrumental activities of daily living; ICE, ifosfamide/carboplatin/etoposide; IPI, International Prognostic Index; LoE, level of evidence; LVEF, left ventricular
ejection fraction; MCL, mantle cell lymphoma; O, ofatumumab; PEP-C, prednisone/etoposide/procarbazine/cyclophosphamide; PET, positron emission
tomography; PRO, patient-reported outcome; PS, performance status; PTCL, peripheral T-cell lymphoma; R, rituximab; VR-CAP, bortezomib/rituximab/cyclo-
phosphamide/doxorubicin/prednisone.
25]. Thus, it is recommended that prospective clinical trials

Table 3. ESMO criteria to define fitness of patients
should include a geriatric and comorbidity score as an obligatory
component of the diagnostic work-up. This will be an important Category
first step to correlate toxicity and efficacy of a treatment with
fitness in patients with lymphoma in controlled clinical trials. Fit patient Application of the standard treatment, including more
In the end, a precise definition of fitness based on scores of dose-intense approaches, is not associated with an
geriatric and/or comorbidity tests will be difficult to achieve, increased risk of treatment-related or treatment-
since prognosis and treatment-related toxicity differ from disease unrelated AEs compared with a young fit patient
to disease, and as such, the criteria to define fit, vulnerable and Vulnerable A high risk for treatment-related AEs or treatment-
terminally ill patients also varies. Different study groups have patient unrelated AEs when receiving standard treatment.
suggested different criteria, e.g. CLL, myelodysplastic syndrome, Vulnerable patients present a continuum, ranging
acute myeloid leukaemia and multiple myeloma [26–29]. To from those who are just at the border of not being
able to tolerate standard treatment to those who
overcome this, we propose the following ESMO criteria, in which
are close to being considered as terminally ill
‘fitness’ is defined in the context of treatment feasibility with
Terminally Has a short life expectancy (around 3 months only,
three different categories discriminating between ‘fit’, ‘vulner-
ill patient due to the lymphoma or competing lethal comor-
able’ and ‘terminally ill’ patients: (i) for an elderly ‘fit’ patient, it
bidities) and will therefore not benefit from any
would be anticipated that application of the standard treatment,
anti-lymphoma treatment, but only from best
including more dose intense approaches, would not be associated supportive care
with a high risk of treatment-related or treatment-unrelated
adverse events (AEs) compared with a young ‘fit’ patient; in AE, adverse event; ESMO, European Society for Medical Oncology.
contrast (ii) a ‘vulnerable’ patient would carry a high risk
for treatment-related AEs or treatment-unrelated AEs when
receiving standard treatment. Vulnerable patients would present
a continuum, ranging from those who are just at the border of
not being able to tolerate standard treatment to those who Reproducibility and standardisation: Assessment instruments
are close to being considered as terminally ill. Finally, (iii) the are validated, standardised and reproducible.
‘terminally ill’ patient is one with a short life expectancy (around
Clarity of reporting system: In different countries, different scales
three months only due to the lymphoma or competing lethal
[e.g. to assess ADL, or instrumental ADL (IADL), are available].
comorbidities), who will therefore not benefit from any anti-
lymphoma treatment, but from best supportive care only [30]
(Table 3). Prognostic value. Poor performance according to geriatric
assessment is associated with increased toxicity and shorter
survival of elderly patients with malignant lymphoma.
Methodological considerations.
Broad availability: Geriatric screening is available in all clinic Panel recommendations for assessing patient fitness in clinical
settings caring for patients with malignant lymphoma. Geriatric trials.
assessment is a multi-professional approach; some institutions Recommendation 1.1: The panel suggests that geriatric
will not have a geriatrician, but most of the work-up can be done assessment should be included in the diagnostic process of
by different professional groups within a therapeutic team. clinical trials in order to assess patient fitness.

by guest
on 05 February 2018
Level of evidence: II [42], and cut-offs for significant variations have been identified
Strength of recommendation: B through a comparison with patients’ perceived changes [43].
Consensus: 100% yes (23 voters)
Clarity of reporting system: Quality of life reporting has been
Panel recommendations for assessing patient fitness in routine standardised [44].
clinical practice.
Recommendation 1.2: The panel suggests that a geriatric Panel recommendations for assessing quality of life in clinical
assessment is included in the diagnostic process to assess patient trials.
fitness in routine clinical practise. In cases when geriatric Recommendation 2.1: Quality of life should be considered as
assessment is not possible, geriatric screening (e.g. G-8) can be a prognostic indicator of survival [33].
carried out. Level of evidence: I
Level of evidence: II Strength of recommendation: A
Strength of recommendation: B Consensus: 100% yes (23 voters)
Recommendation 2.2: Quality of life should be included as a
major end point in clinical trials in the elderly, either alone (e.g.
2. Assessing ‘quality of life’ in the elderly in the palliative setting) or in combination with a survival end
lymphoma patient point (as a co-primary or composite end point) [7].
In the management of haematological malignancies, the typical Level of evidence: II
dilemma faced by haematologists when making treatment Strength of recommendation: B
decisions is to determine the balance between efficacy and toxic- Consensus: 100% yes (23 voters)
ity. However, in the elderly, another balance between quantity
and quality of life should also be considered, although this is more Panel recommendations for assessing quality of life in routine
difficult to assess since it is highly dependent on the individual clinical practice.
patient’s views [31]. Indeed, quality of life and other patient-reported Recommendation 2.3: Other PROs can be considered, includ-
outcomes (PROs) are a routine part of treatment decision making, ing maintenance of functional capacity/dependence, either alone
most strikingly for older patients, but unexpectedly are rarely [7, 34, 35] or in combination with survival end points. This type
included among major end points of clinical trials in the elderly. In of end point is encouraged in clinical trials but methodological
a review of recent literature (2005–2011), among 87 randomised questions remain to be solved. Standardised instruments such as
controlled trials conducted in patients with lymphomas, none ADL and IADL are available [45, 46] and their use during treat-
considered quality of life as a primary end point and only 5% ment in clinical practice is encouraged.
included it as a secondary end point [32]. This was also observed in Level of evidence: III
a review of 1207 trials in haematological malignancies, with 8% of Strength of recommendation: C
trials including quality of life as an end point; however, this paper Consensus: 100% yes (23 voters)
also showed that this proportion rises to 31% in phase III trials and
appears higher in the elderly (18% versus 7% in all other studies)
3. Diagnostic work-up and treatment of elderly
[5]. Finally, there is strong evidence to show that quality of life may
patients with CLL
have prognostic value [33].
Other approaches to PROs should also be considered, including CLL is the most common leukaemia in Western countries. Most
Q-TWIST (quality-adjusted time without symptoms of disease or patients newly diagnosed with CLL are of advanced age. The
toxicity of treatment) which measures quality-adjusted survival median age at diagnosis is over 70 years, and almost one fifth of
with three possible consecutive health states (time with toxicity patients are 80 years old or older when CLL is diagnosed [3].
resulting from treatment, time without symptoms of Major comorbidities such as coronary heart disease, heart failure,
disease or toxicity, time from progression/relapse to death) [7] peripheral artery disease, chronic obstructive lung disease or
and evaluation of functional capacity/independence according diabetes mellitus are present in approximately half of all patients
to time, either alone or in combination with survival end points newly diagnosed with CLL [47], but the prevalence of geriatric
[34, 35]. However, methodological difficulties with this approach syndromes such as dementia, delirium, depression, falls, sarcope-
are not always easy to solve. nia or frailty has not been specifically examined in older patients
with CLL. There is growing evidence, however, that comorbid-
Methodological considerations. ities and geriatric syndromes unfavourably impact on treatment
Broad availability: Multiple quality of life questionnaires are feasibility and overall prognosis of such patients [48–51].
available but their metrics should be evaluated cautiously. Treatment differs substantially between young and elderly
Among others, including the MOS-SF36 [36–38], the EORTC patients with CLL; for example, fludarabine/cyclophosphamide/
QLQ-C30 is strongly validated in multiple languages [39], and its rituximab (FCR) is still the standard first-line treatment in young
complement for the elderly, named QLQ-ELD14, is now available fit patients without a TP53 deletion/mutation [9, 52].
[40, 41].
Diagnostic work-up of CLL in elderly patients. Procedures to
Reproducibility and standardisation: Recommendations have diagnose and stage CLL are outlined by pre-existing guidelines
been provided for the use of EORTC QLQ-C30 in clinical trials [9, 53]. There are no data to suggest that methodology should

by guest
on 05 February 2018
differ between younger and older patients. Diagnosis of CLL is with bendamustine/rituximab (BR) or dose-attenuated FCR
made by history taking, physical examination, blood count includ- appear less toxic than full-dose FCR and suggest preserved
ing differential, blood smear microscopy and flow cytometry of the efficacy in fit older patients [58–61]. These regimens may
blood. Staging of CLL (i.e. Binet or Rai classification) requires a therefore be considered as alternatives to full-dose FCR in this
physical examination and blood count. Various studies have shown patient group. Combinations of chlorambucil with a monoclo-
that the presence of a deletion of the short arm of chromosome nal anti-CD20 antibody (obinutuzumab, G-CLB; ofatumumab,
17 [Del(17p)] and/or a mutation of the tumour suppressor gene O-CLB; rituximab, R-CLB) have been demonstrated as feasible
p53 (TP53mut) is associated with poor response to conventional and beneficial in previously untreated vulnerable older patients
chemo-immunotherapy approaches and poor survival, independ- with CLL [25, 62]. Chemo-immunotherapy with BR or dose-
ent of age. For proper choice of therapy, knowledge of Del(17p) attenuated FCR is also feasible and efficacious in these patients
and/or TP53mut status is therefore a prerequisite not just in [63, 64]. Yet, evidence for such treatments in vulnerable older
younger but also in older patients. patients is lower than in fit older patients. Irrespective of being
Older patients with CLL differ in fitness and thus there is no fit or vulnerable, older patients harbouring Del(17p) or TP53mut
uniform benefit from a given treatment regimen. In a pragmatic are candidates for frontline treatment with the kinase inhibitor
approach, these patients can be categorised into three groups: (i) fit ibrutinib [65]. Ibrutinib may also represent a suitable frontline
patients who appear fit for full-dose standard therapy (aiming treatment for older patients without Del(17p) or TP53mut
for symptom-control, complete remission (CR) of the disease and [66]. However, studies comparing ibrutinib with chemo-
prolongation of survival); (ii) vulnerable patients who appear unfit immunotherapy are lacking and short follow-up currently
for full-dose standard therapy but eligible for alternative therapy precludes final conclusions regarding the benefits and risks
plus geriatric interventions (aiming for symptom-control, (i.e. specific toxicities, drug interactions, non-adherence) of such
long-term disease-control and eventually prolongation of treatment in older patients with CLL. With advancing age,
survival); and (iii) terminally ill patients who appear ineligible however, treatment with the kinase inhibitors ibrutinib or
for any anti-leukaemic therapy and who should therefore receive idelalisib (the latter in combination with rituximab) have been
best supportive care (aiming for symptom palliation only). Clinical shown to be safe and highly efficacious as salvage therapy for fit or
judgement is the current standard to assign individual patients vulnerable older patients with refractory or early relapsing CLL.
to one of these categories. However, additional geriatric assessment, Importantly, these treatments proved active in heavily pre-
including a systematic scoring of comorbidity, function and treated patients without Del(17p) or TP53mut and also for patients
autonomy, is able to unmask vulnerability that otherwise may with Del(17p) or TP53mut who have failed chemo-immunotherapy
remain undetected [20, 54]. [67, 68]. For patients who have failed kinase inhibitor therapy, one
treatment option is the pro-apoptotic drug venetoclax (ABT-199)
[69]. Other salvage therapies are available (e.g. ofatumumab, ster-
Panel recommendations for diagnostic work-up.
oids plus rituximab), but evidence in older patients with CLL is gen-
Recommendation 3.1: Similar methodology should be used in
erally limited. Recapitulation of chemo-immunotherapy remains a
older compared with younger patients to diagnose and to stage
therapeutic option in older patients with late relapse of CLL and
CLL (i.e. history taking, physical examination, differential blood
lack of Del(17p) or TP53mut.
count, blood smear microscopy, flow cytometry of blood).
Level of evidence: V
Strength of recommendation: B Panel recommendations for treatment in routine clinical practice.
Consensus: 100% (23 voters) Recommendation 3.4: Only a small minority of carefully
selected and very fit older patients with untreated CLL not harbour-
Recommendation 3.2: Older patients with CLL should to be
ing Del(17p) or TP53mut might be treated with full-dose FCR.
screened for Del(17p) and/or TP53mut whenever treatment is
Level of evidence: I
planned.
Strength of recommendation: B
Level of evidence: III
Consensus: 100% (23 voters)
Consensus: 100% (23 voters) Recommendation 3.5: Fit older patients with untreated CLL
not harbouring Del(17p) and TP53mut should be evaluated for
Recommendation 3.3: Clinical judgement supported by
alternative treatments such as BR or dose-attenuated FCR.
geriatric assessments should be used to stratify older patients
Level of evidence: II
with CLL for fitness and treatment goals (fit versus unfit versus
terminally ill).
Strength of recommendation: B Recommendation 3.6: Vulnerable older patients with un-
Consensus: 100% (23 voters) treated CLL not harbouring Del(17p) or TP53mut should be
treated with G-CLB, O-CLB or R-CLB.
Level of evidence: I
Treatment of CLL in elderly patients. Full-dose chemo-
Strength of recommendation: A
immunotherapy with FCR is the standard frontline treatment in
younger patients with CLL [55]. With advancing age, however,
the risk of FCR-related toxicity and treatment discontinuation Recommendation 3.7: Vulnerable older patients with
generally increases [55–58]. Frontline chemo-immunotherapy untreated CLL not harbouring Del(17p) or TP53mut may be

by guest
on 05 February 2018
considered for alternative treatments such as BR or dose- thoroughness in elderly patients as in younger patients.
attenuated FCR or ibrutinib. Whenever possible, excisional lymph node biopsy or at least a
Level of evidence: II core biopsy should be carried out in patients without easily
Strength of recommendation: B accessible lymph nodes. Fine-needle aspirations should be
Consensus: 100% (23 voters) avoided as they are insufficient for an appropriate diagnosis.
Staging should be with a view to exclude localised disease since
Recommendation 3.8: Older patients with untreated CLL
treatment differs between localised and advanced stages of FL.
harbouring Del(17p) and/or TP53mut should be considered for
This implies that as soon as advanced stage disease is confirmed,
treatment with ibrutinib.
staging procedures should be avoided which do not impact treat-
ment decisions. Initial work-up normally includes a computed
tomography (CT) scan of the neck, thorax, abdomen and pelvis,
and a bone marrow (BM) aspirate and biopsy. However, in indi-
Recommendation 3.9: Older patients with relapsed or vidual patients with palpable lymph nodes (e.g. of the neck or
refractory CLL should be considered for treatment with ibrutinib axilla), physical examination combined with an ultrasound of the
or idelalisib plus rituximab (irrespective of Del[17p] and abdomen may be sufficient. In terminally ill patients with
TP53mut status). confirmed advanced stage disease and normal peripheral blood
Level of evidence: I (PB) cell counts, BM biopsy might be omitted. Positron emission
Strength of recommendation: A tomography CT (PET-CT) scan is not recommended on a regular
Consensus: 100% (23 voters) basis, but can be helpful to confirm localised disease in patients
for whom local radiotherapy is an option. A complete blood
Recommendation 3.10: Older patients with relapsed or
count, routine blood chemistry including LDH, b2-microglobulin
refractory CLL not harbouring Del(17p) or TP53mut could be
and uric acid, as well as screening tests for human immunodefi-
evaluated for alternative treatments (e.g. CD20 antibody).
ciency virus (HIV) and hepatitis B and C, are also recommended
in elderly patients. Immunophenotyping of PB and BM cytology,
Strength of recommendation: C
and detection of BCL2 rearrangements by polymerase chain reac-
tion assay, are not routinely recommended. Prognostication by the
Recommendation 3.11: Older patients with late relapse ‘Follicular Lymphoma-specific International Prognostic Index’
CLL not harbouring Del(17p) or TP53mut may be evaluated for (FLIPI) or the revised FLIPI 2 (incorporating b2-microglobulin,
re-administration of chemo-immunotherapy. diameter of largest lymph node, BM involvement and haemoglo-
Level of evidence: V bin level) is recommended, whenever possible. There are no data
Strength of recommendation: B demonstrating the need to adjust treatment according to the prog-
Consensus: 100% (23 voters) nostic scores of these indices. Geriatric assessment has not been
established specifically for FL patients, but has its value to assess
4. Diagnostic work-up and treatment of elderly fitness in older lymphoma patients as demonstrated for patients
patients with FL with CLL [49].
FL occurs in the elderly with a median age of diagnosis of 61 years Panel recommendations for diagnostic work-up.
[70]. Registry data have shown that elderly non-Hodgkin’s Recommendation 4.1: Elderly patients should be diagnosed
lymphoma (NHL) patients, including those with FL, suffer from based on lymph node histology whenever possible. The aim of
comorbidities with increasing age, with a prevalence of serious staging is to discriminate between patients with limited disease
comorbidity in 43% and 61% for patients aged 60–69 and those with advanced stage disease. Any diagnostics that do not
and >70 years old, respectively [71]. The vast majority of these impact on treatment decisions should be avoided, particularly in
elderly patients present at an advanced stage and are therefore in terminally ill patients.
a palliative situation. Thus, the aim of treatment in this group is Level of evidence: V
to control disease while maintaining the best quality of life Strength of recommendation: A
possible. Following this approach, many patients with advanced Consensus: 100% (23 voters)
stage FL achieve disease control with long-lasting disease-free
intervals without compromising their quality of life. There are Treatment of FL in elderly patients.
differences in management between young fit and elderly patients Asymptomatic patients: For treatment-naı̈ve patients, as well
with comorbidities; for instance, in young fit patients with as those in relapse, the ‘watch and wait’ strategy is standard in the
advanced-stage disease, six cycles of fully dosed (90 mg/m2) management of elderly patients with FL.
bendamustine in combination with rituximab are standard [11],
whereas in elderly patients, a reduction in both the number Symptomatic patients: If the elderly patient develops
of cycles and dose (e.g. four cycles, 70 mg/m2) is considered symptoms due to lymphoma progression, treatment should be
appropriate for many patients, as noted below. initiated.
Diagnostic work-up of FL in elderly patients. Diagnostic work-up Patients with mild symptoms: For elderly patients with mild
in advanced stage FL is defined in the ESMO guidelines [11]. In symptoms and low tumour burden, it is recommended to avoid
general, diagnostics should be carried out with the same chemotherapy. Rituximab has anti-lymphoma activity and is a

by guest
on 05 February 2018
valid treatment option as single-agent induction (4 weekly, Panel recommendations for treatment in routine clinical practice.
375 mg/m2) or single-agent induction followed by maintenance Recommendation 4.2: Asymptomatic elderly patients should
[72]. In relapse, dose reduced BR (50 or 70 mg/m2 bendamustine undergo a watch and wait strategy.
on day 1 and day 2 of 28-day cycles) or rituximab/cyclophospha- Level of evidence: I
mide/vincristine/prednisone (R-CVP) are treatment options. Strength of recommendation: A
Another possibility is to adjust rituximab/chemotherapy to the Consensus: 100% (23 voters)
fitness of the patient in order to limit the number of cycles to
Recommendation 4.3: Symptomatic patients with mild
three or four. Idelalisib is appropriate for patients with relapsed
symptoms should be offered a chemotherapy-free approach such
FL who are no longer responding to rituximab/chemotherapy.
as rituximab single agent, if possible.
However, although this is a chemotherapy-free approach, recent
data demonstrated considerable toxicity, which included
diarrhoea and opportunistic infections, and so this drug should
be used with caution and Pneumocystis jiroveci pneumonia
prophylaxis and cytomegalovirus monitoring [73]. Recommendation 4.4: For patients with high tumour burden
tolerating chemotherapy, a rituximab/chemotherapy regimen
Patients with high tumour burden: For elderly patients with a such as BR is recommended (with bendamustine dose reduction
high tumour burden according to the GELF (Groupe d’Etude des or fewer treatment cycles, if necessary. Be aware of bendamustine-
Lymphomes Folliculaires) criteria [74], more rapid lymphoma associated infections; consider antibacterial/antiviral prophylaxis).
debulking is necessary. In this situation, BR and R-CVP are well- Level of evidence: I
established and well-tolerated treatment options. However, Strength of recommendation: A
BR was just recently associated with a higher rate of partly fatal Consensus: 100% (23 voters)
infections, probably due to prolonged T-cell suppression.
Recommendation 4.5: For relapsed patients, rituximab/
Therefore, bendamustine in combination with an anti-CD20
chemotherapy adjusted to the fitness of the patient is also
antibody should be given with caution and anti-bacterial/antivi-
standard in the elderly. Rituximab maintenance is optional both
ral prophylaxis should be considered. In elderly fit patients,
first line and in relapse. Idelalisib should be used with caution
rituximab plus CHOP (R-CHOP) is also an effective treatment
in relapsed patients not responding to rituximab/chemotherapy
regimen. Rituximab maintenance in first remission has
because of its toxicity profile.
been shown to prolong progression-free survival (PFS), but it is
associated with toxicities such as neutropaenia and an increased
rate of infections. Although these complications are manageable
in the majority of patients, rituximab maintenance should be
considered optional. In relapse with a long remission duration
(>18–24 months), the initial rituximab/chemotherapy can be 5. Diagnostic work-up and treatment of elderly
repeated. In cases with shorter remission duration, alternate patients with MCL
R-chemotherapies (e.g. R-CVP after BR and vice versa) should
be used. Rituximab maintenance is a well-established treatment Treatment goals in elderly patients with MCL: cure versus disease
option for relapsed patients and may also be considered for control/palliation? Advanced MCL is generally considered an
patients who did not receive rituximab maintenance as part of incurable disease. Nonetheless, even within the subgroup of older
their first-line therapy. Idelalisib is a treatment option patients not eligible for intensive induction therapy and stem cell
for patients who are no longer responding to rituximab/ transplantation, treatment goals differ according to biological age,
chemotherapy combinations, but should be used with the PS and comorbidities. A reasonable and commonly used approach
precautions mentioned above. Radio-immunotherapy has a is to stratify elderly patients into three categories: (i) elderly fit, (ii)
different mode of action compared with all other mentioned elderly vulnerable and (iii) terminally ill individuals [77].
treatment approaches as it exploits the high radiosensitivity The treatment goal for elderly fit patients is similar to that for
of lymphomas. Ibritumomab tiuxetan combines anti-CD20 younger patients [12] (i.e. to achieve long-term remissions).
targeting with the b-emitter Yttrium-90; it has shown high Therefore, achievement of a CR by intensive immune-chemotherapy
single-agent activity in relapsed FL and as consolidation therapy is the therapeutic goal.
after first-line treatment of FL, which led to the approval of this For vulnerable patients with comorbidities, the aim of treat-
agent in these two indications in Europe [75, 76]. ment is to control the disease. However, tolerability of standard
treatment might be compromised by comorbidities and impaired
Elderly vulnerable patients: A particular challenge is the man- organ function. Therefore, adaptation of treatment intensity is
agement of elderly vulnerable patients in need of treatment. mandatory with the aim of balancing therapeutic efficacy
Treatment options correspond to the treatment options and toxicity; thus, the therapeutic approach differs from those
mentioned above for patients with low tumour burden and pursued in young fit patients [12].
comprise rituximab single agent with or without a shortened ritux- In contrast, symptom control while preserving quality of life is
imab maintenance (e.g. four infusions every 2 months). Rituximab the main goal in terminally ill patients.
is approved for subcutaneous application if the patient tolerates the
first intravenous application. Another treatment option for elderly How do treatment goals impact on our diagnostic work-up?
vulnerable patients, avoiding intravenous applications, is R-CLB. Diagnostic work-up does not substantially differ in elderly

by guest
on 05 February 2018
patients compared with patients of a younger age. As in younger the bortezomib-containing induction regimen, bortezomib/rit-
patients, a reliable diagnosis is of major importance to distinguish uximab/cyclophosphamide/doxorubicin/prednisone (VR-CAP),
MCL from other lymphoproliferative entities. Therefore, histolo- has been shown to be more effective than R-CHOP in patients
gical confirmation by excisional lymph node biopsy or at least with newly-diagnosed MCL in a large phase III trial, but this is
core biopsy is mandatory. Fine-needle biopsy is considered at the cost of increased haematological toxicity [82]. The add-
inappropriate. In addition to immunophenotypic characterisation, ition of dose-adapted high-dose cytarabine to BR (i.e. R-BAC)
detection of cyclin D1 overexpression or chromosomal transloca- in elderly patients has shown excellent response and PFS rates in
tion t(11;14) is essential to ensure an unequivocal diagnosis. In rare a phase II trial, but it is also associated with considerable
cyclin D1-negative cases, the use of SOX11 may help to identify at haematological toxicity [83]. Preliminary results from a follow-
least some cyclin D1-negative MCL variants [12]. up trial using a reduced cytarabine dose have recently been
A CT scan with iodine contrast of the neck, chest, abdomen presented demonstrating reduced toxicity while preserving effi-
and pelvis is mandatory for all patients. The newest consensus cacy [84]. Therefore, this regimen might be considered in very
report of the International Conference on Malignant fit older patients. Currently, a large randomised phase III trial
Lymphomas Imaging Working Group recommends the use comparing R-CHOP versus alternating R-CHOP and rituxi-
of PET-CT for staging and remission assessment at the end of mab/cytarabine/dexamethasone (R-HAD) in older patients is
therapy in 18F-fluorodeoxyglucose (FDG)-avid lymphoma in open for accrual [85].
clinical practice [78]. Therefore, PET-CT scanning is considered For vulnerable patients with comorbidities and/or impaired organ
optional for fit elderly patients. (PET)-CT is particularly useful for function, dose-adapted chemo-immunotherapeutic regimens are
the minority of patients with stage I–II disease who are candidates appropriate. Options include dose-reduced BR, R-CVP or R-CLB.
for localised radiotherapy to confirm early-stage disease and is For vulnerable patients with more severe comorbidities, no
recommended in the latest ESMO CPG [12]. For terminally ill standard of care exists. A number of less intensive therapies
patients, the use of PET-CT scanning is not recommended. are available including R-CLB, (dose-reduced) BR or the oral
BM aspirate and biopsy should be included in the diagnostic metronomic combination of prednisone/etoposide/procarbazine/
work-up of all elderly fit patients with MCL. In vulnerable cyclophosphamide (PEP-C) [86–88]. Rituximab monotherapy is
patients, a BM examination is considered optional, whereas in another option for elderly patients unable to tolerate chemo-
terminally ill patients, it might be dispensable at least if the PB therapy. However, the objective response rate with single-agent
count is normal or only slightly altered. rituximab is low (27%) and, in contrast to findings in FL, patients
receiving prolonged treatment did not show an improved clinical
Panel recommendations for diagnostic work-up in routine clinical response [89].
practice. Recently, chemotherapy-free regimens (e.g. lenalidomide
Recommendation 5.1: Diagnostic work-up in elderly patients plus rituximab) have shown promising results as a first-line
should generally not differ from patients of a younger age. regimen for MCL, with good tolerability [90]. In addition,
Histological confirmation by excisional lymph node biopsy or high response rates and long-lasting remissions have been re-
at least core biopsy is mandatory. Detection of cyclin D1 overex- ported with the bruton tyrosine kinase inhibitor, ibrutinib,
pression or chromosomal translocation t(11;14) is essential. in relapsed or refractory MCL, with a favourable safety profile
Imaging should include at least a CT scan with iodine contrast of [91, 92]. It is tempting to speculate whether vulnerable patients
the neck, chest, abdomen and pelvis. The use of PET-CT imaging may particularly benefit from these new drugs. Unfortunately,
is considered optional for fit elderly patients. BM aspirate and to-date no data are available from clinical trials for this subset
biopsy should be carried out in elderly fit patients, whereas BM of patients and therefore clear recommendations cannot be
examination is not required in vulnerable or terminally ill made.
patients.
Level of evidence: V Relapsed/refractory disease: As with first-line treatment, the
Strength of recommendation: A choice of second-line and subsequent treatment should be
Consensus: 100% yes (23 voters) adapted to the age and PS of the patient with relapsed or refractory
disease. After a long-lasting remission, repetition of the first-line
Treatment of MCL in elderly patients. It is the strong belief of all treatment can be considered. Otherwise, in elderly fit patients,
authors that the best option for all elderly patients – either for non-cross-resistant drugs and combination regimens should be
first-line treatment or in relapse – is enrolment in a clinical trial. preferred as salvage treatment. In subsequent relapse or in elderly
vulnerable patients, monotherapies with targeted drugs
First-line treatment: Based on the results of a large phase III (in particular ibrutinib [91, 93], lenalidomide 6 rituximab [94],
trial, R-CHOP chemo-immunotherapy followed by rituximab temsirolimus 6 rituximab [95] or bortezomib [96]) as well as
maintenance is considered a standard first-line option well-tolerated, dose-adapted chemotherapy combinations and
for elderly fit patients outside a clinical trial [79]. The use of rit- palliative radiotherapy should be considered.
uximab/fludarabine/cyclophosphamide (R-FC) as first-line
treatment is discouraged due to early failures and insufficient Panel recommendations for treatment in routine clinical practice.
haematopoietic recovery. BR has been reported to be at least as Recommendation 5.2: For elderly fit patients, the following
effective as R-CHOP in two phase III trials and is considered an chemo-immunotherapeutic regimens are recommended as the
alternative induction regimen for older MCL patients [80, 81]. preferred first-line treatment options in routine clinical
Both trials did not include rituximab maintenance. In addition, practice:

by guest
on 05 February 2018
1. R-CHOP followed by rituximab maintenance diseases is typically over 60 years [97], the only exception being
Level of evidence: I anaplastic lymphoma kinase (ALK)þ ALCL. Treatment goals for
Strength of recommendation: A elderly patients with PTCL may be stratified according to fitness
2. BR at presentation and thus often differ from treatment of young fit
Level of evidence: I patients, for whom dose intensification and transplant play a
Strength of recommendation: A major role [13]. In elderly fit patients, the aim of treatment is to
3. VR-CAP induce a CR with induction therapy since attainment of a CR
Level of evidence: I correlates with the best outcomes for PTCL [98]. For vulnerable
Strength of recommendation: A patients, the aim of treatment is to control disease using
4. R-BAC treatment-adapted regimens according to end organ deficit and
Level of evidence: V comorbid conditions. For terminally ill patients, it may be appro-
Strength of recommendation: B priate to offer palliative control to maintain a reasonable quality
Consensus: 100% yes (23 voters) of life.
Recommendation 5.3: For vulnerable elderly patients, dose-
adapted chemo-immunotherapeutic regimens are considered Diagnostic work-up for elderly patients with PTCL. Due to the
appropriate. Options include dose-reduced BR, R-CVP or R-CLB. heterogeneity in disease subtypes of PTCL, histological confirm-
Level of evidence: V ation by excision biopsy is mandatory, and in some circumstances,
Strength of recommendation: B incisional core biopsy may be sufficient. The histological diagnosis
Consensus: 100% yes (23 voters) of PTCL should be made according the World Health
Organization (WHO) Classification 2008 [99] and requires an
Recommendation 5.4: For vulnerable patients with severe expert haematopathologist to gather the histological, immunophe-
comorbidities, mild chemo-immunotherapeutic regimens like notypical and molecular (if required in difficult cases) results and
R-CLB, (dose-reduced) BR or PEP-C are considered appropriate. combine these with the clinical presenting features to categorise
Level of evidence: V the final diagnosis. Recently, advances in our understanding of the
Strength of recommendation: B biology of some specific subtypes have revealed unique hallmark
Consensus: 100% yes (23 voters) identifications, such as expression of CD10, CXCL13 and
Recommendation 5.5: While newer targeted drugs like programmed cell death ligand-1 (PD-L1), suggesting a follicular
ibrutinib and lenalidomide might offer benefits, particularly in helper T-cell of origin [100, 101].
vulnerable patients, no data are available from clinical trials for The diagnostic work-up requires full staging investigations,
this subset of patients and therefore clear recommendations including CT with iodine staging scans of the neck, chest, abdomen
cannot be given. For relapsed or refractory disease, treatment and pelvis. PET-CT scanning should be considered for elderly fit
should be adapted to the age and PS of the patient. Besides patients when curative intent is considered possible, since PET
non-cross-resistant combination regimens, treatment options appears to be particularly useful for identifying extranodal disease
include: such as in the gastrointestinal tract which is often observed in
1. Ibrutinib PTCL [102]. The recent consensus guidelines recommend PET-
Level of evidence: II CT for FDG-avid lymphomas which encompass the three most
Strength of recommendation: A common PTCL subtypes [78]. BM examination should only be
2. Lenalidomide 6 rituximab undertaken in those patients for whom the treatment approach
Level of evidence: II is curative. For all patients where anthracycline therapy is being
Strength of recommendation: B considered, a baseline electrocardiogram and left ventricular
3. Temsirolimus 6 rituximab ejection fraction (LVEF) assessment are recommended.
Strength of recommendation: B Panel recommendations for diagnostic work-up.
4. Bortezomib Recommendation 6.1: The final histological diagnosis
Level of evidence: V requires full analysis and integration of the clinical context and
Strength of recommendation: B expert haematopathological review.
Consensus: 100% yes (23 voters) Level of evidence: IV
Grade of recommendation: A
6. Diagnostic work-up and treatment of elderly
patients with PTCL Treatment of elderly patients with PTCL. Unlike high-grade B-cell
PTCL comprises a very heterogeneous group of diseases and the lymphomas, PTCL is characterised by a higher incidence of early
incidence rates of each subtype varies by geographical region. In relapse and refractory disease. Although several trials have
Europe, the top three subtypes are PTCL not otherwise specified addressed the need to optimise CHOP, this regimen still remains
(34.3%), angioimmunoblastic (28.7%) and anaplastic large-cell the first-line treatment of choice for elderly PTCL [103], although
lymphoma (ALCL) (15.8%) [97]; treatment recommendations in remissions may not be durable. In order to maintain dose
this article refer to these three subgroups, which are largely the intensity, all patients should receive growth factor support.
most common forms and, importantly, are treated similarly in Agents such as etoposide and alemtuzumab have been added to
the elderly population. The median age of presentation of these the CHOP-21 backbone, but were proven to be either too toxic

by guest
on 05 February 2018
[104] or non-beneficial [105] for more elderly patients. New trials 7. Diagnostic work-up and treatment of
are urgently required to test less toxic agents either in addition to elderly patients with DLBCL
the CHOP backbone or as a maintenance strategy to prevent
DLBCL is the most common subtype of NHL, with an incidence
relapse, and the authors strongly recommend entering patients
of 45 per 100 000 in persons aged 60–64 years, rising to 112 per
into clinical trials whenever possible. In the younger population,
100 000 in those of 80–84 years [109]. In contrast to the other
consolidation with autologous stem cell transplantation (ASCT)
lymphoma subtypes discussed so far, DLBCL is a curative disease.
in patients achieving a CR has been shown to improve long-term
The treatment of elderly patients with DLBCL is unsatisfactory,
outcomes [106]; however, the applicability of this approach,
with decreasing OS rates observed with increasing age [110, 111].
although desirable, is seldom achieved in the elderly population
Reasons for this are multifactorial and include an increased
except for in the fittest patients. The assessment of elderly patients
incidence of comorbid conditions, decreased physiological
for suitability to a more intensive approach based on optimal
reserve and decreased functional capacity, all of which are associ-
organ function and the presence of comorbid disease is key to
ated with advancing age. Given this, in contrast to young
decision making. In these circumstances, dose attenuation of
fit patients, dose intense regimens are often not feasible in
chemotherapy or entry into a clinical trial should be considered.
this elderly population [10]. Recent seminal publications
To aid treatment decision making, the conventional
have confirmed that R-CHOP is the gold standard in patients
International Prognostic Index (IPI) may be helpful in selecting
aged 60–80 years [112], but it should be noted that these studies
high-risk patients for more intensive approaches if fit enough
did not include patients of >80 years. Optimal categorisation of
[97, 105, 107]. Although most patients with PTCL present
elderly patients who are fit to receive curative regimens is
with advanced disease, for patients who present with early stage
required to ensure a consistent approach to decision making for
disease, an attenuated chemotherapy schedule followed by local
the elderly patient diagnosed with DLBCL.
radiotherapy may be appropriate. For terminally ill patients,
management should be with a view to preserving quality of life
and symptomatic control. Diagnostic work-up for elderly patients with DLBCL. The
The occurrence of relapse or refractory disease in the elderly diagnosis of elderly patients with DLBCL who are fit for curative
patient is associated with a very poor prognosis, and the treatment should follow standard diagnostic guidelines used
subsequent treatment decision is dictated by patient fitness to for younger patients [10]. This requires resource from a haema-
receive a salvage regimen with gemcitabine or platinum-containing topathology laboratory with expertise in morphology and the
agents. In the rare instances when patients are considered fit ability to carry out immunophenotypic [immunohistochemistry
enough, ASCT may be appropriate providing a satisfactory (IHC) or fluorescence-activated cell sorting (FACS)] and mo-
response is attained, ideally a CR. Patents unsuitable for lecular investigations, should they be required [level of evidence
this intensive approach should be offered novel agents when (LoE) V, grade of recommendation (GoR) A]. A minimum set of
appropriate, such as brentuximab in ALCL [108] or other B-cell IHC markers should be carried out to include CD20,
CD30þ T-cell lymphomas, or be entered into clinical trials CD22, CD79a and CD10. A cell of origin phenotype may be
testing novel agents. obtained based on IHC, but it is not recommended to base
clinical decisions on these results. Epstein–Barr virus (EBV)
Panel recommendations for first-line treatment in routine clinical confirmation by EBER-1 staining is helpful to confirm a diagno-
practice. sis of EBV-positive DLBCL. The concurrent IHC expression of
Recommendation 6.2: Whenever possible, patients should be both MyC and BCL-2 is associated with a poor prognosis [113].
entered into clinical trials. First-line regimes for elderly patients Patients should be carefully examined for evidence of comor-
should be based on a CHOP induction backbone. bid illnesses, impaired PS and functional deficits. A comprehen-
Level of evidence: III sive geriatric assessment (CGA) is desirable and recommended to
Strength of recommendation: B aid categorisation into fit, vulnerable and terminally ill patients.
Consensus: 100% (23 voters) Routine haematological and biochemical blood investigations,
including LDH assessment, are required. Serology screening
Panel recommendations for salvage treatment in routine clinical should be carried out for hepatitis B [hepatitis B core antibody
practice. (anti-HBc) and hepatitis B surface antigen (HBsAg)], hepatitis C
Recommendation 6.3: Whenever possible, patients should and HIV status. Vitamin D level evaluation could be taken into
be entered into clinical trials testing novel agents. However, for consideration, since low levels may result in an inferior treatment
elderly relapsed patients considered unsuitable for clinical trials, outcome.
treatment options include: Staging should be carried out according to the Ann Arbor
1. Salvage chemotherapy with gemcitabine or platinum-containing classification [I, A]. PET-CT should be carried out in patients
agents who are candidates for curative treatment. If the PET-CT demon-
Level of evidence: IV strates BM involvement, BM biopsy is not required. However, if
Strength of recommendation: C the PET-CT fails to demonstrate this, in cases of early stage
2. Novel agents such as brentuximab monotherapy for patients disease, a BM biopsy may be appropriate. Further imaging with
with CD30þ T-cell lymphoma magnetic resonance imaging is indicated if central nervous
Level of evidence: III system disease is suspected, and a lumbar puncture may be car-
Strength of recommendation: B ried out, if feasible, to detect leptomeningeal disease. All patients
Consensus: 100% (23 voters) who are candidates for anthracycline-based treatment require

by guest
on 05 February 2018
baseline LVEF estimation. Prognostic index scoring should be Regarding the group of vulnerable elderly patients with comor-
carried out, and the IPI should be calculated [I, A], but it should bidities, especially cardiac, which may contraindicate anthracy-
be recognised that this index allocates one point for an age of cline use, substitution of doxorubicin by drugs such as
>60 years, but does not stratify the very elderly (>80 years). The gemcitabine, etoposide or a liposomal formulation of doxorubicin
recently devised NCCN-IPI, as well as other elderly prognostic may be beneficial [III, C] [122, 123]. In cases where a high tumour
indexes [114, 115], have refined their categorisation according to load is evident, a steroid pre-phase is recommended to optimise
age, with higher weighting allocated according to advanced age PS and decrease the risk of tumour lysis syndrome [I, A]. The use
[18]. This is important as newer trials focussing specifically of radiotherapy to sites of bulky disease (>7.5 cm) should also be
on elderly patients will require more refined risk scoring to aid considered, since a recent study has shown that it improves out-
stratification to more intensive treatment approaches. comes for elderly patients in this setting [II, C] [124].
In clinical practice, it is often difficult to discriminate who
Panel recommendations for diagnostic work-up. should receive full-dose curative approaches. CGAs are helpful
Recommendation 7.1: For patients treated with curative intent, in the decision-making process regarding who should receive
diagnosis should be carried out in an expert haematopathology full-dose R-CHOP, and their use is recommended [III, C] [125,
laboratory with full diagnostic capabilities (immunophenotypic 126]. Further prospective studies are required to test and opti-
and molecular) and staging should be with PET-CT. mise these strategies, preferably with time-efficient assessments
Level of evidence: V to inform DLBCL risk stratification, both within the clinical trial
Strength of recommendation: A setting and in routine clinical practice. Repletion with vitamin D
Consensus: 100% yes (23 voters) is generally recommended in patients with vitamin D deficiency,
Recommendation 7.2: Cardiac assessment (LVEF) is required as rituximab-mediated cellular cytotoxicity is reduced in vitamin
for patients treated with curative intent. D deficient patients [IV, B] [127].
Level of evidence: V Further emerging reports regarding the mechanisms of treatment
Strength of recommendation: A failure with chemo-immunotherapy have been described,
Consensus: 100% yes (23 voters) with enhanced drug clearance of rituximab observed in elderly male
patients [128]; however, there is currently insufficient evidence to
Recommendation 7.3: The IPI score should be calculated. modify rituximab dosage for these patient subgroups until further
Level of evidence: I prospective data are available. For terminally ill patients, the goal
Strength of recommendation: A of treatment is to achieve a response and control of symptoms in a
Consensus: 100% yes (23 voters) palliative approach, and to maintain quality of life as much as
possible.
Panel recommendations for treatment in routine clinical practice.
De novo disease: Six to eight cycles of R-CHOP-21 are a widely Relapsed DLBCL: For rare cases of patients who are <70 years
used standard treatment of elderly fit patients aged 60–80 years. old and fit enough to be considered for stem cell transplant, salvage
However, there are no prospective data to show that six cycles of treatments should be selected with the best chance of inducing
R-CHOP-21 are as effective as eight cycles. remission, such as rituximab/dexamethasone/high-dose cytarabine/
A large randomised phase III trial demonstrated equivalent cisplatin (R-DHAP), rituximab/etoposide/methylprednisolone/
efficacy and similar toxicity with eight cycles of R-CHOP-21 and cytarabine/cisplatin (R-ESHAP), rituximab/gemcitabine/dexa-
eight cycles of R-CHOP-14 in treatment-naı̈ve elderly patients of methasone/cisplatin (R-GDP) or rituximab/ifosfamide/carboplatin/
this age group [116]. In the German RICOVER-60 trial, six cycles etoposide (R-ICE) [129, 130]. If such patients cannot tolerate full
of R-CHOP-14 were superior to eight cycles of R-CHOP-14 in doses, dose attenuation may be necessary. For patients who are
patients of the same age group [112]. Six cycles of R-CHOP-14 transplant ineligible, salvage treatment may be offered with similar
has the advantage of a shorter duration of chemotherapy regimens as above, or other regimens which may be administered in
compared with eight cycles R-CHOP-21 (71 versus 149 days) and the outpatient setting, such as rituximab/gemcitabine/oxaliplatin
reduced exposure (only three quarters of the total chemotherapy (R-Gem-Ox) [III, B] [131]. Other agents, such as bendamustine
dose). Thus, six cycles of R-CHOP-14 with eight doses of rituxi- [132] and pixantrone [133], have been associated with some effi-
mab are an alternative treatment regimen for elderly patients cacy in these patients. Ideally, and if suitable, patients should be con-
who tolerate this dose dense regimen. sidered for clinical trials testing novel small molecules, novel
All patients above 65 years should receive prophylactic immunotherapy approaches or maintenance strategies, as recently
granulocyte-colony stimulating factor since the highest incidence of reported for lenalidomide [134]. For those patients with primary
treatment-related mortality occurs within the first two cycles [117]. refractory disease, the outlook is very poor; however, consideration
Emerging data in the fit very elderly (>80 years) patients with no for clinical trials is an option for suitable patients.
existing comorbidities suggest that full-dose-intensity R-CHOP may
be detrimental [118, 119], with reports suggesting inferior results in Panel recommendations for treatment in routine clinical practice.
this age group. This is supported by the encouraging results for First-line treatment:
the >80-year-old group treated with the mini R-CHOP regimen, Recommendation 7.4: A CGA is recommended to guide treat-
who achieved a 2-year OS rate of 59% with mini R-CHOP [120], ment choice.
and 2-year OS rate of 64.7% with ofatumumab and reduced-dose Level of evidence: III
CHOP (OFA-mini-CHOP) [III, B] [121]. However, carefully Strength of recommendation: A
selected patients could still receive full-dose CHOP. Consensus: 100% yes (23 voters)

by guest
on 05 February 2018
Recommendation 7.5: The aim of treatment in fully fit academic trials from Celgene, Janssen, Mundipharma, Pfizer and
patients who are <80 years old should be curative, with a full- Roche; speaker’s honoraria from Celgene, Janssen,
dose anthracycline-based regimen preferred. R-CHOP is the Mundipharma, Pfizer and Roche. A.J.M.F.: advisory boards for
recommended first-line treatment choice. Celgene, Gilead and Mundipharma; speaker honoraria from
Level of evidence: I Italfarmaco, Mundipharma and Gilead; research support from
Strength of recommendation: A Celgene, Roche, Mundipharma and Rhizen. G.G.: advisory
Consensus: 100% yes (23 voters) boards for Roche, Janssen, Amgen, Novartis, GlaxoSmithKline,
Karyopharm and Morphosys; speaker honoraria from Roche,
Recommendation 7.6: For fully fit patients who are >80 years
Janssen, Gilead and GlaxoSmithKline. V.G.: speaker honoraria,
old without comorbidities, dose-attenuated R-CHOP may be
advisory boards, consultancy and travel grants from Roche;
appropriate.
speaker honoraria from Mundipharma, GlaxoSmithKline and
Bristol-Myers Squibb. M.H.: speaker honoraria, advisory boards
and consultancy from Takeda; advisory boards and consultancy
from Janssen, Celgene and Roche/Genentech. M.L.: honoraria
from Celgene, Janssen-Cilag, Roche, Amgen, Mundipharma and
Treatment of relapsed patients:
Teva; research contracts with Celgene, Pfizer, Mundipharma and
Recommendation 7.7: For relapsed fit (no organ dysfunction,
Roche; research grants from Amgen, Roche and Takeda. S.L.G.:
PS 0–1, no comorbidities), transplant-eligible patients, appropri-
speaker honoraria, advisory boards, consultancy and travel grants
ate salvage treatment with R-DHAP, R-ESHAP, R-ICE or R-GDP
from Roche, Janssen-Cilag and Celgene; research programme
is indicated. In the event of an adequate response, ASCT is
funding from Roche, Janssen and Servier. U.M.: honoraria for
recommended.
advisory board activities from Roche, Celgene, Amgen and
Janssen-Cilag; research grants from Mundipharma and Celgene.
P.dN.B.: speaker’s bureau/lectures for Roche and Bayer. M.P.:
advisory boards for Celgene, Roche and Spectrum; research
Recommendation 7.8: For transplant-ineligible patients, dose support from Amgen, Roche and Spectrum. N.S.: travel grants,
attenuated R-DHAP, R-ESHAP, R-ICE, or less intense regimens advisory boards and research support from Roche; travel grants
such as R-Gem-Ox, are appropriate. and advisory boards from Celgene; travel grants and honoraria for
Level of evidence: III company symposia from Riemser Pharma and CTI Life Sciences.
Strength of recommendation: B P.S.: advisory boards for Celgene and Teva; honoraria from
Consensus: 100% yes (23 voters) Spectrum Pharmaceuticals and Pierre Fabre Oncology; travel
grants from Hospira, Celgene and Teva. M.S.: speaker’s bureau
Recommendation 7.9: For transplant-ineligible patients,
and advisory boards for Teva, Mundipharma, Janssen-Cilag,
single-agent chemotherapies such as bendamustine or pixantrone
Gilead, CTI BioPharma and Servier; research support, consultant,
may be considered.
speaker’s bureau and advisory boards for Menarini; speaker’s bur-
Level of evidence: I (pixantrone), II (bendamustine)
eau for Roche and Takeda. R.S.: research support from Celgene,
Novartis and Teva; honoraria from Celgene, Novartis and Teva;
scientific advisory board for Celgene. A.S.B.: consultancy fees from
Takeda. M.T.: consultancy for Roche, Gilead, Janssen, Takeda and
Acknowledgements Celgene. J.W.: advisory boards for Roche, Celgene, Janssen-Cilag
and Takeda; research grants from Roche, Celgene, Janssen-Cilag,
The authors thank Jennifer Lamarre, Claire Bramley, Matthew Takeda, GlaxoSmithKline and Seattle Genetics. U.W.: speaker
Wallace, Aude Galli and all ESMO staff for their support honoraria from Amgen, Novartis, Roche and Chugai. E.Z.:
throughout the whole consensus process. Angela Corstorphine advisory honoraria and/or support for investigator-initiated stud-
of Kstorfin Medical Communications Ltd provided medical ies (for the institution) from Celgene, Johnson and Johnson/
writing support with the preparation of this manuscript. This Janssen, Gilead, Mundipharma, Roche and Bayer. All remaining
support was funded by ESMO. authors have declared no conflicts of interest.
Funding References
All costs relating to the consensus conference were covered from 1. Eurostat. News release: Nearly 27 million people aged 80 or over in the
the ESMO central funds. There was no external funding of the European Union 2016; http://ec.europa.eu/eurostat/documents/
event or manuscript production. 2995521/7672228/3-29092016-AP-EN.pdf/4b90f6bb-43c1-45ed-985b-
dfbe9564157a (25 November 2016, date last accessed).
2. European Commission, The 2015 Ageing Report 2015; http://europa.
eu/epc/sites/epc/files/docs/pages/ageing_report_2015_en.pdf (8 July
Disclosure 2016, date last accessed).
3. Howlader N, Noone AM, Krapcho M et al. SEER Cancer Statistics
C.B.: honoraria from Roche, Pfizer, Celgene, Pharmacyclics and Review, 1975-2013, National Cancer Institute. Bethesda, MD, based on
Janssen; research grants from Roche and Janssen. M.D.: advisory November 2015 SEER data submission 2016; https://seer.cancer.gov/
boards for Bayer, Celgene, Gilead, Janssen and Pfizer; support for csr/1975_2013/ (22 November 2016, date last accessed).

by guest
on 05 February 2018
4. Tirol Kliniken Tumor register Tyrol 2016; https://www.iet.at/page.cfm? 26. Stauder R, Eichhorst B, Hamaker ME et al. Management of chronic
vpath¼index (22 November 2016, date last accessed). lymphocytic leukemia (CLL) in the elderly: a position paper from an
5. Hamaker ME, Stauder R, van Munster BC. On-going clinical trials for International Society of Geriatric Oncology (SIOG) Task Force. Ann
elderly patients with a hematological malignancy: are we addressing the Oncol 2017; 28: 218–227.
right end points? Ann Oncol 2014; 25: 675–681. 27. Burgstaller S, Wiesinger P, Stauder R. Myelodysplastic syndromes in the
6. Hurria A, Cirrincione CT, Muss HB et al. Implementing a geriatric as- elderly: treatment options and personalized management. Drugs Aging
sessment in cooperative group clinical cancer trials: CALGB 360401. 2015; 32: 891–905.
J Clin Oncol 2011; 29: 1290–1296. 28. Klepin HD, Rao AV, Pardee TS. Acute myeloid leukemia and myelodys-
7. Wildiers H, Mauer M, Pallis A et al. End points and trial design in geri- plastic syndromes in older adults. J Clin Oncol 2014; 32: 2541–2552.
atric oncology research: a joint European organisation for research and 29. Larocca A, Palumbo A. Optimizing treatment for elderly patients with
treatment of cancer–Alliance for Clinical Trials in Oncology– newly diagnosed multiple myeloma: a personalized approach. J Clin
International Society Of Geriatric Oncology position article. J Clin Oncol 2016; 34: 3600–3604.
Oncol 2013; 31: 3711–3718. 30. Cherny NI, de Vries EG, Emanuel L et al. Words matter: distinguishing
8. Dykewicz CA. Summary of the guidelines for preventing opportunistic “personalized medicine” and “biologically personalized therapeutics”.
infections among hematopoietic stem cell transplant recipients. Clin J Natl Cancer Inst 2014; 106.
Infect Dis 2001; 33: 139–144. 31. Yellen SB, Cella DF, Leslie WT. Age and clinical decision making in on-
9. Eichhorst B, Robak T, Montserrat E et al. Chronic lymphocytic leukae- cology patients. J Natl Cancer Inst 1994; 86: 1766–1770.
mia: ESMO Clinical Practice Guidelines for diagnosis, treatment and 32. Bellera C, Praud D, Petit-Monéger A et al. Barriers to inclusion of
follow-up. Ann Oncol 2015; 26 (Suppl 5): v78–v84. older adults in randomised controlled clinical trials on Non-
10. Tilly H, Gomes da Silva M, Vitolo U et al. Diffuse large B-cell lymph- Hodgkin’s lymphoma: a systematic review. Cancer Treat Rev 2013; 39:
oma (DLBCL): ESMO Clinical Practice Guidelines for diagnosis, treat- 812–817.
ment and follow-up. Ann Oncol 2015; 26 (Suppl 5): v116–v125. 33. Quinten C, Coens C, Mauer M et al. Baseline quality of life as a prog-
11. Dreyling M, Ghielmini M, Rule S et al. Newly diagnosed and relapsed fol- nostic indicator of survival: a meta-analysis of individual patient data
licular lymphoma: ESMO Clinical Practice Guidelines for diagnosis, from EORTC clinical trials. Lancet Oncol 2009; 10: 865–871.
treatment and follow-up. Ann Oncol 2016; 27 (Suppl 5): v83–v90. 34. Hoppe S, Rainfray M, Fonck M et al. Functional decline in older pa-
12. Dreyling M, Campo E, Hermine O et al. Newly diagnosed and relapsed tients with cancer receiving first-line chemotherapy. J Clin Oncol 2013;
mantle cell lymphoma: ESMO Clinical Practice Guidelines for diagno- 31: 3877–3882.
sis, treatment and follow-up. Ann Oncol 2017; 28 (Suppl 4): iv62–iv71. 35. Brain EG, Mertens C, Girre V et al. Impact of liposomal doxorubicin-
13. d’Amore F, Gaulard P, Trümper L et al. Peripheral T-cell lymphomas: based adjuvant chemotherapy on autonomy in women over 70 with
ESMO Clinical Practice Guidelines for diagnosis, treatment and follow- hormone-receptor-negative breast carcinoma: a French Geriatric
up. Ann Oncol 2015; 26 (Suppl 5): v108–v115. Oncology Group (GERICO) phase II multicentre trial. Crit Rev Oncol
14. Murawski N, Pfreundschuh M, Zeynalova S et al. Optimization of rituxi- Hematol 2011; 80: 160–170.
mab for the treatment of DLBCL (I): dose-dense rituximab in the DENSE- 36. Hays RD, Sherbourne CD, Mazel RM. The RAND 36-Item Health
R-CHOP-14 trial of the DSHNHL. Ann Oncol 2014; 25: 1800–1806. Survey 1.0. Health Econ 1993; 2: 217–227.
15. Pallis AG, Papamichael D, Audisio R et al. EORTC Elderly Task Force 37. Ware JE. Health Survey. Manual & Interpretation Guide. Boston, MA:
experts’ opinion for the treatment of colon cancer in older patients. The Health Institute, 1993.
Cancer Treat Rev 2010; 36: 83–90. 38. Aaronson NK, Acquadro C, Alonso J et al. International Quality of Life
16. Ziepert M, Schmits R, Trümper L et al. Prognostic factors for hemato- Assessment (IQOLA) project. Qual Life Res 1992; 1: 349–351.
toxicity of chemotherapy in aggressive non-Hodgkin’s lymphoma. Ann 39. Aaronson NK, Ahmedzai S, Bergman B et al. The European
Oncol 2008; 19: 752–762. Organization for Research and Treatment of Cancer QLQ-C30: a
17. Reeve BB, Mitchell SA, Dueck AC et al. Recommended patient-reported quality-of-life instrument for use in international clinical trials in on-
core set of symptoms to measure in adult cancer treatment trials. J Natl cology. J Natl Cancer Inst 1993; 85: 365–376.
Cancer Inst 2014; 106: pii: dju129. 40. Johnson C, Fitzsimmons D, Gilbert J et al. Development of the
18. Zhou Z, Sehn LH, Rademaker AW et al. An enhanced International European Organisation for Research and Treatment of Cancer quality
Prognostic Index (NCCN-IPI) for patients with diffuse large B-cell of life questionnaire module for older people with cancer: the EORTC
lymphoma treated in the rituximab era. Blood 2014; 123: 837–842. QLQ-ELD15. Eur J Cancer 2010; 46: 2242–2252.
19. Wedding U, Pientka L, Höffken K. Quality-of-life in elderly patients 41. Wheelwright S, Darlington AS, Fitzsimmons D et al. International val-
with cancer: a short review. Eur J Cancer 2007; 43: 2203–2210. idation of the EORTC QLQ-ELD14 questionnaire for assessment of
20. Wildiers H, Heeren P, Puts M et al. International Society of Geriatric health-related quality of life elderly patients with cancer. Br J Cancer
Oncology consensus on geriatric assessment in older patients with can- 2013; 109: 852–858.
cer. J Clin Oncol 2014; 32: 2595–2603. 42. Cocks K, King MT, Velikova G et al. Evidence-based guidelines for de-
21. Pallis AG, Ring A, Fortpied C et al. EORTC workshop on clinical trial termination of sample size and interpretation of the European
methodology in older individuals with a diagnosis of solid tumors. Ann Organisation for the Research and Treatment of Cancer Quality of Life
Oncol 2011; 22: 1922–1926. Questionnaire Core 30. J Clin Oncol 2011; 29: 89–96.
22. Soubeyran P, Bellera C, Goyard J et al. Screening for vulnerability in 43. Osoba D, Rodrigues G, Myles J et al. Interpreting the significance of
older cancer patients: the ONCODAGE Prospective Multicenter Cohort changes in health-related quality-of-life scores. J Clin Oncol 1998; 16:
Study. PLoS One 2014; 9: e115060. 139–144.
23. Decoster L, Van Puyvelde K, Mohile S et al. Screening tools for multidi- 44. Velikova G, Coens C, Efficace F et al. Health-related quality of life in
mensional health problems warranting a geriatric assessment in older EORTC clinical trials—30 years of progress from methodological devel-
cancer patients: an update on SIOG recommendations. Ann Oncol opments to making a real impact on oncology practice. Eur J Cancer
2015; 26: 288–300. 2012; 10 (Suppl): 141–149.
24. Bellera CA, Rainfray M, Mathoulin-Pélissier S et al. Screening older 45. Katz S, Ford AB, Moskowitz RW et al. Studies of illness in the aged. The
cancer patients: first evaluation of the G-8 geriatric screening tool. Ann index of ADL: a standardized measure of biological and psychosocial
Oncol 2012; 23: 2166–2172. function. JAMA 1963; 185: 914–919.
25. Goede V, Fischer K, Busch R et al. Obinutuzumab plus chlorambucil in 46. Lawton MP, Brody EM. Assessment of older people: self-maintaining
patients with CLL and coexisting conditions. N Engl J Med 2014; 370: and instrumental activities of daily living. Gerontologist 1969; 9:
1101–1110. 179–186.

by guest
on 05 February 2018
47. Thurmes P, Call T, Slager S et al. Comorbid conditions and survival in 64. Michallet A-S. Rituximab in combination with bendamustine or chlor-
unselected, newly diagnosed patients with chronic lymphocytic leuke- ambucil for the treatment of chronic lymphocytic leukaemia: primary
mia. Leuk Lymphoma 2008; 49: 49–56. results from the randomised phase IIIb MABLE study. In XVI
48. Baumann T, Delgado J, Santacruz R et al. Chronic lymphocytic leuke- International Workshop on Chronic Lymphocytic Leukaemia. Sydney,
mia in the elderly: clinico-biological features, outcomes, and proposal Australia: 2015.
of a prognostic model. Haematologica 2014; 99: 1599–1604. 65. Farooqui MZ, Valdez J, Martyr S et al. Ibrutinib for previously untreated
49. Goede V, Cramer P, Busch R et al. Interactions between comorbidity and relapsed or refractory chronic lymphocytic leukaemia with TP53 ab-
and treatment of chronic lymphocytic leukemia: results of German errations: a phase 2, single-arm trial. Lancet Oncol 2015; 16: 169–176.
Chronic Lymphocytic Leukemia Study Group trials. Haematologica 66. Burger JA, Tedeschi A, Barr PM et al. Ibrutinib as initial therapy for pa-
2014; 99: 1095–1100. tients with chronic lymphocytic leukemia. N Engl J Med 2015; 373:
50. Goede V, Bahlo J, Chataline V et al. Evaluation of geriatric assessment 2425–2437.
in patients with chronic lymphocytic leukemia: results of the CLL9 trial 67. Furman RR, Sharman JP, Coutre SE et al. Idelalisib and rituximab in
of the German CLL study group. Leuk Lymphoma 2016; 57: 789–796. relapsed chronic lymphocytic leukemia. N Engl J Med 2014; 370: 997–1007.
51. Satram-Hoang S, Reyes C, Hoang KQ et al. Treatment practice in the 68. Byrd JC, Brown JR, O’Brien S et al. Ibrutinib versus ofatumumab in
elderly patient with chronic lymphocytic leukemia-analysis of the com- previously treated chronic lymphoid leukemia. N Engl J Med 2014; 371:
bined SEER and Medicare database. Ann Hematol 2014; 93: 1335–1344. 213–223.
52. ESMO Guidelines Committee. Appendix 4: Chronic lymphocytic leu- 69. Stilgenbauer S, Eichhorst B, Schetelig J et al. Venetoclax in relapsed or
kaemia: eUpdate published online 27 June 2017 (www.esmo.org/ refractory chronic lymphocytic leukaemia with 17p deletion: a multi-
Guidelines/Haematological-Malignancies). Ann Oncol 2017; 28 (Suppl centre, open-label, phase 2 study. Lancet Oncol 2016; 17: 768–778.
4): iv149–iv152. 70. Friedberg JW, Taylor MD, Cerhan JR et al. Follicular lymphoma in the
53. Hallek M, Cheson BD, Catovsky D et al. Guidelines for the diagnosis United States: first report of the national LymphoCare study. J Clin
and treatment of chronic lymphocytic leukemia: a report from the Oncol 2009; 27: 1202–1208.
International Workshop on Chronic Lymphocytic Leukemia updating 71. van Spronsen DJ, Janssen-Heijnen ML, Breed WP, Coebergh JW.
the National Cancer Institute-Working Group 1996 guidelines. Blood Prevalence of co-morbidity and its relationship to treatment among un-
2008; 111: 5446–5456. selected patients with Hodgkin’s disease and non-Hodgkin’s lymph-
54. Hamaker ME, Prins MC, Stauder R. The relevance of a geriatric assess- oma, 1993-1996. Ann Hematol 1999; 78: 315–319.
ment for elderly patients with a haematological malignancy--a system- 72. Martinelli G, Schmitz SF, Utiger U et al. Long-term follow-up of pa-
atic review. Leuk Res 2014; 38: 275–283. tients with follicular lymphoma receiving single-agent rituximab at two
55. Hallek M, Fischer K, Fingerle-Rowson G et al. Addition of rituximab to different schedules in trial SAKK 35/98. J Clin Oncol 2010; 28:
fludarabine and cyclophosphamide in patients with chronic lympho- 4480–4484.
cytic leukaemia: a randomised, open-label, phase 3 trial. Lancet 2010; 73. Miller BW, Przepiorka D, de Claro RA et al. FDA approval: idelalisib
376: 1164–1174. monotherapy for the treatment of patients with follicular lymphoma
56. Keating MJ, O’Brien S, Albitar M et al. Early results of a chemoimmu- and small lymphocytic lymphoma. Clin Cancer Res 2015; 21:
notherapy regimen of fludarabine, cyclophosphamide, and rituximab as 1525–1529.
initial therapy for chronic lymphocytic leukemia. J Clin Oncol 2005; 23: 74. Brice P, Bastion Y, Lepage E et al. Comparison in low-tumor-burden
4079–4088. follicular lymphomas between an initial no-treatment policy, predni-
57. Robak T, Dmoszynska A, Solal-Céligny P et al. Rituximab plus fludara- mustine, or interferon alfa: a randomized study from the Groupe
bine and cyclophosphamide prolongs progression-free survival com- d’Etude des Lymphomes Folliculaires. Groupe d’Etude des Lymphomes
pared with fludarabine and cyclophosphamide alone in previously de l’Adulte. J Clin Oncol 1997; 15: 1110–1117.
treated chronic lymphocytic leukemia. J Clin Oncol 2010; 28: 75. Witzig TE, Gordon LI, Cabanillas F et al. Randomized controlled trial
1756–1765. of yttrium-90-labeled ibritumomab tiuxetan radioimmunotherapy ver-
58. Eichhorst B, Fink AM, Bahlo J et al. First-line chemoimmunotherapy sus rituximab immunotherapy for patients with relapsed or refractory
with bendamustine and rituximab versus fludarabine, cyclophospha- low-grade, follicular, or transformed B-cell non-Hodgkin’s lymphoma.
mide, and rituximab in patients with advanced chronic lymphocytic J Clin Oncol 2002; 20: 2453–2463.
leukaemia (CLL10): an international, open-label, randomised, phase 3, 76. Morschhauser F, Radford J, Van Hoof A et al. 90Yttrium-ibritumomab
non-inferiority trial. Lancet Oncol 2016; 17: 928–942. tiuxetan consolidation of first remission in advanced-stage follicular
59. Fischer K, Cramer P, Busch R et al. Bendamustine in combination with non-Hodgkin lymphoma: updated results after a median follow-up of
rituximab for previously untreated patients with chronic lymphocytic 7.3 years from the International, Randomized, Phase III First-Line
leukemia: a multicenter phase II trial of the German Chronic Indolent trial. J Clin Oncol 2013; 31: 1977–1983.
Lymphocytic Leukemia Study Group. J Clin Oncol 2012; 30: 3209–3216. 77. Dreyling M, Ferrero S. How to treat old MCL patients: one size fits it
60. Mulligan SP, Gill D, Turner P et al. A randomised dose de-escalation all? Blood 2014; 124: 1207–1208.
study of oral fludarabine, 6oral cyclophosphamide and intravenous rit- 78. Barrington SF, Mikhaeel NG, Kostakoglu L et al. Role of imaging in the
uximab as first-line therapy of fit patients with chronic lymphocytic leu- staging and response assessment of lymphoma: consensus of the
kaemia (CLL) aged 65 years: final analysis of response and toxicity. International Conference on Malignant Lymphomas Imaging Working
Blood 2014; 124: 3325. Group. J Clin Oncol 2014; 32: 3048–3058.
61. Lew TE, Cheah CY, Carney DA et al. Dose-reduced fludarabine, cyclo- 79. Kluin-Nelemans HC, Hoster E, Hermine O et al. Treatment of older pa-
phosphamide and rituximab is well tolerated in older patients with tients with mantle-cell lymphoma. N Engl J Med 2012; 367: 520–531.
chronic lymphocytic leukemia and has preserved therapeutic efficacy. 80. Rummel MJ, Niederle N, Maschmeyer G et al. Bendamustine plus ritux-
Leuk Lymphoma 2016; 57: 1044–1053. imab versus CHOP plus rituximab as first-line treatment for patients
62. Hillmen P, Robak T, Janssens A et al. Chlorambucil plus ofatumumab with indolent and mantle-cell lymphomas: an open-label, multicentre,
versus chlorambucil alone in previously untreated patients with chronic randomised, phase 3 non-inferiority trial. Lancet 2013; 381: 1203–1210.
lymphocytic leukaemia (COMPLEMENT 1): a randomised, multi- 81. Flinn IW, van der Jagt R, Kahl BS et al. Randomized trial of
centre, open-label phase 3 trial. Lancet 2015; 385: 1873–1883. bendamustine-rituximab or R-CHOP/R-CVP in first-line treatment of
63. Smolej L, Brychtova Y, Doubek M et al. Low-dose FCR is a safe and ef- indolent NHL or MCL: the BRIGHT study. Blood 2014; 123:
fective treatment option for elderly/comorbid patients with chronic 2944–2952.
lymphocytic leukemia/small lymphocytic lymphoma. Updated results of 82. Robak T, Huang H, Jin J et al. Bortezomib-based therapy for newly
project Q-Lite By Czech CLL Study Group. Blood 2014; 124: 4670. diagnosed mantle-cell lymphoma. N Engl J Med 2015; 372: 944–953.

by guest
on 05 February 2018
83. Visco C, Finotto S, Zambello R et al. Combination of rituximab, benda- 103. Weisenburger DD, Savage KJ, Harris NL et al. Peripheral T-cell lymph-
mustine, and cytarabine for patients with mantle-cell non-Hodgkin oma, not otherwise specified: a report of 340 cases from the
lymphoma ineligible for intensive regimens or autologous transplant- International Peripheral T-cell Lymphoma Project. Blood 2011; 117:
ation. J Clin Oncol 2013; 31: 1442–1449. 3402–3408.
84. Visco C, Chiappella A, Nassi L et al. Rituximab, bendamustine, and 104. Kim JG, Sohn SK, Chae YS et al. Alemtuzumab plus CHOP as front-line
low-dose cytarabine as induction therapy in elderly patients with man- chemotherapy for patients with peripheral T-cell lymphomas: a phase II
tle cell lymphoma: a multicentre, phase 2 trial from Fondazione Italiana study. Cancer Chemother Pharmacol 2007; 60: 129–134.
Linfomi. Lancet Haematol 2017; 4: e15–e23. 105. Schmitz N, Trümper L, Ziepert M et al. Treatment and prognosis of
85. NCT01865110. R-CHOPþR-HAD vs R-CHOP followed by mainten- mature T-cell and NK-cell lymphoma: an analysis of patients with
ance lenalidomide þ rituximab vs rituximab for older patients with T-cell lymphoma treated in studies of the German High-Grade Non-
MCL 2016; https://clinicaltrials.gov/ct2/show/NCT01865110?term¼ Hodgkin Lymphoma Study Group. Blood 2010; 116: 3418–3425.
NCT01865110&rank¼1 (22 November 2016, date last accessed). 106. d’Amore F, Relander T, Lauritzsen GF et al. Up-front autologous stem-
86. Bauwens D, Maerevoet M, Michaux L et al. Activity and safety of com- cell transplantation in peripheral T-cell lymphoma: NLG-T-01. J Clin
bined rituximab with chlorambucil in patients with mantle cell lymph- Oncol 2012; 30: 3093–3099.
oma. Br J Haematol 2005; 131: 338–340. 107. Suzumiya J, Ohshima K, Tamura K et al. The International Prognostic
87. Sachanas S, Pangalis GA, Vassilakopoulos TP et al. Combination of rit- Index predicts outcome in aggressive adult T-cell leukemia/lymphoma:
uximab with chlorambucil as first line treatment in patients with mantle analysis of 126 patients from the International Peripheral T-Cell
cell lymphoma: a highly effective regimen. Leuk Lymphoma 2011; 52: Lymphoma Project. Ann Oncol 2009; 20: 715–721.
387–393. 108. Pro B, Advani R, Brice P et al. Brentuximab vedotin (SGN-35) in pa-
88. Coleman M, Ruan G, Elstrom RL et al. Metronomic therapy for refrac- tients with relapsed or refractory systemic anaplastic large-cell lymph-
tory/relapsed lymphoma: the PEP-C low-dose oral combination oma: results of a phase II study. J Clin Oncol 2012; 30: 2190–2196.
chemotherapy regimen. Hematology 2012; 17 (Suppl 1): S90–S92. 109. Yancik R, Ries LA. Cancer in older persons: an international issue in an
89. Ghielmini M, Schmitz SF, Cogliatti S et al. Effect of single-agent rituxi- aging world. Semin Oncol 2004; 31: 128–136.
mab given at the standard schedule or as prolonged treatment in pa- 110. Lowry L, Smith P, Cunningham D, Linch DC. Factors affecting survival
tients with mantle cell lymphoma: a study of the Swiss Group for in patients aged 60 and over with diffuse large B cell lymphoma failing
Clinical Cancer Research (SAKK). J Clin Oncol 2005; 23: 705–711. first-line therapy. J Geriatr Oncol 2013; 4: 134–140.
90. Ruan J, Martin P, Shah B et al. Lenalidomide plus rituximab as initial 111. Maartense E, Kluin-Nelemans HC, le Cessie S et al. Different age limits
treatment for mantle-cell lymphoma. N Engl J Med 2015; 373: for elderly patients with indolent and aggressive non-Hodgkin lymph-
1835–1844. oma and the role of relative survival with increasing age. Cancer 2000;
91. Wang ML, Blum KA, Martin P et al. Long-term follow-up of MCL pa- 89: 2667–2676.
tients treated with single-agent ibrutinib: updated safety and efficacy re- 112. Pfreundschuh M, Schubert J, Ziepert M et al. Six versus eight cycles of
sults. Blood 2015; 126: 739–745. bi-weekly CHOP-14 with or without rituximab in elderly patients with
92. Wang ML, Rule S, Martin P et al. Targeting BTK with ibrutinib in aggressive CD20þ B-cell lymphomas: a randomised controlled trial
relapsed or refractory mantle-cell lymphoma. N Engl J Med 2013; 369: (RICOVER-60). Lancet Oncol 2008; 9: 105–116.
507–516. 113. Johnson NA, Slack GW, Savage KJ et al. Concurrent expression of MYC
93. Dreyling M, Jurczak W, Jerkeman M et al. Ibrutinib versus temsirolimus and BCL2 in diffuse large B-cell lymphoma treated with rituximab plus
in patients with relapsed or refractory mantle-cell lymphoma: an interna- cyclophosphamide, doxorubicin, vincristine, and prednisone. J Clin
tional, randomised, open-label, phase 3 study. Lancet 2016; 387: 770–778. Oncol 2012; 30: 3452–3459.
94. Wang M, Fayad L, Wagner-Bartak N et al. Lenalidomide in combin- 114. Advani RH, Chen H, Habermann TM et al. Comparison of conventional
ation with rituximab for patients with relapsed or refractory mantle-cell prognostic indices in patients older than 60 years with diffuse large B-cell
lymphoma: a phase 1/2 clinical trial. Lancet Oncol 2012; 13: 716–723. lymphoma treated with R-CHOP in the US Intergroup Study (ECOG
95. Hess G, Herbrecht R, Romaguera J et al. Phase III study to evaluate tem- 4494, CALGB 9793): consideration of age greater than 70 years in an eld-
sirolimus compared with investigator’s choice therapy for the treatment erly prognostic index (E-IPI). Br J Haematol 2010; 151: 143–151.
of relapsed or refractory mantle cell lymphoma. J Clin Oncol 2009; 27: 115. Procházka V, Pytlı́k R, Janı́ková A et al. A new prognostic score for eld-
3822–3829. erly patients with diffuse large B-cell lymphoma treated with R-CHOP:
96. Fisher RI, Bernstein SH, Kahl BS et al. Multicenter phase II study of the prognostic role of blood monocyte and lymphocyte counts is absent.
bortezomib in patients with relapsed or refractory mantle cell lymph- PLoS One 2014; 9: e102594.
oma. J Clin Oncol 2006; 24: 4867–4874. 116. Delarue R, Tilly H, Mounier N et al. Dose-dense rituximab-CHOP
97. Vose J, Armitage J, Weisenburger D, International TCLP. International compared with standard rituximab-CHOP in elderly patients with dif-
peripheral T-cell and natural killer/T-cell lymphoma study: pathology fuse large B-cell lymphoma (the LNH03-6B study): a randomised phase
findings and clinical outcomes. J Clin Oncol 2008; 26: 4124–4130. 3 trial. Lancet Oncol 2013; 14: 525–533.
98. Abramson JS, Feldman T, Kroll-Desrosiers AR et al. Peripheral T-cell 117. Olszewski AJ, Mantripragada KC, Castillo JJ. Risk factors for early death
lymphomas in a large US multicenter cohort: prognostication in the after rituximab-based immunochemotherapy in older patients with diffuse
modern era including impact of frontline therapy. Ann Oncol 2014; 25: large B-cell lymphoma. J Natl Compr Canc Netw 2016; 14: 1121–1129.
2211–2217. 118. Eyre TA, Salisbury R, Eyre DW et al. Results of a large retrospective ana-
99. Campo E, Swerdlow SH, Harris NL et al. The 2008 WHO classification lysis of the effect of intended dose intensity of R-CHOP on outcome in
of lymphoid neoplasms and beyond: evolving concepts and practical a cohort of consecutive, unselected elderly patients with de novo diffuse
applications. Blood 2011; 117: 5019–5032. large B cell lymphoma. Br J Haematol 2016; 173: 487–491.
100. Dupuis J, Boye K, Martin N et al. Expression of CXCL13 by neoplastic 119. Carson KR, Riedell P, Lynch R et al. Comparative effectiveness of
cells in angioimmunoblastic T-cell lymphoma (AITL): a new diagnostic anthracycline-containing chemotherapy in United States veterans age 80 and
marker providing evidence that AITL derives from follicular helper T older with diffuse large B-cell lymphoma. J Geriatr Oncol 2015; 6: 211–218.
cells. Am J Surg Pathol 2006; 30: 490–494. 120. Peyrade F, Jardin F, Thieblemont C et al. Attenuated immunochemo-
101. Rodrı́guez-Pinilla SM, Atienza L, Murillo C et al. Peripheral T-cell therapy regimen (R-miniCHOP) in elderly patients older than 80 years
lymphoma with follicular T-cell markers. Am J Surg Pathol 2008; 32: with diffuse large B-cell lymphoma: a multicentre, single-arm, phase 2
1787–1799. trial. Lancet Oncol 2011; 12: 460–468.
102. Feeney J, Horwitz S, Gönen M, Schöder H. Characterization of T-cell 121. Peyrade F, Bologna S, Delwail V et al. Combination of ofatumumab
lymphomas by FDG PET/CT. AJR Am J Roentgenol 2010; 195: 333–340. and reduced-dose CHOP for diffuse large B-cell lymphomas in patients

by guest
on 05 February 2018
aged 80 years or older: an open-label, multicentre, single-arm, phase 2 128. Pfreundschuh M, Müller C, Zeynalova S et al. Suboptimal dosing of rituxi-
trial from the LYSA group. Lancet Haematol 2017; 4: e46–e55. mab in male and female patients with DLBCL. Blood 2014; 123: 640–646.
122. Fields PA, Townsend W, Webb A et al. De novo treatment of diffuse 129. Gisselbrecht C, Glass B, Mounier N et al. Salvage regimens with autolo-
large B-cell lymphoma with rituximab, cyclophosphamide, vincristine, gous transplantation for relapsed large B-cell lymphoma in the rituxi-
gemcitabine, and prednisolone in patients with cardiac comorbidity: a mab era. J Clin Oncol 2010; 28: 4184–4190.
United Kingdom National Cancer Research Institute trial. J Clin Oncol 130. Crump M, Kuruvilla J, Couban S et al. Randomized comparison of
2014; 32: 282–287. gemcitabine, dexamethasone, and cisplatin versus dexamethasone,
123. Luminari S, Montanini A, Caballero D et al. Nonpegylated liposomal cytarabine, and cisplatin chemotherapy before autologous stem-cell
doxorubicin (MyocetTM) combination (R-COMP) chemotherapy in transplantation for relapsed and refractory aggressive lymphomas:
elderly patients with diffuse large B-cell lymphoma (DLBCL): results NCIC-CTG LY.12. J Clin Oncol 2014; 32: 3490–3496.
from the phase II EUR018 trial. Ann Oncol 2010; 21: 1492–1499. 131. Mounier N, El Gnaoui T, Tilly H et al. Rituximab plus gemcitabine and
124. Held G, Murawski N, Ziepert M et al. Role of radiotherapy to bulky dis- oxaliplatin in patients with refractory/relapsed diffuse large B-cell lymph-
ease in elderly patients with aggressive B-cell lymphoma. J Clin Oncol oma who are not candidates for high-dose therapy. A phase II
2014; 32: 1112–1118. Lymphoma Study Association trial. Haematologica 2013; 98: 1726–1731.
125. Morrison VA, Hamlin P, Soubeyran P et al. Diffuse large B-cell lymph- 132. Arcari A, Chiappella A, Spina M et al. Safety and efficacy of rituximab
oma in the elderly: impact of prognosis, comorbidities, geriatric assess- plus bendamustine in relapsed or refractory diffuse large B-cell lymph-
ment, and supportive care on clinical practice. An International Society oma patients: an Italian retrospective multicenter study. Leuk
of Geriatric Oncology (SIOG) expert position paper. J Geriatr Oncol Lymphoma 2016; 57: 1823–1830.
2015; 6: 141–152. 133. Pettengell R, Coiffier B, Narayanan G et al. Pixantrone dimaleate versus
126. Spina M, Balzarotti M, Uziel L et al. Modulated chemotherapy accord- other chemotherapeutic agents as a single-agent salvage treatment in
ing to modified comprehensive geriatric assessment in 100 consecutive patients with relapsed or refractory aggressive non-Hodgkin lymph-
elderly patients with diffuse large B-cell lymphoma. Oncologist 2012; oma: a phase 3, multicentre, open-label, randomised trial. Lancet Oncol
17: 838–846. 2012; 13: 696–706.
127. Bittenbring JT, Neumann F, Altmann B et al. Vitamin D deficiency im- 134. Ferreri AJ, Sassone M, Zaja F et al. Lenalidomide maintenance in patients
pairs rituximab-mediated cellular cytotoxicity and outcome of patients with relapsed diffuse large B-cell lymphoma who are not eligible for autolo-
with diffuse large B-cell lymphoma treated with but not without rituxi- gous stem cell transplantation: an open label, single-arm, multicentre phase
mab. J Clin Oncol 2014; 32: 3242–3248. 2 trial. Lancet Haematol 2017; 4: e137–e146.
Appendix Peter de Nully Brown, Department of Hematology,

Rigshospitalet, Copenhagen, Denmark; Michael Pfreundschuh,
ESMO Lymphoma Consensus Conference Panel Innere Medizin I, University Klinik des Saarlandes, Hamburg,
Members Germany; Christiane Pott, Second Medical Department,
Christian Buske, Comprehensive Cancer Center Ulm, Institute of University Hospital Schleswig-Holstein, Kiel, Germany; Norbert
Experimental Cancer Research, Department of Internal Medicine Schmitz, Department of Hematology, Oncology and Stem Cell
III, University Hospital, Ulm, Germany; Martin Dreyling, Transplantation, Asklepios Klinik St. Georg, Hamburg,
Medizinische Klinik III, Klinikum der Universität München/ Germany; Pierre Soubeyran, Department of Medical Oncology,
LMU, Munich, Germany; Andrés J.M. Ferreri, Unit of Lymphoid Institut Bergonié, Comprehensive Cancer Centre, Bordeaux,
Malignancies, Department of Onco-Hematology, IRCCS San France; Michele Spina, Division of Medical Oncology A,
Raffaele Scientific Institute, Milan, Italy; Paul Fields, Department National Cancer Institute, Aviano, Italy; Reinhard Stauder,
of Haematology, Guys and St Thomas’ and King’s College Haematology and Oncology department, Innsbruck Medical
Hospitals, London, United Kingdom; Gianluca Gaidano, University, Innsbruck, Austria; Anna Sureda Balari, Servei
Division of Hematology, Department of Translational Medicine, d’Hematologia, Institut Català d’Oncologia - Hospital Duran i
Amedeo Avogadro University of Eastern Piedmont, Novara, Reynals, Barcelona, Spain; Marek Trneny , Institute of
Italy; Valentin Goede, Department of Internal Medicine, Hematology and Blood Transfusion, Ist Department of
University Hospital Cologne, Germany; Martin Hutchings, Medicine, 1st Faculty Medicine, Charles University, General
Department of Hematology, Rigshospitalet, Copenhagen, Hospital, Prague, Czech Republic; Gustaaf van Imhoff, Section of
Denmark; Marco Ladetto, Hematology Division, Azienda Hematology, University of Groningen, Groningen, The
Ospedaliera Santi Antonio e Biagio e Cesare Arrigo, Alessandria, Netherlands; Jan Walewski, Department of Lymphoid
Italy; Steven Le Gouill, Centre Hospitalo-Universitaire de Malignancies, Maria Sklodowska-Curie Memorial Cancer Centre
Nantes, UMR892 team 10, CIC Nantes, France; Stefano and Institute of Oncology, Warsaw, Poland; Ulrich Wedding,
Luminari, Hematology, Arcispedale S. Maria Nuova, IRCCS Department of Palliative Care, University Hospital, Jena,
Reggio Emilia, and Department of Diagnostic, Clinical and Germany; Alberto Zamò, Department of Diagnostics and Public
Public Health Medicine, University of Modena and Reggio Health, University of Verona, Verona, Italy; Emanuele Zucca,
Emilia, Modena, Italy; Ulrich Mey, Department of Oncology and Lymphoma Unit, Oncology Institute of Southern Switzerland,
Haematology, Kantonsspital Graubünden, Chur, Switzerland; Ospedale San Giovanni, Bellinzona, Switzerland.

by guest
on 05 February 2018
High-grade glioma: ESMO Clinical Practice Guidelines

R. Stupp1, M. Brada2, M. J. van den Bent3, J.-C. Tonn4 & G. Pentheroudakis5 on behalf of the
1
Department of Oncology and Cancer Centre, University Hospital Zurich, Zurich, Switzerland; 2Department of Molecular and Clinical Cancer Medicine, University of
Liverpool, Clatterbridge Cancer Centre, Wirral, UK; 3Department of Neuro-Oncology, Erasmus MC Cancer Center, Rotterdam, Netherlands; 4Department of Neurosurgery,
Ludwig-Maximilians-University, Munich, Germany; 5Department of Medical Oncology, Medical School, University of Ioannina, Ioannina, Greece
incidence and epidemiology team is recommended [5, 6]. Molecular markers are useful add-
itional tools for diagnosis and treatment guidance (see below),
The yearly incidence of malignant glioma is ∼3–5/100 000 with and are of increasing importance in daily practice. Adequate
a slight predominance in males. Malignant glioma may develop tissue collection and preservation (e.g. sufficient material, fresh
at all ages, with the peak incidence being in the fifth and sixth frozen tumour tissue) should be planned prospectively (see
decades of life [1–3]. Exposure to ionising irradiation has been Table 1).
associated with increased risk of development of glioma, while
association with the use of cell phones could not be confirmed
clinical practice
in epidemiological studies. Rare hereditary syndromes carry an molecular markers
guidelines
increased risk for glioma: Cowden-, Turcot-, Lynch-, Li- Genetic loss on chromosomes 1p/19q (co-deletion or loss of
Fraumeni syndrome and neurofibromatosis type 1. heterozygosity [LOH] 1p/19q) is a consequence of a chromo-
somal translocation and describes a distinct tumour entity
characterised by a prolonged natural history irrespective of treat-
diagnosis and pathology ment, and increased sensitivity both to radiotherapy (RT) and
The commonly used World Health Organization (WHO) clas- to chemotherapy [7]. LOH 1p/19q should be evaluated to
sification distinguishes tumours according to their presumed support a diagnosis of oligodendroglioma.
cell of origin (astrocytes or oligodendrocytes), and grades Mutations of the isocitrate dehydrogenase gene (IDH) were
them from grade I–IV [4]. Grade I tumours occur mainly in recognised in 2008 as an early event in glioma genesis [8].
childhood, and grade II (or low-grade) glioma are slow-growing Mutations in the IDH gene 1 or 2 are hallmarks of low-grade
tumours but will almost invariably transform over time to a glioma; when observed in high-grade glioma, it suggests that the
more malignant phenotype. Grade III tumours (also commonly tumour has developed from a lower grade precursor lesion (sec-
referred to as anaplastic glioma) comprise anaplastic astrocy- ondary GBM) [9, 10]. Less than 10% of individuals with adult
toma, mixed anaplastic oligoastrocytoma and anaplastic GBM carry an IDH mutation, while ∼60% of grade III tumours
oligodendroglioma, while glioblastoma (GBM) represents WHO are IDH-mutated [8, 11]; IDH-mutated tumours are associated
grade IV. Tissue diagnosis is mandatory, and usually obtained by with a more favourable prognosis [12, 13]. Indeed, the survival
stereotactic biopsy or after tumour resection. GBM carries the of IDH-mutated GBM is more favourable than for non-mutated
worst prognosis, while pure oligodendroglioma has a protracted grade III astrocytoma, thus underscoring the strong prognostic
natural history and better outcome, and excellent response to value of this finding [10, 12]. The more frequent IDH1 mutation
therapy. Prognosis of mixed anaplastic oligoastrocytoma and accounts for ∼90% of all IDH mutations and can be demon-
anaplastic astrocytoma is intermediate between GBM and pure strated by immunohistochemistry, while IDH2 mutations and
anaplastic oligodendroglioma. Concordance between local diag- IDH1 mutations at other sites can only be identified by sequen-
nosis and central neuropathology review can be as low as 50%, cing. In lower grade tumours or suspected transformed glioma,
thus careful review of the histology by an expert neuropathology IDH sequencing should be carried out if staining by immuno-
histochemistry with the anti-IDH antibody (which recognises
*Correspondence to: ESMO Guidelines Working Group, ESMO Head Office, Via L. Taddei
R132H mutation) is negative.
4, CH-6962 Viganello-Lugano, Switzerland. Epigenetic silencing of the methyl-guanine methyl transferase
E-mail: clinicalguidelines@esmo.org (MGMT) gene promoter by gene promoter methylation suggests
†
a partial inability of the tumour to repair the chemotherapy-
Approved by the ESMO Guidelines Working Group: December 2004, last update March
2014. This publication supersedes the previously published version—Ann Oncol 2010; induced DNA damage [14]. In retrospective analyses, MGMT
21(Suppl 5): v190–v193. methylation has been correlated with a response to or benefit

by guest
on 05 February 2018
Table 1. Expected survival of glioma

limited to the central nervous system and distant metastases are
virtually non-existent; thus, staging focuses on imaging of the
Grade and cell type Median survival brain, ideally by magnetic resonance imaging (MRI). The spine
Grade II
and cerebrospinal fluid (CSF) are not routinely assessed in the
Astrocytoma 7–10 years absence of clinical symptoms. Ideally, the extent of tumour re-
Oligodendrogliomaa >10–15 years section and determination of residual disease should be assessed
Grade III after surgery; however, this assessment must be carried out
Anaplastic astrocytoma 3.5 years within 24–48 h in order to distinguish post-surgical contrast en-
Anaplastic oligodendrogliomaa >10 years hancement from residual tumour. Lower tumour grade, radical
Grade IV tumour resection, younger age (<50 years), good performance
Glioblastoma 15 months, 2-year survival 27% status and an intact neurological function are favourable prog-
MGMT nostic factors. Determination of molecular markers (discussed
Methylated 23 months, 2-year survival: 49% above) will identify patients with a more favourable prognosis or
Unmethylated 13 months, 2-year survival: 12% better chance of response to alkylating agent chemotherapy.
a
With LOH 1p/19q. MGMT, methyl-guanine methyl transferase
disease management and treatment plan
general management
of alkylating agent chemotherapy [II, B] [15, 16]. MGMT deter- Patients should be evaluated and the treatment plan determined
mination by immunohistochemistry lacks standardisation, by a specialised multidisciplinary team including neurosur-
reproducibility and, most importantly, correlation with clinical geons, medical and radiation oncologists, but also an expert
outcome [III, C], while MGMT status determination by neuropathologist and neuroradiologist. Special consideration
methylation-specific polymerase chain reaction (MSP) has been should be given to performance status and neurological func-
largely standardised and allows for reproducible results [17, 18]. tion. Corticosteroids (usually dexamethasone 8–16 mg/day, but
A predictive value of a methylated MGMT status for response to lower doses may be just as effective) allow for rapid reduction of
or benefit from alkylating agent chemotherapy and thus tumour-associated oedema and improve clinical symptoms.
improved outcome has been demonstrated repeatedly in GBM Patients’ glucose levels need to be monitored. Steroids are not
[19, 20]. However, in lower grade IDH-mutated tumours, MGMT necessary in patients without increased intracranial pressure or
methylation identifies a sub-population with a better prognosis in the absence of oedema-associated neurological deficits. There
irrespective of the applied treatment [18]. is no need for prolonged steroid therapy after tumour resection
The determination of these markers is now part of routine or for prophylaxis during radiotherapy in asymptomatic patients.
practice; however, whether MGMT methylation status deter- Rapid tapering and discontinuation of corticosteroids is recom-
mination is carried out for an individual patient depends on the mended in order to avoid toxicity associated with prolonged ex-
diagnostic and therapeutic context (Table 2). In the absence of posure to steroids, e.g. myopathy and weakness, lymphopenia
clinical consequences and better alternative treatments, MGMT and risk of infection, osteoporosis and Cushing syndrome.
status assessment may not be necessary for every patient [V, D]; Anti-epileptic therapy is indicated in patients presenting with
nevertheless, in elderly patients improved outcome with adapted seizures; however, prophylactic use of anticonvulsants outside
treatment strategy [i.e. temozolomide (TMZ) chemotherapy for the perioperative phase is not indicated [III, C]. After tumour
MGMT-methylated tumours, radiotherapy for unmethylated resection, the indication for anti-seizure therapy should be revis-
tumours] has been demonstrated in two randomised trials [II, ited only if seizures occur [22, 23]. First generation anti-epileptic
A] [19, 20]. In current clinical trials, MGMT status is always drugs ( phenytoin, carbamazepine, phenobarbital and their deri-
required as a stratification factor. For a detailed discussion, the vatives) are strong inducers of the hepatic metabolism, and may
reader is referred to recent extensive reviews [18]. interfere with medications including many commonly used
These markers are neither mutually exclusive nor entirely in- chemotherapy agents (but not with TMZ). Agents such as lamo-
dependent. MGMT promoter methylation is highly associated trigine, levetiracetam, pregabalin or valproic acid are preferred.
with LOH 1p/19q and IDH mutations [11]. MGMT methylation Glioma patients are at increased risk of thromboembolic events
is predictive for benefit from alkylating agent chemotherapy in due to a tumour-induced hyper-coagulable state, but also as a
GBM, while it confers a largely prognostic value in anaplastic consequence of neurological deficits, immobilisation and steroid
grade III tumours. A recent report from the Neuro-Oncology use [24]. Prophylactic anticoagulation is not recommended;
Working Group (NOA) of the German Cancer Society however, a low threshold for excluding deep vein thrombosis and
confirmed a predictive value of MGMT methylation for benefit pulmonary emboli is indicated when suspicious symptoms occur.
from chemotherapy in patients with a wild-type IDH, independ- The presence of a brain tumour is not a contraindication for
ent of tumour grade [21]. the use of standard anticoagulants in patients with proven
thrombosis.
staging and risk assessment newly diagnosed patients

Although glioma are invasive tumours with a strong propensity Surgery is commonly the initial therapeutic approach for
of glioma cells to migrate, tumour dissemination remains tumour debulking and obtaining tissue for diagnosis. Tumour
iii | Stupp et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
Annals of Oncology
Table 2. Clinically relevant molecular markers (adapted from [70] with permission from Elsevier)
IDH1/2 mutation 1p/19q co-deletion MGMT promoter methylation
Biological consequence Increased levels of 2-hydroxyglutarate, link to Unclear, candidate genes CIC and FUBP1 Reduced DNA repair, association with G-CIMP phenotype
G-CIMP phenotype under investigation in IDH1/2-mutant tumours
Methods of assessment Immunohistochemistry for IDH1-R132H; if FISH, micro-satellite analysis for loss of MSP, MS or bisulphite (pyro)sequencing
negative, gene sequencing heterozygosity
Frequency
WHO grade II
Diffuse astrocytoma 70%–80% 15% 40%–50%
Oligodendroglioma/ 70%–80% 30%–60% 60%–80%
oligoastrocytoma
WHO grade III
Anaplastic astrocytoma 50%–70% 15% 50%
Anaplastic 50%–80% 50%–80% 70%

Oligodendroglioma/
oligoastrocytoma
WHO grade IV
Glioblastoma 5%–10% <5% 35%
Diagnostic role DD glioma versus gliosis Pathognomonic for oligodendroglioma None
Typical for transformed low-grade glioma
Prognostic role Protracted natural history in IDH-mutated Protracted natural history in 1p/19q codeleted Prognostic for anaplastic glioma patients (possibly with IDH
tumours tumours mutations) treated with radiotherapy or alkylating drugs
Predictive role Absence of mutation suggests predictive role Prolongation of survival with early chemotherapy Predictive in GBM for benefit from alkylating chemotherapy
for MGMT promoter methylation in 1p/19-co-deleted oligodendrogliomas Elderly GBM: MGMT-methylated → TMZ MGMT-
unmethylated → RT
doi:10.1093/annonc/mdu050 | iii
WHO, World Health Organization; G-CIMP, glioma CpG island methylator phenotype; IDH, isocitrate dehydrogenase; DD, differential diagnosis; MGMT, methyl-guanine methyl transferase; CIC,
capicua transcriptional repressor; FUBP1, far upstream element (FUSE) binding protein 1; FISH, fluorescence in situ hybridisation; DNA, deoxyribonucleic acid; MSP, methylation-specific PCR; MS,
methylation-specific; GBM, glioblastoma; TMZ, temozolomide; RT, radiotherapy.

by guest
on 05 February 2018
resection is of prognostic value; it may be beneficial to attempt Fractionated localised radiotherapy (60 Gy, 30–33 fractions of
maximal tumour resection provided that neurological function is 1.8–2 Gy, or equivalent doses/fractionations) is part of the stand-
not compromised by the extent of resection [II, C] [25]. An in- ard treatment after resection or biopsy [I, A] [31]. Escalating
crease in the complete resection rate and improvement in progres- doses beyond 60 Gy have not been shown to be of value. In
sion-free survival (PFS) have been shown when surgery is carried elderly patients or patients with a low performance status, shorter
out under blue light with the use of fluorescent marking of the hypo-fractionated regimens (e.g. 40 Gy in 15 fractions) are com-
tumour by 5-amino-laevulinic acid (5-ALA) [26]. When micro- monly proposed [II, B] [32]. Radiotherapy (28 × 1.8 Gy, 50 Gy)
surgical resection is not safely feasible (e.g. due to location of the in patients >70 years of age was superior to best supportive care
tumour or impaired clinical condition of the patient), a biopsy alone in a randomised phase III trial [II, B] [33]. Exclusive TMZ
should be carried out. In experienced hands, the diagnostic yield chemotherapy has shown an improved outcome compared with
is >95% [18]. A sufficient amount of tissue should be obtained for radiotherapy alone in elderly patients with a methylated MGMT
molecular analyses, and open biopsies may yield more tissue to gene promoter in two randomised trials [II, A] [19, 20] (see
allow this compared with stereotactic biopsies. On freshly frozen Figure 1).
specimens, molecular genetic analyses (LOH 1p/19q, MGMT pro-
moter methylation) can be carried out even with small amounts
of tissue, and this should be planned beforehand [27]. glioblastoma (WHO grade IV)
Implantation of chemotherapy-impregnated wafers (carmus- Concomitant and adjuvant TMZ chemotherapy in addition to
tine polymers) into the resection cavity before radiotherapy has radiotherapy (TMZ/RT → TMZ) significantly improved
been shown to marginally improve median survival compared median, 2- and 5-year survival in a large randomised trial, and
with radiotherapy alone [II, B] [28]; however, no prospective is the current standard of care for patients with GBM up to age
data are available when compared with current standard TMZ/ 70 [34, 35] [I, A], or fit elderly patients older than 70 years [II,
RT (see below) [29]. An increase in wound healing and infec- B] [36]. TMZ is administered daily (7 days a week) during radio-
tious complications has been reported. The combination of car- therapy and for 5 days every 4 weeks for six cycles as maintenance
mustine wafers and TMZ/RT has not been assessed in (adjuvant) treatment after the end of radiation. MGMT gene
prospective trials, a retrospective comparison failed to demon- promoter methylation has been demonstrated as the strongest
strate additive efficacy [IV, D] [30]. prognostic marker for outcome, and the added benefit of TMZ
Maximal Resection or Biopsy

Molecular markers Treatment Options
Tissue diagnosis
Alternative strategies in
unmethyl. investigational protocols
Rapid onset:
1° GBM MGMT
methyl.
TMZ/RT TMZ
V
eI
Prior lower grade

ad
IDH mutation
glioma/symptoms > 6 IDH
Gr
High-Grade Glioma
mo (MGMT) Prognosis
2° GBM
RT†
Anaplastic IDH IDH mutation
or
Astrocytoma (MGMT) Prognosis
Gr
Chemo†¶
ad
n
tio
e
ele
III
d
co
No
Anapl. Oligo- 1p19q RT Chemo§ or
codeletion +
dendroglioma (IDH) Chemo†¶
Figure 1. Treatment algorithm for high-grade glioma. GBM, glioblastoma; 1°, primary; 2°, secondary; MGMT, methyl-guanin methyltransferase gene pro-
moter methylation status; IDH, isocitrate dehydrogenase gene mutation; LOH 1p/19q, loss of heterozygosity of chromosomes 1 and 19; RT, radiotherapy;
Chemo, chemotherapy with either PCV or temozolomide; PCV, procarabine, lomustine [CCNU], vincristine; TMZ, temozolomide. †, RT or chemotherapy
first, and at progression chemo or RT, respectively (according to NOA-04 study). ¶, no randomised data for oligodendroglioma, only class III + V evidence.
§
, subgroup analysis and long-term follow-up of RTOG and EORTC randomised studies demonstrated prolonged survival for patients treated with RT → PCV.
iii | Stupp et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
chemotherapy appears largely restricted to this subgroup [16]. to 4.5 years, time to failure of both chemotherapy and radiation
Feasibility of upfront testing and stratification by MGMT status was similar whether patients were treated initially with chemo-
has been demonstrated in large prospective trials, and alternative therapy (and received RT at first progression) or were treated with
strategies for patients with unmethylated MGMT are proposed to initial RT (and received chemotherapy at progression) [I, A] [42].
patients as part of ongoing clinical trials. In the absence of effect- An updated report is awaited. No difference of efficacy was appar-
ive better alternatives, TMZ/RT → TMZ remains the standard of ent between PCV or TMZ chemotherapy [II, B].
care for the majority of GBM patients [II, B].
Owing to the overall inferior prognosis of elderly patients, diffuse glioma (WHO grade II)
shorter, simpler and less intensive treatment strategies have been
Diffuse gliomas are low-grade tumours and comprise the histo-
investigated in frail or elderly patients approximately >65 years of
logical subtypes of astrocytoma, oligoastrocytoma and oligo-
age. Although hypo-fractionated RT (10 × 3.4 or 15 × 2.66 Gy)
dendroglioma. Although they have a protracted natural history
has been shown equivalent to standard fractionated irradiation
of years or occasionally decades, and may not need immediate
with 30 fractions (60 Gy) in one trial [II, A] [32], the 6-week
therapy, these tumours invariably recur and often transform
regimen was inferior compared with hypo-fractionated RT in
into a higher histological grade.
patients >70 years in a subsequent larger randomised trial [19, 32],
The diagnosis is often made after an initial seizure is observed
supporting the practice of administering RT as a short hypo-
as a presenting symptom, or when imaging of the brain is
fractionated course in elderly patients. Primary chemotherapy
carried out to investigate neurological deficits, changes in mood
versus radiotherapy has been compared in two randomised phase
or personality. At that time, the tumour may already have grown
III trials reported in 2012 [19, 20]. Consistently, both trials have
to a substantial size without causing evident neurological dys-
demonstrated a poor median overall survival (OS) of 7–9 months.
function, due to its slow growth over many years.
MGMT promoter methylation was strongly predictive for a
Radical surgery both for diagnostic and therapeutic purposes
benefit from TMZ chemotherapy, suggesting that patients with a
is the backbone of low-grade glioma management. It is assumed
methylated MGMT are better treated with chemotherapy, while
that, by resection of all visible tumour tissue, the risk of subse-
patients with an unmethylated MGMT should be managed
quent malignant transformation is substantially reduced [III, B].
with radiotherapy alone [II, A] [19, 20]. In the absence of informa-
In an exemplary randomised trial, investigators from the
tion on the MGMT status, the appropriate treatment is hypo-
EORTC have demonstrated that an ‘expectative’ or cautious ap-
fractionated radiotherapy [II, A] [33]. A randomised, controlled
proach—delaying additional radiotherapy until symptoms prevail
trial (clinicaltrials.gov: NCT#00482677) investigating hypo-
and the tumour has grown to >5–6 cm—will substantially delay
fractionated RT ± concomitant and maintenance TMZ completed
or eliminate the need for irradiation in a significant number of
recruitment in 2013; first results are expected in late 2014.
patients [I, A] [43]. Adjuvant radiotherapy for a total dose of
Recent large phase III trials of novel treatment approaches
50.4 Gy (28 × 1.8 Gy) is the accepted standard of care [I, A]. Two
failed to improve survival of patients with newly diagnosed
randomised trials investigated lower (45–50.4 Gy) versus higher
GBM. Dose intensification of the adjuvant or maintenance
dose (59.4–64.8 Gy) of irradiation; the lower dose was equivalent
TMZ therapy [36], the addition of cilengitide [37] or the add-
to the higher dose with less toxicity [44, 45]. Based on these trials,
ition of bevacizumab [38, 39] did not translate into prolongation
a number of negative prognostic factors were identified: tumour
of OS. The clinical value of the observed prolonged PFS with
size >5–6 cm, tumours crossing the mid-line, absence of oligo-
bevacizumab remains controversial. The reported median OS
dendroglial histology, age >40 years and neurological deficits
from surgery in these clinical trials is ∼16–18 months for both
before surgery [46]. Three or more of these risk factors are con-
the control and experimental arms, and no substantial improve-
sidered an indication for adjuvant radiotherapy. A contemporary
ment over the results of the initial European Organisation for
randomised trial compared adjuvant single-agent TMZ chemother-
Research and Treatment of Cancer (EORTC)/National Cancer
apy (without radiation) with standard radiotherapy. Preliminary
Institute of Canada TMZ trial [median 15 months, 95% confi-
results were presented at the American Society of Clinical Oncology
dence interval (CI) 13–17 months] was demonstrated.
meeting in 2013, and demonstrated that PFS analysis favours radio-
therapy; however, follow-up is too early for firm conclusions and
anaplastic astrocytoma, oligoastrocytoma and survival results are not yet mature [47]. Molecular characterisation
oligodendroglioma (WHO grade III) of tumours may allow identification of subgroups of patients benefi-
tting from one of the other treatment modalities, with 1p/19q co-
Anaplastic astrocytoma and oligoastrocytoma have a better prog-
deleted tumours having a similar PFS after TMZ chemotherapy
nosis compared with glioblastoma. Standard therapy consists of
compared with radiotherapy. The Radiation Therapy Oncology
adjuvant radiotherapy up to 60 Gy after surgery. The value of con-
Group (RTOG) has conducted a randomised trial in a similar
comitant and/or maintenance chemotherapy with TMZ has not
population exploring the efficacy of adjuvant PCV chemotherapy
been tested prospectively [V, D]. Long-term follow-up (10 years)
after radiotherapy. At the time of first presentation, the follow-up
of randomised clinical trials demonstrated prolonged survival
data were immature; thus, definitive conclusions are not yet
with (neo)adjuvant PCV chemotherapy (procarbazine, lomustine
possible [48].
[CCNU], vincristine) in newly diagnosed anaplastic oligoastrocy-
toma and oligodendroglioma [40, 41], but benefit was mainly
present in prognostically favourable molecular subgroups of recurrent disease
patients (in particular co-deletion of 1p/19q) [II, B]. In a more Some benefit of chemotherapy has been shown for patients with
recent randomised trial with follow-up data currently available up an adequate performance status who have not received prior
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu050 | iii

by guest
on 05 February 2018
adjuvant cytotoxic therapy. Relapsing low-grade astrocytoma, (FLAIR)-weighted MRI are evaluated [67]. Furthermore, an as-
anaplastic astrocytomas and oligodendrogliomas are more likely sessment of neurological function and corticosteroid use is
than GBM to respond to TMZ chemotherapy [III, B] [49, 50]. For included. The recent introduction of anti-angiogenic and vascu-
patients progressing after prior chemotherapy, there is no estab- lar modifying agents led to a reassessment of response criteria,
lished chemotherapy regimen available and patients are best in which, in addition to contrast enhancement, tumour exten-
treated within investigational clinical protocols. Chemotherapy sion on T2- and FLAIR-weighted MRI are to be evaluated [67].
with PCV or single-agent nitrosourea therapy may achieve similar In cases of doubtful differential diagnosis between tumour re-
tumour control rates compared with TMZ [51–53]. Randomised currence and treatment-induced unspecific changes (especially
trials in recurrent glioblastoma have failed to demonstrate measur- after multimodal therapy), magnetic resonance spectroscopy
able anti-tumour efficacy of epidermal growth factor receptor and positron-emission tomography investigation using an
(EGFR) inhibition by erlotinib or platelet-derived growth factor amino acid tracer (e.g. methionine, fluoro-ethyl-tyrosine) may
receptor inhibition by imatinib in an unselected patient popula- be helpful [III, B] [68]. However, often both residual tumour
tion [II, C] [54, 55]. High response rates and a steroid-sparing and necrotic and inflammatory changes are present in the same
effect have been observed with the administration of bevacizumab lesion.
(± irinotecan); however, the effect is frequently short-lived and
may be due exclusively to changes in vascular permeability; the
effect on life expectancy remains unknown [III, C] [56, 57]. personalised medicine
Randomised trials evaluating the pan-vascular endothelial growth Recent investigation of molecular markers, gene expression,
factor receptor inhibitor cediranib or protein kinase C inhibitor whole genome sequencing and epigenetics has enabled identifi-
enzastaurin failed to demonstrate improved outcome, and provide cation of patient subgroups according to pathway activation,
contemporary data on the limited but confirmed single-agent specific aberrations or pathognomonic molecular characteristics
efficacy of lomustine (CCNU) [52, 53]. Applying alternating [69]. The clinical utility of IDH mutations, LOH 1p/19q and
electric fields—tumour-treating fields (TTFs)—using a battery- MGMT promoter methylation in prognostication and their
powered device connected to electrodes placed on the patient’s predictive value and role in clinical decision-making have been
scalp—was compared with physicians’ choice of chemotherapy in discussed in the relevant subheadings above. ATRX mutations
a randomised trial in recurrent disease [58]. TTF failed to prolong have been identified to be specific for astrocytic lineage. Other
survival compared with second-line chemotherapy [I, A]. markers like EGFR overexpression and EGFR variant III
There may be an indication for a re-operation [IV, C], in par- (EGFRviii) mutation are characteristic for GBM that carry an in-
ticular in patients where the recurrent tumour exerts an acute ferior prognosis (reviewed in [70]). Targeting the EGFR with
mass effect. However, there are no prospective data available on specific inhibitors and vaccination against EGFRviii are current-
the impact of repeat surgery on OS. Retrospective analyses on ly under clinical investigation [71]. Rare activating BRAF muta-
selected patients did not identify surgery for recurrent disease as tions can be identified in 3%–5% of GBM; whether the BRAF
a significant prognostic factor for prolonged survival [59, 60]. inhibitors are effective (alone or in combination) in this sub-
Repeat surgery and implantation of carmustine-impregnated group of tumours is subject of ongoing clinical research proto-
polymers may lead to marginal prolongation of survival com- cols. All planned investigational protocols are being designed
pared with placebo, as demonstrated in a randomised trial for specific subgroups, selecting and enriching for specific
conducted in the early 1990s [II, B] [61]. molecular aberrations when targeting activated pathways.
Re-irradiation is being considered increasingly for recurrent
small tumours [IV, C], although there is considerable doubt
about its benefit and the literature lacks prospective and com- follow-up and long-term implications
parative trials [62, 63]. The few limited size case series do not
allow for any conclusion. Follow-up consists of a clinical evaluation with particular atten-
tion to neurological function, seizures and corticosteroid use.
Patients should be tapered off steroid use as early as possible.
response evaluation Venous thrombotic events occur frequently in patients with
MRI is the preferred imaging method. Increased contrast en- residual or recurrent tumours. Laboratory tests are not indicated
hancement and presumed tumour progression on imaging 4–12 unless the patient is receiving chemotherapy (blood counts),
weeks after the end of radiotherapy may be due to a reactive corticosteroids (glucose) or anti-epileptic drugs (blood count,
process following radiotherapy ( pseudo-progression) [64, 65]. liver function tests). MRI every 3–4 months is standard practice
Early presumed progression after the end of radiotherapy outside clinical trials, unless more frequent monitoring is clinic-
should raise the possibility of pseudo-progression, and chemo- ally indicated.
therapy should be continued as planned with repeat imaging
after 6–8 weeks. An incidence of >60% pseudo-progression has
been reported among patients with early radiological progres-
note
sion after radiochemotherapy [66]. A summary of recommendations is provided in Table 3. Levels
Response to chemotherapy is currently evaluated according to of evidence and grades of recommendation have been applied
the 2D Response Assessment in Neuro-Oncology (RANO) cri- using the system shown in Table 4. Statements without grading
teria, in which in addition to contrast enhancement, tumour ex- were considered justified standard clinical practice by the expert
tension on T2- and fluid-attenuated inversion recovery authors and the ESMO faculty.
iii | Stupp et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
Table 3. Conclusions and recommendations conflict of interest
Histological diagnosis is mandatory and should include sufficient RS has reported that he is the president of EORTC, and has
tissue for molecular tumour characterisation acted the principal investigator in pivotal studies investigating
temozolomide, cilengitide or NovoTTF. He has received honor-
Methyl-guanine methyl transferase (MGMT) promoter methylation, aria for advisory board participation from Roche/Genentech,
isocitrate dehydrogenase gene (IDH) mutations and co-deletions and MSD/Merck & Co. He also received remuneration for par-
of 1p/19q are commonly determined depending on the
ticipation in Advisory Boards organised by Roche, Amgen,
histological and clinical context
AstraZeneca, Celgene and Sanofi. GP has reported research
Surgery to the extent feasible is the first therapeutic intervention for
sponsored by Amgen, Teva, Roche, Boehringer Ingelheim and
all malignant glioma
Sanofi. JCT has reported speaker’s honoraria and Advisory
For glioblastoma, combined modality therapy with temozolomide
(TMZ) and radiotherapy remains the standard of care
boards of Roche, MerckSereno and Medac. MV has reported
In elderly patients and a methylated MGMT promoter, exclusive
that he is conducting research sponsored by Roche and by
TMZ chemotherapy may be considered, while (hypo-fractionated) AbbVie. He is a member of the speaker’s bureau of MSD, and
radiotherapy is the treatment of choice for patients with an he has also received honoraria from Roche, AMGEN, Actelion,
unmethylated gene promoter Merck AG, Celldex and AbbVie. MB has reported no potential
Anaplastic oligodendroglioma should receive radiotherapy and conflicts of interest.
adjuvant chemotherapy, while no benefit has been demonstrated
for adjuvant PCV chemotherapy (procarbazine, lomustine
[CCNU], vincristine) in anaplastic astrocytoma references
In anaplastic glioma, treatment with adjuvant chemotherapy alone
and irradiation only at the time of tumour progression is 1. Preusser M, de Ribaupierre S, Wöhrer A et al. Current concepts and management
equivalent to the standard sequence of initial radiotherapy and of glioblastoma. Ann Neurol 2011; 70: 9–21.
salvage chemotherapy at the time of progression 2. Wen PY, Kesari S. Malignant gliomas in adults. N Engl J Med 2008; 359:
492–507.
3. Ferlay J, Parkin DM, Steliarova-Foucher E. Estimates of cancer incidence and
mortality in Europe in 2008. Eur J Cancer 2010; 46: 765–781.
4. Louis DN, Ohgaki H, Wiestler O et al. WHO Classification of Tumours of the Central
Nervous System, 4th Edn. Lyon: International Agency for Research on Cancer
2007.
5. Kros JM, Gorlia T, Kouwenhoven MC et al. Panel review of anaplastic
States Public Health Service Grading Systema) oligodendroglioma from European Organization For Research and Treatment of
Cancer Trial 26951: assessment of consensus in diagnosis, influence of 1p/19q
Levels of evidence
loss, and correlations with outcome. J Neuropathol Exp Neurol 2007; 66:
I Evidence from at least one large randomised, controlled trial of 545–551.
good methodological quality (low potential for bias) or 6. Hegi ME, Janzer RC, Lambiv WL et al. Presence of an oligodendroglioma-like
meta-analyses of well-conducted randomised trials without component in newly diagnosed glioblastoma identifies a pathogenetically
heterogeneity heterogeneous subgroup and lacks prognostic value: central pathology review of
the EORTC_26981/NCIC_CE.3 trial. Acta Neuropathol 2012; 123: 841–852.
7. Jenkins RB, Blair H, Ballman KV et al. A t(1;19)(q10;p10) mediates the combined
deletions of 1p and 19q and predicts a better prognosis of patients with
oligodendroglioma. Cancer Res 2006; 66: 9852–9861.
heterogeneity
8. Yan H, Parsons DW, Jin G et al. IDH1 and IDH2 mutations in gliomas. N Engl J
III Prospective cohort studies Med 2009; 360: 765–773.
IV Retrospective cohort studies or case–control studies 9. Sturm D, Witt H, Hovestadt V et al. Hotspot mutations in H3F3A and IDH1 define
V Studies without control group, case reports, experts opinions distinct epigenetic and biological subgroups of glioblastoma. Cancer Cell 2012;
Grades of recommendation 22: 425–437.
10. Hartmann C, Hentschel B, Simon M et al. Long-term survival in primary
A Strong evidence for efficacy with a substantial clinical benefit, glioblastoma with versus without isocitrate dehydrogenase mutations. Clin Cancer
strongly recommended Res 2013; 19: 5146–5157.
B Strong or moderate evidence for efficacy but with a limited 11. Sanson M, Marie Y, Paris S et al. Isocitrate dehydrogenase 1 codon 132 mutation
clinical benefit, generally recommended is an important prognostic biomarker in gliomas. J Clin Oncol 2009; 27:
C Insufficient evidence for efficacy or benefit does not outweigh 4150–4154.
the risk or the disadvantages (adverse events, costs, ...), 12. Hartmann C, Hentschel B, Wick W et al. Patients with IDH1 wild type anaplastic
optional astrocytomas exhibit worse prognosis than IDH1-mutated glioblastomas, and IDH1
D Moderate evidence against efficacy or for adverse outcome, mutation status accounts for the unfavorable prognostic effect of higher age:
generally not recommended implications for classification of gliomas. Acta Neuropathol 2010; 120: 707–718.
E Strong evidence against efficacy or for adverse outcome, never 13. Schittenhelm J, Mittelbronn M, Meyermann R et al. Confirmation of R132H
recommended mutation of isocitrate dehydrogenase 1 as an independent prognostic factor in
anaplastic astrocytoma. Acta Neuropathol 2011; 122: 651–652.
a 14. Hegi ME, Liu L, Herman JG et al. Correlation of O6-methylguanine
methyltransferase (MGMT) promoter methylation with clinical outcomes in
glioblastoma and clinical strategies to modulate MGMT activity. J Clin Oncol 2008;
26: 4189–4199.
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu050 | iii

by guest
on 05 February 2018
15. Hegi ME, Diserens AC, Godard S et al. Clinical trial substantiates the predictive 37. Stupp R, Hegi ME, Gorlia T et al. Cilengitide combined with standard treatment for
value of O-6-methylguanine-DNA methyltransferase promoter methylation in patients with newly diagnosed glioblastoma and methylated O6-methylguanine-
glioblastoma patients treated with temozolomide. Clin Cancer Res 2004; 10: DNA methyltransferase (MGMT) gene promoter: key results of the multicenter,
1871–1874. randomized, open-label, controlled, phase III CENTRIC study. ASCO Meeting
16. Hegi ME, Diserens AC, Gorlia T et al. MGMT gene silencing and benefit from Abstracts. J Clin Oncol 2013; 31: LBA2009.
temozolomide in glioblastoma. N Engl J Med 2005; 352: 997–1003. 38. Gilbert MR, Dignam JJ, Armstrong TS et al. A randomized trial of bevacizumab for
17. Preusser M, Charles Janzer R, Felsberg J et al. Anti-O6-methylguanine- newly diagnosed glioblastoma. N Engl J Med 2014; 370: 699–708.
methyltransferase (MGMT) immunohistochemistry in glioblastoma multiforme: 39. Chinot OL, Wick W, Mason W et al. Bevacizumab plus radiotherapy-temozolomide
observer variability and lack of association with patient survival impede its use as for newly diagnosed glioblastoma. N Engl J Med 2014; 370: 709–722.
clinical biomarker. Brain Pathol 2008; 18: 520–532. 40. van den Bent MJ, Brandes AA, Taphoorn MJ et al. Adjuvant procarbazine,
18. Weller M, Stupp R, Reifenberger G et al. MGMT promoter methylation in malignant lomustine, and vincristine chemotherapy in newly diagnosed anaplastic
gliomas: ready for personalized medicine? Nat Rev Neurol 2010; 6: 39–51. oligodendroglioma: long-term follow-up of EORTC Brain Tumor Group study
19. Malmström A, Grønberg BH, Marosi C et al. Temozolomide versus standard 26951. J Clin Oncol 2013; 31: 344–350.
6-week radiotherapy versus hypofractionated radiotherapy in patients older than 41. Cairncross G, Wang M, Shaw E et al. Phase III trial of chemoradiotherapy for
60 years with glioblastoma: the Nordic randomised, phase 3 trial. Lancet Oncol anaplastic oligodendroglioma: long-term results of RTOG 9402. J Clin Oncol
2012; 13: 916–926. 2013; 31: 337–343.
20. Wick W, Platten M, Meisner C et al. Temozolomide chemotherapy alone versus 42. Wick W, Hartmann C, Engel C et al. NOA-04 randomized phase III trial of
radiotherapy alone for malignant astrocytoma in the elderly: the NOA-08 sequential radiochemotherapy of anaplastic glioma with procarbazine, lomustine,
randomised, phase 3 trial. Lancet Oncol 2012; 13: 707–715. and vincristine or temozolomide. J Clin Oncol 2009; 27: 5874–5880.
21. Wick W, Meisner C, Hentschel B et al. Prognostic or predictive value of MGMT 43. van den Bent MJ, Afra D, de Witte O et al. Long-term efficacy of early versus
promoter methylation in gliomas depends on IDH1 mutation. Neurology 2013; 81: delayed radiotherapy for low-grade astrocytoma and oligodendroglioma in adults:
1515–1522. the EORTC 22845 randomised trial. Lancet 2005; 366: 985–990.
22. Rossetti AO, Stupp R. Epilepsy in brain tumor patients. Curr Opin Neurol 2010; 23: 44. Karim AB, Maat B, Hatlevoll R et al. A randomized trial on dose-response in
603–609. radiation therapy of low-grade cerebral glioma: European Organization for
23. Weller M, Stupp R, Wick W. Epilepsy meets cancer: when, why, and what to do Research and Treatment of Cancer (EORTC) study 22844. Int J Radiat Oncol Biol
about it? Lancet Oncol 2012; 13: e375–e382. Phys 1996; 36: 549–556.
24. Perry JR. Thromboembolic disease in patients with high-grade glioma. Neuro 45. Shaw E, Arusell R, Scheithauer B et al. Prospective randomized trial of low- versus
Oncol 2012; 14(Suppl. 4): iv73–iv80. high-dose radiation therapy in adults with supratentorial low-grade glioma: initial report
25. Lacroix M, Abi-Said D, Fourney DR et al. A multivariate analysis of 416 patients of a North Central Cancer Treatment Group/Radiation Therapy Oncology Group/Eastern
with glioblastoma multiforme: prognosis, extent of resection, and survival. J Cooperative Oncology Group study. J Clin Oncol 2002; 20: 2267–2276.
Neurosurg 2001; 95: 190–198. 46. Pignatti F, van den Bent M, Curran D et al. Prognostic factors for survival in adult
26. Stummer W, Pichlmeier U, Meinel T et al. Fluorescence-guided surgery with 5- patients with cerebral low-grade glioma. J Clin Oncol 2002; 20: 2076–2084.
aminolevulinic acid for resection of malignant glioma: a randomised controlled 47. Baumert BG, Mason WP, Ryan G et al. Temozolomide chemotherapy versus
multicentre phase III trial. Lancet Oncol 2006; 7: 392–401. radiotherapy in molecularly characterized (1p loss) low-grade glioma: a randomized
27. Stupp R, Hegi ME, Gilbert MR, Chakravarti A. Chemoradiotherapy in malignant phase III intergroup study by the EORTC/NCIC-CTG/TROG/MRC-CTU (EORTC
glioma: standard of care and future directions. J Clin Oncol 2007; 25: 22033–26033). ASCO Meeting Abstracts. J Clin Oncol 2013; 31: abstr 2007.
4127–4136. 48. Shaw EG, Wang M, Coons SW et al. Randomized trial of radiation therapy plus
28. Westphal M, Hilt DC, Bortey E et al. A phase 3 trial of local chemotherapy with procarbazine, lomustine, and vincristine chemotherapy for supratentorial adult low-
biodegradable carmustine (BCNU) wafers (Gliadel wafers) in patients with primary grade glioma: initial results of RTOG 9802. J Clin Oncol 2012; 30: 3065–3070.
malignant glioma. Neuro Oncol 2003; 5: 79–88. 49. Yung WK, Prados MD, Yaya-Tur R et al. Multicenter phase II trial of temozolomide
29. Gutenberg A, Lumenta CB, Braunsdorf WE et al. The combination of carmustine in patients with anaplastic astrocytoma or anaplastic oligoastrocytoma at first
wafers and temozolomide for the treatment of malignant gliomas. A relapse. Temodal Brain Tumor Group. J Clin Oncol 1999; 17: 2762–2771.
comprehensive review of the rationale and clinical experience. J Neurooncol 2013; 50. Taal W, Dubbink HJ, Zonnenberg CB et al. First-line temozolomide chemotherapy
113: 163–174. in progressive low-grade astrocytomas after radiotherapy: molecular characteristics
30. Noël G, Schott R, Froelich S et al. Retrospective comparison of chemoradiotherapy in relation to response. Neuro Oncol 2011; 13: 235–241.
followed by adjuvant chemotherapy, with or without prior gliadel implantation 51. Brada M, Stenning S, Gabe R et al. Temozolomide versus procarbazine, lomustine,
(carmustine) after initial surgery in patients with newly diagnosed high-grade and vincristine in recurrent high-grade glioma. J Clin Oncol 2010; 28: 4601–4608.
gliomas. Int J Radiat Oncol Biol Phys 2012; 82: 749–755. 52. Wick W, Puduvalli VK, Chamberlain MC et al. Phase III study of enzastaurin
31. Walker MD, Green SB, Byar DP et al. Randomized comparisons of radiotherapy compared with lomustine in the treatment of recurrent intracranial glioblastoma.
and nitrosoureas for the treatment of malignant glioma after surgery. N Engl J Med J Clin Oncol 2010; 28: 1168–1174.
1980; 303: 1323–1329. 53. Batchelor TT, Mulholland P, Neyns B et al. Phase III randomized trial comparing the
32. Roa W, Brasher PM, Bauman G et al. Abbreviated course of radiation therapy in efficacy of cediranib as monotherapy, and in combination with lomustine, versus
older patients with glioblastoma multiforme: a prospective randomized clinical trial. lomustine alone in patients with recurrent glioblastoma. J Clin Oncol 2013; 31:
J Clin Oncol 2004; 22: 1583–1588. 3212–3218.
33. Keime-Guibert F, Chinot O, Taillandier L et al. Radiotherapy for glioblastoma in the 54. van den Bent MJ, Brandes AA, Rampling R et al. Randomized phase II trial of
elderly. N Engl J Med 2007; 356: 1527–1535. erlotinib versus temozolomide or carmustine in recurrent glioblastoma. EORTC
34. Stupp R, Mason WP, van den Bent MJ et al. Radiotherapy plus concomitant Brain Tumor Group study 26034. J Clin Oncol 2009; 27: 1268–1274.
and adjuvant temozolomide for glioblastoma. N Engl J Med 2005; 352: 55. Dresemann G, Weller M, Rosenthal MA et al. Imatinib in combination with
987–996. hydroxyurea versus hydroxyurea alone as oral therapy in patients with progressive
35. Stupp R, Hegi ME, Mason WP et al. Effects of radiotherapy with concomitant and pretreated glioblastoma resistant to standard dose temozolomide. J Neurooncol
adjuvant temozolomide versus radiotherapy alone on survival in glioblastoma in a 2010; 96: 393–402.
randomised phase III study: 5-year analysis of the EORTC-NCIC trial. Lancet Oncol 56. Friedman HS, Prados MD, Wen PY et al. Bevacizumab alone and in combination
2009; 10: 459–466. with irinotecan in recurrent glioblastoma. J Clin Oncol 2009; 27: 4733–4740.
36. Gilbert MR, Wang M, Aldape KD et al. Dose-dense temozolomide for newly 57. Kreisl TN, Kim L, Moore K et al. Phase II trial of single-agent bevacizumab followed
diagnosed glioblastoma: a randomized phase III clinical trial. J Clin Oncol 2013; by bevacizumab plus irinotecan at tumor progression in recurrent glioblastoma.
31: 4085–4091. J Clin Oncol 2009; 27: 740–745.
iii | Stupp et al. Volume 25 | Supplement 3 | September 2014

by guest
on 05 February 2018
58. Stupp R, Wong ET, Kanner AA et al. NovoTTF-100A versus physician’s choice 65. Brandes AA, Tosoni A, Spagnolli F et al. Disease progression or pseudoprogression
chemotherapy in recurrent glioblastoma: a randomised phase III trial of a novel after concomitant radiochemotherapy treatment: Pitfalls in neurooncology. Neuro
treatment modality. Eur J Cancer 2012; 48: 2192–2202. Oncol 2008; 10: 361–367.
59. Gorlia T, Stupp R, Brandes AA et al. New prognostic factors and calculators for 66. Brandes AA, Franceschi E, Tosoni A et al. MGMT promoter methylation status can
outcome prediction in patients with recurrent glioblastoma: a pooled analysis of predict the incidence and outcome of pseudoprogression after concomitant
EORTC Brain Tumour Group phase I and II clinical trials. Eur J Cancer 2012; 48: radiochemotherapy in newly diagnosed glioblastoma patients. J Clin Oncol 2008;
1176–1184. 26: 2192–2197.
60. Clarke JL, Ennis MM, Yung WK et al. Is surgery at progression a prognostic marker 67. Wen PY, Macdonald DR, Reardon DA et al. Updated response assessment criteria
for improved 6-month progression-free survival or overall survival for patients with for high-grade gliomas: response assessment in neuro-oncology working group.
recurrent glioblastoma? Neuro Oncol 2011; 13: 1118–1124. J Clin Oncol 2010; 28: 1963–1972.
61. Brem H, Piantadosi S, Burger PC et al. Placebo-controlled trial of safety and 68. Hottinger AF, Levivier M, Negretti L et al. PET imaging in glioma. The neuro-
efficacy of intraoperative controlled delivery by biodegradable polymers of oncologist’s expectations. PET Clin 2013; 8: 117–128.
chemotherapy for recurrent gliomas. The Polymer-brain Tumor Treatment Group. 69. Phillips HS, Kharbanda S, Chen R et al. Molecular subclasses of high-grade glioma
Lancet 1995; 345: 1008–1012. predict prognosis, delineate a pattern of disease progression, and resemble stages
62. Combs SE, Thilmann C, Edler L et al. Efficacy of fractionated stereotactic in neurogenesis. Cancer Cell 2006; 9: 157–173.
reirradiation in recurrent gliomas: long-term results in 172 patients treated in a 70. Weller M, Pfister SM, Wick W et al. Molecular neuro-oncology in clinical practice: a
single institution. J Clin Oncol 2005; 23: 8863–8869. new horizon. Lancet Oncol 2013; 14: e370–e379.
63. Fogh SE, Andrews DW, Glass J et al. Hypofractionated stereotactic radiation 71. Brandes AA, Franceschi E, Tosoni A et al. Epidermal growth factor receptor
therapy: an effective therapy for recurrent high-grade gliomas. J Clin Oncol 2010; inhibitors in neuro-oncology: hopes and disappointments. Clin Cancer Res 2008;
28: 3048–3053. 14: 957–960.
64. Brandsma D, Stalpers L, Taal W et al. Clinical features, mechanisms, and 72. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
management of pseudoprogression in malignant gliomas. Lancet Oncol 2008; 9: among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
453–461. 139–144.
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu050 | iii

by guest
on 05 February 2018
EANO–ESMO Clinical Practice Guidelines for

diagnosis, treatment and follow-up of patients with
leptomeningeal metastasis from solid tumours†
E. Le Rhun1,2,3, M. Weller4, D. Brandsma5, M. Van den Bent6, E. de Azambuja7, R. Henriksson8,9,

T. Boulanger10, S. Peters11, C. Watts12, W. Wick13,14, P. Wesseling15,16, R. Ruda17 & M. Preusser18,
on behalf of the EANO Executive Board and ESMO Guidelines Committee*
1
Neuro-Oncology, Department of Neurosurgery, Lille University Hospital, Lille; 2Neurology, Medical Oncology Department, Oscar Lambret Center, Lille; 3Lille
University, Inserm U-1192, Villeneuve d’Ascq, France; 4Department of Neurology and Brain Tumour Center, University Hospital, Zurich, Switzerland; 5Department of
Neuro-Oncology, Netherlands Cancer Institute, Amsterdam; 6The Brain Tumour Center at the Erasmus MC Cancer Institute, Rotterdam, The Netherlands; 7Medical
Oncology Department, Institut Jules Bordet and L’Université Libre de Bruxelles (U.L.B), Brussels, Belgium; 8Regional Cancer Center, Stockholm; 9Department of
Radiation Sciences and Oncology, University, Umea, Sweden; 10Neuroradiology, Imaging Department, Oscar Lambret Center, Lille, France; 11Department of
Oncology, University Hospital, Lausanne, Switzerland; 12Division of Neurosurgery, Department of Clinical Neurosciences, University of Cambridge, Cambridge, UK;
13
Neurology Clinic, Heidelberg University Hospital, Heidelberg; 14Clinical Cooperation Unit Neuro-Oncology, German Consortium for Translational Cancer Research
(DKTK), German Cancer Research Centre (DKFZ), Heidelberg, Germany; 15Department of Pathology, VU University Medical Centre and Brain Tumour Center,
Amsterdam; 16Department of Pathology, Princess Maxima Center for Paediatric Oncology and University Medical Centre Utrecht, Utrecht, The Netherlands;
17
Department of Neuro-Oncology, City of Health and Science Hospital, University of Turin, Turin, Italy; 18Clinical Division of Oncology, Department of Medicine 1,
CNS Unit Comprehensive Cancer Centre (CCC-CNS), Medical University, Vienna, Austria
*Correspondence to: ESMO Guidelines Committee, ESMO Head Office, Via L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland. E-mail: clinicalguidelines@esmo.org or
EANO Office, c/o WMA GmbH | Alser Strasse 4, 1090 Vienna, Austria; E-mail: office@eano.eu
†
These Guidelines were developed by the European Society for Medical Oncology (ESMO) and the European Association of Neuro-Oncology (EANO).
The two societies nominated authors to write the guidelines as well as reviewers to comment on them. These guidelines were approved by the EANO Executive Board and
the ESMO Guidelines Committee in May 2017.
Introduction
LM to provide guidance when to treat (as opposed to when to in-
These EANO–ESMO joint recommendations for the diagnosis tensify diagnostic efforts) and on which patients to include in
and treatment of leptomeningeal metastasis (LM) from solid can- clinical trials. Given the low level of evidence, the recommenda-
cers represent the first European guideline initiative on this topic. tions are based more on expert opinion and consensus than on
LM is defined as the spread of tumour cells within the leptomen- evidence from informative clinical trials. Still, the EANO–ESMO
inges and the subarachnoid space, is synonymous with ‘neoplas- multidisciplinary recommendations shall serve as a valuable
tic meningitis’ and can be further denoted by primary tumour as source of information for physicians, other health care providers,
leptomeningeal carcinomatosis, gliomatosis or lymphomatosis. as well as informed patients and relatives.
The recommendations address LM from solid tumours, but nei-
ther LM from primary brain tumours nor LM from lymphoma or
leukaemia. They cover prevention, diagnosis, therapy and follow-
up, but not differential diagnosis, adverse effects of therapeutic Epidemiology
measures or supportive or palliative care. We propose diagnostic LM may be diagnosed in approximately 10% of patients with
criteria as well as a neuroimaging- and cytopathology-based metastatic cancer in the course of disease [1]. The incidence is
classification of LM syndromes to derive pragmatic treatment probably underestimated because of non-specific symptoms and
algorithms. We also assign levels of certainty to the diagnosis of signs, lack of sensitivity to diagnostic procedures and limited
V

by guest
on 05 February 2018
therapeutic options. Breast cancer, lung cancer and melanoma primary tumour type, cerebrospinal fluid (CSF) protein levels,
represent the three most common causes of LM, but LM may be administration of combined modality treatment, systemic treat-
observed with all malignant tumours. LM occurs in the context of ment or intra-CSF treatment, and initial clinical or CSF responses
progressive systemic disease in approximately 70% of solid cancer to treatment [2–5, 9–11, 35–39]. In contrast, an association of
patients with LM, in around 20% at the time of first progression whole brain radiotherapy (WBRT) with overall survival (OS) has
after initial treatment, but in up to 10% already at the time of diag- not been consistently reported [13–16, 28].
nosis. In recent large cohorts of LM patients, brain metastases
were associated with LM in 33%–54% of breast cancer, 56%–82%
of lung cancer and 87%–96% of melanoma patients [2–20].
Risk factors for the development of LM include opening of
Pathogenesis
the ventricular system during brain metastasis surgery or resec- The invasion of the leptomeninges by tumour cells may occur by
tion of cerebellar metastases, especially when using a piecemeal haematogenous spread through the arterial or venous circulation,
resection [21–26]. In a large cohort of patients with brain meta- or endoneural, perineural, perivascular or lymphatic spread, es-
stasis, the incidence of LM was higher in patients treated with pecially from breast and lung cancers. Furthermore, there may be
surgery followed by stereotactic radiosurgery than in patients a direct invasion from brain or spinal parenchymal metastases in
treated with radiosurgery alone [27]. contact with the CSF, the choroid plexus and from subependymal
Patients with lobular subtype and triple-negative tumours metastases [40–42]. Iatrogenic spread may occur after neurosur-
have a relatively higher risk of LM than patients with other types gical interventions, notably when lesioning the ventricles [22]. De
of breast cancer [28]. In recent cohorts, primary breast tumours novo tumours originating in the leptomeninges with melanoma
causing LM were ductal carcinoma in 51%–78%, lobular carcin- histology are also observed, but represent a distinct disease entity
oma in 26%–35%, human epidermal growth factor receptor-2 [43]. Once seeded in the meninges, tumour cells may disseminate
(HER2)-positive in 10%–29% (up to 47% in only one series) and along the meningeal and ependymal surfaces or with the CSF
triple negative in 22%–40.5% [2–7, 9–12, 28]. flow, with a predilection of colonising regions of slow CSF flow
Lung tumours causing LM were adenocarcinomas in 84%– and gravity-dependent locations, e.g. posterior fossa, basilar cis-
96% [13–16, 19]. Epidermal growth factor receptor (EGFR)-mu- terns and lumbar cistern [40]. Molecular factors facilitating seed-
tant lung cancer may have a central nervous system (CNS) trop- ing of the leptomeninges by tumour cells have not been
ism [29–31]. Metastatic CNS involvement has also been identified.
recognised as an emerging complication in patients with anaplas-
tic lymphoma kinase (ALK)-positive non-small-cell lung cancer
(NSCLC) [32, 33]. In a second-line trial for ALK-positive disease,
approximately 35% of ALK-positive patients had brain metasta-
Clinical presentation
sis at the time of study entry [34]. Symptoms and signs are related to the specific CNS areas
Only a few large cohorts of melanoma patients with LM have been involved by LM and thus are typically multifocal. The most fre-
reported and risk factors, including LM risk-associated molecular quent manifestations at presentation [2, 3, 5–7, 11, 13, 15, 18–20,
profiles (e.g. BRAF mutation status), have not been identified. 44] are as follows:
The role of cerebrospinal magnetic resonance imaging (MRI) • headache;
in addition to standard extracerebral staging during the follow- • nausea and vomiting;
up of patients at high risk of LM (e.g. with triple-negative or lob- • mental changes;
ular breast cancer) has not been evaluated. • gait difficulties;
• cranial nerve palsies, e.g. with diplopia or visual disturbance
Recommendation: (cranial nerve VI, III, IV, II) and hearing loss (cranial nerve
• LM should be considered in particular in patients with breast VIII);
or lung cancer or melanoma who present with neurological • radicular signs including weakness, voiding and cauda equina
symptoms or signs [EANO: III, C; ESMO: III, B]. problems; and
• focal or irradiating (radicular) neck and back pain.
Some of these symptoms and signs are in part or largely related
to increased intracranial pressure due to CSF circulation disturb-
Prognosis ances and can be rapidly alleviated by lowering intracranial pres-
Median survival is poor and limited to 6–8 weeks without sure through CSF drainage. Patients may also present with subtle
tumour-specific treatment whereas survival may be prolonged to isolated symptoms and signs. Bladder, sexual and bowel dysfunc-
a few months with LM-directed treatment, including targeted tion are possibly underreported and should be explored at pres-
therapy and immunotherapy: 1.75–4.5 months in breast cancer, entation and during the course of the disease. To ensure
3–6 months in lung cancer and 1.7–2.5 months in melanoma (co- appropriate clinical management strategies, symptoms or signs
horts of more than 30 patients published within the last 10 years) due to parenchymal metastases, extracranial disease, side-effects
[2–7, 9–20, 28]. Published survival rates at 1 year were 16%–24% of treatments or non-cancer comorbidities should be distin-
for breast cancer [5, 12], 19% for lung cancer [13, 18] and 7% for guished from LM-related neurological symptoms and signs. A
melanoma patients [19]. Performance status at diagnosis of LM detailed neurological examination is required and a standard
is the most important prognostic factor, as shown by multivariate evaluation form should be used for the clinical evaluation of pa-
analysis. Other frequently reported prognostic factors include tients at diagnosis and during follow-up [1]. The Neurologic

by guest
on 05 February 2018
Assessment in Neuro-Oncology (NANO) criteria used to assess comprehensive neuroradiological assessment of patients with
neurological function in brain tumour patients [45] are unlikely suspected LM (Table 1) [46]. Characteristic MRI findings include
to be useful in LM patients because they do not address the sulcal enhancement or obliteration, linear ependymal enhance-
multi-level involvement of the CNS typically seen in LM. ment, cranial nerve root enhancement and leptomeningeal
enhancing nodules, notably of the cauda equina. Prospective sys-
Recommendations: tematic studies addressing sensitivity and specificity of MRI in
• Typical clinical signs of LM such as headache, nausea and distinguishing LM from other diseases with a similar MRI pattern
vomiting, mental changes, gait difficulties, cranial nerve pal- have not been conducted. Sensitivity and specificity of cerebro-
sies with diplopia, visual disturbances, hearing loss, sensori- spinal MRI remain difficult to appreciate due to a limited number
motor deficits of extremities and cauda equine syndrome, of publications in patients with a suspicion of LM and improve-
radicular, neck and back pain, notably in a patient with can- ment of technique over time, but have been estimated in the
cer, should alert clinicians to consider LM [EANO: IV, n/a; range of 66%–98% and 77%–97.5%, respectively [47–50]. In re-
ESMO: V, n/a]. cently reported cohorts, 68%–97% of patients with a diagnosis of
• A detailed neurological examination using a standard evalu- LM based on presence of tumour cells in the CSF or on typical
ation form, e.g. as proposed by the Leptomeningeal clinical and MRI findings had radiological evidence of LM on
Assessment in Neuro-Oncology (LANO) group, should be cerebrospinal MRI, indicating that the diagnosis of LM is infre-
carried out at diagnosis [EANO: IV, n/a; ESMO: V, n/a]. quently made in patients with a normal MRI [2–5, 7, 9–11, 14,
16–20, 39, 44].
The neuroradiological assessment of LM is challenging.
Contrast enhancement can be complex in geometry and small in
Diagnostic procedures volume [1]. Several technical issues, such as slice positioning and
thickness, time interval between contrast injection and image
Neuroimaging acquisition influence the sensitivity and specificity of neuro-
imaging. Contrast-enhanced T1-weighted and fluid-attenuated
Cerebrospinal MRI without and with contrast enhancement inversion recovery (FLAIR) sequences are probably the most sen-
using at least 1.5-T field strength is the ‘gold standard’ for the sitive for the detection of LM [51, 52]. Gadolinium should be
Table 1. Evaluation of suspected LM

Recommended protocols of evaluation Results
Clinical Standardised neurological evaluation [1] Presence of typical clinical signs of LMa
evaluation Any other neurological abnormality
Normal neurological evaluation
Neuroimaging Brain: axial T1-weighted, axial FLAIR, axial diffusion, axial T2-weighted Typical MRI findings of linear LM (type A)b
post-gadolinium 3D T1-weighted and post-gadolinium 3D FLAIR Typical MRI findings of nodular leptomeningeal disease (type B)
sequences.
Spinal axis: post-gadolinium sagittal T1-weighted sequences. Spine sag- Both (type C)
ittal T1-weighted sequences without contrast and sagittal fat suppres- No neuroimaging evidence of LM, except possibly
sion T2-weighted sequences, combined with axial T1-weighted hydrocephalus (type D)
images with contrast of regions of interest, may also be considered.
CSF cytology Fresh CSF samples should be processed within 30 min after sampling Positive: presence of tumour cells
when feasible; alternatively, fresh CSF samples can be fixed with Equivocal: suspicious or atypical cells
ethanol/Carbowax (CSF/fixative ratio 1 : 1) Negative: absence of tumour cells
CSF volume is ideally > 10 mL but at least 5 mL
Routine staining for cytological analysis: Pap/Papanicolaou and (in
freshly processed CSF samples) Giemsa
Additional immunocytochemical staining (upon indication and avail-
ability of material) for epithelial and melanocytic markers; in case of
haematological malignancy in the differential diagnosis, consider
immunostainings for lymphoid cells and/or flow cytometry analysis
of the CSF sample
A second CSF sample should be analysed if the initial CSF sample is
negative
a
Typical clinical signs of LM include headache, nausea and vomiting, mental changes, gait difficulties, cranial nerve palsies with diplopia, visual disturb-
ances, hearing loss, sensorimotor deficits of extremities and cauda equine syndrome, and radicular neck and back pain.
b
See Table 2 and text.
CSF, cerebrospinal fluid; FLAIR, fluid-attenuated inversion recovery; LM, leptomeningeal metastasis; MRI, magnetic resonance imaging.
iv86 | Le Rhun et al. Volume 28 | Supplement 4 | August 2017

by guest
on 05 February 2018
injected 10 min before data acquisition at a dose of 0.1 mmol/kg. 2. equivocal, corresponding to the detection of ‘suspicious’ or
The slice thickness should be 1 mm. As recommended by the ‘atypical’ cells in the CSF; or
LANO group, nodules should be defined as 5 10 mm in or- 3. negative, defined as the absence of malignant or equivocal
thogonal diameters [1] and should be distinguished from linear cells in the CSF.
contrast enhancement. As meningeal contrast enhancement may
A distinction of suspicious and atypical as proposed by the
also be observed after lumbar puncture or ventricular shunt place-
LANO group [1] is difficult to use in clinical practice. Staining of
ment, cerebrospinal MRI should be obtained before such proced-
neoplastic cells for specific alterations such as the HER2 protein in
ures whenever feasible. Communicating hydrocephalus is
breast cancer or BRAFV600E protein in melanoma by immuno-
observed in 11%–17% of patients [9, 16]. Cerebrospinal MRI also
cytochemistry may be useful in selected equivocal cases. The fol-
detects brain metastasis, epidural spinal cord compression and
lowing simple measures may improve the sensitivity of CSF
intramedullary spinal cord metastases. Cranial computed tomog-
studies: obtaining sufficient volumes of CSF (ideally > 10 mL, but
raphy (CT) should be limited to patients with contraindications
at least 5 mL), processing CSF within 30 min after sampling and
for MRI and mainly helps to identify nodular disease. 18F-fluoro-
avoiding haemorrhagic contamination [1, 56–58]. A higher sensi-
deoxyglucose positron emission tomography-computed tomog-
tivity was reported with thin-layer preparations (Thinprep) than
raphy (FDG-PET–CT) is rarely useful for the diagnosis of LM
with cytospin-coupled Wright–Giemsa stains [59]. In patients
[53]. CSF flow studies using intra-CSF application of tracers, such
suspected of having LM, CSF analysis should be carried out under
as 111Indium-diethylene triamine pentaacetic acid (DTPA) or
99 optimal conditions. If the first CSF analysis is negative, a second
Technetium macro-aggregated albumin, have been recom-
lumbar puncture should be carried out under optimised condi-
mended for patients considered candidates for intra-CSF pharma-
tions as outlined above, potentially increasing the sensitivity to
cotherapy, as obstruction to CSF flow may impede coverage of the
80%. The yield of further CSF assessments remains doubtful. CSF
target volume of drug distribution [1]. CSF flow abnormalities
fixation in dedicated tubes as established for haematological dis-
have been observed in 61%–70% of patients in small cohorts of
orders [60], such as ethylenediaminetetraacetic acid (EDTA) CSF
non-selected patients with LM [54, 55]. Blocks can be partial or
sample storage tubes or CellSaveTM preservative tubes, may di-
complete and can occur at the base of the brain, in the spinal sub-
minish the need for rapid processing of the sample, but reports on
arachnoidal space and over the cerebral convexities.
the validation of this approach for solid tumours are lacking.
Novel techniques using epithelial cell adhesion molecule (EpCAM)
Recommendations:
antibodies or other tumour-specific antibody-covered magnetic
• The diagnostic work-up should include cerebrospinal MRI. nanoparticles to identify circulating tumour cells have shown promis-
Brain MRI should include axial T1-weighted, axial FLAIR, ing results using various adaptations of the device initially designed
axial diffusion, axial T2-weighted, post-gadolinium 3D for peripheral blood studies, combined with flow cytometry or tu-
T1-weighted and post-gadolinium 3D FLAIR sequences. mour marker immunofluorescent in situ hybridisation (TM-iFISH).
Spinal MRI should include post-gadolinium sagittal Such approaches may improve the sensitivity for tumour cell detec-
T1-weighted sequences. Spine sagittal T1-weighted sequences tion in the CSF but need validation in prospective studies [45, 61–67].
without contrast and sagittal fat suppression T2-weighted se- Genomic alterations can be detected in the CSF by micro-arrays [68],
quences, combined with axial T1-weighted images with con- digital or real-time polymerase chain reaction (qPCR) and targeted
trast of regions of interest, may also be considered [EANO: amplicon sequencing, and whole exome sequencing [69–72]. Yet,
III, C; ESMO: II, B]. there are still insufficient data to substitute a positive CSF cytology by
• CSF flow studies should be considered for patients in whom
the detection in the CSF of tumour-specific mutations at DNA level,
CSF flow obstruction may be present, e.g. hydrocephalus, e.g. BRAFV600E or EGFRT790M. It remains unclear whether tumour
large nodules potentially reducing the CSF circulation on DNA detection in the CSF compartment always reflects the local pres-
MRI, unexpected toxicity of intra-CSF treatment, and who ence of cells or whether this DNA may be derived from tumour cells
are candidates for intra-CSF pharmacotherapy [EANO: IV, circulating in the blood or even from distant extracerebral metastases.
n/a; ESMO: IV, C]. Future studies need to address the question of which quantitative cut-
off of tumour DNA in CSF truly represents clinically relevant LM.
CSF analysis Numerous CSF biomarkers of LM have been explored in a tumour-
Non-diagnostic pathological findings upon routine CSF analysis specific manner, including b-glucuronidase, lactate dehydrogenase
are observed in more than 90% of LM patients and include (LDH), b2-microglobulin, cancer antigen (CA) 15-3, CA 125, CA 19-
increased opening pressure (> 200 mm H2O) in 21%–42% [16, 9, a-foetoprotein (AFP), neuron-specific enolase (NSE), or Cyfra 21-1
44], increased leukocyte counts (> 4/mm3) in 48%–77.5% [2, 10, as direct tumour markers, or molecules thought to be involved in the
16, 44], elevated protein (> 50 mg/dL) in 56%–91% [2, 3, 10, 16, metastatic process, specifically angiogenesis [e.g. vascular endothelial
44] and decreased glucose (<60 mg/dL) in 22%–63% [2, 10, 16, growth factor (VEGF), tissue plasminogen activator (tPA), stromal
44]. However, only the identification of malignant cells in the CSF cell-derived factor (SDF)-1], migration and invasion [e.g. matrix met-
or in a leptomeningeal biopsy establishes the diagnosis of LM (gold alloprotease (MMP)-2 and -9, cathepsins B, C and H], or adhesion
standard). The results of CSF cytology studies are commonly quali- and inflammation [e.g. C-X-C motif chemokine ligand 8 (CXCL-8)/
tative and sensitivity is low, although specificity is high. In recent interleukin-8, CXCL-10/interferon-inducible protein-10 and CCL18].
large cohorts of LM patients, CSF cytology was considered positive The role of these CSF biomarkers in clinical practice is limited, except
in 66%–90% [3]. CSF cytology should be reported as: for AFP or b-human chorionic gonadotropin (hCG) in patients with
1. positive, defined by the presence of malignant cells in the CSF; germ cell tumours, which are not covered by this guideline.

by guest
on 05 February 2018
Recommendation: diagnosis (‘possible’, ‘no evidence for’). Patients meeting criteria
• CSF studies with optimised analysis conditions must be car- for ‘no evidence for’ should not receive LM-directed tumour-spe-
ried out as part of the diagnostic work-up. One repeat lumbar cific treatment, but should receive a follow-up evaluation if the
puncture with optimised analysis conditions should be car- clinical suspicion of LM remains. Furthermore, we propose that
ried out in patients with suspected LM and initial negative or clinical trials should only enrol patients with confirmed or prob-
equivocal CSF studies [EANO: IV, n/a; ESMO: V, n/a]. able LM, and should stratify for this certainty level of diagnosis.
Biopsy
Rarely, leptomeningeal biopsies may be required to confirm the Therapeutic strategies
diagnosis of LM. It may be useful when CSF cytology is repeatedly The aim of treatment of LM is to prolong survival with acceptable
negative, when there is no history of cancer or if there are doubts quality of life, and to prevent or delay neurological deterioration.
about the cause of the clinical and imaging features and if thera- Several tumour-specific approaches can be used in isolation or
peutic interventions are clinically indicated. combination. Recommendations for the treatment modalities for
LM described below are not supported by data from randomised
Diagnostic criteria for LM clinical trials; they are based on uncontrolled case series and ex-
pert opinion and current management strategies vary widely
The diagnosis of LM may be challenging, and several subtypes of across Europe [73].
syndromes collectively referred to as LM can be distinguished
based on clinical findings, neuroimaging features and CSF ana-
lysis. In every case of suspected LM, it should be assessed whether Pharmacotherapy: general considerations
any clinical abnormalities are causally related to LM detected by Based on the assumption that intravenous (i.v.) antitumour
neuroimaging or CSF analysis. In most contemporary clinical tri- agents will distribute in the same way as i.v. administered contrast
als, LM is diagnosed based on the detection of malignant cells in agents, there is a priori no reason to believe that systemic pharma-
the CSF or on suggestive clinical and neuroimaging findings in cotherapy for contrast-enhancing manifestations of LM should
patients with cancer. We propose to classify LM by using two be less efficient than for other systemic manifestations of cancer.
major criteria: Moreover, increased CSF protein levels in most LM patients con-
1. Has the diagnosis been verified cytologically or histologically: firm that the blood–CSF barrier is commonly disrupted in LM
yes (type I) or no (type II)? and that there must be, therefore, increased levels of systemically
2. What are the neuroimaging findings: linear leptomeningeal administered drugs in the CSF of most patients with LM.
disease (type A), nodular leptomeningeal disease (type B), However, floating tumour cells in the CSF in the setting of little
both (type C) or neither nor, e.g. no neuroimaging evidence or no blood–CSF barrier dysfunction or diffuse leptomeningeal
of LM except possibly hydrocephalus (type D)? Type C or ependymal spread not yet accompanied by blood–brain bar-
should be assigned if the less prevalent phenotype (A or B) rier dysfunction may be poorly covered by systemic pharmaco-
still accounts for an estimated disease burden of at least 20%. therapy. Importantly, in the absence of blood–CSF barrier
Based on these considerations, the likelihood of LM can be as- disturbances, drug distribution into the CSF depends mainly on
signed ‘confirmed’, ‘probable’, ‘possible’ or ‘no evidence for’ drug transport across the choroid plexus and not across the
(Table 2). This classification provides guidance when to treat blood–brain barrier [74]. No specific prospective trials have been
with relative confidence (‘confirmed’, ‘probable’) and when to re- reported on systemic treatment of LM, but retrospective series
consider intensified diagnostic efforts at establishing a firm suggest some activity of systemic chemotherapy [75–77]. As there
Table 2. Diagnostic criteria for LM

Cytology/biopsy MRI Confirmed Probablea Possiblea Lack of evidence
Type I: positive IA þ Linear yes n/a n/a n/a

CSF cytology IB þ Nodular yes n/a n/a n/a
or biopsy IC þ Linear þ nodular yes n/a n/a n/a
ID þ Normal yes n/a n/a n/a
Type II: clinical IIA or equivocal Linear n/a With typical clinical signs Without typical clinical signs n/a
findings and IIB or equivocal Nodular n/a With typical clinical signs Without typical clinical signs n/a
neuroimaging IIC or equivocal Linear þ nodular n/a With typical clinical signs Without typical clinical signs n/a
only IID or equivocal Normal n/a n/a With typical clinical signs Without typical
clinical signs
Type A: LM with typical linear MRI abnormalities; Type B: LM with nodular disease only as type B; Type C: LM with both linear and nodular disease; Type
D: LM without MRI abnormalities except possibly hydrocephalus.
a
Requires a history of cancer.
þ, positive; , negative; CSF, cerebrospinal fluid; LM, leptomeningeal metastasis; MRI, magnetic resonance imaging; n/a, not applicable.

by guest
on 05 February 2018
are very limited data on the efficacy of targeted agents in LM, it re- used after progression under crizotinib according to expert opin-
mains uncertain whether these agents improve the prognosis of ion. Accumulating evidence suggests a better CNS penetration of
LM. these compounds; however, only a few reports on the efficacy of
As for patients with brain metastasis [78], the best systemic second- or third-generation TKIs on LM are available [106].
treatment of LM is determined by the primary tumour, its mo- HER2, MET and BRAF mutations as well as RET and ROS re-
lecular characteristics or the molecular characteristics of tumour arrangement have rarely been reported in lung cancer patients
cells of the CSF (when available) and prior treatment of the with LM. Appropriate inhibitors could represent therapeutic op-
underlying malignancy. tions once such a target has been identified. Only preliminary
data are available on the efficacy of anti-programmed cell death 1
(PD-1) agents such as nivolumab [107, 108] or pembrolizumab
LM from breast cancer [109] or anti-PD ligand 1 such as atezolizumab [110] in NSCLC
Common treatments administered in breast cancer patients with patients with brain metastasis [111]. Data on the efficacy of im-
CNS metastases include capecitabine, cyclophosphamide, 5-fluo- munotherapy for brain metastasis or LM are limited because
rouracil (5-FU), methotrexate (MTX), vincristine, cisplatin, eto- such patients are commonly excluded from clinical trials.
poside, vinorelbine and gemcitabine. No trial has specifically
evaluated the role of trastuzumab in treating brain metastasis;
however, a clinical benefit has been reported in patients with LM from melanoma
HER2-positive tumours treated for newly diagnosed brain meta- Systemic chemotherapy using classical agents such as temozolo-
stasis [79, 80]. The efficacy of lapatinib was limited in brain meta- mide, dacarbazine or fotemustine has only limited efficacy in
stasis patients when used alone [81–83], but promising when melanoma patients with brain metastasis or LM. Ipilimumab, an
combined with capecitabine [82–86]. Only very limited data are antibody to anti-cytotoxic T-lymphocyte antigen 4, has shown ef-
available on the efficacy of trastuzumab-emtansine (T-DM1) in ficacy in patients with brain metastasis [112, 113]. Anti-PD-1
treating brain metastasis [87–89]. antibodies have shown activity against extracerebral disease.
There are hardly any systematic studies of LM treatment in pa- However, only preliminary data are available for brain metastasis
tients with breast cancer [90]. A concordance of 94% has been re- [111].
ported between the HER2 status of primary tumours and the BRAF mutations are observed in 40%–60% of melanomas and
tumour cells in the CSF [91]. In HER2-positive tumours with a concordance of BRAF status of 100% has been observed be-
LM, anti-HER2 treatment in combination with chemotherapy tween primary tumour and brain metastasis [114]. Responses
should be considered. More data are also needed on the efficacy have been reported in brain metastasis or LM patients with BRAF
of bevacizumab in combination with various chemotherapies in inhibitors such as vemurafenib [115, 116] or dabrafenib [117].
LM patients [92]. Occasional responses of LM have been reported Although some emerging approaches have shown promising re-
with hormonal agents. However, tumours are often resistant to sults in the control of extracerebral disease, only limited data are
hormonal therapy at the time of LM diagnosis. available on the efficacy of MAPK/ERK kinase (MEK) inhibitors
(trametinib, selumetinib, pimarsertib) or the combination of
LM from lung cancer BRAF and MEK inhibitors (dabrafenib þ trametinib,
vemurafenib þ cobimetinib) in patients with CNS metastases or,
A platinum based-combination (including pemetrexed, gemcita- specifically, LM from melanoma [20].
bine, paclitaxel or vinorelbine for non-squamous and unselected
NSCLC, respectively) can be chosen for treating brain metastasis
from lung cancer [93, 94]. The addition of bevacizumab to clas- Other solid tumours
sical chemotherapy might be beneficial for the treatment of brain
No specific data are available for the treatment of LM caused by
metastasis in NSCLC [95].
other tumours, and treatments should be selected according to
EGFR mutations are observed in 11% of the Caucasian popula-
the primary tumour and its prior treatment.
tion with lung cancer [96] and predict benefit from first- (gefiti-
nib, erlotinib) and second-generation (afatinib) EGFR tyrosine
kinase inhibitors (TKIs) [13, 14, 16, 18, 29, 97–101], as well as the
High-dose systemic chemotherapy
third-generation inhibitor osimertinib in the case of TKI resist-
ance [102]. In several small cohorts, a preservation of the EGFR Cytotoxic CSF concentrations of MTX, cytarabine or thiotriethy-
status was noted in the CSF relative to the primary tumour [18, lenephosphoramide (thioTEPA) may be achieved using high-
70, 72, 98]. Erlotinib may achieve higher CSF concentrations dose systemic administration, and these agents have induced re-
than gefitinib [103]. New EGFR TKIs such as osimertinib have sponses in LM from various solid tumours [76, 118–120]. The
shown promising results in a phase I trial on LM in patients who major limitations of these approaches are haematological toxicity
had progressed on prior EGFR TKI therapy [104]. Combinations and their incompatibility with other systemic regimens poten-
of EGFR TKIs with other agents await exploration. tially needed for the control of systemic disease.
ALK rearrangements and MET mutations are found in 3%–5%
and 4% of NSCLC, respectively. Crizotinib, which targets ALK, Recommendation:
proto-oncogene tyrosine-protein kinase (ROS) and MET, repre- • Systemic pharmacotherapy based on primary tumour and
sents the first targeted option for ALK-rearranged NSCLC previous treatment should be considered for most patients
patients [105]. Ceritinib, alectinib, lorlatinib or brigatinib may be with type B/C LM [EANO: IV, n/a; ESMO: V, n/a].

by guest
on 05 February 2018
Intra-CSF pharmacotherapy Three agents are commonly used for the intrathecal treatment
of LM: MTX, cytarabine, including liposomal cytarabine, or
Although no randomised trial has demonstrated that intra-CSF
thioTEPA. Thus, the compounds routinely used for intra-CSF
chemotherapy prolongs survival in LM patients, intra-CSF
treatment do not have a key role as single agents for systemic
pharmacotherapy is used by the majority of physicians treating
treatment of most common cancers causing LM. Different sched-
LM patients across Europe [73]. Due to its limited penetration
ules have been proposed for these agents, without consensus on
into solid tumour lesions, e.g. up to 1–3 mm [121, 122], intra-
CSF pharmacotherapy is mainly considered for patients with type optimal dose, frequency of administration or optimal duration of
A LM and a significant tumour cell load in the CSF (Table 2). treatment (Table 3). No intra-CSF agent has shown a significant
Furthermore, intra-CSF pharmacotherapy should not be admin- survival advantage over another [130, 131]. Combined intra-CSF
istered to patients with symptomatic hydrocephalus who require agents have not demonstrated superiority over single intra-CSF
ventriculoperitoneal shunt placement or with a ventricular device agents [132], yet, longer time to neurological progression was re-
without on/off option. ported in LM patients with liposomal cytarabine than in those pa-
When intra-CSF pharmacotherapy is used, the administration tients treated with MTX [131]. Toxicities of the various intra-CSF
may be done through repeated lumbar punctures or preferably agents differed. More mucositis or neurological complications such
through a subgaleal reservoir and intraventricular catheter. An as headache were observed with MTX than with thioTEPA [130].
equivalent or larger volume of CSF should be removed before For MTX and liposomal cytarabine, Cancer and Leukemia Group B
each intra-CSF injection. After lumbar injection, patients should (CALGB)-expanded Common Toxicity Criteria treatment-related
remain in a flat position for 1 h [123]. grade 3 toxicity was similar [131]. In a subanalysis, a greater
The conceptual advantages of the ventricular route include the quality-adjusted survival, measured by quality-adjusted time with-
certainty that the drug is not delivered into the epidural or sub- out symptoms or toxicity (Q-TWIST), was observed in patients
dural space instead of the CSF compartment, a more uniform dis- treated with liposomal cytarabine than with MTX [133].
tribution of the agent, greater patient comfort and a faster The efficacy and tolerance of intra-CSF trastuzumab in HER2-
procedure, which improves compliance and safety of drug ad- positive breast cancer is under evaluation in two trials
ministration. The latter is particularly true for patients requiring (NCT01325207, NCT01373710). A small pooled analysis of 17
anticoagulation [124–126]. The safety of ventricular devices has patients with HER2-positive breast cancer from 13 publications
been shown in several cohorts of patients using different technol- treated by intra-CSF trastuzumab reported clinical improvement
ogies and several devices, but careful handling is required to en- in 69% of patients, with a median OS of 13.5 months [134].
sure aseptic puncture and drug application to minimise the risk A critical review of the randomised trials on LM, five of which
of infectious complications [126–128]. The best surgical proced- enrolled patients with solid tumours [75, 125, 130–132]
ure has to be defined by the neurosurgeon in charge of the pa- (Supplementary Table S1, available at Annals of Oncology online),
tient. No benefit of ventricular versus lumbar route has been all assessing the response to intra-CSF therapy, revealed a lack
demonstrated; however, a longer progression-free survival for of standardisation for the evaluation of response and methodo-
MTX was observed in a sub-study of a randomised trial using the logical limitations with respect to the type of tumour (haemato-
ventricular route. In contrast, there was no significant difference logical versus solid), baseline evaluation, response to treatment
using liposomal cytarabine, presumably due to the different half- and evaluation of safety, and all experienced long times for ac-
lives of these agents [124]. New approaches such as ventriculo- crual [46]. Moreover, all were open-label studies. Only cohort
lumbar perfusion are interesting, but require further study [129]. studies have been reported in recent years.
Table 3. Characteristics and schedules of administration of intra-CSF therapy

Agent Description Half-life in Recommended schedules Prophylaxis of adverse events
the CSF of administration
Methotrexate Folate anti-metabolite, cell 4.5–8 h 10–15 mg twice weekly (total, 4 weeks), Folinic acid rescue, 25 mg 6 h
cycle specific then 10–15 mg once weekly (total, 4 weeks) for 24 h starting 6 h after
then 10–15 mg once monthly administration
Cytarabine Pyrimidine nucleoside ana- < 1h 10 mg twice weekly (total, 4 weeks) None
logue, cell cycle specific then 10 mg once weekly (total, 4 weeks)
then 10 mg once a month
Liposomal Pyrimidine nucleoside ana- 14–21 days 50 mg every 2 weeks (total, 8 weeks) Oral steroids, e.g. 6 mg dexa-
cytarabine logue, cell cycle specific then 50 mg once a month methasone equivalent daily
(d1–d4)
ThioTEPA Alkylating ethyleneimine 3–4 h 10 mg twice weekly (total, 4 weeks) None
compound, cell cycle non- then 10 mg once weekly (total, 4 weeks)
specific then 10 mg once a month
CSF, cerebrospinal fluid; thioTEPA, thiotriethylenephosphoramide.

by guest
on 05 February 2018
One single trial tried to explore the value of adding intra- concomitant intra-CSF MTX plus dexamethasone with focal RT
ventricular MTX to systemic therapy and involved-field radio- for patients with LM from various solid tumours [137]. Intra-
therapy (RT), but the trial was prematurely closed [75]. A total of CSF administration of radioisotopes or radiolabelled monoclonal
35 breast cancer patients with LM were evaluated based on clinical antibodies should be explored in the context of clinical trials.
findings only. No differences for clinical response or OS were
observed, but more treatment-related neurotoxicity, scored ac- Recommendations:
cording to a local scale, was noted in the intra-CSF chemotherapy • Focal RT should be considered for circumscribed, notably
arm (47% versus 6%). The complication rate in the intra-CSF symptomatic lesions [EANO: IV, n.a.; ESMO: V, n.a.].
chemotherapy arm, with 18% of reservoir revisions, was also high • WBRT can be considered for extensive nodular or sympto-
compared with other cohorts (< 7.3% of reservoir revision) [126– matic linear LM [EANO: IV, n/a; ESMO: V, n/a].
128]. The role of adding intra-CSF chemotherapy using liposomal
cytarabine to systemic therapy in breast cancer patients with LM is
readdressed in an ongoing phase III trial (NCT01645839).
The optimal duration of intra-CSF treatment has also not been Individualised approach to LM
adequately explored. Most patients nowadays are treated until Diagnosis and management of patients with LM should follow
progression or for 1 year, if tolerated. In the absence of evidence multidisciplinary tumour board recommendations throughout
from appropriate clinical trials, clinical symptoms and MRI and the disease course. The therapeutic strategy should consider gen-
CSF findings, as well as tolerance of treatment, guide individual eral health and the neurological status of the patient, histological
decisions on the duration of treatment. Notably, the role of per- and molecular subtype of the primary cancer, extent and available
sisting positive CSF cytology alone for decision-making regarding therapeutic options for extra-CNS disease, clinical and imaging
the continuation of treatment remains controversial [1]. presentation of LM, and presence of concomitant brain metasta-
sis. The therapeutic recommendations summarised in Figure 1
Recommendations: and Table 4 are largely based on retrospective cohort data or ex-
• Intra-CSF pharmacotherapy should be considered for most pert agreements and must be considered as having a low level of
patients with type IA/C LM [EANO: IV, n/a; ESMO: V, n/a]. evidence. Randomised clinical trials with adapted methodology
• Intra-CSF pharmacotherapy should be administered through and standardised criteria for diagnosis and response are needed
the ventricular rather than lumbar route whenever feasible to better define the role of all therapeutic interventions in LM.
[EANO: IV, n/a; ESMO: V, n/a].
Radiotherapy Monitoring and follow-up

No randomised clinical trial to assess the efficacy and tolerance of No robust data are available and monitoring recommendations
RT in LM has been conducted. Focal RT administered in fractio- today are still based on consensus and expert opinion. The LANO
nated regimens such as involved-field or stereotactic RT or ad- group has made efforts to determine criteria to evaluate the re-
ministered in single fractions (radiosurgery) can be used to treat sponse to treatment in LM patients [1] (Supplementary Table S2,
nodular disease and symptomatic cerebral or spinal sites. available at Annals of Oncology online). Although the proposal by
In exceptional cases, focal RT can be carried out for cauda equina the LANO group remains controversial largely because of its
syndrome or cranial nerve palsies after exclusion of other causes, complexity, we concur with the view that response should be
even in the absence of corresponding MRI findings. The presence of evaluated according to a complete neurological assessment, a
CSF flow interruptions is associated with decreased survival [54, 55, neuroimaging evaluation and standard CSF cytology. CSF par-
135]. Restoration of CSF flow obstruction can be obtained by focal ameters other than the detection of tumour cells are not con-
RT in 30% of patients with spinal blocks and in 50% of patients sidered (Table 5). Symptoms and signs related to concomitant
with intracranial blocks [136] and has been proposed to reduce the brain metastasis, extracerebral progression, toxicity induced by
toxicity from, and enhance the efficacy of, intra-CSF therapy. treatments or concurrent diseases must not be considered for the
Typical target volumes for RT in the presence of cranial neuro- evaluation of LM response. As most neurological deficits in LM
pathies include, by expert consensus, the skull base, the interpe- patients are irreversible, the best anticipated clinical response is
duncular cistern and the two first cervical vertebrae. Typical usually achievement of stable disease. A complete cerebrospinal
target volumes for RT in the presence of a cauda equina syn- MRI is required for the evaluation of response. During follow-up,
drome include the lumbosacral vertebrae. WBRT may be con- MRI should preferably be repeated using the same scanner or at
sidered for extensive nodular or symptomatic linear LM or co- least a scanner of identical field strength (Table 1). Other imaging
existing brain metastasis. Yet, no association of WBRT with sur- modalities such as MR spectroscopy, MR perfusion or PET have
vival was observed in retrospective studies of LM patients [13, 14, no role so far in the assessment of LM during follow-up. The
15, 18, 28]. Cerebrospinal RT is rarely an option for adult patients LANO group has proposed an MRI grid for the assessment of
with LM from solid cancers because of risk of bone marrow tox- radiographic response, which needs to be validated. Due to small
icity, enteritis and mucositis, and the usual co-existence of sys- volume and geometric complexity, a quantitative assessment is
temic disease. Concomitant craniospinal RT and systemic or often not possible, and LM lesions may be difficult to measure re-
intra-CSF treatment should be avoided to prevent severe toxicity, liably. Thus, it has been proposed to distinguish ‘measurable’ LM
notably myelosuppression. Up to 20% of grade 3–4 adverse disease defined by at least one nodular lesion of more than 5
events have been reported in a phase II trial evaluating 10 mm from ‘non-measurable’ disease which encompasses all

by guest
on 05 February 2018
by guest
Prognosis
on 05 February 2018
Life expectancy Life expectancy
< 1 month ≥ 1 month
Palliative
iv92 | Le Rhun et al.

Type I LM Type II LM
approach positive CSF clinical findings and
or biopsy neuroimaging only
CSF cytology
CSF cytology
negative
positive
(LM confirmed
by biopsy)
No active BM Active BM No active BM Active BM
Stable ECD Progressive ECD Stable ECD Progressive ECD Stable ECD Progressive ECD Stable ECD Progressive ECD

Type IA Type IA Type IA Type IA Type IIA Type IIA Type IIA Type IIA
· IT therapy + · IT therapy + · IT therapy + · IT therapy + · IT therapy (+) · IT therapy (+) · IT therapy (+) · IT therapy (+)
· Modification of · Modification of · Modification of
· Modification of · Modification of · Modification of · Modification of · Modification of
systemic therapy (+) systemic therapy + systemic therapy systemic therapy + systemic therapy systemic therapy + systemic therapy or systemic therapy +
· WBRT (+) or WBRT or both + · WBRT (+)
· WBRT (+) or WBRT (+) · WBRT (+) WBRT or both + · WBRT (+)
Type IB Type IB Type IB Type IB Type IIB Type IIB Type IIB Type IIB
· IT therapy + · IT therapy + · IT therapy +
· IT therapy + · IT therapy - · IT therapy - · IT therapy - · IT therapy -
· Modification of · Modification of · Modification of · Modification of · Modification of · Modification of · Modification of · Modification of
systemic therapy + systemic therapy + systemic therapy +
systemic therapy (+) systemic therapy (+) systemic therapy + systemic therapy + systemic therapy +
· Focal RT+ · Focal RT (+) · Focal RT + · Focal RT (+) · Focal RT + · Focal RT + · Focal RT + · Focal RT (+)
Type IC Type IC Type IC Type IC Type IIC Type IIC Type IIC Type IIC
· IT therapy + · IT therapy + · IT therapy + · IT therapy + · IT therapy (+) · IT therapy (+) · IT therapy (+) · IT therapy (+)
· Modification of · Modification of · Modification of · Modification of · Modification of · Modification of · Modification of · Modification of
systemic therapy (+) systemic therapy + systemic therapy + systemic therapy + systemic therapy (+) systemic therapy + systemic therapy + systemic therapy +
· WBRT and/or focal · WBRT and/or · WBRT and/or
· Focal RT +, WBRT (+) · WBRT and/or focal · WBRT and/or focal · WBRT and/or focal · WBRT and/or focal
RT (+) focal RT + focal RT + RT + RT + RT + RT (+)
Type ID
· IT therapy + Type ID Type ID Type ID
· IT therapy + · IT therapy + · IT therapy +
· Modification of
systemic therapy (+) · Modification of · WBRT and/or · Modification of
· RT - systemic therapy + modification of systemic therapy +
· RT - systemic therapy + · WBRT (+)
Figure 1. Therapeutic approach to LM.

This decision tree is based on expert agreement. When selecting therapeutic options, systemic pharmacotherapy should always be considered. A modification of systemic pharmacotherapy is recom-
mended in patients with progressive extracranial disease as well as in patients with progressive BM, unless RT alone is a preferred option. Intra-CSF pharmacotherapy is recommended in case of floating
tumour cell load in the CSF and is optional in case of linear metastatic meningeal disease. Intra-CSF pharmacotherapy is not recommended in patients with symptomatic hydrocephalus who require ven-
triculoperitoneal shunt placement or with a ventricular device without on/off option or in patients with nodular meningeal metastases only. Focal RT is recommended for the treatment of nodular disease
and symptomatic metastatic cerebral or spinal sites. WBRT may be an option for extensive nodular or symptomatic linear LM or co-existing BM. Note that RT options refer to treatment of LM only.
Annals of Oncology

BM, brain metastases; CSF, cerebrospinal fluid; ECD, extracranial disease; IT, intrathecal; LM, leptomeningeal metastases; RT, radiotherapy; WBRT, whole brain radiotherapy; þ, recommended; (þ), optional;
, not recommended.
Table 4. Key recommendations for the diagnosis and treatment of LM from solid tumours
EANO [146] ESMO [147]
Level of Grade of Level of Grade of

evidence recommendation evidence recommendation
Diagnostic procedures
LM should be considered in particular in patients with breast or lung cancer or III C III B
melanoma who present with neurological symptoms or signs
Typical clinical signs of LM such as headache, nausea and vomiting, mental changes, IV n/a V n/a
gait difficulties, cranial nerve palsies with diplopia, visual disturbances, hearing loss,
sensorimotor deficits of extremities and cauda equine syndrome, radicular, neck
and back pain, notably in a patient with cancer, should alert clinicians to consider LM
A detailed neurological examination using a standard evaluation form, e.g. as proposed IV n/a V n/a
by the LANO group, should be carried out at diagnosis
The diagnostic work-up should include cerebrospinal MRI. Brain MRI should include axial III C II B
T1-weighted, axial FLAIR, axial diffusion, axial T2-weighted, post-gadolinium 3D
T1-weighted and post-gadolinium 3D FLAIR sequences. Spinal MRI should include
post-gadolinium sagittal T1-weighted sequences. Spine sagittal T1-weighted sequences
without contrast and sagittal fat suppression T2-weighted sequences, combined with axial
T1-weighted images with contrast of regions of interest, may also be considered
CSF flow studies should be considered for patients in whom CSF flow obstruction may IV n/a IV C
be present, e.g. hydrocephalus, large nodules potentially reducing the CSF circulation
on MRI, unexpected toxicity of intra-CSF treatment, and who are candidates for
intra-CSF pharmacotherapy
CSF studies with optimised analysis conditions must be carried out as part of the diagnostic IV n/a V n/a
work-up. One repeat lumbar puncture with optimised analysis conditions should be
carried out in patients with suspected LM and initial negative or equivocal CSF studies
Therapeutic strategies
Systemic pharmacotherapy based on primary tumour and previous treatment should IV n/a V n/a
be considered for most patients with type B/C LM
Intra-CSF pharmacotherapy should be considered for most patients with type IA/C LM IV n/a V n/a
Intra-CSF pharmacotherapy should be administered through the ventricular rather than IV n/a V n/a
lumbar route whenever feasible
Focal RT should be considered for circumscribed, notably symptomatic lesions IV n/a V n/a
WBRT can be considered for extensive nodular or symptomatic linear LM IV n/a V n/a
Monitoring and follow-up
A detailed neurological examination using a standard evaluation form should be carried IV n/a V n/a
out every 2–3 months or at radiological progression or when new neurological
symptoms or signs are reported
Cerebrospinal MRI should be carried out every 2–3 months or at any instance of IV n/a V n/a
suspected clinical progression
CSF studies should be carried out every 2–3 months in patients undergoing IV n/a V n/a
intra-CSF pharmacotherapy
Level of evidence and grade of recommendation according to Brainin (EANO recommendations) or Dykewicz (ESMO recommendations) [146, 147].
CSF, cerebrospinal fluid; EANO, European Association of Neuro-Oncology; ESMO, European Society for Medical Oncology; FLAIR, fluid-attenuated inver-
sion recovery; LANO, Leptomeningeal Assessment in Neuro-Oncology; LM, leptomeningeal metastasis; MRI, magnetic resonance imaging; n/a, not applic-
able; RT, radiotherapy; WBRT, whole brain radiotherapy.
other MRI abnormalities [1]. Changes in size, but not changes in MRI grid. In clinical trials, concomitant brain or extradural spinal
intensity of contrast enhancement, should be considered. Changes metastases are evaluated separately for response.
in hydrocephalus should be considered as part of the response CSF cell counts could, in principle, be obtained specifically for
evaluation in clinical practice, but were not included in the LANO tumour, as opposed to non-neoplastic cells, but this has

by guest
on 05 February 2018
Table 5. EANO–ESMO response assessment in LMa
Clinical Imaging CSF Response determination Action
Improved or stable Improved Improved or stable Response Continue treatment

Stable Stable Stable Stable Continue treatment
Worse Improved or Improved or stable Suspicion of progression Consider alternative neurological diagnoses or other
stable reasons for clinical deterioration; change treatment
only if there is no other explanation and if there is
significant worsening of clinical signs for more than
2 weeks
Improved or stable Improved or Worse Suspicion of progression, or progres- Continue treatment, change treatment if appearance
stable sion in case of de novo appearance of tumour cells is confirmed on two consecutive
of tumour cells in the CSFb CSF studies from the same CSF site (lumbar or ven-
tricular) at least 4 weeks apart
Worse Improved or Worse Suspicion of progression, or progres- Consider alternative neurological diagnoses, continue
stable sion in case of de novo appearance treatment; change treatment if there is worsening
of tumour cells in the CSFb of clinical signs for more than 2 weeks and if ap-
pearance of tumour cells is confirmed on two con-
secutive CSF studies from the same CSF site
(lumbar or ventricular) at least 4 weeks apart
Improved or stable Worse Improved or stable Progression Change treatment
Improved or stable Worse Worse Progression Change treatment
Worse Worse Improved or stable Progression Change treatment

or worse
a
Differences from the LANO recommendations are illustrated in Supplementary Table S2, available at Annals of Oncology online.
b
De novo detection of tumour cells as an indicator of progressive disease requires that there were at least two adequately performed negative CSF
analyses.
CSF, cerebrospinal fluid; EANO, European Association of Neuro-Oncology; ESMO, European Society for Medical Oncology; LANO, Leptomeningeal
Assessment in Neuro-Oncology; LM, leptomeningeal metastasis.
remained challenging and would require more sophisticated response. Evaluations should be planned every 2 months for the
techniques than commonly available. As discussed above, the first 6 months and every 3 months thereafter in stable patients,
proposal by the LANO group to classify the standard CSF cy- but should be carried out earlier whenever there is suspicion of
tology results into two groups of patients, negative or atypical, progression based on clinical assessment.
versus positive or suspicious, appears too complex. Three catego-
ries of positive, equivocal and negative seem to be more feasible Recommendations:
in clinical practice (see above). A complete CSF cytological re- • A detailed neurological examination using a standard evalu-
sponse requires a conversion of a previously positive to a negative ation form should be carried out every 2–3 months or at
CSF response maintained for at least 4 weeks. If only lumbar CSF radiological progression or when new neurological symptoms
was positive and the patient is treated through a ventricular reser- or signs are reported [EANO: IV, n/a; ESMO: V, n/a].
voir, the CSF response cannot be evaluated unless further lumbar • Cerebrospinal MRI should be carried out every 2–3 months
CSF samples are obtained. An unequivocal de novo appearance of or at any instance of suspected clinical progression [EANO:
malignant cells in the CSF after repeated negative CSF cytologies IV, n/a; ESMO: V, n/a].
carried out under optimised conditions should be considered as • CSF studies should be carried out every 2–3 months in pa-
progression and does not require a confirmatory analysis. In con- tients undergoing intra-CSF pharmacotherapy [EANO: IV,
trast, a change from negative to equivocal is not considered rele- n/a; ESMO: V, n/a].
vant for clinical decision-making. CSF cytology may remain
positive in patients with stable or improved clinical or imaging
features [138]. The levels of CSF protein, glucose or lactate, or
novel biomarkers or new methodologies for the identification of Supportive care
tumour cells in the CSF have not been integrated into routine re- Although this guideline does not aim at comprehensively describ-
sponse determination at present. ing palliative and supportive care, a few points deserve consider-
Clinical, imaging and CSF evaluations should be carried out at ation. The role of steroids has not been specifically studied in LM
baseline and at defined time points thereafter to assess the patients, notwithstanding their role for associated brain metastasis,

by guest
on 05 February 2018
Table 6. Levels of evidence and grades of recommendation for a thera- Table 7. Levels of evidence and grades of recommendation for a diagnostic
peutic intervention (using the European Federation of Neurological measure (using the European Federation of Neurological Societies criteria
Societies criteria as recommended by EANO) [146] as recommended by EANO) [146]
Evidence classification Evidence classification
Class I: An adequately powered prospective, randomised, controlled Class I: A prospective study in a broad spectrum of persons with the sus-
clinical trial with masked outcome assessment in a representative pected condition, using a ‘gold standard’ for case definition, where the
population or an adequately powered systematic review of prospect- test is applied in a blinded evaluation, and enabling the assessment of
ive randomised controlled clinical trials with masked outcome assess- appropriate tests of diagnostic accuracy
ment in representative populations. The following are required: Class II: A prospective study of a narrow spectrum of persons with the
a. Randomisation concealment suspected condition, or a well-designed retrospective study of a broad
b. Primary outcome(s) is/are clearly defined spectrum of persons with an established condition (by ‘gold standard’)
c. Exclusion/inclusion criteria are clearly defined compared with a broad spectrum of controls, where test is applied in
d. Adequate accounting for dropouts and crossovers with numbers suf- a blinded evaluation, and enabling the assessment of appropriate tests
ficiently low to have minimal potential for bias of diagnostic accuracy
e. Relevant baseline characteristics are presented and substantially Class III: Evidence provided by a retrospective study where either per-
equivalent among treatment groups or there is appropriate statis- sons with the established condition or controls are of a narrow spec-
tical adjustment for differences trum, and where test is applied in a blinded evaluation
Class II: Prospective matched-group cohort study in a representative Class IV: Any design where test is not applied in blinded evaluation OR
population with masked outcome assessment that meets a–e above evidence provided by expert opinion alone or in descriptive case ser-
or a randomised, controlled trial in a representative population that ies (without controls)
lacks one criteria a–e Rating of recommendations
Class III: All other controlled trials (including well-defined natural history Level A rating (established as useful/predictive or not useful/predict-
controls or patients serving as own controls) in a representative popu- ive) requires at least one convincing class I study or at least two con-
lation, where outcome assessment is independent of patient sistent, convincing class II studies
treatment Level B rating (established as probably useful/predictive or not useful/
Class IV: Evidence from uncontrolled studies, case series, case reports or predictive) requires at least one convincing class II study or over-
expert opinion whelming class III evidence
Rating of recommendations Level C rating (established as possibly useful/predictive or not useful/
Level A rating (established as effective, ineffective or harmful) requires predictive) requires at least two convincing class III studies
at least one convincing class I study or at least two consistent, con-
Reprinted from [146] by permission of European Journal of Neurology.
vincing class II studies
Copyright V
C 2004, John Wiley and Sons.
Level B rating (probably effective, ineffective or harmful) requires at
EANO, European Association of Neuro-Oncology.
least one convincing class II study or overwhelming class III evidence
Level C (possibly effective, ineffective or harmful) rating requires at
least two convincing class III studies
enriched for molecular genetic signatures where feasible, and
Reprinted from [146] by permission of European Journal of Neurology. with adequate criteria of evaluation are required to improve the
Copyright V
C 2004, John Wiley and Sons. outcome of LM in a primary cancer-specific manner. Important
EANO, European Association of Neuro-Oncology. questions to address include the role of intrathecal pharmaco-
therapy and of novel systemic therapies, notably targeted agents
and immunotherapy.
chemical meningitis or other systemic complications of cancer.
Steroids may also alleviate meningeal irritation and radicular pain.
When required clinically, the lowest dose of steroids should be
used for the shortest time possible. Seizures should be managed Methodology
using drugs that do not interact with systemic treatments. Primary References were identified through searches of PubMed with the
prophylaxis is not recommended [139, 140]. Ventriculoperitoneal search terms ‘leptomeningeal metastasis’, ‘neoplastic meningitis’,
shunting may provide durable relief from symptomatic hydro- ‘intrathecal’, ‘intra-CSF’, ‘CNS’, ‘brain’, ‘metastasis’, ‘trial’, ‘clin-
cephalus [141–144]. National and institutional guidelines may ical’, ‘radiotherapy’ and ‘chemotherapy’ in various combinations
provide further guidance [145]. from 1 January 1976 to 30 November 2016, because of the paucity
of data on this topic. Articles were also identified through
searches of the authors’ own files. Only papers in English were re-
viewed. Data available only in Abstract form were only exception-
Outlook ally included. The final reference list was generated by consensus
Guidelines reflect knowledge and consensus at a given timepoint. of the authors and based on originality and relevance to the broad
Updates on these recommendations will be announced on the scope of this guideline. Levels of evidence and grades of recom-
website of EANO (www.eano.eu) and ESMO (www.esmo.org). mendation were applied using the European Federation of
Randomised trials based on well-defined diagnostic and inclu- Neurological Societies criteria as recommended by EANO
sion criteria, in appropriately selected sub-groups of LM patients, (Tables 6 and 7) [146] as well as using an adapted version of the

by guest
on 05 February 2018
Table 8. Levels of evidence and grades of recommendation as recom- Merck Sharp & Dohme and Roche and honoraria for lectures,
mended by ESMO (adapted from the Infectious Diseases Society of consultation or advisory board participation from Bristol-Myers
America–United States Public Health Service Grading Systema) Squibb, Novartis, Gerson Lehrman Group, CMC Contrast,
GlaxoSmithKline, Mundipharma, Roche and AstraZeneca. DB,
Levels of evidence
RH, TB, SP, CW, WW, PW and RR have declared no potential
conflicts of interest.
heterogeneity
References
II Small randomised trials or large randomised trials with a suspicion
of bias (lower methodological quality) or meta-analyses of such 1. Chamberlain M, Junck L, Brandsma D et al. Leptomeningeal metasta-
trials or of trials demonstrated heterogeneity ses: a RANO proposal for response criteria. Neuro Oncol 2017; 19:
484–492.
2. Rudnicka H, Niwi nska A, Murawska M. Breast cancer leptomeningeal
metastasis—the role of multimodality treatment. J Neurooncol 2007;
V Studies without control group, case reports, expert opinions 84: 57–62.
Grades of recommendation 3. Gauthier H, Guilhaume MN, Bidard FC et al. Survival of breast cancer pa-
A Strong evidence for efficacy with a substantial clinical benefit, tients with meningeal carcinomatosis. Ann Oncol 2010; 21: 2183–2187.
4. Lee S, Ahn HK, Park YH et al. Leptomeningeal metastases from breast
cancer: intrinsic subtypes may affect unique clinical manifestations.
B Strong or moderate evidence for efficacy but with a limited clin-
Breast Cancer Res Treat 2011; 129: 809–817.
ical benefit, generally recommended 5. de Azevedo CR, Cruz MR, Chinen LT et al. Meningeal carcinomatosis
C Insufficient evidence for efficacy or benefit does not outweigh the in breast cancer: prognostic factors and outcome. J Neurooncol 2011;
risk or the disadvantages (adverse events, costs, . . .), optional 104: 565–572.
D Moderate evidence against efficacy or for adverse outcome, gen- 6. Lara-Medina F, Crismatt A, Villarreal-Garza C et al. Clinical features
erally not recommended and prognostic factors in patients with carcinomatous meningitis sec-
E Strong evidence against efficacy or for adverse outcome, never ondary to breast cancer. Breast J 2012; 18: 233–241.
recommended 7. Meattini I, Livi L, Saieva C et al. Prognostic factors and clinical features
in patients with leptominengeal metastases from breast cancer: a single
a
By permission of the Infectious Diseases Society of America [147]. center experience. J Chemother 2012; 24: 279–284.
8. Kim HJ, Im SA, Keam B et al. Clinical outcome of central nervous sys-
tem metastases from breast cancer: differences in survival depending on
systemic treatment. J Neurooncol 2012; 106: 303–313.
9. Niwi nska A, Rudnicka H, Murawska M. Breast cancer leptomeningeal
Infectious Disease Society of America–United States Public metastasis: propensity of breast cancer subtypes for leptomeninges and
Health Service Grading System as recommended by ESMO the analysis of factors influencing survival. Med Oncol 2013; 30: 408.
(Table 8) [147]. Statements without grading were considered jus- 10. Yust-Katz S, Garciarena P, Liu D et al. Breast cancer and leptomeningeal dis-
tified standard clinical practice by the experts. This manuscript ease (LMD): hormone receptor status influences time to development of
has been subjected to an anonymous peer review process. LMD and survival from LMD diagnosis. J Neurooncol 2013; 114: 229–235.
11. Le Rhun E, Taillibert S, Zairi F et al. A retrospective case series of 103
consecutive patients with leptomeningeal metastasis and breast cancer.
J Neurooncol 2013; 113: 83–92.
Funding 12. Niwi nska A, Rudnicka H, Murawska M. Breast cancer leptomeningeal
metastasis: the results of combined treatment and the comparison of
No external funding has been received for the preparation of methotrexate and liposomal cytarabine as intra-cerebrospinal fluid
these guidelines. Production costs have been covered by ESMO chemotherapy. Clin Breast Cancer 2015; 15: 66–72.
from central funds. 13. Morris PG, Reiner AS, Szenberg OR et al. Leptomeningeal metastasis
from non-small cell lung cancer: survival and the impact of whole brain
radiotherapy. J Thorac Oncol 2012; 7: 382–385.
14. Park JH, Kim YJ, Lee JO et al. Clinical outcomes of leptomeningeal me-
Disclosure tastasis in patients with non-small cell lung cancer in the modern
ELR has received research support from Mundipharma and chemotherapy era. Lung Cancer 2012; 76: 387–392.
15. Gwak HS, Joo J, Kim S et al. Analysis of treatment outcomes of intraven-
Amgen. MW has received research grants from Acceleron, tricular chemotherapy in 105 patients for leptomeningeal carcinomatosis
Actelion, Bayer, Isarna, MSD, Merck EMD, Novocure, Piqur and from non-small-cell lung cancer. J Thorac Oncol 2013; 8: 599–605.
Roche and honoraria for lectures or advisory board participation 16. Lee SJ, Lee JI, Nam DH et al. Leptomeningeal carcinomatosis in non-
or consulting from Bristol-Myers Squibb, Celldex, small-cell lung cancer patients: impact on survival and correlated prog-
Immunocellular Therapeutics, Isarna, Magforce, MSD, Merck nostic factors. J Thorac Oncol 2013; 8: 185–191.
EMD, Northwest Biotherapeutics, Novocure, Pfizer, Roche, Teva 17. Riess JW, Nagpal S, Iv M et al. Prolonged survival of patients with non-
small-cell lung cancer with leptomeningeal carcinomatosis in the mod-
and Tocagen. MVdB has received research support from BBB
ern treatment era. Clin Lung Cancer 2014; 15: 202–206.
therapeutics. EdA has received advisory board from Roche; trav- 18. Kuiper JL, Hendriks LE, van der Wekken AJ et al. Treatment and sur-
els grant from Roche and GlaxoSmithKline; research grant from vival of patients with EGFR-mutated non-small cell lung cancer and
Roche. WW has received research support from Pfizer, Apogenix, leptomeningeal metastasis: a retrospective cohort analysis. Lung Cancer
MSD and Roche and has reported being a member of the speak- 2015; 89: 255–261.
er’s bureau of Brystol-Myers Squibb and MSD. MP has received 19. Harstad L, Hess KR, Groves MD. Prognostic factors and outcomes in patients
research support from Böhringer-Ingelheim, GlaxoSmithKline, with leptomeningeal melanomatosis. Neuro Oncol 2008; 10: 1010–1018.

by guest
on 05 February 2018
20. Geukes Foppen MH, Brandsma D, Blank CU et al. Targeted treatment 42. Kokkoris CP. Leptomeningeal carcinomatosis. How does cancer reach
and immunotherapy in leptomeningeal metastases from melanoma. the pia-arachnoid? Cancer 1983; 51: 154–160.
Ann Oncol 2016; 27: 1138–1142. 43. Küsters-Vandevelde HV, Küsters B, van Engen-van Grunsven AC et al.
21. Roelz R, Reinacher P, Jabbarli R et al. Surgical ventricular entry is a key Primary melanocytic tumors of the central nervous system: a review
risk factor for leptomeningeal metastasis of high grade gliomas. Sci Rep with focus on molecular aspects. Brain Pathol 2015; 25: 209–226.
2015; 5: 17758. 44. Kwon J, Chie EK, Kim K et al. Impact of multimodality approach for
22. Ahn JH, Lee SH, Kim S et al. Risk for leptomeningeal seeding after re- patients with leptomeningeal metastases from solid tumors. J Korean
section for brain metastases: implication of tumor location with mode Med Sci 2014; 29: 1094–1101.
of resection. J Neurosurg 2012; 116: 984–993. 45. Nayak L, DeAngelis LM, Brandes AA et al. The Neurologic Assessment
23. Elliott JP, Keles GE, Waite M et al. Ventricular entry during resection of in Neuro-Oncology (NANO) scale: a tool to assess neurologic function
malignant gliomas: effect on intracranial cerebrospinal fluid tumor dis- for integration into the Response Assessment in Neuro-Oncology
semination. J Neurosurg 1994; 80: 834–839. (RANO) criteria. Neuro Oncol 2017; 19: 625–635.
24. Norris LK, Grossman SA, Olivi A. Neoplastic meningitis following sur- 46. Chamberlain M, Soffietti R, Raizer J et al. Leptomeningeal metastasis: a
gical resection of isolated cerebellar metastasis: a potentially preventable Response Assessment in Neuro-Oncology critical review of endpoints
complication. J Neurooncol 1997; 32: 215–223. and response criteria of published randomized clinical trials. Neuro
25. van der Ree TC, Dippel DW, Avezaat CJ et al. Leptomeningeal metasta- Oncol 2014; 16: 1176–1185.
sis after surgical resection of brain metastases. J Neurol Neurosurg 47. Freilich RJ, Krol G, DeAngelis LM. Neuroimaging and cerebrospinal
Psychiatry 1999; 66: 225–227. fluid cytology in the diagnosis of leptomeningeal metastasis. Ann
26. Suki D, Hatiboglu MA, Patel AJ et al. Comparative risk of leptomenin- Neurol 1995; 38: 51–57.
geal dissemination of cancer after surgery or stereotactic radiosurgery 48. Singh SK, Agris JM, Leeds NE, Ginsberg LE. Intracranial leptomenin-
for a single supratentorial solid tumor metastasis. Neurosurgery 2009; geal metastases: comparison of depiction at FLAIR and contrast-
64: 664–674; discussion 674–676. enhanced MR imaging. Radiology 2000; 217: 50–53.
27. Johnson MD, Avkshtol V, Baschnagel AM et al. Surgical resection of 49. Straathof CS, de Bruin HG, Dippel DW, Vecht CJ. The diagnostic ac-
brain metastases and the risk of leptomeningeal recurrence in patients curacy of magnetic resonance imaging and cerebrospinal fluid cytology
treated with stereotactic radiosurgery. Int J Radiat Oncol Biol Phys in leptomeningeal metastasis. J Neurol 1999; 246: 810–814.
2016; 94: 537–543. 50. Zeiser R, Burger JA, Bley TA et al. Clinical follow-up indicates differen-
28. Abouharb S, Ensor J, Loghin ME et al. Leptomeningeal disease and tial accuracy of magnetic resonance imaging and immunocytology of
breast cancer: the importance of tumor subtype. Breast Cancer Res the cerebral spinal fluid for the diagnosis of neoplastic meningitis—a
Treat 2014; 146: 477–486. single centre experience. Br J Haematol 2004; 124: 762–768.
29. Liao BC, Lee JH, Lin CC et al. Epidermal growth factor receptor tyro- 51. Singh SK, Leeds NE, Ginsberg LE. MR imaging of leptomeningeal meta-
sine kinase inhibitors for non-small-cell lung cancer patients with lepto- stases: comparison of three sequences. AJNR Am J Neuroradiol 2002;
meningeal carcinomatosis. J Thorac Oncol 2015; 10: 1754–1761. 23: 817–821.
30. Matsumoto S, Takahashi K, Iwakawa R et al. Frequent EGFR mutations 52. Mahendru G, Chong V. Meninges in cancer imaging. Cancer Imaging
in brain metastases of lung adenocarcinoma. Int J Cancer 2006; 119: 2009; 9: Spec No A:S14-S21.
1491–1494. 53. Lombardi G, Zustovich F, Farina P et al. Neoplastic meningitis from
31. Iuchi T, Shingyoji M, Itakura M et al. Frequency of brain metastases in solid tumors: new diagnostic and therapeutic approaches. Oncologist
non-small-cell lung cancer, and their association with epidermal growth 2011; 16: 1175–1188.
factor receptor mutations. Int J Clin Oncol 2015; 20: 674–679. 54. Grossman SA, Trump DL, Chen DC et al. Cerebrospinal fluid flow
32. Gainor JF, Ou SH, Logan J et al. The central nervous system as a sanctu- abnormalities in patients with neoplastic meningitis. An evaluation using
ary site in ALK-positive non-small-cell lung cancer. J Thorac Oncol 111indium-DTPA ventriculography. Am J Med 1982; 73: 641–647.
2013; 8: 1570–1573. 55. Glantz MJ, Hall WA, Cole BF et al. Diagnosis, management, and sur-
33. Lee DW, Lee KH, Kim JW, Keam B. Molecular targeted therapies for vival of patients with leptomeningeal cancer based on cerebrospinal
the treatment of leptomeningeal carcinomatosis: current evidence and fluid-flow status. Cancer 1995; 75: 2919–2931.
future directions. Int J Mol Sci 2016; 17: pii E1074. 56. Glantz MJ, Cole BF, Glantz LK et al. Cerebrospinal fluid cytology in pa-
34. Shaw AT, Kim DW, Nakagawa K et al. Crizotinib versus chemotherapy tients with cancer: minimizing false-negative results. Cancer 1998; 82:
in advanced ALK-positive lung cancer. N Engl J Med 2013; 368: 733–739.
2385–2394. 57. Rogers LR, Duchesneau PM, Nunez C et al. Comparison of cisternal
35. Clarke JL, Perez HR, Jacks LM et al. Leptomeningeal metastases in the and lumbar CSF examination in leptomeningeal metastasis. Neurology
MRI era. Neurology 2010; 74: 1449–1454. 1992; 42: 1239–1241.
36. Oechsle K, Lange-Brock V, Kruell A et al. Prognostic factors and treat- 58. Dux R, Kindler-Röhrborn A, Annas M et al. A standardized protocol
ment options in patients with leptomeningeal metastases of different for flow cytometric analysis of cells isolated from cerebrospinal fluid. J
primary tumors: a retrospective analysis. J Cancer Res Clin Oncol 2010; Neurol Sci 1994; 121: 74–78.
136: 1729–1735. 59. Pan Z, Yang G, Wang Y et al. Thinprep plus Papanicolaou stain method
37. Herrlinger U, Wiendl H, Renninger M et al. Vascular endothelial is more sensitive than cytospin-coupled Wright Giems stain method in
growth factor (VEGF) in leptomeningeal metastasis: diagnostic and cerebrospinal fluid cytology for diagnosis of leptomeningeal metastasis
prognostic value. Br J Cancer 2004; 91: 219–224. from solid tumors. PLoS One 2015; 10: e0122016.
38. Clatot F, Philippin-Lauridant G, Ouvrier MJ et al. Clinical improve- 60. Quijano S, L opez A, Manuel Sancho J et al. Identification of leptomen-
ment and survival in breast cancer leptomeningeal metastasis correlate ingeal disease in aggressive B-cell non-Hodgkin’s lymphoma: improved
with the cytologic response to intrathecal chemotherapy. J Neurooncol sensitivity of flow cytometry. J Clin Oncol 2009; 27: 1462–1469.
2009; 95: 421–426. 61. Patel AS, Allen JE, Dicker DT et al. Identification and enumeration of
39. Hyun JW, Jeong IH, Joung A et al. Leptomeningeal metastasis: clinical circulating tumor cells in the cerebrospinal fluid of breast cancer pa-
experience of 519 cases. Eur J Cancer 2016; 56: 107–114. tients with central nervous system metastases. Oncotarget 2011; 2:
40. Boyle R, Thomas M, Adams JH. Diffuse involvement of the leptomenin- 752–760.
ges by tumour—a clinical and pathological study of 63 cases. Postgrad 62. Le Rhun E, Massin F, Tu Q et al. Development of a new method for
Med J 1980; 56: 149–158. identification and quantification in cerebrospinal fluid of malignant cells
41. Gonzalez-Vitale JC, Garcia-Bunuel R. Meningeal carcinomatosis. from breast carcinoma leptomeningeal metastasis. BMC Clin Pathol 2012;
Cancer 1976; 37: 2906–2911. 12: 21.

by guest
on 05 February 2018
63. Lee JS, Melisko ME, Magbanua MJ et al. Detection of cerebrospinal 84. Bachelot T, Romieu G, Campone M et al. Lapatinib plus capecitabine in
fluid tumor cells and its clinical relevance in leptomeningeal metastasis patients with previously untreated brain metastases from HER2-
of breast cancer. Breast Cancer Res Treat 2015; 154: 339–349. positive metastatic breast cancer (LANDSCAPE): a single-group phase
64. Milojkovic Kerklaan B, Pluim D, Bol M et al. EpCAM-based flow 2 study. Lancet Oncol 2013; 14: 64–71.
cytometry in cerebrospinal fluid greatly improves diagnostic accuracy 85. Metro G, Foglietta J, Russillo M et al. Clinical outcome of patients with
of leptomeningeal metastases from epithelial tumors. Neuro Oncol brain metastases from HER2-positive breast cancer treated with lapati-
2016; 18: 855–862. nib and capecitabine. Ann Oncol 2011; 22: 625–630.
65. Ma C, Lv Y, Jiang R et al. Novel method for the detection and quantifi- 86. Sutherland S, Ashley S, Miles D et al. Treatment of HER2-positive meta-
cation of malignant cells in the CSF of patients with leptomeningeal me- static breast cancer with lapatinib and capecitabine in the lapatinib ex-
tastasis of lung cancer. Oncol Lett 2016; 11: 619–623. panded access programme, including efficacy in brain metastases–the
66. Subira D, Sim o M, Illan J et al. Diagnostic and prognostic significance UK experience. Br J Cancer 2010; 102: 995–1002.
of flow cytometry immunophenotyping in patients with leptomeningeal 87. Bartsch R, Berghoff AS, Vogl U et al. Activity of T-DM1 in Her2-positive
carcinomatosis. Clin Exp Metastasis 2015; 32: 383–391. breast cancer brain metastases. Clin Exp Metastasis 2015; 32: 729–737.
67. Tu Q, Wu X, Le Rhun E et al. CellSearch technology applied to the de- 88. Keith KC, Lee Y, Ewend MG et al. Activity of trastuzumab-emtansine
tection and quantification of tumor cells in CSF of patients with lung (TDM1) in HER2-positive breast cancer brain metastases: a case series.
cancer leptomeningeal metastasis. Lung Cancer 2015; 90: 352–357. Cancer Treat Commun 2016; 7: 43–46.
68. Magbanua MJ, Roy R, Sosa EV et al. Genome-wide copy number ana- 89. Jacot W, Pons E, Frenel JS et al. Efficacy and safety of trastuzumab
lysis of cerebrospinal fluid tumor cells and their corresponding archival emtansine (T-DM1) in patients with HER2-positive breast cancer with
primary tumors. Genomics Data 2014; 2: 60–62. brain metastases. Breast Cancer Res Treat 2016; 157: 307–318.
69. Li Y, Pan W, Connolly ID et al. Tumor DNA in cerebral spinal fluid re- 90. Dudani S, Mazzarello S, Hilton J et al. Optimal management of lepto-
flects clinical course in a patient with melanoma leptomeningeal brain meningeal carcinomatosis in breast cancer patients—a systematic re-
metastases. J Neurooncol 2016; 128: 93–100. view. Clin Breast Cancer 2016; 16: 456–470.
70. Sasaki S, Yoshioka Y, Ko R et al. Diagnostic significance of cerebro- 91. Park IH, Kwon Y, Ro JY et al. Concordant HER2 status between meta-
spinal fluid EGFR mutation analysis for leptomeningeal metastasis in static breast cancer cells in CSF and primary breast cancer tissue. Breast
non-small-cell lung cancer patients harboring an active EGFR mutation Cancer Res Treat 2010; 123: 125–128.
following gefitinib therapy failure. Respir Investig 2016; 54: 14–19. 92. Wu PF, Lin CH, Kuo CH et al. A pilot study of bevacizumab combined
71. Shingyoji M, Kageyama H, Sakaida T et al. Detection of epithelial with etoposide and cisplatin in breast cancer patients with leptomenin-
growth factor receptor mutations in cerebrospinal fluid from patients geal carcinomatosis. BMC Cancer 2015; 15: 299.
with lung adenocarcinoma suspected of neoplastic meningitis. J Thorac 93. Li M, Zhang Q, Fu P et al. Pemetrexed plus platinum as the first-
Oncol 2011; 6: 1215–1220. line treatment option for advanced non-small cell lung cancer: a meta-
72. Yang H, Cai L, Zhang Y et al. Sensitive detection of EGFR mutations in analysis of randomized controlled trials. PLoS One 2012; 7: e37229.
cerebrospinal fluid from lung adenocarcinoma patients with brain 94. Zimmermann S, Dziadziuszko R, Peters S. Indications and limitations
metastases. J Mol Diagn 2014; 16(5): 558–563. of chemotherapy and targeted agents in non-small cell lung cancer brain
73. Le Rhun E, Ruda R, Devos P et al. Diagnosis and treatment patterns for metastases. Cancer Treat Rev 2014; 40: 716–722.
patients with leptomeningeal metastasis from solid tumors across 95. Besse B, Le Moulec S, Mazières J et al. Bevacizumab in patients with
Europe. J Neuro-Oncol 2017 [Epub ahead of print]. nonsquamous non-small cell lung cancer and asymptomatic, untreated
74. Pardridge WM. CSF, blood-brain barrier, and brain drug delivery. brain metastases (BRAIN): a nonrandomized, phase II study. Clin
Expert Opin Drug Deliv 2016; 13: 963–975. Cancer Res 2015; 21: 1896–1903.
75. Boogerd W, van den Bent MJ, Koehler PJ et al. The relevance of intra- 96. Barlesi F, Mazieres J, Merlio JP et al. Routine molecular profiling of pa-
ventricular chemotherapy for leptomeningeal metastasis in breast can- tients with advanced non-small-cell lung cancer: results of a 1-year na-
cer: a randomised study. Eur J Cancer 2004; 40: 2726–2733. tionwide programme of the French Cooperative Thoracic Intergroup
76. Glantz MJ, Cole BF, Recht L et al. High-dose intravenous methotrexate (IFCT). Lancet 2016; 387: 1415–1426.
for patients with nonleukemic leptomeningeal cancer: is intrathecal 97. Umemura S, Tsubouchi K, Yoshioka H et al. Clinical outcome in pa-
chemotherapy necessary? J Clin Oncol 1998; 16: 1561–1567. tients with leptomeningeal metastasis from non-small cell lung cancer:
77. Bokstein F, Lossos A, Siegal T. Leptomeningeal metastases from solid Okayama Lung Cancer Study Group. Lung Cancer 2012; 77: 134–139.
tumors: a comparison of two prospective series treated with and without 98. Kawamura T, Hata A, Takeshita J et al. High-dose erlotinib for refrac-
intra-cerebrospinal fluid chemotherapy. Cancer 1998; 82: 1756–1763. tory leptomeningeal metastases after failure of standard-dose EGFR-
78. Soffietti R, Abacioglu U, Baumert B et al. Diagnosis and treatment of brain TKIs. Cancer Chemother Pharmacol 2015; 75: 1261–1266.
metastases from solid tumors: guidelines from the European Association 99. Jackman DM, Cioffredi LA, Jacobs L et al. A phase I trial of high dose
of Neuro-Oncology (EANO). Neuro Oncol 2017; 19: 162–174. gefitinib for patients with leptomeningeal metastases from non-small
79. Church DN, Modgil R, Guglani S et al. Extended survival in women cell lung cancer. Oncotarget 2015; 6: 4527–4536.
with brain metastases from HER2 overexpressing breast cancer. Am J 100. Tetsumoto S, Osa A, Kijima T et al. Two cases of leptomeningeal meta-
Clin Oncol 2008; 31: 250–254. stases from lung adenocarcinoma which progressed during gefitinib
80. Brufsky AM, Mayer M, Rugo HS et al. Central nervous system metasta- therapy but responded to erlotinib. Int J Clin Oncol 2012; 17: 155–159.
ses in patients with HER2-positive metastatic breast cancer: incidence, 101. Hoffknecht P, Tufman A, Wehler T et al. Efficacy of the irreversible ErbB fam-
treatment, and survival in patients from registHER. Clin Cancer Res ily blocker afatinib in epidermal growth factor receptor (EGFR) tyrosine kin-
2011; 17: 4834–4843. ase inhibitor (TKI)-pretreated non-small-cell lung cancer patients with brain
81. Lin NU, Carey LA, Liu MC et al. Phase II trial of lapatinib for brain metastases or leptomeningeal disease. J Thorac Oncol 2015; 10: 156–163.
metastases in patients with human epidermal growth factor receptor 2- 102. Mok TS, Wu Y-L, Ahn M-J et al. Osimertinib or platinum-pemetrexed in
positive breast cancer. J Clin Oncol 2008; 26: 1993–1999. EGFR T790M-positive lung cancer. N Engl J Med 2017; 376: 629–640.
82. Lin NU, Diéras V, Paul D et al. Multicenter phase II study of lapatinib 103. Togashi Y, Masago K, Masuda S et al. Cerebrospinal fluid concentration
in patients with brain metastases from HER2-positive breast cancer. of gefitinib and erlotinib in patients with non-small cell lung cancer.
Clin Cancer Res 2009; 15: 1452–1459. Cancer Chemother Pharmacol 2012; 70: 399–405.
83. Toi M, Iwata H, Fujiwara Y et al. Lapatinib monotherapy in patients 104. Yang JCH, Kim DW, Kim SW et al. Osimertinib activity in patients (pts)
with relapsed, advanced, or metastatic breast cancer: efficacy, safety, with leptomeningeal (LM) disease from non-small cell lung cancer
and biomarker results from Japanese patients phase II studies. Br J (NSCLC): Updated results from BLOOM, a phase I study. J Clin. Oncol
Cancer 2009; 101: 1676–1682. 2016; 34, suppl: 9002.

by guest
on 05 February 2018
105. Solomon BJ, Cappuzzo F, Felip E et al. Intracranial efficacy of crizotinib ver- 126. Zairi F, Le Rhun E, Bertrand N et al. Complications related to the use of
sus chemotherapy in patients with advanced ALK-positive non-small-cell an intraventricular access device for the treatment of leptomeningeal
lung cancer: results from PROFILE 1014. J Clin Oncol 2016; 34: 2858–2865. metastases from solid tumor: a single centre experience in 112 patients.
106. Zhang I, Zaorsky NG, Palmer JD et al. Targeting brain metastases in ALK- J Neurooncol 2015; 124: 317–323.
rearranged non-small-cell lung cancer. Lancet Oncol 2015; 16: e510–e521. 127. Kennedy BC, Brown LT, Komotar RJ, McKhann GM 2nd. Stereotactic
107. Brahmer J, Reckamp KL, Baas P et al. Nivolumab versus docetaxel in catheter placement for Ommaya reservoirs. J Clin Neurosci 2016; 27: 44–47.
advanced squamous-cell non-small-cell lung cancer. N Engl J Med 128. Morgenstern PF, Connors S, Reiner AS, Greenfield JP. Image guidance for
2015; 373: 123–135. placement of Ommaya reservoirs: comparison of fluoroscopy and frameless
108. Borghaei H, Paz-Ares L, Horn L et al. Nivolumab versus docetaxel in stereotactic navigation in 145 patients. World Neurosurg 2016; 93: 154–158.
advanced nonsquamous non-small-cell lung cancer. N Engl J Med 129. Gwak HS, Joo J, Shin SH et al. Ventriculolumbar perfusion chemother-
2015; 373: 1627–1639. apy with methotrexate for treating leptomeningeal carcinomatosis: a
109. Herbst RS, Baas P, Kim DW et al. Pembrolizumab versus docetaxel for Phase II Study. Oncologist 2014; 19: 1044–1045.
previously treated, PD-L1-positive, advanced non-small-cell lung can- 130. Grossman SA, Finkelstein DM, Ruckdeschel JC et al. Randomized pro-
cer (KEYNOTE-010): a randomised controlled trial. Lancet 2016; 387: spective comparison of intraventricular methotrexate and thiotepa in
1540–1550. patients with previously untreated neoplastic meningitis. Eastern
110. Rittmeyer A, Barlesi F, Rittmeyer A, Barlesi F, Waterkamp D et al. Cooperative Oncology Group. J Clin Oncol 1993; 11: 561–569.
Atezolizumab versus docetaxel in patients with previously treated non- 131. Glantz MJ, Jaeckle KA, Chamberlain MC et al. A randomized controlled
small-cell lung cancer (OAK): a phase 3, open-label, multicentre rando- trial comparing intrathecal sustained-release cytarabine (DepoCyt) to
mised controlled trial. Lancet 2017; 389: 255–265. intrathecal methotrexate in patients with neoplastic meningitis from
111. Goldberg SB, Gettinger SN, Mahajan A et al. Pembrolizumab for pa- solid tumors. Clin Cancer Res 1999; 5: 3394–3402.
tients with melanoma or non-small-cell lung cancer and untreated 132. Hitchins RN, Bell DR, Woods RL, Levi JA. A prospective randomized
brain metastases: early analysis of a non-randomised, open-label, phase trial of single-agent versus combination chemotherapy in meningeal
2 trial. Lancet Oncol 2016; 17: 976–983. carcinomatosis. J Clin Oncol 1987; 5: 1655–1662.
112. Margolin K, Ernstoff MS, Hamid O et al. Ipilimumab in patients with 133. Cole BF, Glantz MJ, Jaeckle KA et al. Quality-of-life-adjusted survival
melanoma and brain metastases: an open-label, phase 2 trial. Lancet comparison of sustained-release cytosine arabinoside versus intrathecal
Oncol 2012; 13: 459–465. methotrexate for treatment of solid tumor neoplastic meningitis.
113. Konstantinou MP, Dutriaux C, Gaudy-Marqueste C et al. Ipilimumab in Cancer 2003; 97: 3053–3060.
melanoma patients with brain metastasis: a retro-spective multicentre 134. Zagouri F, Sergentanis TN, Bartsch R et al. Intrathecal administration
evaluation of thirty-eight patients. Acta Derm Venereol 2014; 94: 45–49. of trastuzumab for the treatment of meningeal carcinomatosis in
114. Gibney GT, Forsyth PA, Sondak VK. Melanoma in the brain: biology HER2-positive metastatic breast cancer: a systematic review and pooled
and therapeutic options. Melanoma Res 2012; 22: 177–183. analysis. Breast Cancer Res Treat 2013; 139: 13–22.
115. Dummer R, Goldinger SM, Turtschi CP et al. Vemurafenib in patients 135. Chamberlain MC, Kormanik PA. Prognostic significance of
with BRAF(V600) mutation-positive melanoma with symptomatic 111indium-DTPA CSF flow studies in leptomeningeal metastases.
brain metastases: final results of an open-label pilot study. Eur J Cancer Neurology 1996; 46: 1674–1677.
2014; 50: 611–621. 136. Chamberlain MC. Radioisotope CSF flow studies in leptomeningeal
116. Fennira F, Pagès C, Schneider P et al. Vemurafenib in the French tem- metastases. J Neurooncol 1998; 38: 135–140.
porary authorization for use metastatic melanoma cohort: a single- 137. Pan Z, Yang G, He H et al. Concurrent radiotherapy and intrathecal
centre trial. Melanoma Res 2014; 24: 75–82. methotrexate for treating leptomeningeal metastasis from solid tumors
117. Long GV, Trefzer U, Davies MA et al. Dabrafenib in patients with with adverse prognostic factors: a prospective and single-arm study. Int
Val600Glu or Val600Lys BRAF-mutant melanoma metastatic to the J Cancer 2016; 139: 1864–1872.
brain (BREAK-MB): a multicentre, open-label, phase 2 trial. Lancet 138. Chamberlain MC, Johnston SK. Neoplastic meningitis: survival as a
Oncol 2012; 13: 1087–1095. function of cerebrospinal fluid cytology. Cancer 2009; 115: 1941–1946.
118. Siegal T. Leptomeningeal metastases: rationale for systemic chemother- 139. Weller M, Stupp R, Wick W. Epilepsy meets cancer: when, why, and
apy or what is the role of intra-CSF-chemotherapy? J Neurooncol 1998; what to do about it? Lancet Oncol 2012; 13: e375–e382.
38: 151–157. 140. Tremont-Lukats IW, Ratilal BO, Armstrong T, Gilbert MR.
119. Lassman AB, Abrey LE, Shah GD et al. Systemic high-dose intravenous Antiepileptic drugs for preventing seizures in people with brain tumors.
methotrexate for central nervous system metastases. J Neurooncol 2006; Cochrane Database Syst Rev 2008; CD004424.
78: 255–260. 141. Jung TY, Chung WK, Oh IJ. The prognostic significance of surgically
120. Chahal J, Stopeck A, Clarke K et al. Intravenous thiotepa for treatment treated hydrocephalus in leptomeningeal metastases. Clin Neurol
of breast cancer-related leptomeningeal carcinomatosis: case series. Neurosurg 2014; 119: 80–83.
Neurol Sci 2015; 36: 1691–1693. 142. Nigim F, Critchlow JF, Kasper EM. Role of ventriculoperitoneal shunt-
121. Benjamin JC, Moss T, Moseley RP et al. Cerebral distribution of immuno- ing in patients with neoplasms of the central nervous system: an analysis
conjugate after treatment for neoplastic meningitis using an intrathecal of 59 cases. Mol Clin Oncol 2015; 3: 1381–1386.
radiolabeled monoclonal antibody. Neurosurgery 1989; 25: 253–258. 143. Zhang XH, Wang XG, Piao YZ et al. Lumboperitoneal shunt for the treat-
122. Burch PA, Grossman SA, Reinhard CS. Spinal cord penetration of intra- ment of leptomeningeal metastasis. Med Hypotheses 2015; 84: 506–508.
thecally administered cytarabine and methotrexate: a quantitative auto- 144. Yamashiro S, Hitoshi Y, Tajiri S et al. Palliative lumboperitoneal shunt
radiographic study. J Natl Cancer Inst 1988; 80: 1211–1216. for leptomeningeal metastasis-related hydrocephalus: a case series.
123. Blaney SM, Cole DE, Godwin K et al. Intrathecal administration of top- Palliat Med 2017; 31: 93–96.
otecan in nonhuman primates. Cancer Chemother Pharmacol 1995; 36: 145. Pace A, Linda Dirven L, Koekkoek JAF et al., on behalf of the European
121–124. Association of Neuro-Oncology palliative care task force. EANO guidelines
124. Glantz MJ, Van Horn A, Fisher R, Chamberlain MC. Route of intracere- for palliative care in adult glioma patients. Lancet Oncol 2017; 18: e330–e340.
brospinal fluid chemotherapy administration and efficacy of therapy in 146. Brainin M, Barnes M, Baron JC et al. Guidance for the preparation of
neoplastic meningitis. Cancer 2010; 116: 1947–1952. neurological management guidelines by EFNS scientific task forces—
125. Shapiro WR, Schmid M, Glantz M, Miller JJ. A randomized phase III/ revised recommendations 2004. Eur J Neurol 2004; 11: 577–581.
IV study to determine benefit and safety of cytarabine liposome injec- 147. Dykewicz CA. Summary of the guidelines for preventing opportunistic
tion for treatment of neoplastic meningitis. J Clin Oncol 2006; 24 (18 infections among hematopoietic stem cell transplant recipients. Clin
Suppl): 1528. Infect Dis 2001; 33: 139–144.

by guest
on 05 February 2018
Bone sarcomas: ESMO Clinical Practice Guidelines

The ESMO/European Sarcoma Network Working Group*
incidence further immediate investigation. Swelling will only be present

if the tumour has progressed through the cortex and distended
Primary bone tumours are rare, accounting for <0.2% of malignant the periosteum. Regarding differential diagnosis, malignant
tumours registered in the EUROCARE database [1]. Different bone tumours in children may be confused with benign
tumours have distinct patterns of incidence. Osteosarcoma and tumours or (in adults/children) with metastatic disease, both of
Ewing sarcoma (ES) have a relatively high incidence in the second which outnumber primary malignant bone tumours [6–8]. The
decade of life, whereas chondrosarcomas are more common in most likely diagnosis of a suspected bone tumour is related to
older age groups (Figure 1). age. Before 5 years of age, a destructive bone lesion is most com-
Osteosarcoma is the most frequent primary cancer of bone monly metastatic neuroblastoma or eosinophilic granuloma;
(incidence: 0.2–0.3/100 000/year). The incidence is higher in ado- above 5 years, it is often a primary bone sarcoma; after 40 years
lescents (0.8–1.1/100 000/year at age 15–19), where it accounts of age, it tends to be metastasis or myeloma [9].
for >10% of all solid cancers. The male : female ratio is 1.4 : 1. Bone sarcomas are frequently difficult to recognise as malig-
Risk factors for the occurrence of osteosarcoma include previous nant by clinicians, radiologists, and pathologists. Therefore, all
clinical practice
radiation therapy, Paget disease of bone [2], and germline ab- patients with a suspected primary malignant bone tumour
guidelines
normalities, such as Li–Fraumeni syndrome, Werner syndrome, should be referred to a bone sarcoma reference centre or an in-
Rothmund–Thomson syndrome, Bloom syndrome, and heredi- stitution belonging to a specialised bone sarcoma network
tary retinoblastoma [3]. before biopsy [10–13] [III, A].
ES is the third most common primary malignant bone- The medical history should focus on symptoms such as dur-
associated sarcoma. It occurs most frequently in children and ation, intensity, and timing of complaints, for example night
adolescents, but is also seen in adults. Median age at diagnosis is pain or fracture. Moreover, specific events for bone tumours
15 years and there is a male predilection of 1.5 : 1. In white include prior benign/malignant lesions, family history, and
Caucasians under the age of 25, ES has an incidence of 0.3/ previous radiotherapy. A recent injury does not rule out a ma-
100 000 per year [4], but it is very uncommon in the African and lignant tumour and must not prevent appropriate diagnostic
Asian population. About 25% of patients have ES of the pelvic procedures. All patients should have a full physical examin-
bones, whereas 50% have extremity tumours. The ribs and verte- ation. Specific attention should be given to the size, consist-
bral column are also frequently affected. ES may involve any bone ency of the swelling, its location and mobility, the relation of
and (less commonly in children) soft tissues. swelling to the involved bone, and the presence of regional/
Chondrosarcoma is the most frequently occurring bone local lymph nodes. Conventional radiographs in two planes
sarcoma of adulthood. The incidence is about 0.2/100 000/year, should always be the first investigation. When the diagnosis of
with the most common age at diagnosis being between 30 and malignancy cannot be excluded with certainty on radiographs,
60 years and the male : female ratio is ∼1 : 1 [5]. the next imaging step is magnetic resonance imaging (MRI) of
Chordomas are rare, arising with an incidence of ∼0.5 per the whole compartment with adjacent joints, which is the best
million population per year. modality for local staging of extremity and pelvic tumours
[14]. Computed tomography (CT) should be used only in the
diagnosis case of diagnostic problems or doubt, to visualise more clearly
calcification, periosteal bone formation or cortical destruction.
The presence of persistent non-mechanical pain in any bone The biopsy of a suspected primary malignant bone tumour
lasting more than a few weeks should cause concern and lead to should be carried out at the reference centre by the surgeon
who is to carry out the definitive tumour resection or a radiolo-
*Correspondence to: ESMO Guidelines Working Group, ESMO Head Office, Via gist member of the team [10–13]. The principles of the biopsy
L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland. are:
†
• there should be minimal contamination of normal tissues;
Approved by the ESMO Guidelines Working Group: March 2010, last update July 2014.
This publication supersedes the previously published version—Ann Oncol 2012; 23 • in many situations, core-needle biopsies (taken under imaging
(Suppl 7): vii100–vii109. control) are an appropriate alternative to open biopsy;

by guest
on 05 February 2018
8
7
Age specific rates

6
(per million)
5
4
3
2
1
0
0–4
5–9
10–14
15–19
20–24
25–29
30–34
35–39
40–44
45–49
50–54
55–59
60–64
65–69
70–74
75–79
80–84
85+
Age band
(Years)
Osteosarcoma Chondrosarcomas Ewing’s sarcoma
Figure 1. Variations in age-specific incidence rates with morphology, England, 1998–2007. http://www.ncin.org.uk/publications/data_briefings/bone_
sarcomas_incidence_and_survival.aspx.
• adequate sampling of representative areas for histology must pathologist should receive information regarding the clinical/
be assured; radiological context in which the tumour has arisen, relevant
• samples should always be sent for microbiological culture in observations made at the time of surgery and whether the
all cases entailing a potential differential diagnosis; patient has received preoperative chemotherapy. The size (mea-
• samples must be interpreted by an experienced pathologist sured in three dimensions in mm) of the tumour in the resected
with the collaboration of a radiologist; bone should be noted.
• the request form should contain sufficient details for the path- The histological features of the tumour should be described
ologist including: the site of the tumour, the patient’s age, and and the tumour type (and subtype) specified according to the
the radiological imaging. 2013 World Health Organization (WHO) Classification [15].
The pathology report should describe the extent of local
If an open biopsy is done, it should be carried out using a lon- tumour spread, including involvement of specific anatomical
gitudinal incision. To be sure that the biopsy location is ad- compartments. It should be noted whether the resection margins
equate and the tissue is representative, it is recommended to are clear or involved by tumour and the distance (in mm) of
make X-rays of the biopsy location and sometimes undertake a tumour from the nearest resection margin measured. The results
frozen section in case more material is required. In aggressive of relevant ancillary investigations (e.g. immunohistochemistry)
and malignant tumours of bone, the biopsy tract must be con- should be recorded [16]. The tumour should be classified
sidered to be contaminated with tumour and must be removed using Systematic Nomenclature of Medicine (SNOMED) or
together with the resection specimen to avoid local recurrences, International Classification of Diseases for Oncology (ICD-O)
including the possible channels through which drains have been codes.
placed. Biopsy tracts should be clearly marked by means of a
small incision or ink tattoo to ensure that the location can be
recognised at the time of the definitive procedure. In cases of
spinal column involvement, laminectomy or decompression
should be avoided unless necessary to relieve spinal cord com-
stage classification and risk assessment
pression. Samples should be quickly submitted for pathological All cases of suspected bone tumours should be discussed on a
assessment, ideally within half an hour; upon arrival, and before multidisciplinary basis by the radiologist who has interpreted
formalin fixation, tumour imprints (touch preps) can be taken the imaging, the pathologist who has reviewed the biopsy mater-
(useful for tumour-specific translocation by fluorescence in situ ial, the surgeon, the radiation therapist, and the medical oncolo-
hybridisation, FISH), and tissue/cell suspensions should be kept gist. This will minimise the risk of errors in diagnosis, staging,
frozen in cryomolds. A further option is to establish primary risk assessment, and treatment.
cell cultures for cytogenetics and other studies. The collection of Several staging systems for bone tumours are in use [17, 18].
fresh frozen tissue and tumour imprints (touch preps) is However, none of them is perfect or generally accepted.
encouraged, because new molecular pathology assessments Generally, tumour burden and the presence of detectable metas-
could be made at a later stage in the patient’s interest. Informed tases are the two main factors which are taken into consider-
consent for tumour banking should be sought, enabling later ation in the clinical staging of these diseases.
analyses and research, as long as this is allowed by existing regu- General staging should be carried out to assess the extent of
lations. The nature of the bone specimen received for pathology distant disease, including bone scintigraphy, chest radiographs,
reporting should be recorded, i.e. needle biopsy, curettage, and and CT [19]. Whole-body MRI and positron emission tomog-
excision (e.g. segmental resection, limb salvage amputation, or raphy (PET)/CT or PET/MRI are under evaluation both for
other complex resection, such as a hemipelvectomy). It is staging and treatment response evaluation [20]. Additional ap-
usually necessary to decalcify a bone tumour biopsy. The propriate imaging studies and biopsies can be taken from
iii | The ESMO/European Sarcoma Network Working Group Volume 25 | Supplement 3 | September 2014
by guest
on 05 February 2018
suspicious sites, as the exact staging of the disease has an impact formal proof that giving chemotherapy preoperatively improves
on treatment and outcome [III, B]. the outcome per se is lacking. The extent of histological response
No specific laboratory tests for the diagnosis of bone sarcoma to preoperative chemotherapy predicts survival [21, 23, 24].
are available. However, some are useful in the follow-up in ES Low-grade central and parosteal osteosarcoma are malignancies
and osteosarcoma and may also be of prognostic value, such as with a lower metastatic potential, which are treated by surgery
alkaline phosphatase (AP) and lactate dehydrogenase (LDH) alone [III, B]. Although chemotherapy has been used for perios-
[21, 22]. teal osteosarcomas, no benefit for chemotherapy was shown in
A pathological fracture may lead to the dissemination of two retrospective analyses [25, 26]. Current prospective trials
tumour cells into surrounding tissues and increase the risk of evaluate whether altering postoperative chemotherapy in poor
local recurrence. In the case of an existing pathological fracture responders to preoperative systemic therapy improves treatment
in a possible primary malignant bone tumour, adequate imaging outcome.
should be carried out, including MRI followed by biopsy. Surgery should be carried out by a surgical team familiar with
In cases of fracture, internal fixation is contraindicated as it the wide range of surgical reconstructive options. The goal of
disseminates tumour further into both bone and soft tissues surgery is to safely remove the tumour and yet preserve as much
and increases the risk of local recurrence. External splintage is function as possible, striving to obtain adequate surgical margins
recommended, along with appropriate pain control. In patients as narrower margins are associated with an increased risk of
with weakened bone apparent at presentation, there may be a local recurrence [24]. Most patients should be considered candi-
strong case for immobilising the part following biopsy, usually dates for limb salvage. In principle, intralesional or marginal
by application of an external splint. Chemotherapy treatment margins increase the local relapse rate, which is associated with
can result in renal, cardiac, and auditory dysfunction, and reduced overall survival (OS). Thus, good margins are the first
patients undergoing this treatment must have baseline renal goal of surgery [III, B]. Areas where there is suspicion of close
function testing and assessment of cardiac function as well as an margins should be marked on the surgical specimen sent to
audiogram (in the case of treatment with platinum derivatives). pathology.
Sperm storage is recommended for male patients of reproduct- Pathological fracture does not necessarily require amputation.
ive age. For female patients, a fertility physician should be con- In chemosensitive tumours, primary neoadjuvant chemother-
sulted for available options. apy can be used with the expectation that a good response will
allow the fracture haematoma to contract and allow subsequent
resection of the tumour and the involved soft tissues. In patients
treatment with a poor response to chemotherapy or in tumours unlikely to
As malignant primary bone tumours are rare cancers, and as respond to chemotherapy, early surgery obtaining wide margins
management is complex, the accepted standard is treatment at should be considered; in some cases, this may require amputa-
reference centres and/or within reference networks able to tion [27].
provide access to the full spectrum of care [IV, A]. In these Doxorubicin, cisplatin, high-dose methotrexate, ifosfamide,
centres/networks, therapy is usually given within the framework and etoposide have anti-tumour activity in osteosarcoma [28–31]
of prospective, often collaborative, clinical studies, or established [I, A]. Doxorubicin, cisplatin, and high-dose methotrexate are
treatment protocols. In the case of high-grade osteosarcoma, ES, most frequently used as the basis of treatment [31] [II, A]. These
or pleomorphic sarcoma, following biopsy proven-diagnosis, drugs should be administered with adequate supportive care by
primary chemotherapy is indicated, preferably within the frame- experienced paediatric oncologists or medical oncologists in
work of (inter)national trials. reference institutions with appropriate infrastructure and a multi-
disciplinary treatment approach [29]. A variety of pre- and post-
operative combinations are used in common practice and in
osteosarcoma clinical trials. Most current protocols include a period of pre-
Osteosarcoma usually arises in the metaphysis of a long bone, operative chemotherapy, to facilitate local surgical treatment and
most commonly around the knee. Involvement of the axial skel- allow the assessment of tumour response, although this has not
eton and craniofacial bones is primarily observed in adults. been proven to entail a survival benefit over postoperative chemo-
Conventional osteosarcoma, a high-grade malignancy, accounts therapy alone [32, 33] [I, B]. Treatment is commonly given over
for 75% of all high-grade osteosarcomas. Low-grade central and periods of 6–10 months [31]. Whenever possible, patients with
parosteal osteosarcoma are low-grade malignancies, whereas osteosarcoma should receive chemotherapy in the context of
periosteal osteosarcoma is an intermediate-grade chondroblastic prospective trials. Immune modulation has been attempted
osteosarcoma. Adverse prognostic or predictive factors include de- with some agents, e.g. interferon [34] and muramyl tripeptide.
tectable primary metastases, axial or proximal extremity tumour Muramyl tripeptide added to postoperative chemotherapy was
site, large tumour size, elevated serum AP or LDH, and older age associated with a substantial advantage in OS and a non-signifi-
[21] [III, B]. Staging should include local imaging studies, as out- cant trend in event-free survival in one large randomised trial
lined in the following text. Curative treatment of high-grade osteo- [35, 36] [II,C]. Muramyl tripeptide has been approved in Europe
sarcoma consists of chemotherapy and surgery [I, A]. for patients <30 years of age with completely resected localised
Compared with surgery alone, multimodal treatment of high- osteosarcoma. There is no consensus in the sarcoma community
grade localised osteosarcoma increases disease-free survival on the use of this drug, because of weaknesses of the single trial
probabilities from only 10%–20% to >60%. In general, chemo- available [35]. Further studies are definitely needed to identify the
therapy is administered before and after surgery, although a subgroup of patients who could benefit.
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu256 | iii

by guest
on 05 February 2018
When tumour response assessment before surgery is clinically recurrent disease has remained poor, with long-term post-
doubtful and relevant for clinical decision-making, dynamic relapse survival of <20%.
MRI is reliable, but requires sequential scans to evaluate change
in tumour vascularity [37, 38] [III B]. Tumour response is often
apparent only after several cycles of chemotherapy. The assess-
ment of MRI peritumoural oedema is helpful: its disappearance Ewing sarcoma
is a sign of good treatment response [37]. ES is a small, blue, round-cell tumour, periodic acid-Schiff
In general, there is no indication for radiation therapy, but positive, and CD99(MIC2)-positive. All ESs are high-grade
there are anatomical locations in which the possibility of com- tumours. They can arise both from bone and soft tissues, dis-
plete surgical resection is limited. In these cases, radiation playing the same behaviour in principle.
therapy may be an option to try to extend the progression-free The definitive diagnosis is made by biopsy, providing a suffi-
interval. New radiation therapy techniques (e.g. proton beam cient material for conventional histology, immunohistochemis-
and carbon ion therapy) may extend indications. try, molecular pathology, and biobanking. Molecular biology
The multimodal treatment principles detailed above were studies have shown that almost all of these tumours share a
generated in children, adolescents, and young adults with high- common gene rearrangement involving the EWS gene on
grade central osteosarcoma, but also relate to adults at least up chromosome 22 [44, 45]. In most cases, this involves a recipro-
to the age of 60 [39] [III, B]. Older patients (>40 years) may cal translocation t(11;22)(q24;q12) [46], but t(21;22)(q22;q12)
require tailored regimens, especially as far as high-dose metho- [47, 48] and others may also occur [t(7;22), t(17;22), and t(2;22)
trexate is concerned. Doxorubicin and cisplatin are the most translocations and inv(22)]. Although most ES can be recog-
active drugs, with the cumulative dose of anthracycline being a nised with classical haematoxylin and eosin (H&E) and immu-
critical factor. nohistochemistry including CD99, EWS translocation detection
High-grade craniofacial osteosarcoma should be treated the is mandatory when the clinical–pathological presentation is
same way as high-grade osteosarcoma of other locations, al- unusual, or the histological diagnosis is doubtful [II, B]. A refer-
though evidence is lacking due to the absence of selective clinic- ence laboratory for ES diagnosis should have both FISH and
al studies in this patient population [V, B]. Proton beam/carbon reverse transcription polymerase chain reaction (RT–PCR)
ion radiation therapy may be considered within clinical studies available [48]. The laboratory is strongly recommended to be
when complete surgery is unfeasible. Primary metastatic osteo- enrolled in an external quality assurance programme. RT–PCR
sarcoma patients are treated with a curative intent along the is the investigation of choice when frozen tissue is available, and
principles of non-metastatic osteosarcomas [40]. In fact, there FISH is a good choice only when formalin-fixed paraffin-em-
are subsets of patients who can have a very similar or even iden- bedded tissue or touch preps (imprints) are available. There are
tical prognosis to that of localised disease, provided surgical several commercial sources for EWS break-apart probes. Assays
removal of all known metastatic deposits is achievable [41] using EWS break-apart probes do not detect EWS–FLI1 fusions,
[III, B]. Approximately 25% of all patients with primary meta- but only EWS rearrangements, which should not be a problem
static osteosarcoma and >40% of those who achieve a complete when interpreted in the appropriate clinical and pathological
surgical remission become long-term survivors. context. However, differential diagnosis versus other sarcomas
The management of recurrent osteosarcoma needs to take carrying EWS rearrangements may be challenging.
into account the timing of recurrence/metastases, number of Bone marrow biopsies and aspirates from sites distant to the
metastases, and site of metastases. The treatment of recurrent primary or known metastatic lesions may be considered, in the
osteosarcoma is primarily surgical in the case of isolated lung face of a very low incidence of bone marrow metastases in loca-
metastases. Complete removal of all metastases must be lised disease, especially if PET scan was carried out. The added
attempted [III, B], as the disease is otherwise almost universally prognostic value of molecular positivity over light microscopic
fatal, while more than a third of patients with a second surgical evaluation has not yet been proven [IV, C].
remission survive for >5 years [42]. Even patients with multiple Between 20% and 25% of patients are diagnosed with meta-
recurrences may be cured as long as recurrences are resectable, static disease (10%: lung—10%: bones/bone marrow—5%: com-
and repeated thoracotomies are often warranted [42] [III, B]. binations, or others) [49, 50]. Staging must be oriented to detect
CT scan can both over- and underestimate the number of lung, bone, and bone marrow metastases. Multiple bone metas-
metastases. tases confer a poorer outcome than lung/pleural metastases
The role of second-line chemotherapy for recurrent osteosar- (<20% compared with 20%–40% 5-year survival). Other known
coma is much less well defined than that of surgery and there is prognostic factors are tumour size or volume, serum LDH
no accepted standard regimen. Treatment choice may take into levels, axial localisation, or older age (>15 years). A poor histo-
account the prior disease-free interval, and often includes logical response to preoperative chemotherapy and incomplete
ifosfamide ± etoposide ± carboplatin, and other active drugs or no surgery for local therapy are further adverse prognostic
(e.g. gemcitabine and docetaxel; sorafenib). In the two largest factors [II, B] [22, 51–55]. Molecular structure of fusion tran-
reported series, the use of second-line chemotherapy correlated scripts has not been shown to be of prognostic value with
with limited prolongation of survival in patients with inoperable current treatment protocols. Genomic analysis with the assess-
metastatic recurrences, while a positive correlation in operable ment of copy number variation has been shown to be of prog-
disease was observed in only one of the two [41, 42]. Radiation nostic value [56, 57]. With surgery or radiotherapy alone, 5-year
therapy (including Samarium) may have a role in palliation survival was <10%. With treatment in current multimodality
[43]. In general, despite second-line treatment, the prognosis of trials including chemotherapy, survival is ∼60%–70% in localised
iii | The ESMO/European Sarcoma Network Working Group Volume 25 | Supplement 3 | September 2014
by guest
on 05 February 2018
and ∼20%–40% in metastatic disease, depending on metastatic cumulative doses. Chemotherapy regimens in relapse situations
sites and burden. are not standardised and are commonly based on alkylating
All current trials employ 3–6 cycles of initial combination agents (cyclophosphamide and high-dose ifosfamide) [68] in
chemotherapy after biopsy, followed by local therapy, and combination with topoisomerase inhibitors (etoposide and
another 6–10 cycles of chemotherapy usually applied at 2- to 3- topotecan), or irinotecan with temozolomide [III,B] [69, 70] or
week intervals. Treatment duration is thus 10–12 months. gemcitabine + docetaxel.
Agents considered most active include doxorubicin, cyclophos-
phamide, ifosfamide, vincristine, dactinomycin, and etoposide high-grade spindle/pleomorphic sarcomas of bone
[58–62]. Almost all active protocols are based on six-drug com-
Pleomorphic sarcomas of bone comprise a diagnostically hetero-
binations of these substances [I, A]. Chemotherapy intensity is
geneous group of malignant tumours including undifferentiated
positively associated with outcome. High-dose chemotherapy
pleomorphic sarcoma [16, 71, 72]. They arise in a similar age
with haematopoietic stem cell transplantation is still investiga-
group to chondrosarcoma, but the skeletal distribution is more
tional in high-risk localised ES [63].
like osteosarcoma. They typically present with pain and have a
Complete surgical excision, where feasible, is regarded as the
high incidence of fractures at presentation. They represent
best modality of local control, given the higher risk of local recur-
between 2% and 5% of primary bone malignancies. The true inci-
rence when radiotherapy is used as the sole treatment of the
dence is hard to establish, as the two entities [malignant fibrous
primary tumour. Radiotherapy alone (in the range of 45–60 Gy,
histiocytoma (MFH)/fibrosarcoma] exhibit a substantial degree of
depending on location) should be applied if complete surgical ex-
morphological overlap, also reflected by an inconsistent use of
cision is impossible. Postoperative radiotherapy should be given
terminology. Males are more frequently affected than females. An
in cases of inadequate surgical margins and discussed when histo-
association with pre-existing disease (Paget’s disease or bone
logical response in the surgical specimen was poor (i.e. >10%
infarct) or history of previous irradiation has been reported. It is
viable tumour cells) [53] [IV, B]. The dose of postoperative radi-
not unusual for a spindle cell sarcoma to be found to be either a
ation therapy is also 45–60 Gy, depending on margins, response,
dedifferentiated chondrosarcoma or osteosarcoma after examin-
and location. Intralesional surgery must be avoided, as there is no
ing further different sections of the resection.
benefit when compared with radiation therapy alone [53]. For
Pleomorphic sarcomas typically present in older patients with
extraskeletal ES, postoperative radiation therapy is generally used,
a lytic lesion in bone. In many, the differential diagnosis will be
with the exception of good prognosis, superficial ES. Change in
against metastases. Full staging and biopsy are required to reach
the size of the soft tissue mass is easily evaluated on MRI and is a
a diagnosis. Pathological fractures are common and should not
good predictor of tumour response [37, 38]. Dynamic MRI is not
undergo fixation before biopsy.
as reliable as in osteosarcoma [38], as remaining small tumour
Treatment strategies mimic those of osteosarcoma, with chemo-
foci may not be detected. Sequential FDG-PET evaluation might
therapy and complete en bloc resection including any soft tissue
be of additional value [64].
component. Their sensitivity to chemotherapy is poorly known
The treatment of adult patients follows the same principles as
and studies on specific histologies, as currently defined (espe-
for ES in typical age groups. However, tolerability of therapies in
cially after reappraisal of histologies previously known as MFH),
older patients needs to be taken into account when transferring
are highly required.
treatment protocols conceived for children and patients of age
Radiation therapy may be considered in inoperable lesions.
≤40–50 years. The treatment of patients with extraskeletal ES
A global effort to collect these cases would be helpful to estab-
follows the same principles as for bone ES, thus incorporating
lish diagnostic and prognostic criteria as well as recommended
chemotherapy in all cases as well as postoperative radiation
treatments.
therapy in most cases, with the possible exception of superficial
lesions.
Patients with metastases at diagnosis are treated with the same chondrosarcoma
treatment approach as patients with localised disease, although Most chondrosarcomas arise as primary malignant tumours.
the disease definitely has a worse prognosis. Several non-rando- The majority are low grade, locally aggressive, non-metastasising
mised trials have assessed the value of more intensive, time-com- tumours (grade I) rather than high grade (grades II–III) [73].
pressed, or high-dose chemotherapy approaches, followed by Grade 1 chondrosarcomas can be labelled atypical chondrogenic
autologous stem cell rescue, with promising results, but evidence tumours, since they usually do not metastatise. One should be
of benefit, resulting from trials, is pending [65] [III, C]. aware that grade I chondrosarcomas may be treated with radi-
In patients with lung metastases, whole-lung irradiation may ation therapy when located at critical sites such as the skull base.
confer a survival advantage [III, B] [54]. The role of surgical Most chondrosarcomas arise centrally in the metaphyseal region
resection of residual metastases is less well defined. of long bones, but they can also develop in flat bones such as
Patients with multiple bone or bone marrow metastases and pelvis, rib, and scapula. High-grade chondrosarcoma frequently
those with recurrent disease still fare poorly, with 5-year survival arises in the axial skeleton and long bones. Chondrosarcoma
rates of ∼20%. Despite this, local control of bone metastases can arise in pre-existing benign lesions such as enchondroma
with either surgery or radiation therapy is recommended [66]. and osteochondroma. In these circumstances, they are referred
The only prognostic factor identified in relapsed patients seems to as secondary chondrosarcoma and secondary peripheral
to be time to relapse: patients relapsing later than 2 years from chondrosarcomas, respectively. The majority of chondrosarco-
initial diagnosis have a better outcome [67] [III, B]. Doxorubicin mas are of the conventional subtype, but rarer subtypes include
therapy is usually no longer feasible due to previously achieved mesenchymal and clear-cell chondrosarcoma [74, 75]. In rare
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu256 | iii

by guest
on 05 February 2018
circumstances, conventional chondrosarcomas can ‘dedifferenti- There are recent reports of activity of gemcitabine in combin-
ate’ into a very high-grade tumour with a dismal prognosis: the ation with taxotere [82].
so-called de-differentiated chondrosarcoma [74, 75]. Most
chondrosarcomas are solitary, but they can occur as multiple
lesions in patients with multiple osteochondromas and enchon- giant cell tumour of bone
dromatosis. Giant cell tumour (GCT) of bone is a relatively rare, benign
Most chondrosarcomas present with a painless mass. Pain at tumour of the skeleton. Although classified as benign, GCT can
the site of a cartilaginous lesion may be an indicator of malig- be aggressive and recurs locally in up to 50% of cases. Up to 5%
nancy. In the case of chondrosarcoma, a contrast-enhanced of GCTs metastasise to the lungs and spontaneous transform-
MRI can reveal high-grade areas. This provides a useful guide to ation to a high-grade malignancy occurs in 1%–3% of patients.
the site of biopsy [76]. The differentiation between benign Treatment options include intralesional curettage with or
enchondroma or osteochondroma and malignant grade I chon- without adjuvant or en bloc excision. Recent work has suggested
drosarcoma can be difficult. In the phalanges of the hands and that denosumab, a human monoclonal antibody to RANKL that
feet, malignancy is extremely rare, but in the other long bones is overexpressed in GCT, obtains substantial tumour responses
central cartilaginous lesions should be considered low-grade in large or unresectable or metastatic GCT. Thus, the agent can
chondrosarcoma till proven otherwise [74]. be used to achieve cytoreduction allowing potentially curative
Inoperable, locally advanced, and metastatic high-grade chon- surgery. It can be used in unresectable disease and rare metastat-
drosarcomas have a poor prognosis because of resistance to con- ic disease, where treatment interruption is usually followed by
ventional treatments, such as radiotherapy and chemotherapy progression, so that treatment needs to be maintained [83].
[74, 75]. Prognosis depends on histological grade. However, histo-
logical classification is subject to variability in interpretation, with
grade II and III chondrosarcomas often grouped together even chordoma
though there is a wide spectrum of outcome [73]. Also, grade I Chordomas are very rare tumours, arising from the remnants of
tumours (atypical chondrogenic tumours) are not necessarily the notochord into the sacrum (50%), skull base (30%), mobile
curable in all cases, mainly due to problematic local recurrence or spine (20%); extraskeleton cases have also been reported but are
progression to high grade. In particular, dedifferentiated chondro- extremely rare.
sarcomas are aggressive and frequently metastasise [74]. Median age is 60 years, but skull base presentations can also
Assessing the grade of chondrosarcomas is difficult and varia- affect a younger population, including children and adolescents.
tions in opinions even among experts are common [73]. Low- Chordoma is a low-grade, but locally invasive malignancy.
grade cartilage tumours are unlikely to metastasise, but may Dedifferentiated cases are observed in 5% of patients. The meta-
recur locally. Grade I central chondrosarcomas in the long static potential of chordoma is ∼30%. Metastases usually appear
bones of the limbs can be managed by curettage with or without late in the natural history of disease, mostly after local recurrence.
adjuvant (e.g. phenol, cement, and cryotherapy) with a high Chordoma prognosis is more related to local aggressiveness than
chance of success. Low-grade peripheral chondrosarcomas to metastases. Chordoma is a tumour showing notochordal dif-
(arising from osteochondromas) should be surgically excised, ferentiation. Brachyury is a transcription factor involved in noto-
aiming to excise the tumour with a covering of normal tissue chord differentiation and is the diagnostic hallmark for
over it. Higher grade chondrosarcomas and all chondrosarco- conventional chordoma [84]. Dedifferentiated chordomas may
mas of the pelvis or axial skeleton should be surgically excised lose brachyury expression. Immunohistochemistry positivity for
with wide margins. brachyury is strongly recommended to confirm diagnosis.
Recent evidence suggests that mesenchymal chondrosarcoma Due to the rarity and long natural history of the disease, the
may be chemotherapy sensitive, and may be considered for ad- quality of evidence available for more common tumour types is
juvant or neoadjuvant therapy [77, 78] [V, B]. Most authorities currently beyond reach for chordoma. In fact, only a few phase
suggest an Ewing-type chemotherapy regime. There remains II trials are available and most published data are from case
uncertainty about chemotherapy sensitivity of dedifferentiated series and/or retrospective.
chondrosarcoma, which is often treated as a high-grade bone Chordomas need a multidisciplinary approach in referral
sarcoma, with therapies which need to be adapted to patient’s centres and/or referral networks, with a multidisciplinary team
age [79, 80] [V, C]. There is a very high risk of local recurrence including expert pathologists and radiologists, surgeons familiar
following excision of dedifferentiated chondrosarcoma, particu- with musculo-skeletal tumours and site of surgery, expert radi-
larly in the presence of a pathological fracture. If wide margins ation oncologists with access to hadron facilities, expert medical
cannot be reliably achieved with limb salvage, then amputation oncologists, and a palliative care team. All diagnostic and thera-
should be considered. peutic procedures should be discussed in the multidisciplinary
The role of radiotherapy in chondrosarcoma is limited, but expert team.
may be appropriate in highly selected cases or for palliation. MRI is the best modality for local staging. CT scan should be
Excellent outcomes have been reported for skull base chondro- used in the case of diagnostic doubt. Chordoma should be dif-
sarcomas with high-dose radiation therapy, including proton ferentiated from benign notochordal lesions and, if radiological
beam or carbon ion radiotherapy, achieving 80%–90% local appearance is typical for these, biopsy is not recommended
control rates [81]. unless the lesion changes with time [85].
With regard to chemotherapy, drugs active in sarcomas may Preoperative core-needle biopsy is recommended. The biopsy
prove active in chondrosarcoma, especially high-grade lesions. track needs to be included in the surgical resection. In the case
iii | The ESMO/European Sarcoma Network Working Group Volume 25 | Supplement 3 | September 2014
by guest
on 05 February 2018
of skull base chordoma, preoperative biopsy can be avoided in follow-up
selected cases.
Tumour location is the most important variable to define the Follow-up is designed to detect either local recurrence or meta-
primary tumour treatment. The quality of surgical margins is static disease at a time when early treatment is still possible and
the most important prognostic factor. En bloc R0 resection is might be effective. Follow-up of high-grade tumours should
standard treatment, when it is feasible and sequelae are accept- include both a physical examination of the tumour site and as-
able/accepted by the patient, with an expected 5-year recur- sessment of the function and possible complications of any re-
rence-free survival = 50%. If en bloc R0 resection is not feasible, construction. Local imaging and chest X-ray/CT should be the
definitive radiation therapy alone should always be considered norm. Though strict rules cannot be provided in the absence of
as a valid alternative. Local relapse has extremely poor survival any formal validation, a recommended follow-up policy may
rates and local control is rarely achievable. Supportive care foresee intervals between checks after the completion of chemo-
should be incorporated into the treatment from the beginning. therapy every 2–3 months for the first 2 years; every 2–4 months
For skull base and upper cervical tract chordoma, R1–R2 for years 3–4; every 6 months for years 5–10, and thereafter
surgery plus high-dose radiation therapy is the treatment of every 6–12 months according to local practice and other factors.
choice [86–88]. In the case of low-grade bone sarcoma, the frequency of
For sacral chordoma, surgery should definitely be offered as follow-up visits may be lower (e.g. 6 months for 2 years and
the first choice in case chordoma arises from S4 and below. then annually). Late metastases as well as local recurrences
Surgery should always be discussed in the context of other alter- and functional deficits may occur >10 years after diagnosis
natives for tumours originating above S3, since surgery is always and there is no universally accepted stopping point for tumour
followed by important neurological sequelae. Surgery is the surveillance.
primary standard choice for tumours originating from S3, espe- In ES, where osseous metastases are likely, isotope bone scan-
cially if the preservation of S2 roots is possible, as it may result ning can be used in addition. More modern techniques (e.g.
in some neurological recovery (40% of cases) [89–91]. PET or whole-body MRI) require further evaluation. It is im-
Hadrons, i.e. high-dose protons or carbon ions, are superior portant to evaluate the long-term toxicity effect of chemother-
to photons physically and in terms of irradiation of non-target apy and radiotherapy if appropriate. Monitoring for late effect
lesions, although no randomised trials are available to assess should be continued for >10 years after treatment, depending
the benefit of hadrons compared with photons in chordoma. on the chemotherapy protocol and radiation used and in con-
Since hadrons allow lower doses to be given to normal tissues, junction with late effects services when available.
they should be considered the treatment of choice. Advanced Secondary cancers may arise in survivors of bone sarcomas,
technology photons could be used in the case of unavailability either related to, or independent of, irradiation. Secondary leu-
or non-accessibility of protons and ions, and every time they kaemia, particularly acute myeloid leukaemia, may rarely be
show similar dose distribution to the target and critical struc- observed following chemotherapy, as early as 2–5 years after
tures. Due to the relative radiation resistance of chordomas, a treatment.
high-dose up to at least 74 GyE in conventional fractionation
(1.8–2 GyE) for photon- and proton-therapy is required
[81, 92, 93].
note
Indications for definitive radiation therapy are: unresectable Levels of evidence and grades of recommendation have been
disease; inoperable patients; or neurological impairment not applied using the system shown in Table 1. Statements without
accepted by the patient. Radiation therapy should be considered grading were considered justified standard clinical practice by
in the case of R2 or R1 resections. The use of adjuvant/neoadju- the panel members.
vant radiation therapy needs to be discussed with the single
patient and prospective studies encouraged.
Patients who have local recurrence are unlikely to be cured by consensus panel ESMO Guidelines 2014
any local salvage treatment. In the case of local relapse, the These Clinical Practice Guidelines have been developed follow-
choice of treatment can include surgery and/or radiation ing a consensus process based on a consensus event organised
therapy and/or systemic treatment, balancing morbidity and by ESMO in Milan, Italy, in December 2013 and refined by July
quality of life. 2014. This involved experts from the community of the
For oligometastatic disease, surgery/radiofrequency ablations/ European sarcoma research groups and ESMO faculty. Their
stereotactic radiation of metastases can be considered in selected names are indicated hereafter. The text reflects an overall con-
cases. Chemotherapy is inactive. An exception can be high- sensus among them, although each of them may not necessarily
grade dedifferentiated chordoma (anecdotal responses to find it consistent with his/her own views. The panel worked on
chemotherapy have been reported). the text of ESMO Guidelines of previous years, whose author-
There is uncontrolled evidence that imatinib is beneficial in ship should also be credited.
advanced chordoma in terms of progression-free survival and
mainly non-dimensional tumour responses [94]. Its role within • Paolo G. Casali, Italy (Moderator)
the treatment strategy deserves further evaluation. • Jean-Yves Blay, France (Moderator)
There are data on the activity of epidermal growth factor re- • Alexia Bertuzzi, Ireland
ceptor and vascular endothelial growth factor receptor inhibi- • Stefan Bielack, Germany
tors. Prospective studies are ongoing. • Bodil Bjerkehagen, Norway
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu256 | iii

by guest
on 05 February 2018
Levels of evidence
II Small randomised trials or large randomised trials with a suspicion of bias (lower methodological quality) or meta-analyses of such trials
or of trials with demonstrated heterogeneity
a
• Sylvie Bonvalot, France A Consensus meeting was specifically held on chordoma,

• Ioannis Boukovinas, Greece whose output was a separate position paper but which also con-
• Paolo Bruzzi, Italy tributed to the Chordoma paragraph of the ESMO Guidelines.
• Angelo Paolo Dei Tos, Italy In addition to some of the above mentioned experts, it was also
• Palma Dileo, UK made up by the following panellists:
• Mikael Eriksson, Sweden
• Alexander Fedenko, Russian Federation • Stephanie Bolle, France
• Andrea Ferrari, Italy • Rodolfo Capanna, Italy
• Stefano Ferrari, Italy • Thomas Delaney, USA
• Hans Gelderblom, Belgium • Francesco Doglietto, Italy
• Robert Grimer, UK • Piero Fossati, Italy
• Alessandro Gronchi, Italy • Lee Jeys, UK
• Rick Haas, Netherlands • Bernd Kasper, Germany
• Kirsten Sundby Hall, Norway • Andreas Leithner, Austria
• Peter Hohenberger, Germany • Ole-Jacob Norum, Norway
• Rolf Issels, Germany • Stefano Radaelli, Italy
• Heikki Joensuu, Finland • Susanne Scheipl, Austria
• Ian Judson, UK • Elena Tamborini, Italy
• Axel Le Cesne, France • Mathias Uhl, Germany
• Saskia Litière, Belgium • Carmen L.A. Vleggert-Lankamp, Netherlands
• Javier Martin-Broto, Spain
• Ofer Merimsky, Israel
• Michael Montemurro, UK acknowledgements
• Carlo Morosi, Italy We deeply thank Barbara Doré and Estelle Lecointe (from
• Piero Picci, Italy SPAEN), and Hans Keulen (from Chordoma Foundation), who
• Isabelle Ray-Coquard, France observed the consensus conference as patient representatives.
• Peter Reichardt, Germany
• Piotr Rutkowski, Poland
• Marcus Schlemmer, Germany
• Silvia Stacchiotti, Italy Stefan Bielack declared: consultancy/advisory board/speakers
• Valter Torri, Italy bureau from Merck, IDM/Takeda, Roche, Celgene, Bayer, and
• Annalisa Trama, Italy Chugai. Sylvie Bonvalot declared: travel grants from PharmaMar,
• Frits Van Coevorden, Netherlands Nanobiotix, and honoraria from Novartis. Paolo G. Casali
• Winette Van der Graaf, Netherlands declared: consultancy/honoraria: Amgen Domplé, ARIAD,
• Daniel Vanel, Italy Bayer, GlaxoSmithKline, Infinity, Janssen Cilag, Merck Sharp &
• Eva Wardelmann, Germany Dohme, Novartis, Pfizer, PharmaMar, Sanofi. Angelo Paolo
iii | The ESMO/European Sarcoma Network Working Group Volume 25 | Supplement 3 | September 2014
by guest
on 05 February 2018
Dei Tos declared: speakers’ bureau: Novartis Oncology, Pfizer, 2. Hansen MF, Seton M, Merchant A. Osteosarcoma in Paget’s disease of bone.
GlaxoSmithKline, and PharmaMar. Thomas Delaney has J Bone Miner Res 2006; 21(Suppl 2): 58–63.
reported that he owned common stock in GlaxoSmithKline that 3. Fuchs B, Pritchard DJ. Etiology of osteosarcoma. Clin Orthop Relat Res 2002;
40–52.
was sold in early 2014; membership of the Amgen Giant Cell
4. Cotterill SJ, Parker L, Malcolm AJ et al. Incidence and survival for cancer in
Tumor Global Scientific Advisory Board and received compensa-
children and young adults in the North of England, 1968–1995: a report from the
tion for attending a meeting of this advisory board in 2012. Northern Region Young Persons’ Malignant Disease Registry. Br J Cancer 2000;
Mikael Eriksson declared: honoraria from Novartis, Swedish 83: 397–403.
Orphan Biovitrum, GlaxoSmithKline, Merck Sharp & Dohme, 5. Bovée JV, Cleton-Jansen AM, Taminiau AH, Hogendoorn P. Emerging pathways in
and Pfizer. Alexander Fedenko declared: speakers’ bureau: the development of chondrosarcoma of bone and implications for targeted
Johnson & Johnson, GlaxoSmithKline, and Roche. Stefano treatment. Lancet Oncol 2005; 6: 599–607.
Ferrari declared: speakers’ honoraria from Takeda; advisory 6. van den Berg H, Kroon HM, Slaar A, Hogendoorn P. Incidence of biopsy-proven
board for Amgen; research grants from Mulmed, Amgen, and bone tumors in children: a report based on the Dutch pathology registration
‘PALGA’. J Pediatr Orthop 2008; 28: 29–35.
Morphotek. Hans Gelderblom declared research grants from
7. Hauben EI, Hogendoorn PCW. Epidemiology of primary bone tumors and economical
Novartis, Pfizer, PharmaMar, GlaxoSmithKline, Eisai, and Bayer. aspects of bone metastases. In Heymann D (ed), Bone Cancer. Progression and
Robert Grimer declared: research grant: Amgen. Alessandro Therapeutic Approaches, 1st edition. London: Academic Press 2009; 3–8.
Gronchi declared: advisory board: Novartis; honoraria: Novartis 8. Malhas AM, Grimer RJ, Abudu A et al. The final diagnosis in patients with a
and Pfizer. Rolf Issels declared: consultancy/honoraria: suspected primary malignancy of bone. J Bone Joint Surg Br 2011; 93:
PharmaMar, Bayer, and Therm Med. Heikki Joensuu declared: 980–983.
research funding to institute from Novartis. Bernd Kasper 9. Grimer RJ, Briggs TW. Earlier diagnosis of bone and soft-tissue tumors. J Bone
declared: consultancy/honoraria: Novartis, GlaxoSmithKline, and Joint Surg Br 2010; 92: 1489–1492.
PharmaMar; research grants: PharmaMar. Axel Le Cesne 10. Enneking WF. The issue of the biopsy. J Bone Joint Surg Am 1982; 64:
1119–1120.
declared: honoraria: Novartis, PharmaMar, GlaxoSmithKline,
11. Andreou D, Bielack SS, Carrle D et al. The influence of tumor- and treatment
and Pfizer. Javier Martin-Broto declared advisory boards for
related factors on the development of local recurrence in osteosarcoma after
GlaxoSmithKline, Novartis, and PharmaMar. Ofer Merimsky adequate surgery. An analysis of 1355 patients treated on neoadjuvant Cooperative
declared: speakers’ honoraria: GlaxoSmithKline and Lilly; advis- Osteosarcoma Study Group protocols. Ann Oncol 2011; 22: 1228–1235.
ory board: Boehringer Ingelheim and Medison; research grant: 12. van den Berg H, Slaar A, Kroon HM et al. Results of diagnostic review in pediatric
Roche. Piero Picci declared advisory board for Takeda. Piotr bone tumors and tumorlike lesions. J Pediatr Orthop 2008; 28: 561–564.
Rutkowski declared honoraria from Novartis, Pfizer, Bristol- 13. Mankin HJ, Lange TA, Spanier SS. The hazards of biopsy in patients with
Myers Squibb, Roche, and GlaxoSmithKline; advisory board for malignant primary bone and soft-tissue tumors. J Bone Joint Surg Am 1982; 64:
Novartis, GlaxoSmithKline, Merck Sharp & Dohme, and Bayer. 1121–1127.
14. Meyer JS, Nadel HR, Marina N et al. Imaging guidelines for children with Ewing
Susanne Scheipl declared grants and personal fees from the
sarcoma and osteosarcoma: a report from the Children’s Oncology Group
Chordoma Foundation, as well as non-financial support from BoneTumor Committee. Pediatr Blood Cancer 2008; 51: 163–170.
GlaxoSmithKline, and non-financial support from PharmaMar 15. WHO Classification of Tumors. In Fletcher CDM, Bridge JA, Hogendoorn PCW,
in course of her current research project on chordoma at the UCL Mertens F (eds), WHO Classification of Tumours of Soft Tissue and Bone. Lyon:
Cancer Institute in London. Marcus Schlemmer declared: honor- IARC 2013.
aria from Novartis, Pfizer, and Teva; research grants from 16. Abdul-Karim FW, Bauer TW, Kilpatrick SE et al. Recommendations for the
Novartis. Silvia Stacchiotti declared: research grants: Novartis, reporting of bone tumors. Association of Directors of Anatomic and Surgical
Pfizer, PharmaMar, GlaxoSmithKline, Amgen, and Bayer. Frits Pathology. Hum Pathol 2004; 35: 1173–1178.
Van Coevorden declared travel grants from Novartis and 17. Enneking WF, Spanier SS, Goodman MA. A system for the surgical staging of
musculoskeletal sarcoma. Clin Orthop Relat Res 1980; 153: 106–120.
PharmaMar. Carmen Vleggeert-Lankamp declared research
18. Sobin LH, Gospodarowicz MK, Wittekind CH (eds). TNM Classification of Malignant
grants from Medtronic, B Braun, Paradigm Spine and the
Tumors, 7th edition. Oxford: Wiley-Blackwell 2009.
Eurospine Foundation, the National Health Organization, 19. Picci P, Vanel D, Briccoli A et al. Computed tomography of pulmonary metastases
Netherlands, and the Dutch Brain Foundation. Winette Van der from osteosarcoma: the less poor technique. A study of 51 patients with
Graaf declared: research funding from GlaxoSmithKline, Novartis histological correlation. Ann Oncol 2001; 12: 1601–1604.
and Pfizer. The following authors have declared no potential con- 20. Benz MR, Tchekmedyian N, Eilber FC et al. Utilization of positron emission
flicts of interest: Alexia Bertuzzi, Bodil Bjerkehagen, Stephanie tomography in the management of patients with sarcoma. Curr Opin Oncol 2009;
Bolle, Ioannis Boukovinas, Rodolfo Capanna, Palma Dileo, 21: 345–351.
Francesco Doglietto, Andrea Ferrari, Piero Fossati, Rick Haas, 21. Bramer JA, van Linge JH, Grimer RJ, Scholten RJ. Prognostic factors in localized
extremityosteosarcoma: a systematic review. Eur J Surg Oncol 2009; 35: 1030–1036.
Kirsten Sundby Hall, Lee Jeys, Andreas Leithner, Saskia Litière,
22. Leavey PJ, Collier AB. Ewing sarcoma: prognostic criteria, outcomes and future
Michael Montemurro, Carlo Morosi, Stefano Radaelli, Isabelle
treatment. Expert Rev Anticancer Ther 2008; 8: 617–624.
Ray-Coquard, Elena Tamborini, Valter Torri, Daniel Vanel and
23. Bielack SS, Kempf-Bielack B, Delling G et al. Prognostic factors in high-grade
Eva Wardelmann. The other authors have not reported any po- osteosarcoma of the extremities or trunk: an analysis of 1702 patients treated on
tential conflicts of interest. neoadjuvant cooperative osteosarcoma study group protocols. J Clin Oncol 2002;
20: 776–790.
24. Picci P, Sangiorgi L, Rougraff BT et al. Relationship of chemotherapy-induced
references necrosis and surgical margins to local recurrence in osteosarcoma. J Clin Oncol
1. Stiller CA, Craft AW, Corazziari I. EUROCARE Working Group. Survival of children 1994; 12: 2699–2705.
with bone sarcoma in Europe since 1978: results from the EUROCARE study. Eur J 25. Grimer RJ, Bielack S, Flege S et al. Periosteal osteosarcoma—a European review
Cancer 2001; 37: 760–766. of outcome. Eur J Cancer 2005; 41: 2806–2811.
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu256 | iii

by guest
on 05 February 2018
26. Cesari M, Alberghini M, Vanel D et al. Periosteal osteosarcoma: a single institution 48. Machado I, Noguera R, Pellin A et al. Molecular diagnosis of Ewing sarcoma family
experience. Cancer 2011; 117: 1731–1735. of tumors: a comparative analysis of 560 cases with FISH and RT-PCR. Diagn Mol
27. Bramer JA, Abudu AA, Grimer RJ et al. Do pathological fractures influence Pathol 2009; 18: 189–199.
survival and local recurrence rate in bony sarcomas? Eur J Cancer 2007; 43: 49. Cangir A, Vietti TJ, Gehan EA et al. Ewing’s sarcoma metastatic at diagnosis.
1944–1951. Results and comparisons of two intergroup Ewing’s sarcoma studies. Cancer
28. Ferrari S, Smeland S, Mercuri M et al. Neoadjuvant chemotherapy with high dose 1990; 66: 887–893.
ifosfamide, high-dose methotrexate, cisplatin, and doxorubicin for patients with 50. Bernstein ML, Devidas M, Lafreniere D et al. Intensive therapy with growth factor
localized osteosarcoma of the extremity: a joint study by the Italian and support for patients with Ewing tumor metastatic at diagnosis: Pediatric Oncology
Scandinavian Sarcoma Groups. J Clin Oncol 2005; 23: 8845–8852. Group/Children’s Cancer Group Phase II Study 9457—a report from the Children’s
29. Arndt CA, Crist WM. Common musculoskeletal tumors of childhood and Oncology Group. J Clin Oncol 2006; 24: 152–159.
adolescence. N Engl J Med 1999; 341: 342–352. 51. Bacci G, Ferrari S, Bertoni F et al. Prognostic factors in non metastatic Ewing’s
30. Lewis IJ, Nooij MA, Whelan J et al. Improvement in histologic response but not sarcoma of bone treated with adjuvant chemotherapy: analysis of 359 patients at
survival in osteosarcoma patients treated with intensified chemotherapy: a the Istituto Ortopedico Rizzoli. J Clin Oncol 2000; 18: 4–11.
randomized phase III trial of the European Osteosarcoma Intergroup. J Natl Cancer 52. Bacci G, Forni C, Longhi A et al. Long-term outcome for patients with non
Inst 2007; 99: 112–128. metastatic Ewing’s sarcoma treated with adjuvant and neoadjuvant
31. Carrle D, Bielack SS. Current strategies of chemotherapy in osteosarcoma. Int chemotherapies. 402 patients treated at Rizzoli between 1972 and 1992. Eur J
Orthop 2006; 30: 445–451. Cancer 2004; 40: 73–83.
32. Bielack SS, Machatschek JN, Flege S, Jürgens H. Delaying surgery with 53. Cotterill SJ, Ahrens S, Paulussen M et al. Prognostic factors in Ewing’s tumor of
chemotherapy for osteosarcoma of the extremities. Expert Opin Pharmacother bone: analysis of 975 patients from the European Intergroup Cooperative Ewing’s
2004; 5: 1243–1256. Sarcoma Study Group. J Clin Oncol 2000; 18: 3108–3114.
33. Goorin AM, Schwartzentruber DJ, Devidas M et al. Presurgical chemotherapy 54. Paulussen M, Ahrens S, Craft AW et al. Ewing’s tumors with primary lung
compared with immediate surgery and adjuvant chemotherapy for non metastatic metastases: survival analysis of 114 (European Intergroup) Cooperative Ewing’s
osteosarcoma: Pediatric Oncology Group Study POG-8651. J Clin Oncol 2003; Sarcoma Studies patients. J Clin Oncol 1998; 16: 3044–3052.
21: 1574–1580. 55. Pinkerton CR, Bataillard A, Guillo S et al. Treatment strategies for metastatic
34. Whelan J, Patterson D, Perisoglou M et al. The role of interferons in the treatment Ewing’s sarcoma. Eur J Cancer 2001; 37: 1338–1344.
of osteosarcoma. Pediatr Blood Cancer 2010; 54: 350–354. 56. Le Deley MC, Delattre O, Schaefer KL et al. Impact of EWS-ETS fusion type on
35. Meyers PA, Schwartz CL, Krailo MD et al. Osteosarcoma: the addition of muramyl disease progression in Ewing’s sarcoma/peripheral primitive neuroectodermal
tripeptide to chemotherapy improves overall survival—a report from the Children’s tumor: prospective results from the cooperative Euro-E.W.I.N.G. 99 trial. J Clin
Oncology Group. J Clin Oncol 2008; 26: 633–638. Oncol 2010; 28: 1982–1988.
36. Hunsberger S, Freidlin B, Smith MA. Complexities in interpretation of 57. van Doorninck JA, Ji L, Schaub B et al. Current treatment protocols have
osteosarcoma clinical trial results. J Clin Oncol 2008; 26: 3103–3104. eliminated the prognostic advantage of type 1 fusions in Ewing sarcoma: a report
37. van der Woude HJ, Bloem JL, Hogendoorn PC. Preoperative evaluation and from the Children’s Oncology Group. J Clin Oncol 2010; 28: 1989–1994.
monitoring chemotherapy in patients with high-grade osteogenic and Ewing’s 58. Schuck A, Ahrens S, Paulussen M et al. Local therapy in localized Ewing tumors:
sarcoma: review of current imaging modalities. Skeletal Radiol 1998; 27: results of 1058 patients treated in the CESS 81, CESS 86, and EICESS 92 trials.
57–71. Int J Radiat Oncol Biol Phys 2003; 55: 168–177.
38. van der Woude HJ, Bloem JL, Verstraete KL et al. Osteosarcoma and Ewing’s 59. Bernstein M, Kovar H, Paulussen M et al. Ewing’s sarcoma family of tumors:
sarcoma after neoadjuvant chemotherapy: value of dynamic MR imaging in detecting current management. Oncologist 2006; 11: 503–519.
viable tumor before surgery. AJR Am J Roentgenol 1995; 165: 593–598. 60. Grier HE, Krailo MD, Tarbell NJ et al. Addition of ifosfamide and etoposide to
39. Grimer RJ, Cannon SR, Taminiau AM et al. Osteosarcoma over the age of forty. standard chemotherapy for Ewing’s sarcoma and primitive neuroectodermal tumor
Eur J Cancer 2003; 39: 157–163. of bone. N Engl J Med 2003; 348: 694–701.
40. Kager L, Zoubek A, Potschger U et al. Primary metastatic osteosarcoma: 61. Nesbit ME, Jr, Gehan EA, Burgert EO, Jr et al. Multimodal therapy for the
presentation and outcome of patients treated on neoadjuvant Cooperative management of primary, nonmetastatic Ewing’s sarcoma of bone: a long term
Osteosarcoma Study Group protocols. J Clin Oncol 2003; 21: 2011–2018. follow-up of the first intergroup study. J Clin Oncol 1990; 8: 1664–1674.
41. Ferrari S, Briccoli A, Mercuri M et al. Postrelapse survival in osteosarcoma of the 62. Paulussen M, Craft AW, Lewis I et al. Results of the EICESS-92 study: two
extremities: prognostic factors for long-term survival. J Clin Oncol 2003; 21: randomized trials of Ewing’s sarcoma treatment—cyclophosphamide compared
710–715. with ifosfamide in standard-risk patients and assessment of benefit of etoposide
42. Kempf-Bielack B, Bielack SS, Jurgens H et al. Osteosarcoma relapse after added to standard treatment in high-risk patients. J Clin Oncol 2008; 26:
combined modality therapy: an analysis of unselected patients in the Cooperative 4385–4393.
Osteosarcoma Study Group (COSS). J Clin Oncol 2005; 23: 559–568. 63. Ferrari S, Sundby Hall K, Luksch R et al. Nonmetastatic Ewing family tumors: high-
43. Berger M, Grignani G, Giostra A et al. 153 Samarium-EDTMP administration dose chemotherapy with stem cell rescue in poor responder patients. Results of
followed by hematopoietic stem cell support for bone metastases in osteosarcoma the Italian Sarcoma Group/Scandinavian Sarcoma Group III protocol. Ann Oncol
patients. Ann Oncol 2012; 23: 1899–1905. 2011; 22: 1221–1227.
44. Aurias A, Rimbaut C, Buffe D et al. Translocation involving chromosome 22 in 64. Shapeero LG, Vanel D. Imaging evaluation of the response of high-grade
Ewing’s sarcoma. A cytogenetic study of four fresh tumors. Cancer Genet osteosarcoma and Ewing sarcoma to chemotherapy with emphasis on dynamic
Cytogenet 1984; 12: 21–25. contrast-enhanced magnetic resonance imaging. Semin Musculoskelet Radiol
45. Turc-Carel C, Philip I, Berger MP et al. Chromosome study of Ewing’s sarcoma 2000; 4: 137–146.
(ES) cell lines. Consistency of a reciprocal translocation t (11;22)(q24;q12). 65. Ladenstein R, Pötschger U, Le Deley MC et al. Primary disseminated multifocal
Cancer Genet Cytogenet 1984; 12: 1–19. Ewing sarcoma: results of the Euro-EWING 99 trial. J Clin Oncol 2010; 28:
46. Zoubek A, Pfleiderer C, Salzer-Kuntschik M et al. Variability of EWS chimaeric 3284–3291.
transcripts in Ewing tumors: a comparison of clinical and molecular data. Br J 66. Haeusler J, Ranft A, Boelling T et al. The value of local treatment in patients with
Cancer 1994; 70: 908–913. primary, disseminated, multifocal Ewing sarcoma (PDMES). Cancer 2010; 116:
47. Sorensen PH, Lessnick SL, Lopez-Terrada D et al. A second Ewing’s sarcoma 443–450.
translocation, t(21;22), fuses the EWS gene to another ETS-family transcription 67. Stahl M, Ranft A, Paulussen M et al. Risk of recurrence and survival after relapse
factor, ERG. Nat Genet 1994; 6: 146–151. in patients with Ewing sarcoma. Pediatr Blood Cancer 2011; 57: 549–553.
iii | The ESMO/European Sarcoma Network Working Group Volume 25 | Supplement 3 | September 2014
by guest
on 05 February 2018
68. Ferrari S, del Prever AB, Palmerini E. Response to high-dose ifosfamide in patients 82. Fox E, Patel S, Wathen JK et al. Phase II study of sequential gemcitabine followed
with advanced/recurrent Ewing sarcoma. Pediatr Blood Cancer 2009; 52: by docetaxel for recurrent Ewing sarcoma, osteosarcoma, or unresectable or
581–584. locally recurrent chondrosarcoma: results of Sarcoma Alliance for Research
69. Wagner LM, McAllister N, Goldsby RE et al. Temozolomide and intravenous through Collaboration Study 003. Oncologist 2012; 17: 321.
irinotecan for treatment of advanced Ewing sarcoma. Pediatr Blood Cancer 2007; 83. Chawla S, Henshaw R, Seeger L et al. Safety and efficacy of denosumab for adults
48: 132–139. and skeletally mature adolescents with giant cell tumour of bone: interim analysis of
70. Hunold A, Weddeling N, Paulussen M et al. Topotecan and cyclophosphamide in an open-label, parallel-group, phase 2 study. Lancet Oncol 2013; 14: 901–908.
patients with refractory or relapsed Ewing tumors. Pediatr Blood Cancer 2006; 47: 84. Vujovic S, Henderson S, Presneau N et al. Brachyury, a crucial regulator of notochordal
795–800. development, is a novel biomarker for chordomas. J Pathol 2006; 209: 157–165.
71. Souhami RL, Tannock I, Hohenberger JC (eds) et al. Oxford Textbook of Oncology. 85. Rodallec MH, Feydy A, Larousserie F et al. Diagnostic imaging of solitary tumours
Oxford: Oxford University Press 2002. of the spine: what to do and say. RadioGraphics 2008; 28: 1019–1041.
72. Pakos EE, Grimer RJ, Peake D et al. The ‘other’ bone sarcomas: prognostic factors 86. Di Maio S, Rostomily R, Sekhar LN. Current surgical outcomes for cranial base
and outcomes of spindle cell sarcomas of bone. J Bone Joint Surg Br 2011; 93: chordomas: cohort study of 95 patients. Neurosurgery 2012; 70: 1355–1360.
1271–1278. 87. Yasuda M, Bresson D, Chibbaro S et al. Chordomas of the skull base and cervical
73. Eefting D, Schrage YM, Geirnaerdt MJ et al. Assessment of interobserver spine: clinical outcomes associated with a multimodal surgical resection combined
variability and histologic parameters to improve reliability in classification and with proton-beam radiation in 40 patients. Neurosurg Rev 2012; 35: 171–182.
grading of central cartilaginous tumors. Am J Surg Pathol 2009; 33: 50–57. 88. Schulz-Ertner D, Karger CP, Feuerhake A et al. Effectiveness of carbon ion
74. Gelderblom H, Hogendoorn PC, Dijkstra SD et al. The clinical approach towards radiotherapy in the treatment of skull-base chordomas. Int J Radiat Oncol Biol
chondrosarcoma. Oncologist 2008; 13: 320–329. Phys 2007; 68: 449–457.
75. Riedel RF, Larrier N, Dodd L et al. The clinical management of chondrosarcoma. 89. Fuchs B, Dickey ID, Yaszemski MJ et al. Operative management of sacral
Curr Treat Options Oncol 2009; 10: 94–106. chordoma. J Bone Joint Surg Am 2005; 87: 2211–2216.
76. Geirnaerdt MJ, Hogendoorn PC, Bloem JL et al. Cartilaginous tumors: fast 90. Stacchiotti S, Casali PG, Lo Vullo S et al. Chordoma of the mobile spine and
contrast-enhanced MR imaging. Radiology 2000; 214: 539–546. sacrum: a retrospective analysis of a series of patients surgically treated at two
77. Cesari M, Bertoni F, Bacchini P et al. Mesenchymal chondrosarcoma. An analysis referral centers. Ann Surg Oncol 2010; 17: 211–219.
of patients treated at a single institution. Tumori 2007; 93: 423–427. 91. Clarke MJ, Dasenbrock H, Bydon A et al. Posterior-only approach for en bloc
78. Dantonello TM, Int-Veen C, Leuschner I et al. Mesenchymal chondrosarcoma of sacrectomy: clinical outcomes in 36 consecutive patients. Neurosurgery 2012;
soft tissues and bone in children, adolescents, and young adults: experiences of 71: 357–364; discussion 364.
the CWS and COSS study groups. Cancer 2008; 112: 2424–2431. 92. Ares C, Hug EB, Lomax AJ et al. Effectiveness and safety of spot scanning proton
79. Dickey ID, Rose PS, Fuchs B et al. Dedifferentiated chondrosarcoma: the role of radiation therapy for chordomas and chondrosarcomas of the skull base: first long-
chemotherapy with updated outcomes. J Bone Joint Surg Am 2004; 86-A: term report. Int J Radiat Oncol Biol Phys 2009; 75: 1111–1118.
2412–2418. 93. Delaney TF, Liebsch NJ, Pedlow FX et al. Phase II study of high-dose photon/
80. Grimer RJ, Gosheger G, Taminiau A et al. Dedifferentiated chondrosarcoma: proton radiotherapy in the management of spine sarcomas. Int J Radiat Oncol Biol
prognostic factors and outcome from a European group. Eur J Cancer 2007; 43: Phys 2009; 74: 732–739.
2060–2065. 94. Stacchiotti S, Longhi A, Ferraresi V et al. A phase II study on imatinib in advanced
81. Noël G, Feuvret L, Ferrand R et al. Radiotherapeutic factors in the management of chordoma. J Clin Oncol 2012; 30: 914–920.
cervical-basal chordomas and chondrosarcomas. Neurosurgery 2004; 55: 95. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections among
1252–1260. hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33: 139–144.
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu256 | iii

by guest
on 05 February 2018
Gastrointestinal stromal tumours: ESMO Clinical

and follow-up†
incidence remains unclear. Therefore, the standard approach to these

patients is endoscopic ultrasound assessment and then annual
Gastrointestinal stromal tumours (GISTs) are rare tumours, follow-up, reserving excision for patients whose tumour
with an estimated unadjusted incidence of around 1/100 000/ increases in size or becomes symptomatic. Alternatively, the de-
year [1]. This only covers clinically relevant GISTs, since it is cision can be shared with the patient to make a histological
likely that a much higher number of microscopic lesions could assessment, also depending on age, life expectancy, and co-mor-
be found pathologically, if looked for. bidities. If follow-up is the choice, an evidence-based optimal
The median age is around 60–65 years, with a wide range. surveillance policy is lacking. A logical choice may be to have a
Occurrence in children is very rare, although paediatric GIST short-term first control (e.g. at 3 months), and then, in the case
represents a distinct subset, marked by female predominance, of no evidence of growth, a more relaxed follow-up schedule
absence of KIT/platelet-derived growth factor alpha (PDGFRA)
clinical practice
may be selected.
guidelines
mutations, gastric multicentric location, and possible lymph In a histologically proven small GIST, standard treatment is
node metastases [2]. excision, unless major morbidity is expected. Alternatively, in
Some syndromes are linked to GISTs: the case of a low-risk GIST, the decision can be shared with the
• the Carney triad syndrome in succinate dehydrogenase patient to follow-up the lesion. However, the standard approach
subunit B (SDHB)-deficient GIST, marked by gastric GISTs, to rectal (or recto-vaginal space) nodules is biopsy/excision after
paraganglioma, and pulmonary chondromas (these may ultrasound assessment, regardless of the tumour size, because
occur at different ages) [3]. the risk of a GIST at this site is higher and the local implications
• Carney-Stratakis syndrome, marked by germ-line mutations for surgery are more critical. A follow-up policy may be an
of SDH subunits A, B, C, and D, leading to a dyad of GIST option, to be shared with the patient, in the case of small lesions
and paraganglioma [4, 5]. and in specific clinical contexts.
• neurofibromatosis type 1, marked by wild-type, often multi- The standard approach to nodules ≥2 cm in size is biopsy/ex-
centric GIST, predominantly located to the small bowel [6]. cision, because, if GIST, they are associated with a higher risk. If
there is an abdominal nodule not amenable to endoscopic
Families with germ-line autosomal dominant mutations of KIT assessment, laparoscopic/laparotomic excision is the standard
are a rare finding, presenting with multiple GISTs at an early approach. If there is a mass, especially if surgery is likely to be a
age. multivisceral resection, multiple core needle biopsies are the
standard approach. They should be obtained through endoscop-
ic ultrasound guidance, or through an ultrasound/computed
diagnosis tomography (CT)-guided percutaneous approach. This may
allow the surgeon to plan the best approach according to the
When small oesophago-gastric or duodenal nodules <2 cm in
histological diagnosis and may avoid surgery for diseases that
size are detected, endoscopic biopsy may be difficult and laparo-
do not merit it (e.g. lymphomas, mesenteric fibromatosis, and
scopic/laparotomic excision may be the only way to make a
germ cell tumours). The risk of peritoneal contamination is neg-
histological diagnosis. Many of these small nodules, if diagnosed
ligible if the procedure is properly carried out. Moreover, lesions
as GISTs, will be low risk, or entities whose clinical significance
at risk in this regard (e.g. cystic masses) should be biopsied only
in specialised centres. Immediate laparoscopic/laparotomic exci-
L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland. sion is an alternative on an individualised basis, especially if
E-mail: clinicalguidelines@esmo.org surgery is limited. If a patient presents with obvious metastatic
†
disease, then a biopsy of the metastatic focus is sufficient and the
Approved by the ESMO Guidelines Working Group: December 2006, last update July
2014. This publication supersedes the previously published version—Ann Oncol 2012; patient usually does not require a laparotomy for diagnostic
23(Suppl 7): vii49–vii55. purposes. The tumour sample should be fixed in 4% buffered

by guest
on 05 February 2018
formalin (Bouin fixation should not be used, since it prevents and follow-up. Magnetic resonance imaging (MRI) may be an
molecular analysis). alternative. For rectal GISTs, MRI provides better preoperative
Pathologically, the diagnosis of GIST relies on morphology staging information. Chest CT scan or X-rays and routine la-
and immunohistochemistry, the latter being positive for CD117 boratory testing complement the staging work-up of the asymp-
and/or DOG1 [7, 8]. A proportion of GISTs (in the range of tomatic patient. The evaluation of fluorodeoxyglucose (FDG)
5%) are CD117-negative. The mitotic count has a prognostic value uptake using an FDG-positron emission tomography (PET)
and should be expressed as the number of mitoses on a total scan, or FDG-PET–CT/MRI, is useful mainly when early detec-
area of 5 mm2 (which replaces the former 50 high-power field tion of the tumour response to molecular-targeted therapy is of
area). Mutational analysis for known mutations involving KIT special interest.
and PDGFRA genes can confirm the diagnosis of GIST, if doubtful
(particularly in CD117/DOG1-negative suspect GIST). Mutational
analysis has a predictive value for sensitivity to molecular-targeted treatment
therapy, and prognostic value, so that its inclusion in the diagnostic Multidisciplinary treatment planning is needed (involving
work-up of all GISTs should be considered standard practice pathologists, radiologists, surgeons, and medical oncologists, as
(with the possible exclusion of <2 cm non-rectal GISTs, which are well as gastroenterologists, nuclear medicine specialists, etc., as
very unlikely ever to be candidates for medical treatment). applicable), such as that which is available in reference centres
Centralisation of mutational analysis in a laboratory enrolled in an for sarcomas and GISTs, and/or within reference networks
external quality assurance programme and with expertise in the sharing multidisciplinary expertise and treating a high number
disease may be useful. In KIT/PDGFRA wild type (WT) GIST, of patients annually.
immunohistochemistry for SDHB is done. The diagnosis should
be made or confirmed by an expert pathologist at a reference
centre. The collection of fresh/frozen tissue is encouraged, because localised disease
new molecular pathology assessments could be made at a later The standard treatment of localised GISTs is complete surgical
stage in the patient’s interest. Informed consent for tumour excision of the lesion, with no dissection of clinically negative
banking should be sought, enabling later analyses and research, as lymph nodes [III, A]. If laparoscopic excision is planned, the
long as this is allowed by local and international guidelines. technique needs to follow the principles of oncological surgery
[13] [III, A]. A laparoscopic approach is clearly discouraged in
patients who have large tumours, because of the risk of tumour
stage classification and risk assessment rupture, which is associated with a very high risk of relapse. R0
The TNM classification has several limitations and its use is excision is the goal (i.e. an excision whose margins are clear of
therefore not recommended. tumour cells). When R0 surgery implies major functional seque-
Prognostic factors are the mitotic rate, tumour size and tumour lae, and preoperative medical treatment has not helped or
site (gastric GISTs have a better prognosis than small bowel or cannot be exploited, the decision can be shared with the patient
rectal GISTs). Tumour rupture is an additional adverse prognos- to accept possible R1 (microscopically positive) margins (i.e. ex-
tic factor and should be recorded, whether it took place before or cision margins containing tumour cells) [IV, B]. This is all the
during surgery. Mutational status has not been incorporated in more acceptable for low-risk lesions, given the lack of any
any risk classification at the moment, although some genotypes formal demonstration that R1 surgery is associated with a worse
have a distinct natural history, and, above all, KIT/PDGFRA WT overall survival (OS).
GISTs have peculiar clinical presentations and course. If R1 excision was already carried out, re-excision may be an
Several risk classifications have been proposed. A widely used option, provided the original site of lesion can be found, and
risk classification was proposed by the Armed Forces Institute of major functional sequelae are not foreseen.
Pathology, which incorporates the primary tumour site, mitotic The risk of relapse can be substantial, as defined by available
count, and tumour size, i.e. the three main prognostic factors in risk classifications. Adjuvant treatment with imatinib for 3 years
localised GISTs [9, 10]. A nomogram utilising all three criteria was associated with a relapse-free survival and OS advantage in
has been developed on another series [11]. When using these a randomised trial in comparison with 1 year of therapy in
tools, it is important to appreciate that the mitotic index and high-risk patients [14]. Previously, a placebo-controlled trial
tumour size are non-linear continuous variables, so that thresh- demonstrated that imatinib dosed for a planned duration of
olds are interpreted wisely. Prognostic contour maps were gen- 1 year is able to prolong relapse-free survival in localised GISTs
erated through a pool of series of GIST patients not treated with having a diameter of 3 cm or more with a macroscopically com-
adjuvant therapy, which incorporate the mitotic index and plete resection [15]. Therefore, adjuvant therapy with imatinib
tumour size as continuous non-linear variables, while tumour for 3 years is the standard treatment of patients with a signifi-
rupture is considered in addition to tumour site [12]. They have cant risk of relapse [I, A]. Adjuvant therapy should not be
been validated against a reference series. considered when the risk is low. There is room for shared
decision-making when the risk is intermediate [16].
Mutational analysis is critical to making a clinical decision
staging procedures about adjuvant therapy. In fact, there is consensus that PDGFRA
Staging procedures take into account the fact that most relapses D842V-mutated GISTs should not be treated with any adjuvant
affect the peritoneum and the liver. Contrast-enhanced abdom- therapy, given the lack of sensitivity of this genotype both in
inal and pelvic CT scan is the investigation of choice for staging vitro and in vivo [IV, A]. Given the data supporting the use of a
iii | The ESMO/European Sarcoma Network Working Group Volume 25 | Supplement 3 | September 2014
by guest
on 05 February 2018
higher dose of imatinib (800 mg daily) in the case of an exon 9 excised [28] [II, B]. When treatment is started, the patient should
KIT mutation in advanced GIST, many clinicians prefer to use be alerted to the importance of compliance with therapy, as well
this dose even in the adjuvant setting for this genotype [17–19]. as of interactions with concomitant medications and foods, and of
Regulatory problems may limit this practice, which is not backed the best ways to handle side-effects. Dose intensity should be
by any controlled trial in the adjuvant setting. There is consensus maintained by proper management of side-effects, and a correct
on avoiding adjuvant treatment in neurofibromatosis 1-related policy of dose reductions and interruptions should be applied in
GISTs, which are insensitive to imatinib in the advanced setting. the case of excessive, persistent toxicity. Retrospective data suggest
On the other hand, a consensus is lacking among experts about that suboptimal plasma levels of imatinib are associated with a
whether wild-type SDH-negative GISTs should be treated with worse outcome, although a correlation with the outcome has not
adjuvant therapy. This reflects their lower sensitivity to imatinib, been established prospectively [29]. Aside from its potential use to
as well as their peculiar natural history, which is often more in- tailor the imatinib dose, assessment of plasma level may be useful
dolent, but subgroup analyses of available randomised trials are in the case of: (i) patients receiving concomitant medications that
too limited to provide sufficient evidence. European and put them at a risk of major interactions or patients with previous
International cooperation would be vital to determine best prac- surgical resections able to decrease plasma levels; (ii) unexpected
tices in the exceedingly rare paediatric GIST. observed toxicities; and (iii) progression on 400 mg, to rationally
In case of tumour rupture at the time of surgery, there has lead the physician to increase the dose to 800 mg daily.
been spillage of tumour cells into the peritoneal cavity, and Close monitoring of the tumour response should be carried
therefore occult peritoneal disease can be assumed to exist. This out in the early phases of treatment. Follow-up should be contin-
puts the patient at a very high risk of peritoneal relapse [20]. ued throughout the treatment, since the risk of secondary pro-
Therefore, these patients should be considered for imatinib gression persists over time. Complete excision of residual
therapy. The optimal duration of treatment in these cases is metastatic disease has been shown to be related to a good progno-
unknown, given the uncertainty as to whether they should be sis, provided the patient is responding to imatinib, but it has
viewed as virtually metastatic. never been demonstrated prospectively whether this is due to
If R0 surgery is not feasible, or it could be achieved through less surgery or to patient selection [30–32]. Randomised trials did not
mutilating/function sparing surgery in the case of cytoreduction prove feasible, with the exception of a small positive trial, in
(this includes total gastrectomy and all other major procedures), which all patients had peritoneal disease [33]. Thus, at the
pre-treatment with imatinib is standard [21, 22] [IV, A]. This may present time, the surgical option should be individualised after
also be the case if the surgeon believes that the surgical conduct is sharing the decision with the patient in the case of uncertainty
safer after cytoreduction (e.g. the risk of bleeding and tumour [III, C]. Surgical excision of progressing disease has not been
rupture is decreased). Following maximal tumour response, gener- rewarding in published series, but surgery of limited progression,
ally after 6–12 months, surgery is carried out. Mutational analysis such as the ‘nodule within a mass’, has been associated with a
is crucial because it helps to exclude less sensitive or resistant progression-free interval in the same range as for second-line
genotypes (e.g. PDGFRA D842V mutations) from therapy with treatment with sunitinib. Therefore, this may be a palliative
imatinib and allows the use of proper dosing for KIT exon 9 muta- option in the individual patient with limited progression, while
tions. Early tumour response assessment is mandatory, so that continuing imatinib [V, C]. Non-surgical procedures (local treat-
surgery is not delayed in the case of non-responding disease. ment, such as ablations, etc.) may be selected. In the case of
Functional imaging makes it possible to assess the tumour re- tumour progression on 400 mg, an option may be to increase the
sponse very rapidly, within a few weeks, particularly in the lack of imatinib dose to 800 mg daily [23–26] [III, B], with the possible
a mutational analysis. There are limited data to guide the physician exception of insensitive mutations (if treated with the lower
on when to stop imatinib before surgery; however, it can be safely dose). Dose escalation is particularly useful in the case of a KIT
stopped a few days or even 1 day before surgery and it can be exon 9 mutated GIST (if a higher dose was not selected from the
resumed promptly when the patient recovers from surgery. beginning), possibly in the case of changes in drug pharmacokin-
etics over time, or perhaps in the case of some molecular second-
ary alterations. False progression on imaging should be ruled out,
metastatic disease due to the response patterns (see below). Also, patient non-com-
In locally advanced inoperable and metastatic patients, imatinib pliance should be ruled out as a possible cause of tumour progres-
is standard treatment [23–26] [III, A], even if the patient previ- sion, as well as drug interactions with concomitant medications.
ously received the drug as adjuvant therapy without relapsing In the case of confirmed progression or rare intolerance on
during it. This applies also to metastatic patients who have been imatinib (after attempts to manage side-effects also through
completely relieved of all lesions surgically, though surgery as a expert advice, also exploiting dose reductions and possibly
primary approach to metastatic GIST is not recommended. The plasma level assessment), standard second-line treatment is
standard dose of imatinib is 400 mg daily [I, A]. However, data another tyrosine kinase inhibitor, sunitinib [34] [I, B]. The drug
have shown that patients with KIT exon 9 mutations fare better was proved effective in terms of PFS following a ‘4 weeks on–2
in terms of progression-free survival (PFS) on a higher dose weeks off’ regimen. Data have been provided that a continuously
level, i.e. 800 mg daily, which is therefore the standard treatment dosed daily oral regimen with a lower daily dose (37.5 mg) is ef-
in this subgroup [27] [III, A]. fective and well tolerated, although no formal comparison has
Treatment should be continued indefinitely, since treatment been carried out within a randomised clinical trial. This sched-
interruption is generally followed by relatively rapid tumour pro- ule can therefore be considered an alternative on an individua-
gression, even when lesions have been previously surgically lised basis [35] [III, B].
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu255 | iii

by guest
on 05 February 2018
After confirmed progression on sunitinib, a prospective affects the speed at which relapses take place. Risk assessment
placebo-controlled randomised trial proved that regorafenib, at based on the mitotic count, tumour size and tumour site may be
the dose of 160 mg daily for 3 every 4 weeks, is able to significant- useful in choosing the routine follow-up policy. High-risk
ly prolong PFS [36]. This therapy, as it becomes routinely avail- patients generally have a relapse within 1–3 years from the end
able, is therefore standard for the third-line targeted therapy of of adjuvant therapy. Low-risk patients may have a relapse later,
patients progressing on or failing to respond to imatinib and although this is much less likely. That said, routine follow-up
sunitinib [I, B]. schedules differ across institutions.
Patients with a metastatic GIST should be considered for par- The optimal follow-up schedules are not known. As an
ticipation in clinical trials on new therapies or combinations. example, in some institutions, high-risk patients undergo a
There is controlled evidence that patients who have already pro- routine follow-up with an abdominal CT scan or MRI every 3–6
gressed on imatinib may benefit when re-challenged with the months for 3 years during adjuvant therapy (with a tighter clin-
same drug [37]. Likewise, there is evidence that maintaining ical follow-up due to the need to manage the side-effects of adju-
treatment with an anti-tyrosine kinase agent, even in the case of vant therapy), unless contraindicated, then on cessation of
progressive disease, may slow down progression as opposed to adjuvant therapy every 3 months for 2 years, then every 6
stopping it (if no other option is available at the time). months until 5 years from stopping adjuvant therapy, and annu-
Therefore, re-challenge or continuation treatment with an anti- ally for an additional 5 years.
tyrosine kinase agent to which the patient has already been For low-risk tumours, the usefulness of a routine follow-up is
exposed is an option in patients with progression [V, B]. On the not known; if selected, this is carried out with abdominal CT
other hand, the use of combinations of anti-tyrosine kinase scan or MRI, every 6–12 months for 5 years.
agents outside of clinical studies should be discouraged, because Very low-risk GISTs probably do not deserve routine follow-
of the potential for considerable toxicity. up, although one must be aware that the risk is not nil.
X-ray exposure is a factor to take into account, especially in
low-risk GIST, with abdominal MRI being an option as an alter-
response evaluation
native.
Response evaluation is complex, and early progression, in par-
ticular, should be confirmed by an experienced team. Anti-
tumour activity translates into tumour shrinkage in the majority Table 1. Levels of evidence and grades of recommendation
of patients, but some patients may show changes only in tumour (adapted from the Infectious Diseases Society of America-United
density on CT scan, or these changes may precede delayed States Public Health Service Grading Systema)
tumour shrinkage. These changes in tumour radiological
appearance should be considered as the tumour response. Even Levels of evidence
increase in the tumour size, in particular, may be indicative of the
tumour response if the tumour density on CT scan is decreased
of good methodological quality (low potential for a bias)
[38]. Even the ‘appearance’ of new lesions may be due to their
or meta-analyses of well-conducted randomised trials
being more evident when becoming less dense. Therefore, both
without heterogeneity
tumour size and tumour density on CT scan, or consistent
changes in MRI or contrast-enhanced ultrasound, should be con- suspicion of bias (lower methodological quality) or
sidered as criteria for tumour response. An FDG-PET scan has meta-analyses of such trials or of trials with
proved to be highly sensitive in early assessment of tumour re- demonstrated heterogeneity
sponse and may be useful in cases where there is doubt, or when III Prospective cohort studies
early prediction of the response is particularly useful (e.g. pre- IV Retrospective cohort studies or case–control studies
operative cytoreductive treatments). A small proportion of GISTs V Studies without control group, case reports, and experts
have no FDG uptake, however. The absence of tumour progres- opinions
sion at 6 months [39] after months of treatment also amounts to
a tumour response. On the other hand, tumour progression may
not be accompanied by changes in the tumour size. In fact, some A Strong evidence for efficacy with a substantial clinical
increase in the tumour density within tumour lesions may be benefit, strongly recommended
indicative of tumour progression. A typical progression pattern is B Strong or moderate evidence for efficacy but with a limited
the ‘nodule within the mass’, by which a portion of a responding clinical benefit, generally recommended
lesion becomes hyperdense [40]. C Insufficient evidence for efficacy or benefit does not
outweigh the risk or the disadvantages (adverse events,
costs, ...), optional
follow-up D Moderate evidence against efficacy or for adverse outcome,
There are no published data to indicate the optimal routine generally not recommended
E Strong evidence against efficacy or for adverse outcome,
follow-up policy of surgically treated patients with localised
never recommended
disease. Relapses most often occur to the liver and/or periton-
eum (other sites of metastases, including bone lesions and other a
sites, may be less rare along the course of metastatic disease
treated with several lines of therapy). The mitotic rate likely
iii | The ESMO/European Sarcoma Network Working Group Volume 25 | Supplement 3 | September 2014
by guest
on 05 February 2018
note • Daniel Vanel, Italy
• Eva Wardelmann, Germany
Levels of evidence and grades of recommendation have been
applied using the system shown in Table 1. Statements without acknowledgements
grading were considered justified standard clinical practice by
the panel members. We deeply thank Barbara Dore and Estelle Lecointe (from
SPAEN), and Hans Keulen (from Chordoma Foundation), who
observed the consensus conference as patient representatives.
consensus panel ESMO Guidelines 2014
These Clinical Practice Guidelines have been developed follow- conflict of interest
ing a consensus process based on a consensus event organised
Stefan Bielack declared: consultancy/advisory board/speakers
by ESMO in Milan, Italy, in December 2013 and refined by July
bureau from Merck, IDM/Takeda, Roche, Celgene, Bayer, and
2014. This involved experts from the community of the
Chugai. Sylvie Bonvalot declared: travel grants from Pharmamar,
European sarcoma research groups and ESMO faculty. Their
Nanobiotix, and honoraria from Novartis. Paolo G. Casali
names are indicated hereafter. The text reflects an overall con-
declared: consultancy/honoraria: Amgen Domplé, ARIAD, Bayer,
sensus among them, although each of them may not necessarily
GlaxoSmithKline, Infinity, Janssen Cilag, Merck Sharp & Dohme,
find it consistent with his/her own views. The panel worked on
Novartis, Pfizer, PharmaMar, Sanofi. Angelo Paolo Dei Tos
the text of ESMO Guidelines of previous years, whose author-
declared: speakers’ bureau: Novartis Oncology, Pfizer,
ship should also be credited.
GlaxoSmithKline, and PharmaMar. Mikael Eriksson declared:
• Paolo G. Casali, Italy (Moderator) honoraria from Novartis, Swedish Orphan Biovitrum,
• Jean-Yves Blay, France (Moderator) GlaxoSmithKline, Merck Sharp & Dohme, and Pfizer. Alexander
• Alexia Bertuzzi, Ireland Fedenko declared: speakers’ bureau: Johnson & Johnson,
• Stefan Bielack, Germany GlaxoSmithKline, Roche. Stefano Ferrari declared speakers’ hon-
• Bodil Bjerkehagen, Norway oraria from Takeda; advisory board for Amgen; research grants
• Sylvie Bonvalot, France from Mulmed, Amgen, and Morphotek. Hans Gelderblom
• Ioannis Boukovinas, Greece declared research grants from Novartis, Pfizer, PharmaMar,
• Paolo Bruzzi, Italy GlaxosmithKline, Eisai, and Bayer. Robert Grimer declared: re-
• Angelo Paolo Dei Tos, Italy search grant: Amgen. Alessandro Gronchi declared: advisory
• Palma Dileo, UK board: Novartis; honoraria: Novartis and Pfizer. Rolf Issels
• Mikael Eriksson, Sweden declared: consultancy/honoraria: PharmaMar, Bayer, and Therm
• Alexander Fedenko, Russian Federation Med. Heikki Joensuu declared: research funding to institute
• Andrea Ferrari, Italy from Novartis. Axel Le Cesne declared: honoraria: Novartis,
• Stefano Ferrari, Italy PharmaMar, GlaxoSmithKline, and Pfizer. Javier Martin-Broto
• Hans Gelderblom, Belgium declared advisory boards for GlaxoSmithKline, Novartis, and
• Robert Grimer, UK PharmaMar. Ofer Merimsky declared: speakers’ honoraria:
• Alessandro Gronchi, Italy GlaxoSmithKline and Lilly; advisory board: Boehringer
• Rick Haas, Netherlands Ingelheim, and Medison; research grant: Roche. Piero Picci
• Kirsten Sundby Hall, Norway declared advisory board for Takeda. Piotr Rutkowski declared
• Peter Hohenberger, Germany honoraria from Novartis, Pfizer, Bristol-Myers Squibb, Roche,
• Rolf Issels, Germany and GlaxoSmithKline; advisory board for Novartis,
• Heikki Joensuu, Finland GlaxoSmithKline, Merck, Sharp & Dohme, and Bayer. Marcus
• Ian Judson, UK Schlemmer declared: honoraria from Novartis, Pfizer, and Teva;
• Axel Le Cesne, France Research grants from Novartis. Silvia Stacchiotti declared: re-
• Saskia Litière, Belgium search grants: Novartis, Pfizer, PharmaMar, GlaxoSmithKline,
• Javier Martin-Broto, Spain Amgen, and Bayer. Frits Van Coevorden declared travel grants
• Ofer Merimsky, Israel from Novartis and PharmaMar. Winette Van der Graaf declared:
• Michael Montemurro, UK research funding from GlaxoSmithKline, Novartis and Pfizer.
• Carlo Morosi, Italy The folllowing authors have declared no potential conflicts of
• Piero Picci, Italy interest: Alexia Bertuzzi, Bodil Bjerkehagen, Ioannis Boukovinas,
• Isabelle Ray-Coquard, France Palma Dileo, Andrea Ferrari, Rick Haas, Kirsten Sundby Hall,
• Peter Reichardt, Germany Saskia Litière, Michael Montemurro, Carlo Morosi, Isabelle Ray-
• Piotr Rutkowski, Poland Coquard, Valter Torri, Daniel Vanel and Eva Wardelmann. The
• Marcus Schlemmer, Germany other authors have not reported any potential conflicts of interest.
• Silvia Stacchiotti, Italy
• Valter Torri, Italy
• Annalisa Trama, Italy references
• Frits Van Coevorden, Netherlands 1. Nilsson B, Bumming P, Meis-Kindblom JM et al. Gastrointestinal stromal tumors:
• Winette Van der Graaf, Netherlands the incidence, prevalence clinical course, and prognostication in the pre-imatinib
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu255 | iii

by guest
on 05 February 2018
mesylate era—a population based study in western Sweden. Cancer 2005; 103: 22. Bauer S, Rutkowski P, Hohenberger P et al. Long-term follow-up of patients with
821–829. GIST undergoing metastasectomy in the era of imatinib—analysis of prognostic
2. Pappo AS, Janeway KA. Pediatric gastrointestinal stromal tumors. Hematol Oncol factors (EORTC-STBSG collaborative study). Eur J Surg Oncol 2014; 40: 412–419.
Clin North Am 2009; 23: 15–34. 23. Blanke CD, Demetri GD, von Mehren M et al. Long-term results from a randomized
3. Zhang L, Smyrk TC, Young WF, Jr et al. Gastric stromal tumors in Carney triad are phase II trial of standard- versus higher-dose imatinib mesylate forpatients with
different clinically, pathologically, and behaviorally from sporadic gastric gastrointestinal unresectable or metastatic gastrointestinal stromal tumors expressing KIT. J Clin
stromal tumors: findings in 104 cases. Am J Surg Pathol 2010; 34: 53–64. Oncol 2008; 26: 620–625.
4. Pasini B, McWhinney SR, Bei T et al. Clinical and molecular genetics of patients 24. Blanke CD, Rankin C, Demetri GD et al. Phase III randomized, intergroup trial
with the Carney-Stratakis syndrome and germline mutations of the genes coding assessing imatinib mesylate at two dose levels in patients with unresectable or
for the succinate dehydrogenase subunits SDHB, SDHC, and SDHD. Eur J Hum metastatic gastrointestinal stromal tumors expressing the kit receptor tyrosine
Genet 2008; 16: 79–88. kinase: S0033. J Clin Oncol 2008; 26: 626–632.
5. Gaal J, Stratakis CA, Carney JA et al. SDHB immunohistochemistry: a useful tool 25. Verweij J, Casali PG, Zalcberg J et al. Progression-free survival in gastrointestinal
in the diagnosis of Carney-Stratakis and Carney triad gastrointestinal stromal stromal tumors with high-dose imatininb: randomized trial. Lancet 2004; 364:
tumors. Mod Pathol 2011; 24: 147–151. 1127–1134.
6. Miettinen M, Fetsch JF, Sobin LH, Lasota J. Gastrointestinal stromal tumors in 26. Zalcberg JR, Verveij J, Casali PG et al. Outcome of patients with advanced gastro-
patients with neurofibromatosis 1: a clinicopathologic and molecular genetic study intestinal stromal tumours crossing over to a daily imatinib dose of 800 mg after
of 45 cases. Am J Surg Pathol 2006; 30: 90–96. progression on 400 mg. Eur J Cancer 2005; 41: 1751–1757.
7. Rubin BP, Blanke CD, Demetri GD et al. Protocol for the examination of specimens 27. Gastrointestinal Stromal Tumor Meta-Analysis Group (MetaGIST). Comparison of two
from patients with gastrointestinal stromal tumor. Arch Pathol Lab Med 2010; doses of imatinib for the treatment of unresectable or metastatic gastrointestinal stromal
134: 165–170. tumors: a meta-analysis of 1640 patients. J Clin Oncol 2010; 28: 1247–1253.
8. Novelli M, Rossi S, Rodriguez-Justo M et al. DOG1 and CD117 are the antibodies 28. Le Cesne A, Ray-Coquard I, Bui BN et al. Discontinuation of imatinib in patients
of choice in the diagnosis of gastrointestinal stromal tumours. Histopathology with advanced gastrointestinal stromal tumors after 3 years of treatment: an open-
2010; 57: 259–270. label multicentre randomised phase 3 trial. Lancet Oncol 2010; 11: 942–949.
9. Miettinen M, Lasota J. Gastrointestinal stromal tumors: review on morphology, 29. Demetri GD, Wang Y, Wehrle E et al. Imatinib plasma levels are correlated with
molecular pathology, prognosis, and differential diagnosis. Arch Pathol Lab Med clinical benefit in patients with unresectable/metastatic gastrointestinal stromal
2006; 130: 1466–1478. tumors. J Clin Oncol 2009; 27: 3141–3147.
10. Miettinen M, Lasota J. Gastrointestinal stromal tumors: pathology and prognosis at 30. Raut CP, Posner M, Desai J et al. Surgical management of advanced
different sites. Semin Diagn Pathol 2006; 23: 70–83. gastrointestinal stromal tumors after treatment with targeted systemic therapy
11. Gold JS, Gönen M, Gutierrez A et al. Development and validation of a prognostic using kinase inhibitors. J Clin Oncol 2006; 24: 2325–2331.
nomogram for recurrence-free survival after complete surgical resection of 31. Wang D, Zhang Q, Blanke CD et al. Phase II trial of neoadjuvant/adjuvant imatinib
localised primary gastrointestinal stromal tumour: a retrospective analysis. Lancet mesylate for advanced primary and metastatic/recurrent operable gastrointestinal
Oncol 2009; 10: 1045–1052. stromal tumors: long-term follow-up results of Radiation Therapy Oncology Group
12. Joensuu H, Vehtari A, Riihimäki J et al. Risk of recurrence of gastrointestinal 0132. Ann Surg Oncol 2012; 19: 1074–1080.
stromal tumor after surgery: an analysis of pooled population-based cohorts. 32. Mussi C, Ronellenfitsch U, Jakob J et al. Post-imatinib surgery in advanced/
Lancet Oncol 2012; 13: 265–274. metastatic GIST: is it worthwhile in all patients? Ann Oncol 2010; 21: 403–408.
13. Ohtani H, Maeda K, Noda E et al. Meta-analysis of laparoscopic and open surgery 33. Du CY, Zhou Y, Song C et al. Is there a role of surgery in patients with recurrent or
for gastric gastrointestinal stromal tumor. Anticancer Res 2013; 33: 5031–5041. metastatic gastrointestinal stromal tumours responding to imatinib: a prospective
14. Joensuu H, Eriksson M, Sundby Hall K et al. One vs three years of adjuvant randomised trial in China. Eur J Cancer 2014; 50: 1772–1778.
imatinib for operable gastrointestinal stromal tumor: a randomized trial. JAMA 34. Demetri GD, van Oosterom AT, Garrett CR et al. Efficacy and safety of sunitinib in
2012; 307: 1265–1272. patients with advanced gastrointestinal stromal tumor after failure of imatinib: a
15. Dematteo RP, Ballman KV, Antonescu CR et al. Adjuvant imatinib mesylate after randomised controlled trial. Lancet 2006; 368: 1329–1338.
resection of localised, primary gastrointestinal stromal tumor: a randomised, 35. George S, Blay JY, Casali PG et al. Clinical evaluation of continuous daily dosing of
double-blind, placebo-controlled trial. Lancet 2009; 373: 1097–1104. sunitinib malate in patients with advanced gastrointestinal stromal tumor after
16. Gronchi A, Judson I, Nishida T et al. Adjuvant treatment of GIST with imatinib: solid imatinib failure. Eur J Cancer 2009; 45: 1959–1968.
ground or still quicksand? A comment on behalf of the EORTC Soft Tissue and Bone 36. Demetri GD, Reichardt P, Kang YK et al. on behalf of all GRID study investigators.
Sarcoma Group, the Italian Sarcoma Group, the NCRI Sarcoma Clinical Studies Efficacy and safety of regorafenib for advanced gastrointestinal stromal tumours
Group (UK), the Japanese Study Group on GIST, the French Sarcoma Group and the after failure of imatinib and sunitinib (GRID): an international, multicentre,
Spanish Sarcoma Group (GEIS). Eur J Cancer 2009; 45: 1103–1106. randomised, placebo-controlled, phase 3 trial. Lancet 2013; 381: 295–302.
17. Debiec-Rychter M, Sciot R, Le Cesne A et al. KIT mutations and dose selection for 37. Kang YK, Ryu MH, Yoo C et al. Resumption of imatinib to control metastatic or
imatinib in patients with advanced gastrointestinal stromal tumors. Eur J Cancer unresectable gastrointestinal stromal tumours after failure of imatinib and sunitinib
2006; 42: 1093–1103. (RIGHT): a randomised, placebo-controlled, phase 3 trial. Lancet Oncol 2013; 14:
18. Heinrich MC, Owzar K, Corless CL et al. Correlation of kinase genotype and clinical 1175–1182.
outcome in the North American Intergroup Phase III Trial of imatinib mesylate for 38. Choi H, Charnsangavej C, Faria SC et al. Correlation of computed tomography and
treatment of advanced gastrointestinal stromal tumor: CALGB 150105 Study by positron emission tomography in patients with metastatic gastrointestinal stromal
Cancer and Leukemia Group B and Southwest Oncology Group. J Clin Oncol tumor treated at a single institution with imatinib mesylate: proposal of new
2008; 26: 5360–5367. computed tomography response criteria. J Clin Oncol 2007; 25: 1753–1759.
19. Heinrich MC, Corless CL, Demetri GD et al. Kinase mutations and imatinib 39. Le Cesne A, Van Glabbeke M, Verweij J et al. Absence of progression as assessed
response in patients with metastatic gastrointestinal stromal tumor. J Clin Oncol by response evaluation criteria in solid tumors predicts survival in advanced GI
2003; 21: 4342–4349. stromal tumors treated with imatinib mesylate: the intergroup EORTC-ISG-AGITG
20. Hohenberger P, Ronellenfitsch U, Oladeji O et al. Pattern of recurrence in patients with phase III trial. J Clin Oncol 2009; 27: 3969–3974.
ruptured primary gastrointestinal stromal tumour. Br J Surg 2010; 97: 1854–1859. 40. Shankar S, van Sonnenberg E, Desai J et al. Gastrointestinal stromal tumor: new
21. Eisenberg BL, Harris J, Blanke CD et al. Phase II trial of neoadjuvant/adjuvant nodule-within-a-mass pattern of recurrence after partial response to imatinib
imatinib mesylate (IM) for advanced primary and metastatic/recurrent operable mesylate. Radiology 2005; 235: 892–898.
gastrointestinal stromal tumor (GIST): early results of RTOG 0132/ACRIN 6665. 41. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections among
J Surg Oncol 2009; 99: 42–47. hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33: 139–144.
iii | The ESMO/European Sarcoma Network Working Group Volume 25 | Supplement 3 | September 2014
by guest
on 05 February 2018
Soft tissue and visceral sarcomas: ESMO Clinical

and follow-up†
Extraskeletal Ewing sarcoma is covered by other ESMO tumour boards discussing new cases), volume of patients, avail-
Guidelines: in general, the same principles for these tumours in ability of facilities needed to properly apply clinical practice guide-
children apply to adults. This is also the case for embryonal and lines, recording and publication of outcomes.
alveolar rhabdomyosarcoma, which are exceedingly rare in In soft tissue tumours, magnetic resonance imaging (MRI) is
adults. On the other hand, pleomorphic rhabdomyosarcoma is the main imaging modality in the extremities, pelvis and trunk.
viewed as a high-grade adult-type soft tissue sarcoma. Standard radiographs may be useful to rule out a bone tumour,
Gastrointestinal stromal tumours are covered by the dedicated to detect bone erosion with a risk of fracture and to show calcifi-
ESMO Clinical Practice Guidelines. Kaposi’s sarcoma is excluded. cations. Computed tomography (CT) has a role in calcified
lesions to rule out a myositis ossificans, and in retroperitoneal
tumours, where the performance is identical to MRI. Ultrasound
incidence
may be the first exam, but it should be followed by CT or MRI.
Adult soft tissue and visceral sarcomas (excluding gastrointes- Following appropriate imaging assessment, the standard ap-
clinical practice
tinal stromal tumour) are rare tumours, with an estimated inci- proach to diagnosis consists of multiple core needle biopsies, pos-
guidelines
dence averaging 4–5/100 000/year in Europe [1]. sibly by using ≥14–16 G needles. However, an excisional biopsy
may be the most practical option for <3 cm superficial lesions. An
diagnosis open biopsy may be another option in selected cases, as decided
within reference centres. An immediate evaluation of tissue viabil-
Soft tissue sarcomas (STSs) are ubiquitous in their site of origin ity may be considered, to ensure that the biopsy is adequate at the
and are often managed with multimodality treatment. A multi- time it is carried out. However, a frozen-section technique for im-
disciplinary approach is therefore mandatory in all cases mediate diagnosis is not encouraged, because it generally does not
(involving pathologists, radiologists, surgeons, radiation thera- allow a complete diagnosis, particularly when preoperative treat-
pists, medical oncologists and paediatric oncologists, as well as ment is planned. Fine needle aspiration is used only in some insti-
nuclear medicine specialists, organ-based specialists, as applic- tutions, which have developed specific expertise on this procedure,
able). Management should be carried out in reference centres for and is not recommended outside these centres. A biopsy may
sarcomas and/or within reference networks sharing multidiscip- underestimate the tumour malignancy grade. Therefore, when pre-
linary expertise and treating a high number of patients annually. operative treatment is an option, radiological imaging (including
These centres are involved in ongoing clinical trials, in which positron emission tomography, PET) may be useful, in addition to
sarcoma patients’ enrolment is common. This centralised referral pathology, in providing the clinician with information that helps to
should be pursued as early as at the time of the clinical diagnosis estimate the malignancy grade (i.e. necrosis). The biopsy should be
of a suspected sarcoma. In practice, referral of all patients with a carried out by a surgeon or a radiologist, after multidisciplinary
lesion likely to be a sarcoma would be recommended. This would discussion, as needed, within reference centres. It should be
mean referring all patients with an unexplained deep mass of soft planned in such a way that the biopsy pathway and the scar can be
tissues, or with a superficial lesion of soft tissues having a diam- safely removed by definitive surgery (except for retroperitoneal sar-
eter of >5 cm. Quality criteria are needed for sarcoma reference comas, RPS). The biopsy entrance point can be tattooed. The
centres and, all the more, reference networks. These criteria may tumour sample should be fixed in 4% buffered formalin in due
vary from country to country but, among others, should be based time (Bouin fixation should not be used, since it prevents molecu-
on: multidisciplinarity (incorporating tools such as weekly lar analysis). The collection of fresh/frozen tissue and tumour
imprints (touch preps) is encouraged, because new molecular path-
*Correspondence to: ESMO Guidelines Working Group, ESMO Head Office, Via L. Taddei ology assessments could be made at a later stage in the patient’s
4, CH-6962 Viganello-Lugano, Switzerland. interest. In this perspective, the availability of a blood sample could
add to the value of tumour tissues. Informed consent for biobank-
†
Approved by the ESMO Guidelines Working Group: August 2003, last update July ing should be sought, enabling later analyses and research, as long
2014. This publication supersedes the previously published version—Ann Oncol 2012; as this is allowed by local and international rules.
23(Suppl 7): vii92–vii99.

by guest
on 05 February 2018
Histological diagnosis should be made according to the 2013 External quality assurance programmes are encouraged for
World Health Organization (WHO) classification [2]. A patho- laboratories performing molecular pathology assessments.
logical expert validation is required in all cases when the original
diagnosis was made outside a reference centre/network [3]. stage classification and risk assessment
The malignancy grade should be provided in all cases in
which this is feasible based on available systems, because it has Available staging classifications have limited relevance and
prognostic and predictive meaning. The ‘Federation Nationale should be improved. The American Joint Committee on Cancer
des Centres de Lutte Contre le Cancer’ (FNCLCC) grading (AJCC)/International Union against Cancer (UICC) stage clas-
system is generally used, which distinguishes three malignancy sification system stresses the importance of the malignancy
grades based on differentiation, necrosis and mitotic rate [4]. grade in sarcoma [5]. In general, in addition to grading, other
Whenever possible, the mitotic rate should be provided inde- prognostic factors are tumour size and tumour depth for limb
pendently. An effort should be made to improve the reliability of sarcomas. Of course, site, tumour resectability and presence of
mitotic count as actually recorded. Grading cannot be assigned metastases are also important (Table 2).
after preoperative medical treatment, by which the tumour
tissue undergoes major therapy-related changes (Table 1). staging procedures
Tumour site should be properly recorded. Tumour size and
tumour depth (in relation to the superficial fascia) should also be A chest spiral CT scan is mandatory for staging purposes.
recorded, since they entail a prognostic value, along with the ma- Regional lymph node metastases are rare, with the exception
lignancy grade. The pathology report after definitive surgery of some histologies, e.g. epithelioid sarcoma and clear cell sarcoma,
should mention whether the tumour was intact and should for which regional assessment through CT/MRI may be added
include an appropriate description of tumour margins (i.e. the to the usual staging procedures.
status of inked margins and the distance between tumour edge Likewise, an abdominal CT scan may be added for limb
and the closest inked margins). This allows the assessment of myxoid liposarcoma. The brain CT scan may be added for al-
margin status (i.e. whether the minimum margin is intralesional, veolar soft part sarcoma, clear cell sarcoma and angiosarcoma.
marginal or wide and distances from surrounding tissues). The Bone scan, whole-body MRI and PET scan are optional.
pathological assessment of margins should be made in collabor- Cost-effectiveness studies on their incorporation into the staging
ation with the surgeon. procedures are required.
If preoperative treatment was carried out, the pathology The surgical report, or patient chart, should provide details on:
report should include an assessment of the histological response preoperative and intraoperative diagnosis; the surgical conduct, in-
of the tumour. In contrast to osteosarcoma and Ewing sarcoma, cluding possible contaminations (i.e. it should mention whether
however, no validated system is available at present in this regard, the tumour was opened and was ‘seen’ during the excision, etc.);
and no percentage of residual ‘viable cells’ is considered to have a surgical actual completeness vis-a-vis planned quality of margins.
specific prognostic significance. This depends on several factors,
including the presence of non-treatment-related necrosis and treatment
haemorrhage and the heterogeneity of post-treatment changes. A
multidisciplinary judgement is recommended, involving the path- localised disease
ologist and the radiologist. Surgery is the standard treatment of all patients with an adult
Pathological diagnosis relies on morphology and immunohis- type, localised STS. It must be carried out by a surgeon specific-
tochemistry. It should be complemented by molecular pathology, ally trained in the treatment of this disease [III, A]. The standard
especially when: surgical procedure is a wide excision with negative margins
(R0). This implies removing the tumour with a rim of normal
• the specific histological diagnosis is doubtful; tissue around it [III, A]. The cut-off of the minimal margin on
• the clinical pathological presentation is unusual; fixed tissue to be considered adequate may depend on several
• it may have prognostic and/or predictive relevance. factors, including histological subtype, preoperative therapies
and the presence of resistant anatomical barriers, such as mus-
cular fasciae, periostium and epineurium. As an individualised
Table 1. Federation Nationale des Centres de Lutte Contre le option, marginal excision can be acceptable in carefully selected
Cancer histological grading criteria cases, in particular for extracompartmental atypical lipomatous
Tumour Necrosis (macro Mitotic count (n/10 tumours [IV, B].
differentiation and micro) high-power fields) Radiation therapy is not given in the case of a truly compart-
mental resection of a tumour entirely contained within the com-
1: Well 0: Absent 1: n < 10 partment [IV, A]. A wide excision is followed by radiation
2: Moderate 1: <50% 2: 10–19 therapy as the standard treatment of high-grade (G2–3), deep
3: Poor 2: ≥50% 3: n ≥ 20
>5 cm lesions [II, B] [6–8]. Exceptions may be made after multi-
disciplinary discussion taking into account several variables [9].
The sum of the scores of the three criteria determines the grade of
With exceptions to be discussed in a multidisciplinary setting,
malignancy. Grade 1: 2 and 3; Grade 2: 4 and 5; Grade 3: 6, 7 and 8.
and faced with a lack of consensus across reference centres,
Reprinted from [4] with permission of John Wiley & Sons, Inc.
high-grade, deep, <5 cm lesions are also treated with surgery,
followed by radiation therapy [IV, A].
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu254 | iii

by guest
on 05 February 2018
Table 2. American Joint Committee on Cancer (AJCC)/International Union against Cancer (UICC) TNM staging system [5]
Primary tumour (T)

T1 Tumour 5 cm or less in greatest dimensiona
T1a Superficial tumour
T1b Deep tumour
T2 Tumour >5 cm in greatest dimensiona
T2a Superficial tumour
T2b Deep tumour
N1b Regional lymph node metastasis
Stage IA
T1a N0 M0 G1, GX
T1b N0 M0 G1, GX
Stage IB
T2a N0 M0 G1, GX
T2b N0 M0 G1, GX
Stage IIA
T1a N0 M0 G2, G3
T1b N0 M0 G2, G3
Stage IIB
T2a N0 M0 G2
T2b N0 M0 G2
Stage III
T2a, T2b N0 M0 G3
Any T N1 M0 Any G
Stage IV
Any T Any N M1 Any G
Used with the permission of the American Joint Committee on Cancer (AJCC), Chicago, Illinois. The original source for this material is the AJCC
Cancer Staging Handbook, Seventh Edition (2010) published by Springer Science and Business Media LLC, www.springerlink.com.
a
Superficial tumour is located exclusively above the superficial fascia without invasion of the fascia; deep tumour is located either exclusively beneath
the superficial fascia, superficial to the fascia with invasion of or through the fascia, or both superficial yet beneath the fascia.
b
Presence of positive nodes (N1) in M0 tumours is considered Stage III.
Radiation therapy is added in selected cases in the case in combination with chemotherapy [10] to a total dose of 50 Gy
of low- or high-grade, superficial, >5 cm and low-grade, deep, in 1.8–2 Gy fractions, followed by surgery may be considered. In
<5 cm STSs [II, B]. addition, by means of modern radiotherapy techniques such as
In the case of low-grade, deep, >5 cm STSs, radiation therapy image guided radiotherapy and intensity modulated radiother-
should be discussed in a multidisciplinary fashion, considering apy the anticipated incidence of wound complications after pre-
the anatomical site and the related expected sequelae versus the operative radiotherapy is lower than historically published. The
histological aggressiveness. main advantage of preoperative radiotherapy is that, with pro-
Local control and survival are not influenced by the timing of longed follow up, late morbidity (fibrosis, bone fracture, etc.) is
radiotherapy, but early and late complications are. If it is antici- lower, translating into improved functional outcome and quality
pated that wound complications will be severe, surgery followed of life.
by adjuvant radiotherapy may be the best option. Radiation Reoperation in reference centres must be considered in the case
therapy should then be administered, with the best technique of R1 resections, if adequate margins can be achieved without
available, to a total dose of 50 Gy in 1.8–2 Gy fractions, possibly major morbidity, taking into account tumour extent and tumour
with a boost up to 66 Gy, depending on presentation and resec- biology (e.g. re-excision can be spared in extracompartmental
tion margins. If it is anticipated that wound complications will atypical lipomatous tumours, etc.) [IV, A]. In the case of R2
be a manageable problem, neoadjuvant radiotherapy, possibly surgery, reoperation in reference centres is mandatory, possibly
iii | The ESMO/European Sarcoma Network Working Group Volume 25 | Supplement 3 | September 2014
by guest
on 05 February 2018
following preoperative treatments if adequate margins cannot be clinical trials. A randomised trial showed no differences between
achieved, or surgery is mutilating. In the latter case, the use of three (preoperative) and five (pre- and postoperative) courses of
multimodal therapy with less radical surgery requires a shared de- full-dose chemotherapy [16]. A trial is ongoing comparing stand-
cision-making with the patient in cases of uncertainty. Plastic ard preoperative chemotherapy versus histology-driven chemo-
repairs and vascular grafting should be used as needed, and the therapy. Adjuvant chemotherapy should never be intended to
patient should be properly referred as necessary. rescue inadequate surgery. In any case, adjuvant chemotherapy is
Radiation therapy will follow marginal or R1–R2 excisions, if not used in histological subtypes known to be insensitive to
these cannot be rescued through re-excision, tailoring the deci- chemotherapy. If the decision is made to use chemotherapy as
sion depending on further considerations, including impact on upfront treatment, it may well be used preoperatively, at least in
future surgeries, etc. part [III, B]. A local benefit may be gained, facilitating surgery.
Mutilating surgery may be of choice in some cases. Options When employed, adjuvant chemotherapy should consist of the
for limb-preserving surgery can be discussed with the patient, combination chemotherapy regimens proved to be most active in
including chemotherapy and/or radiotherapy [III, A], or iso- advanced disease. Radiation therapy should not delay the start of
lated hyperthermic limb perfusion with tumour necrosis factor- chemotherapy and can be used preoperatively. In one large rando-
alpha + melphalan [III, A], if the tumour is confined to an mised phase III study (in patients with G2–3, deep, >5 cm STSs),
extremity, or regional hyperthermia combined with chemother- regional hyperthermia in addition to systemic chemotherapy was
apy [I, B] [11]. These options are resorted to in non-resectable associated with a local progression-free survival (PFS) and DFS
tumours as well, i.e. in the truly locally advanced disease. advantage [I, B] [11].
Regional lymph node metastases should be distinguished The standard approach to local relapses parallels the approach
from soft tissue metastases involving lymph nodes. They are to primary local disease, except for a wider resort to preoperative
rare and constitute an adverse prognostic factor in adult-type or postoperative radiation therapy and/or chemotherapy, if not
STSs. More aggressive treatment planning is therefore felt to be previously carried out.
appropriate for these patients, although there is a lack of formal
evidence to indicate that this improves clinical results. Surgery
through wide excision (mutilating surgery is exceptionally done
given the prognosis of these patients) may be coupled with adju- advanced disease
vant radiation therapy and adjuvant chemotherapy for sensitive The decision-making is complex, depending on diverse presenta-
histological types, as the standard treatment of these presentations and histologies, and should always be multidisciplinary.
tions [IV, B]. Chemotherapy may be administered as preopera- Metachronous (disease-free interval ≥1 year) resectable lung me-
tive treatment, at least in part. Given the paucity of published tastases without extrapulmonary disease are managed with
data on adjuvant radiotherapy after lymph node dissections in surgery as standard treatment, if complete excision of all lesions is
regional metastatic STS, the indication should probably be feasible [17] [IV, B]. A minimally invasive thoracoscopic approach
reserved for patients with a relatively large number of tumour- can be resorted to in selected cases. Other appropriate local tech-
positive lymph nodes and/or extranodal spread in the absence niques can be resorted to, though surgery is the standard and data
of haematogenic metastases. The increase in local control are required on alternative less invasive options. Decisions must
should be balanced against toxicity (especially peripheral lym- also consider the feasibility of the various options. When surgery
phoedema). These treatment modalities adding to surgery of lung metastases is selected, an abdominal CT scan and a bone
should not be viewed as truly ‘adjuvant’, the context being in scan or a fluorodeoxyglucose (FDG)-PET are mandatory to
fact that of a likely systemic disease. In one large randomised confirm that lung metastases are ‘isolated’.
phase III study (in patients with G2–3, deep, >5 cm STSs), re- Chemotherapy may be added to surgery as an option, taking
gional hyperthermia in addition to systemic chemotherapy was into account the prognostic factors (a short previous recurrence-
associated with a local and disease-free survival (DFS) advan- free interval and a high number of lesions are adverse factors,
tage when compared with chemotherapy alone [I, B]. Isolated encouraging the addition of chemotherapy), although there is a
limb perfusion may be an option in this patient population. In lack of formal evidence that this improves outcome [IV, B].
itself, this modality has obviously no impact on systemic control Chemotherapy is preferably given before surgery in order to
(but it can be combined with other modalities) [III, A] [12]. assess tumour response and thus modulate treatment.
There is no consensus on the current role of adjuvant chemo- In cases where lung metastases are synchronous, in the
therapy. Study results are conflicting, in the presence of negative absence of extrapulmonary disease, standard treatment is
results from the largest studies, though data are available from chemotherapy [III, B]. Surgery of completely resectable residual
smaller studies suggesting that it might improve, or at least lung metastases may be offered as an option, especially when a
delay, distant and local recurrence in high-risk patients [13, 14]. tumour response is achieved.
A meta-analysis found a statistically significant limited benefit Extrapulmonary metastatic disease is treated with chemother-
in terms of both survival- and relapse-free survival [15]. It is apy as the standard treatment [I, A].
unknown whether adjuvant chemotherapy may be particularly In highly selected cases, surgery of responding metastases
beneficial in specific subgroups or even detrimental in others. may be offered as an option following a multidisciplinary evalu-
Therefore, adjuvant chemotherapy is not standard treatment in ation, taking into consideration their site and the natural history
adult-type STS. It can be proposed as an option to the high-risk of the disease in the individual patient.
individual patient (high-grade, deep, >5 cm tumour) for a Surgery, or ablations, or radiation therapy, of extrapulmonary
shared decision-making with the patient [II, C] or within metastases may be an option without chemotherapy in highly
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu254 | iii

by guest
on 05 February 2018
selected cases (e.g. some patients with myxoid liposarcoma, soli- Radiation therapy should be used as a palliative resource in all
tary fibrous tumour, etc.) [7]. cases as appropriate to the clinical need (e.g. bone lesions at risk
Standard chemotherapy is based on anthracyclines as the of fracture, etc.).
first-line treatment [I, A]. As of today, there is no formal dem- In general, advanced previously treated patients are candi-
onstration that multiagent chemotherapy is superior to single- dates for clinical trials.
agent chemotherapy with doxorubicin alone in terms of overall With reference to selected histological types, there is anec-
survival (OS). However, a higher response rate can be expected, dotal evidence of activity of several molecular targeted agents,
in particular in a number of sensitive histological types, accor- building on consistent preclinical data. Examples are:
ding to several, although not all, randomised clinical trials
[18, 19]. Therefore, multiagent chemotherapy with adequate- • mammalian target of rapamycin inhibitors in malignant peri-
dose anthracyclines plus ifosfamide may be the treatment of vascular epithelioid cell tumours (PEComas), which are often
choice, particularly when a tumour response is felt to be poten- associated with the loss of tuberous sclerosis complex 1 (TSC1)/
tially advantageous and patient performance status is good. TSC2 [32];
In angiosarcoma, taxanes are an alternative option, given their • crizotinib in inflammatory myofibroblastic tumour associated
high antitumour activity in this specific histological type [20] [III, with anaplastic lymphoma kinase translocations [33];
B]. An alternative option is gemcitabine ± docetaxel [21] [V, B]. • sunitinib and cediranib in alveolar soft part sarcoma, where
Doxorubicin plus dacarbazine is an option for multiagent the molecular target is as yet unclear [34, 35]
first-line chemotherapy of leiomyosarcoma, where the activity of • sunitinib in solitary fibrous tumours, where the molecular
ifosfamide is far less convincing in available retrospective evi- target is as yet unclear [36].
dence, or solitary fibrous tumour [22] [V, B]. These patients can be sent to reference centres, to be treated
Imatinib is standard medical therapy for those rare patients accordingly, preferably within clinical studies or prospective
with dermatofibrosarcoma protuberans who are not amenable clinical recordings [III, C].
to non-mutilating surgery or with metastases deserving medical
therapy [23, 24] [III, A].
After failure of anthracycline-based chemotherapy, or the im- follow-up
possibility to use it, the following criteria may apply, although
There are few published data to indicate the optimal routine
high-level evidence is lacking:
follow-up policy of surgically treated patients with localised
• Patients who have already received chemotherapy may be disease [37].
treated with ifosfamide, if they did not progress on it previ- The malignancy grade affects the likelihood and speed at
ously. High-dose ifosfamide (around 14 g/m2) may be an which relapses may occur. The risk assessment based on tumour
option also for patients who have already received standard- grade, tumour size and tumour site therefore helps in choosing a
dose ifosfamide [25, 26] [IV, C]. routine follow-up policy. High-risk patients generally relapse
• Trabectedin is a second-line option [II, B] and is approved for within 2–3 years, whereas low-risk patients may relapse later, al-
advanced previously treated STS in the EU. It has proved ef- though it is less likely. Relapses most often occur to the lungs.
fective in leiomyosarcoma and liposarcoma [27]. In myxoid Early detection of local or metastatic recurrence to the lungs may
liposarcoma, a high antitumour activity was described. A pe- have prognostic implications, and lung metastases are asymptom-
culiar pattern of tumour response has been reported, with an atic at a stage in which they are suitable for surgery. Therefore,
early phase of tissue changes preceding tumour shrinkage the routine follow-up may focus on these sites. Although the use
[28]. Clinical benefit with trabectedin was also obtained in of MRI to detect local relapse and CT to scan for lung metastases
other histological types. is likely to pick up recurrences earlier, it has not been demon-
• One trial showed that gemcitabine + docetaxel is more effective strated that this is beneficial, or cost effective, compared with the
than gemcitabine alone as second-line chemotherapy, with clinical assessment of the primary site and regular chest X-rays.
special reference to leiomyosarcoma and undifferentiated pleo- That said, while prospective studies are needed, a practical ap-
morphic sarcoma, but data are conflicting and toxicity is different proach in place at several institutions is as follows: Surgically-
[29] [II, C]. Gemcitabine was shown to have anti-tumour activity treated intermediate-/high-grade patient may be followed every 3–
in leiomyosarcoma and angiosarcoma also as a single agent. 4 months in the first 2–3 years, then twice a year up to the fifth
• Dacarbazine has some activity as a second-line therapy (mostly year and once a year thereafter; low-grade sarcoma patients may be
in leiomyosarcoma and solitary fibrous tumour). The combin- followed for local relapse every 4–6 months, with chest X-rays or
ation of dacarbazine and gemcitabine was shown to improve the CT scan at longer intervals in the first 3–5 years, then annually.
OS and PFS over dacarbazine in a randomised trial [30] [II, B].
• A randomised trial showed a benefit in PFS averaging 3 months
for pazopanib given up to progression to advanced, previously
special presentations and entities
treated, STS patients (excluding liposarcomas) [31]. Thus, it is retroperitoneal sarcomas
an option in non-adipogenic STS [I, B]. Patients with suspected RPS need to be referred to high-volume
sarcoma centres [38].
Best supportive care alone is an alternative for pre-treated Chest, abdomen and pelvis IV contrast-enhanced CT are
patients with advanced STS, especially if further-line therapies standard for staging. IV contrast-enhanced MRI is an option, es-
have already been used in the patient. pecially for pelvic tumours, to assess specific aspects of tumour
iii | The ESMO/European Sarcoma Network Working Group Volume 25 | Supplement 3 | September 2014
by guest
on 05 February 2018
extent. Functional assessment of contralateral kidney is necessary. significant short- and long-term toxicities. A therapeutic radiation
Pre-treatment biopsy for pathological diagnosis should be carried treatment dose can be achieved in a minority of patients following
out, to allow tailored present and future therapeutic decisions, resection. In selected cases, it may be an option in well-defined ana-
unless otherwise indicated by a sarcoma tumour board. A mul- tomical areas felt to be at high risk. Brachytherapy is of unproven
tiple core biopsy with an adequate coaxial needle of sufficient size value and is associated with significant short- and long-term com-
(14–16 G) is the standard procedure. Risk of needle track seeding plications. Intraoperative radiotherapy is of unproven value.
is minimal and should not be a reason to avoid a biopsy. The value of adjuvant chemotherapy is not established.
Nonetheless, the pathway of the biopsy should be carefully However, one can make the same considerations which apply to
planned to minimise contamination and complications, and extremity STSs.
should not be carried out transperitoneally. Open or laparo- Surgery of local recurrences could be offered on an individua-
scopic biopsies must be avoided. lised basis, especially to patients affected by well-differentiated
Comprehensive imaging evaluation is critical, to accurately liposarcoma and having a long disease-free interval between
assess extent of tumour. Certain areas (e.g. inguinal canal, initial resection and subsequent recurrence, or to patients ex-
retrohepatic vena cava, diaphragm, neural foramina, etc.) are periencing a response to medical therapies.
particularly challenging to evaluate and may require additional
specialised radiological input. Specific appreciation of the well-
differentiated versus the dedifferentiated component(s) of lipo- uterine sarcomas
sarcoma is critical to surgical decision-making. The group of uterine sarcomas includes leiomyosarcomas, endo-
The best chance of cure is at the time of primary presentation, metrial stromal sarcomas (ESS, formerly low-grade ESSs) and un-
and an individualised management plan should be made, fol- differentiated endometrial sarcomas (UESs). Carcinosarcomas
lowing discussion at a multidisciplinary sarcoma case confer- (malignant Müllerian mixed tumours) are currently viewed as
ence on both imaging and pathological findings. The standard epithelial cancers, and treatment should be tailored accordingly.
treatment of primary lesions is surgery, to be carried out by a Thus, before a final diagnosis of sarcoma is made, the pathologist
surgeon with specific sarcoma expertise. Surgery should be should be certain that an epithelial component is absent, through
aimed at achieving macroscopically complete resection in one proper immunohistochemical analysis.
specimen bloc and minimising microscopically positive At the moment, we do not have clinical and radiological cri-
margins. This is best done by resecting the tumour en bloc with teria to differentiate leiomyomas from malignant uterine
adherent structures, even if not overtly infiltrated (III, A) [39–42]. tumours. Thus, procedures resulting in potential tumour cell
Preservation of specific organs (i.e. kidney, head of the pancreas spillage, such as morcellation out of endobags, entail a high risk
and/or liver) should be considered on an individualised basis and of worsening patient prognosis if malignancy is the post-
mandates a specific expertise in the disease to make the right deci- operative pathological diagnosis [45, 46].
sions. Judgement must be used in deciding which neurovascular Smooth tumours of undefined malignant potential constitute
structures to sacrifice, weighing the potential for local control a negative definition, which is resorted to when both a leio-
against the potential for long-term dysfunction. myoma and a leiomyosarcoma cannot be diagnosed with cer-
Grossly incomplete resection of RPSs is of questionable tainty [47]. There are remarkable variations with this diagnosis
benefit and potentially harmful, and can only be regarded as po- amongst pathologists. Some of these lesions might represent
tentially palliative in carefully selected patients. Grossly incom- ‘low-grade’ leiomyosarcomas. Due to the uncertainty, hysterec-
plete resection is to be avoided by imaging review, thoughtful tomy is an option, but there is room for individualised decision-
planning and referral to appropriate centres. making with an informed patient.
Although no randomised trials of neoadjuvant therapy versus Standard local treatment of uterine leiomyosarcoma, ESS and
resection alone for RPS have been reported to date, neoadjuvant UES (when localised) is en bloc total hysterectomy (including
therapy, in the form of chemotherapy, external beam radiation, re- laparoscopy/assisted or robotic surgery, provided the tumour is
gional hyperthermia or combinations, is safe in well-selected resected with the same criteria as for open surgery). With a
patients and may be considered after careful review by a multidis- diagnosis of sarcoma, fertility-preserving surgery in young
ciplinary sarcoma tumour board [43]. This is particularly relevant women is not supported by any evidence and should not be
in the case of technically unresectable/borderline resectable RPS regarded as standard, though of course it may be the choice
that could potentially be rendered resectable by downsizing, and made by an informed patient. The added value of bilateral
also in chemosensitive histologies such as synovial sarcoma. The salpingo-oophorectomy is not established, particularly in pre-
sensitivity of solitary fibrous tumour to radiation therapy should menopausal women, and lymphadenectomy has not been
also be considered. In one large randomised phase III study (in demonstrated to be useful in the lack of macroscopic involve-
patients with G2–3, deep, >5 cm STSs), regional hyperthermia in ment. In ESS, however, lymph nodes may be positive in roughly
addition to systemic chemotherapy was associated with a local 10% of cases. Although in uterine leiomyosarcoma retrospective
PFS and DFS advantage [44] [I, B]. studies suggested a possible decrease in local relapses, radiation
Preoperative radiation therapy in resectable tumours is being therapy has not improved survival and relapse-free survival in a
investigated in a currently accruing prospective randomised clinic- prospective randomised trial, and therefore is not recommended
al trial. Preoperative treatments are not intended to change the [48]. The use of radiation therapy as an adjuvant to surgery can
extent of surgery, but to improve the quality of surgical margins. be an option in selected cases, after shared decision-making
Postoperative/adjuvant external beam radiation following com- with the patient, following multidisciplinary discussion taking
plete gross resection is of limited value, and is associated with into account special risk factors, including: local relapse, cervical
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu254 | iii

by guest
on 05 February 2018
involvement, parametral involvement, serosal involvement and desmoid-type fibromatosis

UES histology [IV, C]. The value of adjuvant chemotherapy in While principles for the diagnosis of STS apply also to des-
uterine leiomyosarcoma is undetermined. Uncontrolled studies moids, beta-catenin mutational analysis may be useful when the
suggest a benefit in comparison with external controls for pathological differential diagnosis is difficult.
gemcitabine + docetaxel × 4 courses followed by doxorubicin × 4 Given the unpredictable natural history of the disease (with
courses, as well as for gemcitabine + docetaxel × 4 courses [49, the possibility of long-lasting stable disease and even occasional
50]. A prospective randomised trial with a no-treatment control spontaneous regressions, along with a lack of metastatic poten-
arm versus gemcitabine + docetaxel × 4 courses followed by tial) and functional problems implied by some tumour anatom-
doxorubicin × 4 courses is ongoing. ical locations, an initial watchful waiting policy can be proposed
The medical treatment of advanced leiomyosarcomas, UES [54, 55] [III, B]. This should follow a shared decision-making
and adenosarcoma with sarcomatous overgrowth parallels that with the patient, with careful monitoring of potentially life-
for adult-type STSs. It should be kept distinct from malignant threatening extra-abdominal locations (e.g. head and neck
Müllerian mixed tumours, which are currently treated with region) and intra-abdominal desmoids (mesenteric fibromato-
therapies for epithelial tumours. As for all leiomyosarcomas, sis). Under such a policy, treatment is reserved for progressing
doxorubicin, dacarbazine, trabectedin and pazopanib are active cases. The preferred imaging modality is MRI, taking into con-
agents, and may be used in a stepwise fashion. There is retro- sideration that the tumour signal is not meaningful with regard
spective evidence that ifosfamide may be less active as a single to the disease evolution.
agent in leiomyosarcomas. For progressing cases, optimal strategy needs to be individua-
ESSs are low-grade tumours, with a consistent pathological ap- lised on a multidisciplinary basis and may consist of surgery
pearance. The diagnosis is supported by typical cytogenetics, (without any adjuvant therapy), radiation therapy, observation,
marked by a chromosomal translocation (7;17), with JAZF1- isolated limb perfusion (if the lesion is confined to an extremity)
SUZ12 or the rare EPC1-PHF1 or JAZF1-PHF1 transcripts. or systemic therapy (see below) [56] [V, B]. Systemic therapies
Adjuvant hormonal therapy is not standard, though it may be an include: hormonal therapies (tamoxifen, toremifene and Gn-RH
option, given retrospective evidence suggesting its role in decreas- analogues), non-steroidal anti-inflammatory drugs; low-dose
ing relapses. However, the sensitivity of the advanced disease to chemotherapy, such as methotrexate + vinblastine or methotrexate
hormones makes the benefit questionable overall. The systemic + vinorelbine; sorafenib; imatinib; interferon; full-dose chemother-
treatment of metastatic low-grade ESS exploits their sensitivity to apy (using regimens active in sarcomas, including liposomal doxo-
hormonal therapies [V, B]. Therefore, progestins, aromatase inhi- rubicin) [57–63]. It is reasonable to employ the less toxic therapies
bitors and Gn-RH analogues (for pre-menopausal patients) can be before the more toxic ones in a stepwise fashion. A comprehen-
used [51]. Tamoxifen is contraindicated due to a possible agonist sive clinical judgement of progression should be used. Hormonal
activity, as is hormonal replacement therapy (HRT) containing contraception should be discussed with the patient, and definitely
oestrogens. Chemotherapy may be an option when hormonal stopped in case of progressing disease.
therapy has failed. Surgery of lung metastases is an option, even in
presentations which might not be surgically approached in other
STS, given the long natural history of the disease. This may apply breast sarcomas
to pelvic disease as well, even in the presence of metastatic disease. These patients should be referred to sarcoma units.
Currently, a subgroup of high-grade ESS is recognised, which is Breast sarcomas encompass radiation- and non-radiation-
defined by specific cytogenetics, marked by the (10;17) transloca- induced sarcomas. Therefore, sarcomas of the skin of the breast
tion, carrying the YWHAE-FAM22 transcript [52]. Their behav- area should be conceptually distinguished from mammary
iour is more aggressive. Currently, their sensitivity to hormonal gland sarcomas. Angiosarcoma has a more aggressive behaviour
therapies is not defined, so cytotoxic chemotherapy is considered than other histological types, while malignant phyllodes
appropriate in the metastatic setting, though data are lacking. tumours (i.e. those having >10 mitoses/10 HPF and marked
High-grade ESS, adenosarcoma with sarcomatous overgrowth stromal overgrowth) have a 20%–30% metastatic rate. On the
and UES are high-grade malignancies. There are no data on the other hand, metaplastic breast carcinomas, also known as carci-
value of adjuvant chemotherapy, though their high-risk status nosarcomas, are epithelial neoplasms, whose treatment should
may justify a shared decision with the patient in conditions be tailored to their mainly epithelial nature.
of uncertainty, especially in UES. Hyperthermic peritoneal The best treatment of breast sarcomas is far from being
chemotherapy has not been shown to be effective and is an defined, given their rarity and heterogeneity. In general, breast-
experimental-only option. conserving surgery may be resorted to, depending on the quality
For benign metastasising leiomyomas, clinical observation is of margins versus the size of the tumour and the breast, along
the treatment of choice at diagnosis, with hormonal therapy (as with the feasibility of radiation therapy. In addition, angiosarco-
for ESS) a resource for progressing disease and surgery. The mas of the mammary gland have such a tendency to recur that
same applies to peritoneal leiomyomatosis, if non-mutilating mastectomy (involving the muscular fascia) is recommended in
surgery is not feasible. most cases, even in combination with postoperative radiation
For pelvic aggressive angiomyxoma, surgery is the treatment of therapy. Lymphadenectomy is not carried out in the absence of
choice if not mutilating, with observation thereafter. In progres- clinical evidence of involvement.
sing disease, hormonal therapy, or interruption of any ongoing As far as adjuvant and neoadjuvant chemotherapy is con-
stimulation with oestrogens, may allow mutilating surgery to be cerned, the same principles of STS apply. One may consider in
avoided and the disease to be kept under control [53]. particular the high risk of angiosarcoma to develop local and
iii | The ESMO/European Sarcoma Network Working Group Volume 25 | Supplement 3 | September 2014
by guest
on 05 February 2018
Service Grading System)a
Levels of evidence
I Evidence from at least one large randomised, controlled trial of good methodological quality (low potential for bias) or meta-analyses
of well-conducted randomised trials without heterogeneity
II Small randomised trials or large randomised trials with a suspicion of bias (lower methodological quality) or meta-analyses of such trials
or of trials with demonstrated heterogeneity
V Studies without control group, case reports and experts opinions
a
systemic relapses, so that preoperative treatments may be resorted • Alexia Bertuzzi, Ireland
to, including chemotherapy, radiation therapy, etc. Re-irradiation • Stefan Bielack, Germany
should be considered in radiation-induced angiosarcomas. • Bodil Bjerkehagen, Norway
• Sylvie Bonvalot, France
• Ioannis Boukovinas, Greece
note • Paolo Bruzzi, Italy
Levels of evidence and grades of recommendation have been • Angelo Paolo Dei Tos, Italy
applied using the system shown in Table 3. Statements without • Palma Dileo, UK
grading were considered justified standard clinical practice by • Mikael Eriksson, Sweden
the panel members. The above recommendations apply to • Alexander Fedenko, Russian Federation
adult-type STSs arising from limbs and the superficial trunk. • Andrea Ferrari, Italy
Guidelines on retroperitoneal sarcomas, desmoid-type fibroma- • Stefano Ferrari, Italy
tosis, uterine sarcomas, and breast sarcomas are provided separ- • Hans Gelderblom, Belgium
ately at the end of the chapter with regard to those main aspects • Robert Grimer, UK
by which they differ from more frequent STSs. In general, the • Alessandro Gronchi, Italy
main principles of diagnosis and treatment may well apply to all • Rick Haas, Netherlands
STSs, including the rarest presentations (e.g. visceral sarcomas • Kirsten Sundby Hall, Norway
other than gastrointestinal stromal tumour), which are therefore • Peter Hohenberger, Germany
not specifically covered. Specific histological types, however, • Rolf Issels, Germany
may deserve specific approaches, not necessarily covered here- • Heikki Joensuu, Finland
after, given the scope of these guidelines. • Ian Judson, UK
• Axel Le Cesne, France
• Saskia Litière, Belgium
consensus panel ESMO Guidelines 2014 • Javier Martin-Broto, Spain
• Ofer Merimsky, Israel
These Clinical Practice Guidelines have been developed following a
• Michael Montemurro, UK
consensus process based on a consensus event organised by ESMO
• Carlo Morosi, Italy
in Milan, Italy, in December 2013 and refined by July 2014. This
• Piero Picci, Italy
involved experts from the community of the European sarcoma re-
• Isabelle Ray-Coquard, France
search groups and ESMO faculty. Their names are indicated here-
• Peter Reichardt, Germany
after. The text reflects an overall consensus among them, although
• Piotr Rutkowski, Poland
each of them may not necessarily find it consistent with his/her
• Marcus Schlemmer, Germany
own views. The panel worked on the text of ESMO Guidelines of
• Silvia Stacchiotti, Italy
previous years, whose authorship should also be credited.
• Valter Torri, Italy
• Paolo G. Casali, Italy (Moderator) • Annalisa Trama, Italy
• Jean-Yves Blay, France (Moderator) • Frits Van Coevorden, Netherlands
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu254 | iii

by guest
on 05 February 2018
• Winette Van der Graaf, Netherlands Novartis and PharmaMar. Ofer Merimsky declared: speakers’ hon-
• Daniel Vanel, Italy oraria: GlaxoSmithKline, Lilly; advisory board: Boehringer
• Eva Wardelmann, Germany Ingleheim, Medison; research grant: Roche. Rosalba Miceli
declared: consultancy/Honoraria/Travel grants: Bayer, Novartis,
A consensus meeting was specifically held on retroperitoneal Pfizer. Piero Picci declared advisory board for Takeda. Piotr
sarcoma, whose output was a separate position paper but which Rutkowski declared honoraria from Novartis, Pfizer, Bristol-Myers
also contributed to the retroperitoneal sarcoma paragraph of the Squibb, Roche, GlaxoSmithKline; advisory board for Novartis,
ESMO Guidelines. In addition to some of the above mentioned GlaxoSmithKline, Merck Sharp & Dohme and Bayer. Marcus
experts, it was also made up by the following panellists: Schlemmer declared: honoraria from Novartis, Pfizer and Teva; re-
• Chiara Colombo, Italy search grants from Novartis. Silvia Stacchiotti declared: research
• Marco Fiore, Italy grants: Novartis, Pfizer, PharmaMar, GlaxoSmithKline, Amgen,
• Luigi Mariani, Italy Bayer. Frits Van Coevorden declared travel grants from Novartis
• Rosalba Miceli, Italy and PharmaMar. Winette Van der Graaf declared: research
• Raphael E. Pollock, USA funding from GlaxoSmithKline, Novartis and Pfizer. The following
• Chandrajit P. Raut, USA authors have declared no potential conflicts of interest: Alexia
• Dirk Strauss, UK Bertuzzi, Bodil Bjerkehagen, Ioannis Boukovinas, Chiara
• Carol J. Swallow, Canada Colombo, Palma Dileo, Andrea Ferrari, Angiolo Gadducci, Rick
Haas, Kirsten Sundby Hall, Saskia Litière, Michael Montemurro,
A consensus meeting was specifically held on Uterine sarcoma, Carlo Morosi, Raphael E. Pollock, Chandrajit P. Raut, Isabelle Ray-
whose output was a separate position paper but which also con- Coquard, Dirk Strauss, Valter Torri, Daniel Vanel and Eva
tributed to the Uterine sarcoma paragraph of the ESMO Wardelmann. The other authors have not reported any potential
Guidelines. In addition to some of the above mentioned experts, conflicts of interest.
it was also made up by the following panellists:
• Angiolo Gadducci, Italy references
• Suzanne George, USA 1. Stiller CA, Trama A, Serraino D et al. Descriptive epidemiology of sarcomas in
• Martee L. Hensley, USA Europe: report from the RARECARE project. Eur J Cancer 2013; 49: 684–695.
• Roberta Sanfilippo, Italy 2. Fletcher CDM, Bridge JA, Hogendoorn PCW, Mertens F (eds). WHO Classification
of Tumours of Soft Tissue and Bone. Lyon: IARC 2013.
acknowledgements 3. Ray-Coquard I, Montesco MC, Coindre JM et al. Sarcoma: concordance between
initial diagnosis and centralized expert review in a population-based study within
We deeply thank Barbara Dore and Estelle Lecointe (from three European regions. Ann Oncol 2012; 23: 2442–2449.
SPAEN), and Hans Keulen (from Chordoma Foundation), who 4. Trojani M, Contesso G, Coindre JM et al. Soft-tissue sarcomas of adults: study of
observed the consensus conference as patient representatives. pathological prognostic variables and definition of a histopathological grading
system. Int J Cancer 1984; 33: 37–42.
5. Soft tissue sarcoma. In Edge SB, Byrd DR, Compton CC et al. (eds), AJCC Cancer
conflict of interest Staging Manual, 7th edition. New York, NY: Springer 2010; 291–296.
6. Yang JC, Chang AE, Baker AR et al. Randomized prospective study of the benefit
Stefan Bielack declared: consultancy/advisory board/speakers
of adjuvant radiation therapy in the treatment of soft tissue sarcomas of the
bureau from Merck, IDM/Takeda, Roche, Celgene, Bayer and extremity. J Clin Oncol 1998; 16: 197–203.
Chugai. Sylvie Bonvalot declared: travel grants from PharmaMar, 7. Beane JD, Yang JC, White D et al. Efficacy of adjuvant radiation therapy in the
Nanobiotix and honoraria from Novartis. Paolo G. Casali declared: treatment of soft tissue sarcoma of the extremity: 20-year follow-up of a
consultancy/Honoraria: Amgen Domplé, ARIAD, Bayer, randomized prospective trial. Ann Surg Oncol 2014; 21: 2484–2489.
GlaxoSmithKline, Infinity, Janssen Cilag, Merck Sharp & Dohme, 8. Pisters PW, Harrison LB, Leung DH et al. Long-term results of a prospective
Novartis, Pfizer, PharmaMar, Sanofi. Angelo Paolo Dei Tos randomized trial of adjuvant brachytherapy in soft tissue sarcoma. J Clin Oncol
declared: speakers’ bureau: Novartis Oncology, Pfizer, 1996; 14: 859–868.
GlaxoSmithKline, PharmaMar. Mikael Eriksson declared: honor- 9. Cahlon O, Brennan MF, Jia X et al. A postoperative nomogram for local recurrence
risk in extremity soft tissue sarcomas after limb-sparing surgery without adjuvant
aria from Novartis, Swedish Orphan Biovitrum, GlaxoSmithKline,
radiation. Ann Surg 2012; 255: 343–347.
Merck Sharp & Dohme, and Pfizer. Alexander Fedenko declared:
10. O’Sullivan B, Davis AM, Turcotte R et al. Preoperative versus postoperative
speakers’ bureau: Johnson & Johnson, GlaxoSmithKline, Roche. radiotherapy in soft-tissue sarcoma of the limbs: a randomized trial. Lancet 2002;
Stefano Ferrari declared speakers’ honoraria from Takeda; advisory 359: 2235–2241.
board for Amgen; research grants from Mulmed, Amgen and 11. Issels RD, Lindner LH, Verweij J et al. Neo-adjuvant chemotherapy alone or with
Morphotek. Hans Gelderblom declared research grants from regional hyperthermia for localized high-risk soft-tissue sarcoma: a randomized
Novartis, Pfizer, PharmaMar, GlaxoSmithKline, Eisai and Bayer. phase 3 multicentre study. Lancet Oncol 2010; 11: 561–570.
Robert Grimer declared: research grant: Amgen. Alessandro 12. Deroose JP, Eggermont AM, van Geel AN et al. Long-term results of tumor
Gronchi declared: advisory board: Novartis; honoraria: Novartis necrosis factor alpha- and melphalan-based isolated limb perfusion in locally
advanced extremity soft tissue sarcomas. J Clin Oncol 2011; 29: 4036–4044.
and Pfizer. Rolf Issels declared: consultancy/honoraria:
13. Woll PJ, Reichardt P, Le Cesne A et al. EORTC Soft Tissue and Bone Sarcoma
PharmaMar, Bayer, Therm Med. Heikki Joensuu declared: re- Group and the NCIC Clinical Trials Group Sarcoma Disease Site Committee.
search funding to institute from Novartis. Axel Le Cesne declared: Adjuvant chemotherapy with doxorubicin, ifosfamide, and lenograstim for resected
honoraria: Novartis, PharmaMar, GlaxoSmithKline, Pfizer. Javier soft-tissue sarcoma (EORTC 62931): a multicentre randomised controlled trial.
Martin-Broto declared advisory boards for GlaxoSmithKline, Lancet Oncol 2012; 13: 1045–1054.
iii | The ESMO/European Sarcoma Network Working Group Volume 25 | Supplement 3 | September 2014
by guest
on 05 February 2018
14. Frustaci S, Gherlinzoni F, De Paoli A et al. Adjuvant chemotherapy for adult soft 33. Butrynski JE, D’Adamo DR, Hornick JL et al. Crizotinib in ALK-rearranged
tissue sarcomas of the extremities and girdles: results of the Italian randomized inflammatory myofibroblastic tumor. N Engl J Med 2010; 363: 1727–1733.
cooperative trial. J Clin Oncol 2001; 19: 1238–1247. 34. Kummar S, Allen D, Monks A et al. Cediranib for metastatic alveolar soft part
15. Pervaiz N, Colterjohn N, Farrokhyar F et al. A systematic meta-analysis of sarcoma. J Clin Oncol 2013; 31: 2296–2302.
randomized controlled trials of adjuvant chemotherapy for localized resectable 35. Stacchiotti S, Tamborini E, Marrari A. Response to sunitinib malate in advanced
soft-tissue sarcoma. Cancer 2008; 113: 573–581. alveolar soft part sarcoma. Clin Cancer Res 2009; 15: 1096–1104.
16. Gronchi A, Frustaci S, Mercuri M et al. Short, full-dose adjuvant chemotherapy in 36. Stacchiotti S, Negri T, Libertini M et al. Sunitinib malate in solitary fibrous tumor
high-risk adult soft tissue sarcomas: a randomized clinical trial from the Italian (SFT). Ann Oncol 2012; 23: 3171–3179.
Sarcoma Group and the Spanish Sarcoma Group. J Clin Oncol 2012; 30: 37. Rothermundt C, Whelan JS, Dileo P et al. What is the role of routine follow-up for
850–856. localised limb soft tissue sarcomas? A retrospective analysis of 174 patients. Br J
17. Blackmon SH, Shah N, Roth JA et al. Resection of pulmonary and extrapulmonary Cancer 2014; 110: 2420–2426.
sarcomatous metastases is associated with long-term survival. Ann Thorac Surg 38. Bonvalot S, Gronchi A, Hohenberger P et al. Management of primary
2009; 88: 877–884. retroperitoneal sarcoma (RPS) in the adult. A consensus approach from the Trans-
18. Antman K, Crowley J, Balcerzak SP et al. An intergroup phase III randomized Atlantic RPS Working group. Ann Surg Oncol 2014; in press.
study of doxorubicin and dacarbazine with or without ifosfamide and mesna in 39. Bonvalot S, Rivoire M, Castaing M et al. Primary retroperitoneal sarcomas: a
advanced soft tissue and bone sarcomas. J Clin Oncol 1993; 11: 1276–1285. multivariate analysis of surgical factors associated with local control. J Clin Oncol
19. Judson I, Verweij J, Gelderblom H et al.; European Organisation and Treatment of 2009; 27: 31–37.
Cancer Soft Tissue and Bone Sarcoma Group. Doxorubicin alone versus intensified 40. Bonvalot S, Miceli R, Berselli M et al. Aggressive surgery in retroperitoneal soft
doxorubicin plus ifosfamide for first-line treatment of advanced or metastatic soft- tissue sarcoma carried out at high-volume centers is safe and is associated with
tissue sarcoma: a randomised controlled phase 3 trial. Lancet Oncol 2014; 15: improved local control. Ann Surg Oncol 2010; 17: 1507–1514.
415–423. 41. Gronchi A, Miceli R, Colombo C et al. Frontline extended surgery is associated with
20. Penel N, Bui BN, Bay JO et al. Phase II trial of weekly paclitaxel for unresectable improved survival in retroperitoneal low- to intermediate-grade soft tissue
angiosarcoma: the ANGIOTAX Study. J Clin Oncol 2008; 26: 5269–5274. sarcomas. Ann Oncol 2012; 23: 1067–1073.
21. Stacchiotti S, Palassini E, Sanfilippo R et al. Gemcitabine in advanced 42. Singer S, Antonescu CR, Riedel E, Brennan MF. Histologic subtype and margin of
angiosarcoma: a retrospective case series analysis from the Italian Rare Cancer resection predict pattern of recurrence and survival for retroperitoneal liposarcoma.
Network. Ann Oncol 2012; 23: 501–508. Ann Surg 2003; 238: 358–370.
22. Lorigan P, Verweij J, Papai Z et al. European Organisation for Research and 43. Pisters PW, O’Sullivan B. Retroperitoneal sarcomas: combined modality treatment
Treatment of Cancer Soft Tissue and Bone Sarcoma Group Study. Phase III trial of approaches. Curr Opin Oncol 2002; 14: 400–405.
two investigational schedules of ifosfamide compared with standard-dose 44. Angele MK, Albertsmeier M, Prix NJ et al. Effectiveness of Regional Hyperthermia
doxorubicin in advanced or metastatic soft tissue sarcoma: a European with Chemotherapy for High-risk Retroperitoneal and Abdominal Soft-tissue
Organisation for Research and Treatment of Cancer Soft Tissue and Bone Sarcoma after Complete Surgical Resection: A Randomized phase-III Multicenter
Sarcoma Group Study. J Clin Oncol 2007; 25: 3144–3150. Study. Ann Surg 2014; in press.
23. Rutkowski P, Van Glabbeke M, Rankin CJ et al. Imatinib mesylate in advanced 45. Seidman MA, Oduyebo T, Muto MG et al. Peritoneal dissemination complicating
dermatofibrosarcoma protuberans: pooled analysis of two phase II clinical trials. morcellation of uterine mesenchymal neoplasms. PLoS ONE 2012; 7: e50058.
J Clin Oncol 2010; 28: 1772–1779. 46. Kho KA, Nezhat CH. Evaluating the risks of electric uterine morcellation. JAMA
24. Stacchiotti S, Pedeutour F, Negri T et al. Dermatofibrosarcoma protuberans- 2014; 311: 905–906.
derived fibrosarcoma: clinical history, biological profile and sensitivity to imatinib.
47. Ip PP, Cheung AN. Pathology of uterine leiomyosarcomas and smooth muscle
Int J Cancer 2011; 129: 1761–1772.
tumours of uncertain malignant potential. Best Pract Res Clin Obstet Gynaecol
25. Le Cesne A, Antoine E, Spielmann M et al. High-dose ifosfamide: circumvention of 2011; 25: 691–704.
resistance to standard-dose ifosfamide in advanced soft tissue sarcomas. J Clin 48. Reed NS, Mangioni C, Malmström H et al. European Organisation for Research and
Oncol 1995; 13: 1600–1608.
Treatment of Cancer Gynaecological Cancer Group. Phase III randomised study to
26. Martin-Liberal J, Alam S, Constantinidou A et al. Clinical activity and tolerability of evaluate the role of adjuvant pelvic radiotherapy in the treatment of uterine
a 14-day infusional Ifosfamide schedule in soft-tissue sarcoma. Sarcoma 2013; sarcomas stages I and II: an European Organisation for Research and Treatment of
2013: 868973. Cancer Gynaecological Cancer Group Study ( protocol 55874). Eur J Cancer 2008;
27. Demetri GD, Chawla SP, von Mehren M et al. Efficacy and safety of trabectedin in 44: 808–818.
patients with advanced or metastatic liposarcoma or leiomyosarcoma after failure 49. Hensley ML, Ishill N, Soslow R et al. Adjuvant gemcitabine plus docetaxel for
of prior anthracyclines and ifosfamide: results of a randomized phase II study of completely resected stages I-IV high grade uterine leiomyosarcoma: results of a
two different schedules. J Clin Oncol 2009; 27: 4188–4196. prospective study. Gynecol Oncol 2009; 112: 563–567.
28. Grosso F, Jones RL, Demetri GD et al. Efficacy of trabectedin (ecteinascidin-743) 50. Hensley ML, Wathen JK, Maki RG et al. Adjuvant therapy for high-grade, uterus-
in advanced pretreated myxoid liposarcomas: a retrospective study. Lancet Oncol limited leiomyosarcoma: results of a phase 2 trial (SARC 005). Cancer 2013; 119:
2007; 8: 595–602. 1555–1561.
29. Maki RG, Wathen JK, Patel SR et al. Randomized phase II study of gemcitabine 51. Rauh-Hain JA, del Carmen MG. Endometrial stromal sarcoma: a systematic review.
and docetaxel compared with gemcitabine alone in patients with metastatic soft Obstet Gynecol 2013; 122: 676–683.
tissue sarcomas: results of sarcoma alliance for research through collaboration 52. Lee CH, Mariño-Enriquez A, Ou W et al. The clinicopathologic features of
study 002. J Clin Oncol 2007; 25: 2755–2763.
YWHAE-FAM22 endometrial stromal sarcomas: a histologically high-grade and
30. García-Del-Muro X, López-Pousa A, Maurel J et al. Randomized phase II study clinically aggressive tumor. Am J Surg Pathol 2012; 36: 641–653.
comparing gemcitabine plus dacarbazine versus dacarbazine alone in patients with 53. Schwartz PE, Hui P, McCarthy S. Hormonal therapy for aggressive angiomyxoma: a
previously treated soft tissue sarcoma: a Spanish Group for Research on Sarcomas
case report and proposed management algorithm. J Low Genit Tract Dis 2014;
study. J Clin Oncol 2011; 29: 2528–2533.
18: E55–E61.
31. van der Graaf WT, Blay JY, Chawla SP et al. Pazopanib for metastatic soft-tissue
54. Fiore M, Rimareix F, Mariani L et al. Desmoid-type fibromatosis: a front-line
sarcoma (PALETTE): a randomised, double-blind, placebo-controlled phase 3 trial. conservative approach to select patients for surgical treatment. Ann Surg Oncol
Lancet 2012; 379: 1879–1886.
2009; 16: 2587–2593.
32. Wagner AJ, Malinowska-Kolodziej I, Morgan JA et al. Clinical activity of mTOR
55. Bonvalot S, Eldweny H, Haddad V et al. Extra-abdominal primary fibromatosis:
inhibition with sirolimus in malignant perivascular epithelioid cell tumors: targeting aggressive management could be avoided in a subgroup of patients. Eur J Surg
the pathogenic activation of mTORC1 in tumors. J Clin Oncol 2010; 28: 835–840.
Oncol 2008; 34: 462–468.
Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu254 | iii

by guest
on 05 February 2018
56. Gronchi A, Colombo C, Le Péchoux C et al. ISG and FSG. Sporadic desmoid-type 60. Weiss AJ, Horowitz S, Lackman RD. Therapy of desmoid tumors and fibromatosis
fibromatosis: a stepwise approach to a non-metastasising neoplasm—a position using vinorelbine. Am J Clin Oncol 1999; 22: 193–195.
paper from the Italian and the French Sarcoma Group. Ann Oncol 2014; 25: 61. Constantinidou A, Jones RL, Scurr M et al. Pegylated liposomal doxorubicin, an
578–583. effective, well-tolerated treatment for refractory aggressive fibromatosis. Eur J
57. de Camargo VP, Keohan ML, D’Adamo DR et al. Clinical outcomes of systemic Cancer 2009; 45: 2930–2934.
therapy for patients with deep fibromatosis (desmoid tumor). Cancer 2010; 116: 62. Patel SR, Benjamin RS. Desmoid tumors respond to chemotherapy: defying the
2258–2265. dogma in oncology. J Clin Oncol 2006; 24: 11–12.
58. Skapek SX, Anderson JR, Hill DA et al. Safety and efficacy of high-dose tamoxifen 63. Gounder MM, Lefkowitz RA, Keohan ML et al. Activity of sorafenib
and sulindac for desmoid tumor in children: results of a Children’s Oncology Group against desmoid tumor/deep fibromatosis. Clin Cancer Res 2011; 17:
(COG) phase II study. Pediatr Blood Cancer 2013; 60: 1108–1112. 4082–4090.
59. Azzarelli A, Gronchi A, Bertulli R et al. Low-dose chemotherapy with methotrexate 64. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
and vinblastine for patients with advanced aggressive fibromatosis. Cancer 2001; among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
92: 1259–1264. 139–144.
iii | The ESMO/European Sarcoma Network Working Group Volume 25 | Supplement 3 | September 2014
by guest
on 05 February 2018
Nasopharyngeal cancer: EHNS–ESMO–ESTRO Clinical

follow-up†
A.T.C. Chan1, V. Grégoire2, J.-L. Lefebvre3, L. Licitra4, E.P. Hui1, S.F. Leung1 & E. Felip5, on behalf
of the EHNS–ESMO–ESTRO Guidelines Working Group*
1
Department of Clinical Oncology, State Key Laboratory in Oncology in South China, Sir YK Pao Centre for Cancer, The Chinese University of Hong Kong, Hong Kong;
2
Department of Radiation Oncology, St-Luc University Hospital, Brussels, Belgium; 3Department of Head and Neck Surgery, Centre Oscar Lambret, Lille, France;
4
Medical Oncology Head and Neck Unit, Istituto Nazionale dei Tumori, Milan, Italy; 5Medical Oncology Service, Vall d’Hebron University Hospital, Barcelona, Spain
incidence examination including cranial nerve examination, complete

blood cell count, serum biochemistry (including liver
Cancer of the nasopharynx (NPC) is rare in Europe, with an function test), chest X-ray, nasopharyngoscopy, computed
annual crude incidence rate of 1.1 per 100 000. On the tomography (CT) scan or magnetic resonance imaging
European scale, NPC accounts for 4760 new cases per year. (MRI) of nasopharynx and base of skull and neck. Although
Incidence is higher in men than women. [1, 2]. MRI is generally preferred if available, each center will
In Europe, the relative survival for NPC was 76% at 1 year choose the best imaging technique according to their usual
clinical practice
and 50% at 5 years in adults. There were no survival clinical practice and experience [III, B]. Imaging for distant
guidelines
differences between the sexes. The effect of age on survival is metastases including isotope bone scan and CT scan of
marked. Survival at 5 years was 72% for the youngest age chest and upper abdomen could be considered for at-risk
group (15–45 years) and 36% in the oldest group of patients subsets (node positive, especially N3 stage) and for those
(65–74 years) [1, 2]. patients with clinical or biochemical abnormalities detected
[III, B]. The use of positron emission tomography CT scan
can replace the traditional work-up for detection of distant
diagnosis metastatic disease since it has proved to be the most
Definitive diagnosis is made by endoscopic-guided biopsy of sensitive, specific and accurate diagnostic method. Both the
the primary nasopharyngeal tumor. The histological type pre-treatment and post-treatment plasma/serum load of
should be classified according to World Health Organization Epstein-Barr viral DNA has been shown to be of prognostic
(WHO) classification [3]. Since the first disease sign in value [III, B] [4–8].
patients is often the appearance of neck nodes it is not
infrequent that patients undergo neck biopsy and/or neck
nodal dissection. This procedure is not recommended since treatment
it may reduce cure probability and have an impact on late
The optimal treatment strategy of patients with advanced NPC
treatment sequelae.
should be discussed in a multidisciplinary team. Radiation
therapy (RT) is the mainstay of treatment and is an essential
component of curative-intent treatment of non-disseminated
NPC. Stage I disease is treated by RT alone, while stage III,
NPC is clinically staged according to the International IVA, IVB disease are treated by RT with concurrent
Union Against Cancer (UICC) and American Joint chemotherapy [I, A]. Concurrent chemotherapy is
Committee on Cancer (AJCC) staging system (Table 1). recommended for stage II disease [I, B] [9]. Patients should be
Routine staging procedures include history, physical treated by intensity-modulated radiation therapy (IMRT) [II,
A] [10]. RT is targeted to the primary tumor and the adjacent
regions considered at risk of microscopic spread from the
L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland; E-mail: clinicalguidelines@esmo. tumor, and to both the sides of the neck (levels Ib–V, and
org retropharyngeal nodes). For patients with lower neck nodes,
†
the supraclavicular fossa should be included as well. Elective
Approved by the ESMO Guidelines Working Group: December 2006, last update June
2012. This publication supersedes the previously published version—Ann Oncol 2010; nodal irradiation is recommended for N0 stage disease. The
21 (Suppl 5): v187–v189. consensus is that a total dose of 70 Gy is needed for

by guest
on 05 February 2018
Table 1. The International Union Against Cancer (UICC) and American Joint Committee on Cancer (AJCC) staging system for NPC, seventh edition
(2010)
Primary tumor (T)

T1 Tumor confined to the nasopharynx, or extends to oropharynx and/or nasal cavity
without parapharyngeal extension
T2 Tumor with parapharyngeal extension
T3 Tumor involves bony structures of skull base and/or paranasal sinuses
T4 Tumor with intracranial extension and/or involvement of cranial nerves, hypopharynx, orbit,
or with extension to the infratemporal fossa/masticator space
N1 Unilateral metastasis in cervical lymph node(s), ≤6 cm in greatest dimension,
above the supraclavicular fossa, and/or unilateral or bilateral, retropharyngeal
lymph nodes, ≤6 cm, in greatest dimension
N2 Bilateral metastasis in cervical lymph node(s), ≤6 cm in greatest dimension,
above the supraclavicular fossa
N3 Metastasis in a lymph node(s) >6 cm and/or to supraclavicular fossa
N3a >6 cm in dimension
N3b Extension to the supraclavicular fossa
Stage 0 Tis N0 M0
Stage I T1 N0 M0
Stage II T1 N1 M0
T2 N0 M0
T2 N1 M0
Stage III T1 N2 M0
T2 N2 M0
T3 N0 M0
T3 N1 M0
T3 N2 M0
Stage IVA T4 N0 M0
T4 N1 M0
T4 N2 M0
Stage IVB Any T N3 M0
Stage IVC Any T Any N M1
eradication of gross tumor and either 50–60 Gy or 46–60 Gy chemotherapy negatively affect the optimal administration of
for elective treatment of potential risk sites. To minimize the concomitant chemoradiation.
risk of late toxicity ( particularly, to adjacent neurological
structures), fractional dose >2 Gy per daily fraction and
excessive acceleration with multiple fractions >1.9 Gy/fraction follow-up
should be avoided [III, E]. IMRT may offer improvement in Documentation of complete remission in the nasopharynx and
local tumor control [II, A], and reduction in radiation neck through clinical and endoscopic examination and/or
xerostomia in early-stage disease [II, A]. The standard agent imaging studies is important. MRI is often used to evaluate the
used in concurrent chemotherapy–RT is cisplatin [I, A]. This response to RT or chemoradiotherapy, especially for T3 and T4
provides a benefit in terms of overall survival and on both tumors, though distinction between post-irradiation changes
locoregional and distant control [9, 11–15]. While three cycles and recurrent tumors may be difficult. Follow-up for patients
of adjuvant cisplatin-5FU has been a standard part of many includes periodic examination of the nasopharynx and neck,
concurrent chemoradiotherapy regimens, its benefit is cranial nerve function and evaluation of systemic complaints to
uncertain and toxic effect is substantial [16]. Cisplatinum- identify distant metastasis. For T3 and T4 tumors, MRI might
based induction chemotherapy has been shown to improve be used on a 6- to 12-month basis to evaluate the nasopharynx
disease-free survival and may be considered in locally and the base of the skull at least for the first few years after
advanced disease although it is not seen as a standard treatment. Evaluation of thyroid function in patients with
treatment [II, B] [17]. In no case should induction irradiation to the neck is recommended at 1, 2 and 5 years.
vii | Chan et al. Volume 23 | Supplement 7 | October 2012

by guest
on 05 February 2018
treatment of recurrent or metastatic Ingelheim. Dr. Hui has reported: consultancy and advisory
disease board for Sanofi-Aventis and Pfizer. Prof. Grégoire has
reported no conflicts of interest.
Small local recurrences are potentially curable and the main
issue is the choice of the most appropriate therapeutic options,
which include nasopharyngectomy, brachytherapy,
radiosurgery, stereotactic RT, IMRT or a combination of
surgery and RT, with or without concurrent chemotherapy.
references
Treatment decisions are tailored to the specific situation of 1. Curado MP, Edwards B, Shin HR et al. (eds). Cancer Incidence in Five
individual cases, taking into consideration the volume, location Continents, Vol. IX. IARC Scientific Publications No. 160. Lyon: IARC 2007.
and extent of the recurrent tumor [III, A]. Regional recurrence 2. Ferlay J, Bray F, Pisani P, Parkin DM. GLOBOCAN 2002 Cancer Incidence,
is managed by radical neck dissection if resectable [III, A]. Mortality and Prevalence Worldwide. IARC CancerBase No. 5, version 2.0. Lyon:
IARC Press 2004.
In metastatic NPC, palliative chemotherapy should be
3. Barnes L, Eveson JW, Reichart P, Sidransky D. World Health Organization
considered for patients with adequate performance status. Classification of Tumours. Pathology and Genetics of Head and Neck Tumours.
Platinum combination regimens are commonly used as first- Lyon: IARC Press 2005.
line therapy since cisplatin represents the most effective drug. 4. Chua DT, Ji M, Zong Y et al. Screening of nasopharyngeal carcinoma by serology
Other active agents include paclitaxel, docetaxel, gemcitabine, and nasopharyngoscopy and treatment outcome in endemic region. J Clin Oncol
capecitabine, irinotecan, vinorelbine, ifosfamide, doxorubicin 2009; 27: 15s.
and oxaliplatin, which can be used as single agents or in 5. Chua ML, Ong SC, Wee JT et al. Comparison of 4 modalities for distant
combination [III, B]. Polychemotherapy is more active than metastasis staging in endemic nasopharyngeal carcinoma. Head Neck 2009; 31:
346–354.
monotherapy. In this context treatment choice should be based
6. Chan AT, Lo YM, Zee B et al. Plasma Epstein–Barr virus DNA and residual
on previous treatments and the expected toxicity [18].
disease after radiotherapy for undifferentiated nasopharyngeal carcinoma. J Natl
Cancer Inst 2002; 94: 1614–1619.
Table 2 Summary of treatment recommendations for Cancer of the 7. Lin JC, Wang WY, Chen KY et al. Quantification of plasma Epstein-Barr virus
nasopharynx (NPC) DNA in patients with advanced nasopharyngeal carcinoma. N Engl J Med 2004;
350 (24): 2461.
Early stage Stage I Radiation alone 8. Leung SF, Zee B, Ma BB et al. Plasma Epstein-Barr viral deoxyribonucleic acid
quantitation complements TNM staging in nasopharyngeal carcinoma
Intermediate stage Stage II Concurrent chemoradiotherapy
prognostication. J Clin Oncol 2006; 34: 5414–5418.
(I, B)
9. Chen QY, Wen YF, Guo L et al. Concurrent chemoradiotherapy vs radiotherapy
Advanced stage Stage III, IVA, Concurrent chemoradiotherapy
alone in Stage II nasopharyngeal carcinoma: phase III randomized Trial. J Natl
IVB +/− adjuvant chemotherapy Cancer Inst 2011; 103 (23): 1761.
(I, A) 10. Kam MK, Leung SF, Zee B et al. Prospective randomized study of intensity
Problematic radiation Stage IVA, Induction chemotherapy modulated radiotherapy on salivary gland function in early-stage nasopharyngeal
therapy (RT) IVB followed by concurrent carcinoma patients. J Clin Oncol 2007; 25: 4873–4879.
planning (e.g. chemoradiotherapy (II, B) 11. Kwong DL, Sham JS, Au GK et al. Concurrent and adjuvant chemotherapy for
tumor abutting nasopharyngeal carcinoma: a factorial study. J Clin Oncol 2004; 22:
chiasm) 2643–2653.
12. Chan AT, Leung SF, Ngan RK et al. Overall survival after concurrent cisplatin-
radiotherapy compared with radiotherapy alone in locoregionally advanced
nasopharyngeal carcinoma. J Natl Cancer Inst 2005; 97: 536–539.
notes
13. Lin JC, Jan JS, Hsu CY et al. Phase III study of concurrent chemoradiotherapy
Levels of evidence [I–V] and grades of recommendation [A–E] versus radiotherapy alone for advanced nasopharyngeal carcinoma: positive effect
adapted from the Infectious Diseases Society of American- on overall and progression-free survival. J Clin Oncol 2003; 21: 631–637.
United States Public Health Service Grading System are given 14. Wee J, Tan EH, Tai BC et al. Randomized trial of radiotherapy versus concurrent
chemoradiotherapy followed by adjuvant chemotherapy in patients with American
in square brackets. Statements without grading were considered
Joint Committee on Cancer/International Union against cancer stage III and IV
justified standard clinical practice by the experts. nasopharyngeal cancer of the endemic variety. J Clin Oncol 2005; 23:
6730–6738.
conflict of interest 15. Lee AW, Tung SY, Chua DT et al. Randomized trial of radiotherapy plus
concurrent-adjuvant chemotherapy vs radiotherapy alone for regionally advanced
Prof. Lefebvre has reported: lecturer and advisory board for nasopharyngeal carcinoma. J Natl Cancer Inst 2010; 102 (15): 1188.
Merck Serono and Sanofi-Aventis. Dr. Licitra has reported: 16. Chen L, Hu CS, Chen XZ et al. Concurrent chemoradiotherapy plus adjuvant
advisory board of Bristol-Myers Squibb, GlaxoSmithKine, Lilly, chemotherapy versus concurrent chemoradiotherapy alone in patients with
Merck Serono and Amgen; institution has received clinical and locoregionally advanced nasopharyngeal carcinoma: a phase 3 multicentre
randomised controlled trial. Lancet Oncol 2012; 13(2): 163–171.
research support from EISAI, Exelixis, Lilly, Merck-Serono,
17. Chua DT, Ma J, Sham JS et al. Long-term survival after cisplatin-based induction
Amgen; and travel support from Merck Serono. Prof. Chan has
chemotherapy and radiotherapy for nasopharyngeal carcinoma: a pooled data
reported receiving honoraria and research support from analysis of two phase III trials. J Clin Oncol 2005; 23: 1118–1124.
Merck-Serono, Amgen, Novartis, Roche, Pfizer and 18. Ma BB, Hui EP, Chan AT. Systemic approach to improving treatment outcome in
SanofiAventis. Dr. Felip has reported: consultancy/honoraria nasopharyngeal carcinoma: current and future directions. Cancer Sci 2008; 99
from Lilly, GlaxoSmithKline, Merck Serono, Roche, Boehringer (7): 1311–1318.
Volume 23 | Supplement 7 | October 2012 doi:10.1093/annonc/mds266 | vii

by guest
on 05 February 2018
doi:10.1093/annonc/mdq185
Squamous cell carcinoma of the head and neck:

EHNS–ESMO–ESTRO Clinical Practice Guidelines for
diagnosis, treatment and follow-up
V. Grégoire1, J.-L. Lefebvre2, L. Licitra3 & E. Felip4
On behalf of the EHNS–ESMO–ESTRO Guidelines Working Group*
1
Department of Radiation Oncology, St-Luc University Hospital, Brussels, Belgium; 2Department of Head and Neck Surgery, Centre Oscar Lambret, Lille, France;
3
Medical Oncology Head and Neck Unit, Istituto Nazionale dei Tumori, Milan, Italy; 4Medical Oncology Service, Vall d’Hebron University Hospital, Barcelona, Spain
incidence 2-deoxy-D-glucose positron emission tomography (FDG-PET

or PET-CT) at staging is under investigation. In this context it
In 2002, the crude incidence rates of carcinoma of the head and can be stated that in general PET has a lower specificity than
neck (ICD, 10th revision C00–C10, C12–C14, C32) in Europe sensitivity, and that it may be more useful for staging the
were 36/100 000/year in the male population and 7/100 000/ metastasis or synchronous tumours than the neck node.
year for females, while the corresponding mortality rates were Squamous cell head and neck cancer should be staged
18 and 3/100 000/year. On the European scale, head and neck according to the TNM system and grouped into categories
cancer accounts for 139 000 new cases per year. More than 90% shown in Table 1. According to the seventh AJCC classification
of head and neck malignancies are squamous cell carcinomas. T4 tumours are subdivided into T4a moderately advanced and
In Europe the relative survival rate for head and neck cancer T4b very advanced. Stage IV is subdivided into stages IVa and
patients was 72% at 1 year and 42% at 5 years in adults. Five- IVb accordingly, and stage IVc for metastatic disease.
year survival was higher in women (51%) than men (39%). The Modern risk assessment should also include that for
effect of age on survival is marked. Survival at 5 years was 54% oropharyngeal tumour, whether the disease is HPV related,
for the youngest age group (15–45 years) and 35% in the oldest together with the smoking habits.
group of patients (‡75 years old).
treatment plan
diagnosis A multidisciplinary treatment schedule should be established in
Pathological diagnosis should be made according to the World all cases. The patient’s nutritional status must be corrected and
Health Organization classification from a surgical biopsy maintained. Dental rehabilitation is indicated before
sample. radiotherapy. Treatment depends on primary tumour location
Routine staging includes physical examination, chest X-ray, and extension. Rare squamous head and neck cancer
head and neck endoscopy, and head and neck computed originating from paranasal sinuses and nasopharynx are usually
tomography (CT) scan or magnetic resonance imaging (MRI). excluded from trial treatment series supporting evidence-based
MRI is the preferable staging procedure for every tumour recommendations, so they are excluded from these clinical
subsite except laryngeal and hypopharyngeal cancers. A recommendations. In early stage (I–II), either conservative
thoracic CT scan may be performed to rule out metastatic surgery or radiotherapy (external radiotherapy or
disease and or second lung primaries. The role of 2-[18F]fluoro- brachytherapy) gives similar loco-regional control. However,
this is based only on retrospective studies as there are no
*Correspondence to: ESMO Guidelines Working Group, ESMO Head Office, Via L.
randomized trials available for reference. Modern radiotherapy
Taddei 4, CH-6962 Viganello-Lugano, Switzerland; treatment should include 3D conformal radiation therapy or
E-mail: clinicalrecommendations@esmo.org intensity-modulated radiation therapy (IMRT).
Approved by the ESMO Guidelines Working Group: June 2002, last update February
Standard options for locally advanced stage III and IV
2010. This publication supercedes the previously published version-Ann Oncol 2009; 20 tumours are: surgery including reconstruction plus
(Suppl 4): iv121-iv122 postoperative radiotherapy and, for those patients found at
Conflict of interest: Prof. Gregoire has reported no conflicts of interest, Prof. Lefebvre
surgery to have high-risk features (nodal extracapsular
has reported that he is a lecturer and a member of the advisory boards of MerckSerono extension and/or R1 resection), post-operative
and Sanofi-Aventis; Dr Licitra has reported that she is on the Advisory Board of BMS, chemoradiotherapy (CRT) with single-agent platinum [I, A].
Glaxo, Lilly, Merk-Serono and Amgen; her institution has received funds for clinical
However, in resectable patients, when the anticipated
studies and research activities in which she is involved from EISAI, Exelixis, Lilly, Merk-
Serono, Amgen; she has received travel coverages to attend medical meetings from functional outcome and/or the prognosis is so poor that
Merk-Serono; Dr Felip has reported no conflicts of interest. mutilating surgery is not justified, combined concomitant
ª The Author 2010. Published by Oxford University Press on behalf of the European Society for Medical Oncology.
All rights reserved. For permissions, please email: journals.permissions@oxfordjournals.org

by guest
on 05 February 2018
Table 1. TNM categories for squamous cell head and neck cancer (TNM treatment with surgery [II, A], although in general those
seventh edition 2009) patients undergoing such treatments tend to have a reduction
in distant metastasis.
Stage I T1 N0 M0
Stage II T2 N0 M0 local, regional and metastatic
Stage III T3 N0 M0
T1, T2, T3 N1 M0
recurrence
Stage IVA T1, T2, T3 N2 M0 In selected cases of localized recurrence, surgery (if operable) or
T4a N0, N1, N2 M0 re-irradiation can be considered. For most patients palliative
Stage IVB Tb Any N M0 chemotherapy is the standard option. First-line option for fit
Any T N3 M0 patients should include the combination of cetuximab with
Stage IVC Any T Any N M1 cisplatin or carboplatin plus 5-fluorouracil (PF). It resulted in
longer survival than PF alone [II, A]. In patients for which
polychemotherapy tolerability is anticipated to be poor
monochemotherapy should be used. Weekly methotrexate may
be considered as the accepted treatment [I, A]. Since there is no
chemoradiation is preferred. Combined concomitant comparison between taxanes and methotrexate as monotherapy
chemoradiation is also the standard treatment in non- it is difficult to state whether taxanes are useful in this context.
resectable patients [I, A]. Radiotherapy given concomitantly Cetuximab alone has a favourable toxicity profile with activity
with cetuximab has demonstrated a higher response rate, longer that is comparable to methotrexate alone.
disease-free progression and longer overall survival versus
radiotherapy alone [II, B]. There is no formal comparison
between the combination of radiotherapy with cisplatin or follow-up
cetuximab. In this context the therapeutic decision is difficult Treatment response should be evaluated by clinical
to take. However, it should be considered that results of examination and CT scan or MRI of head and neck depending
concomitant chemoradiation are based on thousands of on the initial procedure. FDG-PET (or PET-CT) may be used
patients, that this combination is associated with significant to evaluate the response to radiotherapy or concomitant CRT
toxicity and that its efficacy in the elderly population is at the neck level and decide upon the usefulness of a neck node
questioned. On the other hand results of cetuximab + radiation dissection. The aim of follow-up is the early detection of
are based on 200 patients, the magnitude in effect was similar potentially curable loco-regional recurrence and second
or even better than that achieved by concomitant tumours. Physical examination along with radiological imaging
chemoradiation, it proved to be less toxic and the benefit in the in the case of suspicion of recurrence should be included in
elderly subgroup is also questioned. the follow-up. FDG-PET scanning may be useful in the
The role of induction chemotherapy (ICT) has been presence of doubtful findings, particularly after combined
reconsidered since the introduction of taxane–platinum-based chemoradiation. In such situations its negative predictive value
(TPF) combinations that have proved to be superior to is superior to the positive one. At this time, special attention
platinum–fluorouracil PF schedule in loco-regionally advanced should be paid to the treatment sequelae that include
disease [I, A]. However, at present, induction chemotherapy is swallowing and respiratory impairment. Chest X-ray may be
not considered standard treatment in advanced disease. ICT included on a yearly basis. Evaluation of thyroid function
followed by RT-CT (so-called sequential CT-RT) is still under (serum thyroid-stimulating hormone—TSH—levels) in
evaluation. The overall toxicity of this approach can be patients with irradiation to the neck is recommended at 1, 2
substantial thus compromising the final result. and 5 years.
TPF induction chemotherapy followed by radiotherapy in
responsive patients is an option for organ preservation in note
advanced larynx and hypopharynx cancer in patients otherwise
requiring total laryngectomy [II, A]. CRT is another option. In Levels of Evidence [I–V] and Grades of Recommendation [A–
one randomized trial, concurrent CRT achieved higher larynx D] as used by the American Society of Clinical Oncology are
preservation rates in the first 2 years after treatment completion given in square brackets. Statements without grading were
but this was not associated with improved survival over either considered justified standard clinical practice by the experts.
ICT followed by radiation in responsive patients or
radiotherapy alone [II A]. The choice between either an ICT- literature
based or a CRT-based organ-preserving protocol depends on
1. Barnes L, Eveson JW, Reichart P, Sidransky D. World Health Organization
various factors (anatomical subsite, foreseable compliance/ Classification of Tumours. Pathology and Genetics of Head and Neck Tumours.
tolerance to treatment, perfomance status, etc.). In addition, Lyon: IARC Press 2005.
not every patient and tumour presentation is suitable for organ 2. Bernier J, Domenge C, Ozsahin M et al. Postoperative irradiation with or without
preservation. Patients with massive larynx cartilage invasion concomitant chemotherapy for locally advanced head and neck cancer. N Engl J
should be excluded from this approach. Whether based on ICT Med 2004; 350: 1945–1952.
or CRT, these treatment options have no negative impact on 3. Bonner JA, Harari PM, Giralt J et al. Radiotherapy plus cetuximab for
disease-free or overall survival, due to the successful salvage locoregionally advanced squamous cell head and neck cancer: 5-year survival
Volume 21 | Supplement 5 | May 2010 doi:10.1093/annonc/mdq185 | v185

by guest
on 05 February 2018
data from a phase III randomised trial, and relation between cetuximab-induced 8. Hitt R, Lopez-Pousa A, Martinez-Trufero J et al. Phase III study comparing
rash and survival. Lancet Oncol 2010; 11: 21–28. cisplatin plus fluorouracil to paclitaxel, cisplatin, and fluorouracil induction
4. Cooper JS, Pajak TF, Forastiere AA et al. Postoperative concurrent radiotherapy chemotherapy followed by chemoradiotherapy in locally advanced head and neck
and chemotherapy for high-risk squamous-cell carcinoma of the head and neck. cancer. J Clin Oncol 2005; 23: 8636–8645.
N Engl J Med 2004; 350: 1937–1944. 9. Pignon Aurélie le Maı̂tre JP, Maillard E et al. Meta-analysis of chemotherapy in
head and neck cancer (MACH-NC): an update on 93 randomised trials and
5. Ferlay J, Bray F, Pisani P, Parkin DM. GLOBOCAN 2002 Cancer Incidence,
17,346 patients. Radiother Oncol 2009; 92: 4–14.
Mortality and Prevalence Worldwide IARC CancerBase No. 5, version 2.0. Lyon:
IARC Press 2004. 10. Posner MR, Hershock DM, Blajman CR et al. Cisplatin and fluorouracil alone or
with docetaxel in head and neck cancer. N Engl J Med 2007; 357: 1705–1715.
6. Forastiere AA, Goepfert H, Maor M et al. Concurrent chemotherapy and
radiotherapy for organ preservation in advanced laryngeal cancer. N Engl J Med 11. Sant M, Allemani C, Santaquilani M et al. Survival of cancer patients diagnosed
2003; 349: 2091–2098. in 1995–1999. Results and commentary. Eur J Cancer 2009; 45: 931–991.
7. Forastiere AA, Metch B, Schuller DE et al. Randomized comparison of cisplatin 12. Vermorken JB, Remener E, van Herpen C et al. Cisplatin, fluorouracil and docetaxel
plus fluorouracil and carboplatin plus fluorouracil versus methotrexate in in unresectable head and neck cancer. N Engl J Med 2007; 357: 1695–1704.
advanced squamous-cell carcinoma of the head and neck: a Southwest 13. Vermorken JB, Mesia R, Rivera F et al. Platinum-based chemotherapy plus
Oncology Group Study. J Clin Oncol 1992; 10: 1245–1251. cetuximab in head and neck cancer. N Engl J Med 2008; 359: 1116–1127.
v186 | Grégoire et al. Volume 21 | Supplement 5 | May 2010

by guest
on 05 February 2018
Adrenal cancer: ESMO Clinical Practice Guidelines for

A. Berruti1, E. Baudin2, H. Gelderblom3, H. R. Haak4, F. Porpiglia5, M. Fassnacht6 &
G. Pentheroudakis7 on behalf of the ESMO Guidelines Working Group*
1
Medical Oncology, Department of Clinical and Biological Sciences, University of Torino, Azienda Ospedaliero, Universitaria San Luigi, Orbassano, Italy; 2Service de
Médecine Nucléaire et de Cancerologie Endocrinienne, Institut Gustave-Roussy, Université Paris XI, Villejuif, France; 3Department of Clinical Oncology, Leiden University
Medical Center, Leiden, The Netherlands; 4Department of Internal Medicine, Máxima Medisch Centrum, Eindhoven, The Netherlands; 5Urology, Department of Clinical
and Biological Sciences, University of Torino, Azienda Ospedaliero, Universitaria San Luigi, Orbassano, Italy; 6Department of Internal Medicine I, Endocrine Unit,
University Hospital, University of Würzburg, Germany; 7Department of Medical Oncology, Medical School, University of Ioannina, Greece
incidence and epidemiology among different genetic backgrounds, it is below 10% for most
sporadic pheochromocytomas, except in patients with
Two different primary malignancies can arise from the adrenal mutations in the succinate dehydrogenase B (SDHB) gene and/
gland: the adrenocortical carcinoma (ACC) from the adrenal or extra-adrenal locations, among whom more than 30%–50%
cortex and the malignant pheochromocytoma from the adrenal may develop a malignant tumor.
medulla. Both malignancies are extremely rare. ACC has an
estimated incidence of ∼0.5–2 new cases per million people
per year. It follows a bimodal age distribution, with peaks in diagnosis and pathology/molecular
childhood and in the fourth to fifth decades of life. ACC is biology
more frequent in women than in men (ratio 1.5∶1). The
practice
Somatic mutations of the tumor suppressor gene TP53 are
clinical
majority of ACCs are sporadic; however, sometimes these
observed in one-third of ACCs. In addition, allelic losses
malignancies form part of hereditary syndromes such as the
(LOH) at the TP53 locus (17p13) are observed in >85% of
Li-Fraumeni syndrome, Beckwith-Wiedeman syndrome,
ACCs. The insulin-like growth factor II (IGF-II) locus (11p15)
multiple endocrine neoplasia (MEN) 1, congenital adrenal
is imprinted and IGF-II is overexpressed in 90% of ACCs.
hyperplasia, familial polyposis coli, and B-catenin mutations
About one-third of ACCs harbor somatic activating mutations
[1]. Germline p53 mutations without Li-Fraumeni are frequent
of the B-catenin gene [3].
in southern Brazilian children.
The gene encoding subunit B of the SDHB complex is by far
The frequency of sporadic ACC cases is consistently greater
the most important contributor to a hereditary malignant
in the population of patients with adrenal incidentaloma (2%
pheochromocytoma/paraganglioma. In addition, many
in the most conservative series) and, in our experience, the
sporadic metastatic pheochromocytomas/paragangliomas have
proportion of incidentally discovered ACC is increasing.
similar molecular profiles to those of hereditary tumors.
Pheochromocytomas are catecholamine-producing
Inactivation of mutations in SDHB is thought to reduce
neuroendocrine tumors arising from chromaffin cells of the
function of the SDHB complex, causing a pseudohypoxic state
adrenal medulla or extra-adrenal paraganglia. In the latter
and increased expression of angiogenic, growth and mitogenic
setting, the term paraganglioma is preferred. Their incidence is
factors via stabilization of the hypoxia-inducible pathway. If no
∼2–8 per million adults per year [2]. The incidence of
SDHB mutations are identified, genetic testing for VHL,
pheochromocytoma increases to 0.5% in patients with
SDHD, or succinate dehydrogenase C (SDHC) is indicated and
hypertensive symptoms and can be as high as 4% in patients
may provide future insights into the best therapeutic options to
with adrenal incidentalomas. Up to 30% of
be developed for these patients [2]. Rearranged during
pheochromocytomas are associated with a variety of inherited
transfection (RET) should be tested in case of calcitonine
conditions, including MEN2, Von Hippel–Lindau (VHL)
secretion and may be tested when all other gene mutations are
disease, neurofibromatosis type 1, and heredity paraganglioma
negative.
syndromes. Only a few (10%–17%) pheochromocytomas are
A detailed preoperative endocrine assessment is essential to
malignant. Although the likelihood of malignancy varies
establish the origin of the tumor (cortex versus medulla
versusothers). In all cases of an adrenal mass a comprehensive
hormonal analysis is strongly recommended. In 2005, the ACC
L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland. E-mail: clinicalguidelines@esmo.
working group of the European Network for the Study of
org Adrenal Tumors (ENSAT; www.ensat.org) proposed standards
†
for this situation in patients with suspected or established ACC
Approved by the ESMO Guidelines Working Group: June 2012.

by guest
on 05 February 2018
Table 1. Diagnostic work-up for adrenal cancer pheochromocytoma due to the risk of a hypertensive crisis
after infusion of i.v. contrast medium for CT. Although these
Hormonal work-up methods cannot determine the exact entity of the mass, both
Glucocorticoid excess (minimum 3 of 4 tests) are able to correctly diagnose benign tumors in most cases—
Dexamethasone suppression test (1 mg, 23:00 h) when performed according to the state-of-the-art criteria (e.g.
Excretion of free urinary cortisol (24 h urine) Hounsfield units <10 in unenhanced CT, rapid washout in 15-
Basal cortisol (serum) min delayed contrast-enhanced CT or signal intensity loss
Basal ACTH (plasma) using opposed-phase MRI strongly suggests a benign tumor).
Sexual steroids and steroid precursors Most ACCs and malignant pheochromocytomas are
DHEA-S (serum) inhomogeneous with irregular margins and irregular
17-OH-progesterone (serum) enhancement of solid components after the infusion of i.v.
Androstenedione (serum) contrast medium and, in many cases, it is difficult to
Testosterone (serum) distinguish these tumor entities using conventional imaging.
17-beta-estradiol (serum, only in men and postmenopausal women) The detection of local invasion or tumor extension into the
24-h urine steroid metabolite examination
inferior vena cava, as well as lymph node or other
Mineralocorticoid excess
metastases (lung and liver), is important for planning
Potassium (serum)
surgery. F-18/flourodeoxyglucose-positron emission
Aldosterone/renin ratio (only in patients with arterial hypertension
tomography (FDG-PET) is a useful tool for distinguishing
and/or hypokalemia)
potentially malignant lesions from benign tumors in
Catecholamine excess
Normetanephrine, metanephrine, and methoxytyramine (plasma)
radiologically indeterminate adrenal lesions. In patients with
Alternatively: fractionated metanephrine excretion (24 h urine) established pheochromocytoma, FDG-PET appears to be
Imaging superior to metaiodobenzylguanidine (MIBG) and
CT or MRI of abdomen and CT thorax somatostatin- or F-dopamin-based methods, particularly in
Bone scintigraphy (when suspecting skeletal metastases) patients with SDHB mutation and malignant tumors, and
FDG-PET (optional) should be indicated as the preferred method [6] (IV).
MIBG scintigraphy, DOTA-TATE-PET, Dopa/Dopamine PET or Fine needle biopsy of a suspected ACC is almost never
FDG-PET if pheochromocytoma is proved justified because of anticipated tumor spill, and in suspected
pheochromocytoma, it is contraindicated.
Adapted according to the recommendation of the ACC working group of The pathological differential diagnosis of adrenal neoplasias
the European Network for the Study of Adrenal Tumors (www.ensat.org/ is still largely based on morphological features requiring an
acc.htm), May 2005.
experienced pathologist. Several markers have been introduced
In patients with a clearly established diagnosis of an ACC, one can skip
to establish the adrenocortical origin of adrenal masses, with
the workup on catecholamine excess (and conversely for established
steroidogenesis factor-1 immunohistochemistry being
pheochromocytoma, one can skip the steroid analysis). DHEA,
particularly useful [7], although not yet widely available. The
dehydroepiandrosterone.
differential diagnosis between carcinoma and adenoma is
challenging as no single marker indicates malignancy. The
most widely used diagnostic score has been introduced by
(Table 1). It is important to acknowledge that the evidence Weiss et al. [8] and includes the following parameters: mitosis,
level for this proposition is formally low, although the atypical mitosis, necrosis, venous invasion, sinusal invasion,
diagnostic accuracy is high (V, B). Furthermore, the capsular invasion, nuclear atypia, diffuse architecture, and clear
preoperative hormone pattern may serve as a fingerprint of the cell. A score of ≥3 suggests malignancy. Ki67 as a marker of
tumor during follow-up. Whereas the measurement of proliferative activity can also be useful in this respect.
metanephrine and normetanephrine is well established for any For pheochromocytomas, the situation is similarly
pheochromocytoma, plasma and urine methoxytyramine levels demanding. Several histologic features (local invasiveness,
that provide useful information to assess the likelihood of growth pattern, presence of necrosis, cellularity, spindled
malignancy [4]. However, this marker is not yet widely morphology, nuclear pleomorphism and hyperchromasia,
available. mitotic activity, and atypical mitosis) comprise the
Urine steroid metabolomics revealed a pattern of nine Pheochromocytoma Adrenal gland Scaled Score. However,
predominantly immature, early-stage steroidogenesis in ACC there is currently no consensus on the adoption of a formal
that performed well in differentiating adenomas from ACCs. scoring system for these tumors [2]. According to the current
The use of this technique in clinics is, however, premature [5]. World Health Organization (WHO) classification, malignancy
For best patient care, adequate visualization of the tumor is defined by the presence of metastases to a site where
and potential metastases is essential. For differential diagnosis pheochromocytoma/paraganglionic tissue is not normally
of an adrenal mass computed tomography (CT) and magnetic present, e.g. liver or bone, to avoid confusion with multiple
resonance imaging (MRI) are currently considered equally primary tumors. No single histological finding can predict
effective. CT is less expensive and should be recommended as metastatic disease. However, patients with large tumor size (>5
the first choice in a suspected ACC. Conversely, MRI offers a cm), extra-adrenal location, or SDHB mutation have a higher
preferential application for the suspicion of risk of malignancy.
vii | Berruti et al. Volume 23 | Supplement 7 | October 2012

by guest
on 05 February 2018
Table 2. Disease staging system for adrenocortical carcinoma term survival [16] (V, A). To obtain an R0 resection of a
locally advanced ACC, it is often mandatory to resect ( parts
ENSAT TNM Definition of ) adjacent organs such as the wall of the vena cava, liver,
stage spleen, colon, pancreas and/or stomach (V, B). Locoregional
I T1, N0, M0 Tumor ≤5 cm lymphadenectomy improves tumor staging and seems to lead
II T2, N0, M0 Tumor >5 cm to a favorable oncologic outcome [17] (IV, A). No data on the
III T1–T2, N1, M0 Lymph node involvement and/or prognostic role of radical surgery and lymphadenectomy are
T3–T4, N0–N1, M0 tumor infiltration into surrounding available up to now for malignant pheochromocytoma.
tissue and/or a tumor thrombus in For ACC tumors not invading the kidney, concomitant
the vena cava and/or renal vein nephrectomy seems not to improve disease-free and overall
IV T1–T4, N0–N1, M1 Metastatic disease survival and can be avoided, although additional data are
needed (V, B).
In the case of inoperable local infiltrating or metastatic ACC,
surgical excision of the primary tumor and/or metastasis
should be considered in case of an objective response after
The disease stage and margin-free resection are the most neoadjuvant chemotherapy, when a radical resection seems to
important prognostic factors in ACC. In the assessment of the be feasible [16].
disease stage, we recommend the tumor–node–metastasis Cytoreductive (R2) resection in malignant
(TNM) classification proposed by the ENSAT network pheochromocytoma may improve the quality of life and
(Table 2) [9] (IV, A), since this system seems to be superior to survival by reducing the tumor burden and controlling
the staging system published by the Union Internationale hormonal hypersecretion [2] (V, B). It is also indicated in
Contre le Cancer in 2004. Age, mitotic count, the proliferation selected ACC patients with advanced tumors and severe
marker Ki67, and the glucocorticoid excess are additional symptomatic hormone excess when medical therapy is not able
prognostic parameters to refine the prognostic stratification to control endocrine symptoms, and life expectancy is >6
[10]. New molecular markers for aggressiveness and survival months [18] (V, B).
have been recently proposed, but their use in clinics needs Surgery for the recurrence of ACC (local and/or with a low
validation [11]. There is no staging system for malignant tumor metastatic burden) after primary treatment might be
pheochromocytoma and, in contrast to other neuroendocrine effective in improving survival, if R0 resection is achievable and
tumors, no poorly differentiated category is recognized. The the time to first recurrence was >12 months (IV, B).
survival rate depends mainly on the tumor size, primary tumor The hormone hypersecretion can increase the risk of
location (extra-adrenal location is associated with poor perioperative complications. For ACC patients, it is very
prognosis) [12], and the SDHB mutation status (SDHB important to diagnose any adrenocorticotropic hormone
mutations are predictive of poor prognosis) [13]. The overall 5- (ACTH)-independent glucocorticoid excess prior to surgery to
year survival rate is between 34% and 60%; however, adopt measures to prevent a postoperative adrenal crisis or
heterogeneous survival is a critical characteristic of this tumor insufficiency. In all patients with glucocorticoid excess
[12]. A significant number of patients die of hormone-related hydrocortisone must be administered during surgery (e.g. 150
complications (increased blood pressure, constipation) [12]. mg/days) and postoperatively.
In patients with pheochromocytoma and secreting
paraganglioma, exposure to high levels of circulating
management of local regional disease catecholamines during surgery could cause hypertensive crises
Surgery is of utmost importance in the treatment of both ACC and arrhythmias. Therefore, all patients with
and pheochromocytoma. Adrenal surgery should be performed pheochromocytoma or paraganglioma should receive
only in selected centers with >10 adrenalectomies for adrenal preoperative preparation at least 10–14 days before surgery (V,
cancer per year [14] (IV, A). A) [19]. Blood pressure targets for the treatment are <130/80
Open surgery with transperitoneal access is the standard mmHg in the supine position, and systolic blood pressure
treatment of all patients with localized (stage I–II) and local preferably >90 mmHg in the upright position. Traditionally,
advanced stage (stage III) ACCs when complete resection can the non-competitive α-adrenoreceptor antagonist
be achieved. Laparoscopic adrenalectomy is a safe and effective phenoxybenzamine is frequently used. A standard dose is 10
procedure for pheochromocytoma and a selected group of mg twice daily with adjustments every 2–4 days. Alternatively
patients with small ACCs (<8 cm) without preoperative doxazosine, a competitive and selective α1-adrenoreceptor
evidence for invasiveness and adrenal masses (e.g. antagonist might be as effective with fewer side effects. If the
incidentalomas) that are judged as only potentially malignant. target blood pressure is not reached, Ca-antagonists (nifedipine
This technique must be performed only in centers with a slow release) or metyrosine may be used. Blockade of β-
consolidated experience in laparoscopic adrenal surgery, in adrenergic receptors is indicated in patients developing
which principles of oncologic surgical treatment are strictly tachyarrhythmias, but should never be started before the
respected [15] (IV, B). blockade of α-receptors.
The resection status (R0, R1, R2) is a major predictor of Hypertension during surgery may be treated with magnesium
prognosis for ACC. A margin-free complete resection (R0 sulfate, intravenous α-adrenoreceptor antagonist
resection), in fact, provides the only means to achieve long- (phentolamine), calcium antagonist, and/or nitroprusside or
Volume 23 | Supplement 7 | October 2012 doi:10.1093/annonc/mds231 | vii

by guest
on 05 February 2018
nitroglycerine. Tachycardia can be treated with intravenous β-

adrenergic receptor blocker (esmolol). Postoperative hypotension
should be prevented by saline infusion the day before surgery. If
this adverse event occurs, it should be treated aggressively.
Postoperative care should also focus on glucose levels
because hypoglycemia may occur after the reduction of
catecholamine levels.
Most ACC patients have resectable disease at presentation,
however, more than half of the patients who have undergone
complete removal of the tumor are destined to have a relapse,
often with metastases [1]. Similarly, radical resection is not a
guarantee of cure for pheochromocytomas either. In one series
of 171 patients followed up after surgical resection of a
chromaffin cell tumor, 29 patients (17%) had recurrent or new
tumors, which were malignant in 15 (9%) cases [20].
The aggressive behavior and the high recurrence rate of ACC
provide the rationale for the use of adjuvant therapy. Adjuvant
radiotherapy to the tumor bed should be considered in patients
with incomplete/R1 resection or Rx resection [21] (V, A).
Mitotane has been the reference drug for the management of
ACC for decades. In a case control study involving 177 patients,
the outcome of 47 patients followed in Italian reference centers
that systematically adopted adjuvant mitotane to all radically
operated ACC was significantly higher (in terms of both disease-
free survival and overall survival) than the outcome of 55 Italian
patients and 75 German patients followed in institutions not
administering adjuvant mitotane therapy [22].
Although these data cannot be considered conclusive, they
suggest that adjuvant mitotane can delay and possibly prevent
a recurrence of disease (V, B). On these bases, a panel of
international experts unanimously stated that patients with Figure 1 Algorithm on management according to stage, risk factors, and
potential residual disease (R1 or Rx resection) and/or Ki67 disease characteristics for adrenocortical carcinoma (ACC) (A) and
more than 10% should be offered adjuvant mitotane, whereas pheochromocytoma (B). *Low-risk ACC is defined stage I–II and Ki67
adjuvant therapy was not considered mandatory in patients expression in ≤10% of neoplastic cells, high-risk ACC: stage III or Ki67
fulfilling all of the following criteria: stage I or II disease, expression in >10% of neoplastic cells.
histologically proven R0 resection; and Ki67 expressed in
≤10% of neoplastic cells [23] (V, B) (Figure 1A). For these
patients participation in the first prospective multinational involved in the follow-up of patients during mitotane
randomized trial, testing the efficacy of adjuvant mitotane treatment.
therapy (www.adiuvo-trial.org) is recommended whenever There are no data regarding the optimal duration of
possible. Mitotane is a difficult drug to manage, with a long adjuvant mitotane; we recommend that adjuvant mitotane
half-life, dose-limiting toxicity, and a narrow therapeutic should be administered for at least 2 years since the greatest
window. Although the evidence is limited, the panel frequency of disease recurrence is expected within this
recommends that the mitotane dosing should be guided by timeframe. It is noteworthy that while mitotane is well
plasma measurements, aiming at a concentration between 14 tolerated by a fraction of patients, the majority of patients find
and 20 mg/ml since there is less antitumor activity <14 mg/ml mitotane a difficult therapy that markedly impacts the quality
and the risk of toxicity (gastrointestinal and neurologic) is of their lives. So the treatment duration should be assessed
higher above 20 mg/ml (V, A) [24]. A suggested mitotane individually, taking carefully into account the cost/benefit ratio.
regimen is depicted in Table 3. Several blood analytes should
be monitored during mitotane therapy. Due to the adrenolytic management of advanced/metastatic
activity of mitotane, all patients must receive concomitant
administration of glucocorticoids to cover adrenal insufficiency.
disease
Owing to an increased metabolic clearance rate of Prognosis in locally advanced inoperable and metastatic ACC
glucocorticoids by mitotane therapy, high-dose glucocorticoid patients is poor, the 5-year overall survival being <15%.
replacement is needed. Mineralocorticoid supplementation is However, recent studies have highlighted a greater
necessary only in a subset of patients (Table 3). Testosterone heterogeneity of advanced ACC than previously thought.
and thyroxine supplementation may be also required. Mitotane Extremely long survival has been reported in patients with
can also induce a wide spectrum of hormone and metabolic resectable oligometastatic disease with considerable intervals
derangement and so an experienced endocrinologist should be among recurrences [10] (V, B).
vii | Berruti et al. Volume 23 | Supplement 7 | October 2012

by guest
on 05 February 2018
Table 3. Mitotane dose regimen, glucocorticoid, and mineralocorticoid supplementation, blood level monitoring, and dose adjustment according to toxicity
and blood level monitoring
Mitotane dose regimena • Start with 1.5 g/d and increase dose within 4–6 days to 6 g/days
• After 3 weeks, adjust dosage according tolerability and blood level (see below)
• Maximum dose 12 g/days, but most patients do not tolerate >8 g/days
• Target mitotane blood level 14–20 mg/l. Using this regimen, ∼50% of patients achieve the target level
within 3 months
Glucocorticoid and mineralocorticoid • A total daily dose of 50 mg hydrocortisone (divided as 20–20–10 mg) or 75 mg cortisone acetate and
supplementation more may be needed. Glucocorticoid replacement is monitored best with careful clinical assessment
• Fludrocortisone may be added depending on the blood pressure, serum potassium levels, and plasma
renin activity
Recommended blood monitoring • Mitotane serum levels every 2–3 weeks in the first 3 months. After reaching a plateau, the interval can
during mitotane therapy be extended (i.e. every 6 weeks)
• Glutamate-Oxaloacetate Transaminase (GOT), Glutamate-Pyruvate Transaminase (GPT), bilirubin,
Gamma-Glutamyl-Transferase (GGT). Initially every 4 weeks, after 6 months every 8 weeks. GGT is
invariably elevated without clinical consequences. If other liver enzymes are rapidly increasing (>3-fold
of baseline), there is a risk of liver failure: stop mitotane
• TSH, fT3, fT4 every 3–4 months. Thyroid hormone replacement is recommended in patients with
clinical symptoms of hypothyroidism
• Testosterone, free testosterone, and sexual hormone binding globulin (SHBG) should be tested in male
patients with symptoms of hypogonadism
• Renin every 3 months. If renin increases in the presence of symptoms suggestive of mineralocorticoid
deficiency, fludrocortisones should be added
• Cholesterol (High-Density Lipoprotein (HDL), Low-Density Lipoprotein (LDL)), triglycerides every 3–
4 months (in an adjuvant setting). If LDL/HDL cholesterol consistently increases, consider treatment
with statins not metabolized by CYP3A4 (e.g. pravastatin, rosuvastatin)
• Blood count every 3–4 months
Plasma mitotane level CNS (grade 2)/GI side effects (grade 3/4) Grade 3/4 CNS side effects
Absent Present Present
<14 mg/l Increase daily dose by 1 gb Reduce daily dose by 1 g Stop mitotanec
14–20 mg/l Maintain dose Reduce daily dose by 1.5 g Stop mitotanec
>20 mg/l Reduce daily dose to 50%–75% of the most Stop mitotanec Stop mitotanec
recent dose
Recommended dose adjustment according to the central nervous system (CNS)/gastrointestinal (GI) side effects and plasma mitotane level. HDL, high-
density lipoprotein; LDL, low-density lipoprotein; GOT, glutamate-oxaloacetate transaminase; GPT, glutamate-pyruvate transaminase; GGT, gamma-
glutamyl-transferase.
a
An alternative low-dose regimen is also available with potentially similar efficacy.
b
up to the maximum tolerated dose.
c
Until symptom resolution (grade 0 or 1)
Mitotane is the only drug approved in locally advanced of patients with a low tumor burden and/or more indolent
inoperable and metastatic patients although randomized, disease (Figure 1A). The combination with locoregional
controlled prospective trials are lacking (IV, A). Response therapies such as radiofrequency ablation (RFA) is
rates vary between 13% and 35%, but many of these results recommended. In case of rapidly progressing or life-
were derived from retrospective series since the 1960s with threatening extensive metastatic disease and/or radiological
variability in the response criteria [1, 10]. Patients with progression under mitotane, cytotoxic chemotherapy is
long-term maintenance of mitotane levels in the therapeutic indicated. Over the last 15 years, only 11 prospective single-
range may obtain a survival benefit [25]. Patients with arm chemotherapy studies with a total of 239 patients with or
hormonal excess often experience clinical benefit of this without mitotane have been published. Response rates vary
strategy and continuation of mitotane treatment can be between 7% and 54%, again with variability in the response
indicated in these patients even after radiological progression criteria. The association of mitotane to chemotherapy seems to
when alternative strategies to inhibit the hormonal excess are be more active than chemotherapy alone [1], although no
lacking. randomized trials have formally demonstrated this superiority.
Owing to the latency of mitotane to attain the therapeutic In the First International Randomized trial in advanced or
range, mitotane monotherapy is indicated in the management Metastatic Adrenocortical Carcinoma Treatment (FIRM-ACT
Volume 23 | Supplement 7 | October 2012 doi:10.1093/annonc/mds231 | vii

by guest
on 05 February 2018
trial), the two most active treatment regimens namely radiofrequency ablation. Radionuclide therapy is an effective
Etoposide, Doxorubicin, Cisplatin, and Mitotane (EDP-M) and treatment and 131I-MIBG, in activities ranging from 5.5 to 38
Streptozotocin and Mitotane (Sz-M) were compared with 304 GBq (150–1000 mCi), is the most frequently used.
chemotherapy-naïve patients. Patients with disease progression Approximately 50% of patients are eligible for 131I-MIBG
received the alternate regimen. The results of this trial, recently therapy based on the uptake on diagnostic scans. Several
published [26], indicate to us that EDP-M is the superior studies have been published on the efficacy of 131I-MIBG
regimen. treatment [2]; most are retrospective, and only one is a
Although no statistically substantial increase in overall prospective phase II study [30]. Objective responses were
survival was documented in patients receiving EDP-M as the observed in 22–47% of cases. Long-term survival of responders
first-line therapy, significantly better response rates and of 4.7 years or 72 months was reported but progression at
progression-free survival rates were achieved with EDP-M in study entry was not a prerequisite of most studies. Objective
comparison with Sz-M. The rate of serious adverse events was responses were mainly observed in patients with soft tissue
comparable. Of note, the results of the second-line regimens metastases. Grade 3–4 toxicity was reported in 16%–83% of
replicated the rates observed with the first-line therapy. Since patients.
EDP-M was superior to Sz-M in terms of progression-free 131
I-MIBG therapy should be considered as a first-line
survival either as first-line or second-line therapy, the crossover approach in patients with a good uptake of 123I-MIBG and
design may have attenuated its advantage on overall survival. unresectable, progressive pheochromocytoma/paraganglioma
On these bases, we recommend EDP-M as the first-line or symptomatic patients (not amenable to locoregional
therapy for ACC requiring cytotoxic therapy, and as the control), or patients with a high tumor burden with a low
reference for new therapies (I, A; Figure 1A). number of bone metastases. More recently peptide-
In patients unfit for the EDP-M regimen P-M may radiolabeled radiotherapy has also been developed.
constitute a reasonable alternative [27]. However, the FIRM- Cyclophosphamide- and dacarbazine-based regimens
ACT trial with a median overall survival between 12 and 14.8 combined with vincristine (CVD) or doxorubicin (CVDD or
months clearly indicated that new systemic therapy options are CDD) are the best studied chemotherapy regimens [2]. In the
urgently needed, but thus far positive results are lacking largest published study to date (n = 52 patients), 40% of
despite an increasing number of prospective studies, also patients treated with CVD, CDD, or CVDD experienced
incorporating modern targeted agents. Results of studies with clinical benefits, including a reduction in tumor size in 25% of
oral CYP3A4-mediated drugs may have been hampered by cases [31]. Systemic chemotherapy is debated as a first-line
subtherapeutic systemic exposure due to CYP3A4 induction by therapy in patients with a low uptake of 123I-MIBG and
mitotane [28, 29]. A multicenter prospective randomized, unresectable, rapidly progressive pheochromocytoma/
placebo-controlled clinical trial aimed to test the efficacy of paraganglioma, or patients with high tumor burden or with a
OSI 906–301, an IGR inhibitor, as second-/third-line approach high number of bone metastases. As in well-differentiated
in ACC patients has recently completed patient accrual. pancreatic neuroendocrine tumors, there is a strong rationale
However, the results of this study will not be available before and some first evidence on the potential efficacy of anti-
2013. angiogenic drugs in malignant pheochromocytomas. Thus, the
In case of painful metastasis, palliative radiotherapy is an European ENSAT network is currently launching a
option, particularly in bone lesions (IV, B). Arterial randomized, placebo-controlled trial testing sunitinib in
chemoembolization and radiofrequency ablation may be patients with malignant pheochromocytoma and
beneficial in selected patients (V, C). paraganglioma (FIRST-MAPPP trial).
The therapeutic strategy of metastatic pheochromocytoma/
paraganglioma aims to control excessive catecholamine
secretion and tumor burden, but no curative treatment is
follow-up
achievable. Treatment choices include a wait and see policy, International recommendations on follow-up are lacking both
locoregional therapies, systemic chemotherapy, and for ACC and pheochromocytoma/paraganglioma, and this is a
radiopharmaceutical agents (Figure 1B) [2], and they should be significant problem. The recommendations in these guidelines
discussed case by case in a multidisciplinary specialized setting. were formulated based on data available on the natural history
In the absence of any randomized studies and demonstrated of these diseases, personal experience, and consensus among
impact on survival, the patients’ quality of life should always be panelists.
considered. Indeed, due to the indolent course of subgroups of For patients with ACC after complete resection, we
patients, a wait and see policy coupled with a watchful follow- recommend regular follow-up every 3 months including
up can be considered as an option in asymptomatic patients abdominal CT (or MRI), thoracic CT, and monitoring of
with a low tumor burden. In these patients, an antineoplastic initially elevated steroids (V, B). After 2 years, intervals may be
treatment should be recommended in case of rapid progression gradually increased. In case of long-term persistence of the
and/or symptom onset. In the absence of tumor progression, disease-free status, follow-up should be continued for at least
surgery of the primary tumor or metastases can reduce 10 years.
hormone secretion and may prevent complications related to a For locally advanced or metastatic disease, overall survival
critical anatomical location and improve the efficacy of and time to progression (TTP) are the most important
subsequent therapies [2]. Metastatic disease palliation may also endpoints with the response rate (RR) as a secondary end
benefit from local therapy with embolization and or point. The TTP and RR guide the clinical decision-making in
vii | Berruti et al. Volume 23 | Supplement 7 | October 2012

by guest
on 05 February 2018
Table 4. Summary of Recommendations malignant recurrence and require long-term clinical
(adrenergic symptoms and blood pressure levels) and
• ACC is defined by a Weiss score of 3 or more. Malignant biochemical ( plasma or urinary metanephrine,
pheochromocytomas/paragangliomas are defined by the presence of normetanephrine, chromogranin A, and methoxythyramine)
metastasis. follow-up [2]. The follow-up is especially important for
• Patients suspected to harbor primary adrenal tumors should undergo a patients with extra-adrenal primary disease, tumor size >5 cm,
standardized diagnostic work-up consisting of endocrine assessment for or SDHB mutations. Biochemical testing is repeated ∼14 days
excess hormone production and modern imaging (CT/MRI of abdomen, following surgery to check for remaining disease and thereafter
chest CT, and in selected cases supplemented by isotope functional every 3–4 months for 2–3 years. This should subsequently be
imaging mainly FDG-PET). The diagnostic work-up differs between repeated every 6 months. Patients with new events (high blood
ACC and pheochromocytoma. pressure, adrenergic symptoms, or pain) or pathological
• Guided biopsies of potentially resectable primary adrenal tumors are not endocrine tests and/or elevated circulating chromogranin A
informative in most cases, but are potentially harmful and should be should undergo imaging that includes thorax and abdomen CT
avoided. and best functioning imaging (PET FDG in most cases). In
• The ENSAT TNM staging system should be used for ACC staging.
case of proven malignant disease, SDHB mutation, extra-
• Histological diagnosis should be done by an experienced pathologist and
adrenal primary disease, and in rare cases of
should rely on morphological, mitotic, and immunohistochemical
pheochromocytoma/paraganglioma without relevant
parameters.
preoperative hormone secretion, imaging could be repeated at
• Complete surgical extirpation of localized and locally advanced ACC or
least every 6 months during the first year and yearly afterward,
pheochromocytoma (R0 resection) is the mainstay of potentially curative
approaches. Additionally, a locoregional lymphadenectomy is suggested
irrespective of the negative results of biochemical tests. In these
for ACC. patients, lifelong follow-up is recommended [2].
• In pheochromocytoma cytoreductive surgery might be considered. In
advanced ACC, this approach is only reasonable for patients with severe conflict of interest
hormone excess.
• Meticulous perioperative management of hormonal, glucose, electrolytes, Dr. Berruti has reported: advisory board honorarium from
cardiac and fluid/blood pressure abnormalities is a critical component of Astellas. Prof. Fassnacht has reported Principal Investigator of
patient care. a study funded by HRA Pharma, Research grants: Pfizer,
• Despite the limited literature evidence, adjuvant systemic mitotane is Astellas Pharma.
recommended for patients with ACC and incomplete resection (R1, Rx The other authors have reported no potential conflicts of
stage III) or in the presence of high-risk features (Ki67>10%). R1 and Rx interest.
ACC resections may be followed by additional adjuvant radiotherapy to
the tumor bed.
• Fit patients with inoperable ACC, high tumor volume and rapid disease references
progression should be treated with combination cytotoxic chemotherapy 1. Fassnacht M, Libé R, Kroiss M et al. Adrenocortical carcinoma: a clinician’s
plus mitotane (EDP-M). Less fit patients and/or patients with low tumor update. Nat Rev Endocrinol 2011; 7: 323–335.
burden and slow progression can (first) be managed with mitotane 2. Pacak K, Eisenhofer G, Ahlman H et al. International symposium on
monotherapy combined or not with locoregional options. pheochromocytoma. Pheochromocytoma: recommendations for clinical practice
• Disease and symptom control is the main treatment goal for patients from the first international symposium. October 2005. Nat Clin Pract Endocrinol
with inoperable pheochromocytoma and can be attempted by Me tab 2007; 3: 92–102.
radiopharmaceuticals (131I-MIBG), locoregional ablative procedures, 3. Bertherat J, Bertagna X. Pathogenesis of adrenocortical cancer. Best Pract Res
and/or combination chemotherapy (CVD) in selected cases. Clin Endocrinol Metab 2009; 23: 261–271.
• Wait and see policy is recommended in low tumor burden and 4. Esenhofer G, Lenders JW, Siegert G et al. Plasma methoxytyramine: a novel
asymptomatic malignant pheochromocytoma and paraganglioma. biomarker of metastatic pheochromocytoma and paraganglioma in relation to
established risk factors of tumour size, location and SDHB mutation status. Eur J
• Patients with resected ACC or pheochromocytoma should be followed at
Cancer 2012; 48: 1739–1749.
regular intervals with clinical, imaging and biochemical screens for at
5. Arlt W, Biehl M, Taylor AE et al. Urine steroid metabolomics as a biomarker tool
least 10 years. Lifelong surveillance with an increased interval of time is
for detecting malignancy in adrenal tumors. J Clin Endocrinol Metab 2011; 96:
favored in malignant pheochromocytoma/paraganglioma. 3775–3784.
• The follow-up of patients with inoperable disease should be performed 6. Timmers HJ, Kozupa A, Chen CC et al. Superiority of fluorodeoxyglucose positron
every 2–4 months for ACC and every 3–6 months for emission tomography to other functional imaging techniques in the evaluation of
pheochromocytoma/paraganglioma during the first year of follow-up and metastatic SDHB-associated pheochromocytoma and paraganglioma. J Clin
then adjusted. Oncol 2007; 25: 2262–2269.
7. Sbiera S, Schmull S, Assie G et al. High diagnostic and prognostic value of
steroidogenic factor-1 expression in adrenal tumors. J Clin Endocrinol Metab
2010; 95: E161–E171.
8. Weiss LM, Medeiros LJ, Vickery AL, Jr. Pathologic features of prognostic
an individual patient and should be evaluated at regular significance in adrenocortical carcinoma. Am J Surg Pathol 1989; 13: 202–206.
9. Fassnacht M, Johanssen S, Quinkler M et al. German adrenocortical carcinoma
intervals (every 12 weeks or less depending on the therapy)
registry group; European Network for the Study of Adrenal Tumors. Limited
using CT scans. The role of PET scans is not yet clear in ACC. prognostic value of the 2004 International Union Against Cancer staging
Patients who underwent successful surgery for non- classification for adrenocortical carcinoma: proposal for a Revised TNM
metastatic pheochromocytoma/paraganglioma are at risk of Classification. Cancer 2009; 115: 243–250.
Volume 23 | Supplement 7 | October 2012 doi:10.1093/annonc/mds231 | vii

by guest
on 05 February 2018
10. Baudin E, Leboulleux S, Al Ghuzlan A et al. Therapeutic management of 21. Polat B, Fassnacht M, Pfreundner L et al. Radiotherapy in adrenocortical
advanced adrenocortical carcinoma: what do we know in 2011? Horm Cancer carcinoma. Cancer 2009; 115: 2816–2823.
2011; 6: 363–371. 22. Terzolo M, Angeli A, Fassnacht M et al. Adjuvant mitotane treatment for
11. de Reyniès A, Assié G, Rickman DS et al. Gene expression profiling reveals a adrenocortical carcinoma. N Engl J Med 2007; 356: 2372–2380.
new classification of adrenocortical tumors and identifies molecular predictors of 23. Berruti A, Fassnacht M, Baudin E et al. Adjuvant therapy in patients with
malignancy and survival. J Clin Oncol 2009; 27: 1108–1115. adrenocortical carcinoma: a position of an international panel. J Clin Oncol
12. Ayala-Ramirez M, Feng L, Johnson MM et al. Clinical risk factors for malignancy 2010; 28: e401–e402.
and overall survival in patients with pheochromocytomas and sympathetic 24. Haak HR, Hermans J, van de Velde CJ et al. Optimal treatment of adrenocortical
paragangliomas: primary tumor size and primary tumor location as prognostic carcinoma with mitotane: results in a consecutive series of 96 patients. Br J
indicators. J Clin Endocrinol Metab 2011; 96: 717–725. Cancer 1994; 69: 947–951.
13. Amar L, Baudin E, Burnichon N et al. Succinate dehydrogenase B gene 25. Hermsen IG, Fassnacht M, Terzolo M et al. Plasma concentrations of o,p’DDD, o,
mutations predict survival in patients with malignant pheochromocytomas or p’DDA, and o,p’DDE as predictors of tumor response to mitotane in
paragangliomas. J Clin Endocrinol Metab 2007; 92: 3822–3828. adrenocortical carcinoma: results of a retrospective ENS@T multicenter study. J
14. Lombardi CP, Raffaelli M, Boniardi M et al. Adrenocortical carcinoma: effect of Clin Endocrinol Metab 2011; 96: 1844–1851.
hospital volume on patient outcome. Langenbecks Arch Surg 2012; 397: 201–207. 26. Fassnacht M, Terzolo M, Allolio B et al. Combination chemotherapy in advanced
15. Porpiglia F, Fiori C, Daffara F et al. Retrospective evaluation of the outcome of adrenocortical carcinoma. N Engl J Med 2012; 366: 2189–2197.
open versus laparoscopic adrenalectomy for stage I and II adrenocortical cancer. 27. Bukowski RM, Wolfe M, Levine HS et al. Phase II trial of mitotane and cisplatin in
Eur Urol 2010; 57: 873–878. patients with adrenal carcinoma: a Southwest Oncology Group study. J Clin
16. Icard P, Goudet P, Charpenay C et al. Adrenocortical carcinomas: surgical trends Oncol 1993; 11: 161–165.
and results of a 253-patient series from the French Association of Endocrine 28. van Erp NP, Guchelaar HJ, Ploeger BA et al. Mitotane has a strong and a
Surgeons study group. World J Surg 2001; 25: 891–897. durable inducing effect on CYP3A4 activity. Eur J Endocrinol 2011; 164:
17. Reibetanz J, Jurowich C, Erdogan I et al. Impact of lymphadenectomy on the 621–626.
oncologic outcome of patients with adrenocortical carcinoma. Ann Surg 2012; 29. Ayala-Ramirez M, Feng L, Habra MA et al. Clinical benefits of systemic
255: 363–369. chemotherapy for patients with metastatic pheochromocytomas or sympathetic
18. Schteingart DE, Doherty GM, Gauger PG et al. Management of patients with 675 extra-adrenal paragangliomas: insights from the largest single-institutional
adrenal cancer: recommendations of an international consensus conference. experience. Cancer 2012; 118: 2804–2812.
Endocr Relat Cancer 2005; 12: 667–680. 30. Kroiss M, Quinkler M, Lutz WK et al. Drug interactions with mitotane by induction
19. Pacak K. Preoperative management of the pheochromocytoma patient. J Clin of CYP3A4 metabolism in the clinical management of adrenocortical carcinoma.
Endocrinol Metab 2007; 92: 4069–4079. 665 Clin Endocrinol (Oxf ) 2011; 75(5): 585–591.
20. Amar L, Servais A, Gimenez-Roqueplo AP et al. Year of diagnosis, features 31. Gonias S, Goldsby R, Matthay KK et al. Phase II study of high-dose [131I]
at presentation, and risk of recurrence in patients with pheochromocytoma metaiodobenzylguanidine therapy for patients with metastatic pheochromocytoma
or secreting paraganglioma. J Clin Endocrinol Metab 2005; 90: 2110–2116. and paraganglioma. J Clin Oncol 2009; 27: 4162–4168.
vii | Berruti et al. Volume 23 | Supplement 7 | October 2012

by guest
on 05 February 2018
Cutaneous melanoma: ESMO Clinical Practice

R. Dummer1, A. Hauschild2, N. Lindenblatt3, G. Pentheroudakis4 & U. Keilholz5, on behalf
of the ESMO Guidelines Committee*
1
Department of Dermatology, University Hospital Zürich, Zürich, Switzerland; 2Department of Dermatology, University Hospital Schleswig-Holstein, Kiel, Germany;
3
Division of Plastic and Reconstructive Surgery, University Hospital Zürich, Zürich, Switzerland; 4Ioannina University Hospital, Ioannina, Greece; 5Charité Comprehensive
Cancer Center, Charité-Universitätsmedizin, Berlin, Germany
incidence and epidemiology The histology report should follow the American Joint
Committee on Cancer (AJCC) classification [8], and include: in-
The European incidence of malignant melanoma varies from 3 to formation on the maximum thickness in millimetres (Breslow),
5/100 000/year in Mediterranean countries to 12–25 (and rising) information on mitotic rate in case of a tumour thickness below
in Nordic countries. Increased ultraviolet (UV) light exposure of 1 mm, presence of ulceration, presence and extent of regression
a genetically predisposed population seems to be, at least in part, and clearance of the surgical margins [II, A]. In addition, infor-
responsible for an ongoing increase in incidence with signs of sta- mation on anatomical site (including extra-cutaneous sites, such
bilisation of mortality over recent decades, except in elderly males as mucosa, conjunctiva) and degree of sun damage is necessary.
[1]. There is a disparity in the mortality-to-incidence ratios It should also include the melanoma type (superficial spreading
clinical practice
between Western and Eastern European countries [2], implying a melanoma, lentigo maligna melanoma, acral lentiginous melan-
guidelines
need to improve prevention, especially in Eastern Europe. oma, nodular melanoma and others). In rare situations, melano-
UV irradiation was identified as a major carcinogen involved mas may derive from dermal melanocytes (melanoma arising
in melanoma genesis. Prevention of UV exposure, including the from giant congenital naevus, malignant blue naevus) [9].
regular use of sunscreen, has been shown to diminish the Superficial spreading and nodular melanomas present a higher
incidence of primary cutaneous melanomas in an Australian frequency of BRAF and NRAS mutations than other melanoma
population [3]. types [10]. Acral lentiginous melanoma and mucosal melanomas
of the genital region have a certain probability to present c-Kit
diagnosis mutations [11].
Mutation testing for treatable mutations is mandatory in
Suspicious lesions are characterised by Asymmetry, Border patients with advanced disease (unresectable stage III or stage IV,
irregularities, Colour heterogeneity, Dynamics, (dynamics or evo- and highly recommended in high-risk resected disease stage IIc,
lution in colours, elevation or size) (‘ABCD rule’) [4]. Today, stage IIIb–IIIc) [V, A]. If the tumour is BRAF-wild type, testing
many primary melanomas have a diameter of <5 mm [5]. for NRAS mutations c-kit mutation should be considered [V, A].
The ugly duckling ‘concept’ [6] helps to identify melanomas, Mutational testing of primary tumours without metastases
because naevi in the same individual tend to resemble one is not recommended. Mutation analysis must be carried out in
another and melanomas often do not fit the individual’s naevus accredited (certified) institutes that have careful quality controls.
pattern.
Dermoscopy by an experienced physician enhances the diag-
nostic accuracy [II, B] [7]. An automated video-dermoscopy
system can provide improved diagnostic accuracy for patients Physical examination with special attention to other suspicious
with multiple atypical naevi in the follow-up. pigmented lesions, tumour satellites, in-transit metastases, re-
Diagnosis should be based on a full-thickness excisional biopsy gional lymph node (LN) and systemic metastases is mandatory.
with a minimal side margin. Processing by an experienced path- In low-risk melanomas (pT1a) no other investigations are neces-
ology institute is mandatory. sary. In higher tumour stages (pT1b–pT3a), imaging (ultrasound
for locoregional LN metastasis) and, in pT stages >pT3a, computed
tomography (CT) or positron emission tomography (PET) scans
CH-6962 Viganello-Lugano, Switzerland. are recommended before surgical treatment and sentinel node
E-mail: clinicalguidelines@esmo.org biopsy [III, C].
†
The refined version of the AJCC staging and classification
Approved by the ESMO Guidelines Committee: February 2002, last update July 2015.
This publication supersedes the previously published version—Ann Oncol. 2012; 23 system, which includes sentinel node staging, is the only inter-
(Suppl. 7): vii86–vii91 nationally accepted classification system [8, 12] (Table 1).

by guest
on 05 February 2018
Table 1. AJCC staging system of melanoma
T classification Thickness (mm) Ulceration status/mitosis
T1 ≤1.0 a: without ulceration and mitosis <1/mm²

b: with ulceration or mitoses ≥1/mm²
T2 1.01–2.0 a: without ulceration
b: with ulceration
T3 2.01–4.0 a: without ulceration
b: with ulceration
T4 >4.0 a: without ulceration
b: with ulceration
N classification No. of metastatic nodes Nodal metastatic mass
N0 0 N/A
N1 1 node a: micrometastasisa
b: macrometastasisb
N2 2–3 nodes a: micrometastasisa
b: macrometastasisb
c: in transit metastases/satellites ‘without’ metastatic
nodes
N3 4 or more metastatic nodes, or matted nodes, or in
transit metastases/satellites ‘with’ metastatic nodes
M classification Site Serum LDH
M0 No distant metastasis N/A

M1a Distant skin, subcutaneous, or nodal metastases Normal
M1b Lung metastases Normal
M1c All other visceral metastases Normal
Any distant metastasis Elevated
Reprinted with permission. © 2009 American Society of Clinical Oncology. All rights reserved.
a
Micrometastases are diagnosed after sentinel lymph node biopsy and completion lymphadenectomy (if carried out).
b
Macrometastases are defined as clinically detectable nodal metastases confirmed by therapeutic lymphadenectomy or when nodal metastasis exhibits gross
extracapsular extension.
AJCC, American Joint Committee on Cancer; N/A, not applicable; LDH, lactate dehydrogenase.
treatment of localised disease investigated adjuvant treatment with low, intermediate and high
doses of interferon-α (IFN-α) [16, 17].
Wide excision of primary tumours with safety margins of 0.5 cm A meta-analysis of 14 randomised, controlled trials, investigat-
for in situ melanomas, 1 cm for tumours with a thickness of up ing adjuvant IFN therapy involving 8122 patients, showed statis-
to 2 mm, and 2 cm for thicker tumours, is recommended [13] tically significant absolute improvement in both disease-free
[II, B]. Modifications, with reduced safety margins, are acceptable
survival [hazard ratio (HR) 0.82] and OS (HR 0.89), with no clear
for preservation of function in acral and facial melanomas and
indication to recommend a certain dose or treatment duration
should be carried out with micrographic surgery.
[18]. Since pegylated IFN-α (PegIFN-α) is suitable for long-term
Elective lymphadenectomy or irradiation to the regional LNs
therapy, the European Organisation for Research and Treatment
should not be carried out routinely [II, B].
of Cancer (EORTC) has initiated a large prospective randomised
Sentinel LN biopsy in melanoma with a tumour thickness of
>1 mm and >0.75 mm and additional risk factors such as ulcer- trial to investigate the protective effect of PegIFN-α-2b in the ad-
ation or mitotic rate ( pT1b) are recommended for precise juvant setting [19]. A total of 1256 patients with resected stage III
staging [II, B] [14]. A complete lymphadenectomy of regional melanoma were randomised to receive observation or PegIFN-α
LNs must be discussed with the patient, if the sentinel node was therapy [19]. Randomisation was stratified for microscopic (N1)
found positive for metastases [III, C]. However, this procedure versus macroscopic (N2) nodal involvement, number of positive
offers just a relapse-free survival (RFS) benefit without proven nodes, ulceration and tumour thickness. The IFN group received
effect on overall survival (OS) [15]. Sentinel LN biopsy should an induction-IFN-weekly dose of 6 μg/kg for the first 8 weeks,
be carried out only in experienced centres. and the dose was then reduced to 3 μg/kg per week for 5 years
Many well-designed clinical trials have investigated the [19]. At 3.8 years of median follow-up, RFS was significantly
impact of adjuvant therapy in patients with high-risk primary improved by 18% in the PegIFN-α-2b arm, compared with obser-
melanoma (stage IIB/C) or completely resected LN metastases vation; the 4-year RFS rate was 45.6% versus 38.9%. OS was un-
(stage III) [6]. A number of prospective randomised trials have changed in the two groups. In stage III–N1a (micrometastases
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv297 | v

by guest
on 05 February 2018
detected in the sentinel node), both RFS and distant metastases- [4] [III, B]. In unresectable in-transit cases, other locoregional
free survival (DMFS) were prolonged in the PegIFNα-2b arm, approaches, such as electrochemotherapy [29] or intralesional
whereas in stage III-N1b (macroscopic metastases), there was therapy with replicating herpes virus [Talimogene laherparepvec
no benefit [19]. An update of this trial, with median follow-up of (T-Vec)] [30], preferentially in the context of a clinical trial,
7.6 years, has shown that IFN therapy had a significant impact on should be considered.
RFS, DMFS and OS (HR 0.59 = 0.006) in a sub-population of Surgical removal or stereotactic irradiation therapy might be
patients with micrometastases and primary ulcerated melanomas curative in a few patients and is recommended in the case of a
[20]. Therefore, while awaiting the results of prospective rando- single metastasis in parenchymal organs, including the central
mised trials, in this patient population, PegIFN-α can be recom- nervous system.
mended if the individual patient tolerates it well [II, B]. Adjuvant Non-resectable in-transit metastases or inoperable primary
treatment in patients with resected macroscopic node involve- tumours of the limbs, without additional metastases, may be
ment is preferentially applied in the context of randomised clinic- treated with isolated limb perfusion using melphalan and/or
al trials in specialised centres. However, high-dose IFN-α-2b is tumour necrosis factor-α [III, C]. Such treatment requires major
an approved indication for this therapeutic situation. A meta- surgery and should be restricted to centres of excellence. Radiation
analysis on adjuvant therapy of melanoma with IFNs, however, therapy, electrochemotherapy [29] or intralesional therapy, with
did not demonstrate an improved efficacy of high-dose IFN com- replicating T-Vec [30], may also be used [V, D], [27, 28].
pared with low- or intermediate-dose IFNs [18].
Long-term therapy with ipilimumab, an antibody blocking treatment of systemic metastatic disease
CTLA4 and thus activating T lymphocytes to mount an immune
response against tumour cells, has improved RFS (HR 0.75;
(stage IV)
median RFS 26.1 versus 17.1 months, with 3-year RFS rates of New therapeutic strategies, such as immunotherapy, that utilise
46.5% versus 34.8%, P = 0.0013) in the adjuvant setting also for antibodies that bind to checkpoint inhibitors of T-cell activation,
N1b and higher stages. However, the treatment with a dose of 10 have demonstrated impressive efficacy. CTLA-4 blocking agents
mg/kg every 3 weeks for 4 doses, then every 3 months for up to 3 like ipilimumab, the anti PD-1 antibodies, such as nivolumab
years, was associated with a number of severe and long-lasting and pembrolizumab, as well as selective BRAF inhibitors, such as
adverse reactions including colitis and endocrinopathies. vemurafenib, encorafenib and dabrafenib (used alone and/or in
Therefore, additional trials are mandatory and the respective combination with MEK inhibitors like binimetinib, cobimetinib
patient population should be referred to centres offering trial and trametinib [31, 32]), have demonstrated impressive anti-
participation [21]. tumour activity [33–39]. Therefore, immunotherapy and kinase
Adjuvant chemotherapy, mistletoe extracts, viscum album and inhibitors are the backbone of systemic therapy. Chemotherapy is
hormone therapies are not beneficial at all [22]. Adjuvant therapy considered a second-line or bridging treatment option.
with other cytokines including interleukin-2, tumour vaccination, Tumour tissues, preferentially of metastatic lesions, should be
immunochemotherapy and BRAF inhibitors is experimental and screened for mutations of BRAF V600. If that is negative,
not to be used outside controlled clinical trials. The application of further molecular testing can be carried out for NRAS, c-Kit
BRAF inhibitors is associated with cutaneous neoplasms such as (mucosal and acrolentigenous primaries) GNA11 or GNAQ
keratoacanthomas, squamous cell carcinomas and melanomas (uveal primary); this helps to direct patients to the appropriate
[23–25], which precludes use outside carefully monitored clinical targeted treatment or clinical trial. There are early signals from a
trials. Radiotherapy for local tumour control should be consid- phase II clinical trial that patients with metastatic melanomas,
ered in: cases of inadequate resection margins of lentigo maligna carrying NRAS mutation, may benefit from MEK kinase-inhibi-
melanoma [26], in R1 resections of melanoma metastases (when tor therapy [40]. The additional analysis for PDL-1 expression
surgery is not adequate), or after resection of bulky disease helps to enrich the population of patients who benefit from
[III, B]. A prospective randomised trial has demonstrated that anti-PD1 therapy, but is not powerful enough to exclude
postoperative irradiation after LN dissection reduces the risk for patients from anti-PD1 treatment [39, 41].
relapse in the irradiation field by ∼50%, but has no impact on The recommendations for first-line treatment of metastatic
RFS and OS [27]. Treatment decisions should be made in an disease are under debate. Reasonable approaches include anti-
interdisciplinary team. PD1 therapies and, for BRAF-mutated melanomas, combinations
of BRAF inhibitors with MEK inhibitors. BRAFi/MEKi inhibitor
combos offer high response rates (70%) and rapid response in-
treatment of locoregional disease duction associated with symptom control, with a progression-free
In the case of isolated locoregional LN metastases, surgical survival (PFS) of ∼12 months. Anti-PD1 therapy, and to a lesser
removal, including the surrounding LN region is indicated [III, extent ipilimumab, offer lower response rates in the range, but
C]; removal of the tumour-bearing LN alone is insufficient. In many responses are durable [42].
high-risk situations such as multiple bulky LN metastases, post- In patients with BRAF-wild-type (wt) disease, ipilimumab
operative radiotherapy can improve local tumour control, but has been the standard treatment based on a survival benefit with
has no impact on RFS and OS [27, 28]. a ∼10% higher survival rate at 1, 2 and 3 years [36]. Based on
However, before undertaking additional aggressive local surgi- very recent randomised trial results, comparing anti-PD1 antibody
cal treatments, a detailed staging investigation, that includes therapies to ipilimumab, anti-PD1 antibody therapy is the pre-
high-resolution imaging techniques, such as PET, CT or magnetic ferred first-line treatment of patients with BRAF-wt disease [42].
resonance imaging is necessary to exclude distant metastases These therapies also demonstrate efficacy for patients with other
v | Dummer et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
BRAF mutations [37]. Anti-PD1 therapies are also recommended preferentially be referred to centres of excellence that provide a
as a second-line treatment, after ipilimumab failure [43, 44]. comprehensive clinical trial programme.
The anti-PD1 antibody nivolumab was compared with the If clinical trials or new compounds are not available, cytotoxic
reference chemotherapy dacarbazine in a double-blind rando- drugs such as dacarbazine (DTIC), temozolomide, taxanes,
mised clinical trial with BRAF-wt patients. This trial showed a fotemustine, platin derivatives or others, cytokines (IFNs, inter-
1-year survival rate of 72.9% in the nivolumab group, compared leukin-2) or combinations may be applied. DTIC is still consid-
with 42.1% in the dacarbazine group (HR for death, 0.42; ered a reference drug in this situation. In aggressive metastatic
P < 0.001) [39]. Nivolumab and pembrolizumab present an ex- disease, multi-agent polychemotherapy, containing paclitaxel
cellent safety profile, resulting in a favourable risk/benefit ratio. and carboplatin or cisplatin, vindesine, and DTIC may provide
The most frequent adverse events included fatigue, pruritus and mostly short-lived partial responses and/or disease stabilisations
nausea. Both molecules have been compared with standard che- in a meaningful number of patients. Despite a better initial
motherapies in a second-line setting after ipilimumab therapy. control rate, no survival benefit has been shown with polyche-
They demonstrated favourable efficacy, with prolonged PFS and motherapy compared with monochemotherapy.
better response rates, than the chemotherapy option [43–45]. Surgery of visceral metastases may be appropriate for selected
Pembrolizumab (at a dose of 10 mg/kg of body weight) every cases with good performance status and isolated tumour manifes-
2 or 3 weeks was compared with ipilimumab in a randomised tations. In principal, the goal is R0 resection in these patients.
clinical trial. The 6-month PFS rates were ∼47% for pembrolizu- Palliative radiotherapy should be considered, especially for
mab, independent from the dose, and 26.5% for ipilimumab symptomatic brain or localised and painful bone metastases.
(HR for disease progression, 0.58; P < 0.001 for both pembroli- Stereotactic irradiation is preferred to whole brain irradiation in
zumab regimens versus ipilimumab). Estimated 12-month case of brain metastases [28].
survival rates were ∼70% for pembrolizumab, versus 58%. The In general, stage IV melanoma patients need to be treated and
response rate was ∼33% for pembrolizumab, compared with discussed in an interdisciplinary tumour board, within centres
11.9% for ipilimumab [42]. that have broad experience in this disease (Table 2).
In a double-blinded prospective randomised trial, nivolumab
was compared with ipilimumab and the ipilimumab/nivolumab
combination. The anti-PD1 antibody alone or in combination personalised medicine
demonstrated improved PFS (ipilimumab: 2.9, nivolumab: 6.9,
Biomarkers such as mutations (NRAS, c-Kit, BRAF) are already
combination: 11.5 months) and response rates.
indispensable today for proper management of advanced mel-
PDL-1 expression was a relevant marker in this context, because
anoma. Additional mutations and the overall mutation rate
there was no difference in PFS between anti-PD1 antibody therapy
might provide additional molecular predictive markers in the
versus a ipilimumab/nivolumab combination, in the PDL-1 posi-
near future. Based on the recent data of anti-PD1 efficacy in
tive population. The study was not powered to distinguish between
PDL-1-positive advanced melanoma [41], this parameter, which
the efficacy of nivolumab and the ipilimumab/nivolumab combo.
is determined by immunohistochemistry and reflects the pres-
The final clinical implications of this study, including the question
ence of T cells in the tumour microenvironment, might soon be
about the superiority of combined anti-PD1/CTLA-4 therapy
relevant. We assume that treatment algorithms for advanced
versus sequential anti-PD1/CTLA-4 therapy, remain open until
melanoma may evolve in a paradigm for precision medicine in
the survival data are mature [41].
the context of targeted and immunotherapy [42].
If the patient suffers from symptomatic, bulky metastases from
a BRAF-V600-mutated melanoma, a combination of BRAFi and
MEKi is a valid treatment option in first and second lines. It has a
high chance for rapid response and offers improvements in
patient information and follow-up
quality of life [37–39, 41]. There are no mature data to guide deci- Melanoma patients should be advised to avoid sunburn,
sion making regarding the sequencing of checkpoint inhibitors extended unprotected solar or artificial UV exposure, and to
and kinase-inhibitor combinations, in patients with BRAF- have lifelong, regular self-examinations of the skin and periph-
mutant metastatic melanoma. Emerging data suggest that BRAF eral LNs. Patients must be aware that family members have an
inhibition is effective following immunotherapy, and checkpoint increased melanoma risk [III, B]. There is no recommendation
inhibitors are still effective in patients who have progressed on for genetic testing.
kinase-inhibitor therapy. Kinase inhibitors [46] and ipilimumab During melanoma follow-up, patients are clinically moni-
and/or anti-PD1 antibody therapy [47] can be safely used even in tored in order to detect a relapse and to recognise additional
patients with symptomatic brain metastases, in fact it has shown skin tumours, especially secondary melanomas, as early as pos-
significant efficacy in this area [48]. Stereotactic irradiation of pro- sible [4] [III, B]. However, it remains to be determined whether
gressive brain metastases is reasonable if systemic therapy can this strategy leads to improved survival rates, especially in this
achieve partial disease control. new era of systemic therapies for stage IV disease. Eight percent
In the context of new developments and medical progress, there of all melanoma patients develop a secondary melanoma within
are continuously new experimental treatment options for patients 2 years of the initial diagnosis [49]. Melanoma patients also
with advanced metastatic melanoma, including combined therap- have an increased risk for other skin tumours. In patients with
ies with anti-CTLA4 and anti-PD1 antibodies, with intralesional lentigo maligna melanomas, 35% of patients develop another
therapies and small molecules. Therefore, patients should cutaneous malignancy within 5 years [26].
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv297 | v

by guest
on 05 February 2018
Table 2. Treatment modalities for melanoma metastases

Number and localisation of the metastases Treatment modalities Grade of recommendation
1st choice
2nd choice
3rd choice
In-transit metastases (few) (pTXN2cM0) Surgical removal C

Radiotherapy C
In-transit metastases (multiple, >5) (pTXN2cM0) Perfusion of the extremitya D
Radiotherapy D
T-Vec (Talimogene laherparepvec) D
Electrochemotherapy D
Systemic therapya D
Locoregional LNs (pTxN1a, 2a) Discuss Regional lymph node dissection and trial B
participation
Additional Interferon alpha treatmenta B
Locoregional LNs (pTxN2b, 2c, 3) Radical lymphadenectomy, C

in case of incomplete resection:
irradiation,
Consider trial participation C
Solitary central nervous system metastases (pTxNxM3) Neurosurgical removal D
Stereotactic irradiationa D
(according to localisation this could also be the 1st choice)
or other local treatment approaches
Consider clinical trial participation
Solitary lung/liver/kidney and other metastases (pTxNxM1) Surgical removal D
Consider clinical trial participation
Systemic therapya D
Multiple metastases (pTxNxM1a–1c) Consider clinical trial participation
Systemic therapya B
Painful bone metastases (pTxNxM1a–1c) Consider clinical trial participation

Radiotherapy C
Bone-modifying agents
a
These therapies should be preferentially carried out at specialised centres.
LNs, lymph nodes.
There is currently no consensus on the frequency of follow-up therefore the most accurate blood test in the follow-up of melan-
examinations and the use of imaging techniques. Recommendations oma patients [53], if any blood test is recommended at all [IV, D].
vary from follow-up visits every 3 months, during the first 3 years
and every 6–12 months thereafter, to no organised follow-up at all.
We encourage consultation of the respective national guidelines.
methodology
Intervals between controls may be tailored according to the indivi- These clinical practice guidelines were developed in accordance
dual’s risk and the personal needs of the patient [50]. with the ESMO standard operating procedures for clinical prac-
Since patients with a thin primary melanoma have only a small tice guidelines development. The relevant literature has been
risk of relapse, routine imaging techniques are definitively not selected by the expert authors. A summary of recommendations
recommended for this patient population. In high-risk patients, is shown in Table 3. Levels of evidence and grades of recommen-
(e.g. those with thick primary tumours, or following treatment of dation have been applied using the system shown in Table 4.
metastases) ultrasound of LNs, CT or whole-body PET/PET–CT Statements without grading were considered justified standard
scans may lead to an earlier diagnosis of regional or systemic clinical practice by the experts and the ESMO faculty. This manu-
relapses [51]. The impact of radiological exams upon survival has script has been subjected to an anonymous peer review process.
not been demonstrated so far [52]. However, targeted therapy and
immunotherapy demonstrate favourable effects in patients with
low tumour burden, who can be identified by high-resolution
imaging during follow-up. Rising serum S-100 has a higher specifi- RD has reported research funding from Novartis, Merck
city for disease progression than lactate dehydrogenase, and is Sharp & Dohme (MSD), Bristol-Myers Squibb (BMS), Roche,
v | Dummer et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Table 3. Summary of recommendations Table 4. Levels of evidence and grades of recommendation
Diagnosis (adapted from the Infectious Diseases Society of America-United
• Diagnosis should be based on a full-thickness excisional biopsy with a
Levels of evidence
small side margin [II, A].
• The histology report should include at least: information on the type I Evidence from at least one large randomised, controlled trial of
of melanoma, actinic damage, maximum vertical thickness in good methodological quality (low potential for bias) or meta-
millimetres, information on mitotic rate in case of pT1, presence of analyses of well-conducted randomised trials without
ulceration, presence and extent of regression and clearance of the heterogeneity
surgical margins [II, A]. II Small randomised trials or large randomised trials with a suspicion
• Physical examination with special attention to other suspicious of bias (lower methodological quality) or meta-analyses of such
pigmented lesions, tumour satellites, in-transit metastases, regional LN trials or of trials with demonstrated heterogeneity
and distant metastases is mandatory. In low-risk melanomas (pT1a), no III Prospective cohort studies
other investigations are necessary. In higher tumour stages, imaging is IV Retrospective cohort studies or case–control studies
recommended in order to allow proper staging [III, C]. V Studies without control group, case reports, experts opinions
Treatment of localised disease Grades of recommendation
• Wide excision of primary tumours with safety margins of A Strong evidence for efficacy with a substantial clinical benefit,
0.5 cm for in situ melanomas, of 1 cm for tumours with a tumour strongly recommended
thickness up to 2 mm and 2 cm for thicker tumours is recommended B Strong or moderate evidence for efficacy but with a limited clinical
[II, B]. benefit, generally recommended
• Sentinel LN biopsy in melanoma with a tumour thickness of >1 mm C Insufficient evidence for efficacy or benefit does not outweigh the
and/or ulceration is recommended for precise staging [II, B]. It should be risk or the disadvantages (adverse events, costs, … ), optional
discussed in patients with a pT1b with a tumour thickness >0.75 mm. D Moderate evidence against efficacy or for adverse outcome,
• Patients with resected stage III melanomas should be evaluated for generally not recommended
adjuvant interferon therapy [II, B]. Subgroup analyses suggest patients E Strong evidence against efficacy or for adverse outcome, never
with microscopic regional nodal involvement and/or ulcerated recommended
primaries are most likely to benefit from adjuvant IFN. In stage IIIB
a
and higher, participation in clinical trials should be encouraged. By permission of the Infectious Diseases Society of America [54].
• Surgical removal or stereotactic irradiation of locoregional recurrence
or single distant metastasis should be considered in fit patients, as a
therapeutic option, offering potential for long-term disease control
[III, C].
Serono, MSD/Merck, Novartis, On-cosec, Roche Pharma. The
Treatment of systemic metastatic disease (stage IV)
other authors have declared no potential conflicts of interest.
• Patients with metastatic melanoma should have metastasis
(preferably) or the primary tumour screened for detection of BRAF-
V600-mutation. Treatment options for the first- and second-line
setting include anti-PD1 antibodies (pembrolizumab, nivolumab), references
ipilimumab, an anti-CTLA4 antibody, for all patients, and BRAF/ 1. Hollestein LM, van den Akker SA, Nijsten T et al. Trends of cutaneous melanoma
MEK inhibitor combinations for patients with BRAF-mutant in The Netherlands: increasing incidence rates among all Breslow thickness
melanoma [II, B]. categories and rising mortality rates since 1989. Ann Oncol 2012; 23: 524–530.
• If clinical trials or the approved new targeted compounds are not 2. Forsea AM, Del Marmol V, Stratigos A, Geller AC. Melanoma prognosis in Europe:
available, cytotoxic drugs such as DTIC or temozolomide may be far from equal. Br J Dermatol 2014; 171: 179–182.
administered, with modest activity shown [II, C]. 3. Green AC, Williams GM, Logan V, Strutton GM. Reduced melanoma after regular
sunscreen use: randomized trial follow-up. J Clin Oncol 2011; 29: 257–263.
Patient information and follow up
4. Dummer R, Guggenheim M, Arnold AW et al. Updated Swiss guidelines for the
• Melanoma patients should be instructed in the avoidance of sunburns, treatment and follow-up of cutaneous melanoma. Swiss Med Wkly 2011; 141:
w13320.
extended unprotected solar or artificial UV exposure, and in lifelong
5. Bono A, Tolomio E, Trincone S et al. Micro-melanoma detection: a clinical study on
regular self-examinations of the skin and peripheral LNs [III, B].
206 consecutive cases of pigmented skin lesions with a diameter < or = 3 mm.
• There is no consensus on the optimal schedule or frequency of follow-
Br J Dermatol 2006; 155: 570–573.
up visits, or on the utility of imaging and blood tests for patients with
6. Grob JJ, Bonerandi JJ. The ‘ugly duckling’ sign: identification of the common
resected melanoma. characteristics of nevi in an individual as a basis for melanoma screening. Arch
Dermatol 1998; 134: 103–104.
LN, lymph node; IFN, interferon; DTIC, dacarbazine; UV, ultraviolet. 7. Kittler H, Pehamberger H, Wolff K, Binder M. Diagnostic accuracy of dermoscopy.
Lancet Oncol 2002; 3: 159–165.
8. Balch CM, Gershenwald JE, Soong SJ et al. Final version of 2009 AJCC
melanoma staging and classification. J Clin Oncol 2009; 27: 6199–6206.
GlaxoSmithKline (GSK) and consultant or advisory board with 9. Whiteman DC, Pavan WJ, Bastian BC. The melanomas: a synthesis of
Novartis, MSD, BMS, Amgen, Roche, GSK. AH has reported epidemiological, clinical, histopathological, genetic, and biological aspects,
consultancy, honoraria and trial grants from Amgen, BMS, supporting distinct subtypes, causal pathways, and cells of origin. Pigment Cell
Celgene, Eisai, GSK, MedImmune, Me-laSciences, Merck Melanoma Res 2011; 24: 879–897.
Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv297 | v

by guest
on 05 February 2018
10. Bastian BC. The molecular pathology of melanoma: an integrated taxonomy of 32. McArthur GA, Chapman PB, Robert C et al. Safety and efficacy of vemurafenib in
melanocytic neoplasia. Annu Rev Pathol 2014; 9: 239–271. BRAF (V600E) and BRAF (V600K) mutation-positive melanoma (BRIM-3): extended
11. Schoenewolf NL, Bull C, Belloni B et al. Sinonasal, genital and acrolentiginous follow-up of a phase 3, randomised, open-label study. Lancet Oncol 2014; 15:
melanomas show distinct characteristics of KIT expression and mutations. Eur 323–332.
J Cancer 2012; 48: 1842–1852. 33. Chapman PB, Hauschild A, Robert C et al. Improved survival with vemurafenib in
12. Hirakawa S, Kodama S, Kunstfeld R et al. VEGF-A induces tumor and sentinel melanoma with BRAF V600E mutation. N Engl J Med 2011; 364: 2507–2516.
lymph node lymphangiogenesis and promotes lymphatic metastasis. J Exp Med 34. Guo J, Si L, Kong Y, Flaherty KT et al. Phase II, open-label, single-arm trial of
2005; 201: 1089–1099. imatinib mesylate in patients with metastatic melanoma harboring c-Kit mutation
13. Thompson JF, Scolyer RA, Kefford RF. Cutaneous melanoma. Lancet 2005; 365: or amplification. J Clin Oncol 2011; 29: 2904–2909.
687–701. 35. Hodi FS, O’Day SJ, McDermott DF et al. Improved survival with ipilimumab in
14. Han D, Zager JS, Shyr Y et al. Clinicopathologic predictors of sentinel lymph node patients with metastatic melanoma. N Engl J Med 2010; 363: 711–723.
metastasis in thin melanoma. J Clin Oncol 2013; 31: 4387–4393. 36. Robert C, Thomas L, Bondarenko I et al. Ipilimumab plus dacarbazine for previously
15. Morton DL, Thompson JF, Cochran AJ et al. Sentinel-node biopsy or nodal untreated metastatic melanoma. N Engl J Med 2011; 364: 2517–2526.
observation in melanoma. N Engl J Med 2006; 355: 1307–1317. 37. Long GV, Stroyakovskiy D, Gogas H et al. Combined BRAF and MEK inhibition
16. Eggermont AM, Suciu S, Testori A et al. Ulceration and stage are predictive of versus BRAF inhibition alone in melanoma. N Engl J Med 2014; 371:
interferon efficacy in melanoma: results of the phase III adjuvant trials EORTC 1877–1888.
18952 and EORTC 18991. Eur J Cancer 2012; 48: 218–225. 38. Robert C, Karaszewska B, Schachter J et al. Improved overall survival in melanoma
17. Kirkwood JM, Ibrahim JG, Sondak VK et al. High- and low-dose interferon alfa-2b with combined dabrafenib and trametinib. N Engl J Med 2015; 372: 30–39.
in high-risk melanoma: first analysis of intergroup trial E1690/S9111/C9190. 39. Robert C, Long GV, Brady B et al. Nivolumab in previously untreated melanoma
J Clin Oncol 2000; 18: 2444–2458. without BRAF mutation. N Engl J Med 2015; 372: 320–330.
18. Mocellin S, Pasquali S, Rossi CR, Nitti D. Interferon alpha adjuvant therapy in 40. Ascierto PA, Schadendorf D, Berking C et al. MEK162 for patients with advanced
patients with high-risk melanoma: a systematic review and meta-analysis. J Natl melanoma harbouring NRAS or Val600 BRAF mutations: a non-randomised, open-
Cancer Inst 2010; 102: 493–501. label phase 2 study. Lancet Oncol 2013; 14: 249–256.
19. Eggermont AM, Suciu S, Santinami M et al. Adjuvant therapy with pegylated 41. Larkin J, Ascierto PA, Dréno B et al. Combined vemurafenib and cobimetinib in
interferon alfa-2b versus observation alone in resected stage III melanoma: final BRAF-mutated melanoma. N Engl J Med 2014; 371: 1867–1876.
results of EORTC 18991, a randomised phase III trial. Lancet 2008; 372: 117–126. 42. Robert C, Schachter J, Long GV et al. Pembrolizumab versus ipilimumab in
20. Eggermont AM, Suciu S, Testori A et al. Long-term results of the randomized advanced melanoma. N Engl J Med 2015; 372: 2521–2532.
phase III trial EORTC 18991 of adjuvant therapy with pegylated interferon alfa-2b 43. Weber JS, D’Angelo SP, Minor D et al. Nivolumab versus chemotherapy in patients
versus observation in resected stage III melanoma. J Clin Oncol 2012; 30: with advanced melanoma who progressed after anti-CTLA-4-treatment
3810–3818. (CheckMate 037): a randomised, controlled, open-label, phase 3 trial. Lancet
21. Eggermont AM, Chiarion-Sileni V, Grob JJ et al. Adjuvant ipilimumab versus Oncol 2015; 16: 375–384.
placebo after complete resection of high-risk stage III melanoma (EORTC 18071): 44. Dummer R, Daud A, Puzanov I et al. A randomized controlled comparison of
a randomised, double-blind, phase 3 trial. Lancet 2015; 16: 522–530. pembrolizumab and chemotherapy in patients with ipilimumab-refractory
22. Kleeberg UR, Suciu S, Brocker EB et al. Final results of the EORTC 18871/DKG melanoma. J Transl Med 2015; 13(Suppl 1): 05.
80–1 randomised phase III trial. rIFN-alpha2b versus rIFN-gamma versus 45. Robert C, Dummer R, Gutzmer R et al. Selumetinib plus dacarbazine versus placebo
ISCADOR M versus observation after surgery in melanoma patients with either plus dacarbazine as first-line treatment for BRAF-mutant metastatic melanoma: a
high-risk primary (thickness >3 mm) or regional lymph node metastasis. Eur phase 2 double-blind randomised study. Lancet Oncol 2013; 14: 733–740.
J Cancer 2004; 40: 390–402. 46. Long GV, Trefzer U, Davies MA et al. Dabrafenib in patients with Val600Glu or
23. Oberholzer PA, Kee D, Dziunycz P et al. RAS mutations are associated with the Val600Lys BRAF-mutant melanoma metastatic to the brain (BREAK-MB): a
development of cutaneous squamous cell tumors in patients treated with RAF multicentre, open-label, phase 2 trial. Lancet Oncol 2012; 13: 1087–1095.
inhibitors. J Clin Oncol 2012; 30: 316–321. 47. Margolin K, Ernstoff MS, Hamid O et al. Ipilimumab in patients with melanoma
24. Su F, Viros A, Milagre C et al. RAS mutations in cutaneous squamous-cell and brain metastases: an open-label, phase 2 trial. Lancet Oncol 2012; 13:
carcinomas in patients treated with BRAF inhibitors. N Engl J Med 2012; 366: 459–465.
207–215. 48. Dummer R, Goldinger SM, Turtschi CP et al. Vemurafenib in patients with BRAF
25. Zimmer L, Hillen U, Livingstone E et al. Atypical melanocytic proliferations and new (V600) mutation-positive melanoma with symptomatic brain metastases: final
primary melanomas in patients with advanced melanoma undergoing selective results of an open-label pilot study. Eur J Cancer 2014; 50: 611–621.
BRAF inhibition. J Clin Oncol 2012; 30: 2375–2383. 49. Titus-Ernstoff L, Perry AE, Spencer SK et al. Multiple primary melanoma: two-year
26. Farshad A, Burg G, Panizzon R, Dummer R. A retrospective study of 150 patients results from a population-based study. Arch Dermatol 2006; 142: 433–438.
with lentigo maligna and lentigo maligna melanoma and the efficacy of 50. Turner RM, Bell KJ, Morton RL et al. Optimizing the frequency of follow-up visits
radiotherapy using Grenz or soft X-rays. Br J Dermatol 2002; 146: 1042–1046. for patients treated for localized primary cutaneous melanoma. J Clin Oncol 2011;
27. Burmeister BH, Henderson MA, Ainslie J et al. Adjuvant radiotherapy versus observation 29: 4641–4646.
alone for patients at risk of lymph-node field relapse after therapeutic lymphadenectomy 51. Bastiaannet E, Wobbes T, Hoekstra OS et al. Prospective comparison of [18F]
for melanoma: a randomised trial. Lancet Oncol 2012; 13: 589–597. fluorodeoxyglucose positron emission tomography and computed tomography in
28. Hong A, Fogarty G. Role of radiation therapy in cutaneous melanoma. Cancer J patients with melanoma with palpable lymph node metastases: diagnostic
2012; 18: 203–207. accuracy and impact on treatment. J Clin Oncol 2009; 27: 4774–4780.
29. Campana LG, Testori A, Mozzillo N, Rossi CR. Treatment of metastatic melanoma 52. Nieweg OE, Kroon BB. The conundrum of follow-up: should it be abandoned? Surg
with electrochemotherapy. J Surg Oncol 2014; 109: 301–307. Oncol Clin N Am 2006; 15: 319–330.
30. Andtbacka RH, Kaufman HL, Collichio F et al. Talimogene laherparepvec improves 53. Beyeler M, Waldispuhl S, Strobel K et al. Detection of melanoma relapse: first
durable response rate in patients with advanced melanoma. J Clin Oncol 2015 comparative analysis on imaging techniques versus S100 protein. Dermatology
[Epub ahead of print]. 2006; 213: 187–191.
31. Flaherty KT, Hennig M, Lee SJ et al. Surrogate endpoints for overall survival in 54. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
metastatic melanoma: a meta-analysis of randomised controlled trials. Lancet among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
Oncol 2014; 15: 297–304. 139–144.
v | Dummer et al. Volume 26 | Supplement 5 | September 2015

by guest
on 05 February 2018
Neuroendocrine bronchial and thymic tumors: ESMO

and follow-up†
K. Öberg1, P. Hellman2, P. Ferolla3 & M. Papotti4, on behalf of the ESMO Guidelines Working
Group*
1
Department of Endocrine Oncology, University Hospital, Uppsala University, Uppsala; 2Department of Surgery, University Hospital Uppsala, Uppsala, Sweden;
3
Multidisciplinary Group for Diagnosis and Therapy of Neuroendocrine Tumors, Umbria Region Cancer Network, Orbassano; 4Division of Anatomic Pathology,San Luigi
Hospital, University of Turin, Orbassano, Italy
incidence and epidemiology careful evaluation and treatment. NETs of the lung include the
low-grade TC, intermediate-grade AC, the high-grade LCNEC
Neuroendocrine tumors (NETs) of the lung comprise a and SCLC. The incidence of SCLC has been declining the last
heterogeneous population of tumors ranging from well- 35 years in the western world, maybe due to decreasing
differentiated bronchial NETs to highly malignant and poorly smoking habits [5]. Mixed tumors are found in <5% of
differentiated small-cell lung cancer (SCLC) and large-cell patients and are more frequently found in the peripheral areas
neuroendocrine carcinoma (LCNEC). The incidence of of the lung. They consist of a combination of SCLC and
clinical practice
pulmonary NETs is low, although reported to have increased LCNEC, but also mixtures of either SCLC or LCNEC with
guidelines
over the past 30 years [1, 2]. This is mainly due to improved adenocarcinomas and/or squamous cell carcinoma. About 70%
detection methods and diagnostic protocols. Of all NETs of all bronchial NETs are located in the major bronchi and the
∼25% are located in the respiratory tract. Typical carcinoids remainder in the periphery of the lungs. They occur more
(TCs) comprise ∼1%–2% and atypical carcinoids (ACs) only frequently (60%) in the right than in the left lung, and
0.1%–0.2% of pulmonary neoplasms. According to the particularly in the middle lobe [6]. The cell of origin for
surveillance, epidemiology and end results program (SEER) bronchial NETs have been suggested to be pulmonary
database from 2003, the combined incidence has been 1.57/ neuroendocrine cells (PNECs) that usually exist as solitary
100 000 inhabitants [3]. SCLC is the most common bronchial cells, but sometimes aggregate to form small nodules termed
NET reported to account for 15%–20% of invasive lung neuroepithelial bodies (NEBs), which are located within the
cancers. LCNEC comprise 1.6%–3% of resectable lung cancers. ciliated epithelium. PNECs express serotonin and neuron-
The prevalence of thymic NET is ∼3% of the total number of specific enolase (NSE) and also gastrin-releasing peptide
NETs at all sites. In the last SEER database, a reported (GRP) [7]. In adults, NEBs have been described to respond
incidence of thymic NETs is 0.02/100 000 population per year to hypoxia by the secretion of serotonin, thereby inducing
[4]. They constitute ∼5% of all thymic tumors. Both bronchial local vasoconstriction to decrease the bloodstream in poorly
and thymic NETs may be part of multiple endocrine neoplasia ventilated areas of the lung and thereby, direct the blood
type 1 syndrome (MEN-1, 5%–15%). The median age at toward better ventilated areas. Diffuse idiopathic PNEC
diagnosis for bronchial NETs is 64 years and for thymic NETs hyperplasia is a rare preneoplastic condition comprising a
59 years. This review is restricted to typical/atypical NETs and generalized proliferation of PNECs predominantly in women
thymic NETs. and non-smokers [7]. Up to 90% of patients with central
bronchial NETs are symptomatic, presenting with
diagnosis hemoptysis, cough, recurrent pulmonary infection, fever,
chest discomfort and unilateral wheezing, while peripheral
NETs of the lung and thymus should be referred to a center carcinoids are incidentally discovered in most of the cases
with particular interest in and knowledge of the disease for [6]. The carcinoid syndrome is very rare in patients with
bronchial NETs. Nevertheless, a carcinoid crisis may
*Correspondence to: ESMO Guidelines Working Group, ESMO Head Office, Via occasionally occur in previously asymptomatic patients
L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland; E-mail: clinicalguidelines@esmo. following bronchoscopic biopsy laser disobliteration, surgical
org
manipulation or peptide receptors radiotherapy (PRRT). AC
†
Approved by the ESMO Guidelines Working Group: August 2007, last update July syndrome may cause life-threatening bronchostenosis and
2012. This publication supersedes the previously published version—Ann Oncol 2010; should be promptly recognized and treated. In about 2% of
21 (Suppl. 5): v220–v222.

by guest
on 05 February 2018
patients with Cushing’s syndrome, the cause is ectopic addition, the proliferation index as detected by Ki-67
adeno-cortic trophic hormone (ACTH) production from immunostaining is also an extremely useful tool to better
either bronchial or thymic NETs [5, 6]. classify a bronchial NET although actually not included in the
SCLC derives from normal bronchial epithelial cells WHO classification criteria [7, 10].
expressing neuroendocrine characteristics. Alternatively, a Most thymic NET cases are completely asymptomatic and
common pulmonary stem cell may exist giving rise to both imaging performance for other reasons generally incidentally
neuroendocrine stem cells and SCLC cells [7]. discovers thymic NETs. Not infrequently distant metastases are
The World Health Organization (WHO) classification of present at the time of diagnosis. Clinical symptoms usually
bronchial NETs combined architectural growth patterns of tumor occur at a later stage of the disease, such as chest discomfort,
cells (organoid growth versus small-cell diffuse growth) with the superior vena cava syndrome, dyspnea and cough. Thymic
mitotic index and the presence of necrosis (Table 1) [8, 9]. NETs are frequently associated with hormonal hyper-secretion
The identification of the neuroendocrine phenotype and the such as ACTH secretion giving rise to Cushing’s syndrome and
correct NET classification necessarily include the evaluation of growth hormone releasing hormone (GHRH) hyper-secretion
specific neuroendocrine markers. Among these chromogranin with ectopic acromegaly [10]. The tumor seems to have a
A and synaptophysin expression are the most reliable stains predilection for men (man to female ratio 3:1). A characteristic
(Table 2). Other markers helpful to define a neuroendocrine feature of these tumors is the presence of nests of tumor cells
phenotype include PGP 9.5, NSE and CD56. Transcription that are detached from the surrounding stroma and contain
factors driving neuroendocrine cell differentiation during areas of necrosis. The tumors often show architectural features
human development have been described in bronchial NETs. of neuroendocrine differentiation with positive
These include human achaete-scute homolog 1 whose immunohistochemistry for chromogranin A, synaptophysin and
expression has been reported in high-grade bronchial NETs. In CD56. They can be divided into low, intermediate and high-
grade tumors. Low-grade tumors present <10 mitoses/10 high
power field (HPF), intermediate tumors 10-20 mitoses/HPF and
Table 1. Classification of bronchial NET high grade tumors <10 20 mitoses/10 HPF. The tumors may be
associated with MEN-1 in 4%–8% of the cases. Nearly all cases
Histological type Necrosis Mitotic count associated with MEN-1 are male patients and smokers [9].
TC absent <2/10 HPF
AC present focal 2–9/10 HPF
LCNEC present (extensive) >9/10 HPF
SCLC present (extensive) >50/10 HPF A tumor–node–metastasis (TNM) staging is recommended for
bronchial NETs and is included in the 7th edition of the Union
TC: typical carcinoid; AC: atypical carcinoid; LCNEC: large-cell
for International Cancer Control/American Joint Committee on
neuroendocrine carcinoma; SCLC: small-cell carcinoma.
Cancer (UICC/AJCC) TNM staging system (Table 3). TNM
staging of thymic NETs follows the general rules for tumors of
Table 2. Diagnosis bronchial NET the thymus. Biochemical evaluation for both bronchial and
thymic NETs include plasma chromogranin A, plasma-NSE, and
Clinical symptoms bronchial NET in selected cases dU-5-hydroxy indol acetic acid with clinical
Hemoptysis, cough, recurrent pulmonary infections, wheezing symptoms of carcinoid syndrome and urine cortisol with
g
Carcinoid syndrome Cushing's disease, plasma ACTH and those with signs of
Rare acromegaly, plasma GHRH and insulin growth factor (IGF)-I
Cushing’s syndrome (III, A) [10]. Conventional X-ray of the chest may suggest a
Clinical symptoms thymic NET diagnosis of both bronchial and thymic NETs, but computed
Chest discomfort, superior vena cava syndrome, stridor, Cushing’s tomography (CT) scan is the recommended investigation.
syndrome, acromegaly Bronchoscopy, if necessary with additional endoscopic
Pathology ultrasonography with biopsies, is the best procedure to detect
Histopathology central bronchial NETs (III, A). Since 80% of typical bronchial
Immunohistochemistry (chromogranin A, synaptophysin, CD56, Ki-67) carcinoids express somatostatin receptors, somatostatin receptor
Biochemistry (specific) scintigraphy may be informative as well as 68Gallium-
p-Chromogranin A, p-NSE, u-5HIAA, [au-cortisol, p-ACTH, p-GHRH, DOTATATE/TOC (DOTA0, D-Phe1, 8tyr3] Octreotate) positron
IGF-1]
emission tomography (PET) scanning (III, B) [11, 12]. For more
Imaging
aggressive bronchial NETs such as LCNEC and SCLC, fluoro
Multislice CT scan
111 deoxy glucose (FDG) PET is more informative than somatostatin
In-DTPA-scintigraphy: (Octreoscan®)
receptor scintigraphy (III, B) [13, 14]. For thymic NETs contrast
PET: 68Ga-DOTATATE, FDG
enhanced CT or magnetic resonance imaging (MRI) is
Bronchoscopy
Genetic screening
recommended to detect tumor metastases. Somatostatin receptor
MEN-1 genetic screening when suspected scintigraphy may be used for these tumors as well as PET
scanning with 68Gallium-DOTATATE (III, B). Bronchial NETs
a
Clinical symptoms suggesting Cushing’s syndrome alt acromegaly. sometimes present with AC syndrome related to the secretion of
NSE, neuron-specific enolase; 5HIAA, 5-hydroxy indol acetic acid. histamine metabolites. The biochemical profile for thymic NETs
Volume 23 | Supplement 7 | October 2012 doi:10.1093/annonc/mds267 | vii

by guest
on 05 February 2018
Table 3. AJCC staging for lung tumor NETs
Primary tumor (T)

TX Primary tumor cannot be assessed or tumor was proved by the presence of malignant cells in sputum or bronchial washings but not visualized
by imaging or bronchoscopy
T1 Tumor ≤3 cm in greatest dimension, surrounded by lung or visceral pleura, without bronchoscopic evidence of invasion more proximal than
the lobar bronchus (i.e. not in the main bronchus)
T1a Tumor ≤2 cm in greatest dimension
T1b Tumor ≥2 cm but <3 cm in greatest dimension
T2 Tumor ≥3 cm but ≤7 cm or tumor with any of the following features (T2 tumors with these features are classified T2a if ≤5 cm); involves
main bronchus, ≥2 cm distal to the carina; invades visceral pleura (PL1 or PL2); associated with atelectasis or obstructive pneumonitis that
extends to the hilar region but does not involve the entire lung
T2a Tumor ≥3 cm but ≤5 cm in greatest dimension
T2b Tumor ≥5 cm but ≤7 cm in greatest dimension
T3 Tumor ≥7 cm or one that directly invades any of the following: parietal pleural (PL3) chest wall (including superior sulcus tumors),
diaphragm, phrenic nerve, mediastinal pleura, parietal pericardium; tumor in the main bronchus (<2 cm distal to the carina pleura, parietal
pericardium; tumor in the main bronchus (<2 cm distal to the carina but without involvement of the carina); or associated atelectasis or
obstructive pneumonitis of the entire lung or separate tumor nodule(s) in the same lobe
T4 Tumor of any size that invades any of the following: mediastinum, heart, great vessels, trachea, recurrent laryngeal nerve, esophagus, vertebral
body, carina or separate tumor nodule(s) in a different ipsilateral lobe
N1 Metastasis in ipsilateral peribronchial and/or ipsilateral hilar lymph nodes and intrapulmonary nodes, including involvement by direct
extension
N2 Metastasis in contralateral mediastinal, contralateral hilar, ipsilateral or contralateral scalene, or supraclavicular lymph node(s)
M1a Separate tumor nodule(s) in a contralateral lobe tumor with pleural nodules or malignant pleural (or pericardial) effusion
is usually similar to the bronchial NETs. A core biopsy is operative disobliterating procedure. The surgical techniques of
preferred from relevant lesion; Cushing’s syndrome is present in choice are lobectomy or sleeve resection (III, A).
about one-third of the patients, particularly in patients with Pneumonectomy should be avoided except in selected cases.
MEN-1 associated thymic NET [10]. Systemic nodal dissection should be performed since
TCs exhibit a good prognosis with a 5-year survival of 87%– lymphonodal metastases may be present in up to 25% of cases
90%. However, distant metastases from TCs may occur many in TC and >50% in AC [15, 16]. Thymic NETs should
years even after radical resection of the primary tumor. A 15- whenever feasible be subjected to radical surgical resection (III,
year follow-up is therefore recommended. ACs are regarded as A). Unfortunately, the percentage of recurrence remained
intermediate in grade and are associated with poor prognosis remarkably high, higher than in bronchial NET counterparts
and a 5-year survival of 44%–78%. The LCNEC is associated and a protracted follow-up should always be performed also in
with a 5-year survival rate of 15%–57% and finally the 5-year patients radically operated.
survival rate for SCLC is ∼5% [6, 10].
The prognosis for patients with primary thymic NETs
remains poor. This is due to the aggressive nature of tumor management of advanced/metastatic
with a high incidence of recurrence following surgery. Low- disease
grade thymic NETs present a 5-year survival of 50% and a10- Cytotoxic treatment combined with surgical resection when
year survival of 9%, whereas high-grade thymic NETs have a indicated has been the standard for metastatic bronchial and
5-year survival of nearly 0% [10]. thymicNETs, although the available chemotherapy regimens
demonstrate a rather poor effect (III, A) [17, 18].
Chemotherapy for SCLC, which is a chemosensitive but not
management of localized disease curable cancer, is discussed in the appropriate guidelines. For
The main therapy for bronchial NETs is surgical resection. The low proliferating tumors treatment with somatostatin analogs
surgical approach is dependent on the size, location and tissue and alpha interferon might be an option for functional tumors
type. Bronchoscopic laser excision of intraluminal typical with clinical symptoms (III, B). Treatment with these agents
bronchial NETs should be considered a suboptimal treatment has resulted in partial remission (PR) in 5%–10% but stable
and reserved for inoperable patients or performed as pre- disease (SD) in 30%–50% and symptomatic improvement in
vii | Öberg et al. Volume 23 | Supplement 7 | October 2012

by guest
on 05 February 2018
40%–60% of cases. Treatment with novel agents, such as The other authors have reported no potential conflicts of
tyrosine kinase inhibitors (e.g. sunitinib) and the mTOR- interest.
inhibitor everolimus, have been reported in very small series or
in the subgroup analysis of larger studies not designed
references
specifically for lung NET with mainly stabilization of the
disease (objective response rates 5%–10%; III, B). In non- 1. Yao JC, Hassan M, Phan A et al. One hundred years after ‘carcinoid’:
functioning tumors, the use of somatostatin analogs is still epidemiology of and prognostic factors for neuroendocrine tumors in 35,825
cases in the United States. J Clin Oncol 2008; 26(18): 3063–3072.
controversial, but after the PROMID study indicating
2. Skuladottir H, Hirsch FR, Hansen HH et al. Pulmonary neuroendocrine tumors:
antitumor efficacy by octreotide long acting release (LAR) in incidence and prognosis of histological subtypes. A population-based study in
small intestinal NETs, it is now widely accepted also for non- Denmark. Lung Cancer 2002; 37(2): 127–135.
functioning tumors of other origins (III, B). Peptide receptor 3. Gustafsson BI, Kidd M, Chan A et al. Bronchopulmonary neuroendocrine tumors.
radiotherapy is an option in patients with tumors that present Cancer 2008; 113(1): 5–21.
a high content of somatostatin receptors (III, B). Available 4. Gaur P, Leary C, Yao JC. Thymic neuroendocrine tumors: a SEER database
chemotherapy regimens for TC and AC include a combination analysis of 160 patients. Ann Surg 2010; 251(6): 1117–1121.
of streptozotocin plus 5-fluoro-uracil/doxorubicin. 5. Travis WD. Lung tumors with neuroendocrine differentiation. Eur J Cancer 2009;
Temozolomide alone or in combination with capecitabine and 45(Suppl. 1): 251–266.
sometimes bevacizumab has demonstrated clinical benefit (III, 6. Travis WD, Rush W, Flieder DB et al. Survival analysis of 200 pulmonary
neuroendocrine tumors with clarification of criteria for atypical carcinoid and its
B). A combination of cisplatinum and etoposide is mainly used
separation from typical carcinoid. Am J Surg Pathol 1998; 22(8): 934–944.
in high proliferating tumors (III, B). In general chemotherapy, 7. Righi L, Volante M, Rapa I et al. Neuro-endocrine tumors of the lung. A review of
results are discouraging except for temozolomide alone or in relevant pathological and molecular data. Virchows Arch 2007; 451(Suppl. 1):
combination with capecitabine (III, B). No randomized trials S51–S59.
have been performed that could guide the treatment. However, 8. Travis WD, Giroux DJ, Chansky K et al. The IASLC Lung Cancer Staging Project:
the RADIANT-2 trial, which was a randomized trial between proposals for the inclusion of broncho-pulmonary carcinoid tumors in the
everolimus (10 mg) and placebo in NET tumor patients, forthcoming (seventh) edition of the TNM Classification for Lung Cancer. J
included 44 of the 429 patients with bronchial NETs. A clear Thorac Oncol 2008; 3(11): 1213–1223.
benefit of everolimus compared with placebo was noted (II, A). 9. Goudet P, Bonithon-Kopp C, Murat A et al. Gender-related differences in MEN1
lesion occurrence and diagnosis: a cohort study of 734 cases from the Groupe
Symptomatic metastatic disease confined to the liver may be d’etude des Tumeurs Endocrines. Eur J Endocrinol/Eur Feder Endocr Soc 2011;
treated with embolization, radiofrequency ablation (RFA) and 165(1): 97–105.
radio-embolization of liver metastases. External local 10. Phan AT, Oberg K, Choi J et al. NANETS consensus guideline for the diagnosis
irradiation of brain and bone metastases might be beneficial. In and management of neuroendocrine tumors: well-differentiated neuroendocrine
thymic NETs chemotherapy using cisplatinum-based regimens tumors of the thorax (includes lung and thymus). Pancreas 2010; 39(6):
has been of value. However, temozolomide-based treatment is 784–798.
also reported to give some benefit (III, B) [18, 19]. All 11. Granberg D, Sundin A, Janson ET et al. Octreoscan in patients with bronchial
treatment series are very small. Somatostatin analogs can be carcinoid tumors. Clin Endocrinol 2003; 59(6): 793–799.
12. Gabriel M, Decristoforo C, Kendler D et al. 68Ga-DOTA-Tyr3-octreotide PET in
used to control the Cushing’s syndrome related to thymic and
neuroendocrine tumors: comparison with somatostatin receptor scintigraphy and
bronchial NETs. PRRT is promising in thymic NETs but more CT. J Nucl Med 2007; 48(4): 508–518.
studies are needed [20]. 13. Rivera MP, Detterbeck F, Mehta AC. Diagnosis of lung cancer: the guidelines.
Chest 2003; 123(1 Suppl.): 129S–136S.
response evaluation and follow-up 14. Lim E, Goldstraw P, Nicholson AG et al. Proceedings of the IASLC International
Workshop on Advances in Pulmonary Neuroendocrine Tumors 2007. J Thorac
After primary surgery patients with TC and AC should be Oncol 2008; 3(10): 1194–1201.
followed at least yearly up to 15 years (III, B) to detect 15. Daddi N, Ferolla P, Urbani M et al. Surgical treatment of neuroendocrine tumors
surgically manageable recurrences. Biochemical markers, such of the lung. Eur J Cardiothorac Surg 2004; 26(4): 813–817.
as chromogranin A and NSE, should be determined every 3–6 16. Lim E, Yap YK, De Stavola BL et al. The impact of stage and cell type on the
prognosis of pulmonary neuroendocrine tumors. J Thorac Cardiovasc Surg 2005;
months (in cases with elevated values at baseline), and CT
130(4): 969–972.
should be performed once a year in atypical and every 2 or 3
17. Oberg K. Chemotherapy and biotherapy in the treatment of neuroendocrine
years in typical. Patients with metastatic or recurrent disease tumors. Ann Oncol 2001; 12(Suppl. 2): S111–S114.
should be followed during treatment with cytotoxic or 18. Ekeblad S, Sundin A, Janson ET et al. Temozolomide as monotherapy is effective
biological agents more often, at 3–6-month intervals with in treatment of advanced malignant neuroendocrine tumors. Clin Cancer Res
imaging, preferably by CT and biological markers to assess 2007; 13(10): 2986–2991.
possible benefits of the treatment administered [21]. 19. Hamada S, Masago K, Mio T et al. Good clinical response to imatinib mesylate in
atypical thymic carcinoid with KIT overexpression. J Clin Oncol 2011; 29(1):
e9–e10.
conflict of interest 20. van Essen M, Krenning EP, Bakker WH et al. Peptide receptor radionuclide
therapy with 177Lu-octreotate in patients with foregut carcinoid tumors of
Prof. Öberg has reported: speakers’ bureau and advisory board
bronchial, gastric and thymic origin. Eur J Nucl Med Mol Imaging 2007; 34(8):
membership: Ipsen, Novartis, Pfizer. Prof. Rougier has 1219–1227.
reported: honoraria from Sanofi Aventis, Amgen, Keocyte, 21. Warren WH, Gould VE. Long-term follow-up of classical bronchial carcinoid
Merck Serono, Pfizer, Roche and Lilly; advisory board for tumors. Clinicopathologic observations. Scand J Thorac Cardiovasc Surg 1990;
Sanofi Aventis and Keocyte. 24(2): 125–130.
Volume 23 | Supplement 7 | October 2012 doi:10.1093/annonc/mds267 | vii

by guest
on 05 February 2018
Neuroendocrine gastro-entero-pancreatic tumors:

K. Öberg1, U. Knigge2, D. Kwekkeboom3 & A. Perren4 on behalf of the ESMO Guidelines Working
Group*
1
Department of Endocrine Oncology, University Hospital, Uppsala University, Uppsala, Sweden; 2Department of Surgery, Rigshospitalet, Copenhagen, Denmark;
3
Department of Nuclear Medicine, Erasmus Medical Center, Rotterdam, The Netherlands; 4Institute of Pathology, University of Bern, Bern, Switzerland
incidence and epidemiology The family of neuroendocrine GEP-NETs constitutes a

heterogeneous group, but all share a common phenotype with
Neuroendocrine gastroenteropancreatic tumors (GEP-NETs) immunoreactivity for the so-called pan-neuroendocrine’
constitute a heterogeneous group of tumors with their origin in markers including chromogranin A and synaptophysin.
neuroendocrine cells of the embryological gut. Most Neuron-specific enolase (NSE) and CD56 are often positive in
commonly, the primary lesion is located in the gastric mucosa, GEP-NETs, but are not specific for this tumor entity. A
the small and large intestine, the rectum and pancreas. The detailed description of the macroscopic, microscopic and
crude incidence has significantly increased over the last years immunohistochemical findings is mandatory to support the
clinical practice
and is now estimated to be 5.25/100 000/year. The prevalence diagnosis of NETs and to allow a correct classification, staging
guidelines
has recently been calculated to 35/100 000/year. The incidence and grading. Specific staining for hormones, such as serotonin,
for small intestinal neuroendocrine tumor (NETs) (carcinoids) gastrin, insulin and glucagon, can be applied to confirm the
is estimated to be from 0.32/100 000/year (England) to 1.12/ source of a clinical symptomatology, but it must be pointed out
100 000/year (Sweden). The incidence for rectal tumors is 0.86/ that there is no reciprocal agreement, as there can be
100 000/year, for pancreatic 0.32/100 000/year and for gastric production of hormones without secretion. Therefore,
NETs 0.30/100 000/year. Neuroendocrine GEP tumors can immunohistochemical demonstration of a hormone alone is
appear at all ages, with the highest incidence being from the not proof of functionality of a NET. Immunohistochemistry for
fifth decade onward. The exception is the carcinoid of the Ki-67 (MIB-1) is mandatory to grade the tumor according to
appendix, which occurs with the highest incidence at ∼40 the new World Health Organization (WHO) classification.
years of age. There is a slight overall higher incidence of NETs GEP-NETs can be part of familial syndromes such as MEN-1,
for males (5.35) compared with females (4.76). Patients with VHL, tuberosclerosis and neurofibromatosis (NF1,2). Genetic
multiple endocrine neoplasia type 1 (MEN-1) or von Hippel– testing should be done according to the approved methodology
Lindau’s disease (VHL), may have a clinical onset 15–20 years and after genetic counseling. NETs arising at different
earlier than patients with corresponding sporadic type of anatomical sites of the digestive system represent tumor
neuroendocrine tumors [1]. entities that differ in their biology and clinical presentation
(Table 1) [2].
biology staging and risk assessment
Patients with clinical symptoms suggestive of neuroendocrine The new WHO classification presented in 2010 defines the
GEP-NET should be referred to a center with special interest entire group of tumors as neuroendocrine neoplasms and
in, and knowledge of, these diseases. Histological diagnosis is divides the tumors into NET G1, NET G2 and poorly
mandatory in all cases and is usually obtained on surgical or differentiated neuroendocrine carcinoma (NEC G3) (Table 2).
endoscopic biopsies or ultrasonography guided liver biopsies. The European Neuroendocrine Tumor Society has proposed a
tumor–node–metastasis staging and grading system for various
types of GEP-NETs (Tables 3–8) (II, A) [2–4]. Preoperative
L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland. E-mail: clinicalguidelines@esmo. staging should, whenever possible, include somatostatin
org receptor scintigraphy which can nowadays be replaced by
68
†
Gallium-DOTA-TOC/-NOC/-TATE positron emission
2012. This publication supersedes the previously published version—Ann Oncol 2010; tomography (PET) with higher spatial resolution and
21 (Suppl 5): v223–v227. quantification, which causes higher sensitivity and specificity.

by guest
on 05 February 2018
Table 1. Classification of neuroendocrine GEP tumors (GEP-NETs) Table 3. TNM classification for gastric endocrine tumors (European
by site of origin and by hormonal activity Neuroendocrine Tumor Society)
Intestinal neuroendocrine tumors (carcinoids, about 50% of GEP-NETs) T Primary tumor

• with carcinoid syndrome (30% of carcinoids) flushing, diarrhea,
endocardial fibrosis, wheezing caused by release of serotonin T0 No evidence of primary tumor
predominantly from liver metastases Tis In situ tumor/dysplasia (<0.5 mm)
• without carcinoid syndrome (70% of carcinoids) T1 Tumor invades lamina propria or submucosa and ≤1 cm
T2 Tumor invades muscularis propria or subserosa or >1 cm
Pancreatic endocrine tumors (PETs) (∼30% of GEP-NETs) T3 Tumor penetrates serosa
Nonfunctioning (45%–60% of PETs) T4 Tumor invades adjacent structures
Functioning (40%–55% of PETs) For any T, add (m) for multiple tumors
• Gastrinoma, excessive gastrin production, Zollinger–Ellison N Regional lymph nodes
syndrome NX Regional lymph nodes cannot be assessed
• Insulinoma, excessive insulin production, hypoglycemia syndrome N0 No regional lymph node metastases
• Glucagonoma, excessive glucagons production, glucagonoma N1 Regional lymph node metastases
syndrome M Distant metastases
• VIPoma, excessive production of vasoactive intestinal peptide (VIP), MX Distant metastases cannot be assessed
Watery diarrhea, hypokalemia–achlorhydria syndrome M0 No distant metastases
• PPoma, excessive PP production, (generally classified as M1 Distant metastases
nonfunctioning PETs)
TNM, tumor–node–metastasis.
• Somatostatinoma, excessive somatostatin production
• CRHoma, excessive corticotropin-releasing hormones production
• Calcitoninoma, excessive calcitonin production
• GHRHoma, excessive growth hormone-releasing hormone be of value as a general marker. For patients with small
production intestinal NETs (carcinoids), urine 5-hydroxy-indole-acetic-
• Neurotensinoma, excessive neurotensin production
acid is important and should be done in combination with the
• ACTHoma, excessive production of adrenocorticotropic hormone
pCgA assessments. For pancreatic NETs, the specific hormones
• GRFoma, excessive production of growth hormone-releasing factor
should be analysed in relation to clinical symptoms such
• Parathyroid hormone-related peptide tumor
gastrin for patients with Zollinger Ellison’s syndrome, insulin
with hypoglycemic syndromes, glucagon with glucagonoma
GEP-NETs, neuroendocrine gastroenteropancreatic tumors.
syndrome and VIP with the Verner Morrison syndrome.
Nonfunctioning pancreatic endocrine tumors may secrete
increased levels of pCgA as well as pancreatic polypeptide (PP)
Table 2. Gastro entero pancreatic neoplasms: WHO Classification (2010)
[8]. Rectal NETs are usually of the so-called nonfunctioning
type, but they often secrete hormones, such as PP, somatostatin
WHO 1 NET G1, Ki-67 ≤2% and PYY. The largest group of GEP-NETs, well differentiated
WHO 2 NET G2, Ki-67 3%–20%
(NETs) of the small intestine (carcinoids), present with the
WHO 3 NEC G3, Ki-67 >20%
carcinoid syndrome in ∼30% of the patients, including
MANEC
flushing, diarrhea and endocardial fibrosis. The syndrome is
Tumor-like lesions
caused by serotonin and peptide hormones released from liver
NEN, neuroendocrine neoplasms; NET, neuroendocrine tumor; NEC, metastases but not from the primary small intestinal tumor, as
neuroendocrine carcinoma; MANEC, mixed adenocarcinoma and the hormones released to the portal vessels as metabolized in
neuroendocrine carcinoma. the liver at by-pass (II, A). The 5-year survival rate for patients
with midgut carcinoid tumor has been 60% for all stages. In
dedicated centers, the 5-year survival rate now for metastatic
However, all tumors do not express a significant number of carcinoid tumors is ∼75%. The 5-year survival rate for patients
somatostatin type 2 receptors. Therefore, the technique should with endocrine pancreatic tumors is estimated to be 60%–
always be complemented with computed tomography (CT) or 100% for localized disease, 40% for regional and 25% for
magnetic resonance imaging (MRI) depending on the tumor metastatic and 80% for all stages. Similarly, in dedicated
location. PET scanning with specific tracers such as 11C-5HTP, centers, the 5-year survival rate for metastatic pancreatic NETs
18
F-DOPA or 18F-DG can further optimize the staging of the is above 60% (III, A) [9–12].
disease [4–7]. Endoscopy (gastroscopy, endoscopic
ultrasonography, colonoscopy, capsule endoscopy etc.) is often management of local/locoregional
of additional value (III, A). The imaging procedures should
always be complemented with biochemical analysis of relevant
disease
biomarkers such plasma chromogranin A ( pCgA), which is a All patients with small intestinal NETs should be considered
general NET marker. In patients with poorly differentiated G3 potential candidates for curative surgery and should be
tumors, pCgA is often normal, but plasma NSE can sometimes evaluated in an interdisciplinary setting including an
Volume 23 | Supplement 7 | October 2012 doi:10.1093/annonc/mds295 | vii

by guest
on 05 February 2018
Table 4. TNM classification for endocrine tumors of the duodenum/ Table 6. TNM classification for endocrine tumors of lower jejunum
ampulla/proximal jejunum (European Neuroendocrine Tumor Society) and ileum (European Neuroendocrine Tumor Society)
T Primary tumor T Primary tumor

TX Primary tumor cannot be assessed TX Primary tumor cannot be assessed
T0 No evidence of primary tumor T0 No evidence of primary tumor
T1 Tumor invades lamina propria or submucosa and has a size ≤1 cm T1 Tumor invades mucosa or submucosa and has a size ≤1 cm
T2 Tumor invades muscularis propria or size >1 cma T2 Tumor invades muscularis propria or size >1 cm
T3 Tumor invades pancreas or retroperitoneum T3 Tumor invades subserosa
T4 Tumor invades peritoneum or other organs T4 Tumor invades peritoneum/other organs
For any T, add (m) for multiple tumors For any T add (m) for multiple tumors
N Regional lymph nodes N Regional lymph nodes
NX Regional lymph nodes cannot be assessed NX Regional lymph nodes cannot be assessed
N0 No regional lymph node metastases N0 No regional lymph node metastases
N1 Regional lymph node metastases N1 Regional lymph node metastases
M Distant metastases M Distant metastases
MX Distant metastases cannot be assessed MX Distant metastases cannot be assessed
M0 No distant metastases M0 No distant metastases
M1 Distant metastases M1 Distant metastases
a
Tumor limited to ampulla of Vater for gangliocytic paraganglioma. TNM, tumor–node–metastasis.
Table 7. TNM classification for endocrine tumors of colon and rectum

(European Neuroendocrine Tumor Society)
Table 5. TNM classification for endocrine tumors of the pancreas
(European Neuroendocrine Tumor Society)
T Primary tumor
T Primary tumor
T1 Tumor invades mucosa or submucosa
T1a size ≤1 cm
T1 Tumor limited to the pancreas and size ≤2 cm
T1b size 1–2 cm
T2 Tumor limited to the pancreas and size 2–4 cm
T2 Tumor invades muscularis propria or size >2 cm
T3 Tumor limited to the pancreas and size >4 cm or invading
T3 Tumor invades subserosa, pericolic, and perirectal fat
duodenum or bile duct
T4 Tumor directly invades other organs/structures and/or perforates
T4 Tumor invading adjacent organs (stomach, spleen, colon, and
visceral peritoneum
adrenal gland) or the wall of large vessels (celiac axis or superior
For any T add (m) for multiple tumors
mesenteric artery)
N Regional lymph nodes
For any T, add (m) for multiple tumors
N Regional lymph nodes
NX Regional lymph node cannot be assessed
N1 Regional lymph node metastases
M Distant metastases
N1 Regional lymph node metastases
MX Distant metastases cannot be assessed
M Distant metastases
MX Distant metastases cannot be assessed
M1 Distant metastases
M1 Distant metastases TNM, tumor–node–metastasis.
Resection of the primary intestinal NET and regional lymph
node metastases in patients with distant metastases (liver) is
experienced surgeon. Curative resection of the primary tumor generally advocated to prevent later development of mesenteric
and locoregional lymph node metastases improves outcomes in fibrosis, small-bowel obstruction or painful vascular
these patients, resulting in excellent 5- and 10-year survivals of encasement. In addition, survival is prolonged in most studies,
100% in stage 1 and stage 2 patients, and still favorable but survival data are based on retrospective studies, which may
outcomes in stage 3 disease with 5- and 10-year survivals of have a patient selection bias ( patients with the best
more than 95% and 80%, respectively. Surgical procedures performance status are operated). Prospective randomized
include small intestinal resection or right hemicolectomy studies are needed. Large resections of the small intestinal
depending on the localization of the primary. Curative should be avoided as it may cause short-bowel syndrome.
resection also involves clearance of mesenterial and Postoperative mortality should be <1% and significant
retroperitoneal lymph node metastases by dissection around morbidity <10% (II, A). In patients with pancreatic NETs,
the mesentery, preserving the intestinal vascular supply. indications for surgery depend on clinical symptom control,
vii | Öberg et al. Volume 23 | Supplement 7 | October 2012

by guest
on 05 February 2018
Table 8. Gut endocrine tumors respectively. The duration of symptomatic response varied
between 14 and 20 months. The procedure is contraindicated
Grade Mitotic count (10 HPF)a Ki67 index (5)b in patients with complete portal vein thrombosis and poor
Grading proposal for neuroendocrine tumors liver function. Whipple procedure is also a contraindication for
G1 <2 ≤2 TACE/TAE since it increases the risk of morbidity. In the
G2 2–20 3–20 absence of comparative trials, it remains unclear whether
G3 >20 >20 TACE is preferred to TAE alone (III, B). Selective internal
irradiation therapy (SIRT) is still considered investigational.
Tumor grading and classification. ENETS grading proposal. Recent studies with 90Yttrium microspheres in about 200
a
10 HPF: 10 high power fields = 2 mm2, at least 40 fields (at ×40
patients with neuroendocrine tumors show objective response
magnification) evaluated in areas of highest mitotic density.
b
rates of 50%–60% in patients with liver metastases. Most data,
MIB1 antibody; percentage of 2000 tumor cells in areas of highest nuclear
however, are retrospective and derive from small phase II trials
labeling.
(III, B). A randomized trial between SIRT and TACE is
Virchows Arch (2006) 449:395–401.
warranted [16–19].
Virchows Arch (2007) 451:757–762.
medical therapy
tumor size/location, extent, malignancy and metastatic spread.
Curative surgery should be considered whenever possible even The use of somatostatin analogs is standard therapy in
in the presence of metastatic disease, including localized functioning NETs of any size [20, 21]. Interferon alpha may
metastatic disease to the liver if considered potentially also be considered for symptom control in some patients and
resectable and the patient can tolerate the surgery. The type of is usually used as second line therapy due to its less-favorable
surgery, the form of pancreatico-duodenal resection (Whipple’s toxic profile [22]. It has, sometimes, additional value as an
operation), distal pancreatic resection or a nucleation in add-on therapy in patients with clinical syndromes that are not
combination with resection, depends on the location of the controlled with somatostatin analogs. The antitumor efficacy of
primary tumor. As the malignancy is frequent in pancreatic somatostatin analogs appears weak with respect to objective
NETs, adequate lymph node clearance is mandatory (III, A). tumor response (5%–10%). However, disease stabilization of
Laparoscopic resection is not recommended because of the up to 50%–60% has been reported (III, A). In a prospective
need for lymphadenectomy and careful inspection for randomized placebo-controlled trial of octreotide long-acting
invasion/metastases. It is a general agreement not to operate on release (LAR), 30 mg every 4 weeks in small intestinal NET
G3 pancreatic NEC, as these tumors are widely metastasized at (the PROMID trial), an antiproliferative efficacy of octreotide
the timepoint of diagnosis (III, B) [13, 14, 15]. LAR has been confirmed. The median time to tumor
progression was 14.3 months with octreotide LAR versus 6
months with placebo [23]. Based on these results, the use of
management of advanced/metastatic somatostatin analogs, especially octreotide LAR, is
recommended for antiproliferative purposes in functioning and
disease nonfunctioning small intestinal tumors (carcinoids) (II, A).
Cytoreductive surgery should be considered when metastatic Somatostatin analogs are the recommended first line therapy in
disease is localized or if >70% of tumor load is thought nonfunctioning as well as functioning progressive G1/G2
resectable, which may decrease endocrine and local symptoms NETs. In contrast, in metastatic NEC G3 regardless of the site
and might help to improve systemic treatment. There are no of origin somatostatin analog treatment is not recommended
randomized clinical trials comparing the efficacy of (III, B). There is also no indication for adjuvant therapy with
locoregional therapies and palliative liver surgery [15]. The somatostatin analogs in NET G1/G2 irrespective of primary
choice of the ablative or locoregional procedure such as tumor origin and potential microscopic metastases (III, B).
radiofrequency ablation (RFA), laser-induced thermotherapy Other specific therapies in GEP-NETs are the mTOR-inhibitor
or selective hepatic transcatheter arterial embolization (TAE), everolimus, alone or in combination with a somatostatin
chemoembolization (TACE) and selective internal radiotherapy analog. In the RADIANT-2 trial which was a randomized
(SIRT) depends on the local expertise, number and size of phase III trial in patients with NETs (carcinoids), everolimus
lesions and location of liver involvement. These types of demonstrated a significant antitumor effect compared with
locoregional therapies are usually used in combination with placebo by local review but not by central review (I, A).
systemic medical treatment. If bulky disease is present, Clinically beneficial effects have been reported in carcinoid
locoregional therapy is indicated early also in nonfunctioning patients. In patients with pancreatic NETs, totally 410 patients
tumors and is used for down-staging of the disease. RFAs in who were randomized to either everolimus–octreotide or
tumors <5 cm in size have shown 70%–80% symptomatic placebo-octreotide, Radiant 3, significant prolonged PFS, 11
responses with control of symptoms up to 1 year (III, B). versus 4.6 months, was noticed and everolimus is now
Selective hepatic TAE or TACE with hepatic artery occlusion registered for treatment of pancreatic NETs worldwide (I, A)
can be applied in the treatment of liver metastases from all [24, 25]. Tyrosine kinase inhibitors, sunitinib and pazopanib,
types of neuroendocrine G1/G2-tumors. Complete or partial have demonstrated significant antitumor efficacy in pancreatic
response for symptoms, tumor markers and imaging occurred NETs. In a randomized trial, sunitinib (37.5 mg/day) was
in 70%–100%, 50%–90% and 30%–50% of the patients, compared with placebo in 170 patients. The study was
Volume 23 | Supplement 7 | October 2012 doi:10.1093/annonc/mds295 | vii

by guest
on 05 February 2018
terminated early due to a significant difference in efficacy cisplatinum/etoposide is recommended early. There is no
between the treatment and the placebo arms. A significant established second-line therapy for poorly differentiated
longer PFS, 11 versus 5.5 months was noticed in favor of endocrine carcinoma, but recent retrospective studies have
sunitinib (I, A) [26]. Also sunitinib is now registered demonstrated the efficacy of temozolomide alone or in
worldwide for the treatment of pancreatic NETs. Pazopanib combination with capecitabine ± bevacizumab (III, B).
has also demonstrated an antitumor effect in pancreatic NETs Encouraging results using either 5-FU i.v. or capecitabine
in small phase II trials, alone or in combination with orally combined with oxaliplatin or irinotecan may also be an
octreotide or bevacizumab. Pazopanib may be better tolerated option in the future (Table 9) [27–30].
than sunitinib in terms of side-effects.
systemic chemotherapy peptide receptor-targeted radiotherapy

Chemotherapy is recommended in NETs, metastatic NET G2 Promising data have evolved with regard to peptide receptor-
and in NEC G3 of any site. So far, results with systemic targeted radiotherapy (PRRT) in the treatment of NETs with
chemotherapy in classical carcinoid tumors (G1) are poor with liver metastases using 90Yttrium and 177Lutetium labeled
response rates <15% (III, A). Systemic cytotoxics are indicated DOTATOC or DOTATATE. PRRT can be considered in both
in patients with inoperable progressive liver metastases from functioning and nonfunctioning NETs with positive
G1/G2 pancreatic NETs using a combination of streptozotocin somatostatin receptor scintigraphy irrespective of the primary
and 5-fluorouracil (5-FU)/doxorubicin with objective response tumor site. Based on phase II trials, more than 1000 patients in
rates in the order of 35%–40% (II, B). From single total have been treated in Europe with objective response rates
retrospective trials, temozolomide-based chemotherapy is ranging between 20% and 40% (III, A). Response rates are
promising in pancreatic NETs either alone or combined with higher in pancreatic compared with small intestinal NETs (III,
capecitabine giving high partial remissions (40%–70%) (III, B). A). The highest [31] objective response rate has so far been
Prospective randomized trials are warranted. In cases of liver obtained in metastatic rectal NETs. Prospective randomized
metastases involving high-grade NEC G3 regardless of the site trials are still lacking but in progress. A treatment algorithm is
of the primary tumor combination chemotherapy, using depicted in Figure 1.
Table 9. Chemotherapy
Reference Type of tumor Regimen No. of Objective Response duration Median survival
patients response (months) (months)
Moertel et al. Pancreatic STZ 42 36 17 16.5
STZ–5-FU 42 63 17 26
Eriksson et al. Pancreatic STZ–5-FU or DOX 44 45 27.5 –
Moertel et al. Pancreatic STZ–DOX 36 69 18 26
STZ–5-FU 33 45 14 18
Cheng and Pancreatic STZ–DOX 16 6 18 –
Saltz
McCollum Pancreatic STZ–DOX 16 6 3.9 20.2
et al.
Kouvaraki et al. Pancreatic STZ–DOX–5-FU 84 39 9.3 40
Strosberg et al. Pancreatic Temozolomide–capecitabine 30 70 18 –
Moertel and Carcinoids 5-FU–cyclophosphamide 47 33 – –
Hanley STZ–5-FU 42 33 – –
Engstrom et al. Carcinoids STZ–5-FU 80 22 8 16
DOX 81 21 6.5 12
Bukowski et al. Carcinoids STZ–DOX–5-FU–cyclophosphamide 56 31 – –
STZ–5-FU–cyclophosphamide 9 22 – 10.8
Sun et al. Carcinoids DOX–5-FU 25 15.9 4.5 15.7
STZ–5-FU 27 16 5.3 24.3
Moertel et al. Poorly Cisplatin–etoposide 18 67 8 19
differentiated
Mitry et al. Poorly Cisplatin–etoposide 41 42 9 15
differentiated
Fjallskog et al. Poorly Cisplatin–etoposide 36 47 9 –
differentiated
Welin et al. Poorly Temozolomide ± capecitabine ± bevacizumab 25 33 19 22
differentiated
vii | Öberg et al. Volume 23 | Supplement 7 | October 2012

by guest
on 05 February 2018
Figure 1 Treatment algorithm.
response evaluation and follow-up progression or if imaging information is lacking, it may be

necessary to re-biopsy liver metastases to re-assess the
Follow-up investigations should include biochemical proliferative activity. If chromogranin A is not elevated NSE
parameters and conventional imaging. In patients with R0/R1 represents an alternative biomarker.
resected NET G1/G2, it is recommended that imaging is
performed every 3–6 months (CT or MRI), and in NEC G3,
every 2–3 months. Somatostatin receptor imaging, either conflict of interest
Octreoscan or PET/CT using 68Ga-DOTA-TOC/-NOC/-TATE
should be included in the follow-up and is recommended after Prof. Öberg has reported: speakers’ bureau and advisory board
18–24 months if expression of somatostatin receptor 2a has membership: Ipsen, Novartis, Pfizer. Prof. Rougier has
been proven on the tumor cells. In the case of rapid tumor reported: honoraria from Sanofi Aventis, Amgen, Keocyte,
Merck Serono, Pfizer, Roche and Lilly; advisory board for
Sanofi Aventis and Keocyte.
Table 10. Summary of recommendations Dr. Kwekkeboom has not reported any potential conflicts of
interests. The other authors have reported no potential
conflicts of interest.
• The diagnosis of NET should be confirmed by histopathology (CgA,
synaptophysin Ki-67).
• The current classification and staging systems should be applied in the references
clinic.
• Somatostatin receptor imaging besides standard imaging (CT and MRI) 1. Yao JC, Hassan M, Phan A et al. One hundred years after "carcinoid":
is part of standard of care. epidemiology of and prognostic factors for neuroendocrine tumors in 35,825
cases in the United States. J Clin Oncol 2008; 26: 3063–3072.
• Resection of locoregional disease in patients with small intestinal NET
2. Rindi G, Arnold R, Bosman FT et al. Nomenclature and classification of
(carcinoids) is recommended.
neuroendocrine neoplasms of the digestive system. In Bosman FT, Hruban RH,
• Somatostatin analog therapy is first-line therapy in all functional NET
Theise ND (eds), WHO Classification of Tumours of the Digestive System. Lyon:
and small intestinal NET G1/G2. IARC 2010; 13–14.
• Everolimus and sunitinib are registered for pancreatic NETs based on 3. Pape UF, Jann H, Muller-Nordhorn J et al. Prognostic relevance of a novel TNM
two phase III randomized trials. classification system for upper gastroenteropancreatic neuroendocrine tumors.
• Temozolomide alone or in combination with capecitabine is promising Cancer 2008; 113: 256–265.
for treatment of pancreatic NETs. 4. Rockall AG, Reznek RH. Imaging of neuroendocrine tumours (CT/MR/US). Best
Pract Res Clin Endocrinol Metab 2007; 21: 43–68.
Volume 23 | Supplement 7 | October 2012 doi:10.1093/annonc/mds295 | vii

by guest
on 05 February 2018
5. Sundin A, Vullierme MP, Kaltsas G et al. ENETS Consensus Guidelines for the 19. Kennedy AS, Dezarn WA, McNeillie P et al. Radioembolization for unresectable
Standards of Care in Neuroendocrine Tumors: radiological examinations. neuroendocrine hepatic metastases using resin 90Y-microspheres: early results
Neuroendocrinology 2009; 90: 167–183. in 148 patients. Am J Clin Oncol 2008; 31: 271–279.
6. Koopmans KP, Neels OC, Kema IP et al. Improved staging of patients with 20. Oberg K, Kvols L, Caplin M et al. Consensus report on the use of somatostatin
carcinoid and islet cell tumors with 18F-dihydroxy-phenyl-alanine and 11C-5- analogs for the management of neuroendocrine tumors of the
hydroxy-tryptophan positron emission tomography. J Clin Oncol 2008; 26: gastroenteropancreatic system. Ann Oncol 2004; 15: 966–973.
1489–1495. 21. Modlin IM, Pavel M, Kidd M et al. Review article: somatostatin analogues in the
7. Binderup T, Knigge U, Loft A et al. 18F-fluorodeoxyglucose positron emission treatment of gastroenteropancreatic neuroendocrine (carcinoid) tumours. Aliment
tomography predicts survival of patients with neuroendocrine tumors. Clin Cancer Pharmacol Ther 2010; 31: 169–188.
Res 2010; 16: 978–985. 22. Oberg K. Interferon in the management of neuroendocrine GEP-tumors: a review.
8. Eriksson B, Oberg K, Stridsberg M. Tumor markers in neuroendocrine tumors. Digestion 2000; 62(Suppl 1): 92–97.
Digestion 2000; 62(Suppl 1): 33–38. 23. Rinke A, Muller HH, Schade-Brittinger C et al. Placebo-controlled, double-blind,
9. Ekeblad S, Skogseid B, Dunder K et al. Prognostic factors and survival in 324 prospective, randomized study on the effect of octreotide LAR in the control of
patients with pancreatic endocrine tumor treated at a single institution. Clin tumor growth in patients with metastatic neuroendocrine midgut tumors: a report
Cancer Res 2008; 14: 7798–7803. from the PROMID Study Group. J Clin Oncol 2009; 27: 4656–4663.
10. Ahmed A, Turner G, King B et al. Midgut neuroendocrine tumours with liver 24. Pavel ME, Hainsworth JD, Baudin E et al. Everolimus plus octreotide long-acting
metastases: results of the UKINETS study. Endocr Relat Cancer 2009; 16: repeatable for the treatment of advanced neuroendocrine tumours associated
885–894. with carcinoid syndrome (RADIANT-2): a randomised, placebo-controlled, phase
11. Halfdanarson TR, Rabe KG, Rubin J et al. Pancreatic neuroendocrine tumors 3 study. Lancet 2011; 378: 2005–2012.
(PNETs): incidence, prognosis and recent trend toward improved survival. Ann 25. Yao JC, Shah MH, Ito T et al. Everolimus for advanced pancreatic neuroendocrine
Oncol 2008; 19: 1727–1733. tumors. N Engl J Med 2011; 364: 514–523.
12. Townsend A, Price T, Yeend S et al. Metastatic carcinoid tumor: changing 26. Raymond E, Dahan L, Raoul JL et al. Sunitinib malate for the treatment of
patterns of care over two decades. J Clin Gastroenterol 2010; 44: 195–199. pancreatic neuroendocrine tumors. N Engl J Med 2011; 364: 501–513.
13. Akerstrom G, Hellman P. Surgery on neuroendocrine tumours. Best Pract Res 27. Turner NC, Strauss SJ, Sarker D et al. Chemotherapy with 5-fluorouracil,
Clin Endocrinol Metab 2007; 21: 87–109. cisplatin and streptozocin for neuroendocrine tumours. Br J Cancer 2010; 102:
14. Kianmanesh R, O’Toole D, Sauvanet A et al. Surgical treatment of gastric, 1106–1112.
enteric pancreatic endocrine tumors. Part 2. treatment of hepatic metastases. J 28. Strosberg JR, Fine RL, Choi J et al. First-line chemotherapy with capecitabine
Chir (Paris) 2005; 142: 208–219. and temozolomide in patients with metastatic pancreatic endocrine carcinomas.
15. Sarmiento JM, Que FG. Hepatic surgery for metastases from neuroendocrine Cancer 2011; 117: 268–275.
tumors. Surg Oncol Clin N Am 2003; 12: 231–242. 29. Welin S, Sorbye H, Sebjornsen S et al. Clinical effect of temozolomide-based
16. Eriksson J, Stalberg P, Nilsson A et al. Surgery and radiofrequency ablation for chemotherapy in poorly differentiated endocrine carcinoma after progression on
treatment of liver metastases from midgut and foregut carcinoids and endocrine first-line chemotherapy. Cancer 2011; 117: 4617–4622.
pancreatic tumors. World J Surg 2008; 32: 930–938. 30. Sorbye H, Welin S, Langer S et al. NNTG: Nordic Neuroendocrine Tumor
17. Mazzaglia PJ, Berber E, Milas M et al. Laparoscopic radiofrequency ablation of Group. Predictive and prognostic factors for treatment and survival in
neuroendocrine liver metastases: a 10-year experience evaluating predictors of 305 patients with advanced gastrointestinal poorly differentiated neuroendocrine
survival. Surgery 2007; 142: 10–19. carcinoma: The NORDIC NEC study. ASCO 2012: abst #4015.
18. Vogl TJ, Naguib NN, Zangos S et al. Liver metastases of neuroendocrine 31. Kwekkeboom DJ, de Herder WW, Kam BL et al. Treatment with the radiolabeled
carcinomas: interventional treatment via transarterial embolization, somatostatin analog [177 Lu-DOTA 0,Tyr3]octreotate: toxicity, efficacy, and
chemoembolization and thermal ablation. Eur J Radiol 2009; 72: 517–528. survival. J Clin Oncol 2008; 26: 2124–2130.
vii | Öberg et al. Volume 23 | Supplement 7 | October 2012

by guest
on 05 February 2018
Cervical cancer: ESMO Clinical Practice Guidelines for

C. Marth1, F. Landoni2, S. Mahner3, M. McCormack4, A. Gonzalez-Martin5 & N. Colombo2, on behalf of the

1
Department of Obstetrics and Gynecology, Medical University Innsbruck, Innsbruck, Austria; 2Department of Gynecologic Oncology, European Institute of
Oncology, Milan, Italy; 3Department of Gynecology and Obstetrics, University of Munich, Munich, Germany; 4Department of Oncology, University College Hospital,
London, UK; 5Medical Oncology Department, MD Anderson Cancer Center, Madrid, Spain
†
Approved by the ESMO Guidelines Committee: January 2008, last update May 2017. This publication supersedes the previously published version—Ann Oncol 2012;
23(Suppl 7): vii27–vii32.
Incidence and epidemiology 5 more oncogenic HPV viruses (HPV 31, 33, 45, 52 and 58) in
Cervical cancer is the second most commonly diagnosed cancer addition to the types already included in 4vHPV, which cause an
and the third leading cause of cancer death among females in less additional 15% of HPV-related cancers in women and 4% of
developed countries. There were an estimated 527 600 new cer- those in men [4]. Both the 2vHPV and 4vHPV have significant
vical cancer cases and 265 700 deaths worldwide in 2012 [1]. cross-protective activity against other oncogenic viruses. All three
Nearly 90% of cervical cancer deaths occurred in developing parts are efficacious against related infection and cervical, vaginal, vul-
of the world. The large geographic variation in cervical cancer var and anal dysplasia [5–7].
rates reflects differences in the availability of screening (which Post-licensure reports from countries with established HPV
allows for the detection and removal of precancerous lesions) and vaccination programs indicate that HPV vaccination has a bene-
in human papillomavirus (HPV) infection prevalence. ficial effect at the population level as early as 3 years after the
However, cervical cancer still represents a major public health introduction of an HPV vaccination programme, including de-
problem even in developed countries: more than 58 000 new cases creases in the incidence of high-grade cervical abnormalities, the
of cervical cancer are diagnosed and 24 000 patients die in prevalence of vaccine HPV types and the incidence of genital
Europe every year [2]. Five-year relative survival for European warts [8, 9]. Prophylactic administration of HPV vaccine can ef-
women diagnosed with cervical cancer in 2000–2007 was 62%, fectively prevent infection and disease associated with the vaccine
ranging from 57% in Eastern Europe to 67% in Northern HPV types. The effect of vaccination on the burden of cancer re-
Europe. Survival was particularly low (< 55%) in Bulgaria, Latvia mains to be determined but, according to surrogate markers, it is
and Poland and highest in Norway (71%) [3]. Survival decreased expected to prevent > 70% of cervical cancers.
with advancing age at diagnosis, from 81% for 15–44-year olds to For many years, the Papanicolaou (Pap) test has been the
34% for women 75 years. Survival increased significantly from standard method for cervical cancer screening, reducing the inci-
61% in 1999–2001 to 65% in 2005–2007. FIGO stage is one of the dence by 60%–90% and the death rate by 90%. However, the
most important prognostic factors. limitations of this cytology-based test are the sensitivity ( 50%)
The most significant cause of cervical cancer is persistent papil- and significant proportion of inadequate specimens. More re-
lomavirus infection. HPV is detected in 99% of cervical tumours, cently, an HPV test has been introduced as a screening tool as
particularly the oncogenic subtypes such as HPV 16 and 18. HPV deoxyribonucleic acid (DNA) is present in almost all cer-
To date, three HPV vaccines are licensed and available: the bi- vical cancers and it has demonstrated higher sensitivity for high-
valent HPV virus-like particle vaccine (2vHPV), the quadrivalent grade cervical intraepithelial neoplasia (CIN2þ) than that
HPV virus-like particle vaccine (4vHPV) and nine-valent HPV achieved by cytology in several studies. A pooled analysis of four
virus-like particle vaccine (9vHPV). All 3 vaccines provide pro- randomised controlled trials of HPV-based cervical screening
tection against HPV 16 and 18. 4vHPV also includes HPV 6 and versus conventional cytology showed that HPV-based cervical
11 which cause 90% of genital warts. Furthermore, 9vHPV covers screening provides 60%–70% greater protection against invasive
V

by guest
on 05 February 2018
cancer compared with cytology-based screening [10]. Findings Table 1. WHO histological classification of tumours of the uterine cervix
support HPV-based screening with triage at prolonged inter-
Epithelial tumours
vals, starting at age 30 years. Especially in a vaccinated popula-
1. Squamous tumours and precursors
tion when dysplastic lesions will be less frequent, screening
Squamous cell carcinoma, not otherwise specified 8070/3
with Pap tests will be more difficult. Pap cytology has signifi-
Keratinising 8071/3
cant limitations. It is based on the subjective interpretation of Non-keratinising 8072/3
morphological alterations present in cervical samples that must Basaloid 8083/3
be collected with proper attention to sampling cells of the Verrucous 8051/3
transformation zone. Also, the highly repetitive nature of the Warty 8051/3
work of screening many smears leads to fatigue, which invari- Papillary 8052/3
ably causes errors in interpretation. Lymphoepithelioma-like 8082/3
Therefore, primary prevention of cervical cancer is now pos- Squamotransitional 8120/3
sible via immunisation with highly efficacious HPV vaccines [II, Early invasive (microinvasive) squamous cell carcinoma 8076/3
A] and secondary prevention has gained impetus with the advent Squamous intraepithelial neoplasia
of sensitive HPV DNA testing to improve traditional Pap cy- Cervical intraepithelial neoplasia (CIN) 3/ 8077/2
tology screening programmes [II, A]. squamous cell carcinoma in situ 8070/2
Benign squamous cell lesions
Condyloma acuminatum
Squamous papilloma 8052/0
Diagnosis and pathology/molecular biology Fibroepithelial polyp
Abnormal cervical cytology or a positive high-risk HPV test 2. Glandular tumours and precursors
Adenocarcinoma 8140/3
should lead to colposcopy and biopsy or excisional procedures
Mucinous adenocarcinoma 8480/3
such as loop electrosurgical excision and conisation. Early cer-
Endocervical 8482/3
vical cancer is often asymptomatic, while locally advanced disease
Intestinal 8144/3
could cause symptoms including abnormal vaginal bleeding (also
Signet-ring cell 8490/3
after coitus), discharge, pelvic pain and dyspareunia. Gross ap- Minimal deviation 8480/3
pearance is variable. Carcinomas can be exophytic, growing out Villoglandular 8262/3
of the surface, or endophytic with stromal infiltration with min- Endometrioid adenocarcinoma 8380/3
imal surface growth. Some early cancers are not easily detected Clear cell adenocarcinoma 8310/3
and even deeply invasive tumours may be somewhat deceptive on Serous adenocarcinoma 8441/3
gross examination. If examination is difficult or there is uncer- Mesonephric adenocarcinoma 9110/3
tainty about vaginal/parametrial involvement, examination Early invasive adenocarcinoma 8140/3
should preferably be done under anaesthesia by an interdisciplin- Adenocarcinoma in situ 8140/2
ary team including a gynaecological oncologist and a radiation Glandular dysplasia
oncologist. Benign glandular lesions
The World Health Organization (WHO) recognises three cate- Müllerian papilloma
gories of epithelial tumours of the cervix: squamous, glandular Endocervical polyp
(adenocarcinoma) and other epithelial tumours including 3. Other epithelial tumours
adenosquamous carcinoma, neuroendocrine tumours and undif- Adenosquamous carcinoma 8560/3
ferentiated carcinoma (Table 1). Squamous cell carcinomas ac- Glassy cell carcinoma variant 8015/3
count for 70%–80% of cervical cancers and adenocarcinomas Adenoid cystic carcinoma 8200/3
for 20%–25%. Adenoid basal carcinoma 8098/3
Neuroendocrine tumours
Carcinoid 8240/3
Squamous cell carcinoma Atypical carcinoid 8249/3
Small cell carcinoma 8041/3
Squamous carcinomas are composed of cells that are recognisably
Large cell neuroendocrine carcinoma 8013/3
squamous but vary in either growth pattern or cytological Undifferentiated carcinoma 8020/3
morphology. Originally, they were graded using Broders’ grading
system; subsequently, they were classified into keratinising, non- Mesenchymal tumours and tumour-like conditions
keratinising and small-cell squamous carcinomas. In the more re- Mixed epithelial and mesenchymal tumours
cent WHO classification, the term small-cell carcinoma was Melanocytic tumours
Miscellaneous tumours
reserved for tumours of neuroendocrine type. Keratinising squa-
Lymphoid and haematopoetic tumours
mous cell carcinomas are characterised by the presence of keratin
Secondary tumours
pearls. Mitoses are not frequent. Non-keratinising squamous cell
carcinomas do not form keratin pearls by definition, but may Morphology code of the International Classification of Diseases for
show individual cell keratinisation. Clear-cell changes can be Oncology (ICD-O) {921} and the Systematized Nomenclature of
prominent in some tumours and should not be misinterpreted as Medicine (http://snomed.org).
clear-cell carcinoma. WHO, World Health Organization.

by guest
on 05 February 2018
Adenocarcinoma International Cancer Control (UICC) TNM staging classifications
(8th edition) shown in Table 2. Cervical cancer is the only gynaeco-
The arrangement of the invasive glands is highly variable and
logical cancer that is clinically staged based on tumour size, vaginal
some tumours are in part or extensively papillary. About 80% of
or parametrial involvement, bladder/rectum extension and distant
adenocarcinomas of the cervix are of endocervical or usual type;
metastases. It requires examination under anaesthesia, radiological
unlike normal endocervical mucinous epithelium, tumour cells
imaging such as chest X-ray and intravenous pyelogram. These
are not obviously mucinous and show a rather characteristic ap-
have been widely replaced by more timely diagnostic tools. Other
pearance having eosinophilic cytoplasm. The most common type
imaging studies are used routinely to more accurately define the
is the mucinous type which comprises endocervical, intestinal
extent of disease and to allow tailoring of treatment, but do not
and gastric subtypes. The great majority of endocervical-type
affect the clinical stage. Computed tomography (CT) can detect
adenocarcinomas are architecturally well differentiated, but they
pathological lymph nodes, while magnetic resonance imaging
are cytologically grade 2 or 3. Only a subset of papillary or villo-
(MRI) can determine tumour size, degree of stromal penetrations,
glandular adenocarcinoma is considered well differentiated for
parametrial involvement, vaginal extension and corpus extension
their good prognosis when in pure form; tumours with an under-
with high accuracy [14]. More recently, positron emission tomog-
lying component of conventional adenocarcinoma behave as
raphy (PET) has been seen to have the potential to accurately de-
adenocarcinomas of the usual type. Unlike cervical squamous cell
lineate the extent of disease, particularly in lymph nodes that are
carcinomas, differential diagnosis of early invasive adenocarcin-
not macroscopically enlarged and in distant sites, with high sensi-
oma from adenocarcinoma in situ showing somewhat complex
tivity and specificity. In early-stage disease, PET/CT has a sensitiv-
architecture can be difficult. In mucinous adenocarcinoma
ity of 53%–73% and specificity of 90%–97% for the detection of
mucin-rich cells predominate; some show gastric-type features
lymph node involvement, while in more advanced stages the sensi-
and some are of the minimal deviation type (or adenoma
tivity for detecting the involvement of para-aortic nodes increases
malignum). Rare tumours are mixed adenosquamous carcin-
to 75% with 95% specificity [15]. The need for pretreatment
omas and include the so-called glassy cell carcinoma. The other
surgical para-aortic lymph node assessment in locally advanced
rarer types of cervical adenocarcinoma include clear-cell carcin-
cervical cancer (LACC) is still a matter of debate [16].
oma and mesonephric adenocarcinoma.
Tumour risk assessment includes tumour size, stage, depth of tu-
mour invasion, lymph node status, lymphovascular space invasion
Other cervical carcinoma (LVSI) and histological subtype. Lymph node status and number of
Neuroendocrine tumours include carcinoids, atypical carcinoids lymph nodes involved are the most important prognostic factors.
and neuroendocrine carcinomas. Diagnosis is histological and In stages IB–IIA, the 5-year survival rates without lymph node me-
can be confirmed by neuroendocrine markers. tastasis and with lymph node metastasis are 88%–95% and 51%–
78%, respectively [17].
Controversy exists as to whether histological type is an inde-
Pathogenesis—molecular biology pendent prognostic factor for survival. Although some studies
HPV has been recognised as the most important aetiological fac- have shown no differences in survival between adenocarcinoma
tor in cervical cancer. HPV 16/18 account for at least two-thirds and squamous cell carcinoma, the majority have shown that
of cervical carcinomas in all continents; HPV 31, 33, 35, 45, 52 adenocarcinoma carries a worse prognosis with 10%–20% differ-
and 58 are the next most common types of cancers globally. ences in 5-year overall survival (OS) rates.
HPV vaccines have a great impact on cervical cancer and HPV- Cervical small-cell neuroendocrine carcinoma is a rare disease,
associated cancers in males and females, thus leading to an annual re- accounting for only up to 2% of all invasive cervical cancers but
duction of 90% of cervical cancer, 85% of vaginal cancer HPV correl- has a particular propensity to spread distantly, which is similar to
ates, 87% of vulvar cancer HPV correlates, 92% of anal cancer HPV small-cell carcinoma of the lung. As a result, patients can present
correlates and 85% of penile cancer HPV correlates [11–13]. with systemic symptoms such as weight loss. In addition, patients
Squamous cell carcinomas and their precursor, intraepithelial may present with a paraneoplastic syndrome such as the syn-
squamous lesions, are related to HPV infection in almost all cases drome of inappropriate antidiuretic hormone secretion
and the presence of HPV 16 DNA is associated with poor prognosis. (SIADH), Cushing syndrome, hypercalcaemia or a neurological
Adenocarcinomas encompass a heterogeneous group of tumours. disorder. The most commonly involved organs include the liver,
Endocervical adenocarcinoma of usual type and its precursor, the adrenals, bone, bone marrow and the brain.
adenocarcinoma in situ, have been shown to be positive for HPV in
nearly 90% and 100% of cases, respectively. HPV 18 is more com-
mon in adenocarcinomas and adenosquamous carcinomas than in
squamous cell carcinomas. Unlike endocervical adenocarcinoma of
usual type, the other rarer types including clear-cell and mesoneph- (Figure 1)
ric adenocarcinoma seem to be unrelated to HPV.
Primary treatment
Surgery. Surgical therapy in cervical cancer is adapted to the stage

Staging and risk assessment of disease according to FIGO and TNM classification (Table 2).
Cervical tumours are staged using the Féderation Internationale Microinvasive cervical cancer (stage IA1) without LVSI can be
de Gynécologie et d’Obstétrique (FIGO) and the Union for managed with conisation or simple trachelectomy to preserve
iv74 | Marth et al. Volume 28 | Supplement 4 | August 2017

by guest
on 05 February 2018
Table 2. The staging of cervical tumours is by the Féderation Internationale de Gynécologie et d’Obstétrique (FIGO) and TNM classification (Union for
International Cancer Control) [61]
TNM clinical classification
TNM categories FIGO stages Definition
T – Primary Tumour
Tis Carcinoma in situ (preinvasive carcinoma)
T1 I Tumour confined to the cervixa
T1ab,c IA Invasive carcinoma diagnosed only by microscopy. Stromal invasion with a maximal depth of
5.0 mm measured from the base of the epithelium and a horizontal spread of 7.0 mm or lessd
T1a1 IA1 Measured stromal invasion 3.0 mm or less in depth and 7.0 mm or less in horizontal spread
T1a2 IA2 Measured stromal invasion more than 3.0 mm and not more than 5.0 mm with a horizontal
spread of 7.0 mm or lessd
T1b IB Clinically visible lesion confined to the cervix or microscopic lesion greater than T1a/IA2
T1b1 IB1 Clinically visible lesion 4.0 cm or less in greatest dimension
T1b2 IB2 Clinically visible lesion more than 4.0 cm in greatest dimension
T2 II Tumour invades beyond uterus but not to pelvic wall or to lower third of vagina
T2a IIA Tumour without parametrial invasion
T2a1 IIA1 Clinically visible lesion 4.0 cm or less in greatest dimension
T2a2 IIA2 Clinically visible lesion more than 4.0 cm in greatest dimension
T2b IIB Tumour with parametrial invasion
T3 III Tumour involves lower third of vagina, or extends to pelvic wall, or causes hydronephrosis or
non-functioning kidney
T3a IIIA Tumour involves lower third of vagina
T3b IIIB Tumour extends to pelvic wall, or causes hydronephrosis or non-functioning kidney
T4 IVA Tumour invades mucosa of the bladder or rectum, or extends beyond true pelvise
N – Regional Lymph Nodesf
N1 Regional lymph node metastasis
M – Distant Metastasisf
M1 Distant metastasis (includes inguinal lymph nodes and intraperitoneal disease). It excludes
metastasis to vagina, pelvic serosa, and adnexa
a
Extension to corpus uteri should be disregarded.
b
The depth of invasion should be taken from the base of the epithelium, either surface or glandular, from which it originates. The depth of invasion is
defined as the measurement of the tumour from the epithelial–stromal junction of the adjacent most superficial papillae to the deepest point of inva-
sion. Vascular space involvement, venous or lymphatic, does not affect classification.
c
All macroscopically visible lesions even with superficial invasion are T1b/IB.
d
Vascular space involvement, venous or lymphatic, does not affect classification.
e
Bullous oedema is not sufficient to classify a tumour as T4.
f
No FIGO equivalent.
TNM, tumour, node and metastasis.
Reprinted from [61] with permission from John Wiley & Sons, Inc.
fertility [I, B] [18]. Simple hysterectomy can be offered if the invasive approach is gaining increasing relevance and is standard
patient does not wish to preserve fertility. In stage IA1 with LVSI, in most centres, since it appears to offer similar oncological safety
surgical assessment of pelvic lymph nodes should be discussed with favourable surgical morbidity [19].
with the patient, including the sentinel lymph node (SLN, see
below). Sentinel lymph node dissection in cervical cancer. SLN dissec-
In patients with FIGO stage IA2, IB and IIA, radical hysterec- tion (SLND) is standard in the treatment of breast cancer as
tomy with bilateral lymph node dissection (with or without SLN) well as vulvar cancer and increasing evidence also suggests an
is standard treatment, if the patient does not wish to preserve fer- important role for SLND in cervical cancer. While the evidence
tility [I, B]. This can be carried out either by laparotomy or lapar- is still evolving and guideline recommendations are not yet
oscopy (which can be robotically assisted). The minimally clearly defined, it should be considered in FIGO stage I

by guest
on 05 February 2018
by guest
on 05 February 2018
Positive Pap – smear / HPV – high risk positive / suspected cervix
iv76 | Marth et al.

Colposcopy / Biopsy
CIN2 / CIN3 Invasive cervical cancer Locally advanced disease Metastatic disease
Conisation Early disease

FIGO IA1 FIGO IA2 FIGO IB2 + IIA FIGO IB2 / IIB / IIIB FIGO IVA FIGO IVB

No LVSI: simple hysterectomy CRT Chemotherapy Chemotherapy
Frozen section: conisation Radical hysterectomy Radical hysterectomy + Tailoring of radiation (C)RT + bevacizumab
only if negative margins Simple hysterectomy PLND ± PALND according to surgical Pelvic exenteration RT
Trachelectomy + SLN staging or PET-CT
With LVSI: simple PLND ± PALND Frozen section: trachelectomy
hysterectomy + PLND ± SLN CRT Neoadjuvant chemotherapy
PALND followed by surgery
Trachelectomy or RT
Adjuvant treatment Adjuvant treatment

depending on depending on
risk factors risk factors
Figure 1. Treatment algorithm for cervical cancer.

CIN, cervical intraepithelial neoplasia; CRT, chemoradiotherapy; FIGO, Féderation Internationale de Gynécologie et d’Obstétrique; HPV, human papillomavirus; LVSI, lymphovascular space in-
vasion; PALND, para-aortic lymph node dissection; Pap, Papanicolaou; PET-CT, positron emission tomography/computed tomography; PLND, pelvic lymph node dissection; RT, radiotherapy;
SLN, sentinel lymph node.
Annals of Oncology

patients with tumours of 4 cm. Some evidence suggests that These results indicate that NACT may offer a benefit over sur-
the detection rate is highest if the tumour is < 2 cm. Tracer is gery alone in cervical cancer patients (borderline LACC, nodes
injected directly into the cervix, and blue dye, technetium positive, parametrial invasion at MRI), reducing the need for ad-
radiocolloid or fluorescent indocyanine green is used. SLND juvant RT [I, C].
should be done only in centres with enough expertise
and training. Sentinel nodes should be detected on both sides Chemoradiotherapy in locally advanced cervical cancer. CRT has
[II, B] [20]. been the standard of care for patients with bulky IB2–IVA disease
for almost two decades. The near simultaneous publication of five
Surgical therapy of the uterus. Since radiotherapy (RT) and randomised trials, three in LACC, collectively demonstrating an
surgery are equally effective in early stages, surgery should only be improvement in both disease-free survival (DFS) and OS with con-
considered in patients with earlier stages (up to FIGO IIA) with- comitant chemotherapy and RT over standard RT/hydroxyurea
out risk factors necessitating adjuvant therapy, which results in a (endorsed by the National Cancer Institute) changed clinical prac-
multimodal therapy without improvement of survival but tice worldwide [I, A] [29–33]. However, concerns were raised
increased toxicity [I, A]. about the applicability of the results in view of patient selection,
It is important to note that the currently established radical hys- protracted overall treatment time, the lack of a RT-only control
terectomy with extensive parametrial resection most likely consti- arm and the poor outcome in the control group. An individual pa-
tutes overtreatment in many patients, especially those with small tient data meta-analysis was undertaken to address these issues
and locally restricted tumours. Large randomised studies such as [34]. The authors identified 18 randomised trials with an RT-only
the SHAPE study are currently enrolling patients to compare simple control arm from 11 countries with the subsequent analysis limited
hysterectomy with radical hysterectomy in this population [21]. to 13 trials. The analysis confirmed the benefit of CRT but with a
smaller effect. The HR for OS and DFS was 0.81 and 0.78, respect-
Neoadjuvant chemotherapy to surgery. The rationale for the use ively, which translates into an absolute improvement of 6% and
of neoadjuvant chemotherapy (NACT) includes: (i) reduction of 8% in OS and DFS, respectively. The estimated absolute survival
the primary tumour size, allowing operability; (ii) eradication of benefit for CRT compared with RT alone was 10% for those with
micrometastatic disease; and (iii) potential increase in tumour FIGO stage I/II disease, compared with 3% for those with FIGO
vascularisation and reduction of the number of hypoxic cells [22–24]. stage III/IVa. The most commonly used regimen is weekly cisplatin
In a meta-analysis, NACT followed by radical surgery showed a 40 mg/m2, although the meta-analysis also reported significant
highly significant 35% reduction in the risk of death compared with benefits with non-platinum agents [I, A] [34].
RT alone [hazard ratio (HR) ¼ 0.65; P ¼ 0.0004], with an absolute More recently, colleagues in Mexico reported on a large
improvement of 14% in survival at 5 years, increasing from 50% to randomised phase III trial comparing standard CRT with a more
64% [25]. The analysis included data from 872 patients with LACC intensive concomitant approach with gemcitabine/cisplatin fol-
enrolled in five different trials. The largest trial included in a second lowed by an additional two cycles of adjuvant chemotherapy
meta-analysis, enrolled 441 FIGO stage IB2–III cervical cancer pa- [35]. Yet, despite a reported 9% improvement in progression-
tients and compared platinum-based NACT followed by radical free survival (PFS) at 3 years with treatment intensification, this
surgery with conventional RT. The main criticism of this study is approach has not been widely adopted amid concerns about tox-
related to the suboptimal RT administration; almost 27% of pa- icity [II, C]. Meanwhile two international trials of additional
tients did not receive intracavitary RT; 11% of patients received less chemotherapy delivered either before (INTERLACE) or after
than 60 Gy of external pelvic beam radiation total dose at point A CRT (OUTBACK) are ongoing and will hopefully answer the
and the median total dose delivered was 70 Gy, while the optimal question as to whether this approach will improve OS further.
treatment is considered to be 80–90 Gy at point A. Technical advances in imaging and in RT planning have facili-
Moreover, in all of these studies, the control arm, RT alone tated a move towards increased precision in brachytherapy prac-
without concomitant chemotherapy, does not represent the cur- tice. This approach has been championed by groups in Austria,
rent standard of care for LACC. In addition, the RT total dose Denmark and France with the dual aim of improving outcome
and the median time of RT administration were sometimes through dose escalation while reducing the toxicity to the sur-
suboptimal. rounding normal tissues [36]. A recently published multicentre co-
There are two randomised phase III trials that have explored the hort study (RetroEMBRACE) demonstrated excellent local control
role of NACT followed by surgery versus chemoradiotherapy (CRT), rates of 93% and 79% for patients with FIGO stage IIB and IIIB
but the results are not yet available (EORTC Protocol 55994 and disease, respectively, at 3 years [37]. However, the 5-year actuarial
NCT00193739) [26, 27]. OS was 65% and, while this is better than historical controls, it re-
Moreover, a recent meta-analysis comparing NACT followed mains to be seen whether this truly represents an improvement in
by surgery versus surgery alone confirmed that patients treated survival over standard CRT with lower RT doses. With a median
with NACT had higher local control [odds ratio (OR) ¼ 0.67; follow up of 43 months, the actuarial 5-year G3–G5 morbidity was
95% confidence interval (CI): 0.45–0.99; P ¼ 0.04)] [28]. 5%, 7% and 5% to the bladder, gastrointestinal tract and vagina,
Exploratory analysis of pathological response showed a signifi- respectively, confirming that the improved local control was
cant decrease in adverse pathological findings with NACT achieved with a low risk of morbidity [I, B]. Given the rarity of
(OR ¼ 0.54; P < 0.0001 for lymph node status; OR ¼ 0.58; small-cell neuroendocrine carcinoma, there are limited data to
P ¼ 0.002 for parametrial infiltration). However, a significant guide treatment of this type of cervical cancer. Most clinicians
percentage of patients will not have surgery because of treatment favour: the use of combined modality therapy (surgery followed by
toxicity or insufficient response. chemotherapy or combined CRT) for limited-stage potentially

by guest
on 05 February 2018
resectable disease; definitive CRT for locoregionally advanced Table 3. Necessary histopathological parameters for assessment of
unresectable but non-metastatic disease; and palliative chemother- cervical cancer
apy alone for those with metastatic disease, using chemotherapy
Histopathological evaluation
regimens that are typically used for small-cell lung cancer.
Dimensions of the tumour
Stromal invasion/depth of the wall involved
Neoadjuvant chemotherapy and radiotherapy. The concept of Tumour differentiation
delivering chemotherapy before RT (neoadjuvant or induction Lymphovascular space invasion
chemotherapy) has been explored in clinical trials with conflicting Status of resection margins
results. An individual patient data meta-analysis was undertaken Status of parametria and vaginal cuff
of 18 randomised trials involving over 2000 patients [38]. Number and status of lymph nodes
Heterogeneity in trial design precluded a unified analysis.
However, the authors identified cycle length and platinum dose in-
tensity as important factors in determining the impact of NACT recurrence rate alone but, when combined, the risk of recurrence is
on outcome. The trials that delivered short-cycle chemotherapy increased to 15%–20%, similar to that of high-risk factors.
( 14 days) gave a pooled HR of 0.83, equivalent to an improve-
ment of 7% in 5-year survival. In contrast, the trials with longer Intermediate-risk disease. A Gynecologic Oncology Group
chemotherapy cycles (> 14 days) gave a pooled HR of 1.25, equiva- (GOG) trial that randomly assigned 277 women to receive pelvic
lent to an absolute detriment in survival of 8% at 5 years. RT (without chemotherapy) or no further treatment demon-
Accelerated repopulation of resistant cancer cells during prolonged strated a benefit for postoperative RT in women with the follow-
intervals (up to 6 weeks in some studies) between NACT and RT ing features: deep cervical stromal invasion (to the middle or
may account for the detrimental effect seen in some studies. The one-third depth), LVSI and large tumour size (> 4 cm). With a
ongoing INTERLACE trial, which is randomising patients with median follow-up of 10 years, a significant benefit has been
LACC between standard CRT alone and 6 weeks of induction shown for PFS (HR 0.54), but not for OS (HR 0.7; P ¼ 0.07) [42].
chemotherapy followed immediately in week 7 by standard CRT,
seeks to address some of these issues by studying the use of a dose-
High-risk disease. Women with one or more negative prognostic
dense schedule, incorporating a taxane and eliminating the interval
factors such as positive or close surgical margins, positive lymph
between induction chemotherapy and RT.
nodes or microscopic parametrial involvement are considered to
be at high risk of relapse. In this setting of patients, adjuvant CRT
Lymph node staging and radiotherapy. In patients with LACC, is indicated based on a clinical trial that randomly assigned 268
RT treatment planning relies on accurate staging information. women IA2, IB and IIA to adjuvant RT with or without chemo-
Pelvic MRI and clinical examination is essential to determine the therapy (cisplatin–5-fluorouracil) for four courses [33]. The use
local extent of the tumour for both external beam RT and brachy- of chemotherapy was associated with a substantially better 4-year
therapy planning. Information on para-aortic nodal status is also OS (81% versus 71%) and PFS (80% versus 63%) and the out-
essential for treatment planning, particularly in determining the come was better for patients who completed three to four cycles
superior extent of the external beam RT portal. FIGO staging of chemotherapy [I, A].
does not take account of the nodal status and this is one of the Cervical cancer patients with intermediate-risk disease do not
weaknesses of this system. Surgical series suggest that the inci- need further adjuvant therapy [II, B], whereas adjuvant CRT is
dence of para-aortic nodal involvement increases with stage from recommended in high-risk patients [I, A].
about 5% in patients with stage I disease to 25% in those with
stage III disease [39].
There is much debate concerning the best way to assess the Management of advanced/metastatic
para-aortic nodes. In some parts of the world, PET/CT is rou-
tinely used for staging while elsewhere there is a reliance on surgi- disease (Figure 1)
cal exploration of the para-aortic region. It is hoped that the Metastatic or recurrent cervical cancer is usually a symptomatic
ongoing randomised trials will address this issue further [40]. and devastating situation for the patient. Palliative chemotherapy
This is particularly important in the light of the findings from a with the aim of relieving symptoms and improving quality of life
multicentre, randomised trial demonstrating an excellent out- is indicated if the patient has a performance status (PS) 2 and
come in patients with negative PET scans and metastasis 5 mm no formal contraindications. Cisplatin 50 mg/m2 every 3 weeks
detected histologically after surgical removal and subsequently was, for two decades, the standard of care. However, the global ef-
treated with extended-field CRT [41]. ficacy was disappointing due to a low response rate (20%), short
median PFS (2.8–3.2 months) and OS (6.2–8.0 months).
Cisplatin-based doublets with topotecan or paclitaxel have
Adjuvant treatment demonstrated superiority to cisplatin monotherapy in terms of
Women with risk factors on the pathology specimen should re- response rate and PFS [43, 44]. Cisplatin combined with topote-
ceive adjuvant therapy following hysterectomy (Table 3). Two can showed superior OS compared with cisplatin alone. Both tri-
classes of risk are defined: intermediate- and high-risk patients. als also demonstrated that the response rate was clearly inferior in
However, intermediate-risk factors such as LVSI, large tumour size patients previously exposed to CRT. In addition, retrospective
and deep stromal invasion (DSI) do not significantly increase the pooled analysis suggested that black race, pelvic location rather

by guest
on 05 February 2018
than non-pelvic, PS 1 or 2 and first relapse within 1 year of diag-
Table 4. Second-line therapy for metastatic cervical cancer
nosis may also be poor prognostic factors associated with lower
response [45]. Agent N CR1PR PFS OS
The three-drug combination of paclitaxel–ifosfamide–cisplatin (%) (months) (months)
(TIP) has shown promising responses (overall response rate 62%,
Bevacizumab [48] 46 11 3.4 7.3
with complete response 26%) and is regarded as an active regimen
Topotecan [62, 63] 94 13-19 2.1-2.4 6.4-6.6
with acceptable toxicity in advanced/relapsed cervical cancer [46].
Vinorelbine [64] 44 14 - -
A large randomised phase III trial (GOG-204) comparing four Gemcitabine [65] 22 5 2.1 6.5
different cisplatin-based doublets with paclitaxel, topotecan, gem- Albumin-bound paclitaxel [66] 35 29 5.0 9.4
citabine or vinorelbine was unable to demonstrate the superiority Docetaxel [67] 23 9 3.8 7.0
of any regimen. Nevertheless, paclitaxel–cisplatin showed the high- Pemetrexed [68, 69] 72 14-15 2.5-3.1 7.4-8.8
est response rate (29%), median PFS (5.8 months) and median OS Irinotecan [70] 42 21 4.5 6.4
(12.8 months) and was considered the preferred regimen based on Sunitinib [71] 19 0 3.5 -
the balance between efficacy and toxicity profile [II, B] [47]. Erlotinib [72] 28 0 1.9 5.0
Tumour angiogenesis plays a significant role in the progression Lapatinib [73] 78 5 4.2 9.7
of cervical cancer and has been associated with a poor prognosis. Pazopanib [73] 74 9 4.5 12.7
Bevacizumab prevents tumour angiogenesis by blocking vascular Pegylated liposomal 27 11 3.2 8.9
endothelial growth factor and was shown to be active in a phase II doxorubicin [74]
study (GOG-227C) in recurrent cervical cancer [48]. Based on
this observation, the GOG-240 study explored the addition of CR, complete response; OS, overall survival; PFS, progression-free sur-
bevacizumab to chemotherapy in a randomised phase III trial vival; PR, partial response.
with a 2 2 factorial design in which OS was the primary end-
point. Patients with primary stage IVB or recurrent/persistent,
good PS (0 or 1) and measurable disease were randomised to RT can play an important role in patients with recurrent dis-
paclitaxel–cisplatin or paclitaxel–topotecan, both with or with- ease, in the case of oligometastatic disease and for patients with
out bevacizumab. Two main conclusions were obtained from this only nodal metastasis in the pelvic, periaortic and/or supraclavic-
study: first, the median OS is significantly prolonged by the add- ular regions, as high-dose RT often leads to long-term disease
ition of bevacizumab (16.8 versus 13.3 months; HR 0.765; 95% control and a prolonged progression-free interval. Short-course
CI: 0.62–0.95; P ¼ 0.0068) and second, non-platinum doublet is palliative RT is used to treat symptoms from distant metastases.
not superior to cisplatin–paclitaxel, even in the population previ-
ously treated with cisplatin. Patients treated with bevacizumab
had a higher risk of grade 2 hypertension (25% versus 1.8%), Local recurrence of cervical cancer following
grade 3 venous thromboembolic events (8.2% versus 1.8%) radical surgery
and grade 2 fistula (8.6% versus 1%), and these side-effects The therapeutic options for patients who relapse in the pelvis fol-
must be carefully monitored during treatment [49]. lowing primary surgery are either radical RT or pelvic exentera-
The combination of paclitaxel and carboplatin could be con- tion. The reported survival rates range from 6% to 77%; patients
sidered an alternative for patients that are not candidates for cis- with central recurrences have better prognoses than those with
platin. Although a Japanese randomised clinical trial which pelvic side wall recurrence. Patients with central recurrences had
compared the two regimens showed a similar efficacy, the com- a 10-year survival rate of 77%, for those with no palpable tumour,
bination with cisplatin was superior to carboplatin in patients and a 10-year survival rate of 48% if the recurrence was < 3 cm;
without previous exposure to cisplatin [50]. The combination of there were no long-term survivors among patients with bulky
carboplatin/paclitaxel/bevacizumab is being studied in a multi- (> 3 cm) central recurrence in one series. The major prognostic
centre, single-arm, interventional trial (CECILIA), to evaluate factors associated with survival following salvage radiation in
the safety and efficacy of the combination in recurrent and/or patients with recurrent pelvic disease include disease-free inter-
metastatic cervical cancer (NCT02467907). val, site of recurrence (i.e. central versus pelvic side wall), and
Paclitaxel and cisplatin combined with bevacizumab is con- size. Higher doses of RT can be delivered with brachytherapy and
sidered the preferred first-line regimen in metastatic or recurrent increase the likelihood of local control for patients with small vol-
cervical cancer based on the balance between efficacy and toxicity ume central recurrences. Patients with large volume central or
profile [I, A]. pelvic side wall recurrences have poor prognoses, and efforts
In patients progressing following first-line therapy, different should be made to detect pelvic recurrences early to increase the
cytostatic agents, including vinorelbine, topotecan, gemcitabine chance of long-term survival [51].
or nanoparticle albumin-bound paclitaxel have been evaluated
(Table 4). However, response rates are low and duration of re-
sponses is short. Therefore, no recommendation can be given Fertility sparing
about the most effective second-line treatment (Table 4). More than 40% of women with early cervical cancer are affected
Some patients develop small lung metastases only, which do during reproductive age and wish to remain fertile. Thus, many
not rapidly progress and can be managed with stereotactic RT patients demand a more conservative policy for managing these
and/or a watchful waiting policy, frequently delaying systemic lesions in order to increase the chance of having an uneventful
chemotherapy for a significant period of time. pregnancy in the future.

by guest
on 05 February 2018
Incidence and epidemiology
Primary prevention of cervical cancer is now possible via immunisation with highly efficacious HPV vaccines [II, A] and secondary prevention has gained
impetus with the advent of sensitive HPV DNA testing to improve traditional Pap cytology screening programs [II, A].
Tumour risk assessment includes tumour size, stage, depth of tumour invasion, lymph node status, LVSI and histological subtype. Lymph node status and
number of lymph nodes involved are the most important prognostic factors.
Surgery
Surgical therapy in cervical cancer is adapted to the stage of disease according to FIGO and TNM classification (see Table 2).
Microinvasive cervical cancer (stage IA1) without LVSI can be managed with conisation or simple trachelectomy to preserve fertility [I, B]. Simple hysterec-
tomy can be offered if the patient does not wish to preserve fertility.
In stage IA1 with LVSI, surgical assessment of pelvic lymph nodes should be discussed with the patient, including the SLN.
In patients with FIGO stage IA2, IB and IIA, radical hysterectomy with bilateral lymph node dissection (with or without SLN) is standard treatment, if the
patient does not wish to preserve fertility [I, B].
Increasing evidence suggests an important role for SLND in cervical cancer. Sentinel nodes should be detected on both sides [II, B].
Surgery should only be considered in patients with earlier stages of cervical cancer (up to FIGO IIA) without risk factors necessitating adjuvant therapy,
which results in a multimodal therapy without improvement of survival but increased toxicity [I, A].
Study results indicate that NACT may offer a benefit over surgery alone in cervical cancer patients, reducing the need for adjuvant RT [I, C].
Chemoradiotherapy in locally advanced cervical cancer
CRT has been the standard of care for patients with bulky IB2–IVA disease for almost two decades, demonstrating an improvement in both DFS and OS
with concomitant chemotherapy and RT over standard RT/hydroxyurea [I, A].
The most commonly used regimen is weekly cisplatin 40 mg/m2, although the meta-analysis also reported significant benefits with non-platinum agents
[I, A].
Adjuvant treatment
Women with intermediate- and high-risk factors on the pathology specimen should receive adjuvant therapy following hysterectomy (see Table 3).
Cervical cancer patients with intermediate-risk disease do not need further adjuvant therapy [II, B], whereas adjuvant CRT is recommended in high-risk
patients [I, A].
Palliative chemotherapy with the aim of relieving symptoms and improving quality of life is indicated if the patient has a PS < 2 and no formal
contraindications.
Cisplatin-based doublets with topotecan or paclitaxel have demonstrated superiority to cisplatin monotherapy in terms of response rate and PFS.
Paclitaxel and cisplatin combined with bevacizumab is considered the preferred first-line regimen in metastatic or recurrent cervical cancer based on the
balance between efficacy and toxicity profile [I, A].
The combination of paclitaxel and carboplatin could be considered an alternative for patients that are not candidates for cisplatin.
For FIGO Stage IA1 patients, conisation is recommended as a first diagnostic and curative step for microscopic tumours in the presence of negative mar-
gins and the absence of clinical contraindications to surgery. PLND is recommended for patients with LVSI, who have an increased risk of lymph node in-
volvement. Sentinel node biopsy or trachelectomy [II, B] should be considered in some patients.
For FIGO Stage IA2 patients wishing to preserve fertility, cone biopsy or radical trachelectomy with PLND is the standard procedure.
Scientific evidence shows that trachelectomy with pelvic lymphadenectomy is the most appropriate surgical treatment of fertility sparing in patients
with tumours measuring 2 cm in diameter (FIGO Stage IB1 < 2 cm) [II, B]. For tumours > 2 cm, NACT followed by conisation or trachelectomy may
also be a valid choice.
Follow-up visits with a complete physical examination including a pelvic–rectal exam and a patient history should be conducted by a physician experi-
enced in the surveillance of cancer patients.
CT or PET/CT scan should be carried out as clinically indicated. A reasonable follow-up schedule involves follow-up visits every 3–6 months in the first
2 years and every 6–12 months in years 3–5.
Patients should return to annual population-based general physical and pelvic examinations after 5 years of recurrence-free follow-up [III, C].
CRT, chemoradiotherapy; CT, computed tomography; DFS, disease-free survival; DNA, deoxyribonucleic acid; FIGO, Féderation Internationale de
Gynécologie et d’Obstétrique; HPV, human papillomavirus; LVSI, lymphovascular space invasion; NACT, neoadjuvant chemotherapy; OS, overall sur-
vival; Pap, Papanicolaou; PET/CT, positron emission tomography/computed tomography; PFS, progression-free survival; PLND, pelvic lymph node
dissection; PS, performance status; RT, radiotherapy; SLN, sentinel lymph node; SLND, sentinel lymph node dissection; TNM, tumour, node and
metastasis.
FIGO stage IA1 the presence of negative margins and the absence of clinical
contraindications to surgery, the cone biopsy may represent de-
According to most international guidelines, the first diagnostic
finitive treatment. For patients with LVSI, who have an increased
and curative step for microscopic tumours is conisation [52]. In

by guest
on 05 February 2018
Table 6. Levels of evidence and grades of recommendation (adapted from that less radical treatment may be a valid choice for fertility sparing
the Infectious Diseases Society of America-United States Public Health also in these lesions (lower morbidity, higher pregnancy rate).
Service Grading Systema) Some studies have reported in the absence of residual disease in
trachelectomy specimens in the range of 60%–65%, questioning
Levels of evidence
the need for radical surgery in patients with low-risk tumours [56].
Finally, some authors have suggested conisation with/without
NACT in this tumour setting as well [II, C] [55, 57–60].
heterogeneity
II Small randomised trials or large randomised trials with a suspicion FIGO stage IB > 2 cm
For tumours > 2 cm, NACT followed by conisation or trachelec-
tomy may also be a valid choice, but downstaging by NACT in
IB1 and IB2 cervical cancer before fertility-sparing surgery is still
an experimental procedure [I, C].
A Strong evidence for efficacy with a substantial clinical benefit, Personalised medicine
B Strong or moderate evidence for efficacy but with a limited clin- In this disease setting, more research is needed to identify molecular
ical benefit, generally recommended markers which could lead to advances in personalised medicine.
C Insufficient evidence for efficacy or benefit does not outweigh
the risk or the disadvantages (adverse events, costs, . . . ).,
optional Follow-up, long-term implications and
D Moderate evidence against efficacy or for adverse outcome, gen-
erally not recommended
survivorship
E Strong evidence against efficacy or for adverse outcome, never No definitive agreement exists on the best post-treatment surveil-
recommended lance of cervical cancer. At a minimum, follow-up visits with a
a
complete physical examination, including a pelvic–rectal exam
and a patient history, should be conducted by a physician experi-
enced in the surveillance of cancer patients. There is little evidence
to suggest that vaginal vault cytology adds significantly to the clin-
risk of lymph node involvement, pelvic lymph node dissection ical exam in detecting early disease recurrence. Routine use of vari-
(PLND) is recommended [52]. Sentinel node biopsy should be ous other radiological or biological follow-up investigations in
considered. Moreover, for these patients, some authors suggest asymptomatic patients is not advocated, because the role of those
trachelectomy, a surgical procedure in which the uterine cervix investigations has yet to be evaluated in a definitive manner. CT or
and adjacent tissues are removed [II, B]. PET/CT scan should be carried out as clinically indicated. A reason-
able follow-up schedule involves follow-up visits every 3–6 months
in the first 2 years and every 6–12 months in years 3–5. Patients
FIGO stage IA2 should return to annual population-based general physical and pel-
vic examinations after 5 years of recurrence-free follow-up [III, C].
For patients wishing to preserve fertility, cone biopsy or radical
trachelectomy with PLND is the standard procedure [53].
Sentinel node biopsy is under validation but may be considered
[II, B]. Methodology
These Clinical Practice Guidelines were developed in accordance
with the ESMO standard operating procedures for Clinical Practice
FIGO stage IB1 < 2 cm Guidelines development http://www.esmo.org/Guidelines/ESMO-
Scientific evidence shows that trachelectomy with pelvic lympha- Guidelines-Methodology. The relevant literature has been selected
denectomy is the most appropriate surgical treatment of fertility by the expert authors. A summary of recommendations is shown in
sparing in patients with these tumours. Tumours > 2 cm are Table 5. Levels of evidence and grades of recommendation have
clearly associated with a higher risk of recurrence (3% for le- been applied using the system shown in Table 6. Statements without
sions 2 cm versus 17% for lesions > 2 cm); thus, international grading were considered justified standard clinical practice by the
guidelines stress that this procedure is valid mostly for tumours experts and the ESMO Faculty. This manuscript has been subjected
measuring 2 cm in diameter [II, B] [54, 55]. to an anonymous peer review process.
Currently, as documented by most international guidelines, rad-
ical trachelectomy is considered a standard fertility-sparing proced-
ure in patients with early cervical cancer and tumours < 2 cm. Disclosure
However, the low incidence of parametrial involvement reported in CM and MMC have reported honoraria and participation at
patients with tumours < 2 cm and no nodal disease or LVSI suggest advisory boards for Roche; SM has reported consulting for

by guest
on 05 February 2018
Roche, Clovis, Sensor Kinesis, MEDAC, AstraZeneca and has 20. Diab Y. Sentinel lymph nodes mapping in cervical cancer a comprehen-
received grants from Roche, Pharmamar, Tesaro, MEDAC, sive review. Int J Gynecol Cancer 2017; 27: 154–158.
21. Canadian Cancer Trials Group. Radical versus simple hysterectomy and
AstraZeneca and honoraria from Roche, Pharmamar, Clovis,
pelvic node dissection in patients with low-risk early stage cervical cancer
Tesaro, Sensor Kinesis, MEDAC and AstraZeneca; AGM has (SHAPE). https://clinicaltrials.gov/ct2/show/NCT01658930 (29 April
received honoraria from Roche, AstraZeneca and PharmaMar; 2017, date last accessed).
NC is a member of Roche speakers’ bureau; FL has reported no 22. Benedetti-Panici P, Greggi S, Scambia G et al. Long-term survival follow-
conflict of interest. ing neoadjuvant chemotherapy and radical surgery in locally advanced
cervical cancer. Eur J Cancer 1998; 34: 341–346.
23. Sardi JE, di Paola GR, Cachau A et al. A possible new trend in the man-
agement of the carcinoma of the cervix uteri. Gynecol Oncol 1986; 25:
References 139–149.
1. Torre LA, Bray F, Siegel RL et al. Global cancer statistics, 2012. CA 24. Paladini D, Raspagliesi F, Fontanelli R, Ntousias V. Radical surgery after
Cancer J Clin 2015; 65: 87–108. induction chemotherapy in locally advanced cervical cancer. A Feasibility
2. International Agency for Research on Cancer, EUCAN. http://eco.iarc.fr/ Study. Int J Gynecol Cancer 1995; 5: 296–300.
eucan (29 April 2017, date last accessed). 25. Neoadjuvant Chemotherapy for Locally Advanced Cervical Cancer
3. Sant M, Chirlaque Lopez MD, Agresti R et al. Survival of women with Meta-analysis Collaboration. Neoadjuvant chemotherapy for locally
cancers of breast and genital organs in Europe 1999–2007: results of the advanced cervical cancer: a systematic review and meta-analysis of indi-
EUROCARE-5 study. Eur J Cancer 2015; 51: 2191–2205. vidual patient data from 21 randomised trials. Eur J Cancer 2003; 39:
4. Petrosky E, Bocchini JA Jr, Hairi S et al. Use of 9-valent human papillo- 2470–2486.
mavirus (HPV) vaccine: updated HPV vaccination recommendations of 26. European Organisation for Research and Treatment of Cancer – EORTC.
the Advisory Committee on Immunization Practices. MMWR Morb Chemotherapy followed by surgery vs radiotherapy plus chemotherapy in
Mortal Wkly Rep 2015; 64: 300–304. patients with stage Ib or II cervical cancer. EORTC Protocol 55994. https://
5. FUTURE II Study Group. Quadrivalent vaccine against human papillo- clinicaltrials.gov/ct2/show/NCT00039338 (29 April 2017, date last accessed).
mavirus to prevent high grade cervical lesion. N Engl J Med 2007; 356: 27. Gupta S. Neoadjuvant chemotherapy followed by surgery versus concur-
1915–1927. rent chemoradiation in carcinoma of the cervix (NACTcervix).
6. Paavonen J, Naud P, Salmer on J et al. Efficacy of human papillomavirus NCT00193739. https://clinicaltrials.gov/ct2/show/NCT00193739
(HPV) 16/18 AS04-adjuvanted vaccine against cervical infection and precancer NCT00193739 https://clinicaltrials.gov/ct2/show/NCT00193739 (29
caused by oncogenic HPV types (PATRICIA): final analysis of a double-blind, April 2017, date last accessed).
randomised study in young women. Lancet 2009; 374: 301–314. 28. Rydzewska L, Tierney J, Vale CL, Symonds PR. Neoadjuvant chemother-
7. Joura EA, Giuliano AR, Iversen OE et al. A 9-valent HPV vaccine against apy plus surgery versus surgery for cervical cancer. Cochrane Database
infection and intraepithelial neoplasia in women. N Engl J Med 2015; Syst Rev 2012; 12: CD007406.
372: 711–723. 29. Whitney CW, Sause W, Bundy BN et al. Randomized comparison of flu-
8. Markowitz LE, Liu G, Hariri S et al. Prevalence of HPV after introduc- orouracil plus cisplatin versus hydroxyurea as an adjunct to radiation
tion of the vaccination program in the United States. Pediatrics 2016; therapy in stage IIB–IVA carcinoma of the cervix with negative para-
137: e20151968. aortic lymph nodes: a Gynecologic Oncology Group and Southwest
9. Ali H, Donovan B, Wand H et al. Genital warts in young Australians five Oncology Group study. J Clin Oncol 1999; 17: 1339–1348.
years into national human papillomavirus vaccination programme: na- 30. Rose PG, Bundy BN, Watkins EB et al. Concurrent cisplatin-based radio-
tional surveillance data. BMJ 2013; 346: f2032. therapy and chemotherapy for locally advanced cervical cancer. N Engl J
10. Ronco G, Dillner J, Elfström KM et al. Efficacy of HPV-based screening Med 1999; 340: 1144–1153.
for prevention of invasive cervical cancer: follow-up of four European 31. Morris M, Eifel PJ, Lu J et al. Pelvic radiation with concurrent chemo-
randomised controlled trials. Lancet 2014; 383: 524–532. therapy compared with pelvic and para-aortic radiation for high-risk cer-
11. Serrano B, de Sanjosé S, Tous S et al. Human papillomavirus genotype vical cancer. N Engl J Med 1999; 340: 1137–1143.
attribution for HPVs 6, 11, 16, 18, 31, 33, 45, 52 and 58 in female ano- 32. Keys HM, Bundy BN, Stehman FB et al. Cisplatin, radiation, and adjuvant
genital lesions. Eur J Cancer 2015; 51: 1732–1741. hysterectomy compared with radiation and adjuvant hysterectomy for
12. Alemany L, Cubilla A, Halec G et al. Role of human papillomavirus in bulky stage IB cervical carcinoma. N Engl J Med 1999; 340: 1154–1161.
penile carcinomas worldwide. Eur Urol 2016; 69: 953–961. 33. Peters WA, 3rd, Liu PY, Barrett RJ, 2nd et al. Concurrent chemotherapy
13. Alemany L, Saunier M, Alvarado-Cabrero I et al. Human papillomavirus and pelvic radiation therapy compared with pelvic radiation therapy
DNA prevalence and type distribution in anal carcinomas worldwide. Int alone as adjuvant therapy after radical surgery in high-risk early-stage
J Cancer 2015; 136: 98–107. cancer of the cervix. J Clin Oncol 2000; 18: 1606–1613.
14. Wagenaar HC, Trimos JB, Postema S et al. Tumor diameter and volume 34. Chemoradiotherapy for Cervical Cancer Meta-Analysis
assessed by magnetic resonance imaging in the prediction of outcome for Collaboration. Reducing uncertainties about the effects of chemora-
invasive cervical cancer. Gynecol Oncol 2001; 82: 474–482. diotherapy for cervical cancer: a systematic review and meta-analysis
15. Patel CN, Nazir SA, Khan Z et al. 18F-FDG PET/CT of cervical carcin- of individual patient data from 18 randomized trials. J Clin Oncol
oma. AJR Am J Roentgenol 2011; 196: 1225–1233. 2008; 26: 5802–5812.
16. Brockbank E, Kokka F, Bryant A et al. Pre-treatment surgical para-aortic 35. Due~ nas-Gonzalez A, Zarba JJ, Patel F et al. Phase III, open-label,
lymph node assessment in locally advanced cervical cancer. Cochrane randomized study comparing concurrent gemcitabine plus cisplatin and
Database Syst Rev 2011; 4: CD008217. radiation followed by adjuvant gemcitabine and cisplatin versus concur-
17. Kim SM, Choi HS, Byun JS. Overall 5-year survival rate and prognostic rent cisplatin and radiation in patients with stage IIB to IVA carcinoma
factors in patients with stage IB and IIA cervical cancer treated by radical of the cervix. J Clin Oncol 2011; 29: 1678–1685.
hysterectomy and pelvic lymph node dissection. Int J Gynecol Cancer 36. Tanderup K, Lindegaard JC, Kirisits C et al. Image guided adaptive
2000; 10: 305–312. brachytherapy in cervix cancer: a new paradigm changing clinical prac-
18. Dittrich R, Lotz L, Hackl J et al. Fertilit€atserhalt bei Krebserkrankungen. tice and outcome. Radiother Oncol 2016; 120: 365–369.
Frauenarzt 2014; 55: 240–246. 37. Sturdza A, Pötter R, Fokdal LU et al. Image guided brachytherapy in lo-
19. Geetha P, Nair MK. Laparoscopic, robotic and open method of radical cally advanced cervical cancer: improved pelvic control and survival in
hysterectomy for cervical cancer: a systematic review. J Min Access Surg RetroEMBRACE, a multicenter cohort study. Radiother Oncol 2016;
2012; 8: 67–73. 120: 428–433.

by guest
on 05 February 2018
38. Neoadjuvant Chemotherapy for Cervical Cancer Meta-analysis Collaboration 57. Maneo A, Sideri M, Scambia G et al. Simple conization and lymphade-
(NACCCMA). Neoadjuvant chemotherapy for locally advanced cervix can- nectomy for the conservative treatment of stage IB1 cervical cancer. An
cer. Cochrane Database Syst Rev 2004; 2: CD001774. Italian experience. Gynecol Oncol 2011; 123: 557–560.
39. Berman ML, Keys H, Creasman W et al. Survival and patterns of 58. Fagotti A, Gagliardi ML, Moruzzi C et al. Excisional cone as fertility-
recurrence in cervical cancer metastatic to periaortic lymph nodes sparing treatment in early-stage cervical cancer. Fertil Steril 2011; 95:
(a Gynecologic Oncology Group study). Gynecol Oncol 1984; 19: 8–16. 1109–1112.
40. Indonesia University. 10 vs 14 Days Triple Therapy: H. pylori Infection 59. Fanfani F, Landoni F, Gagliardi ML et al. Sexual and reproductive outcomes
Eradication. NCT01566240. https://clinicaltrials.gov/ct2/show/ (29 April in early stage cervical cancer patients after excisional cone as a fertility-
2017, date last accessed). sparing surgery: an Italian experience. J Reprod Infertil 2014; 15: 29–34.
41. Gouy S, Morice P, Narducci F et al. Prospective multicenter study evalu- 60. Choi MC, Jung SG, Park H et al. Photodynamic therapy for management
ating the survival of patients with locally advanced cervical cancer of cervical intraepithelial neoplasia II and III in young patients and ob-
undergoing laparoscopic para-aortic lymphadenectomy before chemora- stetric outcomes. Lasers Surg Med 2013; 45: 564–572.
diotherapy in the era of positron emission tomography imaging. J Clin 61. James D, Brierley JD, Gospodarowicz MK et al. (eds). TNM Classification of
Oncol 2013; 31: 3026–3033. Malignant Tumours, 8th edition. Oxford, UK: John Wiley & Sons, Inc. 2016.
42. Rotman M, Sedlis A, Piedmonte MR et al. A phase III randomized trial 62. Bookman MA, Blessing JA, Hanjani P et al. Topotecan in squamous cell
of postoperative pelvic irradiation in Stage IB cervical carcinoma with carcinoma of the cervix: a phase II study of the Gynecologic Oncology
poor prognostic features: follow-up of a Gynecologic Oncology Group Group. Gynecol Oncol 2000; 77: 446–449.
study. Int J Radiat Oncol Biol Phys 2006; 65: 169–176. 63. Muderspach LI, Blessing JA, Levenback C, Moore JL Jr. A phase II study
43. Moore DH, Blessing JA, McQuellon RP et al. Phase III study of cisplatin of topotecan in patients with squamous cell carcinoma of the cervix: a
with or without paclitaxel in stage IVB, recurrent, or persistent squamous Gynecologic Oncology Group study. Gynecol Oncol 2001; 81: 213–215.
cell carcinoma of the cervix: a Gynecologic Oncology Group Study. 64. Muggia FM. Relevance of chemotherapy dose and schedule to outcomes
J Clin Oncol 2004; 22: 3113–3119. in ovarian cancer. Semin Oncol 2004; 31(6 Suppl 15): 19–24.
44. Long HJ 3rd, Bundy BN, Grendys EC Jr et al. Randomized phase III 65. Schilder RJ, Blessing J, Cohn DE. Evaluation of gemcitabine in previously
trial of cisplatin with or without topotecan in carcinoma of the uterine treated patients with non-squamous cell carcinoma of the cervix: a phase
cervix: a Gynecologic Oncology Group Study. J Clin Oncol 2005; 23: II study of the Gynecologic Oncology Group. Gynecol Oncol 2005; 96:
4626–4633. 103–107.
45. Moore DH, Tian C, Monk BJ et al. Prognostic factors for response to 66. Alberts DS, Blessing JA, Landrum LM et al. Phase II trial of nab-
cisplatin-based chemotherapy in advanced cervical carcinoma: a paclitaxel in the treatment of recurrent or persistent advanced cervix
Gynecologic Oncology Group Study. Gynecol Oncol 2010; 116: 44–49. cancer: a Gynecologic Oncology Group study. Gynecol Oncol 2012;
46. Kosmas C, Mylonakis N, Tsakonas G et al. Evaluation of the paclitaxel– 127: 451–455.
ifosfamide–cisplatin (TIP) combination in relapsed and/or metastatic 67. Garcia AA, Blessing JA, Vaccarello L et al. Phase II clinical trial of doce-
cervical cancer. Br J Cancer 2009; 101: 1059–1065. taxel in refractory squamous cell carcinoma of the cervix: a Gynecologic
47. Monk BJ, Sill MW, McMeekin DS et al. Phase III trial of four cisplatin- Oncology Group Study. Am J Clin Oncol 2007; 30: 428–431.
containing doublet combinations in stage IVB, recurrent, or persistent 68. Miller DS, Blessing JA, Krasner CN et al. Phase II evaluation of peme-
cervical carcinoma: a Gynecologic Oncology Group study. J Clin Oncol trexed in the treatment of recurrent or persistent platinum-resistant
2009; 27: 4649–4655. ovarian or primary peritoneal carcinoma: a study of the Gynecologic
48. Monk BJ, Sill MW, Burger RA et al. Phase II trial of bevacizumab in the Oncology Group. J Clin Oncol 2009; 27: 2686–2691.
treatment of persistent or recurrent squamous cell carcinoma of the cervix: 69. Lorusso D, Ferrandina G, Pignata S et al. Evaluation of pemetrexed
a Gynecologic Oncology Group study. J Clin Oncol 2009; 27: 1069–1074. (Alimta, LY231514) as second-line chemotherapy in persistent or recurrent
49. Tewari KS, Sill MW, Long HJ 3rd et al. Improved survival with bevacizu- carcinoma of the cervix: the CERVIX 1 study of the MITO (Multicentre
mab in advanced cervical cancer. N Engl J Med 2014; 370: 734–743. Italian Trials in Ovarian Cancer and Gynecologic Malignancies) Group.
50. Kitagawa R, Katsumata N, Shibata T et al. Paclitaxel plus carboplatin versus Ann Oncol 2010; 21: 61–66.
paclitaxel plus cisplatin in metastatic or recurrent cervical cancer: the open- 70. Verschraegen CF, Levy T, Kudelka AP et al. Phase II study of irinotecan
label randomized phase III trial JCOG0505. J Clin Oncol 2015; 33: 2129–2135. in prior chemotherapy-treated squamous cell carcinoma of the cervix. J
51. Friedlander M, Grogan M; U.S. Preventative Services Task Force. Clin Oncol 1997; 15: 625–631.
Guidelines for the treatment of recurrent and metastatic cervical cancer. 71. Mackay HJ, Tinker A, Winquist E et al. A phase II study of sunitinib in
Oncologist 2002; 7: 342–347. patients with locally advanced or metastatic cervical carcinoma: NCIC
52. Yoneda JY, Braganca JF, Sarian LO et al. Surgical treatment of microin- CTG Trial IND.184. Gynecol Oncol 2010; 116: 163–167.
vasive cervical cancer: analysis of pathologic features with implications 72. Schilder RJ, Sill MW, Lee YC, Mannel R. A phase II trial of erlotinib in
on radicality. Int J Gynecol Cancer 2015; 25: 694–698. recurrent squamous cell carcinoma of the cervix: a Gynecologic
53. NCCN Guidelines for treatment of cervical cancer. https://www.nccn. Oncology Group Study. Int J Gynecol Cancer 2009; 19: 929–933.
org/professionals/physician_gls/pdf/cervical.pdf (15 May 2017, date last 73. Monk BJ, Mas Lopez L, Zarba JJ et al. Phase II, open-label study of pazo-
accessed). panib or lapatinib monotherapy compared with pazopanib plus lapatinib
54. Lanowska M, Mangler M, Spek A et al. Radical vaginal trachelectomy combination therapy in patients with advanced and recurrent cervical
(RVT) combined with laparoscopic lymphadenectomy: prospective cancer. J Clin Oncol 2010; 28: 3562–3569.
study of 225 patients with early-stage cervical cancer. Int J Gynecol 74. Rose PG, Blessing JA, Lele S et al. Evaluation of pegylated liposomal dox-
Cancer 2011; 21: 1458–1464. orubicin (Doxil) as second-line chemotherapy of squamous cell carci-
55. Landoni F, Parma G, Peiretti M et al. Chemo-conization in early cervical noma of the cervix: a phase II study of the Gynecologic Oncology Group.
cancer. Gynecol Oncol 2007; 107(Suppl 1): S125–S126. Gynecol Oncol 2006; 102: 210–213.
56. Plante M, Gregoire J, Renaud MC et al. Simple vaginal trachelectomy in 75. Dykewicz CA. Summary of the guidelines for preventing opportunistic
early-stage low-risk cervical cancer: a pilot study of 16 cases and review infections among hematopoietic stem cell transplant recipients. Clin
of the literature. Int J Gynecol Cancer 2013; 23: 916–922. Infect Dis 2001; 33: 139–144.

by guest
on 05 February 2018
Published online 1 September 2013
Gestational trophoblastic disease: ESMO Clinical Practice

M. J. Seckl1, N. J. Sebire2, R. A. Fisher1, F. Golfier3, L. Massuger4 & C. Sessa5, on behalf of the
1
Department of Cancer Medicine; 2Department of Histopathology, Charing Cross Gestational Trophoblastic Disease Centre, Charing Cross Hospital Campus of Imperial
College London, London, UK; 3Centre de Référence des Maladie Trophoblastiques, Hospices Civils de Lyon, Lyon, France; 4Department of Obstetrics and Gynaecology,
Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands; 5Oncology Institute of Southern Switzerland, Ospedale San Giovanni, Bellinzona, Switzerland;
incidence and epidemiology

Gestational trophoblastic disease (GTD) comprises a spectrum Box 1. hCG measurement
of disorders from the pre-malignant conditions of complete
HCG comprises an alpha subunit common to all
(CHM) and partial (PHM) hydatidiform moles through to the
glycoprotein hormones including lutenising hormone (LH)
malignant invasive mole, choriocarcinoma (CC) and very rare
and thyroid-stimulating hormone (TSH) and a specific beta
placental site trophoblastic tumour/epithelioid trophoblastic
subunit. Consequently, assays to detect hCG use antibodies
tumour (PSTT/ETT). The malignant forms of the disease are
directed against the beta subunit. In pregnancy, this subunit
also collectively known as gestational trophoblastic tumours or
clinical practice
is usually intact and becomes hyperglycosylated particularly
neoplasia (GTN). In the UK, all GTD cases are nationally
guidelines
during the first trimester. In contrast, cancer-related beta
registered, with central pathology review. The incidence is
hCG can exist in several different forms/fragments including
estimated at 1-3: 1000 pregnancies for CHM and 3: 1000
nicked free beta, c-terminal peptide, hyperglycosylated and
pregnancies for PHM, respectively, with other western countries
so it is essential that the hCG assay used to detect hCG in
reporting similar data [1]. GTD appears to be more frequent in
cancer patients can measure all forms of beta hCG equally
Asia than in North America or Europe. This may be because of
well. There are currently many commercial hCG assays
discrepancies between hospital- and population-based data,
available that are very good for assessing hCG in pregnancy,
availability of central pathological review or may reflect dietary
but their ability to work well in cancer is less clear. Several
and genetic influences. An increased risk of molar pregnancy is
reports indicate that some assays either fail to detect all the
seen in the very young (<16 years), but is most associated with
hCG isoforms/fragments or significantly under or over-read
advanced maternal age (>45 years) [1]. Following a molar
certain isoforms. This can lead to false-negative results and
pregnancy, the risk of a further CHM or PHM increases to ∼1%.
there are also several assays that appear to have particular
After two molar gestations, the risk of a third mole is 15%–20%
problems with false-positive results. Clinicians need to be
and is not decreased by changing partners.
aware of these potential problems and when hCG results do
The frequency of CC and PSTT is less clear, since these can
not fit the clinical picture, they should measure the hCG on
arise after any type of pregnancy. CC develops after around
a different assay. When a false positive is suspected,
1:50 000 deliveries, while recent data suggest that PSTT
assessment of the urine hCG can also be helpful as cross-
represents 0.2% of UK GTD cases [2]. GTN risk may also relate
reactive molecules in the blood that cause false positives
to hormonal factors since women with menarche after 12 years
rarely get into the urine. Consequently, a positive urine hCG
of age, light menstrual flow and prior use of oral contraceptives
excludes a false-positive serum result. Further details on
are at increased risk. Additionally, the subsequent risk of
hCG assays and monitoring in GTN are available in ref. [1].
malignancy following a hydatidiform mole (HM) has been
linked in some but not all series to oral contraceptives, if started
while the human chorionic gonadotrophin (hCG) is still
elevated [1]. This hormone is essential for the diagnosis, diagnosis, genetics/molecular biology
management and subsequent surveillance of GTD and details and pathology
regarding hCG and its measurement are provided in Box 1.
diagnosis
*Correspondence to: ESMO Guidelines Working Group, ESMO Head Office, Via CHMs and PHMs most commonly present with vaginal
L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland. bleeding in the first trimester of pregnancy. Previously reported
features such as anaemia, uterine enlargement, pre-eclampsia,
†
Approved by the ESMO Guidelines Working Group: July 2013. hyperemesis, hyperthyroidism and respiratory distress are now
by guest
on 05 February 2018
rare [3], reflecting the introduction of routine ultrasonography with recurrent AnCHM are likely to have normal live births in
in early pregnancy. Characteristic sonographic findings for subsequent pregnancies and benefit from conventional in vitro
CHM in the second trimester, of a heterogeneous mass fertilisation, women with FRHM are unlikely to achieve a
(‘snowstorm’), without foetal development and with theca- normal pregnancy except through ovum donation from an
lutein ovarian cysts, are not seen in the first trimester, and unaffected individual [7].
ultrasonography is not diagnostically reliable [4]. Indeed, false PHMs are almost always triploid, usually as a result of
positive and negative rates are high with ultrasound, especially fertilisation of an apparently normal ovum by two sperm or
for PHM, and histological examination is essential to achieve a occasionally a diploid sperm (Figure 1D). The existence of
correct diagnosis [4]. All products of conception from non- diploid PHM is unlikely, most reported cases representing
viable pregnancies must undergo histological examination misdiagnosed complete moles, hydropic abortions or twin
regardless of ultrasound findings [5]. pregnancies.
The safest method of evacuation is suction dilation and While most molar pregnancies are diploid CHM or triploid
curettage (D&C) under ultrasound control to ensure adequate PHM, numerical and structural abnormalities have been
emptying of uterine contents and to avoid uterine perforation reported in both CHM and PHM. In addition, CHM, and
[1]. A proportion of women who miscarry or who undergo occasionally PHM, can be associated with a twin pregnancy
medical terminations will have unsuspected molar pregnancies. with a coexistent normal twin [8]. The continuance of such twin
As histological examination is not routinely requested, the pregnancies results in healthy babies in ∼40% of cases, without
diagnosis of GTN can be delayed resulting in significantly an obvious increase in the risk of malignant change [8].
greater morbidity [6]. Histological examination of every Since post-molar GTN is treated on a clinical, rather than
termination is impractical, and perhaps a simple measurement pathological, diagnosis tumour tissue is rarely available for
of the urine or serum hCG level 3–4 weeks post-treatment to genetic analysis. However, where tissue is available from GTN,
ensure return to normal is indicated [6]. All women with a the genotype will reflect that of the causative pregnancy, having
diagnosis of molar pregnancy require careful hCG monitoring both maternal and paternal chromosomes if the tumour
to look for the recurrence of disease, suggesting malignant originated in a term pregnancy, hydropic abortion or PHM but
change indicated by a plateaued or rising hCG on three and two only paternal genes if the causative pregnancy was a CHM.
consecutive samples, respectively (see Box 1 for details about Since the interval from the causative pregnancy to the time of
hCG testing) [1]. Re-biopsy to confirm malignant change is not GTN diagnosis carries prognostic information, genotyping can
advised because of the risk of triggering life-threatening be helpful particularly in patients with multiple pregnancies [1].
haemorrhage. Genetics can also be important in the differential diagnosis
The other malignant forms of GTD, CC and PSTT/ETT can between gestational and non-gestational tumours, such as lung
be much more tricky to diagnose as the disease can develop and gastric cancers, that can occasionally present as CC, but will
months or many years after a prior pregnancy with protean have a genotype reflecting that of the patient [9]. These non-
presentations possible. Although change in menstruation is gestational CC often initially respond to GTN-based therapies,
frequent, it does not always occur. It is therefore essential to but their outcome is invariably poor, reflecting the originating
measure the hCG in any woman of childbearing age who has tissue [1].
unexplained metastatic disease. Biopsy of lesions without the
ability to control bleeding is highly risky in this very vascular
disease and is not essential before commencing chemotherapy. pathology
However, where complete excision is possible this can provide All forms of GTD are derived from components of the normal
useful histological confirmation of the diagnosis and material human placenta; HM plus CC, and PSTT/ETT, representing
for genetic analysis (see below). abnormal counterparts of the villous and extravillous
(interstitial) trophoblast, respectively. Most CHM and PHM
have distinctive morphological characteristics, but it is
genetics/molecular biology recommended that cases of suspected GTD be reported by
CHMs are usually diploid and androgenetic in origin, ∼80% specialist histopathologists. CHMs show a characteristic villous
resulting from duplication of the haploid genome of a single sperm architecture, associated with abnormal trophoblast hyperplasia,
while 20% arise by dispermic fertilisation of an ovum (Figure 1A stromal hypercellularity, stromal karyorrhectic debris and
and B). In either case maternal chromosomes are lost before, or collapsed villous blood vessels (Figure 2A). In contrast, PHMs
shortly after, fertilisation. However, while nuclear DNA is show patchy villous hydropic change with scattered abnormally
entirely paternal in CHM, mitochondrial DNA remains shaped irregular villi with trophoblastic pseudoinclusions and
maternal in origin [1]. patchy trophoblast hyperplasia (Figure 2B) [10]. The
Recent evidence indicates that some patients with recurrent morphological distinction between non-molar miscarriage,
CHM have diploid biparental CHM (BiCHM) rather than the especially when associated with chromosomal abnormality, and
typical androgenetic CHM (AnCHM) (Figure 1C). In these PHM can sometimes be difficult, and ancillary techniques may
cases, the molar phenotype is due to an autosomal recessive be required including immunostaining with p57KIP2 (negative in
condition, familial recurrent HM (FRHM) that predisposes CHM), ploidy analysis by in situ hybridisation or flow
women to recurrent pregnancy loss, most usually CHM. cytometry or molecular genotyping. Genotyping can also be
Mutations in two genes have now been associated with this useful in the identification of BiCHM, associated with FRHM,
condition: NLRP7 and, more rarely, KHDC3L. While women since most are pathologically indistinguishable from typical
vi | Seckl et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
Figure 1. Genetic origins of molar pregnancies. (A) Monospermic CHMs arise as a result of pre- or post-fertilisation loss of the maternal nuclear genome and
duplication of the paternal genome. These androgenetic diploids are 46,XX, 46,YY conceptuses being presumed non-viable. (B) Dispermic CHMs arise as a
result of two sperm fertilising an ovum from which the maternal nuclear genome is lost. These androgenetic diploid conceptuses may be 46,XX or 46,XY. (C)
Biparental CHMs occur in females who are homozygous, or a compound heterozygote, for mutations in NLRP7 or KHDC3L. These biparental conceptuses are
phenotypically CHM and may be 46,XX or 46,XY. (D) Dispermic PHMs arise as a result of fertilisation of a single ovum by two sperms. These diandric triploid
conceptions may be 69,XXX, 69,XXY or 69,XYY.
AnCHM [11]. Unfortunately, there are no histological or syncytiotrophoblast-like areas. Intraplacental CC are rare but
immunohistochemical features that reliably predict which probably represent the source of metastatic CC, which occur
patients will subsequently develop persistent GTD ( pGTD)/ following apparently uncomplicated term pregnancies. PSTT
GTN, and hence all HMs require hCG surveillance. (Figure 2D) is the malignant equivalent of extravillous
CC (Figure 2C) are malignant hCG-producing epithelial interstitial implantation site-like trophoblast and forms uterine
tumours with differentiation towards a villous trophoblast lesions with less haemorrhage and necrosis, and lower hCG
phenotype, usually demonstrating central necrosis and levels, than CC. The histological features show locally
characteristic biphasic architecture recapitulating infiltrating nests and sheets of monomorphic, interstitial-type
cytotrophoblast-like cells and multinucleate, pleomorphic trophoblast, with moderate pleomorphism and mitotic activity,
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt345 | vi

by guest
on 05 February 2018
Figure 2. Photomicrographs demonstrating the various histopathological forms of GTD. (A) Complete hydatidiform mole, (B) partial hydatidiform mole, (C)
choriocarcinoma and (D) placental site trophoblastic tumour. All are characterised by abnormal forms of trophoblast proliferation, associated with dysmorphic
chorionic villi in CHM and PHM, but no villi and abnormal trophoblast invasion in CC and PSTT. (Original magnifications ×40, ×20, ×200 and ×100,
respectively.)
and expression of human placental lactogen (hPL) and other Table 1. UK indications for chemotherapy following the diagnosis of GTD
extravillous trophoblast markers. A specific variant of PSST with
distinctive hyalinisation and a slightly different Indications for chemotherapy
immunohistochemical profile has been reported, ETT which is Plateaued or rising hCG after evacuationa
clinically thought to behave like PSTT [12]. Heavy vaginal bleeding or evidence of gastrointestinal or intraperitoneal
haemorrhage
Histological evidence of choriocarcinoma
staging and risk assessment Evidence of metastases in the brain, liver or gastrointestinal tract, or
radiological opacities of >2 cm on chest X-ray
indications for treatment Serum hCG of ≥20 000 IU/l >4 weeks after evacuation, because of the risk
Following suction curretage of a PHM, patients should have of uterine perforation
anti-Rhesus D prophylaxis. After any HM, the onset of Raised hCG 6 months after evacuation even if still falling (now omitted
malignant change, referred to as pGTD or post-mole GTN, is [15])
nearly always indicated by a plateaued or rising hCG (Table 1). a
Plateaued or rising is defined as four or more equivalent values of hCG
In the UK, this occurs after 15% and 0.5%–1% of CHM and
over at least 3 weeks (days 1, 7, 14 and 21) and two consecutive rises in hCG
PHM, respectively [1]. In other countries, these rates may be
of 10% or greater over at least 2 weeks (days 1, 7 and 14), respectively.
higher, possibly reflecting differences in hCG assays, hCG
criteria for the diagnosis of GTN, lack of whole population
demographics or, less likely, a genuine difference in disease commonest is a plateaued or rising hCG, but others include a
biology. The precise hCG surveillance protocol varies by tissue diagnosis of CC and spread to other organs. However, our
country, but principles are similar. In the UK, serum and urine UK experience indicates that the disease is also unlikely to
hCG is measured two weekly until normal and then monthly in spontaneously remit if the hCG is >20 000 IU/l 1 month after
urine [1]. The durations of monitoring once the hCG is normal HM evacuation (also associated with an increased risk of uterine
also vary between countries, reflecting uncertainty around the perforation) or there are lung or vaginal metastasis of >2 cm
importance of a very low risk of disease recurrence once the (smaller lesions may spontaneously regress) [1]. In addition, in
hCG is normal. Women completing the UK scheme have an the UK, chemotherapy is started to help stop heavy bleeding
estimated 1:2000 chance of missed disease [13], but the risk is that requires transfusion even if the hCG is falling. Interestingly,
already very low with the first normal hCG value even for CHM. recent data have overturned the previous UK and FIGO
The UK indications for commencing chemotherapy are listed in guidance that women who continue to have a falling hCG 6
Table 1 and are broadly similar to those of the International months after uterine evacuation automatically need
Federation of Gynecology and Obstetrics (FIGO) [14]. The chemotherapy. Indeed, the hCG spontaneously normalised in
vi | Seckl et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
Figure 3. Pelvic Doppler ultrasonography of persisting GTN following a HM. (A) Pre-chemotherapy. (B) Post-chemotherapy. (Reprinted from ref. [1],
Copyright 2010, with permission from Elsevier.)
Figure 4. Algorithm of imaging investigations for patients with GTN following a HM on hCG surveillance (left-hand panel) or after any other type of
pregnancy (right-hand panel). USS, ultrasound scan; CT, computerised tomography; MRI, magnetic resonance imaging; PET, positron emission tomography;
−ve, negative; +ve, positive; mets, metastases.
all such individuals left on surveillance [15]. Thus, this (MTX) therapy [16] and is now being evaluated in a prospective
indication for chemotherapy has now been removed from UK trial. Pulmonary metastases are most common, so a chest
guidelines. radiograph is essential [17]. Computed tomography (CT) of the
chest is not required if the chest X-ray (CXR) findings are
normal, since discovery of micrometastases, which may be seen
staging investigations and treatment stratification in ∼40% of patients, does not influence outcome [18]. However,
after a molar pregnancy if lesions are noted on CXR, magnetic resonance imaging (MRI)
Most patients developing GTN post-HM are detected early via of the brain and CT body are indicated (Figure 4) to exclude
hCG monitoring and so extensive investigation is rarely more widespread disease involving, for example, the brain or
required. Information to determine therapy can be obtained liver, which would significantly alter management.
from the clinical history, examination, measurement of serum FIGO reports data on GTN using prognostic scoring and
hCG and a Doppler pelvic ultrasound to confirm the absence of anatomic staging systems (Table 2) [19]. Since 2002, all
a pregnancy, to measure the uterine size/volume, spread of physicians treating GTN should use this system to enable the
disease within the pelvis and its vascularity (Figure 3). The latter comparison of data. The prognostic score predicts the potential
assessed by the Doppler pulsatility index is an independent for developing resistance to single-drug chemotherapy with
prognostic factor for resistance to single-agent methotrexate MTX or actinomycin D (ActD). A score of 0–6 and ≥7 indicates
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt345 | vi

by guest
on 05 February 2018
Table 2. FIGO 2000 scoring system for GTN
Prognostic factor Score

0 1 2 4
Age (years) <40 ≥40 – –
Antecedent pregnancy (AP) Mole Abortion Term –
Interval (end of AP to chemotherapy in months) <4 4–6 7–12 >12
hCG (IU/l) <103 103–104 104–105 >105
Number of metastases 0 1–4 5–8 >8
Site of metastases Lung Spleen and kidney GI tract Brain and liver
Largest tumour mass – 3–5 cm >5 cm
Prior chemotherapy – – Single drug >2 drugs
The total score for a patient is obtained by adding the individual scores for each prognostic factor. Low risk, 0–6; high risk, ≥7. PSTT should not be scored and
instead requires staging. Stage I, disease confined to the uterus; stage II, disease extending into the pelvis; stage III, disease spread to lungs and/or vagina; stage
IV, all other metastatic sites including liver, kidney, spleen and brain. (Reprinted [19] Copyright 2002, with permission from Elsevier for the International
Federation of Gynecology and Obstetrics.)
a low and high risk of resistance, respectively. The latter has need for chemotherapy remains controversial. UK results
almost no chance of being cured with single-drug therapy and indicate that this procedure is only valuable if the hCG is <5000
requires multi-agent treatment. The anatomical staging does not IU/l with disease in the cavity rather than myometrium. Indeed,
help with determining therapy, but provides additional the low efficacy of a second D&C, small risks of introducing
information to help clinicians who compare results between infection, causing haemorrhage and uterine perforation should
centres. The variables that are assessed in the prognostic score be balanced against the almost 100% cure rate and relative safety
include: (i) tumour volume (hCG level, size of metastases and of chemotherapy (reviewed in [1]). Sometimes patients with
number of metastases), (ii) site of involvement, (iii) prior stage I GTN who have completed their families request
chemotherapy resistance and (iv) duration of disease from hysterectomy, which, although possible, may not completely
antecedent pregnancy (Table 2) [19]. obviate the need for chemotherapy.
Consequently, for nearly all low-risk GTN patients, single-
agent chemotherapy with either MTX or ActD is the preferred
staging investigations for CC and PSTT/ETT
treatment. A variety of regimens have been developed, which in
Women who present with an elevated hCG and suspected GTN non-randomised, mostly retrospective, studies demonstrate a
(CC or PSTT/ETT) following a prior pregnancy require much 50%–90% chance of inducing remission [20]. This variability
more extensive staging investigations, which include a contrast reflects differences in dose, frequency and route of
enhanced CT of the chest and abdomen, MRI of the brain and administration as well as criteria used to select patients for
pelvis, a Doppler ultrasound of the pelvis and may benefit from therapy [17]. Some investigators have argued that more intense
a lumbar puncture to assess the cerebrospinal fluid to serum therapies given daily over 5–8 days every 2 weeks are superior to
hCG ratio. The latter if more than 1:60 suggests occult central treatments given once every 2 weeks [21]. Others have suggested
nervous system disease [1]. In addition, where there is doubt that ActD is more likely to induce remission than MTX. The
over the clinical diagnosis, tissue should be obtained and genetic few randomised studies to address some of these issues [22]
analysis undertaken to confirm the gestational origin of the have been underpowered and compared regimens that are not
tumour through the presence of paternal genes. For CC, the frequently used internationally [20]. Consequently, a new larger
FIGO scoring/staging system is the same as described above. international randomised trial has recently commenced
However, PSTT/ETT has a discrete biological behaviour with comparing the more commonly used MTX regimens in Europe/
less hCG production, slower growth, late metastasis and slightly many parts of the world (Table 3) and some centres elsewhere
less chemosensitivity. Consequently, the scoring system is not [MTX 0.4 mg/kg (maximum 25 mg) IV d1–5 every 2 weeks]
valid for PSTT/ETT, but FIGO staging is used to help adapt [23] with ActD 1.25 mg/m2 IV every 2 weeks. Importantly,
treatment intensity (see below). Some investigators have patients failing first-line therapy, usually because of resistance,
recently started using positron emission tomography (PET)/CT can be easily salvaged with second and occasionally third-line
imaging, but experience is still quite limited. It appears that this chemotherapy so that the overall survival (OS) is ∼100%
imaging modality is more helpful in relapsed disease to identify [23–25]. As survival is so high, it seems sensible to start with
sites for resection and, as with other cancers, is prone to both the least toxic therapy first to minimise the exposure of patients
false-positive and false-negative results [1]. to more harmful treatments.
The MTX with folinic acid rescue (MTX/FA) regimen
developed at Charing Cross Hospital (Table 3) is effective, well-
management of low-risk disease tolerated and unlike ActD, does not induce hair loss, so MTX/
About 95% of patients with HM who develop GTN are low risk FA has been widely adopted [24]. After a short stay in hospital
(score 0–6). In women with stage I disease apparently confined to monitor for bleeding complications, most of the patients can
to the uterine cavity, the role of second D&C in reducing the be treated at home, with their general practitioner, or in their
vi | Seckl et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
nearest hospital depending on local health service indicate that women in this category with an hCG of >400 000
arrangements. About 2% of women suffer mouth ulcers, sore IU/l are unlikely to be cured by MTX/FA and so multi-agent
eyes or rarely pleuritic or peritoneal pains from serositis [24]. treatment should be given from the outset [27]. Other
During chemotherapy, the hCG should ideally be measured at promising strategies to identify patients with drug resistance at
least once per week, so that at least two samples with a plateau an early time-point during initial therapy have employed
or rise are available to enable an early decision regarding the normograms and hCG kinetic analyses [28, 29].
onset of resistance indicating a need for a change in therapy. In
those developing resistance to MTX/FA, a switch to ActD or
combination agent chemotherapy depending on whether the management of high-risk GTN
hCG was <300 or >300 IU/l, respectively, will cure nearly all Patients scoring of ≥7 (Table 2) are at high risk of developing
remaining women [25]. Chemotherapy should be continued drug resistance and so are very unlikely to be cured with single-
until the hCG is normal and then for a further 6 weeks agent chemotherapy. Consequently, several different multi-
(Figure 5). The latter helps to eliminate any residual tumour agent therapies have been developed including: MTX, FA and
cells and to minimise the chances of relapse [26]. Indeed, non- ActD (MFA); MTX, ActD, cyclophosphamide, doxorubicin,
randomised data suggest that reducing the consolidation melphalan, hydroxyurea and vincristine (CHAMOCA); MTX,
therapy by just one cycle doubles the risk of relapse [26]. In view ActD and cyclophosphamide (MAC); etoposide, MTX and
of these data, the Dutch have recently moved from giving two to ActD (EMA) and others [30]. At Charing Cross Hospital, after
now using three consolidation cycles. Only 30% of patients many years of progressive experience, a regimen was developed
scoring 5–6 can expect to be cured with low-risk therapy [1]. consisting of EMA alternating weekly with cyclophosphamide
Consequently, it would be helpful to refine the FIGO scoring and vincristine (CO; see Table 4). This has been widely adopted
system, so that the 70% of women in this group who develop worldwide [30], because it is effective with predictable and easily
MTX/FA resistance could be identified initially for more managed short-term toxicity. Indeed, a retrospective
intensive therapy. It is possible that the vascularity seen on comparison from the Korean GTD centre’s experience of MFA,
Doppler ultrasound may help [16]. Moreover, recent data MAC, CHAMOCA with EMA-CO demonstrated a remission
rate of 63.3% (31 of 49), 67.5% (27 of 40), 76.2% (32 of 45) and
Table 3. Methotrexate and folinic acid chemotherapy regimen for low-risk 90.6% (87 of 96), respectively [31]. The EMA/CO regimen
patients requires one overnight stay every 2 weeks and causes reversible
alopecia. It is myelosuppressive but granulocyte colony
stimulating factor (G-CSF) support helps to maintain
Methotrexate (MTX) 50 mg by intramuscular injection repeated every
48 h for a total of four doses
neutrophil count, treatment intensity and avoid neutropenic
Calcium folinate 15 mg orally 30 h after each injection of MTX febrile episodes [1].
(folinic acid) Five-year OS of patients treated with this schedule has been
reported to vary between 75% and 90% [31–33]. In the 272
Courses repeated every 2 weeks, i.e. days 1, 15, 29, etc. cases at Charing Cross Hospital treated between 1980 and 1994,
Figure 5. HCG tumour marker treatment graph demonstrating a patient responding to low-risk chemotherapy. Following uterine evacuation of a CHM, the
hCG remained plateaued indicating persisting GTD/GTN, so the patient was commenced on methotrexate and folinic acid (MTX/FA). Therapy was continued
for 6 weeks after the hCG was normal (<5 IU/l) as indicated. (Reprinted from ref. [1], Copyright 2010, with permission from Elsevier.)
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt345 | vi

by guest
on 05 February 2018
Table 4. EMA/CO chemotherapy regimen for high-risk patients Table 5. TP/TE schedule for relapsed GTN
EMA Regimen Schedule

Day 1 Day 1
Etoposide 100 mg/m2 by i.v. infusion over Dexamethasone 20 mg oral (12 h pre-paclitaxel)
30 min Dexamethasone 20 mg oral (6 h pre-paclitaxel)
Actinomycin D 0.5 mg i.v. bolus Cimetidine 30 mg in 100 ml NS over 30 min i.v.
Methotrexate 300 mg/m2 by i.v. infusion over 12 h Chlorphenamine 10 mg bolus i.v.
Paclitaxel 135 mg/m2 in 250 ml NS over 3 h i.v.
Day 2 Mannitol 10% in 500 ml over 1 h i.v.
Etoposide 100 mg/m2 by i.v. infusion over Cisplatin 60 mg/m2 in 1 l NS over 3 h i.v.
30 min Post-hydration 1 l NS + KCl 20 mmol + 1 g MgSO4 over 2 h i.v.
Actinomycin D 0.5 mg i.v. bolus Day 15
Folinic acid rescue (starting 15 mg i.v. or orally every 12 h for four Dexamethasone 20 mg oral (12 h pre-paclitaxel)
24 h after commencing the doses Dexamethasone 20 mg oral (6 h pre-paclitaxel)
methotrexate infusion) Cimetidine 30 mg in 100 ml NS over 30 min i.v.
Chlorphenamine 10 mg bolus i.v.
CO
Paclitaxel 135 mg/m2 in 250 ml NS over 3 h i.v.
Day 8
Etoposide 150 mg/m2 in 1 l NS over 1 h i.v.
Vincristine 1 mg/m2 i.v. bolus (maximum 2 mg)
Cyclophosphamide 600 mg/m2 i.v. infusion over 30 min NS, normal saline; i.v., intravenous. (Reprinted from ref. [1], Copyright
2010, with permission from Elsevier.)
EMA alternates with CO every week. To avoid extended intervals between
courses caused by myelosuppression, it may occasionally be necessary to
reduce the EMA by omitting the day 2 doses of etoposide and actinomycin Similar to low-risk disease, therapy is continued for 6 weeks
D. i.v., intravenous. (Reprinted from ref. [1], Copyright 2010, with of normal hCG values or 8 weeks if poor prognostic features
permission from Elsevier.) such as liver or brain metastases are present [1]. Patients are
then re-imaged to document the post-treatment appearance for
future comparison. Removal of residual masses is unnecessary
OS was 86.2% [95% confidence interval (CI) 81.9% to 90.5%]
as it does not reduce the risk of recurrence which is less than
[32]. While these results were good, the presence of liver or
∼3% [1].
brain metastases correlated with only 27% or 70% long-term
survival, respectively, and was just 10% with both liver and brain
metastases (reviewed in [1]). Most of the patients with adverse
outcomes did not have a prior HM, were not registered for hCG
management of drug-resistant disease
follow-up and consequently presented with extensive disease. About 20% of high-risk GTN patients will progress on or after
This was associated with death from haemorrhage or metabolic primary chemotherapy, but these individuals still have an
complications of overwhelming disease within 4 weeks of excellent outcome with ∼75%–80% still being salvaged [36].
admission and/or before adequate chemotherapy could be This is partly because relapse is detected early due to hCG
given. If such patients are excluded, survival of patients with monitoring so disease volume is small. Moreover, hCG
brain metastasis is similar to other patients [34]. The situation monitoring enables the early detection of resistance during
with liver metastasis may be similar; of 37 patients with liver therapy, which could potentially be more rapidly detected
metastasis treated between 1977 and 2005 at Charing Cross through the use of normograms and kinetic models [28, 29, 38].
Hospital, OS had increased to ∼50% at 5 years but if early In relapsed patients, fluorine-18 fluorodeoxyglucose-PET
deaths were excluded, survival was nearly 70% [35]. In addition (FDG-PET) scanning may help identify the site of active disease
to disease extent, other factors associated with poor outcome to facilitate surgical resection and cure [39]. The T1/2 for hCG is
include the type of, and duration from, the antecedent ≤48 h after surgery if all the disease has been removed [1].
pregnancy and the prior use of chemotherapy (reviewed in [1]). However, if surgery is not possible or the hCG falls
To reduce early deaths in patients with very advanced disease, inappropriately, several salvage regimens have been either
we have found that commencing chemotherapy gently with created or adopted from the germ cell tumour setting [40]. At
low-dose etoposide 100 mg/m2 and cisplatin 20 mg/m2 on days Charing Cross Hospital, we developed a regimen combining
1 and 2 repeated weekly for 1–3 weeks has virtually eliminated etoposide with cisplatin (EP) alternating weekly with EMA that
this problem. Indeed, low-dose induction etoposide and omitted the second day of etoposide and ActD [41]. Survival
cisplatin combined with genetic testing to exclude non- rates are >80% but toxicity is significant [41], and less toxic
gestational CC has helped to improve long-term OS data to over salvage therapies are required. Several cases of drug-resistant
94% in high-risk patients [36]. Further details on the GTN have been reported to respond and/or be cured by
management and modifications of treatment required for these paclitaxel-based single-agent or combination therapy [42–45],
and other challenging clinical situations such as brain gemcitabine and capecitabine [46, 47]. Of these, an alternating
metastasis and pulmonary failure are beyond the scope of the two weekly doublet of paclitaxel/cisplatin and paclitaxel/
present review, but are contained within the following etoposide (TP/TE; Table 5) appears from non-randomised data
references [34, 37]. to be much better tolerated than EP/EMA and is effective in
vi | Seckl et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
patients with relapsed and/or refractory GTN [45]. In view of personalised medicine
these results, the International Society of the Study of
Trophoblastic Diseases (ISSTD) has recently proposed a GTN is one of the rare examples of a group of related cancers
randomised trial of TE/TP versus EP/EMA to determine the where novel molecularly targeted agents have not been
optimal therapy for patients relapsing after non-cisplatin/ employed, as cure has been achieved through the use of
paclitaxel-based combination therapies such as EMA/CO. conventional chemotherapeutic agents. This is because GTN are
Another approach in patients with refractory disease involves exquisitely sensitive to these drugs and the serial measurement
high-dose chemotherapy with peripheral stem-cell of hCG, a highly sensitive biomarker of the disease [1], has
transplantation. However, cures are not common [48], so enabled early recognition of resistance, so that second- and
improved patient selection may be required to achieve better third-line therapies can be commenced before significant
outcomes from this approach. tumour re-growth has occurred. Very rarely, multi-drug-
resistant disease develops that is not amenable to surgical
resection or any other existing treatment, so it is unclear
whether anything can be done in this case. Since GTN is very
management of PSTT and ETT vascular it is plausible that vascular targeting agents such as
PSTT differs from CC, growing more slowly, metastasising later, bevacizumab might be active. The tumours can also over-
involving lymph nodes more commonly and producing less express epidermal growth factor receptor, leading to the
hCG [1]. However, like CC, it can arise after any type of question whether erlotinib or gefitinib could demonstrate
pregnancy, including PHM, [49] and usually presents with efficacy. Anecdotally, thus far, we have not seen any benefit
abnormal vaginal bleeding [2]. PSTT may be suspected if the from these agents in several multi-drug-resistant patients. The
hCG level is low for the volume of disease present on imaging potential for an anti-hCG targeted therapy has not been
combined with an elevated free beta form of hCG, but none of explored and could be of interest in women who have
these features are diagnostic [50, 51]. Consequently histological completed their families or have run out of other options.
confirmation is essential.
A recent large population-based series of PSTT comprised 62
cases over 30 years, representing 0.2% of UK GTD cases, and
examined prognostic features [2]. On univariate analysis, stage,
hCG, mitotic index and a duration of >4 years from the The risk of relapse after chemotherapy is ∼3% and most occur
preceding pregnancy were prognostic, but the FIGO score was in the first year of follow-up. Therefore, careful hCG monitoring
unhelpful. Only the duration from the prior pregnancy is required and pregnancy should ideally be delayed until
remained predictive of survival on multivariate analysis with beyond this period. Any method of contraception can be used
100% (13 of 13) dying and 98% (48 of 49) surviving for those including the oral contraceptive pill, as long as there are no
≥48 and <48 months, respectively. This effect was not explained other contraindications to their use. In the UK, the hCG is
by differences in disease stage or hCG levels, but may reflect a monitored weekly for 6 weeks post-chemotherapy, and then in
biological switch in the tumours after this time [2]. serum and urine two weekly until 6 months, before switching to
The management of PSTT differs from CC. Patients with just urine assessments, initially monthly, but eventually
metastatic disease require combination chemotherapy with, for decreasing to just six monthly (Table 6). We continue this for
example, EP/EMA continued for 8 weeks of normal hCG levels life as we are currently uncertain when it is safe to stop
[2]. Unlike CC, residual masses are removed surgically, including monitoring and it enables us to collect long-term data
the uterus, as this can harbour microscopic disease. This may concerning late effects of treatment including second cancers.
cause difficulties in the management of stage I disease [52]. The
safest option is hysterectomy with pelvic lymph node sampling
Table 6. UK follow-up protocol of GTN patients who have been treated
and ovarian conservation unless there is a family history of
with chemotherapy
ovarian cancer or the patient is post-menopausal. In the absence
of sufficient data regarding adjuvant therapy, we currently
Low-/high-risk post-chemotherapy
advocate 8 weeks of EP/EMA or TE/TP when there are poor risk
patients, hCG concentration
factors such as disease presenting beyond 4 years of the antecedent sampling
pregnancy. Indeed, in the latter group, a case can be made for
Urine Blood
including high-dose chemotherapy. However, in younger
Year 1
nulliparous women, there is often a strong desire to preserve
Week 1–6 after chemotherapy Weekly Weekly
fertility particularly when there appears to be a focal abnormality
Month 2–6 Two weekly Two weekly
in the uterus. While uterine-sparing surgery is possible [1],
Month 7–12 Two weekly –
multifocal microscopic uterine disease can occur [52], which
Year 2 Four weekly –
could compromise survival and careful counselling is required.
Year 3 Eight weekly –
Currently, it is thought that ETT behaves very similarly to Year 4 Three monthly –
PSTT but in reality, little data are available to be sure of this. Year 5 Four monthly –
PSTT and ETT are so rare that it is unlikely that their treatment After Year 5 Six monthly –
will ever be fully optimised, so that the ISSTD has now launched
an international PSTT/ETT database to pool cases [53]. Reprinted from ref. [1], Copyright 2010, with permission from Elsevier.
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt345 | vi

by guest
on 05 February 2018
Fortunately, apart from EMA/CO bringing forward the as it is highly effective, simple to administer and relatively
menopause date by 3 years, fertility is not otherwise affected non-toxic [IV, A].
with 83% of women becoming pregnant after either MTX/FA or • Patients with high-risk disease should have maintenance
EMA/CO chemotherapy [1]. Moreover, there is no obvious therapy for 6 weeks extended to 8 weeks with poor prognostic
increase in the incidence of congenital malformations. When a features such as liver with or without brain metastasis [IV, A].
patient does become pregnant, it is important to confirm by • Early deaths in ultrahigh-risk GTN can be reduced by
ultrasound and other appropriate means that the pregnancy is induction of low-dose etoposide and cisplatin [IV, A]. Such
normal. Follow-up is then discontinued, but the hCG should be patients may also benefit from substitution of EMA/CO with
rechecked at 6 and 10 weeks after the pregnancy to ensure no EP/EMA [IV, A].
recurrence or new disease. • Residual lung or uterine masses following chemotherapy for
Late sequelae from chemotherapy have been remarkably rare. low-risk or high-risk diseases are not predictive of recurrence
In 15 279 patient-years of follow-up, there was no significant and do not require surgical excision [IV, A].
increase in the incidence of second tumours [54] following • High-risk failures can be frequently salvaged with further
MTX therapy. In contrast, 26 patients receiving combination chemotherapy and most centres use either EP/EMA or TE/TP
chemotherapy for GTN developed another cancer when the (Table 5) [IV, A]. A randomised trial comparing these
expected rate was only 16.45, a significant difference [54]. Most regimens is being developed.
of this risk appears to occur if combination chemotherapy is • Surgery alone can effectively salvage some patients with
continued beyond 6 months. Interestingly, new data in over isolated foci of chemoresistant disease [IV, A].
30 000 patient-years of follow-up now show that, for EMA/CO, • PSTT/ETT is managed according to its stage and risk factors
there is no overall increased risk of second cancers with a slight for poor outcome, the most dominant of which is the interval
but significant excess of leukaemias but reduction in other from last known pregnancy. Hysterectomy with pelvic lymph
cancers including breast cancer risk (data submitted). This node sampling is recommended for stage I disease presenting
emphasises the continued importance of long-term monitoring within 4 years of the last known pregnancy [IV, A]. Multi-
of our treated patient populations. agent chemotherapy with, for example, EP/EMA is
recommended for metastatic disease [IV, A]. Patients
presenting beyond 4 years may benefit from multi-agent and
summary of recommendations subsequent high-dose chemotherapy [IV, B].
Recommendations are largely based on non-randomised
retrospective cohort studies from single centres and/or national
experiences where the level of evidence (LOE) is IV. However, search strategy and selection criteria
because of the measurable large benefit to patients, the grade of
All authors performed a detailed review of published work and
recommendation (GOR) is generally very high at A. LOE and
contributed to the writing, review and editing of the manuscript.
GOR are given in brackets.
MJS had access to all the data used to write the report and had
• Management of GTN is optimised by the centralisation of final responsibility for submission. All authors saw and
care, pathology review and hCG monitoring [IV, A]. approved the final version. Our search strategy was formulated
• Women with singleton molar pregnancies should, in general, to identify any meta-analyses and previous systematic reviews in
have these terminated by suction D&C [IV, A]. Second D&C all aspects of GTD, in addition to all published cohort studies
for recurrence does not usually prevent the subsequent need (and where appropriate, comparison groups) and case-control
for chemotherapy and should only be attempted after studies. We searched the Cochrane Library, Medline (via
discussion with a GTD reference centre [IV, A]. PubMed, Internet Grateful Med, OVID and Knowledgefinder),
• Anti-D prophylaxis is recommended following suction D&C with a combination of keywords including: ‘trophoblastic
of PHM [IV, A]. disease’, ‘GTD’, ‘GTN’, ‘choriocarcinoma’, ‘molar pregnancy’,
• The FIGO scoring system should be used to determine the ‘hydatidiform mole’, ‘placental site trophoblastic tumor’,
risk of GTN becoming resistant to single-agent ‘genetics’, ‘epidemiology’, ‘pathology’, ‘treatment’,
chemotherapy, but is not of value in PSTT/ETT [IV, A]. ‘chemotherapy’, ‘methotrexate’, ‘actinomycin D’,
• Patients with a FIGO score of 0–6 can be treated with either ‘dactinomycin’, ‘cisplatin’, ‘paclitaxel’, ‘high-dose’,
single-agent MTX with or without FA, or ActD [II–IV, A]. In ‘management’, ‘risk factors’, ‘hCG’, ‘imaging’, ‘ultrasound’,
most European centres, MTX/FA (Table 3) is preferred ‘PET’, ‘CT’, ‘MRI’, ‘prognosis’, and ‘staging’. The reference lists
because it is less toxic than MTX alone or single-agent ActD, and bibliographies of all previous publications were scanned to
and all patients can expect to be cured even if first-line find any publications not already identified by our electronic
therapy fails [II–IV, A]. A randomised trial comparing the search strategy.
most frequently used MTX/FA and ActD regimens is
currently underway.
• Chemotherapy for low-risk disease should be continued for
note
6 weeks of maintenance treatment after hCG normalisation Levels of evidence and grades of recommendation have been
[IV, A]. applied using the system shown in Table 7. Statements without
• Patients with a FIGO score of ≥7 should receive multi-agent grading were considered justified standard clinical practice by
chemotherapy and most centres now use EMA/CO (Table 4), the experts and the ESMO faculty.
vi | Seckl et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018
Grading Systema)
Levels of evidence
I Evidence from at least one large randomised, controlled trial of good methodological quality (low potential for bias) or meta-analyses of well-conducted
a
acknowledgements 11. Sebire NJ, Savage PM, Seckl MJ, Fisher RA. Histopathological features of
biparental complete hydatidiform moles in women with NLRP7 mutations. Placenta
MJS, NS and RAF wish to thank the Department of Health, 2013; 34: 50–56.
National Commissioning Group and the Cancer Treatment and 12. Shih IM, Kurman RJ. Epithelioid trophoblastic tumor: a neoplasm distinct from
Research Trust for their continued support. MJS also choriocarcinoma and placental site trophoblastic tumor simulating carcinoma. Am
acknowledges support from the Imperial College Experimental J Surg Pathol 1998; 22: 1393–1403.
Cancer Medicine Centre and Biomedical Research Centre 13. Sebire NJ, Foskett M, Short D et al. Shortened duration of human chorionic
gonadotrophin surveillance following complete or partial hydatidiform mole:
grants. evidence for revised protocol of a UK regional trophoblastic disease unit. BJOG
2007; 114: 760–762.
14. Kohorn EI. Negotiating a staging and risk factor scoring system for gestational
conflict of interest trophoblastic neoplasia. A progress report. J Reprod Med 2002; 47: 445–450.
The authors have declared no potential conflicts of interest. 15. Agarwal R, Teoh S, Short D et al. Chemotherapy and human chorionic
gonadotropin concentrations 6 months after uterine evacuation of molar
pregnancy: a retrospective cohort study. Lancet 2012; 379: 130–135.
16. Agarwal R, Harding V, Short D et al. Uterine artery pulsatility index: a predictor of
references methotrexate resistance in gestational trophoblastic neoplasia. Br J Cancer 2012;
1. Seckl MJ, Sebire NJ, Berkowitz RS. Gestational trophoblastic disease. Lancet 106: 1089–1094.
2010; 376: 717–729. 17. Berkowitz RS, Goldstein DP. Current management of gestational trophoblastic
2. Schmid P, Nagai Y, Agarwal R et al. Prognostic markers and long-term outcome of diseases. Gynecol Oncol 2009; 112: 654–662.
placental-site trophoblastic tumours: a retrospective observational study. Lancet 18. Darby S, Jolley I, Pennington S, Hancock BW. Does chest CT matter in the staging
2009; 374: 48–55. of GTN? Gynecol Oncol 2009; 112: 155–160.
3. Hou JL, Wan XR, Xiang Y et al. Changes of clinical features in hydatidiform mole: 19. FIGO Oncology Committee, FIGO staging for gestational trophoblastic neoplasia
analysis of 113 cases. J Reprod Med 2008; 53: 629–633. 2000. International Journal of Gynecology & Obstetrics 77: 285–287.
4. Fowler DJ, Lindsay I, Seckl MJ, Sebire NJ. Histomorphometric features of 20. Alazzam M, Tidy J, Hancock BW et al. First line chemotherapy in low risk
hydatidiform moles in early pregnancy: relationship to detectability by ultrasound gestational trophoblastic neoplasia. Cochrane Database Syst Rev 2009 Jan 21; (1):
examination. Ultrasound Obstet Gynecol 2007; 29: 76–80. CD007102.
5. Hinshaw K, Fayyad A, Munjuluri P. The management of early pregnancy loss. In 21. Kohorn EI. Is lack of response to single-agent chemotherapy in gestational
Green-top Guideline. London: Royal College of Obstetricians and Gynaecologists, trophoblastic disease associated with dose scheduling or chemotherapy
2006. resistance? Gynecol Oncol 2002; 85: 36–39.
6. Seckl MJ, Gillmore R, Foskett M et al. Routine terminations of pregnancy-should 22. Osborne RJ, Filiaci V, Schink JC et al. Phase III trial of weekly methotrexate or
we screen for gestational trophoblastic neoplasia. Lancet 2004; 364: 705–707. pulsed dactinomycin for low-risk gestational trophoblastic neoplasia: a gynecologic
7. Fisher RA, Lavery SA, Carby A et al. What a difference an egg makes. Lancet oncology group study. J Clin Oncol 2011; 29: 825–831.
2011; 378: 1974. 23. Lurain JR, Chapman-Davis E, Hoekstra AV, Schink JC. Actinomycin D for
8. Sebire NJ, Foskett M, Paradinas FJ et al. Outcome of twin pregnancies with methotrexate-failed low-risk gestational trophoblastic neoplasia. J Reprod Med
complete hydatidiform mole and healthy co-twin. Lancet 2002; 359: 2165–2166. 2012; 57: 283–287.
9. Fisher RA, Savage PM, MacDermott C et al. The impact of molecular genetic 24. McNeish IA, Strickland S, Holden L et al. Low risk persistent gestational
diagnosis on the management of women with hCG-producing malignancies. trophoblastic disease: outcome after initial treatment with low-dose methotrexate
Gynecol Oncol 2007; 107: 413–419. and folinic acid, 1992 to 2000. J Clin Oncol 2002; 20: 1838–1844.
10. Sebire NJ, Seckl MJ. Immunohistochemical staining for diagnosis and prognostic 25. Sita-Lumsden A, Short D, Lindsay I et al. Treatment outcomes for 618 women
assessment of hydatidiform moles: current evidence and future directions. with gestational trophoblastic tumours following a molar pregnancy at the Charing
J Reprod Med 2010; 55: 236–246. Cross Hospital, 2000–2009. Br J Cancer 2012; 107: 1810–1814.
Volume 24 | Supplement 6 | October 2013 doi:10.1093/annonc/mdt345 | vi

by guest
on 05 February 2018
26. Lybol C, Sweep FC, Harvey R et al. Relapse rates after two versus three 40. Lurain JR, Nejad B. Secondary chemotherapy for high-risk gestational
consolidation courses of methotrexate in the treatment of low-risk gestational trophoblastic neoplasia. Gynecol Oncol 2005; 97: 618–623.
trophoblastic neoplasia. Gynecol Oncol 2012; 125: 576–579. 41. Newlands ES, Mulholland PJ, Holden L et al. Etoposide and cisplatin/etoposide,
27. McGrath S, Short D, Harvey R et al. The management and outcome of women with methotrexate, and actinomycin D (EMA) chemotherapy for patients with high-risk
post-hydatidiform mole ‘low-risk’ gestational trophoblastic neoplasia, but hCG gestational trophoblastic tumors refractory to EMA/cyclophosphamide and
levels in excess of 100 000 IU l(-1). Br J Cancer 2010; 102: 810–814. vincristine chemotherapy and patients presenting with metastatic placental site
28. van Trommel NE, Massuger LF, Schijf CP et al. Early identification of resistance to trophoblastic tumors. J Clin Oncol 2000; 18: 854–859.
first-line single-agent methotrexate in patients with persistent trophoblastic 42. Jones WB, Schneider J, Shapiro F, Lewis JL, Jr. Treatment of resistant gestational
disease. J Clin Oncol 2006; 24: 52–58. choriocarcinoma with taxol: a report of two cases. Gynecol Oncol 1996; 61:
29. You B, Harvey R, Henin H et al. Early prediction of treatment resistance in low-risk 126–130.
gestational trophoblastic neoplasia using population kinetic modelling of hCG 43. Osborne R, Covens A, Mirchandani D, Gerulath A. Successful salvage of relapsed
measurements. Br J Cancer 2013; 108: 1810–1816. high-risk gestational trophoblastic neoplasia patients using a novel paclitaxel-
30. Deng L, Yan X, Zhang J et al. Combination chemotherapy for high-risk gestational containing doublet. J Reprod Med 2004; 49: 655–661.
trophoblastic tumour. Cochrane Database Syst Rev 2009 April 15; (2): CD005196. 44. Termrungruanglert W, Kudelka AP, Piamsomboon S et al. Remission of refractory
31. Kim SJ, Bae SN, Kim JH et al. Effects of multiagent chemotherapy and gestational trophoblastic disease with high-dose paclitaxel. Anticancer Drugs
independent risk factors in the treatment of high-risk GTT—25 years experiences 1996; 7: 503–506.
of KRI-TRD. Int J Gynaecol Obstet 1998; 60 (Suppl 1): S85–S96. 45. Wang J, Short D, Sebire NJ et al. Salvage chemotherapy of relapsed or high-risk
32. Bower M, Newlands ES, Holden L et al. EMA/CO for high-risk gestational gestational trophoblastic neoplasia (GTN) with paclitaxel/cisplatin alternating with
trophoblastic tumours: results from a cohort of 272 patients. J Clin Oncol 1997; paclitaxel/etoposide (TP/TE). Ann Oncol 2008; 19:
15: 2636–2643. 1578–1583.
33. Turan T, Karacay O, Tulunay G et al. Results with EMA/CO (etoposide, 46. Pandian Z, Seckl MJ, Smith R, Lees DA. Gestational choriocarcinoma: an unusual
methotrexate, actinomycin D, cyclophosphamide, vincristine) chemotherapy in presentation with response to gemcitabine and surgery. BJOG 2004; 111:
gestational trophoblastic neoplasia. Int J Gynecol Cancer 2006; 16: 1432–1438. 382–384.
34. Newlands ES, Holden L, Seckl MJ et al. Management of brain metastases in 47. Bianconi M, Jankilevich G, Otero S et al. Successful salvage of a relapsed high
patients with high-risk gestational trophoblastic tumors. J Reprod Med 2002; 47: risk gestational trophoblastic neoplasia patient using capecitabine. Gynecol Oncol
465–471. 2007; 106: 268–271.
35. Ahamed E, Short D, North B et al. Survival of women with gestational trophoblastic 48. El-Helw LM, Seckl MJ, Haynes R et al. High-dose chemotherapy and peripheral
neoplasia and liver metastases: is it improving? J Reprod Med 2012; 57: 262–269. blood stem cell support in refractory gestational trophoblastic neoplasia. Br J
36. Alifrangis C, Agarwal R, Short D et al. EMA/CO for high-risk gestational Cancer 2005; 93: 620–621.
trophoblastic neoplasia: good outcomes with induction low-dose etoposide- 49. Palmieri C, Fisher RA, Sebire NJ et al. Placental site trophoblastic tumour arising
cisplatin and genetic analysis. J Clin Oncol 2013; 31: 280–286. from a partial hydatidiform mole. Lancet 2005; 366: 688.
37. Seckl MJ, Newlands ES. Investigation and treatment of patients with persistent 50. Cole LA, Khanlian SA, Muller CY et al. Gestational trophoblastic diseases: 3.
gestational trophoblastic disease and gestational trophoblastic tumours/neoplasia Human chorionic gonadotropin-free beta-subunit, a reliable marker of placental
in the United Kingdom. In: Hancock BW, Seckl MJ, Berkowitz RS, Cole LA (eds) site trophoblastic tumors. Gynecol Oncol 2006; 102: 160–164.
Gestational Trophoblastic Disease, 3rd Edition, 2009; 335–365; ISSTD.org. 51. Harvey RA, Pursglove HD, Schmid P et al. Human chorionic gonadotropin free
ISSTD, London. beta-subunit measurement as a marker of placental site trophoblastic tumors. J
38. Lybol C, Westerdijk K, Sweep FC et al. Human chorionic gonadotropin (hCG) Reprod Med 2008; 53: 643–648.
regression normograms for patients with high-risk gestational trophoblastic 52. Pfeffer PE, Sebire N, Lindsay I et al. Fertility-sparing partial hysterectomy for
neoplasia treated with EMA/CO (etoposide, methotrexate, actinomycin D, placental-site trophoblastic tumour. Lancet Oncol 2007; 8: 744–746.
cyclophosphamide and vincristine) chemotherapy. Ann Oncol 2012; 23: 53. The ISSTD global Placental Site Trophoblastic Tumour database. International
2903–2906. Society of the Study of Trophoblastic Diseases. https://pstt.shef.ac.uk.
39. Dhillon T, Palmieri C, Sebire NJ et al. Value of whole body 18FDG-PET to identify 54. Rustin GJ, Newlands ES, Lutz JM et al. Combination but not single agent
the active site of gestational trophoblastic neoplasia. J Reprod Med 2006; 51: methotrexate chemotherapy for gestational trophoblastic tumours increases the
879–887. incidence of second tumours. J Clin Oncol 1996; 14: 2769–2773.
vi | Seckl et al. Volume 24 | Supplement 6 | October 2013

by guest
on 05 February 2018

ESMO-ESGO-ESTRO Consensus Conference

on Endometrial Cancer: diagnosis, treatment
and follow-up†
N. Colombo1*, C. Creutzberg2, F. Amant3,4, T. Bosse5, A. González-Martín6,7, J. Ledermann8,
C. Marth9, R. Nout10, D. Querleu11,12, M.R. Mirza13, C. Sessa14 & the ESMO-ESGO-ESTRO
Endometrial Consensus Conference Working Group‡
1
Division of Medical Gynecologic Oncology, European Institute of Oncology and University of Milan-Bicocca, Milan, Italy; 2Department of Radiation Oncology, Leiden
University Medical Center, Leiden, The Netherlands; 3Department of Gynecological Oncology, University Hospital Leuven, Leuven, Belgium; 4Center for Gynecological
Oncology Amsterdam (CGOA), Antoni van Leeuwenhoek, Amsterdam; 5Department of Pathology, Leiden University Medical Center, Leiden, The Netherlands; 6Department
of Medical Oncology, GEICO Cancer Center, Madrid; 7Department of Medical Oncology, MD Anderson Cancer Center, Madrid, Spain; 8Department of Oncology and
Cancer Trials, UCL Cancer Institute, London, UK; 9Department of Obstetrics and Gynecology, Innsbruck Medical University, Innsbruck, Austria; 10Department of
Radiotherapy, Leiden University Medical Center, Leiden, The Netherlands; 11Department of Surgery, Institut Bergonié, Bordeaux, France; 12Department of Gynecology and
Obstetrics, McGill University Health Centre, Montreal, Canada; 13Department of Oncology, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark;
14
Department of Medical Oncology, Oncology Institute of Southern Switzerland, Ospedale San Giovanni, Bellinzona, Switzerland
Received 17 July 2015; revised 30 September 2015; accepted 5 October 2015
The first joint European Society for Medical Oncology (ESMO), European SocieTy for Radiotherapy & Oncology (ESTRO)
and European Society of Gynaecological Oncology (ESGO) consensus conference on endometrial cancer was held on
11–13 December 2014 in Milan, Italy, and comprised a multidisciplinary panel of 40 leading experts in the management
of endometrial cancer. Before the conference, the expert panel prepared three clinically relevant questions about endo-
metrial cancer relating to the following four areas: prevention and screening, surgery, adjuvant treatment and advanced
and recurrent disease. All relevant scientific literature, as identified by the experts, was reviewed in advance. During the
consensus conference, the panel developed recommendations for each specific question and a consensus was reached.
Results of this consensus conference, together with a summary of evidence supporting each recommendation, are
detailed in this article. All participants have approved this final article.
Key words: endometrial neoplasms, practice guideline, consensus, treatment, adjuvant, surgery
introduction than 40 years old [2], many of whom still wish to retain their fer-
tility. The majority of endometrial cancers are diagnosed early
Endometrial cancer is the most common gynaecological cancer (80% in stage I), with 5-year survival rates of over 95%. However,
in developed countries. The number of newly diagnosed cases in 5-year survival rates are much lower if there is regional spread or
Europe was nearly 100 000 in 2012, with an age standardised in- distant disease (68% and 17%, respectively) [3].
cidence of 13.6 per 100 000 women. Cumulative risk for a diag- Historically, endometrial carcinoma has been classified into two
nosis of endometrial cancer is 1.71% [1]. main clinicopathological and molecular types: type I is the much
More than 90% of cases of endometrial cancer occur in women more common endometrioid adenocarcinoma (80%–90%) and
>50 years of age, with a median age at diagnosis of 63 years. type II comprises non-endometrioid subtypes such as serous, clear-
However, 4% of women with endometrial cancer are younger cell and undifferentiated carcinomas, as well as carcinosarcoma/
malignant-mixed Müllerian tumour (10%–20%) [4]. Molecular
data in support of this dichotomous classification have become
*Correspondence to: Prof. Nicoletta Colombo, ESMO Guidelines Committee, ESMO
Head Office, Via L. Taddei 4, CH-6962 Viganello-Lugano, Switzerland;
an integral component of pathologic evaluation, as type I car-
E-mail: clinicalguidelines@esmo.org cinomas are preferentially associated with genetic alterations in
†
PTEN, KRAS, CTNNB1 and PIK3CA and MLH1 promoter
These Guidelines were developed by the European Society for Medical Oncology
(ESMO), the European Society of Gynaecological Oncology (ESGO), and the European
hypermethylation, whereas serous carcinomas prototypically
SocieTy of Radiotherapy and Oncology (ESTRO), and are published jointly in the Annals harbour TP53 mutations. However, this dualistic model has
of Oncology, the International Journal of Gynecological Cancer and Radiotherapy & limitations as considerable molecular heterogeneity exists; for
Oncology. The three societies nominated participants who attended the consensus con- example, 25% of high-grade endometrioid carcinomas express
ference and co-authored the final manuscript.
‡
See appendix for members of the ESMO-ESGO-ESTRO Endometrial Consensus
mutated TP53 and behave like serous carcinomas [5]. Extensive
Conference Working Group. work performed by The Cancer Genome Atlas (TCGA) Research

by guest
on 05 February 2018
Network has significantly improved our understanding of the mo- 6. How radical should the surgery be in different stages and
lecular landscape of endometrial cancer, introducing not two, but pathological subtypes of endometrial cancer?
four molecular subtypes including: (i) POLE (ultra-mutated) 7. What is the current best definition of risk groups for adju-
tumours, (ii) microsatellite unstable tumours, (iii) copy-number vant therapy?
high tumours with mostly TP53 mutations and (iv) remaining 8. What are the best evidence-based adjuvant treatment strat-
group without these alterations [6]. Hereditary endometrial adeno- egies for patients with low- and intermediate-risk endomet-
carcinomas are mostly seen in families with hereditary non-polyp- rial cancer?
osis colon cancer [HNPCC, Lynch syndrome (LS)]. Although the 9. What are the best evidence-based adjuvant treatment strat-
majority of endometrial carcinomas related to LS are type I egies for patients with high-risk endometrial cancer?
cancers, the proportion of type II cancers seems to be higher than 10. Does surgery or radiotherapy (RT) have a role in advanced
in the case of sporadic endometrial carcinoma [7]. or recurrent endometrial cancer?
Although the majority of cases of endometrial cancer are diag- 11. What are the optimal systemic therapies for advanced/re-
nosed at an early stage, differences in patient characteristics and current disease?
histopathological features of the disease impact on both patient 12. What are the most promising targeted agents and which study
prognosis and the recommended treatment approach. Given the designs should be used to evaluate their clinical benefit?
large body of literature available that addresses the management
of endometrial cancer, the aim of this consensus conference was Each working group was responsible for reviewing the rele-
to produce multidisciplinary evidence-based guidelines on vant literature in order to draft preliminary recommendations
selected clinically relevant questions in order to complement the relating to each of their assigned questions. No systematic litera-
already available European Society for Medical Oncology ture search was undertaken. During the conference, in parallel
(ESMO) Clinical Practice Guidelines (CPG) for the diagnosis, sessions, the four working groups discussed and reached agree-
treatment and follow-up of patients with endometrial cancer [8]. ment on recommendations relating to each of their assigned
questions. Recommendations from each group were then pre-
sented to the entire panel of experts, where they were discussed
methods and modified as required. An adapted version of the ‘Infectious
In 2014, ESMO decided to update the clinical recommendations Diseases Society of America-United States Public Health Service
for endometrial cancer using a consensus conference approach. Grading System’ was used (Table 1 [9]) to define the level of evi-
The consensus panel comprised 40 experts in the management of dence and strength of each recommendation proposed by the
endometrial cancer, and included representation from the group. Finally, a vote was conducted to determine the level of
European SocieTy for Radiotherapy & Oncology (ESTRO), the agreement among the expert panel for each of the recommenda-
European Society of Gynaecological Oncology (ESGO) and tions. Panel members were allowed to abstain from voting in
ESMO. Each panel member was assigned to one of four working cases where they either had insufficient expertise to agree/
groups, with a working group chair and co-chair appointed for
each group. Three consensus conference chairs (N. Colombo, Table 1. Levels of evidence and grades of recommendation
C. Creutzberg, C. Sessa) were also appointed.
Levels of evidence
Each working group was assigned a subject area as follows:
1. Prevention and screening of endometrial cancer (Chair: F. good methodological quality (low potential for bias) or meta-
Amant; Co-Chair: T. Bosse) analyses of well-conducted, randomised trials without
2. Surgery (Chair: C. Marth; Co-Chair: D. Querleu) heterogeneity
3. Adjuvant treatment (Chair: R. Nout; Co-Chair: M. R. II Small randomised trials or large randomised trials with a suspicion
Mirza)
4. Advanced and recurrent disease (Chair: J. Ledermann;
Co-Chair: A. González-Martín)
The consensus conference was held on 11–13 December 2014 in V Studies without control group, case reports, experts opinions
Milan, Italy. Before this consensus conference, three clinically rele- Grades of recommendation
vant questions were identified for each subject area/working group, A Strong evidence for efficacy with a substantial clinical benefit,
giving a total of 12 clinically relevant questions as follows: strongly recommended
1. Which surveillance should be used for asymptomatic women?
benefit, generally recommended
2. What work-up and management scheme should be under-
taken for fertility-preserving therapy in patients with atypical
risk or the disadvantages (adverse events, costs, …), optional
hyperplasia (AH)/endometrial intraepithelial neoplasia D Moderate evidence against efficacy or for adverse outcome,
(EIN) and grade 1 endometrioid endometrial cancer (EEC)? generally not recommended
3. Which (molecular) markers can help distinguish ( pre)can- E Strong evidence against efficacy or for adverse outcome, never
cerous lesions from benign mimics? recommended
4. How does the medical condition influence surgical treatment?
5. What are the indications for and to what extent is lympha- By permission of the Infectious Diseases Society of America-United
denectomy indicated in the surgical management of endo- States Public Health Service Grading System [9].
metrial cancer?
Volume 27 | No. 1 | January 2016 doi:10.1093/annonc/mdv484 | 

by guest
on 05 February 2018
disagree with the recommendation or if they had a conflict of Other risk factors for endometrial cancer include unopposed
interest that could be considered as influencing their vote. oestrogen therapy, oestrogen-producing tumours and early menar-
Results of this consensus conference, together with a che/late menopause. Unopposed oestrogen therapy increases the
summary of evidence supporting each recommendation, are risk for endometrial cancer 10- to 30-fold if treatment continues 5
detailed in this article, and a summary of all recommendations years or more [18]. Oestrogen-producing tumours, or ovarian gran-
is included in supplementary Table S1, available at Annals of ulosa, and theca cell tumours carry an increased risk for endomet-
Oncology online. However, these additional recommendations rial cancer, with up to 20% of women with these tumours reported
for specific clinical situations should be read in conjunction as having a simultaneous endometrial cancer [19]. Both early me-
with the ESMO CPG for the diagnosis, treatment and follow-up narche and late menopause are associated with a 2-fold increased
of patients with endometrial cancer [8]. risk for endometrial cancer. The RR is 2.4 for women <12 versus
≥15 years [20] and is 1.8 for women ≥55 versus <50 years [21].
results Studies of women with breast cancer taking tamoxifen with
therapeutic or preventive intent have shown that the RR of devel-
prevention and screening of endometrial cancer oping endometrial cancer is 2.53 times higher than that of an age-
risk factors for endometrial cancer. Most patients with matched population. This risk differs depending on menopausal
endometrial cancer have an identifiable source of excess oestrogen status. Premenopausal women treated with tamoxifen have no
and typically display a characteristic clinical profile comprising a known increased risk of endometrial cancer, while this risk in
high body mass index (BMI) that is considered as overweight postmenopausal women is 4.0 (95% CI 1.70–10.90) [22]. The level
(BMI 25–30) or obese (BMI 30), often with other components of of risk of endometrial cancer is also dose and time dependent.
metabolic syndrome (e.g. hypertension, diabetes). The evidence LS or HNPCC is an autosomal dominant inherited disorder
that greater body fatness (reflected by BMI, measures of abdominal caused by germline mutations in DNA mismatch repair genes.
girth and adult weight gain) is a cause of endometrial cancer is Women with mutations in MLH1, MSH2, MSH6 or PMS2 have
convincing. Glycaemic load is probably a cause of endometrial up to a 40%–60% lifetime risk of developing both endometrial
cancer, while the evidence suggesting that sedentary habits and colorectal cancers, as well as a 9%–12% lifetime risk of
(marked by sitting time) and adult attained height are causes of developing ovarian cancer [23].
endometrial cancer is limited [10].
screening and prevention of endometrial cancer. Most cases of
High BMI correlates with good prognostic features of endo-
endometrial cancer cannot be prevented, but reducing the risk
metrial cancer, including low tumour grade, endometrioid hist-
factors and introducing protective factors into the lifestyle
ology and presentation at early stage. In a small subset of
whenever possible, may lower the risk of developing this disease.
patients, the pathogenesis is related to mismatch repair abnor-
All women should be told about the risks and symptoms of
mality and LS. Tumours associated with mismatch repair abnor-
endometrial cancer and be strongly encouraged to engage in
malities and LS appear to be distinct, with worse prognostic
regular physical activity (exercise) and adopt an active lifestyle
factors and worse clinical outcome [11].
which can help to attain and maintain a healthy weight as well
According to a recent meta-analysis involving 6 studies and
as lowering the risk of other risk factors for endometrial cancer
3132 cancer cases, relative risk (RR) for developing endometrial
such as high blood pressure and diabetes. The use of combined
cancer in women with metabolic syndrome is 1.89 [95% confi-
oral contraceptives is significantly associated with a decrease in
dence interval (CI) 1.34–2.67, P ≤ 0.001]. According to individual
endometrial cancer in ever users, a benefit that is greater with
components of metabolic syndrome, obesity is associated with the
increasing duration of use.
greatest increase in RR of 2.21 (P ≤ 0.001) [12]. The strength of as-
sociation between obesity and cancer risk increases with increasing
BMI: RR for overweight is 1.32 (95% CI 1.16–1.50) and for obesity 1. Which surveillance should be used for asymptomatic
is 2.54 (95% CI 2.11–3.06) [13]. Other components of the meta- women?
bolic syndrome linked to endometrial cancer include hyperten-
sion, with a RR of 1.81 (P = 0.024) [12] or an odds ratio (OR) of women with average risk for endometrial cancer. There is no
1.77 (1.34–2.34) [14]. Hypertriglyceridaemia has a weaker but still indication that population-based screening has a role in the early
significant association (RR 1.17, P < 0.001) [12]. detection of endometrial cancer among women who are at
Diabetes mellitus, in particular type II, has long been held as an average endometrial cancer risk and have no symptoms. There is
independent risk factor for endometrial cancer, with an approxi- also no standard or routine screening test for endometrial cancer.
mate doubling of risk (OR 2.1; 95% CI 1.40–3.41) [14]. However, Screening of asymptomatic women for endometrial carcinoma
the fact that people with type II diabetes mellitus (T2DM) tend to has in general been recommended only for those with LS [24, 25].
be obese is a confounding factor, and a recent epidemiological There is no evidence that screening by ultrasonography (e.g.
study from the United States questioned the independent role of endovaginal or transvaginal ultrasound) reduces mortality from
T2DM as a risk factor for endometrial cancer [15]. endometrial cancer. Moreover, cohort studies indicate that
Nulliparity and infertility are also classical risk factors for endo- screening asymptomatic women will result in unnecessary add-
metrial cancer. Among the causes of infertility, polycystic ovarian itional biopsies because of false-positive test results. Risks asso-
syndrome (PCOS) seems to be the most important, with an ciated with false-positive tests include anxiety and complications
almost threefold increase in risk (OR 2.79–2.89) [16]. However, from biopsies [26].
as with diabetes, obesity seems to be a confounding factor, and At the time of menopause, women should be strongly encour-
the BMI-adjusted OR is lower (2.2; 95% CI 0.9–5.7) [17]. aged to report any vaginal bleeding, discharge or spotting to
 | Colombo et al. Volume 27 | No. 1 | January 2016

by guest
on 05 February 2018
their doctor to ensure they receive appropriate treatment of any Recommendation 1.6: Routine screening for endometrial
precancerous disorders of the endometrium. cancer in asymptomatic tamoxifen users is not recommended
Recommendation 1.1: There is no evidence for endometrial Level of evidence: III
cancer screening in the general population Strength of recommendation: B
Level of evidence: II Consensus: 94.6% (35) yes, 5.4% (2) abstain (37 voters)
Consensus: 100% yes (37 voters) women with high risk for endometrial cancer. Women with a
high risk for endometrial cancer include known carriers of
women at increased risk for endometrial cancer. Women at HNPCC-associated genetic mutations, those who have a substantial
increased risk for endometrial cancer due to a history of unopposed likelihood of being a mutation carrier (i.e. a mutation is known to
oestrogen therapy, late menopause, tamoxifen therapy, nulliparity, be present in the family) and those without genetic testing results
infertility or failure to ovulate, obesity, diabetes or hypertension but who are from families with a suspected autosomal dominant
should be informed of the risks and symptoms of endometrial predisposition to colon cancer.
cancer and strongly encouraged to report any unexpected bleeding Findings from a prospective observational cohort study of
or spotting to their physicians. women with LS opting for endometrial cancer screening and
Asymptomatic women with risk factors for endometrial who underwent annual outpatient hysteroscopy and endomet-
cancer who have endometrial thickening and other positive rial sampling (OHES) suggest that, in women with LS, annual
findings on ultrasound, such as increased vascularity, inhomo- OHES is acceptable and has high diagnostic accuracy in screen-
geneity of the endometrium, particulate fluid or thickened endo- ing for endometrial cancer and atypical endometrial hyperplasia
metrium over 11 mm should be managed on a case-by-case (AEH) [29]. However, larger international studies are needed
basis. The potential benefits, risks and limitations of testing for for confirmation.
early endometrial cancer should be explained in order to ensure Women with an HNPCC-associated mutation or with a sub-
informed decision making about testing. stantial likelihood of having an HNPCC-associated mutation
Premenopausal women treated with tamoxifen do not require should be informed of the potential benefits, risks and limita-
additional monitoring beyond routine gynaecological care. tions of testing for early endometrial cancer; they should also be
Postmenopausal women taking tamoxifen should be informed informed that the recommendation for screening is based on
about symptoms of endometrial hyperplasia or cancer [27]. expert opinion in the absence of definitive scientific evidence.
Although findings from a recently published meta-analysis have Although there is insufficient evidence to endorse annual
verified the efficacy of the levonorgestrel intrauterine device screening for endometrial cancer in this group, annual screening
(LNG-IUD) in preventing de novo polyps in breast cancer patients beginning at age 35 is recommended due to the high risk of
treated with tamoxifen, there was insufficient evidence to ascertain endometrial cancer and the potentially life-threatening nature of
whether the LNG-IUD was associated with any benefit in redu- this disease. As screening will be of limited efficacy in gynaeco-
cing the incidence of precancerous or cancerous lesions [28]. logical cancers (endometrial and ovarian), once the family is
Recommendation 1.2: Unopposed oestrogen treatment completed, particularly by age 35–40 years, careful consider-
should not be started or should be discontinued in women with ation must be given to the option of prophylactic hysterectomy
a uterus in situ and bilateral salpingo-oophorectomy [30].
Level of evidence: III In women with LS, the following options are available:
Strength of recommendation: A • Annual screening beginning at age 35 (recommended)
Consensus: 100% yes (37 voters) • Regular hysteroscopy and endometrial biopsies or hysterec-
Recommendation 1.3: Routine surveillance in asymptomatic tomy (current options)
women with obesity, PCOS, diabetes mellitus, infertility, nulli- • The application of local progesterone using the LNG-IUD
parity or late menopause is not recommended • Treatment of premalignant disease (AEH, EIN)
Level of evidence: III • Hysterectomy and bilateral oophorectomy
Consensus: 100% yes (37 voters) Evaluating the likelihood of a patient having a gynaecological
Recommendation 1.4: For women with adult granulosa cell cancer predisposition syndrome enables the physician to provide
tumour, if hysterectomy has not been performed, endometrial individualised assessments of cancer risk, as well as the opportunity
sampling is recommended. If this shows no evidence of ( pre) to offer tailored screening and prevention strategies such as surveil-
malignancy, no further screening for endometrial malignancies lance, chemoprevention and prophylactic surgery that may reduce
is required the morbidity and mortality associated with these syndromes.
Level of evidence: IV Recommendation 1.7: Surveillance of the endometrium by
Strength of recommendation: B gynaecological examination, transvaginal ultrasound and aspir-
Consensus: 100% yes (37 voters) ation biopsy starting from the age of 35 years (annually until
Recommendation 1.5: In patients with epithelial ovarian hysterectomy) should be offered to all LS mutation carriers
cancer undergoing fertility-sparing treatment, endometrial sam- Level of evidence: IV
pling is recommended at the time of diagnosis Strength of recommendation: B
Level of evidence: IV Consensus: 97.3% (36) yes, 2.7% (1) abstain (37 voters)
Strength of recommendation: B Recommendation 1.8: Prophylactic surgery (hysterectomy
Consensus: 100% yes (37 voters) and bilateral salpingo-oophorectomy), preferably using a
Volume 27 | No. 1 | January 2016 doi:10.1093/annonc/mdv484 | 

by guest
on 05 February 2018
minimally invasive approach, should be discussed at the age of Level of evidence: IV

40 as an option for LS mutation carriers to prevent endometrial Strength of recommendation: A
and ovarian cancer. All pros and cons of prophylactic surgery Consensus: 100% yes (37 voters)
must be discussed Recommendation 2.4: Pelvic MRI should be performed to
Level of evidence: IV exclude overt myometrial invasion and adnexal involvement.
Strength of recommendation: B Expert ultrasound can be considered as an alternative
Consensus: 100% yes (37 voters) Level of evidence: III
2. What work-up and management scheme should be under- Consensus: 100% yes (37 voters)
taken for fertility-preserving therapy in patients with AH/ Recommendation 2.5: Patients must be informed that fertility-
EIN and grade 1 EEC? sparing treatment is a non-standard treatment and the pros and
cons must be discussed. Patients should be willing to accept close
work-up for fertility-preserving therapy. The diagnosis of follow-up and be informed of the need for future hysterectomy
endometrial carcinoma in young women of childbearing age is Level of evidence: V
rare. Indeed, only 4% of patients with endometrial carcinoma Strength of recommendation: A
are <40 years of age [2]. Younger and premenopausal women Consensus: 97.3% (36) yes, 2.7% (1) abstain (37 voters)
with endometrial carcinoma seem to have a better prognosis
than older patients, with increased rates of early-stage and low- management schemes for fertility-preserving therapy. Conservative
grade disease reported [2, 31, 32]. medical treatment for endometrial cancer is based on progestins
The standard approach for the management of endometrial with medroxyprogesterone acetate (MPA; 400–600 mg/day) or
cancer in young women of childbearing age is hysterectomy and bi- megestrol acetate (MA; 160–320 mg/day) [33]. Few papers have
lateral salpingo-oophorectomy with or without lymphadenectomy. addressed the use of LNG-IUD but preliminary data using such
Although this is a highly effective approach, carrying a 5-year sur- treatment [added to gonadotropin-releasing hormone (GnRH)
vival rate of 93%, it also results in a permanent loss of reproductive analogues] seem to demonstrate similar remission and recurrence
potential. Conservative management of endometrial carcinoma is rates as oral progestins [37]. Assessment of response must be
based on medical treatment with oral progestins. The most import- performed at 6 months with a new D&C and imaging [38].
ant issues when considering a conservative management approach Response rates associated with the conservative management
are the assessment of clinical and pathological characteristics of the of endometrial carcinoma are ∼75% [39, 40], but recurrence
tumour and selection of the appropriate medical intervention. rates are 30%–40% [39, 41, 42]. Standard surgery with hysterec-
A conservative management approach could be considered in tomy should be proposed to non-responders while maintenance
patients with a histological diagnosis of grade 1 endometrial car- treatment for a further 6 months can be considered in respon-
cinoma (or premalignant disease such as AH) [31]. The optimal ders who wish to delay pregnancy [33].
method to obtain these histologic characteristics is dilatation Although progesterone receptor (PgR) status is a reliable predict-
and curettage (D&C) [33]; this procedure is superior to pipelle ive factor for disease remission, a routine check is not recommended
biopsy in terms of accuracy of the tumour grade [34]. since 50% of PgR-negative patients will respond to treatment [43].
The histological diagnosis should be reviewed by an expert Pregnancy is associated with a reduced risk for endometrial
pathologist to improve the accuracy of histological assessment cancer recurrence [40]. Findings from recent meta-analyses
(endometrial carcinoma or AH) and the reliability of tumour showed that the pooled live birth rate among women receiving
grading [35], whereas the initial stage should be confirmed by fertility-preserving treatment for endometrial cancer was 28%
enhanced pelvic magnetic resonance imaging (MRI) to exclude and reached 39% when assisted reproduction technology was
overt myometrial invasion, as well as adnexal or pelvic nodes in- used [39, 44]. Thus, for patients achieving a complete response
volvement [36]. Patients should be informed that this is a non- at 6 months, conception must be encouraged and these patients
standard approach and they should be willing to accept close should be referred to a fertility clinic.
follow-up during and after the treatment. They should also be For patients with disease recurrence after an initial response, hys-
informed of the need for future hysterectomy in case of failure of terectomy should be proposed as the first option. Moreover, given
the treatment and/or after pregnancies. the high rate of recurrence, after completion of childbearing (or after
Recommendation 2.1: Patients with AH/EIN or grade 1 EEC the age of potential pregnancy), standard treatment with hysterec-
requesting fertility-preserving therapy must be referred to specia- tomy and salpingo-oophorectomy is recommended. Preservation of
lised centres the ovaries can be considered in selected cases, depending on the
Level of evidence: V patient’s age and genetic risk factors.
Strength of recommendation: A Recommendation 2.6: For patients undergoing fertility-preserv-
Consensus: 100% yes (37 voters) ing therapy, MPA (400–600 mg/day) or MA (160–320 mg/day) is
Recommendation 2.2: In these patients, D&C with or without the recommended treatment. However, treatment with LNG-IUD
hysteroscopy must be performed with or without GnRH analogues can also be considered
Level of evidence: IV Level of evidence: IV
Strength of recommendation: A Strength of recommendation: B
Consensus: 97.3% (36) yes, 2.7% (1) abstain (37 voters) Consensus: 100% yes (37 voters)
Recommendation 2.3: AH/EIN or grade 1 EEC must be con- Recommendation 2.7: In order to assess response, D&C, hys-
firmed/diagnosed by a specialist gynaecopathologist teroscopy and imaging at 6 months must be performed. If no
 | Colombo et al. Volume 27 | No. 1 | January 2016

by guest
on 05 February 2018
response is achieved after 6 months, standard surgical treatment Level of evidence: IV
should be performed Strength of recommendation: B
Level of evidence: IV Consensus: 100% yes (37 voters)
Another histological entity that may arise in the differential
diagnosis of AH/EIN is the rare atypical polypoid adenomyoma
Recommendation 2.8: In case of complete response, concep-
(APA), for which there are no IHC stains with practical value.
tion must be encouraged and referral to a fertility clinic is
Recommendation 3.3: IHC is not recommended to distin-
recommended
guish APA from AH/EIN
Level of evidence: IV
Recommendation 2.9: Maintenance treatment should be con-
sidered in responders who wish to delay pregnancy The putative precursor of serous carcinoma, serous endomet-
Level of evidence: IV rial intraepithelial carcinoma (SEIC), is considered a non-invasive
Strength of recommendation: B cancer rather than a precancer since it may be associated with ex-
Consensus: 100% yes (37 voters) tensive extra-uterine disease [9]. Molecular alterations of serous
Recommendation 2.10: Patients not undergoing hysterec- carcinoma are already present in SEIC, which is especially true
tomy should be re-evaluated clinically every 6 months for p53 expression [52–54]. A completely negative immunoreac-
Level of evidence: IV tive pattern for p53 (all or null) is considered a surrogate for p53
Strength of recommendation: B mutation, and is present in almost all SEIC and invasive serous
Consensus: 97.3% (36) yes, 2.7% (1) abstain (37 voters) carcinomas [55].
Recommendation 2.11: After completion of childbearing, a Recommendation 3.4: p53 by IHC is recommended to distin-
hysterectomy and salpingo-oophorectomy should be recom- guish SEIC from its mimics
mended. The preservation of the ovaries can be considered de- Level of evidence: IV
pending on age and genetic risk factors Strength of recommendation: A
Level of evidence: IV Consensus: 100% yes (37 voters)
In selected cases of endometrial cancer, clinical and radiological
work-up may not be conclusive about the endometrial origin of
3. Which (molecular) markers can help distinguish ( pre)can- the uterine tumour. In addition, endocervical, ovarian and endo-
cerous lesions from benign mimics? metrial adenocarcinomas may show histopathological overlap.
Several IHC markers have been proposed for these differential
Differential diagnosis between benign uterine lesions and diagnoses, but these markers lack sensitivity or specificity to be
endometrial ( pre)carcinomas is based mainly on morphological used as single markers. When endocervical origin is considered,
criteria but may be supported by additional immunohistochem- the use of a panel of markers, including carcinoembryonic antigen
ical (IHC) markers and molecular alterations in problematic (CEA), vimentin, oestrogen receptor (ER) and p16 [as surrogate
cases [45]. for human papilloma virus (HPV)], is recommended [56]. In case
Currently, AH/EIN is the preferred terminology of the pre- of p16 positivity, the staining pattern should be taken into account.
cursor lesion of the most common type of endometrial carcin- Diffuse p16 staining is frequently seen in serous, clear-cell and mu-
oma, endometrioid carcinoma, including its variants. cinous carcinoma endometrial cancers [57, 58]. In cases of scanty
Recommendation 3.1: In case of uncertainty, low threshold tissue with serous carcinoma, an ovarian origin of the serous car-
referral to a specialised gynaecopathologist is recommended cinoma should be considered. The most discriminatory marker for
Level of evidence: V this differential diagnosis is Wilms tumour 1 gene (WT-1) [59],
Strength of recommendation: A which is expressed in 80%–100% of high-grade serous carcinomas
Consensus: 100% yes (37 voters) of the ovary [60, 61] compared with 7%–20% in serous endomet-
The differential diagnosis of AH/EIN includes, in particular, rial carcinomas [62, 63]. In general, the expression profile should
endometrial hyperplasia without atypia, but also includes other be interpreted in the context of the morphological subtype. An in-
mimics, such as glandular and stromal breakdown, focal glandular dividual approach, with close correlation between clinical presenta-
crowding and epithelial metaplasias (e.g. hypersecretory changes). tion and morphological subtype, is therefore recommended.
Loss of PTEN expression, mostly by mutation, and loss of PAX-2 Recommendation 3.5: A panel of markers must be used in
by down-regulation [46–48] are the only IHC markers that have cases where endocervical cancer is suspected. This panel should
been sufficiently studied and can be used on curettage material. include at least ER, vimentin, CEA and p16 by IHC and needs
Loss of PTEN occurs in 40%–50% of AH/EIN cases, whereas loss to be assessed in the histologic and clinical context. In addition,
of PAX-2 occurs in 70% of AH/EIN, and a joint loss of PTEN and HPV analysis can be considered
PAX-2 occurs in ∼30% of AH/EIN [49–51]. Level of evidence: IV
Recommendation 3.2: PTEN and PAX-2 IHC is recom- Strength of recommendation: B
mended to distinguish AH/EIN from benign mimics. Other Consensus: 100% yes (37 voters)
markers that can be used in this context are MLH1 and ARID1a Recommendation 3.6: WT-1 by IHC is the recommended
by IHC marker to determine the origin of serous cancer
Volume 27 | No. 1 | January 2016 doi:10.1093/annonc/mdv484 | 

by guest
on 05 February 2018
Level of evidence: IV (of endometrioid tumours; grade 1–3 or a binary system) and
Strength of recommendation: A histotype (endometrioid versus non-endometrioid tumour).
Consensus: 100% yes (37 voters) Recommendation 4.1: Mandatory work-up must include:
Recommendation 3.7: Morphology (and not IHC) should be Family history; general assessment and inventory of comorbid-
used to distinguish AH/EIN from EEC ities; geriatric assessment, if appropriate; clinical examination,
Level of evidence: IV including pelvic examination; transvaginal or transrectal ultra-
Strength of recommendation: A sound; and complete pathology assessment (histotype and
Consensus: 100% yes (37 voters) grade) of an endometrial biopsy or curettage specimen
surgery
4. How does the medical condition influence surgical treat- Recommendation 4.2: Extent of surgery should be adapted to
ment? the medical condition of the patient
mandatory preoperative work-up. The consensus is based on Strength of recommendation: A
current clinical practice. Family history is usually taken to Consensus: 100% yes (37 voters)
identify risk factors associated with LS, including endometrial
cancer, colon cancer and other cancers belonging to the Lynch optional preoperative work-up.
spectrum. General assessment and, if appropriate, geriatric imaging: Additional imaging is considered according to the
assessment are required in patients with comorbidities and clinical situation. Computed tomography (CT) scan and/or
elderly patients, respectively, in order to adapt the surgical positron emission tomography (PET)-CT are options in clinically
strategy. Indeed, endometrial cancer is frequently associated advanced endometrial cancer. In apparent stage I endometrial
with obesity, hypertension and diabetes and, in some patients, cancer, MRI may be useful to complete information regarding
the extent of surgery or staging that is theoretically required myometrial invasion [65]. However, this applies only in institutions
may not be feasible. In such cases, a benefit–risk assessment of where the indication for lymph node dissection (LND) is tailored
surgery may lead to an individualised decision to perform a according to the stratification of patients into low-, intermediate-
‘non-standard’ surgery or a limited staging procedure. and high-risk groups. In this setting, specialised ultrasonography
Pelvic examination and pelvic ultrasonography are manda- and/or intra-operative pathological examination of the uterus may
tory components of clinical staging of endometrial cancer in also be considered [68].
order to establish a tentative International Federation of Recommendation 4.3: In clinical stage I, grade 1 and 2: At
Gynecology and Obstetrics (FIGO) staging before definitive least one of the three following tools should be used to assess
pathology. In addition to being the first imaging technique myometrial invasion if LND is considered: Expert ultrasound
used to evaluate abnormal uterine bleeding, ultrasonography, and/or MRI and/or intra-operative pathological examination
preferably specialised ultrasonography [64], offers the possibil- Level of evidence: IV
ity of evaluating the size of the tumour, ruling out ovarian Strength of recommendation: A
disease and assessing myometrial invasion and cervical stromal Consensus: 100% yes (37 voters)
involvement [65]. Recommendation 4.4: Other imaging methods (thoracic, ab-
Preoperative pathological information is crucial for establishing dominal and pelvic CT scan, MRI, PET scan or ultrasound)
the surgical plan. First, all patients with a risk of cancer, particu- should be considered to assess ovarian, nodal, peritoneal or
larly patients with postmenopausal bleeding and a hyperplastic metastatic disease
endometrium at ultrasound, should be investigated with endo- Level of evidence: IV
metrial biopsy or curettage in order to (i) avoid uterine morcella- Strength of recommendation: C
tion, which poses a risk of spreading unsuspected cancerous Consensus: 94.6% (35) yes, 2.7% (1) abstain, 2.7% (1) no
tissue, notably endometrial carcinomas or sarcomas, beyond the (37 voters)
uterus and may make the pathological assessment of myometrial
invasion extremely difficult; and (ii) prevent the discovery of an serum tumour markers: There is evidence that the serum
unexpected malignancy after inadequate surgery (subtotal hyster- tumour markers cancer antigen 125 (CA 125) and, more recently,
ectomy and/or preservation of the ovaries in a postmenopausal human epididymis protein 4 are significantly correlated with
patient, incomplete staging). Secondly, as grading of EEC has a histological grade, stage, lymph node metastases, myometrial
significant prognostic impact [66] and various histotypes of invasion and cervical involvement [69–71]. However, the
endometrial cancer harbour different natural histories, the appropriate cut-off has not been established and evidence that
primary therapeutic strategy must be adapted to the information serum marker assessment is clinically useful is lacking.
provided by a preoperative pathological examination, despite the Recommendation 4.5: There is no evidence for the clinical
fact that discrepancies between preoperative evaluation and final usefulness of serum tumour markers, including CA 125
pathology exist [67]. Level of evidence: IV
The final therapeutic strategy should be adapted according to Strength of recommendation: B
the information available before surgery, taking into account the Consensus: 91.9% (34) yes, 5.4% (2) abstain, 2.7% (1) no
tentative stage (apparent stage I or more advanced stage), grade (37 voters)
 | Colombo et al. Volume 27 | No. 1 | January 2016

by guest
on 05 February 2018
surgical management of apparent stage I endometrial cancer. patients in the laparoscopy and laparotomy groups, respectively
With the exception of patients managed conservatively, extra- (P < 0.0001), there was no difference in the overall detection
fascial total hysterectomy without colpectomy is the mainstay of of advanced stage disease between the two groups. The major
management for patients with endometrial cancer. The rationale shortcoming of this trial is the high conversion rate related to its
for the additional removal of the adnexae is to prevent ovarian multicentric design. Indeed, 25.8% of patients assigned to the lap-
cancer and rule out ovarian metastases. In premenopausal patients, aroscopic group were converted to laparotomy, with a statement of
however, ovarian preservation may be discussed in selected cases. ‘poor visibility’ reported in 14.6% of cases, reflecting the learning
Younger patients with endometrial cancer often have early-stage, curve of some investigators, particularly for LND. In contrast, a
low-grade tumours. Thus, to avoid the short-term and long-term conversion rate of 10.8%, with poor visibility recorded as the main
consequences of surgical menopause, there is a rationale for factor in 4.9% of cases, was reported in a Dutch randomised trial
ovarian preservation in young women. Several retrospective studies in which no lymphadenectomy was performed [74]. However, as
have recently provided evidence that ovarian preservation has no further training or the use of robotic assistance would likely have
statistically significant impact on the overall survival (OS) of young resulted in even better results with laparoscopic surgery, this high
patients with early-stage endometrial cancer [72]. However, conversion rate reported in LAP2 does not weaken the authors’
extreme care must be taken to rule out synchronous concomitant conclusions, and this trial provides evidence that laparoscopic sur-
ovarian malignancy. gical staging for uterine cancer results in fewer complications and
Recommendation 4.6: Standard surgery is total hysterectomy shorter hospital stay.
with bilateral salpingo-oophorectomy without vaginal cuff According to a meta-analysis of data from eight randomised,
Level of evidence: IV controlled trials (RCTs) conducted by Zullo et al. [75], intra-opera-
Strength of recommendation: A tive complication rates were not different between laparoscopy and
Consensus: 100% yes (37 voters) laparotomy (RR 1.25; 95% CI 0.99–1.56) with no significant het-
Recommendation 4.7: Ovarian preservation can be consid- erogeneity across the studies. Estimated blood loss and haemoglo-
ered in patients younger than 45 years old with grade 1 EEC bin or haematocrit changes were consistently less after laparoscopy
with myometrial invasion <50% and no obvious ovarian or in the six studies where this was reported. Operative time was
other extra-uterine disease higher by 34–74 min in the laparoscopy group. The authors also
Level of evidence: IV found a significant advantage of laparoscopy over laparotomy in
Strength of recommendation: B terms of postoperative complications (RR 0.71; 95% CI 0.63–0.79)
Consensus: 100% yes (37 voters) with significant heterogeneity across the studies.
Recommendation 4.8: In cases of ovarian preservation, sal- Aortic dissection can also be achieved in obese patients using
pingectomy is recommended an extra-peritoneal laparoscopic approach [76].
Level of evidence: IV Taken together, these findings provide definitive evidence of
Strength of recommendation: B the short-term benefit and cost-effectiveness of laparoscopic hys-
Consensus: 100% yes (37 voters) terectomy in patients with gynaecological cancer. This includes
Recommendation 4.9: Ovarian preservation is not recom- patients with comorbidities, obesity or advanced age. Regarding
mended for patients with cancer family history involving comorbidity, Tozzi et al. [77] found that the surgical technique is
ovarian cancer risk (e.g. BRCA mutation, LS etc.). Genetic coun- the only significant parameter associated with complication rate,
selling/testing should be offered regardless of risk group, stressing the fact that patients with
Level of evidence: IV serious comorbidities benefit most from laparoscopy. The issue of
Strength of recommendation: B advanced age has also been addressed in the gynaecological on-
Consensus: 100% yes (37 voters) cology literature. Siesto et al. [78] reported outcomes from a
series of 48 patients aged >65 years who had undergone laparo-
minimally invasive surgical techniques. Hysterectomy and scopic surgery for endometrial cancer. Outcomes from this group
bilateral salpingo-oophorectomy can be carried out using the were comparable with younger patients in terms of operative
open, laparoscopic or vaginal approach. time, blood loss, need for blood transfusions, nodal count and
The largest randomised trial comparing laparoscopy with lapar- intra-operative and postoperative complications. The authors
otomy is the LAP2 study [73], which was designed to compare conclude that, in the absence of absolute anaesthesia contraindi-
laparoscopy versus laparotomy for comprehensive surgical staging cations, laparoscopy is feasible and safe in older women with
and management of stage I–IIA uterine cancer, including hysterec- endometrial cancer. However, as cancer in older women was
tomy, salpingo-oophorectomy, pelvic cytology and pelvic and more frequently upstaged than in younger women, they state that
para-aortic lymphadenectomy. In this trial, patients were randomly comprehensive surgical staging should be offered, regardless of
assigned to laparoscopy (n = 1696) or open laparotomy (n = 920). age, to avoid under-staging and to optimise treatment strategies.
A significantly longer operative time was reported for the laparos- Six randomised trials comparing outcomes after laparotomy
copy group compared with the laparotomy group (204 versus 130 with laparoscopy are currently available, four of which have
min, respectively). Intra-operative complication rates were similar been included in a published meta-analysis [79]. However, only
between groups. However, laparoscopy was associated with signifi- two of these four trials reported data for OS, disease-free sur-
cantly fewer moderate-to-severe postoperative adverse events (14% vival and cancer-related survival. Based on the availability of
versus 21%) and a lower frequency of hospitalisations of more new data, this meta-analysis was subsequently updated by
than 2 days (52% versus 94%) than laparotomy. Although pelvic Palomba et al. in 2009 [80] to include a third trial reporting
and para-aortic lymph nodes were not removed in 8% and 4% of these long-term outcomes, resulting in a sample of 359 patients.
Volume 27 | No. 1 | January 2016 doi:10.1093/annonc/mdv484 | 

by guest
on 05 February 2018
No significant heterogeneity was observed among these trials, 5. What are the indications for and to what extent is lympha-
and there was no significant adverse effect of a laparoscopic ap- denectomy indicated in the surgical management of endo-
proach on the OS, disease-free survival or cancer-related sur- metrial cancer?
vival (OR 0.96, 0.95 and 0.91, respectively).
Long-term outcomes of the randomised, controlled LAP2 trial
were published in 2012 [81]. The primary end point was non-in- surgical staging in apparent stage I EEC. Collection of
feriority of the recurrence-free interval. Non-inferiority was peritoneal cytology was included as a staging procedure in earlier
defined as no more than a 40% increase in the risk of recurrence recommendations, but it is no longer considered mandatory.
with laparoscopy compared with laparotomy. The estimated However, since retrospective studies indicate that positive
hazard ratio (HR) for recurrence-free survival with laparoscopy peritoneal cytology has prognostic value, collection of this
versus laparotomy was 1.14 (90% CI 0.92–1.46). Actual recur- information could be considered, especially in patients with
rence rates were substantially lower than anticipated; the esti- tumours of non-endometrioid histology [83, 84].
mated 3-year recurrence rate was 11.4% with laparoscopy and Recommendation 5.1: Peritoneal cytology is no longer con-
10.2% with laparotomy, and the estimated 5-year OS was almost sidered mandatory for staging
identical in both arms (89.8%). Level of evidence: IV
Recommendation 4.10: Minimally invasive surgery is recom- Strength of recommendation: A
mended in the surgical management of low- and intermediate- Consensus: 100% yes (37 voters)
risk endometrial cancer
Level of evidence: I lymphadenectomy. Lymphadenectomy is an integral part of the
Strength of recommendation: A comprehensive surgical staging of endometrial cancer. However,
Consensus: 100% yes (37 voters) the role of lymphadenectomy in early endometrial cancer is unclear
and controversy remains regarding the indications for, the anatomic
In a retrospective, multi-institutional trial of patients with
extent of, and the therapeutic value of lymphadenectomy in the
high-grade endometrial cancer, outcomes of 191 patients who
management of the disease.
underwent laparotomy were compared with 192 patients
The definition of an adequate lymphadenectomy has not
who underwent minimal invasive surgery. In this trial, women
been standardised: current approaches include pelvic lymphade-
with high-grade endometrial cancer staged by minimally invasive
nectomy, para-aortic lymphadenectomy to the inferior mesen-
techniques experienced fewer complications and similar survival
teric artery (IMA) and para-aortic lymphadenectomy up to the
outcomes compared with those staged by laparotomy [82].
renal vessels. Lymph node counts have become a marker for ad-
Recommendation 4.11: Minimally invasive surgery can be
equacy of lymph node evaluation in a variety of solid tumour
considered in the management of high-risk endometrial cancer
disease sites. In endometrial cancer, two retrospective reviews
have shown that patients had improved survival when at least
10–12 lymph nodes were removed during lymphadenectomy
[85, 86]. Lymph node counts therefore provide a surrogate way
alternative approaches for patients unsuitable for standard of measuring the adequacy of a LND and, as such, more than 10
surgical therapy. Although advances in surgical techniques, nodes should be removed [87, 88].
anaesthesiology and perioperative management mean that the vast Sampling of lymph nodes has a low sensitivity in endometrial
majority of patients with endometrial cancer are amenable cancer [89]. Indeed, it has been shown that para-aortic nodes
to standard surgical therapy, a small proportion of patients are may be positive in the absence of positive pelvic nodes [90, 91],
still medically unfit for laparoscopic surgery or laparotomy. suggesting that para-aortic lymph nodes should be removed in
However, these patients can still be managed either surgically cases where a lymphadenectomy is indicated. In the Mayo
by vaginal hysterectomy, whenever possible, with bilateral Clinic experience of 281 patients with endometrial cancer who
salpingo-oophorectomy, or by definitive RT, combining external underwent lymphadenectomy, 22% of patients with high-risk
beam radiation therapy (EBRT) and brachytherapy, or by disease had lymph node metastases: 51% had both positive
hormonal treatment. In addition, vaginal hysterectomy is an pelvic and para-aortic nodes, 33% had positive pelvic lymph
acceptable minimally invasive surgical option in some low-risk nodes only and 16% had isolated para-aortic lymphadenopathy
patients who do not need LND (see section 4). [92]. As the majority (77%) of patients with para-aortic lymph
Recommendation 4.12: Vaginal hysterectomy with salpingo- node involvement had metastases above the IMA, para-aortic
oophorectomy can be considered in patients unfit for the lymphadenectomy up to the renal vessels is recommended.
recommended surgery and in selected patients with low-risk The concept of sentinel lymph node (SLN) dissection (SLND)
endometrial cancer was first developed in cervical cancer as a tool to select patients
Level of evidence: IV most suitable for surgical management. In low- and intermedi-
Strength of recommendation: C ate-risk endometrial cancer, the rationale is different as the need
Consensus: 100% yes (37 voters) for SLND is controversial. However, SLND could represent a
Recommendation 4.13: In medically unfit patients, RT or compromise between no dissection (leaving a small proportion
hormone treatment can be considered of node-positive patients) and full dissection (adding a useless
Level of evidence: IV procedure for the majority of node-negative patients). In add-
Strength of recommendation: C ition, ultra-staging of the SLNs detects micrometastases other-
Consensus: 100% yes (37 voters) wise undiagnosed by conventional histology, even in patients
 | Colombo et al. Volume 27 | No. 1 | January 2016

by guest
on 05 February 2018
considered at low risk, on the basis of grade and depth myome- the average frequency of LND was 31%, 40%, 47% and 53% for
trial invasion [93]. However, these large series only use the the years 1988–1991, 1992–1995, 1996–1999 and 2000–2003, re-
cervix as the injection site. The question of alternative injection spectively (P < 0.0001) [98]. On multivariate analysis, the pres-
sites in the endometrium or uterine fundus, which are anatom- ence of LND was associated with OS and uterine-specific
ically more logical, is still a topic for investigation. Injection survival benefits with HRs of 0.81 (P < 0.0001) and 0.78
under hysteroscopic, ultrasound, laparoscopic or open guidance (P < 0.0001), respectively, and removal of >11 lymph nodes was
in patients with endometrial cancer has been addressed, without associated with HRs of 0.74 (P < 0.0001) and 0.69 (P < 0.0001),
evidence of benefit of the more demanding and less practical mo- respectively. On the basis of these findings, the authors con-
dalities. Nevertheless, evidence is accumulating that the SLND cluded that the presence of LND and increased number of
may be useful in the management of endometrial cancers [94]. nodes dissected predicted for improved OS and uterine-specific
Recommendation 5.2: If a lymphadenectomy is performed, survival in women with adenocarcinoma of the endometrium.
systematic removal of pelvic and para-aortic nodes up to the Retrospective single-institution studies advocate lymphade-
level of the renal veins should be considered nectomy for all grades of tumour [87, 88]. In contrast, a series
Level of evidence: IV using a US database supports lymphadenectomy for high-grade
Strength of recommendation: B tumours only [99]. This was confirmed by the SEPAL trial in a
Consensus: 91.9% (34) yes, 2.7% (1) abstain, 5.4% (2) no series of intermediate- or high-risk patients with pelvic lympha-
(37 voters) denectomy with or without para-aortic LND [100]. Patients
Recommendation 5.3: SLND is still experimental, but large who underwent para-aortic lymphadenectomy had a superior
series suggest that it is feasible. SLND increases the detection of survival compared with those who did not.
lymph nodes with small metastases and isolated tumour cells; In addition to risk factors, the number of lymph nodes
however, the importance of these findings is unclear removed also seems to be important, with higher node count
Level of evidence: IV associated with improved survival [85, 101]. Kim et al. recently
Strength of recommendation: D analysed data from nine trials (two RCTs and seven observa-
Consensus: 100% yes (37 voters) tional studies) involving 16 995 patients with endometrial
cancer and showed that the efficacy of systematic lymphadenect-
indications for lymphadenectomy. Although the therapeutic omy, defined as removal of ≥10–11 lymph nodes, was associated
effect of lymphadenectomy is unclear, it is an integral part of with limited survival benefit in patients with low-risk endomet-
comprehensive staging. The advantages of comprehensive surgical rial cancer, but resulted in improved OS in patients with inter-
staging are a better definition of prognosis and appropriate triage mediate- or high-risk endometrial cancer [102]. However,
of patients for adjuvant therapy. patients with low-risk disease (i.e. grade 1 and 2 endometrioid
Data from two RCTs do not support the therapeutic benefit lesions with <50% myometrial invasion) have a very low prob-
of lymphadenectomy in early-stage endometrial cancer. ability of lymphadenopathy and therefore derive no benefit
Benedetti Panici et al. randomised 514 women with clinical from a systematic lymphadenectomy [103].
stage I endometrial cancer to either systematic pelvic lymphade- Recommendation 5.4: Lymphadenectomy is a staging pro-
nectomy or no LND and found no improvement in disease-free cedure and allows tailoring of adjuvant therapy
survival or OS between the two groups [95]. Similarly, the Level of evidence: III
ASTEC trial, which included 1408 patients with stage I endo- Strength of recommendation: B
metrial cancer who were randomised to receive surgical staging Consensus: 100% yes (37 voters)
with or without pelvic lymphadenectomy, failed to show a bene- Recommendation 5.5: Patients with low-risk endometrioid
ficial effect of lymphadenectomy [96]. Although these trials re- carcinoma (grade 1 or 2 and superficial myometrial invasion
present the best data available, controversy still exists, partly due <50%) have a low risk of lymph node involvement, and two
to criticisms of the ASTEC trial, in which the number of lymph RCTs did not show a survival benefit. Therefore, lymphadenect-
nodes removed was low and systematic para-aortic lymphade- omy is not recommended for these patients
nectomy was not performed. A mathematical model applied to Level of evidence: II
the ASTEC trial suggested a survival difference of <2% between Strength of recommendation: A
the experimental and control arms under all circumstances [97]. Consensus: 100% yes (37 voters)
This model suggested that, even if LND was therapeutic, this Recommendation 5.6: For patients with intermediate risk
trial would have been negative due to the trial design. In the (deep myometrial invasion >50% or grade 3 superficial myome-
Italian trial [95], median node counts were 26, or 30 for the 26% trial invasion <50%), data have not shown a survival benefit.
of patients who also had para-aortic lymphadenectomy, and Lymphadenectomy can be considered for staging purposes in
there were no differences in relapse rates, disease-free survival these patients
and OS. Level of evidence: II
In contrast, retrospective data, which are prone to selection Strength of recommendation: C
bias and stage migration, suggest that patients who underwent Consensus: 100% yes (37 voters)
systematic lymphadenectomy had improved survival over those Recommendation 5.7: For patients with high risk (grade 3
who had limited or no sampling performed [88]. Data from with deep myometrial invasion >50%), lymphadenectomy
42 184 patients with endometrial cancer, obtained from the should be recommended
Surveillance, Epidemiology and End Results Program of the US Level of evidence: IV
National Cancer Institute for the years 1988–2003, showed that Strength of recommendation: B
Volume 27 | No. 1 | January 2016 doi:10.1093/annonc/mdv484 | 

by guest
on 05 February 2018
Consensus: 73.0% (27) yes, 8.1% (3) abstain, 18.9% (7) no Recommendation 6.5: Multimodality management should be
(37 voters) considered for the treatment of advanced endometrial cancer
Recommendation 5.8: Lymphadenectomy to complete when surgery may significantly impair vaginal function
staging could be considered in previously incompletely operated Level of evidence: IV
high-risk patients to tailor adjuvant therapy Strength of recommendation: B
Level of evidence: V Consensus: 97.3% (36) yes, 2.7% (1) abstain (37 voters)
Consensus: 100% yes (37 voters) surgical management of non-EEC. The standard of surgical
therapy in non-EEC is not different from EEC (see sections 3 and
6. How radical should the surgery be in different stages and 5). Hysterectomy and bilateral salpingo-oophorectomy is the
pathological subtypes of endometrial cancer? mainstay of therapy in apparent stage I disease. Radical
hysterectomy is not recommended in stage II disease, whereas
complete cytoreduction is required in advanced disease stages.
surgical management of stage II–IV endometrial cancer. In a
However, there is no documentation on ovarian preservation.
recent study from Japan, radical surgery in stage II endometrial
Bilateral salpingo-oophorectomy is mandatory.
cancer did not result in any survival benefit compared with simple
Comprehensive surgical staging of more advanced disease
hysterectomy but was associated with more perioperative and late
stages is mandatory (see section 5). Although no data from rando-
adverse events [104]. Another recent study found that parametrial
mised trials are available in non-EEC, the staging of apparent
spread cannot be predicted by cervical involvement alone but
stage I disease are similar to high-risk EEC. Omentectomy is also
may be predicted by various lymphovascular space invasion
considered in apparent stage I papillary serous carcinoma, in
(LVSI)-related histopathological factors [105]. However, radical
which peritoneal implants are not uncommon. However, omen-
hysterectomy is considered in cases of obvious involvement of the
tectomy is not mandatory in cases of clear-cell carcinoma [108],
parametrium. Surgery should then be tailored according to the
but should be considered where there is a serous component since
recent classification of radical hysterectomy [106] in order to
uterine serosal spread has a negative impact on survival [109].
obtain free margins. Lymphadenectomy is recommended.
Recommendation 6.6: In non-EEC (apparent stage I), lym-
Recommendation 6.1: Radical hysterectomy is not recom-
phadenectomy is recommended
mended for the management of stage II endometrial cancer
Consensus: 91.9% (34) yes, 8.1% (3) abstain (37 voters)
Recommendation 6.7: Staging omentectomy is not manda-
Recommendation 6.2: Modified (type B) or type A radical
tory in clear-cell or undifferentiated endometrial carcinoma and
hysterectomy should be considered only if required for obtain-
carcinosarcoma
ing free margins
Recommendation 6.8: Staging omentectomy should be con-
Recommendation 6.3: Lymphadenectomy is recommended
sidered in serous carcinoma
for clinical or intra-operative stage II endometrial cancer
adjuvant treatment
surgical management of stage III–IV endometrial cancer. 7. What is the current best definition of risk groups for
Although there is no evidence from randomised trials for stage adjuvant therapy?
III–IV endometrial cancer, there is consensus that multimodality
therapy is required, generally starting with radical cytoreductive The majority of patients with endometrial cancer have a low
surgery. Several retrospective studies have shown a statistically risk of recurrence and are managed by surgery alone [110]. Risk
significant advantage in progression-free survival (PFS) and OS groups have been devised based on clinicopathological prognos-
when optimal cytoreduction can be achieved [107]. However, not tic factors to identify patients at risk of recurrence who may
all patients are amenable to optimal cytoreduction as a result of benefit from adjuvant therapy.
poor general condition or tumour extent. In addition, the surgical In order to have clinical value, a definition of risk groups
management of metastatic vaginal disease may impair the vaginal should have both prognostic value and consequences for the in-
function. Primary RT is therefore preferable in some cases. dication of adjuvant therapy. Well-defined clinicopathological
Recommendation 6.4: Complete macroscopic cytoreduction prognostic factors include: age, FIGO stage, depth of myometrial
and comprehensive staging is recommended in advanced endo- invasion, tumour differentiation grade, tumour type (endome-
metrial cancer trioid versus serous and clear cell) and LVSI [89]. Compared
Level of evidence: IV with the ESMO risk group classification [8], the adverse prog-
Strength of recommendation: A nostic role of both LVSI and tumour grade 3 within the inter-
Consensus: 100% yes (37 voters) mediate-risk group (stage IA grade 3 or stage IB grade 1–2) has
 | Colombo et al. Volume 27 | No. 1 | January 2016

by guest
on 05 February 2018
Table 2. New risk groups to guide adjuvant therapy use of routine lymphadenectomy for nodal staging purposes in
low- and intermediate-risk endometrial cancer [95, 96]. Given
Risk group Description LOE
the absence of a survival benefit and its associated side-effects,
Low Stage I endometrioid, grade 1–2, <50% I routine lymphadenectomy is not recommended for low- and
myometrial invasion, LVSI negative (high)intermediate-risk disease in most national and internation-
Intermediate Stage I endometrioid, grade 1–2, ≥50% I al guidelines for these patients. The value of lymphadenectomy in
myometrial invasion, LVSI negative high-risk endometrial cancer is the subject of ongoing investiga-
High- Stage I endometrioid, grade 3, <50% I tions. Recommendations regarding what defines adequate (lymph
intermediate myometrial invasion, regardless of LVSI
node) staging are detailed in the chapter on surgery.
status
Compared with the ESMO CPG on endometrial cancer [8],
Stage I endometrioid, grade 1–2, LVSI II
the current recommendations are specified to address both
unequivocally positive, regardless of depth
scenarios that surgical nodal staging is performed and is not
of invasion
performed, and to specifically address non-endometrioid histo-
High Stage I endometrioid, grade 3, ≥50% I
myometrial invasion, regardless of LVSI
logical subtypes. In addition, the roles of vaginal brachytherapy,
status EBRT and chemotherapy or combinations of these treatments
Stage II I have been specified in more detail for each of these situations.
Stage III endometrioid, no residual disease I
Non-endometrioid (serous or clear-cell or I low-risk endometrial cancer. Some patients now considered as
undifferentiated carcinoma, or low risk were included in the large randomised trials of adjuvant
carcinosarcoma) RT, and no benefit of RT was found in this subgroup [116–119].
Advanced Stage III residual disease and stage IVA I A randomised trial of 645 patients with low-risk endometrial
Metastatic Stage IVB I cancer treated with vaginal brachytherapy also showed no
advantage for the use of adjuvant brachytherapy, likely because
FIGO 2009 staging used; molecular factors were considered but not the risk of recurrence after surgery alone is <5% [125].
included; tumour size was considered but not included; nodal status may Therefore, no adjuvant treatment is indicated for patients with
be considered for treatment recommendations. low-risk endometrial cancer.
LOE, level of evidence; LVSI, lymphovascular space invasion. Recommendation 8.1: In patients with low-risk endometrial
cancer (stage I endometrioid, grade 1–2, <50% myometrial inva-
sion, LVSI negative), no adjuvant treatment is recommended
been recognised [111–115]. This has led to a new subdivision of Level of evidence: I
low risk, intermediate risk and high-intermediate risk in the Strength of recommendation: A
current classification, which is different from the risk classifica- Consensus: 100% yes (37 voters)
tion used in many clinical trials. Historically, low-risk endomet-
rial cancer was defined as endometrioid adenocarcinoma FIGO intermediate-risk endometrial cancer. Patients considered
stage I and grade 1 with superficial invasion or grade 2 without intermediate risk in the current classification were included in
invasion, and high-risk as stage I, grade 3 with deep myometrial the large randomised trials evaluating the role of adjuvant RT in
invasion, with other combinations of grade and invasion early-stage endometrial cancer [116–119]. In these trials, patients
defined as intermediate risk. Against this background, the large were randomised after total hysterectomy with bilateral salping-
trials evaluating the role of RT for intermediate-risk endometrial oophorectomy to pelvic EBRT or observation after surgery. All
cancer (PORTEC-1, GOG99, ASTEC/EN5, described below three trials and a meta-analysis by Kong et al. [119] found that
[116–118]) were conducted and, based on the results of these EBRT reduced the risk of pelvic recurrence by threefold (from
trials and a subsequent meta-analysis [119], a refined classifica- 14% to 4%), but did not lead to an OS benefit and came at the
tion of low risk, intermediate risk and high-intermediate risk cost of increased risk of (predominantly gastrointestinal) toxicity.
has been introduced. In contrast to the PORTEC-1 trial [117], surgical staging lym-
Factors such as tumour size and several molecular factors (e.g. phadenectomy was mandatory in the GOG99 trial [118],
TP53, L1CAM) have been reported as having prognostic value in showing that for node-negative disease, EBRT still reduced the
observational studies but have not been incorporated into this risk of recurrence. This risk reduction was mainly caused by pre-
classification since they are still under investigation and currently vention of local (vaginal) recurrence. Both PORTEC-1 and
not in clinical use [120–124]. A definition of risk groups to iden- GOG99 defined a subgroup of patients who derived the greatest
tify patients at risk of recurrence who may benefit from adjuvant benefit of adjuvant EBRT, a so-called high-intermediate-risk
therapy has been devised by the consensus panel and is shown in group. In the PORTEC-1 trial, the definition of risk groups was
Table 2. based on risk factors for locoregional recurrence [age >60 years,
deep (≥50%) myometrial invasion, grade 3], with high-inter-
8. What are the best evidence-based adjuvant treatment strat- mediate-risk patients defined as having two of three of these risk
egies for patients with low- and intermediate-risk endo- factors. In this subgroup, the 5-year risk of locoregional recur-
metrial cancer? rence was 20% for observation versus 5% for adjuvant RT, and
only in this subgroup was the risk of relapse deemed high
Although the 1988 FIGO staging system included surgical enough to consider adjuvant RT [117]. In the GOG99 trial, the
staging, two large randomised trials have since found no benefit definition of risk groups was based on risk factors for overall
Volume 27 | No. 1 | January 2016 doi:10.1093/annonc/mdv484 | 

by guest
on 05 February 2018
recurrence identified in previous Gynecologic Oncology Group In the GOG249 study, both high-intermediate- and high-risk
(GOG) studies, with high-intermediate-risk patients defined as: patients were randomised between pelvic EBRT and vaginal
age <50 years and one risk factor, age 50–70 years and two risk brachytherapy followed by chemotherapy (three cycles of carbo-
factors and age >70 and all three risk factors. Similar results platin and paclitaxel). Results have been presented (abstract
were found in the ASTEC trial, which reported a lower risk of only) that showed no PFS benefit of adjuvant chemotherapy
vaginal and pelvic relapse in the no-EBRT group (7% versus 4% over the standard EBRT [131].
in the EBRT arm). In the ASTEC trial, vaginal brachytherapy Recommendation 8.3: In patients with high-intermediate-
was allowed in both study arms, and more than 50% of patients risk endometrial cancer (stage I endometrioid, grade 3, <50%
in the observation arm received vaginal brachytherapy. myometrial invasion, regardless of LVSI status; or stage I endo-
The randomised PORTEC-2 trial included only patients with the metrioid, grade 1–2, LVSI unequivocally positive, regardless of
high-intermediate-risk factors defined in PORTEC-1, and showed depth of invasion):
that vaginal brachytherapy provided excellent vaginal control com- 1: Surgical nodal staging performed, node negative:
pared with EBRT, and had a more favourable toxicity and quality- A. Adjuvant brachytherapy is recommended to decrease
of-life profile [126]. These results have been confirmed in a Swedish vaginal recurrence
trial in which vaginal brachytherapy was compared with combined Level of evidence: III
EBRT and a vaginal brachytherapy boost [127]. Strength of recommendation: B
Multiple cohort studies have identified LVSI and grade 3 as B. No adjuvant therapy is an option
risk factors for disease recurrence [111–115]. This finding was Level of evidence: III
confirmed in a recent pooled analysis of data from the PORTEC- Strength of recommendation: C
1 and -2 trials, which showed that both LVSI and grade 3 are risk Consensus: 100% yes (37 voters)
factors for regional nodal recurrence and for distant metastasis 2: No surgical nodal staging:
[128]. EBRT decreased the risk of regional nodal recurrence in A. Adjuvant EBRT recommended for LVSI unequivocally
this small subgroup (5%) of patients, while vaginal brachytherapy positive to decrease pelvic recurrence
did not. As the vast majority of patients in PORTEC-2 had grade Level of evidence: III
1–2 tumours with deep (≥50%) myometrial invasion and Strength of recommendation: B
without LVSI, this population is now considered intermediate B. Adjuvant brachytherapy alone is recommended for grade
risk in the current consensus classification. These patients have a 3 and LVSI negative to decrease vaginal recurrence
low risk of regional and distant recurrence, while their risk of Level of evidence: III
local (vaginal) recurrence is significantly decreased with adjuvant Strength of recommendation: B
vaginal brachytherapy. In addition, others have validated the Consensus: 100% yes (37 voters)
added prognostic value of the incorporation of LVSI in the 3: Systemic therapy is of uncertain benefit; clinical studies are
ESMO risk classification [129]. encouraged
Because adjuvant RT does not improve OS and combined Level of evidence: III
EBRT and brachytherapy for recurrent disease is associated with Strength of recommendation: C
a high chance of complete remission, not performing routine Consensus: 94.6% (35) yes, 5.4% (2) abstain (37 voters)
adjuvant RT is also an option [130]. However, combined EBRT
and brachytherapy for recurrent disease is associated with a
higher rate of side-effects compared with adjuvant vaginal 9. What are the best evidence-based adjuvant
brachytherapy alone. treatment strategies for patients with high-risk
Recommendation 8.2: In patients with intermediate-risk endometrial cancer?
endometrial cancer (stage I endometrioid, grade 1–2, ≥50% In general, high-risk endometrial cancer is characterised by
myometrial invasion, LVSI negative): an increased risk of pelvic recurrence and distant metastases
1: Adjuvant brachytherapy is recommended to decrease vaginal that contribute to the inferior outcomes of this group.
recurrence However, high-risk endometrial cancer represents a heteroge-
Level of evidence: I neous group of patients, including both endometrioid and
Strength of recommendation: B non-endometroid tumour types such as serous and clear cell,
2: No adjuvant treatment is an option, especially for patients and ranges from stage IB grade 3 (with or without LVSI and
aged <60 years with or without nodal staging) to more advanced FIGO
Level of evidence: II stages. Regardless of tumour type, the estimated 5-year OS
Strength of recommendation: C according to the 26th FIGO annual report is 85%–90% for
Consensus: 100% yes (37 voters) stage I, 75%–85% for stage II, 50%–65% for stage III and
20%–25% for stage IV [132]. Among FIGO stage I patients,
high-intermediate-risk endometrial cancer. Patients with grade those with deep myometrial invasion and grade 3 histology
1–2 tumours with deep (≥50%) myometrial invasion and are at increased risk of pelvic and distant relapse [133–135].
unequivocally positive (substantial, not focal) LVSI, and those Estimated 5-year OS rates in patients with ≥50% myometrial
with grade 3 tumours with <50% myometrial invasion regardless invasion and grade 3 tumours (without nodal staging) were
of LVSI status are referred to as high-intermediate risk in the only 58%. Regarding non-endometrioid tumour types,
current classification. ∼60%–70% of patients with uterine serous cancer have
 | Colombo et al. Volume 27 | No. 1 | January 2016

by guest
on 05 February 2018
disease outside the uterus at the time of presentation. The 1: Surgical nodal staging performed, node negative:
5-year OS rate for patients with uterine serous cancer is A. Adjuvant EBRT with limited fields should be considered
20%–25% versus 80% for all patients with endometrial cancer to decrease locoregional recurrence
[136]. For these reasons, recommendations were made for the Level of evidence: I
following subgroups: endometrioid stage I, grade 3 and >50% Strength of recommendation: B
myometrial invasion; endometrioid stage II; endometrioid B. Adjuvant brachytherapy may be considered as an alterna-
stage III without residual disease and non-endometrioid tive to decrease vaginal recurrence
tumour types. Recommendations for patients with advanced Level of evidence: III
non-resectable or residual disease are provided separately in Strength of recommendation: B
the ‘Advanced and Recurrent Endometrial Cancer’ section of C. Adjuvant systemic therapy is under investigation
this article. Level of evidence: II
External beam pelvic RT is the standard therapy for high-risk Strength of recommendation: C
patients and is indicated to maximise pelvic control. The add- Consensus: 100% yes (37 voters)
ition of chemotherapy, or replacement of RT by chemotherapy, 2: No surgical nodal staging:
has been studied in several randomised trials. A historic GOG A. Adjuvant EBRT is generally recommended for pelvic
randomised trial that included patients with high-risk stage I control and relapse-free survival
and occult stage II disease found no benefit of adjuvant doxo- Level of evidence: III
rubicin after surgery and postoperative pelvic EBRT [137]. A Strength of recommendation: B
Japanese (JGOG 2033) and an Italian trial randomised patients B. Sequential adjuvant chemotherapy may be considered to
with high-risk endometrial cancer between pelvic EBRT and improve PFS and cancer-specific survival (CSS)
adjuvant cyclophosphamide, doxorubicin, cisplatin (CAP) Level of evidence: II
chemotherapy (three and five cycles, respectively), and both Strength of recommendation: C
trials found no difference in OS or disease-free survival (5-year C. There is more evidence to support giving chemotherapy
OS: 85% versus 87% and 69% versus 66%, respectively) and EBRT in combination rather than either treatment
[138, 139]. modality alone
Results of a combined analysis of the NSGO 9501/EORTC Level of evidence: II
55991 and MaNGO-ILIADE III randomised trials have been Strength of recommendation: B
published [140]. In this pooled analysis, the addition of adjuvant Consensus: 100% yes (37 voters)
chemotherapy (four cycles of platinum-based chemotherapy
given either before or after RT) to adjuvant EBRT was associated high-risk, stage II endometrial cancer. The definition of stage II
with a significant improvement in 5-year PFS (78% versus 69%, endometrial cancer was changed in the most recent FIGO 2009
P = 0.009), and a trend towards improved OS (82% versus 75%, staging system; tumours with endocervical glandular involvement
P = 0.07). Findings from a subgroup analysis suggested that the (previously stage IIA) were moved to stage I as this has no
benefit of adjuvant chemotherapy was restricted to patients prognostic impact [142, 143]. As a result, stage II now only
with endometrioid tumours rather than the 36% with serous or includes tumours with cervical stromal invasion [144].
clear-cell tumours. However, as this was an unplanned and Stage II tumours have been associated with an increased fre-
small subgroup analysis, no definite conclusions can be drawn quency of deep myometrial invasion and grade 3 histology,
on the efficacy of adjuvant chemotherapy for serous or clear-cell making it difficult to conclude if cervical invasion alone is the
cancers. reason for the observed higher risk of recurrence and lower OS
Promising results were found in the RTOG 9708 phase II compared with stage I disease [145]. In a SEER analysis that
study in 46 patients using concurrent pelvic RT and two included 1577 patients with stage II endometrial cancer, of
cycles of cisplatin (50 mg/m2 days 1 and 28) followed by four which half had stromal invasion, a multivariate analysis demon-
additional courses at 28-day intervals of cisplatin (50 mg/m2) strated no OS benefit for radical compared with simple hysterec-
and paclitaxel (175 mg/m2) as a 24-h infusion [141]. tomy, while RT (independently of the type of surgery) was
Reported 4-year OS rates were 85% for the whole group and associated with a survival benefit [146].
77% for stage III patients. This concurrent and adjuvant Controversy exists regarding the role of additional vaginal
chemotherapy schedule formed the rationale for the treat- brachytherapy boost in combination with EBRT [147]. The indi-
ment arms included in recently completed trials that investi- cation for a brachytherapy boost is clear in the rare situation of a
gated the role of combined cisplatin-based chemoradiation plus tumour with positive vaginal margin. However, in the adjuvant
adjuvant chemotherapy compared with either RT alone setting, it has historically been performed largely for stage II
(PORTEC-3) or chemotherapy alone (GOG258) for patients with disease. In randomised trials conducted in patients with inter-
high-risk and advanced stage endometrial cancer. The ongoing mediate-risk stage I endometrial cancer, there is no clear benefit
ENGOT-EN2-DGCG/EORTC55102 trial is evaluating the role of in terms of vaginal control among trials that included a vaginal
chemotherapy versus observation in patients with high-risk, node- brachytherapy boost compared with those that did not, with low
negative endometrial cancer. recurrence rates of ∼2% at 5 years after EBRT or vaginal brachy-
Recommendation 9.1: In patients with high-risk endometrial therapy alone [126, 127]. A SEER analysis conducted in patients
cancer (stage I endometrioid, grade 3, ≥50% myometrial inva- with stage IIIC endometrial cancer suggested a survival benefit
sion, regardless of LVSI status): for patients with ‘direct extension’ of the tumour, but no vaginal
Volume 27 | No. 1 | January 2016 doi:10.1093/annonc/mdv484 | 

by guest
on 05 February 2018
recurrence rates are available [148]. Other studies have found no and chemotherapy, as also evaluated in PORTEC-3, does indeed
difference in local recurrence or OS rates among patients with improve PFS and OS compared with chemotherapy alone.
stage II endometrial cancer treated with or without vaginal Recommendation 9.3: In patients with high-risk, stage III
brachytherapy in addition to EBRT, but it was associated with endometrial cancer and no residual disease:
increased risk of side-effects [149–153]. 1: EBRT is recommended to:
Recommendation 9.2: In patients with high-risk, stage II A. Decrease pelvic recurrence
endometrial cancer: Level of evidence: I
1: Simple hysterectomy, surgical nodal staging performed, node Strength of recommendation: B
negative: B. Improve PFS
A. Grade 1–2, LVSI negative: Recommend vaginal brachy- Level of evidence: I
therapy to improve local control Strength of recommendation: B
Level of evidence: III C. Improve survival
Strength of recommendation: B Level of evidence: IV
B. Grade 3 or LVSI unequivocally positive: Strength of recommendation: B
i. Recommend limited field EBRT 2: Chemotherapy is recommended to improve PFS and CSS
Level of evidence: III Level of evidence: II
Strength of recommendation: B Strength of recommendation: B
ii. Consider brachytherapy boost 3: There is more evidence to give chemotherapy and EBRT in
Level of evidence: IV combination than either alone in stage III disease:
Strength of recommendation: C A. IIIA: Chemotherapy AND EBRT to be considered
iii. Chemotherapy is under investigation B. IIIB: Chemotherapy AND EBRT to be considered
Level of evidence: III C. IIIC1: Chemotherapy AND EBRT to be considered
Strength of recommendation: C D. IIIC2: Chemotherapy AND extended field EBRT to be
Consensus: 97.3% (36) yes, 2.7% (1) abstain (37 voters) considered
2: Simple hysterectomy, no surgical nodal staging: Level of evidence: II
A. EBRT is recommended Strength of recommendation: B
Level of evidence: III Consensus: 94.6% (35) yes, 5.4% (2) abstain (37 voters)
B. Consider brachytherapy boost
Level of evidence: IV high-risk, non-endometrioid cancers. For the purpose of
Strength of recommendation: C these recommendations, serous, clear-cell, carcinosarcoma,
C. Grade 3 or LVSI unequivocally positive: Sequential adju- undifferentiated and mixed (>10%) tumours are regarded as
vant chemotherapy should be considered high-risk non-endometrioid-type cancers. These tumours
Level of evidence: III represent an infrequent subset of patients; hence, most studies are
Strength of recommendation: B retrospective and have included a limited number of patients. The
Consensus: 100% yes (37 voters) largest retrospective study conducted to date suggested a survival
benefit for the combination of chemotherapy and RT in uterine
serous cancer [159]. However, a subgroup analysis of the NSGO
high-risk, stage III endometrial cancer. In patients with stage 9501/EORTC 55991 and MaNGO-ILIADE III trials did not show
IIIC endometrial cancer, pelvic and/or extended field RT have a survival benefit for patients with serous or clear-cell tumours
been associated with increased OS and locoregional control rates, [140]. Given the high rates of distant metastasis observed in
while a higher rate of pelvic recurrence was found after adjuvant patients with uterine serous and clear-cell tumours, adjuvant
chemotherapy alone [154, 155]. In the GOG122 trial [156], chemotherapy can be considered and clinical trials addressing
women with advanced stage III/IV endometrial cancer were these rare subtypes are encouraged [136, 160]. One retrospective
randomised between whole abdominal irradiation and eight cycles study investigated the role of vaginal brachytherapy for stage I
of doxorubicin/cisplatin chemotherapy. Both adjusted PFS and serous or clear-cell cancers. The majority were either non-
OS were higher in the group who received chemotherapy invasive (26%) or had <50% myometrial invasion (58%), and
(predicted 5-year rates of 50% versus 38% and 55% versus 42%, 34% received adjuvant chemotherapy. The 5-year rate of isolated
respectively). However, event rates were high in both arms (50% pelvic recurrence was 4% and locoregional recurrence was 7%; the
and 54%). Patients with up to 2 cm residual disease were included 5-year OS rate was 84%, suggesting that vaginal brachytherapy
in this trial, suggesting that the dose delivered with whole alone is sufficient in patients with stage IA disease [161].
abdominal irradiation is not effective for macroscopic disease and Carcinosarcomas are regarded as metaplastic carcinomas con-
is toxic. In view of findings from the pooled NSGO/EORTC/Iliace taining both sarcomatous and carcinomatous elements [162].
trials [140] as well as results from prospective and retrospective They are rare and aggressive tumours with more than 35% of
trials [141, 154, 155, 157, 158], the use of combined RT and patients presenting with extra-uterine disease at diagnosis and
chemotherapy is recommended as opposed to either alone. are associated with a 5-year OS rate of 50% for patients with
Results of the recently completed GOG258 for stage III–IV stage I disease [163]. In the EORTC-55874 trial, patients with
endometrial cancer are awaited to see if the combination of EBRT stage I–II uterine sarcomas were randomised to receive adjuvant
 | Colombo et al. Volume 27 | No. 1 | January 2016

by guest
on 05 February 2018
RT after surgery. Of the 224 patients included, 91 had carcino- cases, palliative surgery can be performed to alleviate symptoms
sarcoma. In both groups, RT significantly reduced the risk of (e.g. bleeding or bowel obstruction).
local relapse but there was no difference in the rate of distant For patients with oligometastases or isolated retroperitoneal
metastasis and OS [164]. Three analyses of SEER data have been lymph node metastases, surgical resection is an option that can
reported in this setting, which initially showed a survival benefit be considered but the evidence of its benefit is limited.
for patients who received RT but who did not undergo lympha- Recommendation 10.1: For patients with advanced or recur-
denectomy [165, 166]. However, in a subsequent analysis, this rent disease, surgery is recommended only if optimal cytoreduc-
survival benefit was not maintained [167], thus limiting the con- tion (no residual disease) can be achieved. In selected cases,
clusions that can be drawn from these analyses. Finally, the palliative surgery is recommended to alleviate specific symptoms
GOG performed a trial in which whole abdominal irradiation Level of evidence: IV
was compared with three courses of ifosfamide and cisplatin Strength of recommendation: C
after complete resection. In this trial, chemotherapy was asso- Consensus: 100% yes (37 voters)
ciated with a numerically lower risk of recurrence and better Recommendation 10.2: Exenteration can be considered in
survival but the differences were not statistically significant selected patients with locally advanced tumours, and for isolated
[168]. central local relapse after radiation, if clear margins are expected
Recommendation 9.4: In patients with high-risk, non-endo- Level of evidence: IV
metrioid cancers: Strength of recommendation: C
1: Serous and clear cell after comprehensive staging: Consensus: 100% yes (37 voters)
A. Consider chemotherapy; clinical trials are encouraged Recommendation 10.3: Complete resection of distant oligo-
Level of evidence: III metastases and pelvic or retroperitoneal lymph node relapse can
Strength of recommendation: B be considered if technically possible according to localisation of
B. Stage IA, LVSI negative: Consider vaginal brachytherapy disease
only without chemotherapy Level of evidence: V
Level of evidence: IV Strength of recommendation: C
Strength of recommendation: C Consensus: 100% yes (37 voters)
C. Stage ≥IB: EBRT may be considered in addition to
chemotherapy, especially for node-positive disease
histology. Uterine serous cancer and clear-cell cancer account for
∼10% and 3% of advanced endometrial cancer cases, respectively
[170]. Patients with advanced disease have a worse prognosis than
those with endometrioid type, but there is no evidence that
2: Carcinosarcoma and undifferentiated tumours: histology should influence the decision regarding surgery.
A. Chemotherapy is recommended Recommendation 10.4: Histological type should not influ-
Level of evidence: II ence the decision whether or not to proceed with surgery
Strength of recommendation: B Level of evidence: IV
B. Consider EBRT; clinical trials are encouraged Strength of recommendation: B
Level of evidence: III Consensus: 100% yes (37 voters)
RT for isolated vaginal relapse in early-stage endometrial
cancer. RT is an effective therapeutic modality for improving
local disease control. However, fewer patients now receive
advanced and recurrent endometrial cancer adjuvant radiation for early disease. Observation after surgery
10. Does surgery or RT have a role in advanced or recurrent particularly applies to those with early-stage, grade 1–2 disease
endometrial cancer? without LVSI, as salvage RT in those who have a localised vaginal
relapse is associated with good local control [171]. In the
surgical cytoreduction. Patients with advanced disease (defined as PORTEC1 trial, 35 of 39 patients with a vaginal recurrence after
bulky FIGO stage IIIA-IV), or recurrent disease should only be surgery alone were treated with radical intent, mostly with
considered for surgery if it is anticipated that cytoreduction with no combinations of EBRT and brachytherapy, and in some cases with
macroscopic residual disease can be achieved. Cytoreduction also surgery. The complete remission rate was 89%, and 77% remained
includes removal of enlarged lymph nodes, but as there is no disease free with a median follow-up of 44 months [130]. Survival
evidence that a systematic pelvic and para-aortal lymphadenectomy after relapse was better in patients who did not receive primary
will influence PFS or OS, it should not be routinely performed. In a adjuvant RT; among patients who had received adjuvant RT, most
meta-analysis of 14 publications containing retrospective data from relapses were at distant sites. There is currently no evidence to
672 patients, median OS time was positively associated with an suggest that modern techniques of image-guided brachytherapy
increasing proportion of patients with no residual disease (each and intensity-modulated RT (IMRT) are superior to conventional
10% increase improved survival by 9.3 months, P = 0.04); the approaches, although a single-institution retrospective study of RT
change in survival for patients with between 0 and ≤2 cm of disease [EBRT predominantly using an IMRT technique followed by
after surgery was not significant [169]. Exenteration may be image-guided high dose rate (HDR) brachytherapy] for vaginal
considered for FIGO stage IIIA and central local relapse. In selected recurrence has also reported high tumour control rates [172].
Volume 27 | No. 1 | January 2016 doi:10.1093/annonc/mdv484 | 

by guest
on 05 February 2018
Recommendation 10.5: RT with curative intent is indicated in Recommendation 10.10: RT may be indicated for primary
patients with isolated vaginal relapse after surgery tumours that are unresectable, or where surgery cannot be per-
Level of evidence: III formed or is contraindicated for medical reasons
Strength of recommendation: A Level of evidence: IV
Consensus: 100% yes (34 voters) Strength of recommendation: B
chemotherapy with RT for recurrence. RT can be considered for
patients with vaginal or pelvic nodal recurrence. Improvements 11. What are the optimal systemic therapies for advanced/
in RT techniques allow for better means of localised treatment, recurrent disease?
or possibly retreatment of patients who have previously received The majority of patients with advanced or recurrent disease
RT. Whether chemotherapy has an additional benefit is unclear. will be candidates for systemic palliative therapy. The choice
The ongoing randomised phase II GOG0238 (NCT00492778) between hormonal treatment and chemotherapy relies on
trial is comparing pelvic irradiation of 45 Gy in 25 fractions plus several factors, including histopathological and clinical features
either brachytherapy or external beam boost with the same of the individual patient.
schedule plus concomitant cisplatin (40 mg/m2 weekly) in
women with vaginal/pelvic relapse who have not received
prior RT.
Recommendation 10.6: For vaginal or pelvic nodal recur- hormonal therapy: which patient and when? Hormonal therapy
rence, chemotherapy with RT could be considered in patients is indicated for patients with advanced or recurrent endometrial
with high-risk features for systemic relapse cancer and endometrioid histology. This statement is based on
Level of evidence: IV several clinical trials that have shown clinical activity with a
Strength of recommendation: C favourable toxicity profile [176, 177].
Consensus: 97.1% (33) yes, 2.9% (1) abstain (34 voters) Recommendation 11.1: Hormone therapy is indicated in
advanced or recurrent EEC
combined approaches to recurrence and re-irradiation. The use Level of evidence: II
of systemic therapy or surgery before RT for vaginal or pelvic Strength of recommendation: A
node recurrence could be considered in certain patients with Consensus: 100% yes (34 voters)
more bulky disease. As the techniques for image-guided RT
Response to hormonal therapy is quite variable, and a
have improved, there are situations where re-irradiation can be
number of pathological factors contributing to this variation
considered, although evidence from clinical trials is lacking.
have been identified. For example, hormonal therapy is more
Recommendation 10.7: Use of systemic therapy or surgery
likely to be effective in grade 1 or 2 endometrioid tumours. In a
before RT for vaginal or pelvic node recurrence could be consid-
large clinical trial of MPA, the response rate was 37% for grade
ered in certain patients
1, 23% for grade 2 and 9% for grade 3 tumours [176]. Others
have reported similar findings [177]. Patients with hormone re-
ceptor-positive disease have also been shown to have a higher
chance of responding to endocrine therapy. In a randomised
Recommendation 10.8: Re-irradiation could be considered in
trial, the response rate observed in patients with ER- and PgR-
highly selected patients using specialised techniques
positive disease was ∼25% and 37%, respectively, but was only
7%–8% in patients with ER/PgR-negative disease [176, 177].
Based on these results, it seems that positivity of ER and/or PgR
could be a predictive factor of response to endocrine therapy and
so should be determined before initiating hormonal therapy.
palliative RT. RT can be effectively used to palliate symptoms
Recommendation 11.2: Hormone therapy is more likely to be
such as bleeding, bone metastases or painful nodal recurrence.
effective in grade 1 or 2 endometrioid tumours
No randomised trials have been conducted comparing RT with
palliative chemotherapy.
Recommendation 10.9: RT is indicated for palliation of symp-
toms related to local recurrence or systemic disease
Recommendation 11.3: Hormone receptor status should be
determined before hormone therapy is initiated, as it is more
likely to be effective in patients with positive PgR and ER status
radical RT for primary endometrial cancer. RT can be used as a
primary treatment in patients with unresectable disease, or where
there are medical contraindications to surgery [173, 174]. Treatment Biopsy of recurrent disease can be considered, since there may
involves intrauterine brachytherapy alone or in combination with be differences in hormone receptor status in the primary and
EBRT. Image-guided brachytherapy may improve outcomes [175]. metastatic tumour. In a prospective collection of 686 primary
Two-year local control rates of more than 90% can be achieved for endometrial tumours and 171 metastatic lesions, loss of PgR
medically inoperable stage I disease. expression increased with disease progression, with 23% of
 | Colombo et al. Volume 27 | No. 1 | January 2016

by guest
on 05 February 2018
primary tumours and 76% of metastatic lesions demonstrating alone in terms of response rate (43%–41% versus 17%–25%) but
PgR loss [178]. with no benefit in terms of OS [185, 186]. The combination also
Recommendation 11.4: Biopsy of recurrent disease could be resulted in a higher incidence of grade 3–4 myelotoxicity and
considered as there may be differences in hormone receptor nausea/vomiting.
status in the primary and metastatic tumour In another GOG trial, conducted in patients with measurable
Level of evidence: III FIGO III–IV endometrial cancer, the addition of paclitaxel to
Strength of recommendation: C cisplatin and doxorubicin was associated with a higher response
Consensus: 100% yes (34 voters) rate and PFS than cisplatin and doxorubicin alone [objective re-
sponse rate (ORR): 57% versus 34%, respectively, P < 0.01;
Hormone therapy is the preferred front-line systemic therapy
median PFS: 8.3 versus 5.3 months, respectively, P < 0.01], and a
for patients with hormone receptor-positive grade 1 or 2
small but significant improvement in OS (median 15.3 versus
tumours in the absence of rapidly progressive disease, as it pro-
12.3 months, respectively, P = 0.037) [187]. However, toxicity,
vides an excellent benefit/risk ratio and convenient toxicity
especially peripheral neuropathy, was significantly higher (grade
profile. However, patients with visceral involvement and rapidly
2–3: 39% versus 5%, respectively). For this reason, it has not
progressive disease are not candidates for hormone therapy as it
been widely adopted as a standard of care.
is not usually associated with a rapid response.
Finally, GOG209 was a randomised, non-inferiority trial that
Recommendation 11.5: Hormone therapy is the preferred
compared the combination of paclitaxel 160 mg/m2, cisplatin
front-line systemic therapy for patients with hormone receptor-
60 mg/m2 and doxorubicin 50 mg/m2 (TAP) with paclitaxel
positive tumours—grade 1 or 2 and without rapidly progressive
175 mg/m2 and carboplatin AUC 6 (TC), both administered
disease
every 3 weeks. A total of 1305 patients were included in this
trial, and preliminary data (not yet fully published) indicate a
similar response rate (51.3% versus 51.2%) and PFS (median
13.5 versus 13.3 months) [188]. The median OS ( primary study
The progestogens, MPA 200 mg or MA 160 mg, are generally end point) was 40.3 months for TAP and 36.5 months for TC,
recommended. They have shown clear activity for the front-line which met the criteria of non-inferiority. TC had a more favour-
treatment of non-selected patients with recurrent or persistent able toxicity profile than TAP in this trial, with fewer patients
endometrioid tumours not suitable for surgery or RT, with re- discontinuing therapy due to toxicity (12% versus 18%). In add-
sponse rates of ∼25% and PFS times of 3 months [176, 179]. ition, TC can be administered in the outpatient setting whereas
Data from a randomised trial comparing low-(200 mg/day) with TAP is given in the inpatient setting in most countries. This
high-(1000 mg/day) dose MPA in 299 patients with advanced aspect may be important in terms of logistical, financial and
or recurrent endometrial carcinoma showed that low-dose MPA quality-of-life considerations in the palliative setting.
was more active than the high dose in terms of response rate Recommendation 11.8: The standard of care is six cycles of 3-
(25% versus 15%, respectively) and OS (11.0 versus 7.0 months, weekly carboplatin and paclitaxel. This is based on the prelimin-
respectively) [176]. ary communication of a randomised trial showing similar efficacy
Recommendation 11.6: Progestogens (e.g. MPA 200 mg or and less toxicity compared with cisplatin/doxorubicin/paclitaxel
MA 160 mg) are generally recommended Level of evidence: I
Level of evidence: III Strength of recommendation: A
Strength of recommendation: A Consensus: 100% yes (34 voters)
Evidence supporting the use of second-line chemotherapy
Other endocrine therapies have also demonstrated activity in after platinum-containing therapy in patients with endometrial
phase II trials among patients with advanced or recurrent endo- cancer is limited, especially in cases where the treatment-free
metrial cancer, with tamoxifen, anastrozole and fulvestrant all interval following first-line chemotherapy is <6–12 months.
associated with response rates of ∼10% [180–182]. Interestingly, Although various regimens have been evaluated in this setting
patients included in the anastrozole trial had not received prior [189–192], no randomised trials have been published. Therefore,
progestin therapy [182]. The combination of tamoxifen and no specific regimen can be recommended as a standard of care
MPA is associated with response rates and PFS similar to MPA for second-line chemotherapy.
alone [183, 184]. Recommendation 11.9: There is no standard of care for
Recommendation 11.7: Other hormonal agents to consider second-line chemotherapy
after progestins include tamoxifen, fulvestrant and aromatase Level of evidence: V
inhibitors Strength of recommendation: C
Level of evidence: III Consensus: 100% yes (34 voters)
Consensus: 100% yes (34 voters) 12. What are the most promising targeted agents and which
chemotherapy: is there any standard of care? Endometrial study designs should be used to evaluate their clinical
cancer is a relatively chemo-sensitive disease, with anthracyclines, benefit?
platinum-based drugs and taxanes shown to be the most active
agents. Two clinical trials showed that the combination of potentially ‘druggable’ molecular alterations in endometrial
cisplatin and doxorubicin was more active than doxorubicin cancer. According to the WHO classification of endometrial
Volume 27 | No. 1 | January 2016 doi:10.1093/annonc/mdv484 | 

by guest
on 05 February 2018
carcinoma, there are seven different types of tumours; however, usually modest and of short duration. Currently, several
endometrioid carcinoma, grade 3 and serous carcinomas different targeted therapies are undergoing clinical evaluation
account for the vast majority of aggressive tumours. Molecular but none are currently licensed for use. EGFR, human
genetic alterations involved in the development of endometrioid epidermal growth factor receptor-2 (HER2), mTOR and
cancers differ from those of serous tumours and this must be VEGFR inhibitors have been tested in phase I and II trials, with
taken into account when designing clinical trials to evaluate the modest response rates [200–203]. However, since this consensus
efficacy of molecular targeted agents. conference was held, findings from two randomised phase II
Over the last 15 years, it has been demonstrated that endomet- trials evaluating the addition of bevacizumab to TC in advanced
rial cancer shows microsatellite instability (MSI) and mutations or recurrent endometrial cancer suggest that this might be a
in PTEN, PIK3CA and KRAS, and that β-catenin genes are the promising approach worthy of further evaluation in phase III
most common molecular abnormalities in endometrioid carcin- clinical trials [204, 205]. GOG-86P was a three-arm trial
omas, whereas serous tumours have alterations of p53 and loss of evaluating the addition of bevacizumab, temsirolimus or
heterozygosity on several chromosomes, as well as other molecu- ixabepilone to first-line TC in 349 patients with advanced or
lar alterations (STK15, p16, E-cadherin and C-erbB2) [193]. recurrent endometrial cancer [204]. No differences in PFS were
Recently, the TCGA Research Network performed an integrated seen when the three arms were compared with historical data
genomic characterisation of endometrial carcinoma [5]. for TC from GOG 209 [188]. However, bevacizumab appeared
The PI3K/AKT pathway is one of the most frequently altered superior when the median OS results were compared with these
signalling pathways in endometrioid tumours, often resulting historical control data (34.0 versus 22.7 months, P < 0.039). In
from mutations in PTEN, PIK3CA and PIK3RI [194]. Of par- the MITO END-2 trial, which included 108 patients with
ticular interest is the downstream effector, mammalian target of advanced or recurrent endometrial cancer who had received 0
rapamycin (mTOR), and inhibitors of mTOR are now undergo- or 1 prior lines of chemotherapy, bevacizumab was added to six
ing evaluation in clinical trials. The RAS-RAF-MEK-ERK sig- to eight cycles of TC and then continued as maintenance
nalling pathway also plays an important role in these tumours, therapy. This approach resulted in a significant improvement in
with frequent mutations in KRAS, but also inactivation of median PFS (13 versus 8.7 months, P = 0.036) and a numerical
tumour suppressors such as RASF1A [195, 196]. Fibroblast increase in median OS (23.5 versus 18 months, P = 0.24),
growth factor-2 (FGFR2) is mutated in 10%–14% of endome- although these OS data are not yet mature [205].
trioid tumours and is a target for receptor tyrosine kinase inhi- Despite these promising results, few clinical trials of new tar-
bitors [197]. Angiogenesis also plays a role in endometrial geted therapies are molecularly driven [206] and the prevalence
tumorigenesis [198]. In addition, tumour homologous recom- of potential targets in metastatic lesions has been studied less
bination and mismatch repair deficiencies are seen in endome- than in primary tumours [178].
trioid tumours, the latter of which is particularly associated with Taken together, these findings suggest that PI3Kinase, mTOR
LS, and these pathways could be interesting targets. and angiogenesis inhibitors are the most promising classes of
Although there are a large number of specific gene abnormal- drugs to investigate in endometrial cancer [207], and progress in
ities and aberrant signalling pathways that appear to be promis- this area is likely to be faster if studies are biomarker driven with
ing targets, the frequency of each abnormality is small and this biopsy at entry.
presents a challenge to evaluating therapies in clinical trials Recommendation 12.2: Drugs targeting PI3K/mTOR
[199]. Examples include known tumour markers such L1CAM, pathway signalling and angiogenesis have shown modest activity
Anexin 2, other tyrosine kinase receptors [insulin-like growth but no agent has been approved for clinical use, and further bio-
factor receptor (IGFR), epidermal growth factor receptor marker-driven studies are warranted
(EGFR)] and signalling pathways involved in epithelial to mes- Level of evidence: III
enchymal transition [transforming growth factor-beta (TGF-β), Strength of recommendation: A
wnt] or stem cell-ness (Notch). PI3K/PTEN/AKT/mTOR Consensus: 100% yes (34 voters)
pathway, PTEN, MAPK-KRAS, angiogenesis [especially FGFR2
and vascular endothelial growth factor (VEGF)/VEGF receptor clinical trial design. While clinical trial end points such as OS
(VEGFR)], ER/PgR and homologous recombination deficiency and PFS are desirable, it may not be possible to make progress
(HRD)/MSI are altered in endometrial cancer, and the relevance unless novel trial design and end points are used. There should be
of these potential targets should be studied in clinical trials with better selection of patients, using a more systematic approach to
targeted agents. integration of biomarkers as well as earlier characterisation and
Recommendation 12.1: PI3K/PTEN/AKT/mTOR pathway, standardisation of diagnostic imaging and biomarker assessments.
PTEN, RAS-MAPK, angiogenesis (especially FGFR2 and Tumour response to biological agents may not occur to the same
VEGF/VEGFR), ER/PgR and HRD/MSI are altered in endomet- degree as with chemotherapy and alternative early end points,
rial cancer and their relevance should be studied in clinical trials such as the percentage of patients free from progression at 18
with targeted agents weeks [208], have been used. Trial designs that include different
Level of evidence: III gynaecological cancers of the same histotype should also be
Strength of recommendation: B considered, an approach that is being taken in the ongoing phase
Consensus: 100% yes (34 voters) III GOG0261 trial of paclitaxel plus carboplatin versus paclitaxel
plus ifosfamide in patients with different types of gynaecological
new agents in recurrent or metastatic endometrial cancer. The carcinosarcomas (NCT00954174), and a randomised phase II trial
benefit of standard chemotherapy and hormonal therapies is of nintedanib versus chemotherapy in patients with recurrent
 | Colombo et al. Volume 27 | No. 1 | January 2016

by guest
on 05 February 2018
clear-cell carcinoma of the ovary or endometrium (EudraCT property rights for some aspects relating to STMN/pSTMN1 as a
2013-002109-73). There is also an argument for not being too prognostic marker for endometrial cancer [US 127962,946 (HS)
selective, as the presence of a specific biomarker target may not be and US 147155,412 (HS)]). All remaining authors have declared
reflective of the probability of response. In a recent analysis of no conflicts of interest.
phase II studies of mTOR inhibitors, there was no correlation
between response and the presence of mutations in the PI3K/AKT
pathway [209], a result that could be explained by a variety of
reasons, including the presence of multiple mutations, cross-talk
in the signalling pathways involved, and the lack of re-biopsy references
samples to discount discordance between the tumour mutation 1. WHO. GLOBOCAN 2012: Estimated cancer incidence, mortality and prevalence
profile at diagnosis versus recurrence. worldwide in 2012. 2012; http://globocan.iarc.fr/Pages/fact_sheets_population.
Setting up individual trials is both costly and time-consum- aspx (3 April 2015, date last accessed).
ing, although adaptive phase II/III trials may offer some advan- 2. Lee NK, Cheung MK, Shin JY et al. Prognostic factors for uterine cancer in
tages [210]. Alternative strategies such as the ‘basket’ approach, reproductive-aged women. Obstet Gynecol 2007; 109: 655–662.
which includes all patients subdivided by specific histological or 3. National Cancer Institute. Endometrial cancer treatment Physician Data Query
molecular cohorts under the umbrella of a single trial, may be (PDQ). 2015; http://www.cancer.gov/cancertopics/pdq/treatment/endometrial/
the most efficient way forward [211]. Such trials should also in- healthprofessional (1 April 2015, date last accessed).
4. ACOG. ACOG practice bulletin, clinical management guidelines for obstetrician-
corporate novel end points and the design would be strength-
gynecologists, number 65, August 2005: management of endometrial cancer.
ened by the inclusion of sequential and repeated assessments of Obstet Gynecol 2005; 106: 413–425.
biomarkers. 5. Kandoth C, Schultz N, Cherniack AD et al. Integrated genomic characterization of
Recommendation 12.3: Clinical trial designs for new, targeted endometrial carcinoma. Nature 2013; 497: 67–73.
therapy: 6. Reid-Nicholson M, Iyengar P, Hummer AJ et al. Immunophenotypic diversity of
1. Basket studies with multiple cohorts related to histological sub- endometrial adenocarcinomas: implications for differential diagnosis. Mod Pathol
types and/or molecular alterations are considered a priority 2006; 19: 1091–1100.
2. Biomarker-driven clinical trials with biopsy at entry and se- 7. Broaddus RR, Lynch HT, Chen LM et al. Pathologic features of endometrial
carcinoma associated with HNPCC: a comparison with sporadic endometrial
quential biopsies in trials with molecular end points are
carcinoma. Cancer 2006; 106: 87–94.
recommended
8. Colombo N, Preti E, Landoni F et al. Endometrial cancer: ESMO Clinical Practice
3. PFS or PFS at a defined time-point are the preferred primary Guidelines for diagnosis, treatment and follow-up. Ann Oncol 2013; 24(Suppl 6):
end points for early phase trials vi33–vi38.
4. OS is the preferred primary end point in phase III trials, 9. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections
unless crossover is planned or expected among hematopoietic stem cell transplant recipients. Clin Infect Dis 2001; 33:
Level of evidence: V 139–144.
Strength of recommendation: A 10. World Cancer Research Fund/American Institute for Cancer Research.
Continuous Update Project Report. Food, Nutrition, Physical Activity, and the
Prevention of Endometrial Cancer. 2013; http://www.dietandcancerreport.org (2
April 2015, date last accessed).
acknowledgements 11. Garg K, Soslow RA. Endometrial carcinoma in women aged 40 years and
younger. Arch Pathol Lab Med 2014; 138: 335–342.
The authors thank Jennifer Lamarre, Claire Bramley, Matthew 12. Esposito K, Chiodini P, Capuano A et al. Metabolic syndrome and endometrial
Wallace, Aude Galli and all ESMO staff for their support cancer: a meta-analysis. Endocrine 2014; 45: 28–36.
throughout the whole consensus process. Angela Corstorphine 13. Zhang Y, Liu H, Yang S et al. Overweight, obesity and endometrial cancer risk:
of Kstorfin Medical Communications Ltd provided medical results from a systematic review and meta-analysis. Int J Biol Markers 2014; 29:
writing support with the preparation of this manuscript. This e21–e29.
support was funded by ESMO. 14. Rosato V, Zucchetto A, Bosetti C et al. Metabolic syndrome and endometrial
cancer risk. Ann Oncol 2011; 22: 884–889.
15. Luo J, Beresford S, Chen C et al. Association between diabetes, diabetes
funding treatment and risk of developing endometrial cancer. Br J Cancer 2014; 111:
1432–1439.
All costs relating to the consensus conference were covered from 16. Barry JA, Azizia MM, Hardiman PJ. Risk of endometrial, ovarian and breast
the European Society for Medical Oncology central funds. There cancer in women with polycystic ovary syndrome: a systematic review and meta-
was no external funding of the event or manuscript production. analysis. Hum Reprod Update 2014; 20: 748–758.
17. Fader AN, Arriba LN, Frasure HE, von Gruenigen VE. Endometrial cancer and
obesity: epidemiology, biomarkers, prevention and survivorship. Gynecol Oncol
disclosure 2009; 114: 121–127.
FA [senior investigator for the Research Fund Flanders (FWO)], 18. Ali AT. Reproductive factors and the risk of endometrial cancer. Int J Gynecol
Cancer 2014; 24: 384–393.
NC (consultancy - Roche, Astra Zeneca), LCdA (speaker for
19. Peiretti M, Colombo N. Sex cord-stromal tumors of the ovary. In Textbook of
Roche and Takeda), GE (research grants from Aeterna Zentaris
Gynaecological Oncology. Ankara and Istanbul: Günes Publishing, 2012;
and Astra Zeneca), CK (research funding from Roche, speaker for 453–456.
Roche), FL (research grants from Intuitive Surgicals, advisory 20. Brinton LA, Berman ML, Mortel R et al. Reproductive, menstrual, and medical
board for Roche), JL (advisory boards for AstraZeneca, Clovis risk factors for endometrial cancer: results from a case-control study. Am J
Oncology, Merck/MSD, Bayer, Oxigene), HS (pending intellectual Obstet Gynecol 1992; 167: 1317–1325.
Volume 27 | No. 1 | January 2016 doi:10.1093/annonc/mdv484 | 

by guest
on 05 February 2018
21. Zucchetto A, Serraino D, Polesel J et al. Hormone-related factors and 43. Yamazawa K, Hirai M, Fujito A et al. Fertility-preserving treatment with progestin,
gynecological conditions in relation to endometrial cancer risk. Eur J Cancer Prev and pathological criteria to predict responses, in young women with endometrial
2009; 18: 316–321. cancer. Hum Reprod 2007; 22: 1953–1958.
22. Fisher B, Costantino JP, Wickerham DL et al. Tamoxifen for prevention of breast 44. Koskas M, Uzan J, Luton D et al. Prognostic factors of oncologic and
cancer: report of the National Surgical Adjuvant Breast and Bowel Project P-1 reproductive outcomes in fertility-sparing management of endometrial atypical
Study. J Natl Cancer Inst 1998; 90: 1371–1388. hyperplasia and adenocarcinoma: systematic review and meta-analysis. Fertil
23. Lancaster JM, Powell CB, Chen LM, Richardson DL. Society of Gynecologic Steril 2014; 101: 785–794.
Oncology statement on risk assessment for inherited gynecologic cancer 45. Mittal K, Soslow R, McCluggage WG. Application of immunohistochemistry to
predispositions. Gynecol Oncol 2015; 136: 3–7. gynecologic pathology. Arch Pathol Lab Med 2008; 132: 402–423.
24. Smith RA, Cokkinides V, Brawley OW. Cancer screening in the United States, 46. Sun H, Enomoto T, Fujita M et al. Mutational analysis of the PTEN gene in
2009: a review of current American Cancer Society guidelines and issues in endometrial carcinoma and hyperplasia. Am J Clin Pathol 2001; 115: 32–38.
cancer screening. CA Cancer J Clin 2009; 59: 27–41. 47. Orbo A, Nilsen MN, Arnes MS et al. Loss of expression of MLH1, MSH2, MSH6,
25. Jacobs I, Gentry-Maharaj A, Burnell M et al. Sensitivity of transvaginal ultrasound and PTEN related to endometrial cancer in 68 patients with endometrial
screening for endometrial cancer in postmenopausal women: a case-control hyperplasia. Int J Gynecol Pathol 2003; 22: 141–148.
study within the UKCTOCS cohort. Lancet Oncol 2011; 12: 38–48. 48. Hecht JL, Pinkus JL, Pinkus GS. Enhanced detection of atypical hyperplasia in
26. National Cancer Institute. Endometrial Cancer Screening Physician Data endometrial polyps by PTEN expression. Appl Immunohistochem Mol Morphol
Query (PDQ). 2015; www.cancer.gov/cancertopics/pdq/screening/endometrial/ 2004; 12: 36–39.
HealthProfessional/ (2 April 2015, date last accessed). 49. Monte NM, Webster KA, Neuberg D et al. Joint loss of PAX2 and PTEN
27. ACOG. Committee Opinion No. 601: tamoxifen and uterine cancer. Obstet expression in endometrial precancers and cancer. Cancer Res 2010; 70:
Gynecol 2014; 123: 1394–1397. 6225–6232.
28. Fu Y, Zhuang Z. Long-term effects of levonorgestrel-releasing intrauterine system 50. Allison KH, Upson K, Reed SD et al. PAX2 loss by immunohistochemistry occurs
on tamoxifen-treated breast cancer patients: a meta-analysis. Int J Clin Exp early and often in endometrial hyperplasia. Int J Gynecol Pathol 2012; 31:
Pathol 2014; 7: 6419–6429. 151–159.
29. Manchanda R, Saridogan E, Abdelraheim A et al. Annual outpatient hysteroscopy 51. Quick CM, Laury AR, Monte NM, Mutter GL. Utility of PAX2 as a marker for
and endometrial sampling (OHES) in HNPCC/Lynch syndrome (LS). Arch Gynecol diagnosis of endometrial intraepithelial neoplasia. Am J Clin Pathol 2012; 138:
Obstet 2012; 286: 1555–1562. 678–684.
30. Vasen HF, Blanco I, Aktan-Collan K et al. Revised guidelines for the clinical 52. Tashiro H, Isacson C, Levine R et al. p53 gene mutations are common in uterine
management of Lynch syndrome (HNPCC): recommendations by a group of serous carcinoma and occur early in their pathogenesis. Am J Pathol 1997; 150:
European experts. Gut 2013; 62: 812–823. 177–185.
31. Duska LR, Garrett A, Rueda BR et al. Endometrial cancer in women 40 years old 53. Zheng W, Khurana R, Farahmand S et al. p53 immunostaining as a significant
or younger. Gynecol Oncol 2001; 83: 388–393. adjunct diagnostic method for uterine surface carcinoma: precursor of uterine
32. Evans-Metcalf ER, Brooks SE, Reale FR, Baker SP. Profile of women 45 years papillary serous carcinoma. Am J Surg Pathol 1998; 22: 1463–1473.
of age and younger with endometrial cancer. Obstet Gynecol 1998; 91: 54. Jia L, Liu Y, Yi X et al. Endometrial glandular dysplasia with frequent p53 gene
349–354.

ESMO Guidelines

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

ESMO Guidelines

Hochgeladen von

Copyright:

Verfügbare Formate

clinical practice guidelines Annals of Oncology 25 (Supplement 3): iii10–iii20, 2014

Anal cancer: ESMO-ESSO-ESTRO Clinical Practice

aetiology incidence and epidemiology

Downloaded from https://academic.oup.com/annonc/article-abstract/25/suppl_3/iii10/1740790

Figure 1. Anatomy of the anal region.

Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu159 | iii

iii | Glynne-Jones et al. Volume 25 | Supplement 3 | September 2014

Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu159 | iii

Table 3. Diagnostic work-up

Mandatory Optional but often recommended Optional

Biopsy to conﬁrm diagnosis

Table 4. Stage and site-based treatment

5-FU, 5-ﬂuorouracil; RT, radiotherapy.

iii | Glynne-Jones et al. Volume 25 | Supplement 3 | September 2014

Assessment of tumour response

Complete response Persistent / recurrent disease

Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu159 | iii

iii | Glynne-Jones et al. Volume 25 | Supplement 3 | September 2014

Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu159 | iii

iii | Glynne-Jones et al. Volume 25 | Supplement 3 | September 2014

Table 6. Levels of evidence and grades of recommendation

Volume 25 | Supplement 3 | September 2014 doi:10.1093/annonc/mdu159 | iii

iii | Glynne-Jones et al. Volume 25 | Supplement 3 | September 2014

Oesophageal cancer: ESMO Clinical Practice Guidelines

incidence and epidemiology diagnosis and pathology/molecular

Downloaded from https://academic.oup.com/annonc/article-abstract/27/suppl_5/v50/1741562

Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw329 | v

v | Lordick et al. Volume 27 | Supplement 5 | September 2016

Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw329 | v

management of advanced/metastatic human epidermal growth factor receptor 2 (HER2)-positive AC,

Table 3. Summary of recommendations

Diagnosis and pathology/molecular biology

v | Lordick et al. Volume 27 | Supplement 5 | September 2016

Upfront interdisciplinary planning of the treatment is mandatory [III, A].

HER2-positive metastatic AC should be treated with a trastuzumab-containing treatment [II, B].

Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw329 | v

v | Lordick et al. Volume 27 | Supplement 5 | September 2016

Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw329 | v

Gastric cancer: ESMO Clinical Practice Guidelines for

incidence and epidemiology suspected, referral to a geneticist for assessment is recom-

Downloaded from https://academic.oup.com/annonc/article-abstract/27/suppl_5/v38/1741597

Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw350 | v

TX Primary tumour cannot be assessed NX Regional lymph node(s) M0 No distant metastasis

Table 3. Anatomic stage/prognostic groups as per AJCC, 7th

v | Smyth et al. Volume 27 | Supplement 5 | September 2016

Operable Operable Inoperable or

Adjuvant Adjuvant HER2-negative: Consider clinical

Figure 1. Gastric cancer treatment algorithm.

Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw350 | v

v | Smyth et al. Volume 27 | Supplement 5 | September 2016

Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw350 | v

v | Smyth et al. Volume 27 | Supplement 5 | September 2016

Incidence and epidemiology

• Endoscopic resection is appropriate for selected early tumours [III, B]

Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw350 | v

v | Smyth et al. Volume 27 | Supplement 5 | September 2016

Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw350 | v

v | Smyth et al. Volume 27 | Supplement 5 | September 2016

Volume 27 | Supplement 5 | September 2016 doi:10.1093/annonc/mdw350 | v

Cancer of the pancreas: ESMO Clinical Practice

Downloaded from https://academic.oup.com/annonc/article-abstract/26/suppl_5/v56/344501

Volume 26 | Supplement 5 | September 2015 doi:10.1093/annonc/mdv295 | v

(iii) Inactivation of genome maintenance genes, such as hMLH1

v | Ducreux et al. Volume 26 | Supplement 5 | September 2015