Sie sind auf Seite 1von 10

Review

Stereotactic body radiotherapy for oligometastases


Alison C Tree, Vincent S Khoo, Rosalind A Eeles, Merina Ahmed, David P Dearnaley, Maria A Hawkins, Robert A Huddart, Christopher M Nutting,
Peter J Ostler, Nicholas J van As

The management of metastatic solid tumours has historically focused on systemic treatment given with palliative Lancet Oncol 2013; 14: e2837
intent. However, radical surgical treatment of oligometastases is now common practice in some settings. The Royal Marsden NHS
development of stereotactic body radiotherapy (SBRT), building on improvements in delivery achieved by intensity- Foundation Trust, London, UK
(A C Tree FRCR, V S Khoo MD,
modulated and image-guided radiotherapy, now allows delivery of ablative doses of radiation to extracranial sites.
M Ahmed MD, M A Hawkins MD,
Many non-randomised studies have shown that SBRT for oligometastases is safe and eective, with local control rates Prof C M Nutting MD,
of about 80%. Importantly, these studies also suggest that the natural history of the disease is changing, with 25 year N J van As FRCR); Institute of
progression-free survival of about 20%. Although complete cure might be possible in a few patients with Cancer Research, Sutton, UK
(V S Khoo,
oligometastases, the aim of SBRT in this setting is to achieve local control and delay progression, and thereby also
Prof D P Dearnaley FRCR,
postpone the need for further treatment. We review published work showing that SBRT oers durable local control R A Huddart PhD); Oncogenetics
and the potential for progression-free survival in non-liver, non-lung oligometastatic disease at a range of sites. Team, Institute of Cancer
However, to test whether SBRT really does improve progression-free survival, randomised trials will be essential. Research and Royal Marsden
NHS Foundation Trust,
London, UK
Oligometastases: denition and management radiotherapy [SBRT]), the subsequent metastatic load (Prof R A Eeles FRCR); and
The oligometastatic state: how do tumours metastasise? could theoretically be reduced. Cancer Centre, Mount Vernon
Knowledge of cancer biology and behaviour is advancing With the development of next-generation tumour Hospital, Northwood,
Middlesex, UK (P J Ostler FRCR)
constantly, and several theories about cancer progression sequencing, evidence is emerging that key driver
Correspondence to:
have been proposed. In 1907, Halsted suggested that mutations trigger catastrophic molecular changes that
Dr Alison C Tree, Royal Marsden
breast cancer always spreads via the locoregional cause cancer. Stephens and colleagues9 described a process NHS Foundation Trust, London
lymphatic vessels in a stepwise way. Thus, as described that they call chromothripsis, in which up to hundreds of SW3 6JJ, UK
in his surgical series, the disease can sometimes be genomic rearrangements occur in an isolated cellular alison.tree@rmh.nhs.uk

cured if diagnosed at an early stage.1 Towards the end of event. Genetic rearrangements oscillate across aected
the 20th century, an opposing theory gained acceptance. regions and switch between two copy number states. The
This theory, which was also described initially in breast hallmarks of this process are highly improbable if
cancer, proposed that cancer is a systemic disease that, rearrangements accumulate randomly over time; thus,
if it will ever metastasise, will already have done so early these rearrangements probably occur during one cellular
in the disease course, and that local therapies are crisis. Chromothripsis occurs in at least 23% of all
therefore less important than the tumour micro- cancers of many histological types. This process has
environment or systemic therapies.2 A third theory, the important implications for the way in which cancers
spectrum hypothesis, suggests that cancers range develop. Furthermore, if a few metastases (oligometastases)
between disease that remains local throughout its occurred rst and others then seeded from the original
course and disease that is already systemic at the time of metastases when those became genetically unstable, the
diagnosis.3,4 None of these theories has been tested process of oligometastases seeding further extensive
directly in a randomised trial, but they have all aected widespread metastatic disease would have a molecular
oncology practice. basis. This hypothesis is being tested by the International For the International Cancer
None of these paradigms describes the progression of Cancer Genome Consortium at the molecular level by Genome Consortium see
http://icgc.org/
cancer at a biological level.5 Until recently, cancer was sequencing of primary and metastatic tumours.
thought to be driven by somatically acquired point Hellman and Weichselbaum5 rst proposed the idea of
mutations and chromosomal rearrangements, con- an oligometastatic state in 1995. They suggested that for
ventionally believed to accumulate gradually over a long many cancers a few metastases exist at rst, before the
period (ie, many years). malignant cells acquire widespread metastatic potential.
Some cancer characteristics seem to be specic to a The role of radiotherapy in the management of metastatic
metastatic phenotype, such as altered cell adhesion, disease has been mainly limited to palliation, but we now
intravasation, and survival in the circulation.6 However, have an opportunity to challenge this idea. Theoretically,
some primary tumour cells have inadequate capability in if radical intervention (dened here as hypofractionated
one or more of the biological requirements for meta- radiotherapy) could be delivered during an oligometastatic
stasis; tumour dormancy could be one such example. phase, the intervention could change disease progression
Metastases arise through clonal expansion, as shown by in patients who would otherwise have been treated
high-resolution, genome-wide single-nucleotide poly- palliatively in most settings.
morphism and copy number analyses7 and next- Surgical removal of metastases has good outcomes in
generation sequencing;8 clones with selective advantage various settings, including liver and lung metastases.1012
give rise to metastases. If these clones are destroyed Although surgical resection has proved this theory in
before this process occurred (eg, with stereotactic body some situations, now that improved imaging, better

www.thelancet.com/oncology Vol 14 January 2013 e28


Review

patients treated aggressively would have had more


Prescription isodose
A 50% isodose extensive disease than was visible on CT or MRI.
30% isodose Improved imaging will enable better selection of patients.
10% isodose
Various groups have identied three distinct cohorts of
patients with oligometastases:5,13,14 those who present with
oligometastatic disease, those with induced oligo-
metastatic disease after cytoreductive therapy, and those
with relapsed oligometastatic disease after curative
locoregional therapy. These dierent groups probably
have dierent prognoses, so therapeutic approaches
might dier.
Factors that favour a truly oligometastatic state include
a long latent interval between the treatment of the
primary tumour and the appearance of metastases, and a
high ratio of the metastasis growth rate to that of the
primary tumour. Withers and Lee15 summarise this idea
as The more micrometastatic doubling times elapsing
B
between removal of the primary and the clinical detection
of the 'leader' metastasis, the higher the probability of
an oligometastatic distribution. Similarly, if several
metastases develop soon after treatment of the primary
tumour, they are unlikely to indicate oligometastatic
disease, unless they grow very rapidly.
We, and others, hypothesise that a subset of patients
might have oligometastatic disease and that high-dose,
radical treatment could change their prognosis.5,16,17 If this
hypothesis is correct, it would challenge the dogma that
most patients with oligometastatic disease should be
treated only with palliative intent.

Surgical treatment of oligometastatic disease


Figure 1: Stereotactic body radiotherapy for oligometastatic disease Surgery for the management of oligometastatic disease
(A) Radiotherapy plan for a patient with an isolated nodal metastasis from a previously treated prostate cancer, is now standard practice in several settings, with the
which received 30 Gy in three fractions. (B) Treatment of a vertebral metastasis from a previously treated prostate
cancer, which received 27 Gy in three fractions. In both images, the orange line indicates the region that was
aim of improving local control and overall survival.12,1821
administered the prescription dose. In the management of liver metastases from colorectal
cancer, liver resection produced 5-year survival of
methods of planning treatment, and stereotactic 3855%;11,18,19,22 however, this treatment has never been
radiotherapy delivery are available, good outcomes can tested in a randomised trial because randomisation of
be achieved without the risk and morbidity associated patients to a conservative treatment option in this
with surgery. Since stereotactic radiotherapy is associated setting would not be deemed ethical. In patients with
with low toxicity rates, the quality-of-life benet is soft-tissue sarcoma, resection of pulmonary metastases
probably greater. gave long-term (90-month) survival of 35% and, at
Two major diculties have complicated assessment of median follow-up of 37 months, 13 of the 77 patients in
the benet of radical treatment of oligometastases. First, this series were alive with no evidence of disease.23 In
in the past, delivery of truly ablative doses of radiation various other tumours, resection of solitary lung
has been limited by the risk to normal tissue, and the metastases has led to long-term survival,17,21,2427 which
need for extended fractionation. However, new supports the notion that metastasectomy enables good
technologies for delivery of SBRT (gure 1) now allow long-term disease control.
delivery of radical ablative doses of radiation. Second, the
identication of patients with truly oligometastatic SBRT and in-eld control of metastases
disease is inherently dicult. Most published surgical The American Society of Radiation Oncology denes
oligometastatic series describe patients managed in an SBRT as external beam radiotherapy used to deliver a
era before modern imaging techniques such as PET-CT high dose of radiation very precisely to an extracranial
became widely available. Thus, many patients were target within the body, as a single dose or a small number
probably understaged, potentially leading to under- of fractions.28 In addition to increasing the accuracy of
estimation of the eect of aggressive management on treatment delivery, thus reducing the amount of normal
truly oligometastatic disease, since some of those tissue irradiated, the high radiation dose per treatment

e29 www.thelancet.com/oncology Vol 14 January 2013


Review

(fraction) can potentially ablate all tissue in the treated metastasis control rates of around 80% for liver and lung
area.2831 metastases.3639 Toxicity rates seem low, with less than 5%
Published studies of SBRT for metastatic disease can be severe (grade 3) toxicity,33,36,37,39,40 and will probably be
divided into three types: rst, studies in which various reduced further with increasing knowledge, standard
types of primary tumour or a range of metastatic locations dose recommendations, and organ-at-risk constraints.41,42
are treated; second, those in which a single metastatic site The abundance of published studies on this topic, and
is treated, such as the lungs or liver; and third, those in the excellent outcomes reported, mean that the
which investigators focus on one histological type. framework for patients with liver and lung metastases is
Clearly, both primary histology and metastatic site are already changing, with more radical treatment, such as
important determinants of outcome, and the case mix in SBRT, becoming more common.
all these studies should be analysed closely. Some
histological types are probably more amenable to radical SBRT for non-liver, non-lung extracranial
treatment of oligometastases than others.32 metastases
Can SBRT achieve control of treated metastases?
Management of lung and liver metastases Tables 1 and 2 show an overview of published case series,
Emerging evidence suggests that liver and lung and phase 2 and 3 trials with a median follow-up of
metastases can be treated locally with low toxicity and 12 months in patients with oligometastases. Although
excellent outcomes, which supports the hypothesis that the studies reviewed are heterogeneous in terms of site,
radical treatment of oligometastatic disease can improve primary histology, and dose, control rates for treated
outcomes. metastases are mostly around 80%.
Lo, Hyer, Schefter, and colleagues3335 have sum- In both tables 1 and 2, some series included patients
marised data supporting local ablative therapies in lung with metastatic disease outside the irradiated area.
and liver metastases, although some series included Although these patients did not meet the criteria for
patients with stable metastatic disease outside the treated oligometastatic disease, the data for treated metastasis
area. Control rates for lung and liver metastases are control and toxicity in these settings are nonetheless
generally high (mostly about 7090% for 2-year local informative.
control), although many dierent doses and fractionation The largest series of mixed oligometastatic disease is
regimens have been used.33 Several studies have been that of Milano and colleagues.53 The investigators
published within the past 2 years, showing treated recruited 121 patients with ve or fewer metastases and

Study year Number of patients Dose Primary site Treated site(s) Treated metastasis control Toxicity
Bignardi et al 43
2010 19 45 Gy in 6 fractions Mixed Abdominal lymph nodes 778% at 2 years Grade 3 in 1 patient (5%)
(reduced in 6/19 cases)
Casamassima et al44 2011 25 Most common dose Prostate Pelvic, para-aortic, or 3-year local control 90% No grade 2 or higher
30 Gy in 3 fractions mediastinal lymph nodes
Choi et al45 2009 30 Most received 3345 Gy Mostly cervix, Para-aortic nodes 4-year local control 674% 20% grade 3 (but 16%
in 3 fractions some haematological because
endometrial most patients also had
chemotherapy)
Kim et al46 2009 7 Median 48 Gy in Gastric Para-aortic nodes 100% (median follow-up 26 No grade 3 recorded
3 fractions months)
Kim et al47 2008 23 Median 39 Gy in Rectal Pelvic/presacral lymph 4-year local control 743% Grade 3 in 1 patient
3 fractions nodes (4%): rectal perforation
Casamassima et al48 2012 48 Most common dose Mixed (mostly Adrenal 2-year local control 90% No grade 3 recorded
36 Gy in 3 fractions NSCLC and colon) (2-year overall survival
145%)
Chawla et al49 2009 30 patients (only Median dose 40 Gy in Mostly NSCLC Adrenal 2-year local control 27% No grade 2 or higher
14 had 10 fractions
<5 metastases)
Holy et al50 2011 13 with only adrenal Median dose 40 Gy in All NSCLC Adrenal 21-month local control 77% 2 patients had gastric
metastases 5 fractions ulcer (probably grade 2
toxic eect)
Scorsetti et al51 2012 34 Median dose 32 Gy in 64% NSCLC Adrenal 1-year local control 66%, No grade 3, 6% grade 2
4 fractions 2-year local control 33% nausea
Torok et al52 2011 7 patients Median 16 Gy in NSCLC in 4 of 7 Adrenal 1-year local control 63% Not known
(9 metastases) 1 fraction

NSCLC=non-small-cell lung cancer.

Table 1: Stereotactic body radiotherapy for lymph-node or adrenal oligometastases

www.thelancet.com/oncology Vol 14 January 2013 e30


Review

Study Number of patients Dose Primary site Treated site(s) Treated metastasis control Toxicity
year (number of lesions)
Milano 2008 121 (293) Various; median 50 Gy in All (mostly breast Lung, liver, bone, 2-year LLC 77%; 4-year LLC 74% Grade 3 in 1 patient (1%)
et al53,54 10 fractions and colorectal) lymph node, 7 CNS
Salama et al55 2011 61 (113) Increasing from 24 Gy in All (26% NSCLC) Lung, liver, lymph 2-year LLC 667%; 880% if dose Acute grade 3 in 2 (3%), 6 possible
3 fractions to 48 Gy in node, bone 30 Gy in 3 fractions late grade 3 (10%)
3 fractions
Kang et al56 2010 59 (78) 42 Gy in 3 fractions Colorectal Lung, liver, lymph 3-year local control 66% (note No grade 3, 3% grade 4
node, other 69% of patients had PD after (gastrointestinal perforation/
chemotherapy) obstruction)
Inoue et al57 2010 44 (60) 48 Gy in 8 (adrenal), Mostly lung Lung, adrenal, brain 3-year local control 80% 98% grade 2; no grade 3 or higher
3560 Gy in 48 fractions
(see text for details)
Stinauer et al58 2011 30 (53) 4050 Gy in 5 fractions or Renal-cell and Lung, liver, bone 18-month local control 88% One grade 3 hypoxia (3%)
4260 Gy in 3 fractions melanoma
Bae et al59 2012 41 (50) Median 48 Gy in Colorectal Lymph node, lung, 3-year local control 64% No acute grade 3, 7% late grade 3
3 fractions liver
Jereczek-Fossa 2011 34 (38) 30 Gy in 5 fractions to Prostate Lymph node, bone, 88% local control 6% grade 3 urinary, 3% grade 3
et al60 36 Gy in 3 fractions prostate recurrence rectal (all prostate recurrence
patients), 6% grade 3 late urinary
Hoyer et al61 2006 64 (141) 45 Gy in 3 fractions Colorectal Liver, lung, nodes, 2-year local control 63% (86% 30% grade 3: pain, nausea, skin
other LLC) reaction; 9% grade 4
Wersall et al62 2005 58 (162) 3040 Gy in 3 fractions Renal-cell Lung (majority), Local control 90% or higher 40% had grade 1 or higher toxicity,
was most common dose carcinoma renal bed, lymph with a high proportion of grade 3
node, adrenal events (some perhaps in the same
patient); one death (gastric
haemorrhage)
Svedman 2006 30 (82) Various: 40 Gy in Renal-cell Lung (majority), Only 2% documented progression 4% of side-eects were grade 3
et al63 4 fractions was most carcinoma renal bed, adrenal at median follow-up 52 months
common dose
Nuyttens 2007 14 (15) Median 7 Gy/fraction, Mixed Mixed 100% local control at median No grade 3
et al64 median 6 fractions follow-up 18 months
Greco et al65 2011 103 (126) 1824 Gy in 1 fraction Prostate, renal, Majority bone, lymph Local control at 2 years 64% (82% <4% grade 3 late (stricture, neuritis)
colorectal node, soft tissue if >22 Gy, 25% for 1820 Gy)

LLC=lesion local control. NSCLC=non-small cell lung cancer. PD=progressive disease.

Table 2: Stereotactic body radiotherapy for mixed oligometastatic sites

gave them a median dose of 50 Gy in 5-Gy fractions over bleeding, neuritis, and pneumonitis, although causality
2 weeks,53 although the target volume was covered only by from treatment was not conclusive in all cases.
the 80% isodose and thus the actual dose to some parts of Inoue and colleagues57 studied a mixed cohort of
the tumour would be less than 50 Gy. These doses are at patients treated with various radical (ie, lesion-ablative)
the lower end of the dose range presently used for strategies, including surgery, SBRT, and stereotactic
stereotactic radiotherapy. Most patients had lung, liver, or cranial treatment. 44 patients with 60 metastases were
lymph-node metastases. The only grade 3 toxic eect was treated with radical intent. The investigators noted good
a non-malignant pleural and pericardial eusion, and no treated metastasis control at 3 years (80%). They
higher-grade adverse eects occurred. The 2-year and identied the interval from primary tumour to recurrence
4-year treated metastasis control rates were 77% and as an important prognostic factor, with 5-year survival of
74%.54 The investigators found that the net gross tumour 10% for patients with an interval to recurrence of less
volume correlated with overall survival and local control. than 12 months, compared with 40% if that interval was
Salama and colleagues55 did a dose-escalation trial in longer than 12 months. Adverse eects were limited to
61 patients with 113 metastases. The rst patients were three instances of grade 2 brain necrosis and one case of
given 24 Gy in three fractions, increased sequentially to grade 2 intercostal neuralgia.
48 Gy in three fractions. For those who received 24 Gy in In general, the toxicity of SBRT is moderate. With the
three fractions, control of the treated metastases was exception of the study by Hoyer and colleagues61 (see
poor at 457%, whereas in the four patients given 48 Gy in table 2), the proportion of patients with grade 3 acute or
three fractions, 100% control was achieved. Two patients late adverse eects is less than 10%, and in many studies
experienced acute grade 3 toxic eects (vomiting in one no grade 3 eects were recorded.44,48,50
patient and fatigue in the other). Six patients experienced Most in-eld recurrences occur within the rst 2 years
late grade 3 toxic eects, including gastrointestinal after SBRT. The update of Milano and colleagues case

e31 www.thelancet.com/oncology Vol 14 January 2013


Review

series32 indicated that the 2-year and 6-year in-eld control and hip also occurred. All patients were staged by choline
rates were very similar, and were identical for patients PET-CT to exclude other disease (only one to two treatable
with primary breast tumours. For abdominal lymph-node lesions allowed) and were given a median single dose of
oligometastases, no further in-eld failures occurred after 20 Gy. Patients were followed up with regular MRI and
the rst year of follow-up.43 Any further relapses after this choline PET-CT. The in-eld control rate at 24 months
point are more likely to occur at other metastatic sites. In was 955%. Median overall survival had not been reached
the series of patients with lung metastases analysed by at the time of writing (median follow-up 14 months).
Okunie and colleagues,66 all in-eld failures occurred A trial (RTOG 0631) by the Radiation Therapy Oncology
within the rst 15 months. By contrast, in Kang and Group is comparing SBRT with conventional radio-
colleagues study of mixed oligometastatic sites from therapy; patients are randomly assigned 1618 Gy single-
colorectal cancer,56 in-eld control decreased from 66% to fraction SBRT versus 8 Gy in a single fraction. The
24% between 3 and 5 years of follow-up. primary endpoint of this study is pain control which
Table 3 show selected series of published work emphasises the importance of symptom control in these
supporting the use of SBRT to treat spinal metastases patients.
that has recently been reviewed.73 Doses varied, but in
most studies were delivered in between one and ve Does everyone relapse with distant metastases even if
fractions, often 1620 Gy in one fraction,70,74,75 or 27 Gy in in-eld control is achieved?
three fractions,75 and the reported in-eld control rate Despite the heterogeneous nature of the studies listed
was 84100%. Almost all data are from single-institution above, the methods used to categorise oligometastatic
series. Although not frequently referred to in the disease, and the inherent diculties of non-randomised
aforementioned studies, concern remains about the rate studies, the ndings indicate that a substantial
of vertebral fracture after SBRT, which can be as high as proportion of patients, generally around 20%, remain
20%, with the amount of risk related to the patients age, disease-free 24 years after SBRT (gure 2). In the series
pre-existing fracture, and local pain.76 A large recent reported by Inoue and colleagues,57 a mixture of surgery
series published in 2012 has reported the risk of vertebral and radiotherapy was used, and progression-free survival
compression fracture was 11%, with median follow-up of seemed to plateau at about 20% at 5 years, although the
74 months and a median time to fracture of 3 months.77 median follow-up was only 20 months. That study
Myelopathy is a rare but greatly feared complication of included a large number of patients with brain
spinal SBRT. In the largest series of 1075 patients, metastases, which might be associated with lower
retrospective review of case records identied six patients progression-free survival than in a population who had
(06%) who had developed myelopathy, with a median extracranial metastases. In a cohort of patients with
time to onset of 63 months.74 prostate cancer and nodal relapse alone on carbon-11-
Muacevic and colleagues67 reported a series of 64 bony labelled choline PET-CT scan, the 3-year disease-free
metastases in 40 patients with prostate cancer, many of survival after SBRT was only 17%,44 but the treated
which were spinal, but metastases in the skull, pelvis, metastasis control rate was 90%. Khan and colleagues78

Study year Number of patients Dose Primary site Treated site(s) Treated metastasis control Toxicity
(number of lesions)
Muacevic et al67 2011 40 (64) 20 Gy in 1 fraction Prostate Bone 2-year control 955% No grade 3 or higher
(median) (34/64 spine)
Wang et al68* 2012 149 (166) 2730 Gy in 3 fractions Mixed (32% renal) Spine 72% (median follow-up 7% grade 3 (non-cardiac
159 months) chest pain, other pain,
nausea, fatigue)
Yamada et al69* 2008 93 (103) 1824 Gy in 1 fraction Mixed (high Vertebrae 90% at 15 months 1 acute grade 3 (1%), 1 late
proportion of grade 3 (1%)
renal-cell carcinoma)
Gerstzen et al70* 2007 393 (500) Mean maximum dose Mixed Vertebrae 88% at median follow-up No signicant neurological
20 Gy in 1 fraction 21 months (100% for breast and eects recorded
lung primaries, 75% for melanoma)
Zelefsky et al71 2011 105 (105) Varied, but mostly Renal-cell carcinoma 99% bone 3-year local control 44%, but 88% 1 grade 4 skin (1%),
24 Gy in 1 fraction or metastases for 24 Gy in 1 fraction 4 fractures (not graded)
30 Gy in 5 fractions
Nguyen et al72* 2010 48 (55) 24 Gy in 1 fraction, Renal-cell carcinoma Spine (one or 82% 1-year spine progression-free 2% pain, 2% anaemia
27 Gy in 3 fractions, or two sites) survival
30 Gy in 5 fractions

*Percentage of patients with oligometastatic disease is not known for these studies.

Table 3: Stereotactic body radiotherapy for treatment of spinal metastases

www.thelancet.com/oncology Vol 14 January 2013 e32


Review

Factors aecting local control, progression-free survival,


60
Series with <30 patients and overall survival
Series with 3050 patients Histology
Series with >50 patients In Milano and colleagues series,32 patients with breast
cancer were analysed separately from those with other
types of cancer (mainly colorectal or lung), because
40 patients with breast cancer have much better survival:32
Disease-free survival (%)

progression-free survival at 2 years was 36% for patients


with breast cancer compared with 13% for those with
non-breast cancers, and overall survival at 6 years was
47% versus 9%. The rate of local control is also higher
20
in breast cancers (87%) than in non-breast cancers
(74%), which suggests that better treated metastasis
control results in improved progression-free survival,
although breast cancer histology itself confers a better
prognosis than cancers at other primary sites. Takeda
and colleagues80 reported that lung metastases from
0 colorectal cancer had a substantially worse in-eld
8

e 57

di 43

a 44

a 55

46

oi 45

47

r 61

e 59
3,5
an 7

ng 5

control rate than did lung metastases from other


08
09
no 5

ye
ou

lam

Ba
sim
ar

Ch
Kh

Ka

20
20

Ho
In

gn

ila

primary cancers (mostly non-small-cell lung cancer and


as

Sa

m
m
Bi

M
m

Ki
Ki
sa

head and neck primaries). However, in the larger


Ca

Milano series no signicant dierence in outcomes


Figure 2: Disease-free survival in patients with oligometastatic disease at 1748 months follow-up
Dotted line represents mean proportion of patients who were disease free at the reported timepoint, weighted for between colorectal and other non-breast histology was
number of patients in each cohort. Error bars represent 95% condence intervals. found.32

described a retrospective cohort of 23 patients with Disease-free interval


oligometastatic disease from non-small-cell lung cancer. Disease-free survival after SBRT seems to be associated
They received chemotherapy and local therapy, in the with time to recurrence. In Inoue and colleagues' series,57
form of surgery, radioablation, or both. At median follow- 3-year survival after SBRT was 53% for patients with a
up of 28 months for surviving patients, 30% were alive time to recurrence of more than 12 months, compared
with no evidence of disease. with 19% for patients with a recurrence time of less than
In a study in which the patterns of failure after SBRT to 12 months.In a cohort of 71 patients who underwent
treat oligometastases in one organ were assessed, 73% of SBRT for lung metastases, Zhang and colleagues38
patients eventually developed new metastases, at median reported that a disease-free interval of more than 12 months
follow-up of 45 months for surviving patients.79 More was signicantly related to overall survival (hazard
than 80% of new metastases occurred in the index organ. ratio 051, 95% CI 030087).
However, in that series, 27% of patients did not develop
distant metastases during follow-up, and all patients Location of metastases
without progression had been treated for metastases to A retrospective review of patients with metastases
just one organ. Overall, for those treated with SBRT in conned to one organ showed that those with bone or
Salama and colleagues series,55 50% had either no thoracic lymph-node metastases have better survival
metastatic progression or very little progression in terms than those with lung or liver metastases, and that they
of number and site of metastases. This group could be also tend to have a better progression-free survival.79
suitable for further radical intervention to the new Adrenal metastases seem to confer worse overall and
metastatic sites, such as resection or further SBRT. Data progression-free survival than other sites,53 possibly
from the International Registry of Lung Metastases because of the haematological mechanism of spread for
suggest that patients undergoing repeated removal of adrenal metastases. However, progression-free survival
metastases have similar survival to those who need only has been more encouraging in some later series of
one procedure.12 patients with adrenal metastases treated with SBRT.
These ndings support the idea of an oligometastatic Holy and colleagues50 recorded progression-free survival
state in which aggressive local therapy could improve of 12 months for patients with an isolated metastasis in
cause-specic survival. The study by Kang and the adrenal gland. In another series, mean time to
colleagues56 shows progression-free survival of 25% at systemic progression was 103 months after adrenal
3 years and 19% at 5 years. Whether patients in this SBRT,50 and at 2 years after SBRT, 14% do not have
cohort without disease 5 years from SBRT relapse at a distant metastases.81 The eects of metastasis location
later date or whether some of them can be truly cured and of histology are dicult to separate, because some
will be interesting to discover. tumour types have strong tendencies to metastasise to

e33 www.thelancet.com/oncology Vol 14 January 2013


Review

particular organs. For example, prostate cancer double-stranded and potentially repairable single-
metastasises frequently to bones, but rarely to visceral stranded DNA breaks. However, it overestimates cell kill
organs. Non-small-cell lung cancer commonly at high doses per fraction (eg, the doses used in SBRT),
metastasises to the adrenal glands, whereas the adrenals because there is proportionally more lethal damage to
are an uncommon site of isolated metastases for many DNA and less sublethal damage.83 The generalised
other tumour origins, including prostate, colorectal, and linear quadratic model has been suggested to
breast cancers. Some of the eect of metastasis location incorporate the conversion of sublethal to lethal DNA
is associated with the ability to deliver high doses to, for damage at high doses per fraction more accurately.
example, spinal metastases, which are immediately Another model, the universal survival curve,
adjacent to the radiosensitive spinal cord. Conversely, incorporates both the linear quadratic and multi-target
pelvic bony metastases can be given a high dose without models and can thus predict cell kill at both high and
any danger to the surrounding organs. low doses per fraction.84
Ablative radiotherapy doses (ie, those that destroy all
Number of metastases living tissue in an area) need to have a higher biologically
In Salama and colleagues series,55 patients with three or eective dose than do most conventionally fractionated
fewer metastases had better progression-free and overall regimens. Biologically eective dose is a measure of the
survival than those with four to ve metastases, all of eectiveness of dierent dose fractionation regimens,
whom had experienced progression within 9 months of to allow comparison of dierent doses or doses per
treatment. 75% of these patients developed widely fraction. No denition of the doses needed for ablative
metastatic disease compared with 46% of those with one radiotherapy is universally accepted, but regimens
to three metastases at the time of treatment. Bignardi resulting in a biologically eective dose larger than
and colleagues43 reported 2-year progression-free 100 Gy (/=10 Gy) would be deemed ablative. However,
survival of 417% for solitary metastases but 0% for whether an SBRT regimen needs to ablate all tissue in
more than one metastasis after SBRT. Wersall and the treated area to be eective is open to question.
colleagues62 showed that the overall survival of patients Lower doses, such as those used successfully in Milano
treated for oligometastatic disease was associated with and colleagues series,32 might be sucient to eradicate
the number of metastases. The patients with one to viable tumour cells. Results from a series of patients
three metastases had median survival of 37 months, with oligometastatic disease from non-small-cell lung
compared with 19 months in patients with more than cancer suggest that a regimen of 50 Gy in ten fractions
three metastases. is at least as eective as a moderately dosed three-
fraction regimen.85
Size In Stinauer and colleagues study,58 despite the small
Size of metastases is also predictive of in-eld control.45 number of patients included, dose per fraction (<11 Gy vs
Rusthoven and colleagues82 reported that lesions smaller >11 Gy), biologically equivalent dose (<100 Gy vs >100 Gy),
than 3 cm had 100% control at 2 years, compared with single fraction equivalent dose, and number of fractions
77% for lesions larger than 3 cm. In Milano and (three vs ve) were signicant predictors of in-eld
colleagues series,32 for patients with non-breast cancers, control. The investigators modelling studies predicted
gross tumour volume correlated inversely with in-eld that at least 48 Gy in three fractions would be needed to
control, which was similar to Kang and colleagues achieve greater than 90% 2-year control, although the
ndings.56 However, for breast cancer patients, in-eld majority (74%) of lesions treated in this study were lung
control was signicantly better for bone metastases metastases, so this dose might not represent that needed
(100%) than for non-bony metastases (68%), even though for other oligometastases. However, in multivariate
the bony metastases were larger on average. analysis, a dose of 48 Gy or more in three fractions was
However, target volume size does not predict for cancer also associated with better in-eld control in a series of
control in all series: in a series of 105 patients with renal- 41 patients with metastases from colorectal cancer.59
cell carcinoma, the planning target volume (range Additionally, in a large retrospective analysis of a series
1291462 cm) was not related to control.71 These data are of 141 patients with lung or liver metastases,86 a dose of
confounded by the fact that some metastases, such as 54 Gy or higher in three fractions was associated with a
those of the liver and lung, are probably bigger than higher treated metastasis control rate (893%) than a
nodal and bony metastases, so separation of the eects of dose of 36539 Gy (59%).
size and location is dicult. Similarly, from a multivariate analysis on their series,
Zelefsky and colleagues71 reported that a single versus a
Dose hypofractionated dose, and a single dose larger than
Various dose-fractionation models have been proposed 24 Gy compared with a dose smaller than 24 Gy, predicted
to predict the eectiveness of radiotherapy doses and improved in-eld control in renal-cell carcinoma. Any
fractionation regimens. The most widely used is the doseresponse in single-fraction spine treatments is not
linear quadratic model, which models cell kill from both yet clear.70,87

www.thelancet.com/oncology Vol 14 January 2013 e34


Review

Panel: Evidence-based practice for extracranial oligometastases Search strategy and selection criteria
Stereotactic body radiotherapy results in a high control rate of treated metastases Studies were identied in April, 2012, by searches of PubMed,
(~80%) EBSCOhost, and Web of Knowledge (search terms included
About 20% of patients are progression free at 23 years after stereotactic body combinations of oligometastases, stereotactic, radiotherapy,
radiotherapy metastases, radiosurgery, SBRT, and SABR), and by review of
Toxicity is low cited papers in selected articles. We identied 810 papers. We
Stereotactic body radiotherapy should be considered in patients with isolated assessed retrospective and prospective studies published in
metastases, especially if the disease-free interval is longer than 6 months English. We excluded duplicates or studies reporting only liver
Randomised trials are needed to establish whether stereotactic body radiotherapy or only lung metastases, but those including a mixture of sites
improves progression free and/or overall survival including some liver or lung metastases were permitted. We
Patients most likely to benet from stereotactic body radiotherapy have: included mixed studies of primary tumours (eg, lung) and
Long disease-free interval metastases in which more than half the patients had
Breast histology oligometastatic disease. Studies were eligible if the dose per
One to three metastases fraction was 6 Gy or higher for a total dose of more than 36 Gy,
Small metastases or 5 Gy per fraction for a total dose of more than 45 Gy. Studies
Higher radiation dose delivered (biologic eective dose >100 Gy) were excluded if fewer than seven patients were reported or if
median follow-up was less than 12 months. 24 studies met the
eligibility criteria for this Review, and four additional studies of
Conclusions spinal metastases were selected (table 3) for comparison.
The evidence reviewed here allows us to reach some
conclusions (panel) about the patients who could benet
from SBRT. Many non-randomised studies have shown is essential, and are therefore presently designing such a
that SBRT for oligometastases is safe and eective, and trial. The rst phase of the study will test whether
that treated metastasis control rates of about 80% are randomisation is feasible in this cohort of patients; if it
achieved. Toxicity seems to be moderate in most cases, is, we will progress to an international phase 3 study.
although complications that are rare with conventional Contributors
fractionation, such as vertebral fracture, have been VSK, RAE, MA, DPD, MAH, RAH, CMN, and PJO wrote and reviewed
reported. Importantly, these studies also suggest that the the paper. ACT did the literature search, made the gures and tables,
and wrote and reviewed the paper. NJvA designed the Review, and wrote
disease natural history is changing, with 25-year and reviewed the paper.
progression-free survival of about 20%. These early
Conicts of interest
clinical data are supported scientically by accumulating PJO, ACT, and NJvA have received educational travel grants from
molecular evidence that the development of widespread Accuray to facilitate attendance at international conferences. RAE has
metastases occurs because of seeding from oligo- received educational grants from Tepnel (now GenProbe) and Vista
metastases when the genome is deranged. This evidence Diagnostics. The other authors declare no conicts of interest.
indicates that, for at least some patients, we have an Acknowledgments
opportunity to control these oligometastases prior to This work was done in the Royal Marsden NHS Foundation Trust, which
received a proportion of its funding from the NHS Executive. The views
them acquiring the genetic alterations which confer expressed in this Review are those of the authors and not necessarily
widespread metastatic potential. those of the NHS Executive. This work was supported by the Institute of
For most oncological interventions in metastatic Cancer Research, the Bob Champion Cancer Trust, and Cancer Research
disease, cure is not the aim of treatment. Although true UK Section of Radiotherapy, grant number C46/A2131. We acknowledge
NHS funding to the NIHR Biomedical Research Centre.
cure might be an option in a few patients with
References
oligometastases, the objective of SBRT in most patients 1 Halsted WS. I. The results of radical operations for the cure of
with metastatic diseases is to control the treated carcinoma of the breast. Ann Surg 1907; 46: 119.
metastasis and delay progression, thereby also delaying 2 Hellman S. Karnofsky Memorial Lecture. Natural history of small
the need for further treatment. This goal is a useful breast cancers. J Clin Oncol 1994; 12: 222934.
3 Fisher B. Laboratory and clinical research in breast cancera
endpoint in a population whose quality of life is often personal adventure: the David A. Karnofsky memorial lecture.
adversely aected by their treatment as well as the disease. Cancer Res 1980; 40: 386374.
Whether SBRT improves progression-free and overall 4 Hellman S, Weichselbaum RR. Importance of local control in an
era of systemic therapy. Nat Clin Pract Oncol 2005; 2: 6061.
survival can only really be proved by means of randomised
5 Hellman S, Weichselbaum RR. Oligometastases. J Clin Oncol 1995;
trials that compare the standard of care (including all 13: 810.
appropriate systemic and palliative treatments) alone or 6 Gupta GP, Massague J. Cancer metastasis: building a framework.
with the addition of SBRT. As use of SBRT becomes Cell 2006; 127: 67995.
7 Liu W, Laitinen S, Khan S, et al. Copy number analysis indicates
more widespread worldwide, such trials will become monoclonal origin of lethal metastatic prostate cancer. Nat Med
increasingly more dicult to start because the treatment 2009; 15: 55965.
will be given as standard in accordance with patients 8 Navin N, Kendall J, Troge J, et al. Tumour evolution inferred by
single-cell sequencing. Nature 2011; 472: 9094.
best interests. We believe that randomised trial evidence

e35 www.thelancet.com/oncology Vol 14 January 2013


Review

9 Stephens PJ, Greenman CD, Fu B, et al. Massive genomic 33 Lo SS, Moatt-Bruce SD, Dawson LA, et al. The role of local therapy
rearrangement acquired in a single catastrophic event during in the management of lung and liver oligometastases.
cancer development. Cell 2011; 144: 2740. Nat Rev Clin Oncol 2011; 8: 40516.
10 Miller G, Biernacki P, Kemeny NE, et al. Outcomes after resection 34 Hyer M, Swaminath A, Bydder S, et al. Radiotherapy for liver
of synchronous or metachronous hepatic and pulmonary colorectal metastases: a review of evidence. Int J Radiat Oncol Biol Phys 2012;
metastases. J Am Coll Surg 2007; 205: 23138. 82: 104757.
11 Fong Y, Fortner J, Sun RL, Brennan MF, Blumgart LH. Clinical 35 Schefter TE, Kavanagh BD. Radiation therapy for liver metastases.
score for predicting recurrence after hepatic resection for metastatic Semin Radiat Oncol 2011; 21: 26470.
colorectal cancer: analysis of 1001 consecutive cases. Ann Surg 1999; 36 Dhakal S, Corbin KS, Milano MT, et al. Stereotactic body
230: 30918; discussion 31821. radiotherapy for pulmonary metastases from soft-tissue sarcomas:
12 Casiraghi M, De Pas T, Maisonneuve P, et al. A 10-year single-center excellent local lesion control and improved patient survival.
experience on 708 lung metastasectomies: the evidence of the Int J Radiat Oncol Biol Phys 2012; 82: 94045.
international registry of lung metastases. J Thorac Oncol 2011; 37 Ricardi U, Filippi AR, Guarneri A, et al. Stereotactic body radiation
6: 137378. therapy for lung metastases. Lung Cancer 2012; 75: 7781.
13 Christodouleas JP, Marks LB. Analysis of patients with 38 Zhang Y, Xiao JP, Zhang HZ, et al. Stereotactic body radiation
oligometastases undergoing two or more curative-intent stereotactic therapy favors long-term overall survival in patients with lung
radiotherapy courses: in regard to Milano et al. (Int J Radiat Oncol metastases: ve-year experience of a single-institution.
Biol Phys 2009;73:832-837). Int J Radiat Oncol Biol Phys 2009; Chin Med J (Engl) 2011; 124: 413237.
74: 1628; author reply 162829. 39 Vautravers-Dewas C, Dewas S, Bonodeau F, et al. Image-guided
14 Niibe Y, Hayakawa K. Oligometastases and oligo-recurrence: robotic stereotactic body radiation therapy for liver metastases: is
the new era of cancer therapy. Jpn J Clin Oncol 2010; 40: 10711. there a dose response relationship? Int J Radiat Oncol Biol Phys
15 Withers HR, Lee SP. Modeling growth kinetics and statistical 2011; 81: e3947.
distribution of oligometastases. Semin Radiat Oncol 2006; 40 Carey Sampson M, Katz A, Constine LS. Stereotactic body radiation
16: 11119. therapy for extracranial oligometastases: does the sword have a
16 Rubin P, Green J. Solitary metastases. Springeld, IL: double edge? Semin Radiat Oncol 2006; 16: 6776.
Charles C Thomas, 1968. 41 Timmerman RD. An overview of hypofractionation and
17 Rubin P, Brasacchio R, Katz A. Solitary metastases: illusion versus introduction to this issue of seminars in radiation oncology.
reality. Semin Radiat Oncol 2006; 16: 12030. Semin Radiat Oncol 2008; 18: 21522.
18 Simmonds PC, Primrose JN, Colquitt JL, Garden OJ, Poston GJ, 42 Benedict SH, Yenice KM, Followill D, et al. Stereotactic body
Rees M. Surgical resection of hepatic metastases from colorectal radiation therapy: the report of AAPM Task Group 101. Med Phys
cancer: a systematic review of published studies. Br J Cancer 2006; 2010; 37: 4078101.
94: 98299. 43 Bignardi M, Navarria P, Mancosu P, et al. Clinical outcome of
19 Primrose JN. Surgery for colorectal liver metastases. Br J Cancer hypofractionated stereotactic radiotherapy for abdominal lymph
2010; 102: 131318. node metastases. Int J Radiat Oncol Biol Phys 2011; 81: 83138.
20 van Geel AN, Rm van Der Sijp J, Schmitz PI. Which soft tissue 44 Casamassima F, Masi L, Menichelli C, et al. Ecacy of eradicative
sarcoma patients with lung metastases should not undergo radiotherapy for limited nodal metastases detected with choline
pulmonary resection? Sarcoma 2002; 6: 5760. PET scan in prostate cancer patients. Tumori 2011; 97: 4955.
21 Kai JT, Gusani NJ, Deshaies I, et al. Indications and approach to 45 Choi CW, Cho CK, Yoo SY, et al. Image-guided stereotactic body
surgical resection of lung metastases. J Surg Oncol 2010; radiation therapy in patients with isolated para-aortic lymph node
102: 18795. metastases from uterine cervical and corpus cancer.
22 Pawlik TM, Choti MA. Surgical therapy for colorectal metastases to Int J Radiat Oncol Biol Phys 2009; 74: 14753.
the liver. J Gastrointest Surg 2007; 11: 105777. 46 Kim MS, Yoo SY, Cho CK, et al. Stereotactic body radiotherapy for
23 Sardenberg RA, Figueiredo LP, Haddad FJ, Gross JL, Younes RN. isolated para-aortic lymph node recurrence after curative resection
Pulmonary metastasectomy from soft tissue sarcomas. in gastric cancer. J Korean Med Sci 2009; 24: 48892.
Clinics (Sao Paulo) 2010; 65: 87176. 47 Kim MS, Choi C, Yoo S, et al. Stereotactic body radiation therapy in
24 Meimarakis G, Angele M, Staehler M, et al. Evaluation of a new patients with pelvic recurrence from rectal carcinoma.
prognostic score (Munich score) to predict long-term survival after Jpn J Clin Oncol 2008; 38: 695700.
resection of pulmonary renal cell carcinoma metastases. Am J Surg 48 Casamassima F, Livi L, Masciullo S, et al. Stereotactic radiotherapy
2011; 202: 15867. for adrenal gland metastases: University of Florence experience.
25 Garca Franco CE, Torre W, Tamura A, et al. Long-term results after Int J Radiat Oncol Biol Phys 2012; 82: 91923.
resection for bone sarcoma pulmonary metastases. 49 Chawla S, Chen Y, Katz AW, et al. Stereotactic body radiotherapy for
Eur J Cardiothorac Surg 2010; 37: 120508. treatment of adrenal metastases. Int J Radiat Oncol Biol Phys 2009;
26 Watanabe K, Nagai K, Kobayashi A, Sugito M, Saito N. Factors 75: 7175.
inuencing survival after complete resection of pulmonary 50 Holy R, Piroth M, Pinkawa M, Eble MJ. Stereotactic body radiation
metastases from colorectal cancer. Br J Surg 2009; 96: 105865. therapy (SBRT) for treatment of adrenal gland metastases from
27 Rashid OM, Takabe K. The evolution of the role of surgery in the non-small cell lung cancer. Strahlenther Onkol 2011; 187: 24551.
management of breast cancer lung metastasis. J Thorac Dis 2012; 51 Scorsetti M, Alongi F, Filippi AR, et al. Long-term local control
4: 42024. achieved after hypofractionated stereotactic body radiotherapy for
28 Potters L, Kavanagh B, Galvin JM, et al. American Society for adrenal gland metastases: a retrospective analysis of 34 patients.
Therapeutic Radiology and Oncology (ASTRO) and American Acta Oncol 2012; 51: 61823.
College of Radiology (ACR) practice guideline for the performance 52 Torok J, Wegner RE, Burton SA, Heron DE. Stereotactic body radiation
of stereotactic body radiation therapy. Int J Radiat Oncol Biol Phys therapy for adrenal metastases: a retrospective review of a noninvasive
2010; 76: 32632. therapeutic strategy. Future Oncol 2010; 7: 14551.
29 Timmerman RD, Kavanagh BD. Stereotactic body radiation therapy. 53 Milano MT, Katz AW, Muhs AG, et al. A prospective pilot study of
Curr Probl Cancer 2005; 29: 12057. curative-intent stereotactic body radiation therapy in patients with
30 Timmerman RD, Kavanagh BD, Cho LC, Papiez L, Xing L. 5 or fewer oligometastatic lesions. Cancer 2008; 112: 65058.
Stereotactic body radiation therapy in multiple organ sites. 54 Milano MT, Katz AW, Schell MC, Philip A, Okunie P. Descriptive
J Clin Oncol 2007; 25: 94752. analysis of oligometastatic lesions treated with curative-intent
31 Lartigau E. Stereotactic body radiotherapy. BMJ 2011; 343: d4286. stereotactic body radiotherapy. Int J Radiat Oncol Biol Phys 2008;
32 Milano MT, Katz AW, Zhang H, Okunie P. Oligometastases 72: 151622.
treated with stereotactic body radiotherapy: long-term 55 Salama JK, Hasselle MD, Chmura SJ, et al. Stereotactic body
follow-up of prospective study. Int J Radiat Oncol Biol Phys 2012; radiotherapy for multisite extracranial oligometastases: nal report
83: 87886. of a dose escalation trial in patients with 1 to 5 sites of metastatic
disease. Cancer 2011; 118: 296270.

www.thelancet.com/oncology Vol 14 January 2013 e36


Review

56 Kang JK, Kim MS, Kim JH, et al. Oligometastases conned one 72 Nguyen QN, Shiu AS, Rhines LD, et al. Management of spinal
organ from colorectal cancer treated by SBRT. Clin Exp Metastasis metastases from renal cell carcinoma using stereotactic body
2010; 27: 27378. radiotherapy. Int J Radiat Oncol Biol Phys 2010; 76: 118592.
57 Inoue T, Katoh N, Aoyama H, et al. Clinical outcomes of stereotactic 73 Martin A, Gaya A. Stereotactic body radiotherapy: a review.
brain and/or body radiotherapy for patients with oligometastatic Clin Oncol (R Coll Radiol) 2010; 22: 15772.
lesions. Jpn J Clin Oncol 2010; 40: 78894. 74 Gibbs IC, Patil C, Gerszten PC, Adler JR Jr, Burton SA. Delayed
58 Stinauer MA, Kavanagh BD, Schefter TE, et al. Stereotactic body radiation-induced myelopathy after spinal radiosurgery.
radiation therapy for melanoma and renal cell carcinoma: impact of Neurosurgery 2009; 64 (2 suppl): A6772.
single fraction equivalent dose on local control. Radiat Oncol 2011; 75 Chang UK, Youn SM, Park SQ, Rhee CH. Clinical results of
6: 34. cyberknife(r) radiosurgery for spinal metastases.
59 Bae SH, Kim MS, Cho CK, et al. High dose stereotactic body J Korean Neurosurg Soc 2009; 46: 53844.
radiotherapy using three fractions for colorectal oligometastases. 76 Boehling NS, Grosshans DR, Allen PK, et al. Vertebral compression
J Surg Oncol 2012; 106: 13843. fracture risk after stereotactic body radiotherapy for spinal
60 Jereczek-Fossa BA, Beltramo G, Fariselli L, et al. Robotic metastases. J Neurosurg Spine 2012; 16: 37986.
image-guided stereotactic radiotherapy, for isolated recurrent 77 Cunha MV, Al-Omair A, Atenafu EG, et al. Vertebral compression
primary, lymph node or metastatic prostate cancer. fracture (VCF) after spine stereotactic body radiation therapy
Int J Radiat Oncol Biol Phys 2012; 82: 88997. (SBRT): analysis of predictive factors. Int J Radiat Oncol Biol Phys
61 Hoyer M, Roed H, Traberg Hansen A, et al. Phase II study on 2012; 84: e34349.
stereotactic body radiotherapy of colorectal metastases. Acta Oncol 78 Khan AJ, Mehta PS, Zusag TW, et al. Long term disease-free
2006; 45: 82330. survival resulting from combined modality management of patients
62 Wersll PJ, Blomgren H, Lax I, et al. Extracranial stereotactic presenting with oligometastatic, non-small cell lung carcinoma
radiotherapy for primary and metastatic renal cell carcinoma. (NSCLC). Radiother Oncol 2006; 81: 16367.
Radiother Oncol 2005; 77: 8895. 79 Milano MT, Katz AW, Okunie P. Patterns of recurrence after
63 Svedman C, Sandstrm P, Pisa P, et al. A prospective Phase II trial curative-intent radiation for oligometastases conned to one organ.
of using extracranial stereotactic radiotherapy in primary and Am J Clin Oncol 2010; 33: 15763.
metastatic renal cell carcinoma. Acta Oncol 2006; 45: 87075. 80 Takeda A, Kunieda E, Ohashi T, Aoki Y, Koike N, Takeda T.
64 Nuyttens JJ, Prevost JB, Van der Voort van Zijp NC, Hoogeman M, Stereotactic body radiotherapy (SBRT) for oligometastatic lung
Levendag PC. Curative stereotactic robotic radiotherapy treatment tumors from colorectal cancer and other primary cancers in
for extracranial, extrapulmonary, extrahepatic, and extraspinal comparison with primary lung cancer. Radiother Oncol 2011;
tumors: technique, early results, and toxicity. 101: 25559.
Technol Cancer Res Treat 2007; 6: 60510. 81 Scorsetti M, Mancosu P, Navarria P, et al. Stereotactic body
65 Greco C, Zelefsky MJ, Lovelock M, et al. Predictors of local control radiation therapy (SBRT) for adrenal metastases: a feasibility study
after single-dose stereotactic image-guided intensity-modulated of advanced techniques with modulated photons and protons.
radiotherapy for extracranial metastases. Int J Radiat Oncol Biol Phys Strahlenther Onkol 2011; 187: 23844.
2011; 79: 115157. 82 Rusthoven KE, Kavanagh BD, Cardenes H, et al. Multi-institutional
66 Okunie P, Petersen AL, Philip A, et al. Stereotactic body radiation phase I/II trial of stereotactic body radiation therapy for liver
therapy (SBRT) for lung metastases. Acta Oncol 2006; 45: 80817. metastases. J Clin Oncol 2009; 27: 157278.
67 Muacevic A, Kufeld M, Rist C, Wowra B, Stief C, Staehler M. Safety 83 Wang JZ, Huang Z, Lo SS, Yuh WT, Mayr NA. A generalized
and feasibility of image-guided robotic radiosurgery for patients linear-quadratic model for radiosurgery, stereotactic body radiation
with limited bone metastases of prostate cancer. Urol Oncol 2011; therapy, and high-dose rate brachytherapy. Sci Transl Med 2010;
published online April 13. http://dx.doi.org/10.1016/j. 2: 39ra48.
urolonc.2011.02.023. 84 Park C, Papiez L, Zhang S, Story M, Timmerman RD. Universal
68 Wang XS, Rhines LD, Shiu AS, et al. Stereotactic body radiation survival curve and single fraction equivalent dose: useful tools in
therapy for management of spinal metastases in patients without understanding potency of ablative radiotherapy.
spinal cord compression: a phase 12 trial. Lancet Oncol 2012; Int J Radiat Oncol Biol Phys 2008; 70: 84752.
13: 395402. 85 Hasselle MD, Haraf DJ, Rusthoven KE, et al. Hypofractionated
69 Yamada Y, Bilsky MH, Lovelock DM, et al. High-dose, image-guided radiation therapy for patients with limited volume
single-fraction image-guided intensity-modulated radiotherapy for metastatic non-small cell lung cancer. J Thorac Oncol 2012;
metastatic spinal lesions. Int J Radiat Oncol Biol Phys 2008; 7: 37681.
71: 48490. 86 McCammon R, Schefter TE, Gaspar LE, Zaemisch R, Gravdahl D,
70 Gerszten PC, Burton SA, Ozhasoglu C, Welch WC. Radiosurgery Kavanagh B. Observation of a dose-control relationship for lung and
for spinal metastases: clinical experience in 500 cases from a single liver tumors after stereotactic body radiation therapy.
institution. Spine (Phila Pa 1976) 2007; 32: 19399. Int J Radiat Oncol Biol Phys 2009; 73: 11218.
71 Zelefsky MJ, Greco C, Motzer R, et al. Tumor control outcomes 87 Foote M, Letourneau D, Hyde D, et al. Technique for stereotactic
after hypofractionated and single-dose stereotactic image-guided body radiotherapy for spinal metastases. J Clin Neurosci 2011;
intensity-modulated radiotherapy for extracranial metastases from 18: 27679.
renal cell carcinoma. Int J Radiat Oncol Biol Phys 2011; 82: 174478.

e37 www.thelancet.com/oncology Vol 14 January 2013