Sie sind auf Seite 1von 13

The n e w e ng l a n d j o u r na l of m e dic i n e

review article

Medical Progress

Pancreatic Cancer
Manuel Hidalgo, M.D.

eaths from pancreatic ductal adenocarcinoma, also known as From the Centro Nacional de Investi
pancreatic cancer, rank fourth among cancer-related deaths in the United gaciones Oncolgicas and Hospital de
Madrid, Madrid; and Johns Hopkins Uni
States. In 2008, the estimated incidence of pancreatic cancer in the United versity School of Medicine, Baltimore.
States was 37,700 cases, and an estimated 34,300 patients died from the disease.1 Address reprint requests to Dr. Hidalgo
Pancreatic cancer is more common in elderly persons than in younger persons, and at 1650 Orleans St., Rm. 489, Baltimore,
MD 21230, or at
less than 20% of patients present with localized, potentially curable tumors. The
This article (10.1056/NEJMra0901557) was
overall 5-year survival rate among patients with pancreatic cancer is <5%.1,2 updated on July 14, 2010, at
The causes of pancreatic cancer remain unknown. Several environmental fac-
tors have been implicated, but evidence of a causative role exists only for tobacco N Engl J Med 2010;362:1605-17.
Copyright 2010 Massachusetts Medical Society.
use. The risk of pancreatic cancer in smokers is 2.5 to 3.6 times that in nonsmok-
ers; the risk increases with greater tobacco use and longer exposure to smoke.3
Data are limited on the possible roles of moderate intake of alcohol, intake of cof-
fee, and use of aspirin as contributing factors. Some studies have shown an in-
creased incidence of pancreatic cancer among patients with a history of diabetes
or chronic pancreatitis, and there is also evidence, although less conclusive, that
chronic cirrhosis, a high-fat, high-cholesterol diet, and previous cholecystectomy
are associated with an increased incidence.4-7 More recently, an increased risk has
been observed among patients with blood type A, B, or AB as compared with
blood type O.8
Approximately 5 to 10% of patients with pancreatic cancer have a family history
of the disease.9 In some patients, pancreatic cancer develops as part of a well-
defined cancer-predisposing syndrome for which germ-line genetic alterations are
known (see Table 1 in the Supplementary Appendix, available with the full text of
this article at In addition, in some families with an increased risk of
pancreatic cancer, a genetic rather than an environmental cause is suspected. The
risk of pancreatic cancer is 57 times as high in families with four or more affected
members as in families with no affected members.10 The genetic bases for these
associations are not known, although a subgroup of such high-risk kindred carry
germ-line mutations of DNA repair genes such as BRCA2 and the partner and local-
izer of BRCA2 (PALB2).11-13
In recent years, there have been important advances in the understanding of the
molecular biology of pancreatic cancer as well as in diagnosis, staging, and treat-
ment in patients with early-stage tumors. Minimal progress has been made, how-
ever, in prevention, early diagnosis, and treatment in patients with advanced disease.
This review summarizes recent progress in the understanding and management
of pancreatic cancer.

The Biol o gy of Pa ncr e at ic C a ncer

Data suggest that pancreatic cancer results from the successive accumulation of gene
mutations.14 The cancer originates in the ductal epithelium and evolves from pre-
malignant lesions to fully invasive cancer. The lesion called pancreatic intraepithe-

n engl j med 362;17 april 29, 2010 1605

The New England Journal of Medicine
Downloaded from at UNIVERSITY OF OTAGO on November 12, 2016. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.
The n e w e ng l a n d j o u r na l of m e dic i n e

lial neoplasia is the best-characterized histologic

Figure 1 (facing page). Components of Pancreatic Cancer.
precursor of pancreatic cancer.15 The progres-
Pancreatic cancers are composed of several distinct
sion from minimally dysplastic epithelium (pan- elements, including pancreatic-cancer cells, pancreatic-
creatic intraepithelial neoplasia grades 1A and cancer stem cells, and the tumor stroma. A recent
1B) to more severe dysplasia (pancreatic intraepi- analysis of 24 pancreatic cancers suggested that the
thelial neoplasia grades 2 and 3) and finally to mature pancreatic-cancer cell carries on average 63
genetic alterations per cancer; these alterations can be
invasive carcinoma is paralleled by the successive
grouped in 12 core signaling pathways.11 These results,
accumulation of mutations that include activa- if confirmed in larger studies, would indicate that pan
tion of the KRAS2 oncogene, inactivation of the creatic cancer is genetically very complex and hetero
tumor-suppressor gene CDKN2A (which encodes geneous. Thus, effective treatments will probably need
the inhibitor of cyclin-dependent kinase 4 [INK4A]), to attack several targets (with combination regimens)
and may require individualized therapy. A small group
and, last, inactivation of the tumor-suppressor
of cells (5%) appear to have cancer stem-cell features
genes TP53 and deleted in pancreatic cancer 4 that render them capable of asymmetric division, en
(DPC4, also known as the SMAD family member abling them to generate mature cells as well as cancer
4 gene [SMAD4]).16 This sequence of events in stem cells. These stem cells may be identified by the
pancreatic carcinogenesis is supported by studies expression of specific membrane markers and can re
generate into full tumors on implantation in immuno
in genetically engineered mouse models in which
deficient animals. Pancreatic-cancer stem cells are resis
targeted activation of Kras2 with concomitant in- tant to conventional treatment, but they have alterations
activation of Trp53 or Cdkn2A/Ink4A results in the in developmental pathways such as Notch, hedgehog,
development of pancreatic cancer that is identical and wingless in drosophila (Wnt)-catenin that may
to the cognate human disease.17-19 Other prema- result in new therapeutic targets. Pancreatic cancer is
characterized by a dense, poorly vascularized stroma;
lignant lesions of the pancreas, which are less
this microenvironment contains a mixture of interact
well characterized, include intrapancreatic muci- ing cellular and noncellular elements. Autocrine and
nous neoplasia and mucinous cystic neoplasia.20 paracrine secretion of growth factors such as platelet-
Almost all patients with fully established pan- derived growth factor (PDGF) and transforming growth
creatic cancer carry one or more of four genetic factor (TGF-) and cytokines results in continuous
interaction between the stromal and cancer cells. Pan
defects.21 Ninety percent of tumors have activat-
creatic stellate cells are a key cellular element in the
ing mutations in the KRAS2 oncogene. Transcrip- stroma. They are characterized by the expression of
tion of the mutant KRAS gene produces an abnor- desmin, glial fibrillary acidic protein, and intracellular
mal Ras protein that is locked in its activated fat droplets. On stimulation by growth factors, pancre
form, resulting in aberrant activation of prolifera- atic stellate cells express smooth-muscle actin and
produce abundant collagen fibers that contribute to
tive and survival signaling pathways. Likewise,
tumor hypoxia. ALDH+ denotes aldehyde dehydroge
95% of tumors have inactivation of the CDKN2A nase, CTGF connective-tissue growth factor, CXCL-12
gene, with the resultant loss of the p16 protein chemokine 12 ligand, EMSA electrophoretic mobility-
(a regulator of the G1S transition of the cell shift assay, EMT epithelial-to-mesenchymal transition,
cycle) and a corresponding increase in cell pro- FGF fibroblast growth factor, GTPase guanosine triphos
phatase, HGF hepatocyte growth factor, HGF-met hepa
liferation. TP53 is abnormal in 50 to 75% of tu-
tocyte growth factor mesenchymalepithelial transition
mors, permitting cells to bypass DNA damage factor, JNK Jun N-terminal kinase, MMP matrix metallo
control checkpoints and apoptotic signals and proteinase, SPARC secreted protein, acidic, cysteine-rich,
contributing to genomic instability. DPC4 is lost TIMP tissue inhibitor of MMP, TNF- tumor necrosis
in approximately 50% of pancreatic cancers, re- factor , and VEGF vascular endothelial growth factor.
sulting in aberrant signaling by the transform-
ing growth factor (TGF-) cell-surface recep- and the key mutations in each pathway appear
tor. A recent comprehensive genetic analysis of to differ from one tumor to another.
24 pancreatic cancers showed that the genetic A characteristic of pancreatic cancer is the
basis of pancreatic cancer is extremely complex formation of a dense stroma termed a desmo-
and heterogeneous.11 In that study, an average of plastic reaction (Fig. 1).22,23 The pancreatic stel-
63 genetic abnormalities per tumor, mainly point late cells (also known as myofibroblasts) play a
mutations, were classified as likely to be relevant. critical role in the formation and turnover of the
These abnormalities can be organized in 12 func- stroma. On activation by growth factors such as
tional cancer-relevant pathways (Fig. 1). However, TGF1, platelet-derived growth factor (PDGF),
not all tumors have alterations in all pathways, and fibroblast growth factor, these cells secrete

1606 n engl j med 362;17 april 29, 2010

The New England Journal of Medicine

Downloaded from at UNIVERSITY OF OTAGO on November 12, 2016. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.
Medical Progress


Pancreatic-cancer cell Cancer pathways: Selected altered genes:

Apoptosis CASP10, CAD,
DNA damage repair TP53, ERCC4, BRCA2,
Cell-cycle control CDK2NA, APC2,
stem cell
Cell adhesion FAT, PCDH9,
Hedgehog GLI1, GLI3,
Tumor Pancreatic-cancer cell Integrin ILK, LAMA1,
Wnt-catenin MYC, TSC2,
Invasion ADAM11, ADAM19, PRSS23,
Small GTPases PLCB3, RP1

Pancreatic-cancer Cancer stem-cell pathways:

stem cell: CXCR4
Self-renewal Hedgehog
15% of cell population BMI-1
Markers: CD24+, CD44+ Wnt-catenin
TGF- ALDH+ Notch
MMPs Resistant to chemotherapy
Integrins and radiotherapy
Interleukin-1 and
CXCL-12 Stromal cells:
Insulin-like growth Fibroblasts
factor 1 Pancreatic stellate cells
HGF Endothelial cells Stromal pathways:
Immune and inflammatory cells Hedgehog
Adipocytes Nuclear factor
Extracellular matrix: Angiogenesis (VEGF and PDGF)
Collagen types I and III HGF-met
Laminin MMP
Fibronectin FGF
Tumor stroma TIMP

collagen and other components of the extracel- genase-2, PDGF receptor, vascular endothelial
lular matrix; stellate cells also appear to be growth factor, stromal cellderived factor, chemo-
responsible for the poor vascularization that is kines, integrins, SPARC (secreted protein, acidic,
characteristic of pancreatic cancer.24,25 Further- cysteine-rich), and hedgehog pathway elements,
more, stellate cells regulate the reabsorption and among others, which have been associated with
turnover of the stroma, mainly through the a poor prognosis and resistance to treatment.
production of matrix metalloproteinases.26 The However, these proteins may also represent new
stroma is not just a mechanical barrier; rather, it therapeutic targets.27,28
constitutes a dynamic compartment that is criti- The role of angiogenesis in pancreatic cancer
cally involved in the process of tumor formation, remains controversial. Although early data sug-
progression, invasion, and metastasis.22,23 Stromal gested that pancreatic cancer is angiogenesis-
cells express multiple proteins such as cyclooxy- dependent, as are most solid tumors, treatment

n engl j med 362;17 april 29, 2010 1607

The New England Journal of Medicine
Downloaded from at UNIVERSITY OF OTAGO on November 12, 2016. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.
The n e w e ng l a n d j o u r na l of m e dic i n e



Figure 2. Pathological, Radiologic, and Histologic Features of Pancreatic Cancer.

ME: Authors combined Panel D (with arrows and type) is scaled vertically from unmarked
Panel Aorginals.
shows aCurrently,
scaling view of a resected
is proportional tumor affecting
to unmarked version. the head of the pancreas. Panel B shows a contrast-enhanced com
puted tomographic scan from a patient with a T3 pancreatic mass. The tumor invades the splenic superior mesenteric veinportal vein
axis. Panel C shows endoscopic retrograde cholangiopancreatographic imaging of a plastic stent through the ampulla of Vater in a pa
tient with a tumor in the head of the pancreas. Panel D (hematoxylin and eosin) shows microscopical adenocarcinoma of the pancreas
with abundant tumor stroma (black arrows). Smaller
AUTHOR: images show the tumor stroma
Hidalgo at low, medium,
RETAKE: 1st and high magnification. Panel E
shows a peripancreatic lymph node involved with metastatic adenocarcinoma (hematoxylin and 2nd eosin, high magnification). (Courtesy
FIGURE: 2 of 2 3rd
of Emilio de Vicente, M.D., and Elena Garcia, M.D.) Revised
7 col
TYPE: Line Combo 4-C H/T

with angiogenesis inhibitors hasAUTHOR,

failed inPLEASE NOTE: why there is much interest in targeting these
Figure has been redrawn and type has been reset.
with pancreatic cancer. A recent study
Pleasein carefully. specific cells.31,32
a mouse
model showed thatJOB:
the stromal hedge- ISSUE: 4-29-10
hog pathway increases tumor vascularization, Cl inic a l Pr e sen tat ion,
resulting in increased delivery of chemothera- Di agnosis, a nd S taging
peutic agents to pancreatic tumors and greater
efficacy.29 The presenting symptoms of pancreatic cancer
In addition, a subgroup of cancer cells with depend on the location of the tumor within the
cancer stem-cell properties such as tumor initia- gland, as well as on the stage of the disease.
tion have been identified within the tumor.30,31 The majority of tumors develop in the head of the
These cells, which compose just 1 to 5% of the pancreas and cause obstructive cholestasis (Fig.
tumor, are capable of unlimited self-renewal, 2A). Vague abdominal discomfort and nausea are
and through asymmetric division, they give rise also common. More rarely, a pancreatic tumor
to more-differentiated cells (Fig. 1). Pancreatic- may also cause duodenal obstruction or gastroin-
cancer stem cells are resistant to chemotherapy testinal bleeding. Pancreatic cancer often causes
and radiation therapy, which may explain why dull, deep upper abdominal pain that broadly lo-
these treatments do not cure the disease and calizes to the tumor area.

1608 n engl j med 362;17 april 29, 2010

The New England Journal of Medicine

Downloaded from at UNIVERSITY OF OTAGO on November 12, 2016. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.
Medical Progress

Obstruction of the pancreatic duct may lead to tion (Fig. 2C).35 In patients who have large tu-
pancreatitis. Patients with pancreatic cancer often mors, especially in the body and tail of the pan-
have dysglycemia. Indeed, pancreatic cancer should creas, as well as other indications of advanced
be considered in the differential diagnoses of acute disease such as weight loss, an elevated level of
pancreatitis and newly diagnosed diabetes. carbohydrate antigen 19-9 (CA 19-9), ascites, or
At presentation, most patients also have sys- equivocal CT findings, a staging laparoscopy can
temic manifestations of the disease such as asthe- accurately determine metastatic and vascular
nia, anorexia, and weight loss. Other, less com- involvement.36
mon manifestations include deep and superficial There are many potential serum biomarkers
venous thrombosis, panniculitis, liver-function for diagnosis, stratification of a prognosis, and
abnormalities, gastric-outlet obstruction, increased monitoring of therapy.37 CA 19-9 is the only bio-
abdominal girth, and depression. marker with demonstrated clinical usefulness and
Physical examination may reveal jaundice, is useful for therapeutic monitoring and early
temporal wasting, peripheral lymphadenopathy, detection of recurrent disease after treatment in
hepatomegaly, and ascites. Results of routine patients with known pancreatic cancer.37-41 How-
blood tests are generally nonspecific and may ever, CA 19-9 has important limitations. It is not
include mild abnormalities in liver-function tests, a specific biomarker for pancreatic cancer; the
hyperglycemia, and anemia.2,21 level of CA 19-9 may be elevated in other condi-
Evaluation of a patient in whom pancreatic tions such as cholestasis. In addition, patients
cancer is suspected should focus on diagnosis who are negative for Lewis antigen a or b (ap-
and staging of the disease, assessment of resec- proximately 10% of patients with pancreatic can-
tability, and palliation of symptoms. Multiphase, cer) are unable to synthesize CA 19-9 and have
multidetector helical computed tomography (CT) undetectable levels, even in advanced stages of
with intravenous administration of contrast ma- the disease. Although measurement of serum
terial is the imaging procedure of choice for the CA 19-9 levels is useful in patients with known
initial evaluation.33 This technique allows visu- pancreatic cancer, the use of this biomarker as a
alization of the primary tumor in relation to the screening tool has had disappointing results.
superior mesenteric artery, celiac axis, superior Universal primary screening for pancreatic
mesenteric vein, and portal vein and also in cancer is currently not recommended, given the
relation to distant organs (Fig. 2B). In general, tools available and their performance.42 Single-
contrast-enhanced CT is sufficient to confirm a institution studies focusing on surveillance of
suspected pancreatic mass and to frame an initial patients at high risk, such as those with a strong
management plan. Overall, contrast-enhanced CT family history or cancer-predisposition syndromes,
predicts surgical resectability with 80 to 90% have used serial endoscopic ultrasonography and
accuracy.34 Positron-emission tomography can be CT. Pancreatic lesions associated with benign
useful if the CT findings are equivocal. intrapancreatic mucinous neoplasia or pancreatic
Some patients require additional diagnostic intraepithelial neoplasia have been detected in
studies. Endoscopic ultrasonography is useful in approximately 10% of these high-risk patients.
patients in whom pancreatic cancer is suspected However, the cost-effectiveness of this approach
although there is no visible mass identifiable on is unclear, and its use is investigational.43
CT. It is the preferred method of obtaining tis-
sue for diagnostic purposes. Although a tissue S taging of Pa ncr e at ic C a ncer
diagnosis is not needed in patients who are
scheduled for surgery, it is required before the Pancreatic cancer is staged according to the most
initiation of treatment with chemotherapy or recent edition of the American Joint Committee
radiation therapy. Endoscopic retrograde cholan- on Cancer tumornodemetastasis classification,
giopancreatography (ERCP) shows the pancreatic which is based on assessment of resectability by
and bile-duct anatomy and can be used to guide means of helical CT.44 T1, T2, and T3 tumors are
ductal brushing and lavage, which provides tis- potentially resectable, whereas T4 tumors, which
sue for diagnosis. The ERCP technique is espe- involve the superior mesenteric artery or celiac
cially useful in patients with jaundice in whom axis, are unresectable (Table 1). Tumors involv-
an endoscopic stent is required to relieve obstruc- ing the superior mesenteric veins, portal veins, or

n engl j med 362;17 april 29, 2010 1609

The New England Journal of Medicine
Downloaded from at UNIVERSITY OF OTAGO on November 12, 2016. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.
The n e w e ng l a n d j o u r na l of m e dic i n e

Table 1. Staging of Pancreatic Cancer.*

Tumor Nodal Distant Median

Stage Grade Status Metastases Survival Characteristics
IA T1 N0 M0 24.1 Tumor limited to the pancreas, 2 cm in longest dimension
IB T2 N0 M0 20.6 Tumor limited to the pancreas, >2 cm in longest dimension
IIA T3 N0 M0 15.4 Tumor extends beyond the pancreas but does not involve the celiac
axis or superior mesenteric artery
IIB T1, T2, or T3 N1 M0 12.7 Regional lymph-node metastasis
III T4 N0 or N1 M0 10.6 Tumor involves the celiac axis or the superior mesenteric artery
(unresectable disease)
IV T1, T2, T3, or T4 N0 or N1 M1 4.5 Distant metastasis

* N denotes regional lymph nodes, M distant metastases, and T primary tumor.

Data are from Bilimoria et al.45

splenic veins are classified as T3, since these tients who are treated with neoadjuvant therapy
veins can be resected and reconstructed, provid- before resection or who are referred to other
ed that they are patent. centers for treatment.52 Patients with symptoms
of cholangitis require decompression as well as
M a nagemen t of E a r ly Dise a se antibiotic treatment before surgery.
Even if the tumor is fully resected, the out-
Patients with pancreatic cancer are best cared for come in patients with early pancreatic cancer is
by multidisciplinary teams that include surgeons, disappointing. The results of three large ran-
medical and radiation oncologists, radiologists, domized clinical trials, summarized in Table 2 in
gastroenterologists, nutritionists, and pain spe- the Supplementary Appendix, have established
cialists, among others.46,47 For patients with resec- the role of postoperative treatment in patients
table disease, surgery remains the treatment of with resected pancreatic cancer.53-55 The results
choice.48 Depending on the location of the tumor, of the European Study Group for Pancreatic Can-
the operative procedures may involve cephalic cer Trial 1 and Charit Onkologie 1 trial show that
pancreatoduodenectomy (the Whipple procedure), postoperative administration of chemotherapy
distal pancreatectomy, or total pancreatectomy. with either fluorouracil and leucovorin or gem-
A minimum of 12 to 15 lymph nodes should be citabine, a nucleotide analogue commonly used
resected, and every attempt should be made to to treat advanced pancreatic cancer, improves
obtain a tumor-free margin. Data from several progression-free and overall survival. In addi-
randomized clinical trials indicate that a more tion, the Radiation Therapy Oncology Group
extensive resection does not improve survival but trial 97-04 showed that the combination of gem-
increases postoperative morbidity. Recent studies citabine with fluorouracil administered as a
show that the results of vein resection and vascu- continuous infusion and radiation therapy re-
lar reconstruction in patients with limited in- sulted in a trend toward increased overall survival,
volvement of the superior mesenteric vein and although the increase was not significant, among
portal vein are similar to the results in patients patients with tumors in the head of the pancreas.
without vein involvement.49 Poor prognostic fac- These results are similar to those of large single-
tors include lymph-node metastases, a high tumor institution series that incorporated radiation
grade, a large tumor, high levels of CA 19-9, per- therapy.56
sistently elevated postoperative levels of CA 19-9, Notwithstanding differences in patient popu-
and positive margins of resection.38,40,50,51 lations and therapies, the outcome in patients
Up to 70% of patients with pancreatic cancer treated in these trials was similar, with a median
present with biliary obstruction, which can be survival of 20 to 22 months. Large tumor size,
relieved by percutaneous or endoscopic stent high differentiation grade, and involvement of the
placement. Decompression is appropriate for pa- lymph nodes are risk factors for recurrent disease.
tients in whom surgery is delayed, such as pa- The effect of positive resection margins, however,

1610 n engl j med 362;17 april 29, 2010

The New England Journal of Medicine

Downloaded from at UNIVERSITY OF OTAGO on November 12, 2016. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.
Medical Progress

is more controversial.57 Thus, gemcitabine alone patients with advanced disease remains palliative,
or gemcitabine in combination with fluorouracil- although these patients should be offered the
based chemoradiation can be considered the stan- opportunity to participate in clinical trials evalu-
dard of care in this setting. The unequivocal dem- ating new treatments when available. A meta-
onstration that postoperative treatment improves analysis of published findings from clinical trials
the outcome in these patients is one of the most showed an improvement in survival among pa-
important advances that has been made in the tients who received chemotherapy; these findings
management of pancreatic cancer. suggest that active treatment is beneficial.61 For
An emerging strategy in patients with resect- more than a decade, gemcitabine has been the
able pancreatic cancer is the use of preoperative treatment of choice on the basis of the results of
(neoadjuvant) treatment. Nonrandomized, phase the randomized trial of gemcitabine versus fluo-
2 studies suggest that this approach is at least as rouracil, summarized in Table 3 in the Supplemen-
effective as postoperative treatment and may de- tary Appendix.64 Multiple new agents with diverse
crease the rate of local failures and positive resec- mechanisms of action in combination with gem-
tion margins after surgery.58 These findings are citabine have been tested in randomized clinical
particularly relevant for patients who have so- trials, with no improvement in outcome.2,65,66
called borderline-resectable tumors with limited The only agent that, in combination with
vascular involvement; in these patients, preopera- gemcitabine, has shown a small, but statistically
tive treatment may result in tumor-free resection significant improvement in survival among pa-
margins.59 tients with advanced pancreatic cancer is erloti
nib, a small-molecule inhibitor of the epidermal
growth factor receptor (EGFR) (Table 3 in the
M a nagemen t of L o c a l ly
A dva nced a nd S ys temic a l ly Supplementary Appendix).67 As shown in other
A dva nced Dise a se studies of agents targeting the EGFR, patients in
whom drug-induced rashes developed had a better
Approximately 30% of patients with pancreatic outcome. However, the high frequency of KRAS2
cancer receive a diagnosis of advanced loco mutations in pancreatic cancer probably limits
regional disease, and an additional 30% of pa- the benefits of an EGFR inhibitor; this limita-
tients will have local recurrence of tumors after tion is similar to that observed in other cancers
treatment for early disease. The treatment of pa- such as colon cancer. As compared with erlotinib
tients with advanced locoregional disease is palli- alone, the combination of gemcitabine and erlo-
ative; with current treatments, the median over- tinib has more toxicity, particularly gastrointes-
all survival ranges only from 9 to 10 months. tinal symptoms. Together with the rather mod-
Management options range from systemic chemo- est improvement in survival, the toxicity of this
therapy alone to combined forms of treatment combination has limited its wide acceptance as
with chemoradiation therapy and chemotherapy. the standard of care. A recent meta-analysis of
A series of randomized trials conducted over the randomized trials showed that patients with
past two decades established that chemoradia- minimal disease-related symptoms and otherwise
tion therapy was superior to radiation therapy good health may benefit from combination che-
alone in these patients.60,61 The results of more motherapy with gemcitabine and either a plati-
recent studies, summarized in Table 3 in the num agent or a fluoropyrimidine.66,68 Thus, at the
Supplementary Appendix, suggest that chemo- present time, the accepted treatment approach
therapy is indeed the critical component in the for patients with advanced disease is either gem-
treatment approach and that combined treatment citabine given alone or gemcitabine combined
with chemotherapy and chemoradiation therapy with a platinum agent, erlotinib, or a fluoropy
is an effective, though more toxic, approach. How- rimidine.
ever, randomized clinical trials of such combined Once the disease progresses, there is no ac-
treatments have had low enrollment, precluding cepted standard of care; most patients at that
a firm conclusion.60,62,63 point are too sick to receive any other treatment.
The majority of patients with pancreatic can- In a highly selected group of patients with mini-
cer either present with metastatic disease or mally symptomatic disease, second-line chemo-
metastatic disease develops in them, mainly in therapy has modest efficacy, and it can be offered
the liver and peritoneal cavity. The treatment of to patients with good functional reserve (i.e., pa-

n engl j med 362;17 april 29, 2010 1611

The New England Journal of Medicine
Downloaded from at UNIVERSITY OF OTAGO on November 12, 2016. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.
Table 2. Commonly Used Treatment Regimens in Pancreatic Cancer.*

Serious Toxic Effects Occurring

Drug Mechanism of Action Dose in >10% of Patients Reference
First-line therapy
Gemcitabine (2',2'-dFdC) Nucleoside analogue metabolized to 1000 mg/m2 given as 30-min IV infu Neutropenia (26%), elevated alkaline Burris et al.64
triphosphate moiety (dFdC triphos sion either weekly for 7 wk, followed phosphatase level (16%), thrombo

phate) by deoxycytidine kinase is by 1 wk rest, then weekly for 3 of every cytopenia (10%), elevated AST level
incorporated into the nascent DNA 4 wk; or weekly for 3 of every 4 wk (10%)
and blocks DNA replication
Fixed-dose infusion of gemcitabine Prolonged exposure to gemcitabine 1500 mg/m2 given as 10 mg/m2/min IV Neutropenia (49%), thrombocytopenia Tempero et al.71
increases accumulation of dFdC infusion weekly for 3 of every 4 wk (37%), anemia (23%), nausea and
triphosphate vomiting (21%)
Gemcitabine plus oxaliplatin Oxaliplatin is a diaminocyclohexano Gemcitabine: 1000 mg/m2 given as Neutropenia (20%), peripheral sensory Louvet et al.72
platinum analogue that binds 10 mg/m2/min IV infusion on day 1 neuropathy (19%), thrombocytope
and alkylates DNA every other wk; plus oxaliplatin: nia (14%), nausea (10%)
100 mg/m2 given as 120-min IV
infusion on day 2, every other wk
Gemcitabine plus cisplatin Cisplatin is a DNA-binding alkylating Gemcitabine: 1000 mg/m2 given as 30- Nausea and vomiting (22%), anemia Heinemann et al.73
n e w e ng l a n d j o u r na l

agent min IV infusion every other wk; cis (13%), pain (12%), leukopenia

The New England Journal of Medicine


platin: 50 mg/m2 given as IV infu (10%)

sion every other wk
Gemcitabine plus capecitabine Capecitabine is converted in the tumor Gemcitabine: 1000 mg/m2 given as 30- Neutropenia (23%) Bernhard et al.74

n engl j med 362;17 april 29, 2010

to fluorouracil and inhibits thymidi min IV infusion weekly for 3 of every
late synthetase 4 wk; plus capecitabine: 1300 mg/m2

Copyright 2010 Massachusetts Medical Society. All rights reserved.

daily, orally for 14 days every 34 wk,
m e dic i n e

divided in two daily doses

Gemcitabine plus erlotinib Erlotinib is a small-molecule inhibitor Gemcitabine: 1000 mg/m2 given as Neutropenia (24%), infection (17%), Moore et al.67
of the epidermal growth factor 30-min IV infusion, either weekly fatigue (15%), elevated AST level
receptor for 7 wk followed by 1 wk rest, then (11%), thrombocytopenia (10%)
weekly for 3 of every 4 wk, or weekly
for 3 of every 4 wk; plus erlotinib:
100 mg/day orally daily

Downloaded from at UNIVERSITY OF OTAGO on November 12, 2016. For personal use only. No other uses without permission.
Second-line therapy
Modified FOLFOX (combination Fluorouracil is an inhibitor of thymidilate Oxaliplatin: 85 mg/m2 given as 120-min Neutropenia (20%), asthenia (13%), Li and Saif75
regimen including oxaliplatin, synthetase, and leucovorin potenti IV infusion every other wk; leucovor vomiting (10%)
fluorouracil, and leucovorin) ates the inhibition of thymidilate in: 400 mg/m2 on day 1, every other
synthetase by fluorouracil wk; and fluorouracil: 2000 mg/m2
given as 46-hr infusion every
other wk
Modified FOLFIRI.3 (combination Fluorouracil is an inhibitor of thymidilate Irinotecan: 70 mg/m2 given as 60-min Neutropenia (22%), vomiting (10%) Li and Saif75
regimen of irinotecan with synthetase, and irinotecan is a topo IV infusion on day 1; leucovorin: 400
fluorouracil and leucovorin) isomerase I inhibitor; leucovorin po mg/m given as given as 120-min IV
tentiates the inhibition of thymidilate infusion on day 1; and fluorouracil:
synthetase by fluorouracil 2000 mg/m2 given as 46-hr IV infu
sion on day 1; and irinotecan: 70 mg/
mg2 given as 60-min IV infusion at
the end of the infusion of fluoroura
cil, every 2 wk
Oxaliplatin plus capecitabine Oxaliplatin is a diaminocyclohexano Oxaliplatin: 130 mg/m2 given as 120- Oxaliplatin: fatigue (13%), diarrhea Xiong et al.69
platinum analogue that binds and al min IV infusion every 3 wk; and (5%), vomiting (3%); capecitabine:
Medical Progress

kylates DNA; capecitabine is convert capecitabine: 2000 mg/m2 daily, handfoot syndrome (3%), abdomi
ed in the tumor to fluorouracil and orally for 14 days every 34 wk, nal pain (3%)
inhibits thymidilate synthetase divided in two daily doses

n engl j med 362;17 april 29, 2010

Capecitabine plus erlotinib Capecitabine is converted in the tumor Capecitabine: 2000 mg/m2 daily, orally Capecitabine: diarrhea (17%), rash (13%), Kulke et al.70

The New England Journal of Medicine

to fluorouracil and inhibits thymidi for 14 days every 34 wk, divided in handfoot syndrome (13%), mucosi
late synthetase; erlotinib is a small- two daily doses; and erlotinib: 150 mg tis (10%); erlotinib: fatigue (3%), ele
molecular inhibitor of the epidermal orally daily vated bilirubin level (3%), elevated
growth factor receptor alkaline phosphatase level (3%)

Copyright 2010 Massachusetts Medical Society. All rights reserved.

* AST denotes aspartate aminotransferase; dFdC difluorodeoxycytidine; FOLFIRI.3 regimen of fluorouracil, leucovorin, and irinotecan; FOLFOX regimen of folinic acid, fluorouracil, and
oxaliplatin; and IV intravenous.

Downloaded from at UNIVERSITY OF OTAGO on November 12, 2016. For personal use only. No other uses without permission.
Table 3. Selected Strategic Targets in Pancreatic Cancer.*

Target Agent Drug Class Mechanism of Action Phase Reference
SPARC Nanoparticle albumin- Cytotoxic agent SPARC, expressed in cancer cells and stroma in the pancreas, binds nanoparticle 3 Li and Saif75
bound paclitaxel albumin-bound paclitaxel, increasing local drug delivery

IGF-IR MK 0646, AMG 479, R1507 Monoclonal antibody Inhibits ligand binding activation of the IGF-IR and cell proliferation 3 Hewish et al.78
Death receptor AMG 655, CS1008 Monoclonal antibody Agonist antibodies to membrane death receptors induce apoptosis 2 Li and Saif,75
Derosier et al.79
Mucin-1 90Y-hPAM4 Radioimmunoconjugate Targets mucin-1 expressed in pancreatic-cancer cells and delivers radiation load 12 Gold et al.80
Hedgehog GDC-0449, IPI-926 Small-molecule inhibitor Inhibits smoothened receptor, resulting in inhibition of cell proliferation; targets 1 Olive et al.,29 Jimeno
pathway cancer stroma and cancer stem cells in the pancreas et al.32
c-kit, PDGFR, Masitinib Small-molecule inhibitor Multikinase inhibitor targets c-kit, PDGFR, and FGFR3 and affects the FAK 3 Li and Saif75
FGFR pathway; masitinib was shown to enhance the antiproliferative effects of
gemcitabine in preclinical studies
MEK AZD6244 Small-molecule inhibitor Targets and inhibits MEK, decreasing cell proliferation 2 Chung et al.81
n e w e ng l a n d j o u r na l

The New England Journal of Medicine

Src AZD0530, dasatinib Small-molecule inhibitor Targets and inhibits Src kinase, resulting in inhibition of cell proliferation and 2 Rajeshkumar et al.77

RAS Sarilasib Small-molecule inhibitor Dislodges all forms of RAS from the plasma membrane, inhibiting RAS signaling 2 Haklai et al.82

n engl j med 362;17 april 29, 2010

PSCA AGS-1C4D4 Monoclonal antibody Binds membrane PSCA; specific mechanisms of cell killing undetermined 2 Wente et al.83

Copyright 2010 Massachusetts Medical Society. All rights reserved.

Mesothelin MORAb-009 Monoclonal antibody Binds membrane mesothelin; specific mechanisms of cell killing undetermined 2 Hassan et al.84
m e dic i n e

TNF- TNFerade Gene therapy Adenoviral gene therapy increases intratumoral concentration of TNF- 3 Murugesan et al.85

* The abbreviation c-kit denotes stem-cell factor receptor; FAK focal adhesion kinase; FGFR fibroblast growth factor receptor; IGF-IR type I insulin-like growth factor receptor; MEK mitogen-
activated protein kinaseextracellular-signalregulated kinase, PDGFR platelet-derived growth factor; PSCA prostate stem-cell antigen; SPARC secreted protein, acidic, cysteine-rich; and
TNF- tumor necrosis factor .

Downloaded from at UNIVERSITY OF OTAGO on November 12, 2016. For personal use only. No other uses without permission.
Medical Progress

tients who are ambulatory and minimally symp- that the tumor microenvironment and cancer
tomatic).69,70 Table 2 lists commonly used first-line stem cells are critical components of pancreatic
and second-line therapeutic regimens. cancer has led to the development of agents, such
as the hedgehog inhibitors, that target these
F u t ur e Dir ec t ions components.23,29,31,32 The availability of preclini-
cal models that recapitulate the complexity of
There is much room for improvement in all as- this disease will probably help in establishing
pects of treatment for pancreatic cancer. Screen- priorities and strategies for the development of
ing of high-risk persons by means of either in- new treatments.77,86 The complexity of the genome
novative imaging methods or measurements of of pancreatic cancer indicates that it is a hetero-
serum biomarkers for early diagnosis is criti- geneous cancer and that methods to individual-
cal.42,43,76 A better understanding of the biology ize therapy will be required.11,87
of pancreatic cancer is opening new avenues for
Dr. Hidalgo reports receiving grant support from Roche and
treatment, and an increasing number of new tar- Daichi; being the principal investigator and receiving grant sup-
geted agents are in clinical development (Table 3). port for clinical studies of nanoparticle albumin-bound paclitaxel
These agents include small-molecule inhibitors of from American Biosciences, of erlotinib from Roche, of AGS-
1C4D4 from Agensys, and of MORAb-009 from Eisai; and receiv-
oncogenes and signaling pathways such as RAS, ing consulting fees from American Biosciences, OSIGenentech,
Src, and MEK, monoclonal antibodies targeting Merck, and Agensys. No other potential conflict of interest rele-
cell-membrane proteins such as mesothelin and vant to this article was reported.
I thank Wells Messersmith and Anirban Maitra for critical
the so-called death receptors, and new nanotech- comments on an earlier version of the manuscript and Sofia
nology and adenoviral agents. The recognition Perea for editorial assistance.

1. Jemal A, Siegel R, Ward E, et al. Can- atives of patients with familial pancreatic pathway constrain progression of pancre-
cer statistics, 2008. CA Cancer J Clin cancer. Clin Cancer Res 2001;7:738-44. atic adenocarcinoma in the mouse. Proc
2008;58:71-96. 11. Jones S, Zhang X, Parsons DW, et al. Natl Acad Sci U S A 2006;103:5947-52.
2. Li D, Xie K, Wolff R, Abbruzzese JL. Core signaling pathways in human pan- 20. Takaori K. Current understanding of
Pancreatic cancer. Lancet 2004;363:1049- creatic cancers revealed by global genom- precursors to pancreatic cancer. J Hepato-
57. ic analyses. Science 2008;321:1801-6. biliary Pancreat Surg 2007;14:217-23.
3. Hassan MM, Bondy ML, Wolff RA, et al. 12. Couch FJ, Johnson MR, Rabe KG, et al. 21. Maitra A, Hruban RH. Pancreatic can-
Risk factors for pancreatic cancer: case- The prevalence of BRCA2 mutations in cer. Annu Rev Pathol 2008;3:157-88.
control study. Am J Gastroenterol 2007; familial pancreatic cancer. Cancer Epide- 22. Chu GC, Kimmelman AC, Hezel AF,
102:2696-707. miol Biomarkers Prev 2007;16:342-6. DePinho RA. Stromal biology of pancre-
4. Batty GD, Kivimaki M, Morrison D, 13. Jones S, Hruban RH, Kamiyama M, et atic cancer. J Cell Biochem 2007;101:887-
et al. Risk factors for pancreatic cancer al. Exomic sequencing identifies PALB2 907.
mortality: extended follow-up of the orig- as a pancreatic cancer susceptibility gene. 23. Mahadevan D, Von Hoff DD. Tumor-
inal Whitehall Study. Cancer Epidemiol Science 2009;324:217. stroma interactions in pancreatic ductal
Biomarkers Prev 2009;18:673-5. 14. Vogelstein B, Kinzler KW. Cancer adenocarcinoma. Mol Cancer Ther 2007;
5. Landi S. Genetic predisposition and genes and the pathways they control. Nat 6:1186-97.
environmental risk factors to pancreatic Med 2004;10:789-99. 24. Masamune A, Shimosegawa T. Signal
cancer: a review of the literature. Mutat 15. Hruban RH, Maitra A, Goggins M. transduction in pancreatic stellate cells.
Res 2009;681:299-307. Update on pancreatic intraepithelial neo- J Gastroenterol 2009;44:249-60.
6. Lowenfels AB, Maisonneuve P. Epide- plasia. Int J Clin Exp Pathol 2008;1:306- 25. Erkan M, Reiser-Erkan C, Michalski
miology and risk factors for pancreatic 16. CW, et al. Cancer-stellate cell interactions
cancer. Best Pract Res Clin Gastroenterol 16. Feldmann G, Beaty R, Hruban RH, perpetuate the hypoxia-fibrosis cycle in
2006;20:197-209. Maitra A. Molecular genetics of pancre- pancreatic ductal adenocarcinoma. Neo-
7. Genkinger JM, Spiegelman D, Ander- atic intraepithelial neoplasia. J Hepatobil- plasia 2009;11:497-508.
son KE, et al. Alcohol intake and pancre- iary Pancreat Surg 2007;14:224-32. 26. Zhang W, Erkan M, Abiatari I, et al.
atic cancer risk: a pooled analysis of four- 17. Hingorani SR, Wang L, Multani AS, et Expression of extracellular matrix metal-
teen cohort studies. Cancer Epidemiol al. Trp53R172H and KrasG12D cooperate loproteinase inducer (EMMPRIN/CD147)
Biomarkers Prev 2009;18:765-76. to promote chromosomal instability and in pancreatic neoplasm and pancreatic
8. Wolpin BM, Chan AT, Hartge P, et al. widely metastatic pancreatic ductal ade- stellate cells. Cancer Biol Ther 2007;6:218-
ABO blood group and the risk of pancre- nocarcinoma in mice. Cancer Cell 2005; 27.
atic cancer. J Natl Cancer Inst 2009;101: 7:469-83. 27. Mukherjee P, Basu GD, Tinder TL, et al.
424-31. 18. Guerra C, Schuhmacher AJ, Caamero Progression of pancreatic adenocarcino-
9. Shi C, Hruban RH, Klein AP. Familial M, et al. Chronic pancreatitis is essential ma is significantly impeded with a com-
pancreatic cancer. Arch Pathol Lab Med for induction of pancreatic ductal adeno- bination of vaccine and COX-2 inhibition.
2009;133:365-74. carcinoma by K-Ras oncogenes in adult J Immunol 2009;182:216-24.
10. Tersmette AC, Petersen GM, Offer- mice. Cancer Cell 2007;11:291-302. 28. Infante JR, Matsubayashi H, Sato N, et
haus GJ, et al. Increased risk of incident 19. Bardeesy N, Aguirre AJ, Chu GC, et al. al. Peritumoral fibroblast SPARC expres-
pancreatic cancer among first-degree rel- Both p16(Ink4a) and the p19(Arf )-p53 sion and patient outcome with resectable

n engl j med 362;17 april 29, 2010 1615

The New England Journal of Medicine
Downloaded from at UNIVERSITY OF OTAGO on November 12, 2016. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.
The n e w e ng l a n d j o u r na l of m e dic i n e

pancreatic adenocarcinoma. J Clin Oncol families for pancreatic cancer. Pancrea- 56. Herman JM, Swartz MJ, Hsu CC, et al.
2007;25:319-25. tology 2009;9:215-22. Analysis of fluorouracil-based adjuvant
29. Olive KP, Jacobetz MA, Davidson CJ, 43. Larghi A, Verna EC, Lecca PG, Costa- chemotherapy and radiation after pancrea-
et al. Inhibition of Hedgehog signaling magna G. Screening for pancreatic cancer ticoduodenectomy for ductal adenocarci-
enhances delivery of chemotherapy in a in high-risk individuals: a call for endo- noma of the pancreas: results of a large,
mouse model of pancreatic cancer. Science scopic ultrasound. Clin Cancer Res 2009; prospectively collected database at the
2009;324:1457-61. 15:1907-14. Johns Hopkins Hospital. J Clin Oncol
30. Li C, Heidt DG, Dalerba P, et al. Iden- 44. Edge SB, Byrd DR, Compton CC, Fritz 2008;26:3503-10.
tification of pancreatic cancer stem cells. AG, Greene FL, Trotti A, eds. AJCC can- 57. Raut CP, Tseng JF, Sun CC, et al. Im-
Cancer Res 2007;67:1030-7. cer staging manual. 7th ed. New York: pact of resection status on pattern of fail-
31. Hermann PC, Huber SL, Herrler T, Springer, 2010. ure and survival after pancreaticoduo-
et al. Distinct populations of cancer stem 45. Bilimoria KY, Bentrem DJ, Ko CY, et al. denectomy for pancreatic adenocarcinoma.
cells determine tumor growth and meta- Validation of the 6th edition AJCC Pancre- Ann Surg 2007;246:52-60.
static activity in human pancreatic cancer. atic Cancer Staging System: report from 58. Evans DB, Varadhachary GR, Crane
Cell Stem Cell 2007;1:313-23. the National Cancer Database. Cancer CH, et al. Preoperative gemcitabine-based
32. Jimeno A, Feldmann G, Surez-Gau- 2007;110:738-44. chemoradiation for patients with resect-
thier A, et al. A direct pancreatic cancer 46. Pawlik TM, Laheru D, Hruban RH, et able adenocarcinoma of the pancreatic
xenograft model as a platform for cancer al. Evaluating the impact of a single-day head. J Clin Oncol 2008;26:3496-502.
stem cell therapeutic development. Mol multidisciplinary clinic on the manage- 59. Katz MH, Pisters PW, Evans DB, et al.
Cancer Ther 2009;8:310-4. ment of pancreatic cancer. Ann Surg On- Borderline resectable pancreatic cancer:
33. Miura F, Takada T, Amano H, Yoshida col 2008;15:2081-8. the importance of this emerging stage of
M, Furui S, Takeshita K. Diagnosis of pan- 47. Katz MH, Wang H, Fleming JB, et al. disease. J Am Coll Surg 2008;206:833-46.
creatic cancer. HPB (Oxford) 2006;8:337- Long-term survival after multidisciplinary 60. Huguet F, Andr T, Hammel P, et al.
42. management of resected pancreatic adeno- Impact of chemoradiotherapy after dis-
34. Karmazanovsky G, Fedorov V, Kubysh- carcinoma. Ann Surg Oncol 2009;16:836- ease control with chemotherapy in locally
kin V, Kotchatkov A. Pancreatic head can- 47. advanced pancreatic adenocarcinoma in
cer: accuracy of CT in determination of 48. Shaib Y, Davila J, Naumann C, El-Serag GERCOR phase II and III studies. J Clin
resectability. Abdom Imaging 2005;30: H. The impact of curative intent surgery Oncol 2007;25:326-31.
488-500. on the survival of pancreatic cancer pa- 61. Sultana A, Tudur Smith C, Cunning-
35. Dumonceau JM, Vonlaufen A. Pancre- tients: a U.S. population-based study. Am ham D, et al. Systematic review, including
atic endoscopic retrograde cholangiopan- J Gastroenterol 2007;102:1377-82. meta-analyses, on the management of
creatography (ERCP). Endoscopy 2007;39: 49. Tseng JF, Tamm EP, Lee JE, Pisters PW, locally advanced pancreatic cancer using
124-30. Evans DB. Venous resection in pancreatic radiation/combined modality therapy. Br J
36. Mayo SC, Austin DF, Sheppard BC, cancer surgery. Best Pract Res Clin Gas- Cancer 2007;96:1183-90.
Mori M, Shipley DK, Billingsley KG. Evolv- troenterol 2006;20:349-64. 62. Loehrer PJ, Powell ME, Cardenes HR,
ing preoperative evaluation of patients 50. Hernandez JM, Cowgill SM, Al-Saadi et al. A randomized phase III study of
with pancreatic cancer: does laparoscopy S, et al. CA 19-9 velocity predicts disease- gemcitabine in combination with radia-
have a role in the current era? J Am Coll free survival and overall survival after tion therapy versus gemcitabine alone in
Surg 2009;208:87-95. pancreatectomy of curative intent. J Gas- patients with localized, unresectable pan-
37. Harsha HC, Kandasamy K, Rangana- trointest Surg 2009;13:349-53. creatic cancer: E4201. J Clin Oncol 2008;
than P, et al. A compendium of potential 51. Slidell MB, Chang DC, Cameron JL, et 26:Suppl:4506. abstract.
biomarkers of pancreatic cancer. PLoS al. Impact of total lymph node count and 63. Chauffert B, Mornex F, Bonnetain F,
Med 2009;6(4):e1000046. lymph node ratio on staging and survival et al. Phase III trial comparing intensive
38. Berger AC, Garcia M Jr, Hoffman JP, after pancreatectomy for pancreatic ade- induction chemoradiotherapy (60 Gy, in-
et al. Postresection CA 19-9 predicts over- nocarcinoma: a large, population-based fusional 5-FU and intermittent cisplatin)
all survival in patients with pancreatic analysis. Ann Surg Oncol 2008;15:165-74. followed by maintenance gemcitabine with
cancer treated with adjuvant chemoradia- 52. Pausawasdi N, Scheiman J. Endoscop- gemcitabine alone for locally advanced
tion: a prospective validation by RTOG ic evaluation and palliation of pancreatic unresectable pancreatic cancer: definitive
9704. J Clin Oncol 2008;26:5918-22. adenocarcinoma: current and future op- results of the 2000-01 FFCD/SFRO study.
39. Hess V, Glimelius B, Grawe P, et al. tions. Curr Opin Gastroenterol 2007;23: Ann Oncol 2008;19:1592-9.
CA 19-9 tumour-marker response to che- 515-21. 64. Burris HA III, Moore MJ, Andersen J,
motherapy in patients with advanced pan- 53. Neoptolemos JP, Stocken DD, Friess H, et al. Improvements in survival and clini-
creatic cancer enrolled in a randomised et al. A randomized trial of chemoradio- cal benefit with gemcitabine as first-line
controlled trial. Lancet Oncol 2008;9: therapy and chemotherapy after resection therapy for patients with advanced pan-
132-8. of pancreatic cancer. N Engl J Med 2004; creas cancer: a randomized trial. J Clin
40. Ferrone CR, Finkelstein DM, Thayer 350:1200-10. [Erratum, N Engl J Med Oncol 1997;15:2403-13.
SP, Muzikansky A, Fernandez-del Castillo 2004;351:726.] 65. Sultana A, Smith CT, Cunningham D,
C, Warshaw AL. Perioperative CA19-9 levels 54. Oettle H, Post S, Neuhaus P, et al. Ad- Starling N, Neoptolemos JP, Ghaneh P.
can predict stage and survival in patients juvant chemotherapy with gemcitabine vs Meta-analyses of chemotherapy for locally
with resectable pancreatic adenocarcino- observation in patients undergoing cura- advanced and metastatic pancreatic can-
ma. J Clin Oncol 2006;24:2897-902. tive-intent resection of pancreatic cancer: cer. J Clin Oncol 2007;25:2607-15.
41. Ko AH, Hwang J, Venook AP, Abbruz a randomized controlled trial. JAMA 2007; 66. Heinemann V, Boeck S, Hinke A, La-
zese JL, Bergsland EK, Tempero MA. Se- 297:267-77. bianca R, Louvet C. Meta-analysis of ran-
rum CA19-9 response as a surrogate for 55. Regine WF, Winter KA, Abrams RA, domized trials: evaluation of benefit from
clinical outcome in patients receiving et al. Fluorouracil vs gemcitabine chemo- gemcitabine-based combination chemo-
fixed-dose rate gemcitabine for advanced therapy before and after fluorouracil- therapy applied in advanced pancreatic
pancreatic cancer. Br J Cancer 2005;93: based chemoradiation following resection cancer. BMC Cancer 2008;8:82.
195-9. of pancreatic adenocarcinoma: a random- 67. Moore MJ, Goldstein D, Hamm J, et al.
42. Greenhalf W, Grocock C, Harcus M, ized controlled trial. JAMA 2008;299:1019- Erlotinib plus gemcitabine compared with
Neoptolemos J. Screening of high-risk 26. [Erratum, JAMA 2008;299:1902.] gemcitabine alone in patients with ad-

1616 n engl j med 362;17 april 29, 2010

The New England Journal of Medicine

Downloaded from at UNIVERSITY OF OTAGO on November 12, 2016. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.
Medical Progress

vanced pancreatic cancer: a phase III trial 74. Bernhard J, Dietrich D, Scheithauer W, atic cancer. Clin Cancer Res 2003;9:3929S-
of the National Cancer Institute of Canada et al. Clinical benefit and quality of life in 3937S.
Clinical Trials Group. J Clin Oncol 2007; patients with advanced pancreatic cancer 81. Chung EJ, Brown AP, Asano H, et al.
25:1960-6. receiving gemcitabine plus capecitabine In vitro and in vivo radiosensitization with
68. Cunningham D, Chau I, Stocken DD, versus gemcitabine alone: a randomized AZD6244 (ARRY-142886), an inhibitor of
et al. Phase III randomized comparison multicenter phase III clinical trial mitogen-activated protein kinase/extracel-
of gemcitabine versus gemcitabine plus SAKK 44/00-CECOG/PAN.1.3.001. J Clin lular signal-regulated kinase 1/2 kinase.
capecitabine in patients with advanced Oncol 2008;26:3695-701. Clin Cancer Res 2009;15:3050-7.
pancreatic cancer. J Clin Oncol 2009;27: 75. Li J, Saif MW. Any progress in the 82. Haklai R, Elad-Sfadia G, Egozi Y,
5513-8. management of advanced pancreatic can- Kloog Y. Orally administered FTS (salira-
69. Xiong HQ, Varadhachary GR, Blais JC, cer? Highlights from the 45th ASCO an- sib) inhibits human pancreatic tumor
Hess KR, Abbruzzese JL, Wolff RA. Phase nual meeting, Orlando, FL, USA, May 29- growth in nude mice. Cancer Chemother
2 trial of oxaliplatin plus capecitabine June 2, 2009. JOP 2009;10:361-5. Pharmacol 2008;61:89-96.
(XELOX) as second-line therapy for pa- 76. Faca VM, Song KS, Wang H, et al. 83. Wente MN, Jain A, Kono E, et al. Pros-
tients with advanced pancreatic cancer. A mouse to human search for plasma pro- tate stem cell antigen is a putative target
Cancer 2008;113:2046-52. teome changes associated with pancreatic for immunotherapy in pancreatic cancer.
70. Kulke MH, Blaszkowsky LS, Ryan DP, tumor development. PLoS Med 2008;5(6): Pancreas 2005;31:119-25.
et al. Capecitabine plus erlotinib in gem- e123. 84. Hassan R, Ebel W, Routhier EL, et al.
citabine-refractory advanced pancreatic 77. Rajeshkumar NV, Tan AC, De Oliveira Preclinical evaluation of MORAb-009, a
cancer. J Clin Oncol 2007;25:4787-92. E, et al. Antitumor effects and biomarkers chimeric antibody targeting tumor-asso-
71. Tempero M, Plunkett W, Ruiz Van Ha- of activity of AZD0530, a Src inhibitor, in ciated mesothelin. Cancer Immun 2007;
peren V, et al. Randomized phase II com- pancreatic cancer. Clin Cancer Res 2009; 7:20.
parison of dose-intense gemcitabine: 15:4138-46. 85. Murugesan SR, King CR, Osborn R, et
thirty-minute infusion and fixed dose 78. Hewish M, Chau I, Cunningham D. al. Combination of human tumor necro-
rate infusion in patients with pancreatic Insulin-like growth factor 1 receptor tar- sis factor-alpha (hTNF-alpha) gene deliv-
adenocarcinoma. J Clin Oncol 2003;21: geted therapeutics: novel compounds and ery with gemcitabine is effective in mod-
3402-8. novel treatment strategies for cancer medi- els of pancreatic cancer. Cancer Gene Ther
72. Louvet C, Labianca R, Hammel P, et al. cine. Recent Pat Anticancer Drug Discov 2009;16:841-7.
Gemcitabine in combination with oxali 2009;4:54-72. 86. Rubio-Viqueira B, Hidalgo M. Direct in
platin compared with gemcitabine alone in 79. Derosier LC, Vickers SM, Zinn KR, et al. vivo xenograft tumor model for predicting
locally advanced or metastatic pancreatic TRA-8 anti-DR5 monoclonal antibody and chemotherapeutic drug response in can-
cancer: results of a GERCOR and GISCAD gemcitabine induce apoptosis and inhibit cer patients. Clin Pharmacol Ther 2009;85:
phase III trial. J Clin Oncol 2005;23:3509- radiologically validated orthotopic pan- 217-21.
16. creatic tumor growth. Mol Cancer Ther 87. Farrell JJ, Elsaleh H, Garcia M, et al.
73. Heinemann V, Quietzsch D, Gieseler F, 2007;6:3198-207. Human equilibrative nucleoside trans-
et al. Randomized phase III trial of gem- 80. Gold DV, Schutsky K, Modrak D, Car- porter 1 levels predict response to gemcit-
citabine plus cisplatin compared with dillo TM. Low-dose radioimmunotherapy abine in patients with pancreatic cancer.
gemcitabine alone in advanced pancreatic ((90)Y-PAM4) combined with gemcitabine Gastroenterology 2009;136:187-95.
cancer. J Clin Oncol 2006;24:3946-52. for the treatment of experimental pancre- Copyright 2010 Massachusetts Medical Society.

images in clinical medicine

The Journal welcomes consideration of new submissions for Images in Clinical
Medicine. Instructions for authors and procedures for submissions can be found
on the Journals Web site at At the discretion of the editor, images that
are accepted for publication may appear in the print version of the Journal,
the electronic version, or both.

n engl j med 362;17 april 29, 2010 1617

The New England Journal of Medicine
Downloaded from at UNIVERSITY OF OTAGO on November 12, 2016. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.