Neoadjuvant Therapy for Resectable Pancreatic Cancer

Feb 25, 2014



By Sabha Ganai, MD, PhD
Southern Illinois University School ofMedicine

Pancreatic adenocarcinoma is currentlythe fourth leading cause ofcancer death in the United States, withfive-year survival rates for patientswith localized disease remaining under25%.1 Despite improvements in morbidityand mortality from pancreassurgery, as well as a greater focuson preoperative imaging and stagingprior to resection, long-term outcomesremain disappointing. While pancreaticresection has traditionally beenthought to be the best opportunityfor cure and is primarily determinedby anatomic considerations, patientoutcomes from resectable pancreaticadenocarcinoma may ultimately bedictated by tumor biology.

The concept that pancreatic cancer isa systemic disease at presentation isa sensible assumption in light of highdistant-recurrence rates seen after surgery.Such assumptions can be furthersupported by phase III evidence showinga significant benefit in disease-freesurvival after the adjuvant administrationof gemcitabine in patients whounderwent R0 and R1 resections forcurative intent.2

Among surgeons, there is increasingconsensus on the importanceof multimodality therapy in the preoperativesetting for patients whohave "borderline-resectable" disease(venous involvement without celiac orsuperior mesenteric arterial abutment)for pragmatic reasons, as this maypotentially impact technical complexity,margin-positive resection rates,and the need for concurrent vascularreconstruction. However, there isgreater debate regarding the timingof multimodality therapy for "resectable"disease, as defined by a lack ofvascular involvement on preoperativeaxial imaging. Recent analysis ofsingle-institutional data has suggestedthat the administration of multimodalitytherapy in the adjuvant settingmay be prevented by postoperativecomplications and the development ofearly metastases, favoring a neoadjuvantapproach as a method to ensurecompletion of therapy.3

In this Current Controversies in Oncologycolumn, Dr. Christopher Crane, ofMD Anderson Cancer Center, provideshis viewpoint as a radiation oncologiston neoadjuvant therapy for resectableand borderline-resectable pancreaticcancer and the particular roleof chemoradiation. Subsequently, Dr.Margaret Tempero, of the University ofCalifornia, San Francisco, discusses her perspective as a medical oncologist onthe timing of therapy for resectabledisease. Certainly, as there is equipoiseon this controversial subject, neoadjuvantclinical trials are justified, but theywill also require increased participationby hepatopancreaticobiliary surgeonswith focused efforts to initiate multidisciplinarydiscussions prior to attemptingresection.

Dr. Ganai is an Assistant Professor ofSurgery at the Southern Illinois UniversitySchool of Medicine and is the Director ofGastrointestinal Oncology at the SimmonsCancer Institute. An ASCO member since2011, she has participated in the ASCOUniversity® Fellows Committee and serveson the Integrated Media and TechnologyCommittee and the ASCO ConnectionEditorial Board.


  1. Siegel R, Ma J, Zou Z, et al. CA Cancer JClin. 2014; epub Jan 7. PMID: 24399786.
  2. Oettle H, Post S, Neuhaus P, et al. JAMA.2007;297:267-77. PMID: 17227978.
  3. Tzeng CW, Cao HS, Lee JE, et al. J GastrointestSurg. 2014;18:16-25. PMID: 24241967.


Role of Neoadjuvant Therapy for Resectable and Borderline-Resectable Pancreatic Cancer


By Christopher H. Crane, MD
The University of Texas MD AndersonCancer Center

The use of neoadjuvant therapy(NAT) has many unique advantagesin patients with resectable pancreaticcancer (PDAC) owing to its uniquelyaggressive biology and the typicallypoor performance status of patientswith pancreatic cancer. The mostappealing aspect is the introduction ofa time period to observe tumor biology,response to therapy, and patients'performance status, which leads toimproved selection for pancreatic surgery—an operation associated withsignificant patient morbidity even whenperformed by experienced surgeons.

A second appealing rationale is theimmediate use of systemic therapy fora disease that is systemic at diagnosisin the majority of patients. Since up toone-half of patients will not be candidatesfor timely postoperative adjuvanttherapy after pancreaticoduodenectomy,the use of NAT also offers theability to treat virtually all patients atpresentation with therapy that immediatelyaddresses the limitations ofcurability—micrometastatic disease aswell as margin-negative resectability.

Finally, pancreatic surgery appears tobe safer following preoperative chemoradiationdue to a reduced risk ofpancreatic anastomotic leak becauseof the pancreatic fibrosis induced bypreoperative chemoradiation.1

Most of the experience with NAT hasincluded chemoradiation as a componentof therapy, but perhaps theclearest indication for chemoradiationin patients with PDAC is in borderline-resectable(BRPC) disease based onpreoperative imaging as defined by theAmericas Hepato-Pancreato-BiliaryAssociation (AHPBA)/Society for Surgeryof the Alimentary Tract (SSAT)/Society of Surgical Oncology (SSO),and subsequently adopted with modificationby the National ComprehensiveCancer Network (NCCN).2

The use of postoperative chemoradiationis controversial in patients whohave undergone margin-negative pancreaticoduodenectomy.However, thehigh frequency of close and positivemarginresections with upfront surgeryhas led many institutions to investigatethe role of chemoradiation given preoperatively,and this strategy is becomingmore common in the off-protocolsetting.

The rationale for the use of chemoradiationin patients with BRPC is even stronger due to the even higher risk ofnon-therapeutic surgery without NAT.

Patients who undergo macroscopically(R2) or even microscopically (R1)incomplete operations have a survivalsimilar to that of patients with locallyadvanced or metastatic disease. In aretrospective study of 142 patientswith radiographic evidence for venousor arterial involvement by tumor whounderwent resection in the absence ofpreoperative therapy, the median disease-specific survival of patients withtumors which did not appear to involveeither vein or artery (24.4 months)was significantly longer than that ofpatients with tumors which appearedto abut or encase the superior mesentericvein/portal vein (SMV-PV; 14.9months) or superior mesenteric artery(SMA; 12.8 months).3 Chemoradiationappears to increase the expectedmargin-negative resection rate forpatients with BRPC. Patients with arterialinvolvement at diagnosis wouldalmost all be expected to have positivemargins, whereas in the experienceof The University of Texas MD AndersonCancer Center nearly all patientswhose disease remained localizedafter completing chemoradiation withor without initial chemotherapy wereable to undergo margin-negative resection.

The radiographic stage rarelychanged and the median survival was40 months.4 Thus, the use of chemoradiationprior to surgery seems toimprove outcomes for these otherwiseincurable patients. There are no studiesevaluating neoadjuvant chemotherapyalone in BRPC, but this approachshould be tested. The current standardapproach in patients with BRPC is todeliver neoadjuvant chemotherapyfollowed by 3D conformal radiationtherapy (50.4 Gy in 28 fractions) withconcurrent chemotherapy (gemcitabineor capecitabine).

In summary, neoadjuvant strategieshave an excellent rationale and arebeing increasingly adopted. There arelikely subsets of patients with potentiallyresectable or BRPC who wouldtheoretically benefit more from theuse of chemoradiation than others(e.g., extensive arterial involvementor venous encasement) and those inwhom systemic chemotherapy maysuffice. Further investigation shouldfocus on this issue. It seems clear, however,that with the use of neoadjuvantchemoradiation the margin is almostalways negative, even in those with thehighest risk for a positive margin.

Dr. Crane is a Professor of RadiationOncology and Program Director ofGastrointestinal Radiation Oncology atThe University of Texas MD AndersonCancer Center. An ASCO member since2000, he has served on the ScientificProgram Committee, the GastrointestinalCancers Symposium Program Committee,and the Pancreatic Cancer Task Force.


  1. Lowy AM, Lee JE, Pisters PW, et al. AnnSurg. 1997;226:632-41. PMID: 9389397.
  2. Callery MP, Chang KJ, Fishman EK, et al.Ann Surg Oncol. 2009;16:1727-33. PMID:19396496.
  3. Yamada S, Fujii T, Sugimoto H, et al. Pancreas.2013;42:1004-10. PMID: 23532000.
  4. Katz MHG, Pisters PWT, Evans DB, et al.J Am Coll Surg. 2008;206:833-46. PMID:18471707.


Putting the Cart Before the Horse: A Medical Oncologist's Perspective on Neoadjuvant vs.Adjuvant Therapy for Resectable Pancreatic Cancer


By Margaret A. Tempero, MD, FASCO
University of California, San Francisco

Over the last 30 years, multiple adjuvanttrials have been conducted inpatients with pancreatic ductal adenocarcinomawithout any meaningfulimprovement in overall survival. Tobe fair, those trials relied on systemic chemotherapy with either gemcitabineor a fluorinated pyrimidine, agentswhich generally have minimal to modestactivity in the disease. Chemoradiotherapyhas often been incorporatedin the trials, but omitting chemoradiotherapyhas not appeared to be detrimental.

Only two trials (ESPAC1 andCONKO 001) have included an observationcontrol arm compared to chemotherapyalone.1,2 These important trialsproved that even modest systemic chemotherapyclearly provided a survivaladvantage.

Even prior to the advent of moreeffective combination chemotherapyregimens such as FOLFIRINOX3 orgemcitabine and nab-paclitaxel,4 therehas been a trend toward neoadjuvanttreatment as a strategy to improveoutcome. This makes sense on severallevels. First, a subset of patients withbiologically aggressive tumors canrelapse with metastases within weeksof resection. Delaying surgery to providea course of preoperative therapymight allow those more aggressivesystemic malignancies to be declared,thus sparing the patient a surgery thatis not likely to help him or her. Second,one of the worst predictors for a badoutcome is a positive margin (R1 resection).

This is of particular concern inpatients who appear to have borderline-resectable disease. The term "borderlineresectable" is variably applied.One of the first definitions appearedin the National Comprehensive CancerNetwork (NCCN) guidelines5 and furtheriterations have evolved to providemore homogeneity in cohorts for clinicaltrials.6 Effective treatment preoperatively(systemic chemotherapy,chemoradiation, or both) could reduceR1 resections, render more patients tobe surgical candidates, and hopefullyincrease the cure rate. However, thisgoal has to be balanced with treatmenttoxicity, especially myelosuppression,which could add significantly to overallmorbidity. Rigorous clinical trialsin the preoperative setting are sorelyneeded!

One might argue that we should comparepreoperative to postoperativeapproaches. This will probably neverbe done and, if attempted, would befraught with unintended biases. Forexample, preoperative treatmentwould be given to all but postoperativetreatment would be selectivelygiven to those patients with successfulresections (10% to 15% of patientswho appear to have resectable diseasepre-op are found to have unresectabledisease at laparotomy), with adequaterecovery from surgery within about 12weeks and without evidence of earlysystemic metastasis. In a nutshell, youcould not control for these variables.In addition, combination chemotherapywith regimens such as FOLFIRINOX orgemcitabine plus nab-paclitaxel arenow commonly employed in the pre-opsetting but are still undergoing phaseIII evaluation in the post-op setting. Itwould be challenging to select a proper"control arm."

The role of radiation in the adjuvantsetting is unknown. To date, therehave been no strong signals suggestinga major impact. In fact, some datasuggest a detrimental effect. RTOG0848 is designed to address the roleof chemoradiation as an adjunct tochemotherapy in the adjuvant setting.For preoperative treatment, given theintent of deploying effective treatmentto prevent an R1 resection, it wouldseem very important to study the roleof neoadjuvant chemoradiation anddetermine if it is essential to that strategy.The availability of more effectivecombination chemotherapy regimensis transformative and potentially laysthe foundation for greater benefit withchemoradiation.We need to figure this out.

Dr. Tempero is the Director of the Universityof California, San Francisco, PancreasCenter. An ASCO member since 1983, shehas served on the ASCO Board of Directorsand as 2003-2004 ASCO President, amongother volunteer activities.


  1. Neoptolemos JP, Stocken DD, Friess H, etal. N Engl J Med. 2004;350:1200-10. PMID:15028824.
  2. Oettle H, Post S, Neuhaus P, et al. JAMA.2007;297:267-77. PMID: 17227978.
  3. Conroy T, Desseigne F, Ychou M. et al.N Engl J Med. 2011;364:1817-25. PMID:21561347.
  4. Von Hoff DD, Ervin T, Arena FP, et al. NEngl J Med. 2013;369:1691-703. PMID:24131140.
  5. Tempero MA, Arnoletti JP, Behrman SW, etal. J Natl Compr Canc Netw. 2012;10:703-13.PMID: 22679115.
  6. Katz MH, Marsh R, Herman JM, et. al.Ann Surg Oncol. 2013;20:2787-95. PMID:23435609.
Back to Top