Oct 28, 2013
By Stephen Leong, MD
University of Colorado Cancer Center
It is estimated that more than 102,000new colon cancer cases will be diagnosedin the United States this year.1Approximately 40% of these will beclassified as localized (stage I andstage II).1 The five-year survival estimatesfor patients with stage II coloncancer vary widely. According to theAmerican Joint Committee on Cancer(AJCC; Surveillance, Epidemiology,and End Results 1975-2000 database),survival ranges from 67% (stage IIA) to37% (stage IIC)2; in modern clinical trials,survival rates are higher, generallyexceeding 80%.3,4
The big question is whether adjuvantchemotherapy can improve the chanceof cure in resected stage II colon cancer.The beneficial role of adjuvantchemotherapy has been clearly demonstratedin stage III colon cancer.4-7However, evidence in stage II coloncancer is less convincing, and hence,not routinely recommended. Even withthe identification of high-risk clinicaland molecular characteristics, thesehave not been shown to be predictiveof benefit from chemotherapy.
In this issue of ASCO Connection’s CurrentControversies in Oncology series,Dr. Daniel Haller and Dr. NicholasPetrelli discuss the issues facing medicaloncologists and surgical oncologiststreating stage II colon cancer.
Dr. Leong is an Associate Professor in theDivision of Medical Oncology at the Universityof Colorado Cancer Center and a memberof the Developmental Therapeutics andGastrointestinal Malignancies Programs. Hehas been an ASCO member since 2005.
- Siegel R, Naishadham D, Jemal A. CA CancerJ Clin. 2013;63:11-30. PMID: 23335087.
- Colon and rectum. In Edge SB, Byrd DR,Compton CC, et al. (eds). AJCC CancerStaging Manual, 7th ed. New York: Springer,2010;143.
- Quasar Collaborative Group, Gray R, BarnwellJ, et al. Lancet. 2007;370:2020-9.PMID: 18083404.
- Efficacy of adjuvant fluorouracil and folinicacid in colon cancer. International MulticentrePooled Analysis of Colon CancerTrials (IMPACT) investigators. Lancet.1995;345:939-44. PMID: 7715291.
- Kuebler JP, Wieand HS, O’Connell MJ, etal. J Clin Oncol. 2007;25:2198-204. PMID:17470851.
- André T, Boni C, Mounedji-Boudiaf L, et al.N Engl J Med. 2004;350:2343-51. PMID:15175436.
- O’Connell MJ, Mailliard JA, Kahn MJ, etal. J Clin Oncol. 1997;15:246-50. PMID:8996149.
A Medical Oncologist’s Perspective
By Daniel G. Haller, MD, FACP, FASCO,FRCP
Abramson Cancer Center and PerelmanSchool of Medicine at the University of Pennsylvania
Adjuvant therapy for stage II, node-negativecolon cancer remains oneof the most controversial topics ingastrointestinal oncology: no postoperativetreatment, single-agent fluoropyrimidinesalone, or combinationchemotherapy with FOLFOX? Personalizedmedicine requires considerationof both tumor and host factors thatare prognostic and markers that arepredictive of benefit from treatment.Patient choice, comorbidities, and agemust also factor into any decisions forthose with stage II disease.
The American Joint Committee on Cancer’snew Seventh Edition of the AJCCCancer Staging Manual has separatedstage II node-negative disease intothree categories.1 Expanded data setshave shown differential prognosis withinT4 lesions based on extent of disease.Accordingly, T4 lesions are subdivided asT4a (tumor penetrates the surface of thevisceral peritoneum) and as T4b (tumordirectly invades or is histologicallyadherent to other organs or structures).Stage group II is now subdivided into IIA(T3N0), IIB (T4aN0), and IIC (T4bN0).
It is uncertain whether stage II andstage III tumors are biologically differentfrom inception or assume biologicbehaviors as tumors progress throughstages. Stage II tumors clearly recurless often, but also later than stageIII disease. The ACCENT database hasalso shown that patients with stage IIdisease that recurs have better overallsurvival than those who presented withstage III disease.2 Whether this representsless adjuvant treatment or differencesin biology is uncertain. O’Connellet al. evaluated markers in stage II andIII disease from National Surgical AdjuvantBreast and Bowel Project (NSABP)studies and compared the analysis tosimilar analyses from PETACC-3.3 Overall,the data were similar, with defectivemismatch repair status (dMMR) astronger prognostic marker for stage IIdisease than stage III.
For stage II disease, the number ofnodes reported and the nodal rationalso is prognostic. Although at least12 nodes should be examined andreported, in many large trials fewerthan half of patients have more than 12nodes reported. Although clinicians usecertain clinicopathologic “high-risk”features, such as number of nodesexamined, high-grade features, andlymphatic invasion, many of these havenot been validated and some may beprognostic, but not necessarily predictive,of benefit from adjuvant therapy.
The largest trial of adjuvant treatmentreported to date in stage II colon canceris the QUASAR trial, using singleagent5-FU.4 The relative risk reductionin recurrence was similar to that forstage III disease, but the higher baselinecure rate of 75% to 80% of stageII disease translated into an absolutebenefit of about 4%, similar to thatseen in the prior retrospective IMPACTanalysis.5 dMMR (microsatellite instability[MSI]) has attracted attention asboth a prognostic and predictive markerin stage II colorectal cancer. MSIis a marker of Lynch syndrome, butalso occurs in 15% of sporadic tumors.In vitro studies suggest that dMMRtumors may be resistant to 5-FU andmay also have a better natural history,making the marker both predictive andprognostic. The original seminal paperby Ribic et al. showed that patientswith dMMR stage II cancer had a betterrecurrence and survival, and thatpatients with dMMR who received adjuvanttreatment actually did worse thanpatients who received surgery alone.6This was also confirmed in a larger retrospectiveanalysis by Sargent et al.7
O’Connell et al. published a trainingset of a 12-gene recurrence score,which was further explored in a validationanalysis by Kerr et al. from theQUASAR data.8 These data were alsointerrogated by the recurrence scoreestablished by O’Connell et al. Low-,intermediate-, and high-risk groupswere identified. Although the data areclearly prognostic, they do not necessarilypredict who may benefit fromadjuvant treatment with 5-FU, muchless FOLFOX. In a multivariate analysis,the recurrence score, MSI status, andT-stage were the most important prognosticfactors, whereas other factors,such as tumor differentiation, were lessimportant. The worst prognostic groupwere the individuals in the small groupwho had recurrence scores of 41 orhigher, T4 tumors, and proficient MMR.In general, patients with T3 tumors anddMMR have excellent prognosis andwill not likely benefit significantly fromtreatment. Regardless of the recurrencescore, patients with T4 tumorswith proficient MMR have poorer prognosisand should be considered foradjuvant therapy with single-agent fluoropyrimidinesor FOLFOX. The groupin whom recurrence score may assistin decision-making are those with T3tumors and proficient MMR (intermediaterisk).
NSABP data have also been analysedby Yothers et al. for the contributionof oxaliplatin in stage II disease.9There was a strong positive effect foroxaliplatin in stage II for disease-freesurvival, time to recurrence, and overallsurvival, with an additional 3% to4% above the effects seen with 5-FUalone. One could surmise that FOLFOXfor stage II disease should be limitedto patients perceived to be at highestrisk for recurrence by pathology andrecurrence score. Further informationshould be available in the future fromthe study E5202, in which high-riskpatients (proficient MMR and loss ofheterozygosity of 18q) were assignedto FOLFOX. However, they were notrandomly assigned to either surgeryalone or single-agent fluoropyrimidines.An update on age and adjuvantchemotherapy with all-stage coloncancer with contemporary treatments,based on the large ACCENT database,was recently published.10
Dr. Haller is Professor of Medicine emeritusat Abramson Cancer Center and PerelmanSchool of Medicine at the Universityof Pennsylvania. An ASCO member since1978, he currently serves as Editor-in-Chiefof ASCO University® and on the SpecialAwards Selection Committee. He receivedthe ASCO Special Recognition Award andwas inducted as a Fellow of ASCO in 2011.
- Edge SB, Byrd DR, Compton CC, et al. (eds).AJCC Cancer Staging Manual, 7th ed. NewYork: Springer; 2010.
- Sargent DS, Sobrero A, Grothey A, et al. JClin Oncol. 2009;27:872-7. PMID: 19124803.
- O’Connell MJ, Lavery I, Yothers G, et al.J Clin Oncol. 2010;28:3937-44. PMID:20679606.
- Quasar Collaborative Group, Gray R,Barnwell J, McConkey C, et al. Lancet.2007;370:2020-9. PMID: 18083404.
- Efficacy of adjuvant fluorouracil and folinicacid in B2 colon cancer. International MulticentrePooled Analysis of B2 Colon CancerTrials (IMPACT B2) Investigators. J ClinOncol. 1999;17:1356-63. PMID: 10334519.
- Ribic CM, Sargent DJ, Moore MJ, et al.N Engl J Med. 2003;349:247-57. PMID:12867608.
- Sargent DJ, Marsoni S, Monges G, et al.J Clin Oncol. 2010;28:3219-26. PMID:20498393.
- Gray RG, Quirke P, Handley K, et al. J ClinOncol. 2011;29:4611-9. PMID: 22067390.
- Yothers G, O’Connell MJ, Allegra CJ, etal. J Clin Oncol. 2011;29:3768-74. PMID:21859995.
- McCleary NJ, Meyerhardt JA, Green E,et al. J Clin Oncol. 2013;31:2600-6. PMID:23733765.
By Nicholas J. Petrelli, MD
Helen F. Graham Cancer Center atChristiana Care Health System andThomas Jefferson University
If you Google “the adjuvant treatmentof stage II colon cancer,” the searchengine returns 907,000 choices. Youwould think that by the 21st century,we would have figured this out.
What we can say is that we are gettingbetter at selecting the patients withstage II disease who would benefitfrom adjuvant chemotherapy, which Dr.Haller has discussed in detail. Generally,the 2013 National ComprehensiveCancer Network guidelines state thatpatients with low-risk stage II diseasecan be enrolled in a clinical trial versusobservation without adjuvant therapyor considered for systemic chemotherapy.We can best define low-risk individualsby emphasizing the high-riskstage II patients to whom one wouldseriously consider administering systemicchemotherapy: patients whosetumors have poor prognostic features,including T4 tumors (stage IIB/IIC),poorly differentiated histology exclusiveof those cancers that are microsatelliteinstability high, tumors withlymphovascular or perineural invasion,bowel obstruction presentation, andcancers with localized perforation orpositive margins (although it is beyondmy imagination that a well-trained surgicaloncologist would have a positivesurgical margin on any type of coloncancer resection).
One should also consider inadequatelysampled lymph nodes defined as lessthan 12 lymph nodes cleared by thepathologist. Of course, lymph nodestatus is not only the responsibility ofthe pathologist, but also the surgeon.Whether the patient has low-risk orhigh-risk stage II colon cancer, observationwithout adjuvant therapy is anoption in this population.
Many cancer guidelines never mentiontwo very important prognostic factors:the processes of cancer care and thesurgeons themselves. In my opinion,inadequate colon cancer surgery is aninexcusable reason for the administrationof adjuvant chemotherapy. No oneshould deny that surgical and hospitalvolumes in conjunction with surgicalspecialty training indeed influencepatient outcomes. Interest in a specificfield by a surgeon results in furtherspecialization and additional improvementin skills and knowledge. Surgeonswith specific training in a given specialtygenerally make greater use of thefull range of treatment and diagnosticoptions available within a multidisciplinaryteam setting. Specialist trainingmust eventually reduce or indeedprevent extreme variation in surgicaloutcomes. Surgical oncologists need toshare their outcomes with their medicaloncology colleagues. This should notbe a relationship where “you scratchmy back and I’ll scratch yours” withouta review of high-quality outcomes.
Aside from surgeon case volume andoutcomes, it is important to rememberthat there are a variety of hospital factorssuch as the types of services andtechnology, multidisciplinary teams,internal quality programs, and NationalCancer Institute designation that haveall been suggested to be associatedwith superior cancer care. The problemis that the literature rarely considersthese factors and certainly one of thereasons is that they are too detailed toabstract.
In conclusion, aside from medicaloncologists evaluating patients withstage II colon cancer for adjuvanttherapy with the prognostic factorsmentioned above and inclusive of multigeneassays discussed by Dr. Haller,it is important to remember that thesurgeons themselves are a significantprognostic factor that needs to beconsidered.
Dr. Petrelli is the Bank of America EndowedMedical Director at the Helen F. GrahamCancer Center at Christiana Care HealthSystem and a Professor of Surgery at ThomasJefferson University. An ASCO membersince 1983, he currently serves on the CancerResearch Committee. He was inducted as a Fellowof ASCO in 2008.
- Bilimoria KY, Bentrem DJ, Feinglass JM, etal. J Clin Oncol. 2008;26:4626-33. PMID:18574159.
- Lerut T. Ann Surg. 2000;232:729-32. PMID:11088067.
- Hillner BE, Smith TJ, Desch CE. J ClinOncol. 2000;18:2327-40. PMID: 10829054.
- Saltz LB. Surg Oncol Clin N Am.2010;19:819-27. PMID: 20883956.
- André T, Boni C, Mounedji-Boudiaf L, et al.N Engl J Med. 2004;350:2343-51. PMID:15175436.
- André T, Boni C, Navarro M, et al. J ClinOncol. 2009;27:3109-16. PMID: 19451431.