- Email: [email protected]
Mismatch Repair as a Prognostic Marker for Adjuvant Therapy in Colorectal Cancer – How Soon is Now?
Clinical Oncology 25 (2013) 625e629 Contents lists available at SciVerse ScienceDirect
Clinical Oncology journal homepage: www.clinicaloncologyonline.net
Editorial
Mismatch Repair as a Prognostic Marker for Adjuvant Therapy in Colorectal Cancer e How Soon is Now? B. O’Leary, D.C. Gilbert* Sussex Cancer Centre, Royal Sussex County Hospital, Brighton, UK Received 11 July 2013; accepted 15 July 2013
Treatment for colorectal cancer is guided principally by stage in a classification that has changed little over 70 years. Patients presenting with stage II and III resected colorectal cancer have no detectable metastatic disease, but a proportion of these patients will have undeclared micrometastatic disease, which it may be possible to eradicate with adjuvant treatment. Adjuvant chemotherapy with 5-fluorouracil (5-FU)based regimens was the standard of care for the treatment of stage III patients [1e3] and a modest benefit of oxaliplatin has been shown [4,5]. These studies also included high-risk stage II patients, but with a lower risk of relapse, even similar hazard ratios translate into small clinical benefit and the role for adjuvant chemotherapy remains debatable [6e8]. The largest study (QUASAR) suggested possible benefits of adjuvant chemotherapy of the order of 3e4% in overall survival for stage II patients [9], although of only marginal statistical significance. CALGB9581 (albeit investigating immunotherapy in this setting) confirmed the good prognosis of many patients with low-risk, stage II colorectal cancer [10]. Despite numerous emerging biomarkers, decisions on the potential benefits of adjuvant chemotherapy in stage II disease have used subjective analyses of adverse histopathological features (derived from the initial adjuvant 5-FU studies without subsequent validation). The hope is that biomarkers will allow better resolution of the heterogeneity thought to be present in colorectal cancer and, subsequently, more effectively targeted treatment. We believe there is now a strong case for routinely assessing mismatch repair (MMR) status in stage II colorectal cancer.
Author for correspondence: D.C. Gilbert, Sussex Cancer Centre, Royal Sussex County Hospital, Eastern Road, Brighton BN1 5BE, UK. Tel: 44-1273696955. E-mail address: [email protected] (D.C. Gilbert).
A Fork in the Road e Different Pathways to Colorectal Cancer DNA MMR was first identified through the discovery that some colorectal tumours showed variations in ‘microsatellites’, mononucleotide and dinucleotide repeats [11]. These tumours were less likely to feature mutations in p53 and KRAS, less likely to be invasive and more likely to occur proximally, suggesting an alternative to the classic model suggested previously by Fearon and Vogelstein [12] of tumours developing through loss of APC function and activation of KRAS [13]. Subsequently it became clear that deficient MMR (dMMR) was responsible for this ‘microsatellite instability’ (‘MSI’ or ‘MSI-H’). Germline mutations in MMR proteins were found to be the driving mechanism behind tumours seen in Lynch syndrome (hereditary non-polyposis colorectal cancer). A similar phenotype of colorectal cancers with MSI is seen in a proportion of sporadic colorectal cancers with either a somatic mutation in MMR proteins or more commonly gene silencing through promoter methylation [14]. MMR status (dMMR or proficient [pMMR] corresponding to MSI-H or microsatellite stability, respectively, see Figure 1) can be determined by direct analysis of MSI (polymerase chain reaction analysis) or by immunohistochemistry for the MMR proteins (MLH1, MSH1, MSH6 and PMH2). With high sensitivity (92.3%) and specificity (100%) [15,16], immunohistochemistry is cheaper and simpler than polymerase chain reaction analysis, and provides a simple and effective means of establishing MMR status that is deliverable in a conventional setting. Specifically, the absence of immunohistochemical staining for one or more of these proteins implies dMMR (i.e. MSI-H).
Strong Evidence of a Prognostic Effect There has been growing evidence that dMMR status is a strong prognostic biomarker for improved outcomes in
0936-6555/$36.00 Ó 2013 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.clon.2013.07.006
626
B. O’Leary, D.C. Gilbert / Clinical Oncology 25 (2013) 625e629
Mismatch repair status
Abbreviation
Deficient
dMMR
Microsatellite status = Microsatellite instability
(loss of staining for MLH1, MSH1, MSH6 or PMH2)
Proficient
Abbreviation MSI or MSI-H
(detectable via polymerase chain reaction)
pMMR
= Microsatellite stable
MSS
Fig 1. Terminology.
colorectal cancer (Table 1). A meta-analysis in 2005 derived a pooled hazard ratio of 0.65 for overall survival in favour of dMMR (95% confidence interval 0.59e0.71) from 32 studies comprising 7642 patients [17]. This advantage held for stage II and III patients, with a pooled hazard ratio of 0.67 (95% confidence interval 0.58e0.78). A subsequent 2010 metaanalysis found an increased overall survival for dMMR patients from 20 studies comprising 9243 patients with stage IeIV disease [18]. This result was also seen in an analysis of the subset of 10 studies of stage II and II colorectal cancers with an odds ratio for survival at completion of study follow-up of 0.65 (95% confidence interval 0.53e0.79, P < 0.0001). Similar results were seen in the studies where data were available for disease-free survival. Retrospective analysis of 1913 patients enrolled into QUASAR showed this prognostic effect of dMMR in stage II colorectal cancer, highlighting the association with clinical
features [19]. Twenty-six per cent of right-sided tumours were dMMR as opposed to 3% of left-sided tumours and 1% of rectal cancers. dMMR rates were twice as frequent in stage II as opposed to stage III tumours (12% versus 6%). Crucially dMMR tumours showed about half the recurrence risk of their pMMR counterparts, with a recurrence rate of 11% in dMMR versus 26% in pMMR tumours (risk ratio 0.53, 95% confidence interval 0.40e0.70; P < 0.001). Similar results have recently been published from a population-based series from a single institution in Norway [20]. Complete data were available for 613 patients undergoing curative resection for stage II colorectal cancer. Fourteen per cent of cases were dMMR (defined by polymerase chain reaction for MSI), more frequently seen in women (19%) and right-sided tumours (29%). Although outcomes were worse that seen in the QUASAR clinical trial cohort, MSI-positive (dMMR) stage II cases had significantly improved 5 year relapse-free survival.
Does Mismatch Repair Status Predict Response to 5-fluorouracil? Beyond the prognostic effect there are accumulating data that MMR status is predictive of the response to adjuvant chemotherapy (Table 2), where dMMR tumours may not derive benefit from 5-FU and indeed may do worse. Mechanistic data support a role for dMMR tumours exhibiting resistance to 5-FU [25]. Clinical data are derived from adjuvant trials or cohort studies [17,18,24], although large numbers of patients are needed to amass the numbers of
Table 1 Studies reporting mismatch repair status as a prognostic marker in stage II and III colorectal cancer Reference
Design
Number of stage II and III
Prognostic marker
[17]
Meta-analysis
2935
[18]
Meta-analysis
4014
[21]
Retrospective from RCTs
457
[22]
Prospective in RCTs
1852
[19]
Retrospective from RCTs
1913
[23]
Retrospective cohort
787
[20]
Retrospective cohort
613
Yes HR ¼ 0.67 for OS (95% CI 0.58e0.78) in dMMR Yes OR ¼ 0.65 for survival at end of study (95% CI 0.53e0.79) in dMMR, P < 0.001 Yes HR ¼ 0.46 for DFS (95% CI 0.22e0.95) in dMMR, univariate analysis, P ¼ 0.03 Yes HR ¼ 0.77 (95% CI 0.71e0.80) for OS in dMMR, P ¼ 0.029 Yes RR ¼ 0.53 (95% CI 0.40e0.70) for recurrence in dMMR, P < 0.001 Yes HR ¼ 0.40 for OS at 3 years (95% CI 0.19e0.86) in dMMR, P ¼ 0.001 Yes HR ¼ 1.60 for RFS (95% CI 1.01e2.52) in pMMR, P ¼ 0.045
Notes
Benefit not maintained in multivariate analysis
Calculation includes stage I and IV patients Result due to stage II rather than stage III, analysis included stage I
RCT, randomised controlled trial; HR, hazard ratio; RR, risk ratio; OR, odds ratio; OS, overall survival; DFS, disease-free survival; RFS, relapsefree survival; dMMR, deficient mismatch repair; pMMR, proficient mismatch repair.
B. O’Leary, D.C. Gilbert / Clinical Oncology 25 (2013) 625e629
627
Table 2 Studies reporting mismatch repair status as a predictive marker in stage II and III colorectal cancer Reference
Design
Number of stage II and III
[17]
Meta-analysis
963
[24]
Meta-analysis
3690
[18]
Meta-analysis
2863
[21]
Retrospective from RCTs
457
[19]
Retrospective from RCTs
1913
[23]
Retrospective cohort
787
Predictive benefit with 5-FU for dMMR
Predictive benefit with 5-FU for pMMR
No HR ¼ 1.24 for OS (95% CI 0.72e2.14) in dMMR
Yes HR ¼ 0.72 for OS (95% CI 0.57e0.92) in pMMR, P ¼ 0.07 Trend towards benefit in pMMR versus dMMR
No HR ¼ 0.7 for OS (95% CI 0.44e1.09) in dMMR, P ¼ 0.86 No OR for survival at end of study ¼ 1.03 (95% CI 0.25e4.3) in dMMR No HR ¼ 1.39 for DFS (95% CI 0.46e4.15) in dMMR, P ¼ 0.56 No OR 0.72 for survival at end of study (95% CI 0.40e1.27) in dMMR No HR ¼ 0.57 for OS at 3 years (95% CI 0.09e3.75) in dMMR, P ¼ 0.566
Yes OR for survival at end of study ¼ 0.52 (95% CI 0.4e0.6) in pMMR, P < 0.0001 Yes HR ¼ 0.67 for DFS (95% CI 0.48e0.93) in pMMR, P ¼ 0.02 No OR 0.86 for survival at end of study (95% CI 0.73e1.02) in pMMR Yes HR ¼ 0.34 for OS at 3 years (95% CI 0.48e0.57) in pMMR, P < 0.001
Notes
Inadequate sample size for by-stage analysis for dMMR
Analysis included stages IIeIV
5-FU, 5-fluorouracil; RCT, randomised controlled trial; HR, hazard ratio; RR, risk ratio; OR, odds ratio; OS, overall survival; DFS, disease-free survival; RFS, relapse-free survival; dMMR, deficient mismatch repair; pMMR, proficient mismatch repair.
events required to show any association between dMMR and response. A systematic review and meta-analysis [24] identified seven studies comprising 3690 patients comparing untreated patients with those treated with 5-FUbased adjuvant chemotherapy for stage II or III colorectal cancer, all of whom were tested for dMMR status. No significant increase in relapse-free survival with chemotherapy was found in the pooled estimate for dMMR tumours. Three further datasets have since been published examining MMR as a predictive marker, with none able to show an advantage for adjuvant 5-FU in dMMR colorectal cancer, including the large dataset provided by the QUASAR study [19,21,23]. Importantly, by combining their data with their previous analysis, Sargent et al. [21] showed a small, but statistically significant, reduction in overall survival for stage II dMMR patients receiving systemic treatment (hazard ratio 2.95, 95% confidence interval 1.02e8.54, P ¼ 0.04).
Future Questions and Directions Where Now for Stage II, Mismatch Repair-proficient (Microsatellite-stable) Colorectal Cancer?
and predictive biomarkers to identify groups who may have a more tangible benefit from adjuvant treatment? BRAF mutations are associated with poorer outcomes within the setting of pMMR stage II and III colorectal cancer [26], but may not be predictive of a response to chemotherapy. A number of gene expression panels are undergoing validation, but none is in routine use [27]. Is There a Role for Mismatch Repair Status in the Management of Stage III Disease? Although dMMR status may confer resistance to 5-FU, it is unknown how MMR status may affect sensitivity to oxaliplatin. One recent study found dMMR to be of no value in predicting the response to oxaliplatin in stage III colorectal cancer [28]. Interestingly, a retrospective analysis of patients in the otherwise negative CALGB 89803 study showed improved disease-free survival for dMMR versus pMMR receiving adjuvant irinotecan, 5-FU and leucovorin in stage III colorectal cancer, a relationship not seen in the cohort receiving 5-FU/leucovorin [29].
Conclusion If dMMR stage II patients can be spared adjuvant chemotherapy for the prognostic and predictive reasons described above, is there a way to further stratify stage II pMMR patients with further combinations of prognostic
Colorectal cancer is a genetically diverse disease displaying a number of important phenotypes. dMMR patients for the most part have an excellent prognosis (with 6 year
628
B. O’Leary, D.C. Gilbert / Clinical Oncology 25 (2013) 625e629
survival in one study as high as 97% [30]) and can be spared additional treatment, which may even result in worse outcomes. Immunohistochemistry for the MMR proteins (MLH1, MSH1, MSH6, PMH2) is readily available and a robust marker of dMMR. The role of MMR status in decision-making for adjuvant chemotherapy in stage III disease with oxaliplatin or further investigating a role for irinotecan is more uncertain. In the future however, dMMR status will probably be part of genetic signature profiles that will allow more precise allocation of adjuvant treatments according to expected benefit. Finally, if a wider cohort of patients is to be tested for MMR status, clinicians must be aware of the increased likelihood of germline mutations in patients found to be dMMR, and be ready to refer on for genetic counselling, testing for Lynch syndrome and subsequent patient treatment and surveillance. In conclusion, we believe that the time is now here for routine testing of stage II colorectal cancer for MMR status, with dMMR (MSI-H) tumours for the most part being able to avoid additional treatment.
[11]
[12] [13]
[14] [15]
[16]
[17]
[18]
References [1] Moertel CG, Fleming TR, Macdonald JS, et al. Levamisole and fluorouracil for adjuvant therapy of resected colon carcinoma. New Engl J Med 1990;322(6):352e358. [2] O’Connell MJ, Laurie JA, Kahn M, et al. Prospectively randomized trial of postoperative adjuvant chemotherapy in patients with high-risk colon cancer. J Clin Oncol 1998;16(1):295e300. [3] Wolmark N, Rockette H, Fisher B, et al. The benefit of leucovorin-modulated fluorouracil as postoperative adjuvant therapy for primary colon cancer: results from National Surgical Adjuvant Breast and Bowel Project protocol C-03. J Clin Oncol 1993;11(10):1879e1887. [4] Kuebler JP, Wieand SH, O’Connell MJ, et al. Oxaliplatin combined with weekly bolus fluorouracil and leucovorin as surgical adjuvant chemotherapy for stage II and III colon cancer: results from NSABP C-07. J Clin Oncol 2007;25(16):2198e2204. T, Boni C, Navarro M, et al. Improved overall survival [5] Andre with oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment in stage II or III colon cancer in the MOSAIC trial. J Clin Oncol 2009;27(19):3109e3116. [6] IMPACT B2 Investigators. Efficacy of adjuvant fluorouracil and folinic acid in B2 colon cancer. J Clin Oncol 1999;17(5):1356. [7] Mamounas E, Wieand S, Wolmark N, et al. Comparative efficacy of adjuvant chemotherapy in patients with Dukes’ B versus Dukes’ C colon cancer: results from four National Surgical Adjuvant Breast and Bowel Project Adjuvant Studies (C-01, C-02, C-03, and C-04). J Clin Oncol 1999;17(5):1349. [8] Gill S, Loprinzi CL, Sargent DJ, et al. Pooled analysis of fluorouracil-based adjuvant therapy for stage II and III colon cancer: who benefits and by how much? J Clin Oncol 2004;22(10):1797e1806. [9] Quasar Collaberative Group. Adjuvant chemotherapy versus observation in patients with colorectal cancer: a randomised study. Lancet 2007;370(9604):2020e2029. [10] Niedzwiecki D, Bertagnolli MM, Warren RS, et al. Documenting the natural history of patients with resected stage II
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
adenocarcinoma of the colon after random assignment to adjuvant treatment with edrecolomab or observation: results from CALGB 9581. J Clin Oncol 2011;29(23):3146e3152. Ionov Y, Peinado MA, Malkhosyan S, et al. Ubiquitous somatic mutations in simple repeated sequences reveal a new mechanism for colonic carcinogenesis. Nature 1993;363(6429):558e561. Fearon ER, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell 1990;61(5):759e767. Thibodeau S, Bren G, Schaid D. Microsatellite instability in cancer of the proximal colon. Science 1993;260(5109): 816e819. Boland CR, Goel A. Microsatellite instability in colorectal cancer. Gastroenterology 2010;138(6):2073e2087. Lindor NM, Burgart LJ, Leontovich O, et al. Immunohistochemistry versus microsatellite instability testing in phenotyping colorectal tumors. J Clin Oncol 2002;20(4):1043e1048. Ward RL, Turner J, Williams R, et al. Routine testing for mismatch repair deficiency in sporadic colorectal cancer is justified. J Pathol 2005;207(4):377e384. Popat S, Hubner R, Houlston RS. Systematic review of microsatellite instability and colorectal cancer prognosis. J Clin Oncol 2005;23(3):609e618. Guastadisegni C, Colafranceschi M, Ottini L, et al. Microsatellite instability as a marker of prognosis and response to therapy: a meta-analysis of colorectal cancer survival data. Eur J Cancer 2010;46(15):2788e2798. Hutchins G, Southward K, Handley K, et al. Value of mismatch repair, KRAS, and BRAF mutations in predicting recurrence and benefits from chemotherapy in colorectal cancer. J Clin Oncol 2011;29(10):1261e1270. Merok MA, Ahlquist T, Royrvik EC, et al. Microsatellite instability has a positive prognostic impact on stage II colorectal cancer after complete resection: results from a large, consecutive Norwegian series. Ann Oncol 2013;24(5):1274e1282. Sargent DJ, Marsoni S, Monges G, et al. Defective mismatch repair as a predictive marker for lack of efficacy of fluorouracil-based adjuvant therapy in colon cancer. J Clin Oncol 2010;28(20):3219e3226. Bertagnolli MM, Redston M, Compton CC, et al. Microsatellite instability and loss of heterozygosity at chromosomal location 18q: prospective evaluation of biomarkers for stages II and III colon cancerda study of CALGB 9581 and 89803. J Clin Oncol 2011;29(23):3153e3162. Hong SP, Min BS, Kim TI, et al. The differential impact of microsatellite instability as a marker of prognosis and tumour response between colon cancer and rectal cancer. Eur J Cancer 2012;48(8):1235e1243. Des Guetz G, Schischmanoff O, Nicolas P, et al. Does microsatellite instability predict the efficacy of adjuvant chemotherapy in colorectal cancer? A systematic review with meta-analysis. Eur J Cancer 2009;45(10):1890e1896. Carethers JM, Chauhan DP, Fink D, et al. Mismatch repair proficiency and in vitro response to 5-fluorouracil. Gastroenterology 1999;117(1):123e131. Roth AD, Tejpar S, Delorenzi M, et al. Prognostic role of KRAS and BRAF in stage II and III resected colon cancer: results of the translational study on the PETACC-3, EORTC 40993, SAKK 60-00 trial. J Clin Oncol 2010;28(3):466e474. Salazar R, Roepman P, Capella G, et al. Gene expression signature to improve prognosis prediction of stage II and III colorectal cancer. J Clin Oncol 2011;29(1):17e24. Li P, Fang YJ, Li F, et al. ERCC1, defective mismatch repair status as predictive biomarkers of survival for stage III colon
B. O’Leary, D.C. Gilbert / Clinical Oncology 25 (2013) 625e629 cancer patients receiving oxaliplatin-based adjuvant chemotherapy. Br J Cancer 2013;108(6):1238e1244. [29] Bertagnolli MM, Niedzwiecki D, Compton CC, et al. Microsatellite instability predicts improved response to adjuvant therapy with irinotecan, fluorouracil, and leucovorin in stage
629
III colon cancer: Cancer and Leukemia Group B Protocol 89803. J Clin Oncol 2009;27(11):1814e1821. [30] Lanza G, Gafa R, Sanyini A, et al. Immunohistochemical test for MLH1 and MSH2 expression predicts clinical outcome in stage II and III colorectal cancer patients. J Clin Oncol 2006;24(15):2359e2367.
Clinical Oncology journal homepage: www.clinicaloncologyonline.net
Editorial
Mismatch Repair as a Prognostic Marker for Adjuvant Therapy in Colorectal Cancer e How Soon is Now? B. O’Leary, D.C. Gilbert* Sussex Cancer Centre, Royal Sussex County Hospital, Brighton, UK Received 11 July 2013; accepted 15 July 2013
Treatment for colorectal cancer is guided principally by stage in a classification that has changed little over 70 years. Patients presenting with stage II and III resected colorectal cancer have no detectable metastatic disease, but a proportion of these patients will have undeclared micrometastatic disease, which it may be possible to eradicate with adjuvant treatment. Adjuvant chemotherapy with 5-fluorouracil (5-FU)based regimens was the standard of care for the treatment of stage III patients [1e3] and a modest benefit of oxaliplatin has been shown [4,5]. These studies also included high-risk stage II patients, but with a lower risk of relapse, even similar hazard ratios translate into small clinical benefit and the role for adjuvant chemotherapy remains debatable [6e8]. The largest study (QUASAR) suggested possible benefits of adjuvant chemotherapy of the order of 3e4% in overall survival for stage II patients [9], although of only marginal statistical significance. CALGB9581 (albeit investigating immunotherapy in this setting) confirmed the good prognosis of many patients with low-risk, stage II colorectal cancer [10]. Despite numerous emerging biomarkers, decisions on the potential benefits of adjuvant chemotherapy in stage II disease have used subjective analyses of adverse histopathological features (derived from the initial adjuvant 5-FU studies without subsequent validation). The hope is that biomarkers will allow better resolution of the heterogeneity thought to be present in colorectal cancer and, subsequently, more effectively targeted treatment. We believe there is now a strong case for routinely assessing mismatch repair (MMR) status in stage II colorectal cancer.
Author for correspondence: D.C. Gilbert, Sussex Cancer Centre, Royal Sussex County Hospital, Eastern Road, Brighton BN1 5BE, UK. Tel: 44-1273696955. E-mail address: [email protected] (D.C. Gilbert).
A Fork in the Road e Different Pathways to Colorectal Cancer DNA MMR was first identified through the discovery that some colorectal tumours showed variations in ‘microsatellites’, mononucleotide and dinucleotide repeats [11]. These tumours were less likely to feature mutations in p53 and KRAS, less likely to be invasive and more likely to occur proximally, suggesting an alternative to the classic model suggested previously by Fearon and Vogelstein [12] of tumours developing through loss of APC function and activation of KRAS [13]. Subsequently it became clear that deficient MMR (dMMR) was responsible for this ‘microsatellite instability’ (‘MSI’ or ‘MSI-H’). Germline mutations in MMR proteins were found to be the driving mechanism behind tumours seen in Lynch syndrome (hereditary non-polyposis colorectal cancer). A similar phenotype of colorectal cancers with MSI is seen in a proportion of sporadic colorectal cancers with either a somatic mutation in MMR proteins or more commonly gene silencing through promoter methylation [14]. MMR status (dMMR or proficient [pMMR] corresponding to MSI-H or microsatellite stability, respectively, see Figure 1) can be determined by direct analysis of MSI (polymerase chain reaction analysis) or by immunohistochemistry for the MMR proteins (MLH1, MSH1, MSH6 and PMH2). With high sensitivity (92.3%) and specificity (100%) [15,16], immunohistochemistry is cheaper and simpler than polymerase chain reaction analysis, and provides a simple and effective means of establishing MMR status that is deliverable in a conventional setting. Specifically, the absence of immunohistochemical staining for one or more of these proteins implies dMMR (i.e. MSI-H).
Strong Evidence of a Prognostic Effect There has been growing evidence that dMMR status is a strong prognostic biomarker for improved outcomes in
0936-6555/$36.00 Ó 2013 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.clon.2013.07.006
626
B. O’Leary, D.C. Gilbert / Clinical Oncology 25 (2013) 625e629
Mismatch repair status
Abbreviation
Deficient
dMMR
Microsatellite status = Microsatellite instability
(loss of staining for MLH1, MSH1, MSH6 or PMH2)
Proficient
Abbreviation MSI or MSI-H
(detectable via polymerase chain reaction)
pMMR
= Microsatellite stable
MSS
Fig 1. Terminology.
colorectal cancer (Table 1). A meta-analysis in 2005 derived a pooled hazard ratio of 0.65 for overall survival in favour of dMMR (95% confidence interval 0.59e0.71) from 32 studies comprising 7642 patients [17]. This advantage held for stage II and III patients, with a pooled hazard ratio of 0.67 (95% confidence interval 0.58e0.78). A subsequent 2010 metaanalysis found an increased overall survival for dMMR patients from 20 studies comprising 9243 patients with stage IeIV disease [18]. This result was also seen in an analysis of the subset of 10 studies of stage II and II colorectal cancers with an odds ratio for survival at completion of study follow-up of 0.65 (95% confidence interval 0.53e0.79, P < 0.0001). Similar results were seen in the studies where data were available for disease-free survival. Retrospective analysis of 1913 patients enrolled into QUASAR showed this prognostic effect of dMMR in stage II colorectal cancer, highlighting the association with clinical
features [19]. Twenty-six per cent of right-sided tumours were dMMR as opposed to 3% of left-sided tumours and 1% of rectal cancers. dMMR rates were twice as frequent in stage II as opposed to stage III tumours (12% versus 6%). Crucially dMMR tumours showed about half the recurrence risk of their pMMR counterparts, with a recurrence rate of 11% in dMMR versus 26% in pMMR tumours (risk ratio 0.53, 95% confidence interval 0.40e0.70; P < 0.001). Similar results have recently been published from a population-based series from a single institution in Norway [20]. Complete data were available for 613 patients undergoing curative resection for stage II colorectal cancer. Fourteen per cent of cases were dMMR (defined by polymerase chain reaction for MSI), more frequently seen in women (19%) and right-sided tumours (29%). Although outcomes were worse that seen in the QUASAR clinical trial cohort, MSI-positive (dMMR) stage II cases had significantly improved 5 year relapse-free survival.
Does Mismatch Repair Status Predict Response to 5-fluorouracil? Beyond the prognostic effect there are accumulating data that MMR status is predictive of the response to adjuvant chemotherapy (Table 2), where dMMR tumours may not derive benefit from 5-FU and indeed may do worse. Mechanistic data support a role for dMMR tumours exhibiting resistance to 5-FU [25]. Clinical data are derived from adjuvant trials or cohort studies [17,18,24], although large numbers of patients are needed to amass the numbers of
Table 1 Studies reporting mismatch repair status as a prognostic marker in stage II and III colorectal cancer Reference
Design
Number of stage II and III
Prognostic marker
[17]
Meta-analysis
2935
[18]
Meta-analysis
4014
[21]
Retrospective from RCTs
457
[22]
Prospective in RCTs
1852
[19]
Retrospective from RCTs
1913
[23]
Retrospective cohort
787
[20]
Retrospective cohort
613
Yes HR ¼ 0.67 for OS (95% CI 0.58e0.78) in dMMR Yes OR ¼ 0.65 for survival at end of study (95% CI 0.53e0.79) in dMMR, P < 0.001 Yes HR ¼ 0.46 for DFS (95% CI 0.22e0.95) in dMMR, univariate analysis, P ¼ 0.03 Yes HR ¼ 0.77 (95% CI 0.71e0.80) for OS in dMMR, P ¼ 0.029 Yes RR ¼ 0.53 (95% CI 0.40e0.70) for recurrence in dMMR, P < 0.001 Yes HR ¼ 0.40 for OS at 3 years (95% CI 0.19e0.86) in dMMR, P ¼ 0.001 Yes HR ¼ 1.60 for RFS (95% CI 1.01e2.52) in pMMR, P ¼ 0.045
Notes
Benefit not maintained in multivariate analysis
Calculation includes stage I and IV patients Result due to stage II rather than stage III, analysis included stage I
RCT, randomised controlled trial; HR, hazard ratio; RR, risk ratio; OR, odds ratio; OS, overall survival; DFS, disease-free survival; RFS, relapsefree survival; dMMR, deficient mismatch repair; pMMR, proficient mismatch repair.
B. O’Leary, D.C. Gilbert / Clinical Oncology 25 (2013) 625e629
627
Table 2 Studies reporting mismatch repair status as a predictive marker in stage II and III colorectal cancer Reference
Design
Number of stage II and III
[17]
Meta-analysis
963
[24]
Meta-analysis
3690
[18]
Meta-analysis
2863
[21]
Retrospective from RCTs
457
[19]
Retrospective from RCTs
1913
[23]
Retrospective cohort
787
Predictive benefit with 5-FU for dMMR
Predictive benefit with 5-FU for pMMR
No HR ¼ 1.24 for OS (95% CI 0.72e2.14) in dMMR
Yes HR ¼ 0.72 for OS (95% CI 0.57e0.92) in pMMR, P ¼ 0.07 Trend towards benefit in pMMR versus dMMR
No HR ¼ 0.7 for OS (95% CI 0.44e1.09) in dMMR, P ¼ 0.86 No OR for survival at end of study ¼ 1.03 (95% CI 0.25e4.3) in dMMR No HR ¼ 1.39 for DFS (95% CI 0.46e4.15) in dMMR, P ¼ 0.56 No OR 0.72 for survival at end of study (95% CI 0.40e1.27) in dMMR No HR ¼ 0.57 for OS at 3 years (95% CI 0.09e3.75) in dMMR, P ¼ 0.566
Yes OR for survival at end of study ¼ 0.52 (95% CI 0.4e0.6) in pMMR, P < 0.0001 Yes HR ¼ 0.67 for DFS (95% CI 0.48e0.93) in pMMR, P ¼ 0.02 No OR 0.86 for survival at end of study (95% CI 0.73e1.02) in pMMR Yes HR ¼ 0.34 for OS at 3 years (95% CI 0.48e0.57) in pMMR, P < 0.001
Notes
Inadequate sample size for by-stage analysis for dMMR
Analysis included stages IIeIV
5-FU, 5-fluorouracil; RCT, randomised controlled trial; HR, hazard ratio; RR, risk ratio; OR, odds ratio; OS, overall survival; DFS, disease-free survival; RFS, relapse-free survival; dMMR, deficient mismatch repair; pMMR, proficient mismatch repair.
events required to show any association between dMMR and response. A systematic review and meta-analysis [24] identified seven studies comprising 3690 patients comparing untreated patients with those treated with 5-FUbased adjuvant chemotherapy for stage II or III colorectal cancer, all of whom were tested for dMMR status. No significant increase in relapse-free survival with chemotherapy was found in the pooled estimate for dMMR tumours. Three further datasets have since been published examining MMR as a predictive marker, with none able to show an advantage for adjuvant 5-FU in dMMR colorectal cancer, including the large dataset provided by the QUASAR study [19,21,23]. Importantly, by combining their data with their previous analysis, Sargent et al. [21] showed a small, but statistically significant, reduction in overall survival for stage II dMMR patients receiving systemic treatment (hazard ratio 2.95, 95% confidence interval 1.02e8.54, P ¼ 0.04).
Future Questions and Directions Where Now for Stage II, Mismatch Repair-proficient (Microsatellite-stable) Colorectal Cancer?
and predictive biomarkers to identify groups who may have a more tangible benefit from adjuvant treatment? BRAF mutations are associated with poorer outcomes within the setting of pMMR stage II and III colorectal cancer [26], but may not be predictive of a response to chemotherapy. A number of gene expression panels are undergoing validation, but none is in routine use [27]. Is There a Role for Mismatch Repair Status in the Management of Stage III Disease? Although dMMR status may confer resistance to 5-FU, it is unknown how MMR status may affect sensitivity to oxaliplatin. One recent study found dMMR to be of no value in predicting the response to oxaliplatin in stage III colorectal cancer [28]. Interestingly, a retrospective analysis of patients in the otherwise negative CALGB 89803 study showed improved disease-free survival for dMMR versus pMMR receiving adjuvant irinotecan, 5-FU and leucovorin in stage III colorectal cancer, a relationship not seen in the cohort receiving 5-FU/leucovorin [29].
Conclusion If dMMR stage II patients can be spared adjuvant chemotherapy for the prognostic and predictive reasons described above, is there a way to further stratify stage II pMMR patients with further combinations of prognostic
Colorectal cancer is a genetically diverse disease displaying a number of important phenotypes. dMMR patients for the most part have an excellent prognosis (with 6 year
628
B. O’Leary, D.C. Gilbert / Clinical Oncology 25 (2013) 625e629
survival in one study as high as 97% [30]) and can be spared additional treatment, which may even result in worse outcomes. Immunohistochemistry for the MMR proteins (MLH1, MSH1, MSH6, PMH2) is readily available and a robust marker of dMMR. The role of MMR status in decision-making for adjuvant chemotherapy in stage III disease with oxaliplatin or further investigating a role for irinotecan is more uncertain. In the future however, dMMR status will probably be part of genetic signature profiles that will allow more precise allocation of adjuvant treatments according to expected benefit. Finally, if a wider cohort of patients is to be tested for MMR status, clinicians must be aware of the increased likelihood of germline mutations in patients found to be dMMR, and be ready to refer on for genetic counselling, testing for Lynch syndrome and subsequent patient treatment and surveillance. In conclusion, we believe that the time is now here for routine testing of stage II colorectal cancer for MMR status, with dMMR (MSI-H) tumours for the most part being able to avoid additional treatment.
[11]
[12] [13]
[14] [15]
[16]
[17]
[18]
References [1] Moertel CG, Fleming TR, Macdonald JS, et al. Levamisole and fluorouracil for adjuvant therapy of resected colon carcinoma. New Engl J Med 1990;322(6):352e358. [2] O’Connell MJ, Laurie JA, Kahn M, et al. Prospectively randomized trial of postoperative adjuvant chemotherapy in patients with high-risk colon cancer. J Clin Oncol 1998;16(1):295e300. [3] Wolmark N, Rockette H, Fisher B, et al. The benefit of leucovorin-modulated fluorouracil as postoperative adjuvant therapy for primary colon cancer: results from National Surgical Adjuvant Breast and Bowel Project protocol C-03. J Clin Oncol 1993;11(10):1879e1887. [4] Kuebler JP, Wieand SH, O’Connell MJ, et al. Oxaliplatin combined with weekly bolus fluorouracil and leucovorin as surgical adjuvant chemotherapy for stage II and III colon cancer: results from NSABP C-07. J Clin Oncol 2007;25(16):2198e2204. T, Boni C, Navarro M, et al. Improved overall survival [5] Andre with oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment in stage II or III colon cancer in the MOSAIC trial. J Clin Oncol 2009;27(19):3109e3116. [6] IMPACT B2 Investigators. Efficacy of adjuvant fluorouracil and folinic acid in B2 colon cancer. J Clin Oncol 1999;17(5):1356. [7] Mamounas E, Wieand S, Wolmark N, et al. Comparative efficacy of adjuvant chemotherapy in patients with Dukes’ B versus Dukes’ C colon cancer: results from four National Surgical Adjuvant Breast and Bowel Project Adjuvant Studies (C-01, C-02, C-03, and C-04). J Clin Oncol 1999;17(5):1349. [8] Gill S, Loprinzi CL, Sargent DJ, et al. Pooled analysis of fluorouracil-based adjuvant therapy for stage II and III colon cancer: who benefits and by how much? J Clin Oncol 2004;22(10):1797e1806. [9] Quasar Collaberative Group. Adjuvant chemotherapy versus observation in patients with colorectal cancer: a randomised study. Lancet 2007;370(9604):2020e2029. [10] Niedzwiecki D, Bertagnolli MM, Warren RS, et al. Documenting the natural history of patients with resected stage II
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
adenocarcinoma of the colon after random assignment to adjuvant treatment with edrecolomab or observation: results from CALGB 9581. J Clin Oncol 2011;29(23):3146e3152. Ionov Y, Peinado MA, Malkhosyan S, et al. Ubiquitous somatic mutations in simple repeated sequences reveal a new mechanism for colonic carcinogenesis. Nature 1993;363(6429):558e561. Fearon ER, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell 1990;61(5):759e767. Thibodeau S, Bren G, Schaid D. Microsatellite instability in cancer of the proximal colon. Science 1993;260(5109): 816e819. Boland CR, Goel A. Microsatellite instability in colorectal cancer. Gastroenterology 2010;138(6):2073e2087. Lindor NM, Burgart LJ, Leontovich O, et al. Immunohistochemistry versus microsatellite instability testing in phenotyping colorectal tumors. J Clin Oncol 2002;20(4):1043e1048. Ward RL, Turner J, Williams R, et al. Routine testing for mismatch repair deficiency in sporadic colorectal cancer is justified. J Pathol 2005;207(4):377e384. Popat S, Hubner R, Houlston RS. Systematic review of microsatellite instability and colorectal cancer prognosis. J Clin Oncol 2005;23(3):609e618. Guastadisegni C, Colafranceschi M, Ottini L, et al. Microsatellite instability as a marker of prognosis and response to therapy: a meta-analysis of colorectal cancer survival data. Eur J Cancer 2010;46(15):2788e2798. Hutchins G, Southward K, Handley K, et al. Value of mismatch repair, KRAS, and BRAF mutations in predicting recurrence and benefits from chemotherapy in colorectal cancer. J Clin Oncol 2011;29(10):1261e1270. Merok MA, Ahlquist T, Royrvik EC, et al. Microsatellite instability has a positive prognostic impact on stage II colorectal cancer after complete resection: results from a large, consecutive Norwegian series. Ann Oncol 2013;24(5):1274e1282. Sargent DJ, Marsoni S, Monges G, et al. Defective mismatch repair as a predictive marker for lack of efficacy of fluorouracil-based adjuvant therapy in colon cancer. J Clin Oncol 2010;28(20):3219e3226. Bertagnolli MM, Redston M, Compton CC, et al. Microsatellite instability and loss of heterozygosity at chromosomal location 18q: prospective evaluation of biomarkers for stages II and III colon cancerda study of CALGB 9581 and 89803. J Clin Oncol 2011;29(23):3153e3162. Hong SP, Min BS, Kim TI, et al. The differential impact of microsatellite instability as a marker of prognosis and tumour response between colon cancer and rectal cancer. Eur J Cancer 2012;48(8):1235e1243. Des Guetz G, Schischmanoff O, Nicolas P, et al. Does microsatellite instability predict the efficacy of adjuvant chemotherapy in colorectal cancer? A systematic review with meta-analysis. Eur J Cancer 2009;45(10):1890e1896. Carethers JM, Chauhan DP, Fink D, et al. Mismatch repair proficiency and in vitro response to 5-fluorouracil. Gastroenterology 1999;117(1):123e131. Roth AD, Tejpar S, Delorenzi M, et al. Prognostic role of KRAS and BRAF in stage II and III resected colon cancer: results of the translational study on the PETACC-3, EORTC 40993, SAKK 60-00 trial. J Clin Oncol 2010;28(3):466e474. Salazar R, Roepman P, Capella G, et al. Gene expression signature to improve prognosis prediction of stage II and III colorectal cancer. J Clin Oncol 2011;29(1):17e24. Li P, Fang YJ, Li F, et al. ERCC1, defective mismatch repair status as predictive biomarkers of survival for stage III colon
B. O’Leary, D.C. Gilbert / Clinical Oncology 25 (2013) 625e629 cancer patients receiving oxaliplatin-based adjuvant chemotherapy. Br J Cancer 2013;108(6):1238e1244. [29] Bertagnolli MM, Niedzwiecki D, Compton CC, et al. Microsatellite instability predicts improved response to adjuvant therapy with irinotecan, fluorouracil, and leucovorin in stage
629
III colon cancer: Cancer and Leukemia Group B Protocol 89803. J Clin Oncol 2009;27(11):1814e1821. [30] Lanza G, Gafa R, Sanyini A, et al. Immunohistochemical test for MLH1 and MSH2 expression predicts clinical outcome in stage II and III colorectal cancer patients. J Clin Oncol 2006;24(15):2359e2367.