Mismatch repair deficiency confers worse survival in stage IV colon cancer

Introduction

Colon cancer (CC) is a common and lethal disease with an estimated 106,590 cases diagnosed annually in the United States whose mortality is strongly correlated with metastatic disease (1-3). The pathogenesis of CC can be outlined according to two distinct pathways: chromosomal instability (CIN), and microsatellite instability (MSI), which differ according to the functionality of their mismatch repair (MMR) systems. During normal DNA replication, MMR genes such as MSH2, MLH1, MSH6, and PMS2 detect and repair DNA mismatch errors to form corrected DNA strands. CCs that exhibit CIN possess functional MMR systems and are referred to as MMR-proficient (MMR-P), while MSI tumors characteristically exhibit deficits in their MMR systems and are referred to as MMR-deficient (MMR-D).

MMR-P/CIN accounts for approximately 85% of all CCs and is caused by mutations in specific tumor suppressor genes and oncogenes, translating to aneuploidy, loss of heterozygosity, and structural chromosomal changes (4,5). Meanwhile, MMR-D/MSI CCs account for approximately 15% of all CCs; their MSI arises consequent to defective MMR systems rendered unable to repair single nucleotide mismatches (4-6). MMR-D/MSI CCs can manifest via mutations in the germline, as seen in Lynch Syndrome, or as sporadic epigenetic modifications of MMR genes (4,6). Lynch Syndrome comprises roughly 20% of all MMR-D CCs (and roughly 3–5% of all CCs), while the majority (roughly 80%) of MMR-D CCs arise via sporadic epigenetic inactivation of the MMR gene promoters (most commonly, MLH1) via DNA hypermethylation in CpG islands via the CpG Island Methylator Phenotype (CIMP) pathway (5,6). Almost 50% of MMR-D CCs are known to harbor v-raf murine sarcoma viral oncogene homolog B1 (BRAF) mutations (5,7,8).

MMR-D CCs, irrespective of whether in the setting of CIMP or Lynch Syndrome, exhibit distinct pathologic and epidemiologic features. Specifically, MMR-D has been associated with predominance in the proximal colon, poor differentiation, and mucinous histology (5,6). MMR-D is considered a favorable prognostic factor among patients with non-metastatic CC (stages I–III) (8-12), however, its prognostic role in patients with metastatic (stage IV) colon cancer (MCC) remains relatively unknown, and few studies have examined this relationship (8). This consideration is crucial, given that at the earliest diagnosis, distant metastatic disease exists in typically 25% of patients with CC (2,3).

Multiple meta-analyses have established MMR-D’s apparent survival-protective effect when compared to MMR-P, however, the vast majority of patients included in these pooled studies had non-metastatic CC, and the relationship between MMR-D and mortality among patients with MCC remains poorly understood (13-15). While the relatively sparse reporting on MMR-D MCC may be explained by the decreased tendency of MMR-D to exhibit lymph node and distant organ metastasis, as noted by Malesci et al. (16), targeted analysis of MMR status among patients with MCC is warranted to further understand the disease process and course (16). Thus, the aims of this study are to compare sociodemographic and clinicopathologic features and mortality between MMR-P and MMR-D in patients with MCC. We present this article in accordance with the STROBE reporting checklist (available at https://jgo.amegroups.com/article/view/10.21037/jgo-24-387/rc).

Methods

Database description

We performed an 8-year [2010–2017] retrospective analysis of the National Cancer Database (NCDB). The NCDB is jointly sponsored by the American College of Surgeons and the American Cancer Society; it collects data from more than 1,500 Commission on Cancer (CoC) accredited facilities across the United States and Puerto Rico (11). The NCDB is updated with 72% of all new types of cancer diagnoses at the national level with a total of over 40 million records (17). The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). This study was exempted from the Institutional Review Board as the NCDB contains no patient-identifying information. The NCDB is publicly available, and acquisition of consent to participate in the study is not applicable.

Patient selection and stratification

We included adult (age ≥18 years) patients diagnosed with stage IV CC between the years 2010-2017 with reported MMR status. The International Classification of Diseases for Oncology, 3rd edition (ICD-O-3) was used to identify patients with stage IV CC. ICD-O-3 histology codes 8140-8263 (adenocarcinoma), 8480-8481 (mucinous adenocarcinoma), 8490 (signet ring cell carcinoma), 8510-8513 (medullary carcinoma), and remaining codes (other) were reported as tumor characteristics. Patients were then stratified into two groups based on MMR status.

Of the 1,088,765 patients with reported CC in the NCDB from 2010–2017, 255,280 were found to have stage IV disease at the time of diagnosis. MSI testing was ordered and reported for 34,438 of these patients. Patients with reported stage IV CC and MMR status were further excluded from the study population if there was missing data regarding patient sociodemographic characteristics, primary tumor locations and characteristics, treatment modalities, and mortality. There were 23,516 patients that did not have reported information regarding one or more of the aforementioned categories, and were excluded from our analysis, yielding 10,922 patients who met inclusion criteria.

Although all 10,922 patients included had confirmed metastatic disease, data on specific metastatic sites was only made available beginning in 2016 in the NCDB which equated to 2,363 patients in our population.

Parameters assessed and outcomes

The following variables were collected from the patient population: sex, age at MCC diagnosis, race, ethnicity, facility location, primary tumor site, distant lymph node involvement, metastatic sites (when applicable), and treatment modalities. Outcomes included 90-day, 180-day, 1-year, and 2-year overall mortality after MCC diagnosis.

Primary tumor sites were encoded as “right colon” if occurring in either the cecum, ascending colon, or hepatic flexure, and as “left colon” if occurring in the splenic flexure, descending colon, sigmoid colon, or rectosigmoid junction.

Statistical analysis

Our results are reported as percentages for categorical variables and as mean ± standard deviation for continuous variables. We used a Chi-squared test to analyze differences between categorical variables and an independent t-test for differences between the continuous variables.

A bivariate logistic regression was performed to identify factors independently associated with MMR-D. Variables with a P value <0.1 in the univariate analysis were incorporated into the multivariate model. The model fit was assessed using the Hosme-Lemeshow test, which exceeded 0.05. The tolerance was above 0.1 for all independent variables with a variance inflation factor less than 10.0.

Survival and mortality were assessed using Kaplan-Meier and univariate Cox regression analysis for categorical and continuous variables, respectively. Variables with a P value of <0.1 in the univariate analyses were incorporated in a multivariate Cox regression analysis, including baseline sociodemographic and histopathological features regardless of statistical significance. A P value of less than 0.05 was considered statistically significant. All statistical analyses were performed on the Statistical Package for the Social Sciences software (SPSS, version 26; SPSS, Inc.).

Results

A total of 10,922 MCC cases were identified; 8,796 (80.53%) were MMR-P and 2,126 (19.47%) were MMR-D. The overall mean age at MCC diagnosis was 63.25±12.621 years, 49.7% (n=5,435) were female, and 81.7% (8,928) were White. Patients with MMR-P were more likely to have private insurance (45.4% vs. 38.6%, P<0.001) and patients with MMR-D were more likely to have Medicare (48.6% vs. 41.2%, P<0.001). The full breakdown of individual sociodemographic and clinicopathologic characteristics associated with MMR-P and MMR-D are summarized in Tables 1-3.

For 2,363 patients for whom data on metastatic sites was available, MMR-P was associated with higher rates of liver metastasis (73.0% vs. 60.8%, P<0.001). MMR-D was associated with higher rates of distant lymph node metastasis (15.5% vs. 10.2%, P=0.003) and involvement of other unspecified metastatic sites (35.5% vs. 28.2%, P<0.001) as seen in Table 4.

As shown in Table 5, patients with MMR-P were more likely to undergo chemotherapy (78.9% vs. 71.8%, P<0.001) and immunotherapy (23.0% vs. 16.3%, P<0.001) treatment, but both groups received surgical resection of the primary tumor at comparable rates (99.2% vs. 99.4%, P=0.29) and had no significant difference in the timing from MCC diagnosis to initiation of treatment with any intervention (15.82±30.40 vs. 15.27±26.47 days, P=0.44). MMR-D was associated with decreased mean survival after MCC diagnosis (23.96±21.53 vs. 26.44±20.41 months, P<0.001).

As shown in Table 6, factors independently associated with MMR-D included older mean age at diagnosis [65.16 vs. 62.79 years, odds ratio (OR) 1.005, 95% confidence interval (CI): 1.000–1.010, P=0.04], female sex (53.1% vs. 49.0%, OR 1.123, 95% CI: 1.019–1.237, P=0.02), mucinous adenocarcinoma (15.1% vs. 8.8%, OR 1.910, 95% CI: 1.192–3.061, P=0.007), signet ring carcinoma (4.2% vs. 2.2%, OR 1.962, 95% CI: 1.168–3.295, P=0.01), medullary carcinoma (1.4% vs. 0.1%, OR 10.243, 95% CI: 4.506–23.283, P<0.001), and lymph-vascular invasion (67.4% vs. 62.9%, OR 1.211, 95% CI: 1.087–1.350, P=0.001). Factors independently associated with MMR-P included perineural invasion (35.7% vs. 33.8%, OR 0.883, 95% CI: 0.793–0.983, P=0.02), chemotherapy treatment (78.9% vs. 71.8% OR 0.866, 95% CI: 0.766–0.979, P=0.02), immunotherapy treatment (23.0% vs. 16.3% OR 0.746, 95% CI: 0.653–0.851, P<0.001), Black race (14.2% vs. 12.4% OR 0.716, 95% CI: 0.543–0.945, P=0.02), and primary tumor site in the left side of the colon (45.9% vs. 34.0%, OR 0.718, 95% CI: 0.555–0.929, P=0.01).

When adjusted for demographics, histologic features, and treatment modalities, MMR-D was associated with higher overall mortality at 180 days [21.6% vs. 15.0%, hazard ratio (HR) 1.140, 95% CI: 1.023–1.270, P=0.02], 1 year (36.5% vs. 26.0%, HR 1.203, 95% CI: 1.107–1.307, P<0.001), and 2 years (54.8% vs. 46.2%, HR 1.120, 95% CI: 1.048–1.197, P=0.001) after MCC diagnosis. Although mortality rates were higher for MMR-D at 90 days, the difference was significant in the univariate analysis but not the multivariate analysis. The full results for the univariate and multivariate Cox regression analysis are seen in Table 7.

Discussion

Our study, to the best of our knowledge, is the largest to describe patients with MCC according to MMR status and shows that MMR-D is an independent risk factor for mortality. This is significant, given that MMR-D is considered a positive prognostic factor in non-metastatic CC (stages I–III) (8-12). Sociodemographic characteristics associated with MMR-D include older age at MCC diagnosis, female sex, non-Black race and decreased rates of private health insurance compared to MMR-P patients. These findings are generally consistent with prior studies showing that across patients with CC of any stage, MMR-D is associated with right-sided colonic predominance, older age at diagnosis, female sex, and White race (18,19).

MMR-D was also found to be associated with specific clinicopathologic characteristics, including increased rates of mucinous adenocarcinoma, signet ring carcinoma, and medullary carcinoma, and decreased predilection for the distal colon, which are also largely consistent with prior studies in patients with CC of any stage (5,6). Although data on specific metastatic sites was only available for a subgroup of patients in our analysis, MMR-D appears to be associated with decreased rates of metastasis to the liver, and higher rates of metastasis to distant lymph nodes and other unspecified sites not including the lung, bone, or brain. A 2007 retrospective study by Malesci et al. reported that MSI tumors were less likely to exhibit lymph node and distant organ metastasis; 61.8% of MSI and 43.5% of MSS cancers in their study had no lymph node or distant organ metastasis at diagnosis (16).

Although MMR-D was associated with higher mortality at all investigated time points in the univariate analysis, the multivariate analysis adjusted for age, race, treatment modalities, and histologic features found that MMR-D was an independent predictor of mortality at 180 days, 1 year, and 2 years after MCC diagnosis.

One possible reason for the increased mortality associated with MMR-D may be due to the contribution of specific mutations known to be associated with MMR-D, such as BRAF and the type II receptor for transforming growth factor β1 (TGF-β1), which we were unable to examine in this study (7,20). Watanabe et al. found higher 5-year survival in patients with early-stage CC and MSI was higher among those with a mutated gene for the type II receptor for TGF-β1 compared to those with MSI tumors that did not have this mutation (20). A pooled analysis by Venderbosch et al. found that BRAF mutations were present in 34.6% of patients with MMR-D MCC compared to 6.8% of patients with MMR-P and conferred an inferior prognosis regardless of MMR status (21). Since data on these specific mutations were not available in the NCDB, it was not possible to reach a conclusion in this regard.

Patients with MMR-D were less frequently treated with chemotherapy and immunotherapy compared to MMR-P patients, however, nearly all patients across both groups underwent surgical resection of the primary tumor, and no significant differences in timing from diagnosis to treatment initiation were found. One explanation for this may be that many of the patients included in our study developed MCC prior to the Food and Drug Administration approving the use of anti-programmed cell death protein 1 (anti-PD1) antibody pembrolizumab in 2017 following the 2015 KEYNOTE-016 trial (22,23). The advent of newer immune therapies such as dostarlimab in treating MMR-D cancers was also not accounted for in our study timeframe; this is likely to refine our approach to treating MMR-D tumors and highlights the importance of continued surveillance of MMR status among CC patients (24). Although the multivariate Cox regression analysis was adjusted for variables including treatment modalities, it is possible that mortality rates for MMR-D patients with MCC may trend downward in the future given these developments.

It is also possible that MMR-D patients experienced more comorbidities than MMR-P patients, given that they were older at age of diagnosis and were less likely to have private insurance. However, given that information on specific medical comorbidities was not available in the NCDB, a possible component of the increased burden of more comorbidities cannot be definitively excluded. Furthermore, in our study, the incidence of MMR-D was 19.47% among patients with MCC, which is higher than what was found in prior studies from the earlier 2000s, which describe an incidence closer to 5% among patients with MCC (9,21,25). While this may be reflective of the relatively increased incidence of right-sided CCs observed in the United States recently given MMR-D CCs right-sided colonic predominance (26,27), it is possible that MMR-D patients may exhibit disproportionate screening and surveillance barriers, as indicated by their relative lack of private insurance as well as the low percentage of patients in the NCDB with MCC who had reported MMR status (34,438 of 255,280, or 13.5%).

Our study has limitations. Since this was a retrospective study using a large, de-identified database, we were unable to account for missing data and excluded variables. As previously mentioned, this included data on medical comorbidities, and specific mutations. Additionally, a small number of patients with MCC had reported MMR status, which may have skewed our results given the high incidence. A minority of the patient population had available information regarding specific metastatic sites, and these patients needed to be analyzed as a subgroup. Lastly, the mortality variables of 90 days, 180 days, 1 year, and 2 years after MCC diagnosis may not accurately reflect cancer-related mortality. Our study population also included predominantly White patients (81.3%), which may not optimally represent current United States demographics.

Despite these limitations, our study, to the best of our knowledge, is the largest to describe exclusively MCC patients according to MMR-D, and shows that MMR-D is an independent risk factor for overall mortality, despite being a favorable prognostic factor in non-metastatic CC.

The implications for patients given these findings are significant. The association of MMR-D with increasing age may be reflective of age-related accumulation of genomic instability, which may highlight a more aggressive tumor biology in these patients. The longer postoperative length of stay in patients with MMR-D could be indicative of more complex postoperative care needs, possibly due to the presence of multiple comorbidities often seen in older patients. The increased rates of lymph-vascular invasion seen in MMR-D patients may guide clinicians in approaching postoperative adjuvant therapy decisions. The potential for utilizing immune checkpoint inhibitors in this population could offer a promising area for future research.

Conclusions

Despite being a favorable prognostic factor in stages I–III CC, MMR-D is associated with higher mortality at 90 days, 180 days, 1 year, and 2 years from diagnosis in stage IV CC compared to MMR-P in our univariate analysis. When adjusting for age, race, treatment modalities, and histologic features, MMR-D was found to be an independent predictor of mortality at 180 days, 1 year, and 2 years from MCC diagnosis. MMR-D is associated with older age at MCC diagnosis, female sex, non-Black race, decreased rates of private health insurance, more aggressive primary tumor histologic features, increased rates of lymph-vascular invasion, decreased rates of metastasis to the liver and lung and distal colonic primary tumor sites compared to MMR-P. The reasons for observed survival differences possibly entail specific oncogenic mutations associated with MMR-D as well as the aforementioned sociodemographic comorbidities.

Acknowledgments

Funding: None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jgo.amegroups.com/article/view/10.21037/jgo-24-387/rc

Peer Review File: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-24-387/prf

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jgo.amegroups.com/article/view/10.21037/jgo-24-387/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013).

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Siegel RL, Giaquinto AN, Jemal A. Cancer statistics, 2024. CA Cancer J Clin 2024;74:12-49. [Crossref] [PubMed]
2. Vatandoust S, Price TJ, Karapetis CS. Colorectal cancer: Metastases to a single organ. World J Gastroenterol 2015;21:11767-76. [Crossref] [PubMed]
3. Alvarado-Bachmann R, Smith A, Gundara JS, et al. The incidence of mismatch repair gene defects in colorectal liver metastases. Mol Med Rep 2014;10:1003-6. [Crossref] [PubMed]
4. Worthley DL, Leggett BA. Colorectal cancer: molecular features and clinical opportunities. Clin Biochem Rev 2010;31:31-8. [PubMed]
5. Gupta R, Sinha S, Paul RN. The impact of microsatellite stability status in colorectal cancer. Curr Probl Cancer 2018;42:548-59. [Crossref] [PubMed]
6. Sinicrope FA. DNA mismatch repair and adjuvant chemotherapy in sporadic colon cancer. Nat Rev Clin Oncol 2010;7:174-7. [Crossref] [PubMed]
7. Weisenberger DJ, Siegmund KD, Campan M, et al. CpG island methylator phenotype underlies sporadic microsatellite instability and is tightly associated with BRAF mutation in colorectal cancer. Nat Genet 2006;38:787-93. [Crossref] [PubMed]
8. Taieb J, Shi Q, Pederson L, et al. Prognosis of microsatellite instability and/or mismatch repair deficiency stage III colon cancer patients after disease recurrence following adjuvant treatment: results of an ACCENT pooled analysis of seven studies. Ann Oncol 2019;30:1466-71. [Crossref] [PubMed]
9. Koopman M, Kortman GA, Mekenkamp L, et al. Deficient mismatch repair system in patients with sporadic advanced colorectal cancer. Br J Cancer 2009;100:266-73. [Crossref] [PubMed]
10. Gelsomino F, Barbolini M, Spallanzani A, et al. The evolving role of microsatellite instability in colorectal cancer: A review. Cancer Treat Rev 2016;51:19-26. [Crossref] [PubMed]
11. Sinicrope FA, Rego RL, Halling KC, et al. Prognostic impact of microsatellite instability and DNA ploidy in human colon carcinoma patients. Gastroenterology 2006;131:729-37. [Crossref] [PubMed]
12. Jin Z, Sinicrope FA. Prognostic and Predictive Values of Mismatch Repair Deficiency in Non-Metastatic Colorectal Cancer. Cancers (Basel) 2021;13:300. [Crossref] [PubMed]
13. Popat S, Hubner R, Houlston RS. Systematic review of microsatellite instability and colorectal cancer prognosis. J Clin Oncol 2005;23:609-18. [Crossref] [PubMed]
14. Toh JWT, Phan K, Reza F, et al. Rate of dissemination and prognosis in early and advanced stage colorectal cancer based on microsatellite instability status: systematic review and meta-analysis. Int J Colorectal Dis 2021;36:1573-96. [Crossref] [PubMed]
15. Wang J, Deng Z, Lang X, et al. Meta-Analysis of the Prognostic and Predictive Role of the CpG Island Methylator Phenotype in Colorectal Cancer. Dis Markers 2022;2022:4254862. [Crossref] [PubMed]
16. Malesci A, Laghi L, Bianchi P, et al. Reduced likelihood of metastases in patients with microsatellite-unstable colorectal cancer. Clin Cancer Res 2007;13:3831-9. [Crossref] [PubMed]
17. ACS. [cited 2023 Aug 18]. National Cancer Database. Available online: https://www.facs.org/quality-programs/cancer-programs/national-cancer-database/
18. Chen S, Du W, Cao Y, et al. Preoperative contrast-enhanced CT imaging and clinicopathological characteristics analysis of mismatch repair-deficient colorectal cancer. Cancer Imaging 2023;23:97. [Crossref] [PubMed]
19. Gutierrez C, Ogino S, Meyerhardt JA, et al. The Prevalence and Prognosis of Microsatellite Instability-High/Mismatch Repair-Deficient Colorectal Adenocarcinomas in the United States. JCO Precis Oncol 2023;7:e2200179. [Crossref] [PubMed]
20. Watanabe T, Wu TT, Catalano PJ, et al. Molecular predictors of survival after adjuvant chemotherapy for colon cancer. N Engl J Med 2001;344:1196-206. [Crossref] [PubMed]
21. Venderbosch S, Nagtegaal ID, Maughan TS, et al. Mismatch repair status and BRAF mutation status in metastatic colorectal cancer patients: a pooled analysis of the CAIRO, CAIRO2, COIN, and FOCUS studies. Clin Cancer Res 2014;20:5322-30. [Crossref] [PubMed]
22. Le DT, Uram JN, Wang H, et al. PD-1 Blockade in Tumors with Mismatch-Repair Deficiency. N Engl J Med 2015;372:2509-20. [Crossref] [PubMed]
23. Zhang X, Wu T, Cai X, et al. Neoadjuvant Immunotherapy for MSI-H/dMMR Locally Advanced Colorectal Cancer: New Strategies and Unveiled Opportunities. Front Immunol 2022;13:795972. [Crossref] [PubMed]
24. Cercek A, Lumish M, Sinopoli J, et al. PD-1 Blockade in Mismatch Repair-Deficient, Locally Advanced Rectal Cancer. N Engl J Med 2022;386:2363-76. [Crossref] [PubMed]
25. Braun MS, Richman SD, Quirke P, et al. Predictive biomarkers of chemotherapy efficacy in colorectal cancer: results from the UK MRC FOCUS trial. J Clin Oncol 2008;26:2690-8. [Crossref] [PubMed]
26. Islami F, Ward EM, Sung H, et al. Annual Report to the Nation on the Status of Cancer, Part 1: National Cancer Statistics. J Natl Cancer Inst 2021;113:1648-69. [Crossref] [PubMed]
27. Schub R, Steinheber FU. Rightward shift of colon cancer. A feature of the aging gut. J Clin Gastroenterol 1986;8:630-4. [Crossref] [PubMed]