Mismatch Repair Status Fails as an Independent Predictor of Survival in Modern Stage II/III Colorectal Cancer: Evidence from a Propensity Score-Matched Cohort

Purpose: Traditionally, Deficient Mismatch Repair(dMMR) has been associated with a favorable prognosis in early-stage colorectal cancer (CRC). dMMR acts as a predictive biomarker for immunotherapy response in metastatic CRC while its role in earlier-stage disease is still being evaluated in ongoing clinical trials. This study aims to re-evaluate the independent prognostic value of MMR status in stage II/III CRC within the framework of standardized modern clinical care. Methods We identified 972 patients with Stage II/III CRC (2017–2019) who underwent curative resection in a tertiary center. A 1:3 propensity score matching (PSM) was performed to balance age, gender, tumor location, pathological stage and chemotherapy. The endpoint was 5-year overall survival (OS), presented using the Kaplan-Meier analysis, compared via log-rank test and Multivariable Cox proportional hazards models were used to identify independent prognostic factors. Results After PSM, 5-year overall survival (OS) showed no significant difference for stage II(p = 0.79) and stage III (p = 0.99). Multivariate analysis confirmed that MMR status was not an independent predictor of OS (HR0.85, p = 0.70), whereas tumor-infiltrating lymphocytes (TILs), tumor location and patient age remained significant independent factors. Conclusion In the current treatment era, the use of oxaliplatin-based regimens has narrowed the survival gap between dMMR and proficient mismatch repair(pMMR) and dMMR has lost its role as an independent predictor of survival. Risk stratification should incorporate TILs, age and tumor location rather than relying on MMR status alone.