For locally advanced rectal cancer (LARC) with a deficient mismatch repair/microsatellite instability-high (dMMR/MSI-H), particularly in patients not eligible for immunotherapy, the optimal treatment remains undetermined. This study was to evaluate the efficacy and safety of surgery, surgery and chemotherapy, surgery and chemoradiotherapy, in patients with LARC harboring dMMR/MSI-H.Patients included from three university centers between August 1, 2012 and March 1, 2023, were categorized into three treatment groups: surgery vs. surgery + chemotherapy vs. surgery + chemoradiotherapy. The primary endpoint was overall survival (OS), with secondary endpoints of progression-free survival (PFS), local recurrence (LR), distant metastasis (DM), and toxicity. The Kaplan-Meier method was utilized to analyze OS and PFS; competing risk methods were employed to evaluate rates of LR and DM. Adjustments were performed utilizing inverse probability of treatment weighting (IPTW) and overlap weighting (OW) based on propensity score, employing logistic regression model. The Cox proportional hazards model was applied for both univariate and multivariate analyses to assess prognostic factors influencing patient OS and PFS.A total of 119 patients were included, with 45 patients (37.8%) receiving surgery alone, 32 (26.9%) receiving surgery + chemotherapy, and 42 (35.3%) undergoing surgery + chemoradiotherapy. In both the unadjusted cohort and after IPTW and OW adjustments, the surgery alone group (vs. surgery + chemoradiotherapy) had improved OS, PFS, LR, but no significant differences in DM. However, no statistical difference was found between the surgery vs. surgery + chemotherapy groups in OS, PFS, and DM, except for significant differences in LR. Similar results were found in both neoadjuvant and adjuvant treatment cohorts. No adverse events of grade 5 occurred.This study suggests surgery alone (without chemotherapy and/or radiotherapy) may be an optimal treatment for LARC patients with dMMR/MSI-H, particularly in those who cannot tolerate or access immunotherapy. The results of this study may be used to power a randomized trial for the approaches.Copyright © 2025 Wang, Yan, Zeng, Wang, Ni, Yu, Shi, Li, Wang, Yuan, Wen, Zaorsky, Zhang, Zang and Meng.