Neoadjuvant PD-1-Based Immunotherapy in Localized dMMR/MSI-H Colorectal Cancer: A Systematic Review and Arm-Based Meta-Analysis

Abstract

BACKGROUND/OBJECTIVES: Deficient mismatch repair (dMMR) and microsatellite instability-high (MSI-H) colorectal cancer (CRC) are highly sensitive to immune checkpoint blockade. We evaluated the efficacy, safety, and clinical maturity of neoadjuvant PD-1-based immunotherapy in localized disease. METHODS: This PRISMA-compliant systematic review included prospective studies of neoadjuvant PD-1-based therapy in localized dMMR/MSI-H CRC. Proportions were pooled on the logit scale using inverse-variance random-effects models with the Paule-Mandel estimator. Pathologic complete response (pCR) was the primary outcome; major pathologic response (MPR), immune-related adverse events (irAEs), surgery, clinical complete response (cCR), organ preservation, and long-term outcomes were secondary outcomes. RESULTS: Five prospective studies were included. Four studies, comprising 305 treated patients and 292 pathologically evaluable patients, contributed five treatment arms to the quantitative synthesis. The pooled pCR proportion was 0.65 (95% CI, 0.54-0.74), and the pooled MPR proportion was 0.89 (95% CI, 0.79-0.94). Exploratory subgroup analyses suggested higher response proportions with combination regimens, but clinical and methodological differences confounded these predominantly indirect comparisons. Any-grade irAEs occurred in 0.54 (95% CI, 0.36-0.71), whereas grade ≥3 irAEs occurred in 0.06 (95% CI, 0.04-0.10). The pooled surgery proportion was 0.96 (95% CI, 0.86-0.99), although this outcome largely reflected protocol-mandated surgery. cCR and organ preservation were promising in selected rectal-cancer cohorts. Survival and recurrence data were immature and unsuitable for quantitative pooling. CONCLUSIONS: Neoadjuvant PD-1-based immunotherapy is a highly promising investigational strategy for localized dMMR/MSI-H CRC. Nevertheless, predominantly early-phase evidence, indirect comparisons, heterogeneous endpoints, and limited follow-up preclude definitive conclusions regarding routine clinical adoption.