Pathway Mutation Accumulate Perturbation Score: A prognostic and predictive biomarker for immunotherapy in advanced gastric cancer

BACKGROUND: Gastric cancer (GC) remains highly lethal, with limited available biomarkers. The Pathway Mutation Accumulation Perturbation Score (PMAPscore), which leverages pathway-level mutations, provides a novel approach to predicting immunotherapy response and survival outcomes. OBJECTIVES: To evaluate the prognostic and predictive value of the PMAPscore in patients with advanced GC undergoing immunotherapy. MATERIAL AND METHODS: Three cohorts of patients with GC treated with immunotherapy were analyzed: PUCH (Peking University Cancer Hospital; n = 39, training cohort), MSK (Memorial Sloan Kettering; n = 19, validation cohort) and SMC (Samsung Medical Center; n = 43, validation cohort). A PMAPscore-based risk model was developed and validated for survival outcomes. Immune mechanisms in highand low-risk groups were explored using The Cancer Genome Atlas (TCGA) GC data (n = 431). RESULTS: Low-risk patients identified by the PMAPscore model exhibited significantly better progression-free survival (PFS), overall survival (OS) and durable clinical benefit (DCB). In the PUCH cohort, low-risk patients had higher DCB rates (85.7% vs 44.0%, p = 0.02), longer PFS (p < 0.001), and longer OS (p < 0.001). Similar trends were observed in the MSK and SMC cohorts. Multivariate analysis confirmed low-risk status as an independent predictor of improved PFS and OS, outperforming tumor mutation burden (TMB), programmed death-ligand 1 (PD-L1) expression, microsatellite instability (MSI) status, and the gastrointestinal immune prognostic signature (GIPS). The TCGA data indicated enhanced antitumor immune activity in low-risk tumors, with increased human leukocyte antigen (HLA)-related gene expression and greater B-cell and natural killer (NK) cell infiltration. CONCLUSIONS: The PMAPscore-based risk model is a robust tool for predicting survival and immunotherapy responses in patients with advanced GC, supporting its clinical application for treatment stratification.