Construction of a risk model based on folate metabolism-related genes to predict prognosis and immunological characteristics of stomach adenocarcinoma

Abstract

Given the substantial disease burden of stomach adenocarcinoma (STAD)—the dominant and lethal subtype of gastric cancer—and the pivotal role of folate metabolism reprogramming in malignancy and treatment response, this study aimed to construct a prognostic model utilizing folate metabolism-related genes (FMRGs). Such a model is urgently required to supplement traditional TNM staging and to guide personalized precision medicine. Interrogation of TCGA and GEO transcriptomic profiles enabled the identification of differentially expressed FMRGs. These genes subsequently facilitated the construction of a prognostic signature via LASSO-Cox regression, which was then subjected to external validation. We further evaluated the clinical relevance of this risk signature by exploring its correlations with the tumor immune microenvironment, immunotherapy efficacy, and drug sensitivity, employing CIBERSORT and ssGSEA analytical frameworks. A five-gene prognostic model established from 220 FMRGs demonstrated moderate prognostic performance in identifying high-risk stomach adenocarcinoma patients with poorer survival. The high-risk group exhibited immunosuppressive microenvironments with stromal activation, while the low-risk group demonstrated “hot” tumor phenotypes characterized by higher immunogenicity and superior responses to immunotherapy and chemotherapy. Consequently, this model serves as a robust independent prognostic indicator. This five-gene folate metabolism signature shows moderate prognostic value in STAD and may help inform future investigations of the tumor immune microenvironment and therapeutic stratification. This study identified and validated a prognostic risk model consisting of five key genes, which can predict the survival outcomes of patients with STAD. The high-risk group exhibits an immunosuppressive and stroma-remodeling phenotype, while the low-risk group demonstrates an immunologically active state with stronger responsiveness to immunotherapy and chemotherapy. The model not only performs excellently in independent prognostic assessment but also predicts patient sensitivity to immunotherapy and various chemotherapeutic drugs. This study identified and validated a prognostic risk model consisting of five key genes, which can predict the survival outcomes of patients with STAD. The high-risk group exhibits an immunosuppressive and stroma-remodeling phenotype, while the low-risk group demonstrates an immunologically active state with stronger responsiveness to immunotherapy and chemotherapy. The model not only performs excellently in independent prognostic assessment but also predicts patient sensitivity to immunotherapy and various chemotherapeutic drugs.