Document resource
Background Butyrate, a principal short-chain fatty acid derived from gut microbial fermentation of dietary fiber, plays a critical role in maintaining intestinal homeostasis and demonstrates potent anti-tumor properties. Although prior studies have elucidated the anti-cancer mechanisms of butyrate, the prognostic significance of its associated genes in stomach adenocarcinoma (STAD) remains undefined. This study aims to develop a risk model based on butyrate metabolism-related genes to predict STAD prognosis, potentially identifying novel therapeutic targets.Methods RNA sequencing (RNA-Seq) data and clinical survival information from STAD patients were obtained from The Cancer Genome Atlas (TCGA). Butyrate metabolism-related genes were curated from the Gene Set Enrichment Analysis (GSEA) database. The TCGA cohort was partitioned into training and validation sets. Using univariate Cox regression, least absolute shrinkage and selection operator (LASSO) regression, multivariate Cox regression, and stepwise regression, we constructed a prognostic risk model. An independent cohort (GSE84433) from the Gene Expression Omnibus (GEO) repository was utilized for validation. Prognostic performance was assessed via Kaplan-Meier (K-M) survival analysis and time-dependent receiver operating characteristic (ROC) curves. A nomogram integrating risk score, age, sex, and clinical T/N/M stages was developed to predict overall survival (OS), with calibration curves evaluating its accuracy. Functional enrichment analyses (GO, KEGG, Molecular Signatures Database) were performed. The tumor immune microenvironment was characterized using CIBERSORT, Tumor Immune Dysfunction and Exclusion (TIDE), single-sample gene set enrichment analysis (ssGSEA), and immune checkpoint gene profiling. Drug sensitivity (IC50) was also evaluated.Results We identified 345 differentially expressed butyrate metabolism-related genes between STAD tumors and normal tissues. A 9-gene prognostic signature (DOK7, NNMT, IGF2BP1, CSH1, DPP6, PGF, ADTRP, DKK1, TRIB2) was established. The AUC values for 1-, 2-, and 3-year OS were 0.754, 0.821, and 0.854, respectively, confirming superior survival in the low-risk group. Enrichment analyses implicated these genes in cancer-associated pathways. Immune profiling revealed elevated infiltration in high-risk patients, with significant differences in eosinophils (p=0.02p=0.02), follicular helper T cells (p=0.01p=0.01), and CD8+ T cells (p=0.003p=0.003). Risk score inversely correlated with activated CD8+ T cells and memory B cells, aligning with favorable outcomes in low-risk patients. High-risk patients exhibited enhanced sensitivity to dasatinib, docetaxel, erlotinib, and gefitinib ( figure 1).Conclusions The 9-gene risk model serves as an independent prognostic biomarker for STAD, providing insights into therapeutic targets and informing personalized treatment strategies.Abstract 126 Figure 1Graphical abstract