Abstract / Summary
Abstract Gastric cancer (GC) is a heterogeneous malignancy with poor outcomes, and reliable biomarkers are needed to improve prognosis assessment and individualized treatment. Neuroinflammation-related genes (NIRGs) may reflect neural-immune interactions within the tumor microenvironment (TME), but their prognostic value in GC remains unclear. This study integrated bulk RNA sequencing data from The Cancer Genome Atlas-stomach adenocarcinoma (TCGA-STAD) and Gene Expression Omnibus (GEO) with single-cell RNA sequencing (scRNA-seq) data to identify NIRG-related prognostic features. Differential expression analysis, non-negative matrix factorization (NMF), Cox regression, and a random survival forest (RSF) model identified five prognostic genes: GRIA3, NGF, SLC17A7, TGFB2, and TGFB3. The risk model stratified patients by overall survival (OS) and showed validation performance in an independent cohort. The risk score was associated with clinicopathological characteristics, immune infiltration, and predicted drug sensitivity. Functional analyses linked these genes to stemness, invasion, and epithelial-mesenchymal transition (EMT). scRNA-seq analysis highlighted fibroblasts as key cells involved in NIRG-related TME remodeling, and reverse transcription-quantitative polymerase chain reaction (RT-qPCR) validated GRIA3, NGF, and SLC17A7 expression. These findings provide a potential prognostic framework for GC.