Integrating multiple omics and machine learning to reveal the prognostic value of endoplasmic reticulum stress gene FKBP10 in gastric cancer.

BACKGROUNDStomach adenocarcinoma (STAD) remains a leading cause of cancer-related mortality worldwide, with limited prognostic biomarkers and heterogeneous responses to immunotherapy. Endoplasmic reticulum stress (ERS) plays a critical role in tumor progression and immune modulation, yet its comprehensive prognostic value in STAD has not been systematically characterized. This study aims to identify ERS-related genes with prognostic significance and elucidate their role in the tumor microenvironment.

METHODSRNA sequencing (RNA-seq) data from The Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression (GTEx) projects were integrated to identify differentially expressed ERS-related genes. Univariate Cox regression and consensus clustering were applied to define molecular subtypes. A prognostic risk model was constructed using least absolute shrinkage and selection operator (LASSO) Cox regression and validated in independent Gene Expression Omnibus (GEO) cohorts (GSE66229, GSE14208). Immune infiltration, functional enrichment, and mutation landscapes were analyzed. Single-cell RNA sequencing (scRNA-seq) (GSE183904) and CellChat were used to explore FKBP10 expression and intercellular communication. Immunohistochemistry and in vitro assays [Cell Counting Kit-8 (CCK-8), flow cytometry] were performed to validate FKBP10 function.

RESULTSWe identified 33 prognostic ERS-related genes, which classified STAD patients into two subtypes (C1 and C2) with distinct survival outcomes and immune infiltration profiles. A 14-gene risk model was constructed and stratified patients into high- and low-risk groups with significant survival differences [area under the curve (AUC) =0.702]. Risk scores correlated with age, tumor (T), metastasis (M), and overall stage. Single-cell analysis revealed FKBP10 was predominantly upregulated in fibroblasts within the tumor microenvironment, and high FKBP10 expression enhanced intercellular communication, particularly between fibroblasts and epithelial cells. Immunohistochemistry confirmed FKBP10 overexpression in STAD tissues, and FKBP10 silencing suppressed proliferation and induced S-phase arrest in HGC27 and MKN1 cells.

CONCLUSIONSThis study establishes a robust ERS-related prognostic signature for STAD and highlights FKBP10 as a key regulator in the tumor microenvironment. FKBP10 promotes gastric cancer (GC) cell proliferation and reshapes cellular communication, offering a potential biomarker and therapeutic target. These findings provide a foundation for personalized prognostic assessment and immunotherapy strategies in STAD.