Integrative Transcriptomic Analysis Identifies Glycolysis/Ferroptosis-Associated Candidate Genes in Gastric Cancer

Background: The crosstalk between glycolysis, ferroptosis and the gastric cancer (GC) immune microenvironment remains poorly defined. This study aims to uncover glycolysis/ferroptosis-related regulators of the GC immune microenvironment.Methods: Transcriptomic datasets from The Cancer Genome Atlas (TCGA)-Stomach adenocarcinoma (TCGA-STAD) and Gene Expression Omnibus (GEO) were integrated to screen glycolysis/ferroptosis-related differentially expressed genes (DEGs). We evaluated univariate prognostic associations of candidate genes and built a signature for stratifying patients based on Tumor Immune Dysfunction and Exclusion (TIDE)-predicted immunotherapy response and systematically evaluated immune-related molecular signatures based on TIDE and Microsatellite Instability (MSI) scores. Single-cell RNA sequencing combined with CellChat was applied to explore cell-cell communication correlations. The oncogenic function of 5′-nucleotidase domain containing 2 (NT5DC2) was experimentally validated in MKN-7 GC cells and subcutaneous xenograft models.Results: Six candidate metabolic DEGs were selected. Their individual diagnostic performance varied substantially across cohorts, although insulin-like growth factor 2 mRNA-binding protein 3 (IGF2BP3) and fibronectin type III domain containing 5 (FNDC5), together with NT5DC2, exhibited relatively consistent expression trends across independent datasets. Single-cell Gene Set Enrichment Analysis (GSEA) suggested a potential association between epithelial NT5DC2​ and gene signatures indicative of elevated glycolysis coupled with attenuated ferroptosis programs. Whereas B-cell IGF2BP3 was correlated with metabolic and DNA repair gene sets. Bulk transcriptome correlation analyses indicated that elevated NT5DC2 and IGF2BP3 expressions were associated with activated Granulin Precursor (GRN) /Transforming Growth Factor Beta 1 (TGF-β) signaling, decreased CD8+ T cell abundance and T cell exhaustion signatures. Though direct functional evidence confirming this metabolic-immune regulatory axis is absent, correlative observations support its existence. A three-gene signature exhibited exploratory capacity for stratifying patients with distinct TIDE-predicted immune response profiles. Univariate survival analysis demonstrated that high NT5DC2 expression correlated with unfavorable overall survival. This association was no longer statistically significant after adjustment for clinical covariates including age. In vitro and in vivo assays confirmed that NT5DC2 knockdown inhibits GC proliferation and xenograft growth. The statistical significance threshold was set at p < 0.05.Conclusion: We identified candidate glycolysis/ferroptosis-related biomarkers for gastric cancer. Bioinformatic correlation analyses indicated that NT5DC2, FNDC5 and IGF2BP3 expressions are associated with metabolic reprogramming signatures and predicted immunosuppressive immune phenotypes in GC.