Construction of lactate-myeloid-related prognostic risk model and prediction of immune-excluded microenvironment in gastric cancer via integrated scRNA-Seq and bulk RNA-Seq.

BACKGROUND: Gastric cancer (GC) progression entails profound epithelial-stromal remodeling, yet how epithelial metabolic programs are linked to myeloid-dominated immune exclusion and prognostic risk remains unclear. This study aimed to integrate single-cell RNA sequencing (scRNA-seq) and bulk transcriptomic analyses to characterize a lactate-myeloid-related risk axis in GC. METHODS: We integrated six scRNA-seq datasets comprising 116 samples and 355,091 cells spanning paratumor (PT), chronic atrophic gastritis (CAG), moderate intestinal metaplasia (MIM), severe intestinal metaplasia (SIM), early gastric cancer (EGC), and advanced gastric cancer (AGC), and coupled them to bulk cohorts. Epithelial and myeloid states, pseudotime trajectories, ligand-receptor programs, and pathway activities were analyzed. A 19-gene prognostic model anchored on SLC16A3/MCT4 co-expression and macrophage immunoregulatory modules was developed in The Cancer Genome Atlas stomach adenocarcinoma (TCGA-STAD) cohort and externally validated in GSE84437. Mutation burden, tumor immune dysfunction and exclusion (TIDE), and expression-based drug-response predictions were further evaluated. SLC16A3 expression was validated by immunohistochemistry (IHC) in an independent clinical cohort. RESULTS: The integrated atlas showed immune contraction with epithelial expansion in premalignant lesions, followed by enrichment of aberrant epithelium and myeloid lineages in EGC/AGC. Within epithelium, a late-pseudotime SLC16A3+ lactate-export program increased with progression and was associated with extracellular matrix (ECM) remodeling, pro-angiogenic signaling, and myeloid-recruiting chemokines. IHC confirmed marked SLC16A3 enrichment within dense, cohesive neoplastic nests of Lauren intestinal and mixed subtypes. Myeloid decomposition revealed macrophage-centric expansion dominated by SPP1+ and APOE+ tumor-associated macrophage (TAM) states. The model-derived risk score stratified overall survival independently of clinical covariates. Low-risk tumors exhibited higher mutation counts and tumor mutational burden (TMB), whereas high-risk tumors were genomically quieter but TIDE-high. Expression-based analyses nominated differential drug-response patterns across risk groups, which require further validation. CONCLUSIONS: This study identifies an epithelial SLC16A3-associated lactate-myeloid transcriptional program linked to macrophage-rich immune exclusion and poor prognosis in GC. The 19-gene model provides a biologically interpretable framework for risk stratification, while the immune-evasion and drug-response findings should be considered hypothesis-generating and require prospective and functional validation.