Screening and Functional Study of Biomarkers Related to Cell-in-cell Structure in Stomach Adenocarcinoma

A comprehensive study of Cell-In-Cell (CIC) structures in stomach adenocarcinoma (STAD) may facilitate the development of therapeutic strategies.

Enrichment analysis was conducted using ssGSEA. WGCNA was used to identify hub genes, and Differentially Expressed Genes (DEGs) were screened by the DESeq2 package. Gene selected by both LASSO regression (glmnet package) and SVM-RFE algorithms (e1071 package) were intersected to develop a diagnostic model for STAD using the rms package. Immune infiltration of STAD samples was analyzed using CIBERSORT and ESTIMATE algorithms. Single-cell data were processed by Seurat package. Cell subpopulations were identified using the FindClusters function, and their marker genes were subsequently determined by FindAllMarkers, followed by in vitro functional validation.

Five genes (MSR1, PDGFRB, COL8A1, PLA2G7, and FCGR3A) with AUC > 0.8 were identified as potential biomarkers for STAD and combined into a five-gene diagnostic model. Immune infiltration analysis showed that these genes were associated with T-cell and macrophage infiltration. Among the ten cell types identified by single-cell analysis, the five biomarkers were found to be highly expressed specifically in macrophages and fibroblasts. In vitro functional assays confirmed that these markers were upregulated in HGC-27 and AGS cells, and that MSR1 regulated STAD cell proliferation, migration, and invasion.

The five biomarkers, which showed a specifically high expression in macrophages and fibroblasts, were closely associated with heterotypic CIC formation. Further analysis suggested that these biomarkers contributed to STAD pathogenesis, potentially via immune, stromal, and metabolic crosstalk. These findings help clarify the interactions between CIC biology and the STAD tumorimmune microenvironment.

The five biomarkers provide valuable insights for the early screening, diagnosis, and treatment of STAD.