Integrated in silico and in vitro analysis identifies a co-expression pattern associated with cisplatin/5-FU treatment response and metastatic features in gastric cancer.

backgroundGastric cancer remains a major cause of cancer mortality worldwide, and treatment failure is frequently associated with chemotherapy resistance and metastatic progression. Although cisplatin and 5-fluorouracil are commonly used components of gastric cancer chemotherapy, the extent to which gene-miRNA expression patterns associated with treatment response overlap with metastatic features remains incompletely understood.

methodsWe integrated transcriptomic analysis of pretreatment GSE14210 samples with STRING network analysis and target-centric miRNA prioritization. Prioritized genes and miRNAs were evaluated by qRT-PCR in metastatic versus non-metastatic gastric cancer tissues and in MKN-45 3D spheroids exposed to cisplatin/5-FU; one patient-derived spheroid model was included as a feasibility observation.

resultsAnalysis of the pretreatment GSE14210 cohort identified 414 upregulated DEGs, including 63 highly connected hubs (degree ≥ 10); 21 of these were also upregulated in tumor-versus-normal GEPIA comparisons. In metastatic tissues, CXCL8, MMP9, STAT1, miR-17-5p, miR-24-3p, and miR-124-3p were significantly increased, whereas miR-145-5p was significantly decreased; CCL5 was not significantly different. In cisplatin/5-FU-treated MKN-45 spheroids, CXCL8, MMP9, STAT1, miR-17-5p, miR-24-3p, and miR-124-3p showed significant expression changes, whereas CCL5 and miR-145-5p did not. MKN-45 spheroids regrew after cyclic exposure, consistent with an adaptive drug-tolerant phenotype, whereas the single patient-derived model showed limited recovery.

conclusionThese findings identify context-dependent candidate co-expression patterns across metastatic tissues and chemotherapy-exposed 3D models. They do not establish a shared causal mechanism and require functional validation in independent cohorts and additional patient-derived models.