Integrative analysis of EMT and hippo pathway activity in gastric cancer peritoneal metastasis

Gastric cancer (GC) is one of the most aggressive malignancies, with poor patient outcomes often due to frequent peritoneal metastasis. Epithelial-mesenchymal transition (EMT) plays a pivotal role in this metastatic process. Recent studies highlight the involvement of the Hippo signaling pathway in regulating EMT and tumor progression. However, the molecular interplay between Hippo signaling components and ARID1A , a key chromatin remodeling factor frequently mutated or lost in GC, remains poorly understood. This study investigates the relationship between Hippo pathway dysregulation and ARID1A alteration in gastric cancer peritoneal metastasis (GCPM). Genomic and molecular profiling were performed on 47 GCPM cell lines of CIN and GS-like subtypes. Hierarchical clustering classified the cell lines into EMT and non-EMT groups, as well as into Hippo and non-Hippo groups. EMT and Hippo pathway activity scores were calculated using single-sample gene set enrichment analysis (ssGSEA). Expression of genes related to EMT, Hippo pathways, and ARID1A was analyzed at mRNA and protein levels. Invasiveness was evaluated by transwell invasion assays, and sensitivity to the YAP inhibitor CA3 (CIL56) was assessed using cell viability assays. Hierarchical clustering divided the cell lines into EMT (n=34) and non-EMT (n=13) groups. The EMT group was enriched in gene sets related to EMT, YAP, and TGF-beta pathways and exhibited significantly higher EMT and Hippo pathway scores ( p < 0.05). These molecular features were accompanied by markedly higher invasive capacity. CA3 sensitivity varied widely across PMGC cell lines, with a mean IC 50 of 0.65 μM (range, 0.07-1.98 μM). No significant differences in CA3 sensitivity were observed across TCGA molecular subtypes, EMT, or Hippo subgroups. Notably, ARID1A -altered cell lines demonstrated significantly enhanced sensitivity to the YAP inhibitor CA3 compared with ARID1A -wild-type lines, suggesting a dependency on YAP activity in ARID1A -deficient tumors. Together, these findings indicate that ARID1A alterations enhance Hippo-YAP pathway dependence and confer increased sensitivity to YAP inhibition in GCPM. ARID1A -deficient tumors may therefore benefit from therapeutic strategies targeting the Hippo-YAP axis.