Identification of molecular patterns and prognostic models of epithelial-mesenchymal transition- and immune-combined index in the gastric cancer

Background: Epithelial-mesenchymal transition (EMT) and the immune microenvironment play important roles in the progression of gastric cancer (GC), but the joint role of both in GC is not clear. Methods: We identified EMT- and immune-related genes (EIRGs), and the molecular subtypes of EIRGs were identified by unsupervised cluster analysis. Then, we constructed an accurate EIRG_score model by using differential genes of molecular subtypes. The correlation of EIRG_score with prognosis, immune infiltration, gene mutation, chemotherapeutic drug sensitivity, and immunotherapy response was comprehensively analyzed. In addition, we investigated the biological function of EIRG_score via in vitro experiments. Results: A total of 808 GC patients were classified into two molecular subtypes, which were enriched in EMT and immune-related biological pathways and significantly correlated with prognosis and immune infiltration. The constructed EIRG_score had an important role in predicting prognosis and immunotherapeutic response. The higher EIRG_score was associated with worse prognosis, higher abundance of immunosuppressive cell infiltration, lower immune checkpoint genes expression, lower tumor mutation burden, microsatellite instability-high, lower chemotherapeutic drug sensitivity, and poorer immunotherapeutic response. Conclusion: EIRG_score may be used as a biomarker to assess prognosis and guide precise treatment.