Abstract
This study aimed to explore an immune response-related gene signature to predict the clinical prognosis and tumor immunity of stomach adenocarcinomas (STAD). Based on the expression and clinical data of STAD in the TCGA database, the immune cell infiltration status was evaluated using CIBERSORT and ESTIMATE methods. Samples were grouped into "hot" and "cold" tumors based on immune cell infiltration status and consensus clustering. The infiltration abundance of activated memory CD4 T cells and CD8 T cells had a significant effect on the overall survival of STAD patients. Among the three clusters, cluster 2 had a higher immune score and a significantly higher abundance of CD8 T cells and activated memory CD4 T cells were assigned as a hot tumor, while cluster 1 and 3 were assigned as a cold tumor. DEGs between hot and cold tumors were mainly enriched in immune-related biological processes and pathways. Total of 13 DEGs were related to the overall survival (OS). After the univariate and multivariable Cox regression analysis, three signature genes (PEG10, DKK1, and RGS1) was identified to establish a prognostic model. Patients with the high-risk score were associated with worse survival, and the risk score had an independent prognostic value. Based on TIMER online tool, the infiltration levels of six immune cell types showed significant differences among different copy number statuses of PEG10, DKK1, and RGS1. In this study, an immune-related prognostic model containing three genes was established to predict survival for STAD patients.