Identification of an immune checkpoint gene signature that accurately predicts prognosis and immunotherapy response in endometrial carcinoma

Abstract

In this study, we performed a bioinformatics analysis to identify immune checkpoint genes (ICGs) associated with prognosis and the immunotherapeutic response in endometrial carcinoma (EC) patients. We classified 47 ICGs into high, medium, and low expression groups by performing RNA-sequencing data analysis of EC patient samples from The Cancer Genome Atlas (<i>n</i> = 521) and GSE77688 (<i>n</i> = 88) datasets. Univariate Cox regression analysis showed that seven ICGs (<i>VTCN1, TNFRSF18, TNFRSF14, TNFRSF4, CD40LG, TMIGD2,</i> and <i>BTLA</i>) were associated with prognosis in EC patients. Spearman correlation analysis showed that prognosis-related ICGs correlated positively with immunotherapy response factors, including tumor mutation burden (TMB), mismatch repair gene mutations, neoantigens, clinical stages, and adaptive immune resistance pathway genes. We identified a prognostic gene signature of four ICGs (<i>IDO1, CD274, CTLA4,</i> and <i>TNFRSF14</i>) that accurately predicted survival outcomes of EC patients. TIMER database and Kaplan-Meier survival analysis showed that OS among EC patients with low <i>TNFRSF14</i> expression was significantly shorter than among those with high <i>TNFRSF14</i> expression. <i>In vitro</i> experiments showed that <i>TNFRSF14</i> silencing increased the migration and invasiveness of EC cells by promoting epithelial-mesenchymal transition (EMT). Collectively, these findings reveal an immune checkpoint gene signature that accurately predicts survival outcomes and immunotherapeutic responses in EC patients.