Background: The interaction between tumor microenvironment (TME) and tumors offers various targets in mounting anti-tumor immunotherapies. However, the diagnostic and prognostic biomarkers in endometrial carcinoma (EC) are still limited. Here, we aimed to analyze the TME features and identify novel prognostic biomarkers for EC. Methods: ESTIMATE, CIBERSORT, protein-protein interaction (PPI) network, univariate Cox regression, and functional enrichment analysis were performed to identify immune-

more »

... and survival-related hub genes as well as possible molecular mechanisms. The limma package and deconvolution algorithm were adopted to estimate the abundance of tumor-infiltrating immune cells (TICs) and their relationship with the target gene. In validation section, tissue microarrays (TMAs) of EC and multiplex immunohistochemistry (m-IHC) were evaluated to validate the expression of TNFRSF4, and its correlation with immune markers, including CD4, CD8, and FOXP3. The receiver operating characteristic (ROC) curve was used to determine the efficacy of TNFRSF4, CD4, CD8, and FOXP3 in diagnosing EC. Results: Two genes, TNFRSF4 and S1PR4, were screened out from 386 intersection differential expression genes (DEGs) shared by ImmuneScore and StromalScore in EC. Highlighted by TNFRSF4, we found that it was not only positively correlated with the TICs (mainly CD4+ T cells, CD8+ T cells, and Tregs) but significantly related to diagnosis and prognosis in patients of EC, both verified by data from the TCGA database and clinical samples. At the same time, the expression trend of TNFRSF4 was further confirmed by an integrated meta-analysis based on six microarrays from the Gene Expression Omnibus database (GEO). Conclusions: Collectively, TNFRSF4 could serve as a high-profile biomarker to robustly predict immune microenvironment, clinical diagnosis, and prognosis for EC.