使用 STRING、MCODE 和 GEPIA2 构建子宫内膜癌免疫网络及检查点基因的预后评估

Retrospective analysis used TCGA-UCEC (The Cancer Genome Atlas - Uterine Corpus Endometrial Carcinoma) and GTEx datasets and reported immune-related genes. Genes were analyzed in STRING v12.0 (confidence ≥0.7; up to 10 neighbors per node) to generate a protein-protein interaction (PPI) network, exported to Cytoscape v3.10.2, and processed with MCODE to identify functional clusters. Hub genes were evaluated for expression and overall survival (OS) in GEPIA2 using median-based stratification and log-rank tests ( p <0.05). Six immune signatures were assessed in TIMER2.0. PDCD1 and CD274 showed strong interactions with other immune effectors.

CD40 and LGALS9 were down- and upregulated, respectively, without affecting OS. Combined overexpression of CTLA4 , PDCD1 , TIGIT , CD8A , CD8B , GZMB , PRF1 , TBX21 , FOXP3 , CXCL9 , CD28 , and ICOS correlated with improved OS, suggesting direct immune effects and enhanced responses to targeted therapies.

This in silico immune network highlights checkpoint centered hubs and coordinated immune programs with prognostic relevance in endometrial cancer, providing a rationale for biomarker guided immunotherapy development and patient stratification. Validation in independent cohorts and correlation with clinicopathologic and treatment response data are needed to support clinical translation.