Identification of prognostic markers related to triaptosis in endometrial cancer and construction of prognostic model

Abstract Background Endometrial cancer (EC) is a malignant neoplasm of the uterine epithelial lining. Triaptosis is a kind of cell death induced by endosomal dysfunction due to PI3P depletion. Triaptosis is suggested to correlate with EC, but the mechanism remains unclear. We aimed to identify triaptosis-related prognostic markers in EC and reveal their mechanisms via bioinformatics.. Methods Weighted correlation network analysis (WGCNA) screened triaptosis related genes (TRGs) in TCGA-UCEC, defined as WGCNA-TRGs. Candidate genes (CGs) were derived by intersecting WGCNA-TRGs with differentially expressed genes (DEGs) extracted from the TCGA-UCEC. Prognostic markers were screened through univariate Cox proportional hazards model and LASSO regression. Prognostic model was constructed and validated in TCGA-UCEC and validation set. A nomogram estimated independent prognostic factors. Immune infiltration and gene set enrichment analysis (GSEA) profiled tumor-infiltrating immune cells and related pathways.. Results There were 110 CGs were identified and 4 prognostic markers (SDF2L1, STK26, TDRKH and USP18) were determined. Prognostic model performed well in TCGA-UCEC and validation set. Both tumor stage and risk score were validated as independent prognostic determinants, with the nomogram built upon these two factors yielding robust predictive accuracy for patient prognosis. Among immune cell subsets, several populations—most notably activated B cells—exhibited substantial abundance differences across the high- and low-risk strata. There were common 12 pathways shared by prognostic markers. Drugs like ZINC were commonly shared by prognostic markers. Conclusion SDF2L1, STK26, TDRKH, and USP18 were characterized as triaptosis-associated prognostic markers in EC, offering a theoretical foundation for the development of targeted therapies.