Objectives: Uterine corpus endometrial carcinoma (UCEC) is a malignant cancer that exhibits significant molecular heterogeneity, leading to distinct clinical outcomes. The aim of this study is to identify long non-coding RNAs (lncRNAs) with independent and superior prognostic value based on the tumor clinical stages in UCEC patients. Methods: The Cancer Genome Atlas (TCGA) was utilized to acquire clinical data and expression levels of lncRNAs and mRNAs in UCEC patients. Tumor samples were compared with normal samples using R-statistical computing and Cy toscape. Four lncRNA-expression signatures (LINC01224, AC015849.16, LINC00908, and LINC00092) were identified through tenfold cross-validation, t-tests, and univariate COX regression. Results: LINC00908 and LINC00092 exhibited a negative correlation with tumor stages and were downregulated in expression compared to normal samples. Conversely, LINC01224 and AC015849.16 were upregulated in tumor samples and positively correlated with the overall survival of UCEC patients. The lncRNAs-mRNAs network and functional en richment analysis indicated the involvement of these four lncRNA signatures in UCEC tumor progression by modulating pathways such as TGF-? signaling, cell cycle, DNA replication, NF-kB signaling, and Notch signaling. Conclusion: LINC01224, AC015849.16, LINC00908, and LINC00092 could be considered as alternate prognostic mark ers for UCEC prediction, potentially improving overall survival and enabling patient-tailored treatment strategies. Keywords: Uterine corpus endometrial carcinoma, prognosis, lncRNAs, tumor progression, overall survival
Corresponding Author: Ruitai Fan