Construction of a novel ferroptosis-related long non-coding RNAs model for predicting prognosis and tumor immune microenvironment in endometrial cancer.

Background: Endometrial cancer (EC) is the most common gynecological cancer. Ferroptosis is a novel type of programmed cell death that is dependent on iron, and mounting evidence suggests that ferroptosis plays an important role in cancer. Long non-coding RNAs (lncRNAs) are known to regulate ferroptosis; however, little is known about the involvement of ferroptosis-related lncRNAs (FerlncRNAs) in EC. This study aimed to determine a FerlncRNA-based prognostic signature associated with the overall survival (OS) and clinicopathological characteristics of patients with EC. Methods: Tumor transcriptomes and corresponding clinical data from patients with EC were downloaded from The Cancer Genome Atlas (TCGA) database, and the ferroptosis database, FerrDb, was used to identify ferroptosis-related genes (FRGs) (mRNAs). FerlncRNAs in EC were selected based on their correlations with FRGs. Univariate, multivariate, and least absolute shrinkage and selection operator (LASSO) Cox regression analyses were conducted to construct a prognostic model based on the FerlncRNAs signature. The EC patients were grouped into high- and low-risk categories based on the prognostic model risk score. Kaplan-Meier (K-M) survival analysis and time-dependent receiver operating characteristic (ROC) curves were used to evaluate the prognostic value of the risk scores. A predictive nomogram was then established. Gene set enrichment analysis (GSEA) was performed to explore the enriched pathways in the two risk groups. Finally, we compared the proportion of infiltrating immune cells and the expression of potential immune checkpoints between the two groups to understand the tumor immunological microenvironment associated with signature FerlncRNAs. Results: We constructed a FerlncRNAs model to predict the prognosis of patients with EC. K-M analysis demonstrated that patients in the high-risk group had a worse OS. According to the ROC curves, our prognostic model had a better ability to predict the prognosis of patients with EC than other clinical factors. Moreover, the predictive nomogram suggested that our model could offer an independent prognostic evaluation with high accuracy. GSEA identified several enriched pathways in both groups. Finally, the immune microenvironment, including the infiltrating immune cells and immune checkpoints, showed several differences between the two groups. Conclusions: This study revealed that a prognostic model based on 10 ferroptosis-related lncRNAs is useful for predicting the prognosis of patients with EC. Our findings provide novel directions for prognostic assessments, immunotherapies, and targeted treatments of EC.