Construction of prognostic model and molecular subtypes based on endoplasmic reticulum stress-related lncRNAs in endometrial cancer

Background: Recently, TCGA molecular typing of endometrial cancer is a research hotspot and plays an important role in clinical practice. However, microsatellite instability hypermutated and low copy number groups still lack a clear prognostic significance. Here, we established a prognostic model and molecular subtypes on the basis of ten ERS-related lncRNAs in UCEC, which complement TCGA molecular subtypes. Methods Significant ERS-related lncRNAs were identified through co-expression and differential analysis of the TCGA cohort. Using least absolute shrinkage and selection operator (LASSO), univariate and multivariate Cox regression methods, we collected ten ERS-related lncRNAs, developed a prognostic model and molecular subtypes, and analyzed overall survival (OS), tumor microenvironment (TME), and drug susceptibility. Afterwards, we validated the predictive accuracy of the ERS score and developed a nomogram to optimize the ERS prognostic model. Results High ERS score and cluster 1 predicted shorter OS in UCEC. ERS score and molecular subtypes were related to immune responses, checkpoints, and drug sensitivity. To sum up, our research indicates that ERS prognostic model and molecular subtypes contribute to the formation of the diverse and complex TME and predict OS as well as drug sensitivity in UCEC. Conclusions Collectively, we constructed a prognostic model and molecular subtypes based on ten ERS-related lncRNAs. Findings from this study will provide a deeper understanding of how ERS-related lncRNAs function in UCEC, allow for new directions in prognostic assessment, and help develop personalized treatments for UCEC patients.