Related Experiment Videos
MLMP: A Methylation-Associated lncRNA-mRNA Prognostic Signature for Endometrial Carcinoma
1Department of Gynecological Radiotherapy, Harbin Medical University Cancer Hospital, Harbin, China.
Background:
Endometrial carcinoma (EC) is one of the most common gynecological malignancies. Aberrant DNA methylation (DNAm) is an important epigenetic alteration involved in cancer progression. Long non-coding RNAs (LncRNAs) play critical roles in gene regulation and pathological processes. However, the biological and prognostic significance of lncRNA-mRNA pairs associated with methylation remains unclear.
Methods:
Transcriptomic, DNA methylation, and clinical data from The Cancer Genome Atlas were integrated to construct a differentially methylated sites (DMSs)-lncRNA-mRNA network. After mapping DMSs to their corresponding differentially methylated genes (DMGs), the DMS-mediated network was further represented as a DMG-lncRNA-mRNA tripartite network. After removal of DMG nodes, a DMG-associated lncRNA-mRNA background network was constructed, and key lncRNA-mRNA pairs were prioritized according to their topological importance using weighted shortest-path calculation and PageRank-based random walk algorithms. The methylation-associated lncRNA- mRNA pair (MLMP) model was then established based on univariate Cox, LASSO Cox, and multivariate Cox regression analyses.
Results:
Eight DMG-associated lncRNA-mRNA pairs were selected to construct the MLMP model. Under a common analysis framework, the MLMP model achieved higher area under the curve (AUC) values than 17 previously published prognostic signatures for overall survival prediction. Patients in the low-MLMP group had significantly better overall survival than those in the high-MLMP group. Functional enrichment and immune microenvironment analyses indicated that the MLMP model was associated with immune-related pathways and distinct tumor immune microenvironment characteristics. Predictive analyses suggested that patients in the low-MLMP group exhibited higher predicted responsiveness to CTLA-4/PD-1 blockade and differential predicted sensitivity to several therapeutic agents.
Conclusions:
The MLMP model may serve as a potential tool for prognostic stratification in EC and provide insights into methylation-associated lncRNA-mRNA interactions. The predicted associations with immunotherapy response and therapeutic sensitivity may generate hypotheses for future treatment-related studies. These findings require further validation in independent clinical cohorts and experimental studies before clinical application.
