Integrative Analysis Of DNA Methylation, RNA Sequencing, And Variant Dataset Via Machine Learning In Predicting Endometrial Cancer Recurrence

초록

Introduction/Background Endometrial cancer (EC) is classified into four molecular subtypes, each with unique prognostic characteristics. The prognosis within each subtype varies due to histological and molecular factors. This study leverages omics datasets and machine learning to identify biomarkers associated with EC recurrence in different molecular subtypes. Methodology From The Cancer Genome Atlas (TCGA) multi-omics datasets, 116 samples that molecular subtypes (polymerase epsilon (POLE), microsatellite instability-high (MSI-H), copy number-low (CN-L), or copy number-high (CN-H)) were assigned and having both molecular data (DNA methylation, RNA-sequencing, and common variants) and clinicopathological information were used for the analysis. Differentially methylated regions (DMRs) and differentially expressed genes (DEGs) were identified using t-test between recurrence and non-recurrence groups. These were visualized through volcano plots and heat maps, while decision trees and random forests classified and stratified the samples. Results A machine learning analysis showed that in the CN-H recurrence group, PARD6G-AS1 gene showed decreased methylation, CSMD1 gene showed increased methylation, and TESC gene expression was higher than the non-recurrence group. In the CN-L recurrence group, CD44 gene expression was elevated. Further validation using TCGA clinical data confirmed that PARD6G-AS1 hypomethylation in CN-H and CD44 overexpression in CN-L as significant indicators of recurrence (p=0.006 and p=0.02, respectively). In addition, PARD6G-AS1 hypomethylation and CD44 overexpression were significantly linked to advanced stage and lymph node metastasis (p=0.014 and 0.013, respectively). Conclusion The study concludes that PARD6G-AS1 hypomethylation and CD44 overexpression could be potential predictors of recurrence in CN-H and CN-L EC patients, respectively. Advanced stage and lymph node metastasis might have roles in increasing recurrence in conjunction with PARD6G-AS1 hypomethylation and CD44 overexpression.