Identification and validation of an epigenetically regulated long noncoding RNA model for breast cancer metabolism and prognosis.

BACKGROUND

Breast cancer (BC) is the leading cause of death among women, and epigenetic alterations that can dysregulate long noncoding RNAs (lncRNAs) are thought to be associated with cancer metabolism, development, and progression. This study investigated the epigenetic regulation of lncRNAs and its relationship with clinical outcomes and treatment responses in BC in order to identify novel and effective targets for BC treatment.

METHODS

We comprehensively analysed DNA methylation and transcriptome data for BC and identified epigenetically regulated lncRNAs as potential prognostic biomarkers using machine learning and multivariate Cox regression analysis. Additionally, we applied multivariate Cox regression analysis adjusted for clinical characteristics and treatment responses to identify a set of survival-predictive lncRNAs, which were subsequently used for functional analysis of protein-encoding genes to identify downstream biological pathways.

RESULTS

We identified a set of 1350 potential epigenetically regulated lncRNAs and generated a methylated lncRNA dataset for BC, MylnBrna, comprising 14 lncRNAs from a list of 20 epigenetically regulated lncRNAs significantly associated with tumour survival. MylnBrna stratifies patients into high-risk and low-risk groups with significantly different survival rates. These lncRNAs were found to be closely related to the biological pathways of amino acid metabolism and tumour metabolism, revealing a potential tumour-regulation function.

CONCLUSION

This study established a potential prognostic biomarker model (MylnBrna) for BC survival and offered an insight into the epigenetic regulatory mechanisms of lncRNAs in BC in the context of tumour metabolism.