Comprehensive characterization of m6A methylation and its impact on prognosis, genome instability, and tumor microenvironment in hepatocellular carcinoma

Abstract

BackgroundN6-methyladenosine (m6A) RNA regulation was recently reported to be important in carcinogenesis and cancer development. However, the characteristics of m6A modification and its correlations with clinical features, genome instability, tumor microenvironments (TMEs), and immunotherapy responses in hepatocellular carcinoma (HCC) have not been fully explored.MethodsWe systematically analyzed the m6A regulator-based expression patterns of 486 patients with HCC from The Cancer Genome Atlas and Gene Expression Omnibus databases, and correlated these patterns with clinical outcomes, somatic mutations, TME cell infiltration, and immunotherapy responses. The m6A score was developed by principal component analysis to evaluate m6A modifications in individual patients.ResultsM6A regulators were dysregulated in HCC samples, among which 18 m6A regulators were identified as risk factors for prognosis. Three m6A regulator-based expression patterns, namely m6A clusters, were determined among HCC patients by m6A regulators with different m6A scores, somatic mutation counts, and specific TME features. Additionally, three distinct m6A regulator-associated gene-based expression patterns were also identified based on prognosis-associated genes that were differentially expressed among the three m6A clusters, showing similar properties as the m6A regulator-based expression patterns. Higher m6A scores were correlated with older age, advanced stages, lower overall survival, higher somatic mutation counts, elevated PD-L1 expression levels, and poorer responses to immune checkpoint inhibitors. The m6A score was validated as an independent and valuable prognostic factor for HCC.ConclusionM6A modification is correlated with genome instability and TME in HCC. Evaluating m6A regulator-based expression patterns and the m6A score of individual tumors may help identify candidate biomarkers for prognosis prediction and immunotherapeutic strategy selection.