A m6A regulators-related classifier for prognosis and tumor microenvironment characterization in hepatocellular carcinoma

Background: Increasing evidence have highlighted the biological significance of mRNA N-6-methyladenosine (m(6)A) modification in regulating tumorigenicity and progression. However, the potential roles of m(6)A regulators in tumor microenvironment (TME) formation and immune cell infiltration in liver hepatocellular carcinoma (LIHC or HCC) requires further clarification. Method: RNA sequencing data were obtained from TCGA-LIHC databases and ICGC-LIRI-JP databases. Consensus clustering algorithm was used to identify m(6)A regulators cluster subtypes. Weighted gene co-expression network analysis (WGCNA), LASSO regression, Random Forest (RF), and Support Vector Machine-Recursive Feature Elimination (SVM-RFE) were applied to identify candidate biomarkers, and then a m(6)Arisk score model was constructed. The correlations of m(6)Arisk score with immunological characteristics (immunomodulators, cancer immunity cycles, tumor-infiltrating immune cells (TIICs), and immune checkpoints) were systematically evaluated. The effective performance of nomogram was evaluated using concordance index (C-index), calibration plots, decision curve analysis (DCA), and receiver operating characteristic curve (ROC). Results: Two distinct m(6)A modification patterns were identified based on 23 m(6)A regulators, which were correlated with different clinical outcomes and biological functions. Based on the constructed m(6)Arisk score model, HCC patients can be divided into two distinct risk score subgroups. Further analysis indicated that the m(6)Arisk score showed excellent prognostic performance. Patients with a high m(6)Arisk score was significantly associated with poorer clinical outcome, lower drug sensitivity, and higher immune infiltration. Moreover, we developed a nomogram model by incorporating the m(6)Arisk score and clinicopathological features. The application of the m(6)Arisk score for the prognostic stratification of HCC has good clinical applicability and clinical net benefit. Conclusion: Our findings reveal the crucial role of m(6)A modification patterns for predicting HCC TME status and prognosis, and highlight the good clinical applicability and net benefit of m(6)Arisk score in terms of prognosis, immunophenotype, and drug therapy in HCC patients.