Figure 3.

m6A methylation regulatory factors can be used as independent prognostic factors for HCC patients. (A) Univariate Cox regression analysis of m6A methylation regulatory factors. (B) Multivariate Cox regression analysis of the prognosis-related feature genes. (C) Kaplan-Meier OS analysis of the patients in high- and low-risk groups (median risk score as the threshold) in TCGA dataset. (D) ROC curves based on the risk model. (E) Kaplan-Meier OS curves for the patients in high- and low-risk groups in the LIRI-JP dataset from ICGC database. (F) ROC curves based on the risk model in the validation set. (G) The difference in the expression of the m6A methylation regulatory factors in the risk model and the difference in clinicopathological differences between the high- and low-risk groups are exhibited in an heatmap. (H) The relationship between the risk score and different clinical grades, stages and T stages of HCC. (I,J) Kaplan-Meier OS curves and DFS curves of the m6A methylation regulatory factors in the risk model are retrieved from GEPIA database. (K) ROC curves shows the potential of the 4 m6A methylation regulatory factors in the risk model as independent factors for HCC prognosis.