Fig. 3

The disease progression status and gene expression levels of each patient, time‐dependent receiver operating characteristic (ROC) curve, and survival analysis based on the prognostic classifier in the training set and validation set. a The distribution of disease progression status (upper panel) and gene expression levels (lower panel) of each patient between low and high-risk groups in the training set. Each point in the scatterplots represents a patient, and the shape of the point represents whether the disease has progressed or not. The corresponding heatmap below each point vertically represents the expression level of ACAP1 and RASGRP1 genes in this patient. The patients were ordered according to the risk score level. b Time-dependent ROC curves in the training set. The area under the curves (AUC) at 1-, 2-, 3-, and 5-year were used to evaluate the prognostic accuracy. c Survival analysis of the different risk groups in the training set according to the optimal cutoff value (log‐rank test p‐value < 0.001). d The distribution of disease progression status (upper panel) and gene expression levels (lower panel) of each patient between low and high-risk groups in the validation set. The patients were ordered according to the risk score level. e Time-dependent ROC curves in the validation set. f Kaplan–Meier curves show the distinct outcome between low‐ and high‐risk groups in the validation set (log‐rank test p‐value = 0.002)