Figure 5. Construction of a prognostic risk signature-based nomogram and evaluation of its performance in the TCGA dataset. (A, B) The forest plot shows association of clinicopathological parameters including risk score and the OS status of TCGA-LADC dataset as assessed by univariate and multivariate Cox regression models. (C) The nomogram with risk score and AJCC stages to predict 1-year, 3-year and 5-year OS of individual LADC patients. (D) The calibration curves and c-index values for the predicted 1-year, 3-year, and 5-year OS based on the nomogram for the TCGA-LADC patients from the training set. (E) ROC curve analysis shows the variations in the AUC values for the nomogram and AJCC stages in the training and validation cohorts for the 1-year to 5-year follow-up period. (F) The DCA analysis shows 1-year, 3-year and 5-year OS for LADC patients based on the nomogram.