Fig. 9

A nomogram integrating LRG Score and clinical parameters for precision prediction. A Univariate independent prognostic analysis in the TCGA cohort. B Multivariate independent prognostic analysis in the TCGA cohort. C–F Barplot shows the relationship between different clinical factors and LRG riskscore. G Nomogram was constructed by combining clinical features (T stage, N stage) with risk score. H The calibration plots test consistency between the actual OS rates and the predicted survival rates, with the 45°line representing the best prediction. I Decision curve analysis (DCA) showed that nomograms had a significantly higher clinical benefit for patients than other clinical features. J AUC curves were used to evaluate the predictive performance of different clinical characteristics, nomogram scores and risk scores. Statistical significance: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001