Research Paper Volume 12, Issue 1 pp 965—977

Prediction of hepatocellular carcinoma prognosis based on expression of an immune-related gene set

class="figure-viewer-img"

Figure 6. Integration of IPSHCC and clinical characteristics. To further enhance the predictive accuracy of IPSHCC, we integrated coefficients for clinical characteristics generated in the multivariate Cox regression model from the training and validation set. Restricted mean survival (RMS) curves for IPSHCC and the integrated model are shown. C-index values for the clinical characteristics alone and for the integrated model were compared. The p-value represents the difference between the C-index values for the two models. (A) In the training set, the C-index values of the clinical characteristics and the integrated model were 0.647 and 0.743, respectively (p < 0.001). (B) In the validation set, the C-index values of the clinical characteristics and the integrated model were 0.745 and 0.785, respectively (p = 0.011).