Prediction models of postoperative outcomes of patients with hepatocellular carcinoma (HCC) after surgery based on the China liver cancer (CNLC) staging system are rare. This study aimed to compare the prognostic abilities of CNLC, Tumor-Node-Metastasis (TNM) 8th edition, and Barcelona Clinic Liver Cancer (BCLC) staging systems for HCC after curative resection. We developed two nomograms incorporating the CNLC staging system to predict the postoperative recurrence-free survival (RFS) and overall survival (OS) of HCC patients.
The prognostic abilities of the CNLC, TNM and BCLC staging systems for HCC after curative resection were compared using receiver operating characteristic (ROC) curves. Two nomograms incorporating five selected risk factors were constructed based on multivariate Cox regression in the primary cohort of 312 HCC patients. It was validated with an independent validation cohort of 130 HCC patients. The predictive performance and discrimination ability of the two nomograms were further evaluated and compared with those of the TNM and BCLC staging systems.
The CNLC staging system had a higher area under the receiver operating characteristic curve (AUROC) value for both OS (AUC=0.692) and RFS (AUC=0.673) than the TNM (ROC=0.667 for OS and 0.652 for RFS) and BCLC (ROC=0.671 for OS and 0.670 for RFS) staging systems. The independent predictors of OS (cirrhosis, gamma-glutamyl transpeptidase (GGT), tumor differentiation and CNLC staging system) and RFS (α-fetoprotein (AFP) and CNLC staging system) were incorporated into the two nomograms. The OS and RFS nomograms consistently outperformed the TNM and BCLC staging systems in the primary cohort. These results were verified in the validation cohort. In the 442 patients with HCC, the RFS nomogram could predict early recurrence very well.
The two proposed nomograms incorporating the CNLC staging system can predict the outcomes of patients with HCC after curative hepatectomy in clinical practice.