Construction and evaluation of a predictive model for post-operative prognosis of hepatitis B virus-related hepatocelluar carcinoma / 肿瘤

ABSTRACT

Objective: To construct and evaluate a nomogram model for predicting the survival rate of hepatitis B virus (HBV)-related hepatocellular carcinoma (HCC) patients after radical resection. Methods: The clinical parameters of 396 patients with HBV-related HCC who underwent radical resection were enrolled and sorted out as the training cohort of nomogram model. The data of 200 patients with HBV-related HCC after operation were used as the independent validation cohort to construct the a survival predicting nomogram model. Results: After univariate and multivariate COX regression analyses, the five independent risk factors significantly associated with the overall survival (OS) of patients with HBV-related HCC were identified (all P < 0.01). The five variables were portal vein tumor thrombosis, post-operative antiviral treatment, tumor differentiation, tumor capsule and pre-operative HBV DNA. A nomogram model was constructed based on these five variables. The calibration curves of 1-and 3-year OS showed good concordance between the prediction and actual observation. The concordance index (C-index) obtained from training cohort was 0.74 [95% confidence interval (CI) 0.70-0.77], while that obtained from validation cohort was 0.70 (95% CI 0.65-0.76). Based on the nomogram scores, the patients in validation cohort were stratified into high-and low-risk groups. The survival rate of the patients in high-risk group was significantly lower than that in low-risk group (P = 9.79×10-6). Conclusion: The nomogram model constructed with 5 independent factors can predict the post-operative OS of HBV-related HCC patients.