Prognosis predictability of hepatocellular carcinoma according to staging systems in hepatitis B virus-endemic area

Abstract

There is no worldwide consensus on clinical application of staging systems that have been proposed for hepatocellular carcinoma (HCC). This study evaluated the predictors of survival and compared the prognosis predictability according to staging systems of HCC. We analyzed the medical records of 142 patients who were consecutively diagnosed as HCC in hepatitis B virus (HBV)-endemic area. To analyze the survival predictors and probability of staging systems, Kaplan-Meier method and Cox proportional hazard model were used. And to compare the discriminatory ability and predictive power of staging systems for prognosis and survival, likelyhood ratio χ(2) test and Akaike information criterion were applied. Overall median survival of HCC patients was 24 months and 1-, 2-, and 3-year survival rate was 61.3, 49.4, and 45.7%, respectively. Child-Pugh classification (P=0.038) and portal vein thrombosis (PVT) (P=0.022) were ascertained as independent predictors of survival. Although all the staging systems showed a progressive decrease in survival as the tumor stage progressively advanced, the Japan Integrated Staging (JIS) and Chinese University Prognostic Index (CUPI) showed the highest homogeneity (small differences in survival among patients in the same stages), and the best monotonicity of gradient (the survival of patients in earlier stages is longer than the survival of patients in more advanced stages within the same system), respectively. In HBV-endemic area, Child-Pugh classification and PVT were independent predictors for survival, and JIS and CUPI were the most powerful staging systems to predict the prognosis of HCC.