Design and validation of risk prediction model for hepatocellular carcinoma development after sustained virological response in patients with chronic hepatitis C

Objectives Hepatocellular carcinoma can develop after hepatitis C virus eradication. We developed a new hepatocellular carcinoma risk score (HCC-SVR score) based on independent predictors for chronic hepatitis C after sustained virological response. Methods Between 2003 and 2016, a total of 1193 patients with chronic hepatitis C who achieved sustained virological response through antiviral therapy were included (669 for training cohort and 524 for validation cohort). The HCC-SVR score was developed using multivariate Cox proportional hazards regression modelling. Results Hepatocellular carcinoma (n = 19) occurred more frequently in older, male patients and was associated with liver cirrhosis; hypertension; diabetes; lower platelet count; higher alpha-fetoprotein, aspartate, and alanine aminotransferase; lower total cholesterol; and higher fibrosis-4 index (FIB-4) (all P < 0.05). FIB-4 (hazard ratio = 1.080), male gender (hazard ratio = 8.189), and higher alpha-fetoprotein (hazard ratio = 1.060) independently predicted hepatocellular carcinoma (all P < 0.05). HCC-SVR score successfully predicted hepatocellular carcinoma development risk [area under receiver operating characteristic curve (AUC) = 0.771, 0.857, and 0.911 at 2, 4, and 6 years, respectively]. The cumulative incidence rate of hepatocellular carcinoma differed significantly among groups stratified by HCC-SVR risk score (0–2 points, low; 3–7 points, intermediate; 8–9 points, high risk) (all P < 0.05 by log-rank test). HCC-SVR score was maintained in a validation cohort (n = 524) (AUC = 0.728 at 2 years, 0.737 at 4 years, and 0.809 at 6 years). Conclusion The HCC-SVR score enables risk stratification for hepatocellular carcinoma development at sustained virological response in patients with chronic hepatitis C.