Risk stratification of hepatocellular carcinoma incidence using a fibrosis-4-based prediction model in patients with chronic hepatitis C receiving antiviral therapy: a nationwide real-world Taiwanese cohort study.

A total of 1,589 patients who had received interferon-based treatment were enrolled and analyzed for the risk of hepatocellular carcinoma (HCC) in a real-world nationwide Taiwanese chronic hepatitis C cohort (T-COACH). We aimed to stratify HCC risk by non-invasive fibrosis index-based risk model. Of 1589 patients, 1363 (85.8%) patients achieved sustained virological response (SVR). Patients with SVR had 1, 3, 5 and 10-year cumulative HCC incidence rates of 0.55%, 1.87%, 3.48% and 8.35%, respectively. A Cox proportional hazards model revealed that non-SVR (adjusted hazard ratio [aHR]: 1.92, 95% confidence interval [CI]: 1.19-3.12, p = 0.008), diabetes mellitus (aHR: 2.11, 95% CI: 1.25-3.55, p = 0.005), and fibrosis (FIB)-4 at the end of follow-up (EOF; aHR: 5.60, 95% CI: 2.97-10.57, p < 0.0001) were independent predictors of HCC. Risk score models based on the three predictors were developed to predict HCC according to aHR. In model 1, the 10-year cumulative incidence rates of HCC were 43.35% in patients at high risk (score 9-10), 25.48% in those at intermediate risk (score 6-8), and 4.06% in those at low risk (score 3-5) of HCC. In model 2, the 10-year cumulative incidence rates of HCC were 39.64% in patients at high risk (at least two risk predictors), 19.12% in those at intermediate risk (with one risk predictor), and 2.52% in those at low risk (without any risk predictors) of HCC. The FIB-4-based prediction model at EOF could help stratify the risk of HCC in patients with chronic hepatitis C after antiviral treatment.