Risk profiling of hepatocellular carcinoma occurrence after eradication of hepatitis C virus with direct-acting antiviral agents

Background: Treatment with direct-acting antiviral agents (DAAs) have significantly improved the outcomes of patients with HCV-related cirrhosis. However, the risk of developing hepatocellular carcinoma (HCC) still persists in these patients and predictive models are lacking. The aim of this study is to develop a predictive model including clinical and genetic features for individual risk profiling of HCC occurrence after DAA-induced sustained virologic response (SVR) in patients with HCV cirrhosis. Methods Consecutive patients with HCV cirrhosis achieving SVR after DAA treatment without previous history of HCC were prospectively enrolled and followed-up for HCC occurrence. Clinical and biochemical features were evaluated at the time of DAA start and at the time of SVR. MERTK single-nucleotide polymorphism (SNP) rs6726639 was assessed in all patients. Cox regression model was used to identify risk factors for HCC occurrence. Accuracy was assessed by time-dependent area under the receiver operating characteristic curves (AUROCs) and net benefit was measured by decision curve analysis. Results Four-hundred-seven patients (mean age 67 years, 57% male) were included and followed-up for a mean time of 70 months. HCC occurred in 58 (14.2%) patients. At the time of DAA start, PLT count lower than 120x109/L (HR 2.12, 95%CI 1.15-3.93, p=0.017), albumin levels lower than 3.5 g/dL (HR 2.37, 95%CI 1.38-4.08, p=0.002) and MERTK AA genotype (HR 1.98, 95%CI 1.09-3.62, p=0.026) were independent risk factors for HCC occurrence, by multivariate analysis. At the time of SVR, PLT count lower than 130x109/L (HR 2.14, 95%CI 1.10-4.34, p=0.036), albumin levels lower than 3.8 g/dL (HR 2.21, 95%CI 1.07-4.58, p=0.032) and MERTK AA genotype (HR 2.18, 95%CI 1.01-4.70, p=0.046) were confirmed as independent risk factors for HCC occurrence, by multivariate analysis. Time-dependent AUROCs showed good accuracy and net benefit of this model was confirmed by DCA Conclusions The risk of HCC after DAA-induced SVR persists, although low, also after a long-term follow-up. A risk profiling based on PLT, albumin and MERTK genotype is useful to predict the risk of HCC occurrence after SVR and it could identify a subgroup of patients in which the surveillance schedule for HCC can be personalized and extended.