Prediction of Liver-related Outcomes After HCV Cure

Overview

Objectives: To develop and validate a predictive model, applicable to daily practice, of liver complications emergence in hepatitis C virus (HCV)-infected patients and advanced fibrosis, who have achieved sustained viral response (SVR) with direct-acting antivirals (DAA)-based therapy. Methods: Design: Mulsite prospective multicenter cohort study. Study subjects: HCV-monoinfected and HIV/HCV-coinfected individuals recruited from two parallel cohorts (GEHEP-MONO Cohort clinicaltrials.gov ID: NCT02333292(HEPAVIR-DAA Cohort clinicaltrials.gov ID: NCT02057003). These cohorts enrolled patients with HCV infection, treated with DAA-based regimens after October 2011, at the units of infectious diseases of 18 hospitals throughout Spain. Patients who fullfilled the following inclusion criteria are included in this study: 1) Have received a regimen with one or more DAA; 2) Have achieved SVR 12 weeks after treatment; 3) Have an evaluable liver stiffness (LS) of more than 9.5 kPa in the three months prior to the start of treatment. Follow-up: The baseline time point is the date of SVR. All participants are evaluated by a common protocol every six months. At every visit, clinical and laboratory examination focusing on the early detection of liver complications are carried out. LS is assessed by vibration-controlled transient elastography, according to a standardized procedure, every 12 months. In patients with cirrhosis, liver ultrasound and plasma alpha-fetoprotein determination are conducted for hepatocellular carcinoma screening, every six months. Variables and data analysis: The primary outcome variable of the study will be the emergence of liver complication (hepatic decompensation or hepatocellular carcinoma) or liver transplant. Predictive models will be develop with clinical, analytical, and genetic variables independently associated with the primary variable in a Cox regression for competitive risks applied to a developmental subpopulation. The performance of the model will be evaluated using COR curves. Sensitivity, specificity, and positive and negative predictive values will be calculated, both in the developmental population and in a validation population.

Conditions