The classical pegylated interferon α (peg-IFNα) and ribavirin (RBV) treatment of chronic hepatitis C (CHC) is progressively being replaced by new direct acting antivirals, whose costs remain a major barrier to widespread use. Using baseline data and viral kinetics, we developed a predictive algorithm to allocate to DAA patients who are not going to respond to peg-IFNα/RBV. This prospective study evaluated 205 CHC patients treated with peg-IFNα/RBV. HCVRNA kinetics during the initial 3 days of therapy and baseline variables including age, genotype, fibrosis and ALTs were used to construct a prediction rule in terms of sustained virological response (SVR). One hundred and twenty-one patients achieved an SVR (59%). Variables independently associated with SVR were HCVRNA, ALT, glycaemia, viral genotype, and fibrosis. The decline of viremia from baseline to 48/72 h was significantly different in SVR compared to non-SVR patients (2.2 vs. 0.65 log10 IU/mL; p < 0.001), and was influenced by viral genotype, levels of ALT, stage of fibrosis and IL28B polymorphism. In genotype 1, HCVRNA decline <0.8 logs had a negative predictive value of 90%, and in genotype 2, HCVRNA decline >1.2 logs had a positive predictive value of 92%. A combination of HCVRNA kinetics and a score based on pre-treatment parameters was highly accurate in predicting SVR in most patients. Outcome of peg-IFNα/RBV treatment may be predicted combining evaluation of baseline variables and HCVRNA kinetics. This allows to individualize treatment, reserving newer and more expensive DAAs to CHC patients who are in most need of them.
Keywords: Chronic Hepatitis C; Pharmacokinetics; Predictive value of tests; Ribavirin; Sofosbuvir; Viremia.