RESUMEN
BACKGROUND & AIMS: Undetectable HCV RNA at 12 weeks is the stopping rule recommended in HCV patients in whom previous treatment has failed. Whether earlier virological criteria may be useful for deciding treatment discontinuation remains subject of debate. The aim of this study was to identify, in HCV-1 non-responders and relapsers to IFN or Peg-IFN and ribavirin, the earliest and most accurate predictor of failure to respond to a new treatment combining Peg-IFN and ribavirin. METHODS: Prediction of SVR was assessed using the area under the ROC (AUROC) curve of reduction in viral load at different time points. RESULTS: This study included 151 patients (32% with extensive fibrosis or cirrhosis). A SVR was reached in 34% (21% in non-responders and 59% in relapsers). In non-responders, 1 month was the most accurate time point for predicting SVR (AUROC: 0.787 ± 0.075, p = 0.0001). Thirty-seven percent of non-responders did not have a 1-log drop in viral load at 1 month. All these patients had detectable HCV RNA at 3 months (p < 0.0001) and only 4% attained a SVR (p = 0.004). The same high negative predictive value for SVR was found in sensitivity analysis restricted to non-responders to Peg-IFN and ribavirin. In contrast, in relapsers, undetectable HCV RNA at 3 months was the earliest criterion with high negative predictive value (92%, p < 0.0001). CONCLUSIONS: All HCV-1 non-responders who did not have a 1-log drop in viral load at 1 month remained HCV-RNA-detectable at 3 months, and only 4% attained a SVR. This new criterion can be used early on as a first stopping rule.