Treatment of chronic hepatitis C with PegIFN-alfa2b and ribavirin within a large German multicentric observational study: predictors of nonresponse in genotype 1 patients

Aims: The current treatment of choice is a combination of pegylated interferon-alfa (PegIFN-alfa) plus ribavirin. Prediction of IFN efficacy before treatment has been mainly based on viral characteristics such as viral load and genotype. We therefore analyzed data from a large multicentric German surveillance study in order to identify additional host- related predictors of nonresponse. Methods: Between 9/2003 and 8/2008, 279 active sites treated a total of 4039 patients with weight-based PegIFN-alfa2b and ribavirin. As this is an ongoing surveillance trial, we have performed an interim analysis of formerly untreated noncirrhotic patients with HCV-genotype 1 and nonresponse defined as HCV-RNA-positivity at end of treatment (EOT) as well as at the end of follow-up (EOF) 6 months after end of treatment. Results: 455 treated patients with genotype 1 failed to achieve HCV-RNA-negativity at EOT and EOF, while 614 patients experienced sustained virological response after 6 months of follow-up (SVR). Baseline data and rapid virologic response (RVR) in both groups are shown in the table. Because data were not available in all patients the number of patients fluctuates throughout the subgroups analyzed. No dose reductions of either PegIFN-alfa2b or ribavirin were performed in 437 (71.2%) of patients with SVR and 324 (71.2%) in the nonresponder-group. Reductions of ribavirin have been made in 69 (11.2%) of SVR-patients and 31 (6.8%) of patients with nonresponse, respectively. Only 7/209 (3.3%) nonresponding patients, who were tested for HCV-RNA at week 4, experienced an RVR, while 21.4% of all SVR-patients exhibited an RVR. Conclusions: Regarding baseline predictive values, there is a significant difference in age as well as baseline viral load while other factors such as body weight and BMI seem not to predict nonresponse in noncirrhotic patients. In contrast, RVR may predict low nonresponse rates, as break through after week 4 are rare events. Data were analyzed using standard summary statistics and two-sided 95% confidence intervals.