Prediction of treatment outcome in patients with chronic hepatitis C: significance of baseline parameters and viral dynamics during therapy.

In patients with chronic hepatitis C virus (HCV) infection scheduled for a 48-week treatment period, premature discontinuation of treatment was previously recommended if HCV-RNA levels remained detectable at week 24 of therapy. Considering the number of side effects and treatment costs, measurement of initial viral decline during therapy may identify virologic nonresponse earlier than 24 weeks. We retrospectively analyzed 260 European patients treated with standard or pegylated interferon alfa (IFN-alpha) and ribavirin for 24 to 48 weeks. Early prediction of virologic response by HCV-RNA decline at weeks 4 and 12 (Versant Quantitative [branched DNA (bDNA) 3.0]; Bayer Diagnostics, Emeryville, CA; and Qualitative [transcription-mediated amplification (TMA)] HCV RNA assay; Bayer Diagnostics) as well as clinical, biochemical, virologic, and histologic baseline parameters were analyzed by logistic regression and receiver operating characteristic (ROC) curves. A viral load at treatment week 4 above 450,000 IU/mL and at week 12 above 30,000 IU/mL was 100% predictive for virologic nonresponse in all patients. From multivariate logistic regression analysis of all patients, independent predictors for sustained virologic response were: genotypes 2 and 3 (P <.0001), a low baseline gamma-glutamyltransferase (GGT) level (P <.0001), a high baseline alanine aminotransferase level (P =.002), and a low baseline viral load (P =.04). None of the latter 3 factors were predictive for sustained virologic response when analysis was restricted to the subgroup of genotypes 2- and 3-infected patients. In conclusion, virologic nonresponse can be predicted early at week 12 of treatment independent from the applied therapeutic regimen based on a cutoff level for HCV RNA of 30,000 IU/mL. This algorithm recognizes 53.7% of nonresponders previously identified at week 24 of treatment.