Probability of non-response during interferon therapy in patients with chronic hepatitis C.

Background/aims: About 50% of patients with chronic hepatitis C do not respond to interferon therapy and this failure is expensive. The aim of this study was to identify possible predictive factors of biochemical non-response during interferon therapy among biochemical, virological (HCV genotype), histological (Knodell's score) and pharmacokinetic (monoethylglycinexylidide formation test) pre-treatment parameters.
Methodology: Our study included 60 patients with chronic hepatitis C undergoing a course of Interferon therapy. Patients whose serum ALT levels were normal at the 3rd month of therapy and remained so until the end of treatment were regarded as responders.
Results: In univariate analysis, only the gamma-glutamyltransferase (gamma-GT) and the gamma-GT/alanine aminotranferase ratio were significantly higher in non-responder patients. Multivariate logistic analysis showed that high gamma-GT levels, high histological activity index, low monoethylglycinexylidide formation rate and viral genotype 1 were the best combination for the identification of non-responder patients (16.7% error rate). By adding alanine aminotranferase modification at the 1st month of therapy the probability error was reduced to 5%.
Conclusions: These results show that the combination of biochemical, histological, virological and pharmacokinetic pre-treatment variables, associated with alanine aminotranferase modification at the 1st month of therapy, can predict non-response to interferon and allow therapeutic modifications.