Background/Aim: Virus-host interactions may have pathogenetic significance in chronic hepatitis. Thus the humoral immune response was evaluated during the clinical course of HCV-infected patients. Methods: Eighteen selected chronic HCV patients received three doses of 3 or 6 MU interferon-alpha 2a weekly for 6 to 12 months and were followed up for 6 to 60 months. Anti-HCV antibody levels were serially measured either in end-point diluted sera with the Matrix-Assay or with quantitative anti-HC34-IgG and -IgM ELISA, Circulating immune complexes were assessed by flow cytometry and the results were correlated with histology, quantitative HCV-RNA levels and genotypes, Results: Nine complete responders (CR; genotypes la n=4; Ib n=1; 2a n=1; 3a n=3) showing sustained virus elimination and ALT normalisation had low HCV-RNA pretreatment levels (mean 14x10(3) copies/ml) compared to six nonresponders and three partial responders (NR/PR; genotypes la n=2; Ib n=7) who had significantly higher HCV-RNA pretreatment levels (mean 254x18(3) copies/ml; p<0.01). In untreated NR/PR the HC34 core-antigen was most immunogenic, in CR the NS3-derived HC29-antigen. Pre-treatment levels of anti-HC 34-IgG and -IgM antibody levels in NR/PR were higher than in CR (IgM/IgG p=0.05, n.s.) and these differences became significant during or after therapy (3 months therapy: IgM p<0.02/IgG p<0.07; end of therapy: IgM 0.006/IgG p<0.04; 6 months post-therapy: IgM p<0.002/IgG p<0.004). The PR patients showed recurrent anti-HC34 antibody levels that preceded disease reactivation and detectable HCV-RNA in serum. Immune complex formation increased in some patients during treatment but did not correlate with disease activity, quantitative viraemia, antibody levels or therapy outcome. Conclusion: Anti-HC34 antibodies, i.e. of the IgM-subtype, correlated quantitatively with viraemia and disease activity. Monitoring the antibody levels may predict the long-term therapy outcome during interferon-alpha treatment.