Background: An early virological response to interferon-alpha treatment is a strong predictor of sustained response, but it has never been exploited to stratify patients in clinical trials. Aim: To evaluate the efficacy of amantadine plus interferon-a compared with interferon-alpha alone in naive patients with chronic hepatitis C who were randomized on the basis of the early virological response to interferon-alpha. Methods: One hundred and eighty-one patients received recombinant interferon-alpha2a (3 MU three times weekly) for 2 months and 164 were evaluated for early (i.e. month 2) virological response. Hepatitis C virus (HCV) RNA-negative patients (n = 66) were randomized to receive 3 MU of interferon-alpha three times weekly, with or without amantadine (200 mg/day); HCV RNA-positive patients (n = 98) were randomized to receive 6 MU of interferon-alpha three times weekly, with or without amantadine (200 mg/day). HCV RNA-positive patients at 6 months discontinued treatment, and all others completed 12 months. Results: At month 6, HCV RNA-negative patients made up 54.2% of the interferon + amantadine group and 42.0% of the monotherapy group (P = 0.07). At month 12, HCV RNA-negative patients made up 38.5% of the interferon + amantadine group and 28.4% of the monotherapy group (N.S.). The sustained virological response rates were 21.6% and 20.9%, respectively (N.S.). Conclusion: The addition of amantadine does not enhance the sustained virological response to interferon-a in naive patients with chronic hepatitis C; however, an additive effect of amantadine occurs in the first 6 months, mainly in patients without an early response to monotherapy. Early response to interferon-alpha is a strong predictor of sustained virological response.