We have conducted a multicenter randomized controlled trial comparing two doses of recombinant human alpha-interferon for efficacy in 60 patients with chronic non-A, non-B hepatitis. The source of infection appeared to be transfusion in 30 patients, intravenous drug abuse in 16 patients and was unknown in 14 patients. Patients were randomly assigned to no treatment or to treatment with either 1 or 3 MU of alpha-interferon given three times a week for 24 wk. Forty-five patients (75%) were positive for antibody to hepatitis C virus. During the 24-wk treatment period, mean serum ALT levels decreased in both treatment groups, but the decrease was statistically significant only in the 3 MU group. However, at 24 wk, the proportion of patients with normal ALT levels was similar in the 3 MU group (39%) and the 1 MU group (45%), and both were significantly higher than in controls (0%). Repeat liver biopsy specimens showed a significant decrease in the severity of histological changes in the 3 MU group but not in the 1 MU group or in controls. Responses to alpha-interferon did not correlate with patient's age, gender, source of infection, pretreatment serum ALT, presence of anti-hepatitis C virus or cirrhosis. After treatment, the mean ALT levels rose in both treated groups. The proportion of patients with normal ALT levels at wk 48 was 28% in the 3 MU group and 20% in the 1 MU group. In conclusion, a dose of 3 MU was superior to 1 MU of alpha-interferon given three times weekly for 24 wk in inducing improvements in serum ALT levels and liver histological examinations.(ABSTRACT TRUNCATED AT 250 WORDS)