Forty-nine Japanese patients were enrolled in a randomized, placebo-controlled, double-blind trial of alpha-interferon for chronic non-A, non-B hepatitis: 24 patients received 3 million units of recombinant human alpha alpha-interferon (alpha-2a) thrice weekly for eight weeks, and 25 patients received placebo in a similar schedule. The mean serum alanine aminotransferase (ALT) dropped from 155 +/- 91 (SD) to 69 +/- 72 during interferon treatment, but remained unchanged (158 +/- 140 to 147 +/- 130) during placebo treatment (P less than 0.001). Serum ALT level fell to the normal range in 29% of interferon-treated patients, but in only 4% of placebo-treated patients. Pre- and posttreatment liver biopsies were obtained in all but one case. Average histological activity indices (HAI) were markedly improved in the interferon-treated group (9.5 +/- 3.7 to 7.0 +/- 4.3), but were unchanged in the placebo group (8.5 +/- 4.3 to 8.5 +/- 4.9). In addition, we compared the efficacy of interferon treatment between anti-hepatitis C virus (HCV) antibody positive and negative groups. Biochemical and histological improvements were similar and statistically significant in patients with and without antibody to hepatitis C virus. These data indicate that a eight-week course of alpha-interferon induces biochemical and histological improvement in more than half the patients with chronic non-A, non-B hepatitis.