Interferon (IFN) treatment of chronic hepatitis C virus (HCV) is associated with a high rate of relapse. IFN is thought to exert its effect against HCV via direct viral inhibition and immune stimulation. We have hypothesized that relapse following termination of therapy results from the sudden withdrawal of this immune modulatory effect and that gradual reduction in the IFN dose may decrease the incidence of relapse. One hundred six patients with chronic HCV were enrolled into this 24-month controlled, randomized prospective trial. All were treated with 5 mU of interferon-alpha-2b three times a week for 6 months. Patients who achieved biochemical response were randomized to either stop or taper IFN gradually at monthly intervals as follows; 3 mu, 2 mU, 1 mU, and 0.5 mU (all three times a week). 0.5 mU twice weekly and then once weekly. Liver histology was assessed by Knodell index and HCV RNA was measured by a quantitative polymerase chain reaction (PCR) assay. Of the 92 patients who completed the initial 6 months of IFN treatment, 47 (51%) achieved biochemical response. Twenty-one of these patients were randomized to stop IFN treatment and 25 to taper (1 drop-out). At randomization patients were well matched with respect to age, sex, race, serum alanine transaminase (ALT), and liver histology. Biochemical relapse was observed in 19 of 21 (91%) patients who stopped IFN treatment compared with only 60% who tapered IFN (P= .04). Virological relapse occurred in 90% of patients who stopped and only 48% of persons who tapered IFN therapy. At completion of the 24-month study patients who achieved long-term sustained biochemical response had a significantly lower mean Knodell score (3.5 vs. 6.5) and a significantly greater number were HCV RNA negative in serum (85% vs. 18%) compared with relapsers. We conclude that gradual reduction in IFN dose is associated with a significant higher rate of sustained response and clearance of HCV RNA from serum compared with abruptly stopping treatment. This in turn is associated with a significant improvement in hepatic histology supporting the premise that response to IFN therapy can prevent progression to cirrhosis.