To study the antiviral effect of lymphoblastoid alpha interferon (IFN) on hepatitis C virus (HCV) we conducted a randomized, controlled trial on 80 patients with chronic hepatitis C using three different doses. Patients were randomly assigned to treatment with 1, 3 or 6 million units of lymphoblastoid IFN-alpha daily for 2 weeks. To assess the antiviral effect of IFN, the amount of HCV present in the serum was estimated by competitive nested polymerase chain reaction (PCR) before and after 2 weeks of treatment. The multiple logistic analysis was used to evaluate factors associated with virus clearance, adjusting the imbalance in predictive factors among patients. Hepatitis C virus became negative as assessed by nested PCR after therapy in 26, 50 and 63% of patients receiving 1, 3 and 6 mega units, respectively. Hepatitis C virus was cleared more often in patients having initially low (< 10(5)/mL) amounts of virus. No significant decrease in the amount of virus was observed in the untreated, control group. Patients without bridging fibrosis in liver histology and with HCV genotypes other than K1 (type II) tended to respond well. These results indicate that lymphoblastoid IFN-alpha suppresses HCV in a dose dependent manner. Higher initial virus amounts, bridging fibrosis and genotype K1 were factors associated with poor response.