Relapse after interferon (IFN) therapy for chronic hepatitis C virus (HCV) infection occurs in 50% of patients after the initial response. The benefit of retreatment with IFN alone has not been assessed in large controlled studies. To assess the effectiveness and the tolerability of IFN retreatment and to identify the optimal second course regimen, we performed a meta-analysis of individual patient's data on a set of 549 patients (mean age 43.8 years; 12.2 SD, men: 65%) who had an end-of-treatment biochemical response to a first IFN course and then relapsed. Retreatment was started within 24 months after the end of the first course. Biochemical end-of-treatment responses (ETR) and sustained responses (SR) were observed in 405 of 549 (73.8%; 95% confidence interval [CI] 70.1-77.5) and in 124 of 549 (22.6%; CI 19.1-26.1) patients, respectively. One hundred seventy-five of 404 patients (43.3%; CI 38.6-48.2) developed an end-of-treatment, biochemical, and virological response when retreated. A biochemical and virological SR to retreatment occurred in 73 of 494 (14.8%; CI 11.7-18) patients. Thirty-two patients (5.8%; CI 3.5-7.8) stopped retreatment for adverse effects. Biochemical and virological SR was predicted independently by logistic regression analysis using a negative HCV RNA at the end of the first cycle of IFN (P= .01) and by retreatment with a high IFN dose (P= .03). Age, cirrhosis, genotype, and γ-glutamyl transferase levels before retreatment were not significant by multivariate analysis. The excellent tolerability of IFN monotherapy retreatment makes it an option for patients who transiently cleared HCV-RNA during their first IFN course. Patients should be retreated with a high IFN dose regardless of the strength of the dose received during the previous course of treatment.