Recombinant interferon alpha-2b (IFN-alpha2) has direct and indirect antiproliferative effects in lymphoma, and may augment cytotoxicity when combined with chemotherapy. CALGB 8691 is a randomized study of daily oral cyclophosphamide (CPA) at 100 mg/m2 with or without IFN-alpha2 at 2 x 106 IU/m2 three times per week, followed by a second randomization between IFN-alpha2 maintenance (2 x 106 IU/m2 three times weekly) versus observation in treatment-naïve patients with follicular lymphoma (FL). Five hundred eighty-one patients were randomized to either CPA (n = 293) or CPA plus IFN-alpha2 (n = 288). One hundred five responding patients were randomized to observation and 99 to maintenance IFN-alpha2. With a median follow-up of 11.5 years, the median event-free and overall survival (OS) for CPA induction alone were 2.5 years (95% CI 2.2, 3.0) and 9 years (95% CI 7.7, 10.2), compared to 2.4 years (95% CI 2.1, 3.1) and 8.4 years (95% CI 7.5, 11.1) for the combination arm (p = NS). Patients with a partial response (PR) and randomized to observation had the worst outcome (event-free survival (EFS) 1.8 years versus 3.9 years; p = 0.002). Patients with a PR randomized to IFN-alpha2 had a similar EFS to compared to patients with complete response (CR), but this did not translate into a survival advantage. Myelosuppression was increased in IFN-alpha2-containing arms. Despite the small benefit in EFS in patients with PR randomized to IFN-alpha2 maintenance, we conclude that the addition of low dose IFN-alpha2 did not significantly improve the response rate, duration of response, event-free, or OS obtained with single-agent daily oral CPA in patients with previously untreated FL.