Rituximab, a monoclonal antibody that targets the CD20 antigen, has shown single-agent activity in patients with previously untreated and relapsed follicular lymphoma (FL). Mechanisms of rituximab-mediated antineoplastic activity include complement-mediated cell (CDC) lysis, apoptosis, and the induction of antibody-dependent cell-mediated cytotoxicity (ADCC). Granulocyte macrophage colony stimulating factor (GM-CSF) is a biologic agent capable of enhancing proliferation of cell populations and cell surface receptors necessary for rituximab activity. It was hypothesized that the combination of GM-CSF and rituximab would substantially improve the treatment efficacy of rituximab in relapsed, FL patients. The current study assessed the clinical response, biologic response, and safety of combination rituximab and GM-CSF in patients with relapsed FL. Response and toxicity were defined using previously determined standard criteria (Cheson B, et al. J Clin Oncol.1999;17:1244–1253) and World Health Organization standard criteria. A total of 46 patients, with a median age of 60 years (range, 37–80 years), received rituximab and GM-CSF between January 2000 and October 2000. This analysis focuses on 34 patients with relapsed or progressive FL. GM-CSF was administered at 5 mg/kg/day on Days 1–8 and rituximab was administered at 375 mg/m 2 /week on Day 5 of each 21-day cycle for a total of 4 cycles. Throughout the treatment and follow-up periods, all patients underwent clinical examinations, bone marrow biopsies, standard biologic assays (eg, lactate dehydrogenase, β 2 -macroglobulin, creatinine, and liver enzymes), complete blood cell counts, immunophenotyping, and polymerase chain reaction analysis for bcl-2 expression. Response was evaluated at Days 30, 60, 90, and every 4 months until progression. Of the 34 enrolled patients, 1 patient was excluded due to transformation and 4 patients withdrew from the study due to toxicity. The complete response (CR) rate was 45% and the overall response rate was 70% (15 CR + CR unconfirmed and 8 partial responses [PR]) for the 33 intent-to-treat patients with relapsed FL. CR was reached in 63% (7 of 11) patients who received ≥3 lines of previous therapy (range, 3–7 lines). The median progression-free survival was 16.7 months. A significant increase ( P P = 0.012). In conclusion, the relatively high CR rate, combined with the longer than expected duration of response in this treatment refractory population, suggests that the combination of rituximab plus GM-CSF is a safe and effective regimen in the treatment of FL patients and warrants further investigation.