Interleukin-2 (IL-2) has been shown to produce durable complete remission in patients with renal cell carcinoma (RCC). A phase 2 study was conducted to evaluate the potential therapeutic synergy as well as the toxic side effects of the concurrent administration of IL-2 and granulocyte-macrophage colony-stimulating factor (GM-CSF) in patients with advanced stage disease. Twenty-one patients with unresectable or metastatic RCC having an Eastern Oncology Cooperative Group performance status of 0 or 1 were enrolled. Six patients had received prior immunotherapy with interferon (IFN)-alpha, IFN-gamma, and IL-12, whereas the remaining 15 subjects were previously untreated. Thirteen patients were assigned to a moderate-dose bolus of IL-2 at 72,000 IU/kg every 8 hours on days 1 through 5 and days 15 through 19, whereas 8 patients were given IL-2 as an intravenous continuous infusion at a dose of 5 MU/m2/d on days 1 through 5 and days 15 through 19. Subcutaneous GM-CSF at 125 microg/d on days 1 through 21 was administered concomitantly with IL-2. The median number of IL-2 bolus doses was 23 of a scheduled 28 (85%), whereas with the continuous infusion, 93% of planned IL-2 was given. All patients received 100% of GM-CSF doses. There were no complete or partial responses in this study. Of 13 patients treated in the bolus IL-2 arm, 10 had systemic progression of disease at 4 to 8 weeks, 1 developed metastasis in the brain at 4 weeks, and 2 had stable disease for 4 and 17 months. Among the 8 subjects treated with continuous infusion IL-2, 3 progressed with brain lesions at 3 to 8 weeks and 5 had stable disease at 6+, 7, 8+, 15+, and 17+ months. The median survival for the whole group was 10 months, with a range of 0.5 to 40+ months. There were no regimen-related deaths, and most of the observed toxicities were grade 1 and 2. Serious toxicities (grade 3 and 4) included anemia, atrial fibrillation, oliguria, abnormal liver function, and neurologic events like agitation or confusion. The combination of recombinant IL-2 and GM-CSF administered in the designed schedule and doses was not effective in patients with metastatic RCC and may even interfere with the therapeutic potential of moderate-dose IL-2 and increase its adverse events.