Objective. T cells are involved in the pathogenesis of rheumatoid arthritis (RA). In animal models of autoimmune diseases, blockade of costimulatory molecules on antigen-presenting cells has been demonstrated to be effective in preventing or treating this disease by preventing T cell activation. To date, the effect of costimulatory blockade in patients with RA is unknown. The goal of this multicenter, multinational study was to determine the safety and preliminary efficacy of costimulatory blockade using CTLA-4Ig and LEA29Y in RA patients who have been treated unsuccessfully with at least 1 disease-modifying agent.Methods. CTLA-4Ig, LEA29Y (0.5, 2, or 10 mg/ kg), or placebo was administered intravenously to 214 patients with RA. Patients received 4 infusions of study medication, on days 1, 15, 29, and 57, and were evaluated on day 85. The primary end point was the proportion of patients meeting the American College of Rheumatology 20% improvement criteria (ACR20). All patients were monitored for treatment safety and tolerability.Results. CTLA-4Ig and LEA29Y infusions were well tolerated at all dose levels. Peri-infusional adverse events were carefully monitored, and showed similar incidence across all dose groups with the exception of headaches, which were slightly more frequent in the 2 treatment groups. The incidence of discontinuations due to worsening of RA was 19%, 12%, and 9% at 0.5, 2, and 10 mg/kg, respectively, in the CTLA-4Ig-treated patients and 3%, 3%, and 6% at 0.5, 2, and 10 mg/kg, respectively, in the LEA29Y-treated patients (versus 31% in the placebo group). ACR20 responses on day 85 had increased in a dose-dependent manner (23%, 44%, and 53% of CTLA-4Ig-treated patients and 34%, 45%, and 61% of LEA29Y-treated patients at 0.5, 2.0, and 10 mg/kg, respectively, versus 31% of placebo-treated patients).Conclusion. Both of the costimulatory blocking molecules studied were generally safe and well tolerated. As compared with placebo, both CTLA-4Ig and LEA29Y demonstrated efficacy in the treatment of RA.