S ide effects after the first administration of OKT3, a murine anti-CD3 monoclonal antibody (mAb) of the IgG2a class, are largely attributed to the release of cytokines as a result of T cell activation caused by interaction with Fc receptors (FcR) on human monocytes. As human monocytes possess FcR for murine IgG2a but not for IgA, it is expected that an anti-CD3 mAb of the IgA class causes less side-effects than an IgG2a anti-CD3 mAb of the same idiotype. To test this hypothesis we treated 20 renal transplant patients prophylactically with either IgG2a or IgA anti-CD3 mAb in a prospective randomized double-blind study. The patients received 0.5 mg anti-CD3 mAb, either IgA (T3.A) or IgG2a (T3.G2a), twice daily during 10 d. Rejection incidence after T3.A and T3.G2a was not significantly different. Side effects score after the first administration of mAb was significantly less after T3.A than after T3.G2a (0.7 vs 2.7, P = 0.002). IL-6 and gamma IFN levels increased significantly at 3 h after T3.G2a, but not after T3.A. The TNF peak level occurring at 1 h after T3.A was much lower than after T3.G2a. In plasma, complement and neutrophil activation products only increased after T3.G2a and not after T3.A. Both T3.A and T3.G2a resulted in a complete depletion of CD3+ cells, but after T3.A, CD3 depletion was of shorter duration than after IgG2a. Finally, in contrast to T3.G2a, T3.A did not affect coagulation and fibrinolysis. In conclusion, an anti-CD3 mAb of the IgA class causes hardly any cytokine release and less side-effects as compared with its IgG2a switch variant. Provided T3.A is sufficiently immunosuppressive, it is superior to OKT3.