In an attempt to improve therapeutic efficacy and limit antimurine responses to the IgG2a monoclonal antibody OKT4A, humanized complementarity determining region (CDR)-grafted OKT4A mAbs (IgG1 and IgG4 isotypes) were developed from the murine molecule. Preclinical evaluation was undertaken in 12 cynomolgus renal allograft recipients (IgG1, n = 7; IgG4, n = 5). Control animals received either no therapy (n = 2) or OKT3 (n = 2). The mAbs were given as a single 10-mg/kg bolus on the day of transplantation or as multiple 1-mg/kg doses. Mean allograft survival (+/- SEM) was 9 +/- 0.7 days in control animals; 45.2 +/- 6.0 days, P = 0.002 (single dose, n = 5), and 39 +/- 5.0 days, P = 0.053, (multiple doses, n = 2) in IgG1-treated animals; and 35.3 +/- 7.2 days, P = 0.034, (single dose, n = 3) and 15.5 +/- 6.5 days, P = .251 (multiple doses, n = 2) in IgG4-treated animals. Whereas IgG4-treated animals showed coating of peripheral blood CD4+ T cells, significant CD4+ T cell depletion was observed in IgG1-treated animals. These results confirm the retained immunosuppressive efficacy of humanized OKT4A antibodies. In contrast to murine mAbs, the use of these agents would not preclude sequential treatment with mAbs directed against different epitopes. The fact that rejection can occur despite peripheral blood CD4+ cell depletion suggests that indefinite control of allograft rejection in primates may require more intensive therapy than has proved effective in rodent models.