In a 6-month open label, randomized, multicenter trial, we compared the efficacy and safety of mycophenolate mofetil (MMF) with high dose intravenous steroids (IVS) for the treatment of refractory, acute cellular rejection in recipients of first or second cadaveric or living-donor renal allografts. A total of 150 patients were enrolled and randomized in a 1-to-1 ratio to receive oral MMF 1.5 g twice daily (n=77) or i.v. methylprednisolone 5 mg/kg for 5 days (n=73), tapered over the subsequent 5 days to 20 mg/day or the baseline dose of steroid given on the day before the diagnosis of rejection. Patients in both groups generally received cyclosporine and maintenance doses of corticosteroids throughout the study period. The IVS group (but not the MMF group) was generally maintained on azathioprine. The primary efficacy variable was graft and patient survival at 6 months. Graft loss and death were reduced by 45% in the MMF treatment group; 19 patients (26.0%) in the IVS group experienced graft loss or died, compared with 11 patients (14.3%) in the MMF group (P=0.081, sequential probability ratio test analysis). In the IVS group, 64.4% of patients experienced either subsequent biopsy proven rejection, presumptive rejection (presumed rejection clinically diagnosed but not biopsy proven and treated with a full course of immunosuppressive therapy), or treatment failure (premature termination for any reason, including death, graft loss, or an adverse event) compared with 39.0% in the MMF group (P=0.001, Cochran-Mantel-Haenszel [CMH] general association test). One or more full courses of immunosuppressive treatment for subsequent rejection episodes were administered to 35.6% of patients in the IVS group and 24.7% of patients in the MMF group. The number of patients who received full courses of corticosteroids for subsequent episodes of rejection was equal in the 2 groups, but the number of patients who received full courses of antilymphocyte therapy was more than twice as great in the IVS group (n=18) compared with the MMF group (n=8). Adverse events were reported in 74.6% of patients who received IVS and in 93.5% of patients who received MMF. A cerebral lymphoma developed in 1 patient in each group, and a lymphoproliferative disorder developed in 2 patients in the MMF group; in 1 of these patients, the lymphoproliferative disorder was subsequently determined to be present before study entry. Opportunistic infections occurred in 35% of patients in each treatment group. The incidence of cytomegalovirus (CMV) viremia/syndrome was comparable between groups, but tissue-invasive CMV was reported for 7 patients (9.1%) who received MMF, compared with 1 patient (1.4%) who received IVS. At 12 months postenrollment, the cumulative proportions of patients who died or lost their grafts were 31.5% in the IV steroid arm versus 18.2% for the MMF arm (P=0.042, CMH general association test, not adjusted for sequential monitoring). Thus, graft loss and death were reduced by 42% in the MMF treatment group. In summary, MMF is a clinically promising immunosuppressant and was effective for the treatment of refractory acute rejection with continuing administration for prophylaxis of subsequent rejection.