To evaluate the efficacy and tolerance of a calcineurin inhibitor (CNI)-free regimen, 145 renal recipients were prospectively randomized to receive either sirolimus (n = 71) or cyclosporine (CsA; n = 74). All patients received polyclonal antilymphocyte antibodies, mycophenolate mofetil (MMF) and steroids (6 months). The primary endpoint, estimated glomerular filtration rate (eGFR) was not significantly different at 12 months comparing sirolimus- and CsA-treated patients (60 +/- 27 vs. 57 +/- 21 mL/min). At 12 months, patient and graft survival, incidence of biopsy-proven rejection and rates of steroid withdrawal were not statistically different (97% vs. 97%; 90% vs. 93%; 14.3% vs. 8.6% and 82.8% vs. 84.1%, respectively). Delayed and slow graft function (SGF) was not significantly different (18.6% vs. 12.3% and 11.4% vs. 13.7%, respectively). In patients who remained on treatment according to protocol at 12 months, eGFR was significantly higher with sirolimus (69 +/- 19 vs. 60 +/- 14 mL/min, p = 0.01). Overall study drug discontinuation rates were 28.2% with sirolimus and 14.9% with CsA. Adverse events (wound complications, mouth ulcers, diarrhea, hypokalemia, bronchopneumonia) and proteinuria >0.5 g/24h (38.8% vs. 5.6%, p < 0.001) were significantly more frequent in sirolimus-treated patients. Cytomegalovirus (CMV) infections were significantly less frequent with sirolimus (6% vs. 23%, p < 0.01). A CNI-free regimen using sirolimus-MMF can achieve excellent renal function, but patients on sirolimus experienced a high rate of adverse events and study drug discontinuation.