Everolimus 1.5 or 3 mg/day was compared with mycophenolate mofetil (MMF) 2 g/day in a randomized, multicenter 36-month trial in de novo renal allograft recipients (n = 588) receiving cyclosporine microemulsion (CsA) and corticosteroids. The study was double-blind until all patients had completed 12 months, then open-label. By 36 months, graft loss occurred in 7.2, 16.7 and 10.7% of patients in the everolimus 1.5, 3 mg/day, and MMF groups, respectively (p = 0.0048 for everolimus 1.5 mg/day vs. 3 mg/day); efficacy failure (biopsy-proven acute rejection (BPAR), graft loss, death or lost to follow-up) occurred in 33.0, 38.9 and 37.2% of patients (p = 0.455 overall), respectively. Mortality and incidence of BPAR were comparable in all groups. Creatinine values were higher in everolimus groups, requiring a protocol amendment that recommended lower CsA exposure. Diarrhea, lymphocele, peripheral edema and hyperlipidemia were more common among everolimus-treated patients, whereas viral infections, particularly cytomegalovirus infection, increased in the MMF group. Overall safety and tolerability were better with MMF and everolimus 1.5 mg/day than with everolimus 3 mg/day. In conclusion, at 36 months, an immunosuppressive regimen containing everolimus 1.5 mg/day had equivalent patient, and graft survival and rejection rates compared with MMF in de novo renal transplant recipients, whereas everolimus 3 mg/day had inferior graft survival. Renal dysfunction in everolimus cohorts necessitates close monitoring.