Background. Calcineurin inhibitors (CNIs) are essential immunosuppressive drugs after renal transplantation. Because of nephrotoxicity, withdrawal has been a challenge since their introduction. Methods. A randomized multicenter trial included 212 kidney patients transplanted between 1997 and 1999. All patients were initially treated with mycophenolate mofetil (MMF), cyclosporine A (CsA), and prednisone (pred). At 6 months after transplantation, 63 patients were randomized for MMF/pred, 76 for MMF/CsA, and 73 for MMF/CsA/pred. Within 18 months after randomization 23 patients experienced a rejection episode: MMF/pred (27.0%), MMF/CsA (6.8%) and MMF/CsA/pred (1.4%) (P<0.001). Results. During 15 years of follow-up, 73 patients died with a functioning graft, and 43 patients lost their graft. Ninety-six were alive with a functioning graft. Intention-to-treat analysis did not show a significant difference in patient and graft survival. In multivariate analysis, death-censored graft survival was significantly associated with serum creatinine at 6 months after transplantation and maximum PRA but not with the randomization group. CNI withdrawal did not result in a reduced incidence of or death by malignancy or cardiovascular disease. Death-censored graft survival was significantly worse in those patients randomized for CNI withdrawal that had to be reverted to CNI. Independent of randomization group, compared with no rejection, death-censored graft survival was significantly worse in 23 patients with acute rejection after randomization. Conclusion. Fifteen years after conversion to a CNI free regimen, there was no benefit regarding graft and patient survival or regarding prevalence of or death by comorbidities. However, rejection shortly after CNI withdrawal was associated with decreased graft survival.