Markers of epithelial-mesenchymal transition (EMT) may identify patients at high risk of graft fibrogenesis who could benefit from early calcineurin inhibitor (CNI) withdrawal. In a randomized, open-label, 12-month trial,de novokidney transplant patients received cyclosporine, enteric-coated mycophenolate sodium (EC-MPS) and steroids to month 3. Patients were stratified as EMT+ or EMT− based on month 3 biopsy, then randomized to start everolimus with half-dose EC-MPS (720 mg/day) and cyclosporine withdrawal (CNI-free) or continue cyclosporine with standard EC-MPS (CNI). The primary endpoint was progression of graft fibrosis (interstitial fibrosis/tubular atrophy [IF/TA] grade increase ≥1 between months 3–12) in EMT+ patients. 194 patients were randomized (96 CNI-free, 98 CNI); 153 (69 CNI-free, 84 CNI) were included in histological analyses. Fibrosis progression occurred in 46.2% (12/26) CNI-free EMT+ patients versus 51.6% (16/31) CNI EMT+ patients (p = 0.68). Biopsy-proven acute rejection (BPAR, including subclinical events) occurred in 25.0% and 5.1% of CNI-free and CNI patients, respectively (p < 0.001). In conclusion, early CNI withdrawal with everolimus initiation does not prevent interstitial fibrosis. Using this CNI-free protocol, in which everolimus exposure was relatively low and administered with half-dose EC-MPS, CNI-free patients were overwhelmingly under-immunosuppressed and experienced an increased risk of BPAR.