The anti-cytomegalovirus (CMV) activity and safety of oral maribavir in CMV-seropositive allogeneic stem-cell transplant recipients were evaluated in a randomized, double-blind, placebo-controlled, dose-ranging study. After engraftment, 111 patients were randomized to receive CMV prophylaxis with maribavir (100 mg twice daily, 400 mg once daily, or 400 mg twice daily) or placebo. Within the first 100 days after transplantation, the incidence of CMV infection based on CMV pp65 antigenemia was lower in each of the respective maribavir groups (15%, P = .046; 19%, P = .116; 15%, P = .053) compared with placebo (39%). Similarly, the incidence of CMV infection based on plasma CMV DNA was lower in each of the respective maribavir groups (7%, P = .001; 11%, P = .007; 19%, P = .038) compared with placebo (46%). Anti-CMV therapy was also used less often in patients receiving each respective dose of maribavir (15%, P = .001; 30%, P = .051; 15%, P = .002) compared with placebo (57%). There were 3 cases of CMV disease in placebo patients but none in the maribavir patients. Adverse events, mostly taste disturbance, nausea, and vomiting, were more frequent with maribavir. Maribavir had no adverse effect on neutrophil or platelet counts. These results show that maribavir can reduce the incidence of CMV infection and, unlike ganciclovir, does not cause myelosuppression. This trial is registered at www.ClinicalTrials.gov as #NCT00223925.