We compared the efficacy and safety of empirical plus PCR-based vs empirical liposomal amphotericin B treatment after Allo-SCT. Allo-SCT recipients were randomized to receive either PCR-based preemptive therapy (group A; n=198) or empirical antifungal therapy (group B; n=211) with liposomal amphotericin B. In group A, therapy was started after one positive PCR result or after 120 h of febrile neutropenia refractory to broad-spectrum antibacterial therapy. In group B, liposomal amphotericin B was started after 120 h of refractory febrile neutropenia. Demographic and clinical characteristics were well balanced. A total of 112 (57.1%) patients in group A and 76 (36.7%) patients in group B received antifungal therapy (P<0.0001). Twelve patients in group A and 16 patients in group B developed proven invasive fungal infection (IFI). Survival curves showed better survival until day 30 when close PCR monitoring was performed (mortality 1.5 vs 6.3%; P=0.015), but there was no difference at day 100. At day 100, no difference was observed in the incidence of IFI (primary end point) and survival between the two arms. Further studies are required to assess the benefit of using PCR in patients after SCT.