Oral Voriconazole Versus Intravenous Low Dose Amphotericin B for Primary Antifungal Prophylaxis in Pediatric Acute Leukemia Induction

Purpose: Invasive fungal infections (IFI) are a major cause of infection-related mortality during induction chemotherapy of acute leukemia (AL) patients. Data on antifungal prophylaxis (AFP) in children are limited by retrospective design, small sample size, and variability of chemotherapy phases having different risk of IFI. There are no data comparing voriconazole versus amphotericin B (AmB) as AFP in either adult/pediatric AL. The objectives of this study were to compare efficacy and toxicity of AmB and voriconazole as AFP in pediatric AL patients. Patients and Methods: As a pilot study, total 100 children (≤15 y) with denovo acute myeloid leukemia and acute lymphoblastic leukemia were randomized to either oral voriconazole or low dose intravenous AmB as AFP during induction chemotherapy. Results: Failure of prophylaxis occurred in 14/50 patients in voriconazole arm (1 proven mucormycosis, 1 possible IFI, 11 empirical antifungal therapy, and 1 withdrawal owing to hepatotoxicity) and 17/50 patients in AmB arm (3 possible IFI, 13 empirical antifungal therapy, and 1 withdrawal owing to difficult venous access) (P=0.66). Of the 29 patients who had failure of prophylaxis unrelated to drug toxicity, computed tomography of the chest showed infiltrates in 10 patients with 3/12 in voriconazole arm and 7/16 in AmB arm (P=0.43). Drug-related serious adverse events were 6% versus 30% in voriconazole and AmB arms, respectively (P<0.01). Further, total number of toxicities per patient in AmB arm were significantly higher as compared with voriconazole arm (P<0.0001). Conclusion: This is the first randomized study comparing voriconazole with AmB in pediatric AL patients as AFP during induction chemotherapy; our results showed that oral voriconazole seems to be comparable with AmB with less toxicity and more convenience. (ClinicalTrials.gov identifier: NCT00624143).