Abstract

OBJETIVE: To analyse the eficacy and safety of voriconazole as primary prophylaxis in hematologic patients with high risk of invasive fungal infections (IFI) due to severe neutropenia after induction/intensification chemotherapy for acute myeloid leukemia (AML) and allogeneic hematopoietic cell transplantation (allo-HCT).

PATIENTS AND METHODS: between June-04 and May-05 we prospectively included 31 episodes of severe neutropenia (<0,5x109/l, >10 days) after chemotherapy for AML and 24 patients submitted to an allo-HCT in a primary prophylaxis treatment with voriconazole (200 mg po/12 h) from the starting of chemotherapy until the neutrophils recovery (>0,5x109/l) in AML and from the starting of the conditioning regimen until the withdrawal of immunossupression in allo-HCT (by three months after transplantation) and we compared both groups with historic controls (63 episodes of neutropenia after AML and 31 allo-HCT) treated between January-03 and May-04 with fluconazole as primary prophylaxis (400 mg po/24 h) in the same way. Drugs were adjusted to weight in children. In case of severe mucositis voriconazole or fluconazole were temporarily used intravenously. EORTC/MSG criteria for IFI definitions were used and galactomanane antigen determinations were done twice a week. In case of neutropenic fever, empiric antifungal treatment was started in the fith day with lyposomal amfotericin B the primary prophylaxis was stopped.

RESULTS: AML GROUP: We didn’t find statistical differences among groups of VORI (n=31) and FLUCO (n=63) when we analyse age (VORI mean 49y, 18–64;FLUCO mean53y,21–72), sex, FAB classification, disease status, number of broad spectrum antibiotics, bacterial infections, days of neutropenia (19 vs 18) and days of fever (5.8 vs 7.2). The number of detected IFI was superior in the fluconazole group in spite of no statistical differences (VORI 3 vs FLUCO 11; 9,6% vs 17,4%, p=NS) with a sum of proven+probable 2 vs 5 (6,4% vs 8%, p=NS). Interestingly, the number of fatal IFI was statistical superior in the fluconazole group (VORI 0 vs FLUCO 4, p<0,05) and the need for empiric antifungal treatment was superior in the fluconazole group (VORI 19,3% vs FLUCO 50,8%, p=0,004). None patient had to interrupt voriconazole due to toxicity.

ALLO-HCT GROUP: we didn’t find statistical differences among groups of VORI (n=24) and FLUCO (n=31) when we analyse age (VORI mean 37y, 2–64 and FLUCO mean 30y, 5–56), sex, diagnosis and status at transplantation. In spite of differences in the conditioning regimen with more reduced-intensity conditioning regimen in the voriconazole group, both groups were similar when we analyse incidence of aGVHD II–IV grades, steroids use, VOD, number of bacterial infections, CMV infections and cGVHD. The days of fever were superior in the fluconazole group (VORI 4 vs FLUCO 7, p=0,04) and so the need for empiric antifungal treatment (VORI 6 vs FLUCO 16, p=0,042). The number of IFI was superior in the fluconazole group (VORI 2 vs FLUCO 10, p=0,04) and the number of fatal events due to IFI was superior in the fluconazole group (VORI 1 vs FLUCO 5, p<0,05). Voriconazole was temporarily stopped in two patients with hepatic aGVHD and one patient with VOD. Three patients developped hepatic colestasis by day +75 wich was reversible after discontinuating voriconazole one week.

CONCLUSSIONS: voriconazole as primary prophylaxis of IFI is a safety and well tolerated efective drug wich reduces the need for empiric antifungal treatment and the fatal events due to IFI in AML and allo-HCT.

Author notes

Corresponding author