Voriconazole as primary antifungal prophylaxis in patients with neutropenia after hematopoietic stem cell transplantation or chemotherapy for acute myeloid leukemia

To the Editor: Invasive fungal infections (IFI) remain a major cause of illness and death in patients with prolonged and severe neutropenia (PSN) following treatment for hematological malignancies and allogeneic hematopoietic stem cell transplantation (HSCT). Treatment success is higher in early stages of the disease making some clinicians to consider antifungal prophylaxis as an adequate strategy to diminishing the IFI morbidity–mortality in patients under PSN (1, 2). Early evidence-based guidelines recommended fluconazole for the prevention of IFI after HSCT (3). By contrast, initial trials did not show a marked advantage of antifungal prophylaxis in acute myeloid leukemia (AML) patients undergoing chemotherapy, although there was a tendency toward a reduced morbidity and mortality with the use of fluconazole or itraconazole (4). More recently, oral posaconazole is recommended in HSCT recipients with graft-versus-host disease as well as in patients with AML or myelodysplastic syndrome at high risk for invasive aspergillosis (5). Voriconazole, a new generation azole with both oral and intravenous formulations available, has been successfully employed as treatment of IFI. For these reasons, voriconazole is also used as antifungal prophylaxis in patients with PSN (6). We conducted a prospective study to determine the efficacy, and safety of voriconazole for the primary prevention of IFI in 127 high risk patients with 181 evaluate episodes of PSN after chemotherapy for AML or HSCT between June 2004 and July 2007. PSN was defined as an absolute neutrophil count of <500 neutrophils/mm3 with an expected duration of at least 7 d. Antifungal prophylaxis consisted of voriconazole 400 mg in oral or i.v. formulation (200 mg tablet, b.i.d.; 8 mg/kg for patients weighing <50 kg). Diagnostic procedures included twice weekly serum galactomannan levels. When IFI was suspected, a lung or sinus high-resolution CT was performed. Cultures of bronchoalveolar lavage fluid were performed if available. Neutropenic fever was managed according to Infectious Diseases Society of America (IDSA) guidelines. IFI was defined as stated by the Mycoses Study Group/European Organization for Research and Treatment of Cancer (MSG/EORTC). Baseline demographic and clinical characteristics are presented in Table 1. Median duration of prophylaxis with voriconazole was 24 d (range 11–70 d). Antifungal prophylaxis was considered successful when patients completed prophylaxis showing no evidence of proven, probable, or suspected IFI, and empirical or pre-emptive therapy was not used. The overall treatment success rate was 72.9%. Thirty-one of 181 episodes (17.1%) failed voriconazole prophylaxis because of the use of empirical antifungal therapy [with caspofungin or Liposomal amphotericin B (L-AmB)]. Seven proven or probable IFIs were documented (3.9%). Five corresponded to invasive aspergillus pneumonia and four resolved after combined therapy with caspofungin and voriconazole (n = 2) or caspofungin and L-AmB (n = 2). The remaining patient died because of graft failure. Additionally, we documented two cases of breakthrough opportunistic fungal infections: a surgically controlled nasal zygomycosis and another case of systemic fusariosis unsuccessfully treated with L-AmB. Ten (7.8%) of 127 voriconazole-treated patients died during the study period but only one patient died because of IFI (0.7%). Overall survival at 28 d from the end of antifungal prophylaxis was high, 92.8 ± 2.1%. Fourteen patients discontinued voriconazole prophylaxis because of possible–probable drug-related adverse events. Interruptions were mainly because of mild to moderate liver disturbances resolving within a range of 1–7 d after drug discontinuation. Furthermore, treatment interruptions occurred in most instances (all but one) in patients undergoing HSCT and only in one pediatric patient. Two patients needed definitive withdrawal because of visual hallucinations. This study showed that voriconazole effectively prevented IFI during PSN after chemotherapy for AML and HSCT. Because of this heterogeneous population, favorable results would need to be confirmed in a specific group of allogeneic HSCT patients. A drawback of this prospective study was non-randomized design but results are in agreement with those reported by two prior non-randomized retrospective studies (7, 8), which showed low rates of IFI and, most impressively, no cases of aspergillosis among allogeneic HSCT recipients. In contrast, we still documented some cases of invasive aspergillosis. Monitoring of drug levels during voriconazole prophylaxis is highly recommended to rule out that breakthrough IFI were caused by inappropriate low levels. The largest prospective, randomized, multicentric, controlled, double-blind trial of voriconazole vs. fluconazole reported (9) to date did not find differences in fungal-free survival rates in standard risk allogeneic HSCT recipients. Remarkably, the incidence of IFI caused by aspergillosis was significantly higher in the fluconazole group versus voriconazole (P = 0.04). As conclusion, voriconazole appears to be well tolerated with manageable side effects and beneficial in the prophylaxis of IFI in high risk patients. It might be an effective alternative as antifungal prophylaxis in patients intolerant to posaconazole oral suspension.