Antifungal prophylaxis following allogeneic haematopoietic stem cell transplantation: what is the role of voriconazole?

Abstract

Voriconazole is given a provisional A1 rating in the European Conference on Infections in Leukaemia (ECIL)3 guidelines as primary prophylaxis of invasive fungal infections (IFIs) in haematopoietic stem cell transplantation (HSCT) recipients. The rationale for this rating is outlined in a recent review by Hicheri et al. [1]. Interestingly, the main argument supporting the new role of voriconazole in primary prophylaxis is based on the results of one recently published study [2]. In this study, Marks and colleagues demonstrated that voriconazole was superior to itraconazole as antifungal prophylaxis after allogeneic HSCT based on differences in the primary composite endpoint [115/234 (49.1 %) vs. 88/256 (34.5 %), respectively] [2]. However, three considerations should be taken into account when interpreting these results. First, the composite endpoint of the study included not only survival at day 180 post-transplant as well as proven or probable IFI during that same period, but also discontinuation of the study drug for [14 days during the 100-day prophylactic period. No differences were found between the treatment arms in terms of general survival and occurrence of IFI. The only part of the composite endpoint where voriconazole proved to be superior to itraconazole was discontinuation of the study drug—and the results and conclusions were driven by that difference only. In this context it is of interest that even in the voriconazole arm the number of patients who reached sufficient duration of prophylaxis was low (54.3 %; 127/234), but still significantly higher than that in the itraconazole arm (40 %; 102/255). Interestingly, both antifungals were similar for side effect scores. When taking into account both (1) the open label design of the study (i.e., investigators who discontinued antifungals were not blinded to the study drugs) and (2) the fact that discontinuation of study drugs was completely up to the discretion of the treating physician and no reasons for discontinuation were given in the Results section of the paper, we suggest that the conclusion drawn by the authors regarding the superiority of voriconazole compared to itraconazole can, based on these results, not be drawn. Second, the incidence of IFI was very low—1.3 % (3/ 234) in the voriconazole arm and 2.1 % (5/255) in the itraconazole arm. All but two cases of IFIs did not meet the criteria of breakthrough infection as they occurred[40 days after discontinuation of the study drug. In a previous comparable study [3], occurrence of breakthrough IFI in patients receiving itraconazole prophylaxis was consistently [6 % which is threefold higher than that reported by Marks and colleagues. One reason for the low incidence may have been that galactomannan monitoring, which has shown to have a significant impact on the prevalence of invasive mould infection (IMI), was not part of the study design [4]. It may be argued that with a higher IFI incidence significant differences might have been found between voriconazole and itraconazole with respect to the rates of IMI. Third, therapeutic drug monitoring was performed only in a subset of the study cohort. Not surprisingly, the standard oral 200 mg twice-daily voriconazole dosage used in the study led to insufficient trough concentrations in the vast majority of patients; 62 % even had trough levels of \1.0 mg/L. Systematic therapeutic drug monitoring may M. Hoenigl (&) K. Seeber R. Krause Section of Infectious Diseases and Tropical Medicine, Medical University of Graz, Graz, Austria e-mail: martin.hoenigl@medunigraz.at