Evaluation of the in vitro activity of voriconazole as predictive of in vivo outcome in a murine Aspergillus fumigatus infection model.

We have evaluated the in vitro activity of voriconazole against 61 strains of Aspergillus fumigatus by using broth microdilution, disk diffusion, and minimal fungicidal concentration procedures. We observed an excellent correlation between the results obtained with the three methods. Five percent of the strains showed MICs greater than or equal to the epidemiological cutoff value (ECV; 1 mu g/ml). To assess whether MICs were predictive of in vivo outcome, we tested the efficacy of voriconazole at 25 mg/kg of body weight daily in an immunosuppressed murine model of disseminated infection using 10 strains representing various patterns of susceptibility to the drug as determined by the in vitro study. Voriconazole prolonged survival and reduced fungal load in the kidneys and brain in those mice infected with strains with MICs of <= 0.25 mu g/ml, while in mice infected with strains with MICs of 0.5 to 2 mu g/ml, the efficacy was varied and strain dependent and in mice infected with the strain with a MIC of 4 mu g/ml, the antifungal did not show efficacy. Voriconazole reduced galactomannan antigenemia against practically all strains with a MIC of <4 mu g/ml. Our results demonstrate that some relationship exists between voriconazole MICs and in vivo efficacy; however, further studies testing additional strains are needed to better ascertain which MIC values can predict clinical outcome.