Use of a serum-based antifungal susceptibility assay to predict the in vivo efficacy of novel echinocandin compounds.

In vitro susceptibility assays of antifungal activity do not always accurately predict in vivo efficacy. As well as having a clear clinical importance, the ability to predict efficacy is also essential for effective screening of novel drug compounds. Initial screening of novel compounds must often be based on in vitro data. The present report describes the use of serum-MIC, an in vitro test of antifungal susceptibility, to accurately predict in vivo efficacy of echinocandin drugs in a mouse model of disseminated candidiasis. The basis of the serum-MIC method was to measure the inhibitory activity of a test compound against Candida albicans hyphal growth in the presence of pooled mouse serum. For 13 previously uncharacterized echinocandin compounds, as well as for the known echinocandin drugs, micafungin and caspofungin, serum-MIC determinations were shown to give better correlation to efficacy in the animal model than conventional, CLSI standard, in vitro antifungal susceptibility tests. The most accurate prediction of efficacy was obtained when the serum-MIC was adjusted in relation to the serum concentration at 30 min post-treatment. Furthermore, when the efficacy of micafungin was determined by measuring C. albicans kidney burden in the mouse model of infection, the adjusted serum-MIC consistently reflected the effective serum concentrations. Our data indicate that determination of serum-MIC values will facilitate prediction of the in vivo potency of new antifungal compounds such as novel echinocandins.