Development of Interpretive Breakpoints for Antifungal Susceptibility Testing: Conceptual Framework and Analysis of In Vitro-In Vivo Correlation Data for Fluconazole, Itraconazole, and Candida Infections

The availability of reproducible antifungal susceptibility testing methods now permits analysis of data correlating susceptibility in vitro with outcome in vivo in order to define interpretive breakpoints. In this paper, we have examined the conceptual framework underlying interpretation of antimicrobial susceptibility testing results and then used these ideas to drive analysis of data packages developed by the respective manufacturers that correlate fluconazole and itraconazole MICs with outcome of candidal infections. Tentative fluconazole interpretive breakpoints for MICs determined by the National Committee for Clinical Laboratory Standards' M27-T broth macrodilution methodology are proposed: isolates for which MICs are < or = 8 microg/mL are susceptible to fluconazole, whereas those for which MICs are > or = 64 microg/mL appear resistant. Isolates for which the MIC of fluconazole is 16-32 microg/mL are considered susceptible dependent upon dose (S-DD), on the basis of data indicating clinical response when > 100 mg of fluconazole per day is given. These breakpoints do not, however, apply to Candida krusei, as it is considered inherently resistant to fluconazole. Tentative interpretive MIC breakpoints for itraconazole apply only to mucosal candidal infections and are as follows: susceptible, < or = 0.125 microg/mL; S-DD, 0.25-0.5 microg/mL; and resistant, > or = 1.0 microg/mL. These tentative breakpoints are now open for public commentary.