Multicentre validation of a EUCAST method for the antifungal susceptibility testing of microconidia-forming dermatophytes

Abstract

Objectives Terbinafine resistance is increasingly reported in Trichophyton, rendering susceptibility testing particularly important in non-responding cases. We performed a multicentre evaluation of six EUCAST-based methods. Methods Ten laboratories susceptibility tested terbinafine, itraconazole, voriconazole and amorolfine against a blinded panel of 38 terbinafine WT and target gene mutant isolates. E.Def 9.3.1 modifications included: medium with/without addition of chloramphenicol and cycloheximide (CC), incubation at 25°C to 28°C for 5–7 days and three MIC endpoints [visually and spectrophotometrically (90%/50% inhibition)], generating 7829 MICs. Quality control (QC) strains were Aspergillus flavus ATCC 204304 and CNM-CM1813. Eyeball, ECOFFinder (where ECOFF stands for epidemiological cut-off) and derivatization WT upper limits (WT-ULs), very major errors (VMEs; mutants with MICs ≤WT-ULs) and major errors (MEs; WT isolates with MICs >WT-ULs) were determined. Results MICs fell within the QC ranges for ATCC 204304/CNM-CM1813 for 100%/96% (voriconazole) and 84%/84% (itraconazole), respectively. Terbinafine MICs fell within 0.25–1 mg/L for 96%/92%, suggesting high reproducibility. Across the six methods, the number of terbinafine MEs varied from 2 to 4 (2.6%–5.2%) for Trichophyton rubrum and from 0 to 2 (0%–2.0%) for Trichophyton interdigitale. Modes for WT and mutant populations were at least seven 2-fold dilutions apart in all cases. Excluding one I121M/V237I T. rubrum mutant and two mixed WT/mutant T. interdigitale specimens, the numbers of VMEs were as follows: T. rubrum: CC visual, 1/67 (1.5%); CC spectrophotometric 90% inhibition, 3/59 (5.1%); and CC spectrophotometric 50% inhibition, 1/67 (1.5%); and T. interdigitale: none. Voriconazole and amorolfine MICs were quite uniform, but trailing growth complicated determination of itraconazole visual and spectrophotometric 90% inhibition MIC. Conclusions Although none of the laboratories was experienced in dermatophyte testing, error rates were low. We recommend the CC spectrophotometric 50% inhibition method and provide QC ranges and WT-ULs for WT/non-WT classification.