Comparison of CLSI and EUCAST broth microdilution reference methods for antifungal susceptibility testing of Candida (Candidozyma) auris using a representative international panel of isolates: a single-c
Maria Siopi, Sevasti Leventaki, Ioannis Pachoulis, Bram Spruijtenburg, Eelco F J Meijer, Jacques F Meis, Spyros Pournaras, Athanasios Tsakris, Georgia Vrioni, Joseph MeletiadisAbstract
Objectives
The two reference broth microdilution antifungal susceptibility methods, developed by the CLSI and EUCAST, were compared qualitatively and quantitatively using an international collection of 107 Candida (Candidozyma) auris clinical isolates including azole- and echinocandin-resistant isolates.
Methods
Inter-method agreement within ±1 twofold dilution was assessed for all drugs. Categorical agreement (CA) was evaluated for amphotericin B, anidulafungin and micafungin using the CDC tentative resistance breakpoints and the recently proposed EUCAST clinical breakpoints for C. auris.
Results
CLSI-EUCAST agreement was moderate to strong for itraconazole (44%), posaconazole (52%), isavuconazole (70%), voriconazole (72%), micafungin (81%), caspofungin (86%), and anidulafungin (87%), and excellent for 5-flucytosine (93%) and amphotericin B (100%). All isolates were categorized as non-resistant to amphotericin B by both methods (100% CA). CA was 92% for anidulafungin and 94% for micafungin, with 8% and 6% major errors, respectively, with CDC breakpoints classifying 33%–44% of FKS1 mutants as non-resistant. While all FKS1 mutants exceeded EUCAST ECOFF values, 11% and 28% fell at or below the corresponding CLSI ECV values for micafungin (0.5 mg/L) and anidulafungin (1 mg/L), respectively, with MICs 0.5–1 mg/L. For caspofungin, all non-wild-type isolates harboured FKS1 mutations, but 6% of isolates without FKS1 mutations had MICs 1 dilution higher than CLSI ECV for caspofungin (0.5 mg/L).
Conclusions
Despite high overall concordance between CLSI and EUCAST MICs, discrepancies in wild-type classification among FKS1 mutants highlight the need for harmonized interpretive criteria. Azole MIC variability further underscores the need for standardized breakpoints.