Evaluation of the MIC test strips for antifungal susceptibility testing of Candidozyma auris (Candida auris) using a representative international collection of isolates

Abstract

We compared MIC test strips (MTS) with the reference Clinical and Laboratory Standards Institute (CLSI) broth microdilution method using an international panel of 100 Candidozyma auris (Candida auris) isolates belonging to different clades. The agreement (±1 twofold dilution) between the methods and the categorical agreement (CA) based on the Centers for Disease Control and Prevention's (CDC's) tentative resistance breakpoints and MTS-specific wild-type upper limit values (WT-ULVs) were determined. The MTS-CLSI agreement was poor to weak for posaconazole (3%), itraconazole (20%), voriconazole (31%), and 5-flucytosine (37%), and moderate to strong for isavuconazole (58%), anidulafungin (68%), caspofungin (72%), micafungin (77%), and amphotericin B (85%). Most fluconazole MICs were off-scale, precluding a corresponding estimation. Significant interpretation discrepancies were recorded using the CDC's breakpoints for amphotericin B (66% CA, 34% major errors; MaEs), but not for fluconazole (98% CA, 1% MaEs, 1% very major errors; VmEs), anidulafungin (97% CA, 3% MaEs, 0% VmEs), micafungin (99% CA, 1% MaEs, 0% VmEs), and caspofungin (95% CA, 5% MaEs, 0% VmEs). Discrepancies were minimized using the amphotericin B method-specific WT-ULV of 4 mg/L (98% CA, 2% MaEs). The MTS-specific WT-ULVs of echinocandins could help to detect 100% of FKS1 mutants. MTS generated higher MICs than the CLSI for azoles and 5-flucytosine. MTS could accurately detect fluconazole and echinocandin resistance among C. auris isolates. Nevertheless, it overestimated amphotericin B resistance as per the CDC's breakpoint of 2 mg/L. This can be improved by using the MTS-specific WT-ULV of 4 mg/L.IMPORTANCECandidozyma auris (Candida auris) may exhibit resistance to multiple and sometimes even all currently available classes of antifungals. Hence, antifungal susceptibility testing (AFST) is of key importance to guide the clinician in therapeutic decision-making and to detect novel patterns of resistance. Gradient diffusion strips, referred to both Etest and MIC test strip (MTS), are broadly used in laboratory routine for AFST of yeasts. We therefore compared MTS with the reference Clinical and Laboratory Standards Institute (CLSI) broth microdilution method using an international panel of 100 C. auris isolates belonging to different clades. Significant interpretation discrepancies were recorded for amphotericin B (66% categorical agreement, 34% major errors), which could be minimized using the amphotericin B method-specific wild-type upper limit value of 4 mg/L. MTS generated higher MICs than the CLSI for azoles and 5-flucytosine. MTS could accurately detect fluconazole and echinocandin resistance.

Keywords: Candida auris; MIC test strip; antifungal susceptibility testing; resistance; wild type upper limit values.

Fig 1

CLSI MIC distributions of C. auris isolates by clade. The broken and dotted lines indicate the CDC’s tentative resistance breakpoints and CLSI ECVs for C. auris (where available), respectively (19, 21).

Fig 2

MTS MIC distributions of C. auris isolates by clade. The broken lines indicate the CDC’s tentative resistance breakpoints for C. auris (where available) (19).

Fig 3

Representative photos for FKS1 WT C. auris isolates with increasing MTS MICs despite similar CLSI MICs. Note that the largest difference between MTS and CLSI was found for isolate AUH2751, which had the lowest OD_530nm at 24 h, indicating poor growth in liquid media.

Fig 4

Scatter plots of CLSI MICs versus MTS MICs. Numbers represent the number of C. auris isolates (total n = 100) at each MIC pair. The black broken lines indicate the CDC’s tentative resistance breakpoints for C. auris (where available) (19). The green shaded areas represent categorical agreement, while the pink and red areas indicate major error (MaE) and very major error (VmE), respectively.