Etest Cannot Be Recommended for In Vitro Susceptibility Testing of Mucorales

TEXT

Mucormycoses are rapidly progressing and frequently lethal infections caused by fungi associated with the order Mucorales (1). Rhizopus is the most commonly found genus, followed by the genera Lichtheimia and Mucor (1). Antifungal treatment options consist of lipid formulations of amphotericin B (AMB) as the first-line therapy (2) and posaconazole (PSC) as salvage therapy (3, 4). The rising number of breakthrough mucormycoses (5–7) has stimulated interest in antifungal susceptibility testing (AST) of Mucorales. Phylogenetic studies found that Mucorales embrace a very heterogeneous group of fungi, with their evolutionary distances mirrored in various levels of intrinsic antifungal susceptibilities (8, 9). The recommended international standard methods of the CLSI (Clinical and Laboratory Standards Institute) (10), and EUCAST (European Committee for Antimicrobial Susceptibility Testing) (11) are based on broth microdilution assays. However, ready-to-use commercial tests, such as the Etest, are popular for routine application, as they are simpler to perform and more time efficient (12). To use and exchange data generated with the Etest in routine laboratories, an evaluation of this commercial test compared to a gold standard method is needed.

The aims of the present study were 2-fold: to determine amphotericin B and posaconazole susceptibility patterns for the most frequent agents of mucormycoses and to compare the results obtained with Etest and EUCAST.

The strain set comprised 131 Mucorales; the clinical strains were collected between 2008 and 2014 in routine laboratory testing at the University Hospital of Innsbruck (see Table S2 in the supplemental material), and an additional 38 reference strains were gained from the Fungal Biodiversity Centre (CBS, Utrecht, The Netherlands) (see Table S3). The strains were cultured on supplementary minimal medium (SUP) agar (13) at 37°C (except that Mucor spp. were grown at 30°C) for 3 to 5 days and preidentified to genus level by micromorphological (Olympus CX21 microscope; Olympus, USA) and macromorphological (Axioplan microscope; Zeiss, Germany) characteristics (6, 14). Species were determined by internal transcribed spacer (ITS) direct sequencing (15). Mycelium was harvested for genomic DNA extraction as previously described (16). ITS sequences were identified with BLAST comparative nucleic acid sequence analysis of sequences in the NCBI database (www.ncbi.nlm.nih.gov/BLAST).

In vitro antifungal susceptibilities against amphotericin B (Sigma-Aldrich, St. Louis, MO, USA) and posaconazole (Sigma-Aldrich, St. Louis, MO, USA) were tested according to the EUCAST guidelines (11) and by Etest (BioMeriéux, Inc., Craponne, France) according to the manufacturers' instructions. Conidia were harvested according to the method of Lackner et al. (17). Stock solutions of antifungals were prepared according to EUCAST guidelines (12) and stored at −80°C. The drug concentrations tested ranged from 0.03125 mg/liter to 16.00 mg/liter.

For the EUCAST method, the MICs of AMB and PSC were read with a magnifying mirror after 24 h at 37°C (Mucor spp. were grown at 30°C). Etest experiments were performed in parallel to the EUCAST tests (18) for three biological replicates. The MICs were read visually after 24 h and 48 h using four ATCC strains as quality controls (19). For each species and drug, the MIC range, MIC₉₀, and MIC₅₀ of each tested population were calculated according to the EUCAST method (20, 21). For strain sets comprising <9 isolates, MIC₅₀s and MIC₉₀s are not presented. High off-scale MIC results were converted to the next highest concentration, and low off-scale MIC results were left unchanged. Etest MICs were rounded up to the next higher EUCAST concentration for easy comparison. The results of Etest and EUCAST were analyzed for their reproducibility by providing essential agreement values within ±1 and ±2 EUCAST dilution steps; the latter is used for calculating essential agreement, as previously described (22).

Species identification is important for making treatment decisions, as susceptibility patterns differ widely among Mucorales (8). ITS sequencing was confirmed to be ideal for species identification of Mucorales, but it had limited discriminatory power for Lichtheimia corymbifera and Lichtheimia ramosa; these species were differentiated by morphology (data not shown). The characteristic values of in vitro susceptibility tests are given in Table S1 in the supplemental material. Species-specific population MICs gained with EUCAST and Etest for both amphotericin B and posaconazole are presented in the supplemental material (see Fig. S1 and S2, respectively; EUCAST results are shown by black bars and Etest results by gray bars); Mucor racemosus MICs are given in Table S2.

Amphotericin B is currently the first-line recommended drug (2), and posaconazole is recommended as salvage therapy (3, 4). Various studies found good in vitro activities for these two antifungal agents, supporting the observed clinical efficacy (23–25). In the current study, amphotericin B was the antifungal agent most active against Mucorales as a whole in vitro, exhibiting median MICs ranging from 0.5 to 2.0 mg/liter depending on the species tested. The degree of amphotericin B activity in vitro was clearly species dependent, especially for Rhizopus arrhizus, Rhizopus microsporus, and Rhizomucor pusillus (see Table S1 in the supplemental material). In contrast, low MIC values were observed for amphotericin B and Mucor species (see Table S1). The best in vitro activity was found for Mucor racemosus (MIC range, 0.0625 to 0.25 mg/liter) (see Table S1). Posaconazole, however, showed lower activities against most of the Mucorales species but had reasonable activity against Lichtheimia corymbifera (see Table S1). The median MICs of posaconazole were between 1.0 and 8.0 mg/liter. The poorest activity found for posaconazole was against members of the genus Mucor, i.e., Mucor circinelloides (MIC range, 0.5 to 32.0 mg/liter) and M. racemosus (MIC range, 0.125 to 32.0 mg/liter). Previous studies have also demonstrated higher MICs for posaconazole than for amphotericin B (with average MICs of 2.0 mg/liter) (23, 26). Also, M. circinelloides was shown to exhibit higher posaconazole MICs, a trend that can be reinforced by our data (see Table S1). Comparisons of CLSI and EUCAST methods against Etest were performed for filamentous fungi, including a limited number of Mucorales (25, 27, 28). In these previous studies, Mucorales were analyzed only as part of a merged strain set and not separately; therefore, good overall levels of agreement were obtained for filamentous fungi in general (25, 27, 28). In contrast, in the present study, which focuses specifically on Mucorales, low levels of agreement were found. The overall agreement of Etest did not reach the acceptable value of 90.0% (29) for Mucorales as a whole for either amphotericin B (73.0%) or posaconazole (77.0%). Lower levels of agreement were found for M. circinelloides and R. arrhizus (62.5 and 70.7%, respectively). Drogari-Apiranthitou et al. (25) found an agreement level of 100.0% for posaconazole, while the present study found agreement levels of only 62.5 to 85.4%, depending on the species tested (Table 1). The poor reproducibility of data might be partially generated by Etest reading difficulties (see Fig. S3), as hyphae frequently overgrow the Etest strip, particularly when testing amphotericin B against L. corymbifera, as seen by the results shown in Fig. S3B and D; these difficulties were previously reported by Maurer et al. (30).

In conclusion, low levels of agreement (<90.0%) between Etest and EUCAST were found for Mucorales. Therefore, based on our data, Etest cannot be recommended for susceptibility testing. Furthermore, amphotericin B was confirmed to be the most active agent against Mucorales in vitro, while posaconazole showed more restricted activities.

Supplementary Material