|Home | About | Journals | Submit | Contact Us | Français|
The Clinical and Laboratory Standards Institute (CLSI) M38-A2 reference broth microdilution (BMD) method for the antifungal susceptibility testing of filamentous fungi now includes guidelines for testing echinocandin activity using the minimum effective concentration (MEC) as the endpoint measurement. In this study, we compared the caspofungin Etest MIC on RPMI agar and Mueller-Hinton agar (supplemented with glucose and methylene blue [MGM]) to the BMD MEC for 345 clinical Aspergillus isolates, including A. flavus, A. fumigatus, A. nidulans, A. niger, and A. terreus. The essential agreement (±1 log2 dilution) of the Etest on MGM and RPMI agar with the reference BMD MEC was 18 and 26%, respectively. The geometric mean values for BMD MEC and MGM Etest were 0.137 and 0.024 μg/ml, respectively, and the geometric mean values for BMD and RPMI agar were 0.128 and 0.031 μg/ml, respectively. Comparatively, 91% of paired MGM and RPMI Etest results were within 2 log2 dilutions of each other and consistently produced clearly defined endpoints. In conclusion, the caspofungin Etest MIC, like the BMD MEC, is a reproducible endpoint but is markedly lower than the reference BMD. In anticipation of susceptibility breakpoint assignments, optimization studies will be required to improve the concordance of these two assays so that the potential for underreporting echinocandin resistance in Aspergillus is mitigated.
A reference method for the antifungal susceptibility testing of filamentous fungi using broth microdilution (BMD) has been developed by the Clinical and Laboratory Standards Institute (CLSI) and has been updated recently with guidelines for testing echinocandins (4). Unlike other antifungal classes, the approved endpoint for measuring echinocandin activity against molds is the minimum effective concentration (MEC), as opposed to the conventional MIC, which demonstrates more consistent and reproducible susceptibility data (4). Initially described by Kurtz et al. (10), the MEC is the lowest concentration of drug that leads to the growth of small, rounded, compact hyphal forms compared to the hyphal growth visualized in the growth control well of a BMD panel.
The echinocandins are an important class of antifungal agents for the empirical treatment of invasive aspergillosis. The BMD reference method is very cumbersome, which prevents many clinical laboratories from performing the test. The Etest (AB bioMerieux, Solna, Sweden) is an agar-based gradient endpoint susceptibility test that is much simpler to perform than BMD. Etest strips for the susceptibility testing of Candida to amphotericin B, fluconazole, voriconazole, posaconazole, and caspofungin (2, 5, 11, 14-17, 19), and Aspergillus to amphotericin B, voriconazole, and posaconazole (9, 12, 18), have demonstrated good correlation to the BMD. However, little has been published to verify the utility of echinocandin Etests for Aspergillus susceptibility testing (6).
The objective of this study was to determine if the caspofungin Etest could accurately and reliably measure the MEC of a large collection of clinical Aspergillus isolates.
Caspofungin activity against Aspergillus isolates was tested by BMD (CLSI M38-A2) for MEC (visual and microscopic) endpoint determination (4) and by Etest for MICs using both an RPMI agar and a modified Mueller-Hinton agar (MHA) medium formulation. The study aim was to determine the ability of the caspofungin Etest to measure the MEC of Aspergillus.
Clinical isolates of Aspergillus collected and stored at −70°C for the past 5 years were selected for study. Isolates consisted of 105 A. fumigatus, 62 A. flavus, 104 A. niger, 50 A. nidulans, and 24 A. terreus isolates. For reproducibility experiments, ATCC strains A. flavus 204304, A. fumigatus 204305, and A. niger 16404, as well as clinical reference isolates A. flavus 4174, A. fumigatus 4073, and A. niger 5009, each were tested on 10 consecutive days by BMD and Etest.
BMD was performed according to the CLSI M38 guideline for filamentous fungi (4). In brief, conidial suspensions were harvested from mature cultures (5 to 7 days of incubation). Suspensions of ~104 CFU/ml were made (using a colorimeter) and used to inoculate BMD panels containing RPMI-1640 liquid medium (with glutamine and phenol red, buffered with morpholinepropanesulfonic acid [MOPS], pH 7.0, no bicarbonate; Dalynn Biologicals, Calgary, Canada) and caspofungin (reference powder provided by Merck Research Laboratories, Rahway, NJ) ranging in concentration from 0.015 to 32 μg/ml. BMD panels were incubated for 24 h in ambient air at 35°C. Growth and sterility control wells also were set up on each BMD panel. The visual (macroscopic) MEC was measured for all strains as the lowest concentration of caspofungin that led to the growth of small, rounded, compact hyphal forms compared to the hyphal growth visualized in the growth control well. The microscopic MEC was determined for a subset of test isolates and defined as the lowest concentration of caspofungin associated with aberrant, short, hyphal segments compared to the growth control (long, unbranched hyphae). The Etest was performed using a modification of the CLSI M44-A disk diffusion method (3) on MHA supplemented with 2% glucose and 0.5 μg/ml methylene blue (MGM) and on RPMI-1640 agar (with glutamine and phenol red and 2% glucose, buffered with MOPS, pH 7.0, no bicarbonate; Dalynn Biologicals, Calgary, Canada). MGM plates were prepared by flooding MHA plates with 1 ml of glucose-methylene blue solution and were allowed to dry at room temperature for 3 days prior to inoculation. MGM and RPMI plates were inoculated with the same conidial suspension as that prepared for BMD testing and were incubated for 24 h in ambient air at 35°C per the manufacturer's instructions (AB bioMerieux, Solna, Sweden). Etest MICs were recorded as the lowest concentration of caspofungin for which the elliptical zone of growth inhibition intersected with the Etest strip. MICs were rounded up to the next even log2 concentration for the evaluation. Any microcolony growth within a clear zone of growth inhibition was disregarded. Etest values of ±1 log2 dilution of the reference BMD MEC were considered to be in essential agreement (EA).
We evaluated the ability of the caspofungin Etest to measure the MEC against a collection of Aspergillus species. A total of 345 clinical isolates of Aspergillus were tested by BMD. Etest was performed in two phases based on the MGM and RPMI agar formulations being evaluated. Of the 273 isolates tested on MGM, 73 isolates were unavailable for further testing, so additional isolates were selected to supplement the RPMI agar study arm (Table (Table1).1). Much of this work was completed prior to the publication of a standardized definition of visual and microscopic MEC endpoints by the CLSI (4). However, the CLSI endpoint definitions were adopted primarily from the formative findings of previous works (1, 6, 13), which were used for this study. In our hands, the microscopic MEC for 119 Aspergillus isolates showed 83% EA to the visual MEC (±1 log2 dilution); similar results have been reported previously (6). With the exception of two isolates, all of the 17 discordant MEC pairs had a lower microscopic MEC, with 12 of 17 values being out by 2 log2 dilutions relative to the visual MEC.
Table Table22 shows the distribution of BMD visual MEC and Etest MICs, and the accompanying EA calculations are shown in Table Table3.3. Regardless of the Aspergillus species or agar medium tested, the caspofungin Etest consistently produced a markedly lower growth inhibition endpoint (MIC). For the MGM study arm, the overall geometric mean MECs and MICs were 0.137 and 0.024 μg/ml, respectively. Similarly, the MEC and MIC geometric mean values for the RPMI agar study arm were 0.128 and 0.031 μg/ml, respectively. The overall EA, where ≥90% was considered acceptable for comparisons of the two MIC methods, for the MGM and RPMI Etest methods were 18 and 26%, respectively (Table (Table3).3). None of the Aspergillus species comparisons yielded agreement greater than 40%, except for A. terreus on RPMI agar (72%), which included only 18 isolates. More than 50% of the MGM Etest MICs and 38% of RPMI Etest MICs were ≥3 log2 dilutions lower than the visual MEC (Table (Table4).4). The exact agreement between Etest methods for the 200 paired isolates tested was 32%, while agreement within 1 log2 dilution was 76.5% and agreement within 2 log2 dilutions was 91%. Overall, Etest endpoints were clearly defined with few exceptions; some strains of A. flavus and A. terreus tended to yield narrower growth inhibition zones with microcolony growth (these results were disregarded).
Consistently with these findings, the visual MECs and MICs for the ATCC and clinical reference isolates were reproducible and within control levels (within 2 log2 dilutions) for the 10 consecutive days tested. Etest MICs also were repeatedly lower than concurrent visual MEC values for both MGM and RPMI agar test arms (Table (Table55).
There are no reference guidelines available for antifungal agar-based susceptibility testing against molds; medium selections for this study were based on the CLSI guidelines for the disk diffusion testing of azoles against Candida spp. (MGM) (3) and manufacturer guidelines for the caspofungin Etest (RPMI agar). However, developmental disk diffusion studies have characterized test conditions, similar to those used in this study, that yield clear and reproducible growth inhibitory zones for Aspergillus and other molds that correlate nicely with BMD results (7, 8). Contrary to our findings, these studies have suggested that MGM is not ideal for supporting Aspergillus growth after 24 h of incubation, and that unsupplemented Mueller-Hinton agar is optimal. Although we cannot rule out Mueller-Hinton formulations as a source of the discordance observed between our Etest and BMD results, the superimposable RPMI Etest data presented here provide support against this argument. Notwithstanding these findings, MIC endpoints for the ATCC and clinical reference isolates remained unchanged at 48 h of incubation on both agar media compared to the 24-h endpoints for the 10 consecutive days tested (data not shown).
In this study, we were able to show that RPMI basal medium and 24 h of incubation was sufficient for the visual interpretation of the caspofungin MEC against Aspergillus species without the need for cumbersome microscopy, effectively strengthening recent CLSI recommendations (4). However, BMD for filamentous fungi is not an ideal test format for many clinical laboratories, and the simplicity of the Etest may make it a more viable option. Although the results from this study demonstrate a clear discordance between the absolute values generated by the caspofungin BMD and Etest, it is difficult to dismiss the consistent performance of the Etest. An expanded optimization of test conditions hopefully will improve Etest concordance with the BMD MEC without compromising consistency. Including additional isolates of Aspergillus with elevated MECs (>1 μg/ml) also will be important in future studies.
In summary, the caspofungin Etest endpoint does not correlate with the BMD MEC of Aspergillus. However, the MEC endpoint for echinocandins was approved based on its reproducibility, and a correlation of MEC with clinical outcome remains to be established.
This study was supported in part by a research grant from Merck Frosst.
Published ahead of print on 9 December 2009.