|Home | About | Journals | Submit | Contact Us | Français|
We surveyed 497 isolates of Aspergillus fumigatus collected from 2008 to 2009 as part of the ARTEMIS global surveillance study for elevated MIC values to itraconazole, voriconazole, and posaconazole. Sequencing of the cyp51A gene revealed that 8/29 isolates with elevated MIC values to one or more triazoles, all originating in China, contained the TR/L98H mutation associated with resistant European isolates of A. fumigatus. This is the first time the TR/L98H mutation has been identified outside Europe.
Invasive aspergillosis (IA) in immunocompromised patients results in high morbidity and mortality (12, 22). Although more than 30 species of Aspergillus have been implicated in infections in humans, IA is most often caused by the species Aspergillus fumigatus (12). Primary therapy has been largely limited to amphotericin B and the triazole compounds itraconazole and voriconazole (36). The use of amphotericin B has been limited by its toxic side effects, and at least one study showed that voriconazole was superior to amphotericin B. As a result, voriconazole is now first-line therapy for aspergillosis (19, 36).
Resistance to triazoles in A. fumigatus was first reported from U.S. clinical isolates collected in the 1980s but has been infrequently recognized in clinical practice (13). Recently, a number of clinical failures of triazoles against A. fumigatus have been reported (2, 5, 7, 8, 10, 13, 20, 21, 35) and isolates resistant to itraconazole have been isolated from environmental samples (27, 34). The primary mechanisms of triazole resistance in A. fumigatus are changes in the amino acid sequence of the cyp51A gene encoding 14-α-demethylase, a component of the ergosterol pathway and the target of the triazoles (15, 19, 20, 24, 25, 34). Specific mutations in cyp51A may result in resistance to one, two, or all three triazoles (24, 25, 32, 35). Other mechanisms that decrease the intracellular accumulation of triazole compounds, such as overexpression of the efflux pumps and decreased cellular permeability, may play a role in resistance as well (6, 13, 23).
Increases in cross-resistance of A. fumigatus isolates to the triazoles would be detrimental to the clinical treatment of aspergillosis, since they would mean the loss of the primary therapy and of the only available oral treatment. Therefore, an increased vigilance for these isolates is warranted. The ARTEMIS global antifungal susceptibility program was designed to detect trends in antifungal resistance through sentinel surveillance of more than 100 medical centers worldwide. In 2008 to 2009, an increase in A. fumigatus isolates with elevated MIC values to triazole compounds was detected (29). Here, we attempted to look for cyp51A mutations responsible for reduced azole susceptibility in the 2008 to 2009 ARTEMIS A. fumigatus isolates. A total of 497 A. fumigatus isolates submitted from 62 medical centers worldwide were collected. Isolates were confirmed as A. fumigatus sensu stricto using a species-specific microsphere-based Luminex assay (17). Broth microdilution susceptibility testing was performed as described previously according to the CLSI M38-A2 standard (9, 29, 31). Because breakpoints are not available, isolates for which itraconazole, voriconazole, or posaconazole MIC values were greater than the epidemiological cutoff values (ECV) of 1 μg/ml for itraconazole and voriconazole and 0.25 μg/ml for posaconazole (30, 32) and 12 isolates with MIC values below the epidemiological cutoff values were selected for further analysis.
In total, 29 isolates with MIC values above the established ECV for itraconazole, voriconazole, and/or posaconazole were identified (Table 1). Twenty-four of the isolates with elevated triazole MIC values came from different centers in Hangzhou, China, two originated in the Czech Republic, and one each were from Portugal, Brazil, and the United States (Table 1). The cyp51A gene, from nucleotide −462 of the promoter region through nucleotide 1696 in the 3′ untranslated region, was amplified and sequenced in both directions for all isolates with elevated MIC values and 12 randomly chosen controls using the oligonucleotide primers shown in Table 2. Sequences were compared to the reference sequences of isolate CM-237 (26) available in GenBank (accession number AF338659). As seen in Table 1, 8 of the 11 isolates for which itraconazole MIC values were ≥4 μg/ml had the L98H amino acid substitution in addition to the 34-base-pair nucleotide tandem repeat in their promoter region (TR/L98H) that has been previously associated with itraconazole resistance (pan-azole resistance) in European A. fumigatus isolates (25, 34). Interestingly, all of the isolates with TR/L98H also carried the pair of amino acid substitutions S297T and F495I. This pair of mutations may contribute to lowering the voriconazole MIC values in what would otherwise be pan-triazole-resistant isolates (25, 34).
All of the isolates for which itraconazole MIC values were ≤2 μg/ml or isolates which had elevations only in posaconazole MIC values had a wild-type cyp51A amino acid sequence or had the F47Y/M172V/N248T/D225E/E427K amino acid substitutions that have also been either detected here (data not shown) or shown previously to be present in fully susceptible isolates (5, 20). As far as we know, this is the first time that the TR/L98H mutation has been reported in A. fumigatus isolates collected in China and the first time it has been reported outside Europe. All of the isolates with elevated triazole MIC values were genotyped using a panel of eight previously described microsatellite markers (3, 14). Results showed the presence of 28 unique genotypes. The eight isolates with the TR/L98H mutation were distributed throughout the dendrogram, with only two sets of two isolates clustering as nearest neighbors (Fig. 1), ruling out clonal spread of a single isolate between patients within the institutions from which they were cultured.
There are no reported cases of patients with A. fumigatus isolates that developed the TR/L98H mutation under pressure of triazole therapy (2, 4, 7, 8, 10, 21). It has been proposed that the increase in frequency of this particular mutation in clinical isolates from the Netherlands is driven by the use of azole compounds as fungicides in agricultural practice (35). To date, clinical A. fumigatus isolates with this mutation have been reported in patients from the Netherlands, Spain, France, Norway, United Kingdom, and Belgium (20, 25, 27, 34). Environmental isolates of A. fumigatus with the TR/L98H mutation been recovered in the Netherlands and Denmark from agricultural samples of soil (27, 34). Because the TR/L98H mutation is not yet widespread outside Europe and the isolates from China reported here are genetically heterogenous, it is possible that the ARTEMIS isolates from China could stem from the environment due to agricultural practices at the source.
Not all of the isolates with elevated itraconazole MIC values had a cyp51A mutation. This is not unusual. In a 2008 to 2009 survey in England, Bueid and coworkers found that 43% of their isolates with itraconazole resistance had a wild-type cyp51A sequence (5) and Snelders and coworkers (34) found itraconazole-resistant isolates with wild-type cyp51A genes during environmental sampling. As others have found, there is probably more than one mechanism working in these isolates to overcome the pressure of triazole exposure, and efflux pumps likely play a role (11, 13, 23, 28, 33).
We did not detect any amino acid substitutions associated with resistance in the cyp51A genes of the organisms with an isolated elevation of the posaconazole MIC to 0.5 μg/ml. These isolates may just represent a small proportion at the high end of the wild-type distribution. Indeed, in a recent multicenter study using CLSI M38-A2 broth microdilution methods, Espinel-Ingroff et al. (16) reported that the ECV for posaconazole and A. fumigatus should be increased from 0.25 μg/ml to 0.5 μg/ml. Among the 8 isolates of A. fumigatus in the present study for which there was a cyp51A mutation, all had posaconazole MICs of ≥1 μg/ml.
Triazole resistance in A. fumigatus seems to be increasing in Europe, where it is most extensively studied. Two studies from the United Kingdom indicate that triazole resistance is increasing, and the cyp51A mutations identified indicate that it is due primarily to pressure under azole therapy (5, 20). In contrast, Snelders and coworkers in the Netherlands (34) have also noted an increase in triazole-resistant A. fumigatus, but they conclude that it is due primarily to isolates containing the TR/L98H mutation and that these are derived from environmental sources. However, not all of the current literature points toward a developing trend in A. fumigatus triazole resistance. In two other recent studies, one from France (1) and one from Spain (18), the authors were unable to detect any temporal increase in triazole resistance in their isolates. Likewise, data from the ARTEMIS global sentinel surveillance program demonstrated no consistent trend toward decreased susceptibility for any triazole and A. fumigatus over a 9-year period from 2001 through 2009 (29). Susceptibility testing of A. fumigatus is not common in clinical practice, but the reports of increasing resistance in Europe and in this study indicate that it is important to monitor for azole resistance in A. fumigatus. To help understand the full scope of this resistance, a surveillance network for azole-resistant A. fumigatus should be maintained.
The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention.
Published ahead of print on 20 June 2011.