ABSTRACT
Aspergillus fumigatus azole resistance has emerged as a global health problem. We evaluated the in vitro antifungal susceptibility of 221 clinical A. fumigatus isolates according to CLSI guidelines. Sixty-one isolates exhibiting MICs at the epidemiological cutoff value (ECV) for itraconazole or above the ECV for any triazole were checked for CYP51A mutations. No mutations were documented, even for the isolates (1.8%) with high voriconazole MICs, indicating that triazoles may be used safely to treat aspergillosis in Brazil.
TEXT
Aspergillus fumigatus is the leading cause of invasive aspergillosis (IA) worldwide, an opportunistic fungal infection documented mostly among patients with hematological malignant diseases, those with primary immunodeficiencies, and those treated with high doses of steroids. Depending on the severity of the disease and underlying conditions, crude mortality rates of IA may be as high as 27% to 80% (1–6).
Voriconazole is used for first-line therapy of IA, as recommended by the Infectious Diseases Society of America (1). Despite being highly effective against Aspergillus spp., resistance to azoles is becoming an emerging health problem in some parts of the world (7, 8). Two scenarios related to the emergence of azole resistance are the increasing number of non-fumigatus Aspergillus species, such as A. lentulus and A. calidoustus, with primary resistance to triazoles, thus causing human infections (9–11); and the development of secondary resistance in A. fumigatus isolates after environmental or human exposure to triazoles (12–16). Indeed, acquisition of resistant A. fumigatus isolates selected by environmental exposition to triazole fungicides in agriculture was recently reported in 1.7% to 30% of all clinical isolates cultured in medical centers in The Netherlands (16–20).
In regard to Latin America, a study conducted in Cundinamarca, Colombia, found that 20 single colonies of A. fumigatus out of 86 environmental samples tested exhibited CYP51A mutations. Sixty percent of Colombia's flower production is cultivated in Cundinamarca, and the use of triazole fungicides in agriculture was related to this high triazole resistance rate (21). There is a single Latin American report of an A. fumigatus clinical isolate resistant to itraconazole linked to a G54E substitution in CYP51A that was cultured from an Argentinian rural worker with a fungal corneal ulcer. Because this patient had never been exposed to azoles before, it was concluded that an environmental azole-resistant isolate caused the infection (22).
Considering the lack of data on A. fumigatus triazole resistance in Latin America, especially in Brazil, we conducted this retrospective study. A total of 221 A. fumigatus isolates were sequentially obtained from 221 unique patients with possible, probable, or proven aspergillosis admitted to six medical centers in Brazil (four cities) between 1998 and 2017. The isolates were identified as potential true pathogens by clinicians at each center following the criteria suggested by the European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group, National Institute of Allergy and Infectious Diseases Mycoses Study Group (EORTC/MSG) (23). The clinical specimens included sputum (n = 73; 33%), tracheal aspirates (n = 24; 10.9%), bronchoalveolar lavage (n = 39; 17.6%), lung and nasal biopsy samples (n = 18; 8.1%), and others (n = 67; 30.3%). All isolates were identified by calmodulin, β-tubulin genes, and ribosomal DNA (rDNA) internal transcribed spacer region sequencing (9, 24). Antifungal susceptibility tests were performed in duplicate according to the Clinical and Laboratory Standards Institute (CLSI) M38-A2 protocol for itraconazole (ITC), voriconazole (VRC), and posaconazole (PSC) (25). The CYP51A gene plus the promoter were sequenced for all isolates exhibiting MIC values higher than the epidemiological cutoff values (ECV) for any triazole or equal to the ECV for itraconazole (26). To elucidate the mechanisms responsible for azole resistance in A. fumigatus, we compared the generated sequences with those of the wild-type A. fumigatus reference isolate (Af293) (27).
Antifungal susceptibility results for the 221 isolates tested against ITC, VRC, and PSC are shown in Table 1.
In vitro antifungal susceptibility profile of 221 A. fumigatus isolates tested against 3 triazoles by the CLSI microbroth method
In general, 59 isolates (26.7%) showed ITC MIC values equal to the ECV of this antifungal drug (1 μg/ml). In addition, only 4 isolates (1.8%) exhibited VRC MIC values > 1 μg/ml, and all isolates presented PSC MIC values ≤ 0.5 µg/ml. Based on their MIC values, 61 clinical isolates were selected to check for molecular mechanisms of resistance, and no CYP51A gene mutation predictive of triazole resistance was found.
The Artemis global antifungal susceptibility program evaluated 497 A. fumigatus isolates from 62 medical centers worldwide. In this series, not a single A. fumigatus-resistant isolate from Latin America was confirmed (28). A recent multicenter international surveillance study of 3,788 A. fumigatus isolates from patients from 19 countries found a total rate of 3.2% triazole resistance, with large variation between countries. Rates of triazole resistance in Europe ranged from 2.1% in Sweden to 28% in the United Kingdom. Only 64 isolates (0.17%; from 57 patients) were collected from Latin American medical centers, and no resistant isolates were detected. Note that in this particular multicenter study, triazole resistance was documented mostly in medical centers where >100 isolates had been evaluated (3). This finding suggests that A. fumigatus triazole resistance in Latin America and Brazil may be underestimated because no center tested >100 clinical isolates (3, 28).
The present study represents the largest series of A. fumigatus clinical isolates from reference medical centers in Brazil that were tested for triazole resistance by using a reference broth microdilution test and molecular characterization of the CYP51A gene.
Currently, in Brazil, there is no fungal infection in the list of officially reportable diseases, but the incidence of chronic pulmonary aspergillosis (CPA) was estimated to be as high as 6.2/100,000 inhabitants. However, fungal diseases are neglected by both the public and private health systems, and only a few cases are diagnosed and treated (29). The lack of CPA diagnosis and the consequent absence of treatment for these patients may help to explain the lack of resistant isolates in this study.
In regard to resistance related to agricultural exposure to azole fungicides, an important aspect to be considered is that the vast majority (80.1%) of isolates tested in the present study were cultured in medical centers from large urban areas, such as São Paulo, where environmental exposure of A. fumigatus to triazole fungicides is less likely to occur. In this scenario, most of the A. fumigatus isolates we tested likely had no previous exposure to environmental triazoles before infecting our patients.
In conclusion, we did not observe triazole resistance mechanisms among 221 clinical A. fumigatus isolates, including the 4 isolates (1.8%) with VRC MICs of 2 μg/ml, collected from Brazilian medical centers. Apparently, azole resistance in Aspergillus isolates in the southeast region of Brazil may be restricted to a limited number of patients infected by cryptic and rare Aspergillus species with primary resistance to antifungal drugs (9, 10, 24).
We emphasize the importance of performing continuous surveillance studies with a larger number of A. fumigatus isolates and including medical centers located in rural areas of Brazil, where exposure to azole compounds used in the environment may contribute to antifungal resistance to a greater extent.
ACKNOWLEDGMENTS
This work was supported by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Brazil (grant AUX-PE-PNPD-2312/2011); Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), Brazil (grant 2012/01138-4); and Conselho Nacional de Desenvolvimento Científico e Tecnológico, Brazil (CNPq, grant 307510/2015-8).
C.E.N. received a doctoral fellowship from CAPES–PROEX and CAPES–PDSE (grant 99999.008426/2014-07). S.S.G. received a postdoctoral fellowship from CAPES, Brazil (PNPD 23038.007393/2011-11). A.C.P.S. received a master fellowship from CAPES–DS. J.F.M. received grants from Astellas, Basilea, and Merck. He has been a consultant to Astellas, Basilea, and Merck and received speaker's fees from Merck, United Medical, and Gilead Sciences. A.L.C. received grants from CNPQ; educational grants from Astellas, Gilead, Pfizer, and United Medical; and research grants from Astellas and Pfizer. All other authors declare no conflict of interest.
FOOTNOTES
- Received 27 March 2017.
- Returned for modification 9 May 2017.
- Accepted 26 August 2017.
- Accepted manuscript posted online 11 September 2017.
- Copyright © 2017 American Society for Microbiology.
