The environmental mold Aspergillus fumigatus is the primary cause of invasive aspergillosis. In patients with high-risk conditions, including stem cell and organ transplant recipients, mortality exceeds 50%. Triazole antifungals have greatly improved survival (1); however, triazole-resistant A. fumigatus infections are increasingly reported worldwide and are associated with increased treatment failure and mortality (2). Of particular concern are resistant A. fumigatus isolates carrying either TR34/L98H or TR46/Y121F/T289A genetic resistance markers, which have been associated with environmental triazole fungicide use rather than previous patient exposure to antifungals (3,4). Reports of these triazole-resistant A. fumigatus strains have become common in Europe (2,3), but U.S. reports are limited (5). Because of the risk posed to immunocompromised patients, understanding the prevalence of such isolates in patients is important to guide clinical and public health decision-making. In 2011, CDC initiated passive laboratory monitoring for U.S. triazole-resistant A. fumigatus isolates through outreach to clinical laboratories. This system identified five TR34/L98H isolates collected from 2016 to 2017 (6), in addition to two other U.S. isolates collected in 2010 and 2014 and reported in 2015 (5). Four of these seven isolates were reported from Pennsylvania, two from Virginia, and one from California. Three isolates were collected from patients with invasive pulmonary aspergillosis, and four patients had no known previous triazole exposure. A. fumigatus resistant to all triazole medications is emerging in the United States, and clinicians and public health personnel need to be aware that resistant infections are possible even in patients not previously exposed to these medications.
Triazole antifungal medications are the primary treatment for invasive A. fumigatus infections, opportunistic infections that typically affect immunocompromised patients. Invasive aspergillosis is almost universally fatal without antifungal treatment. Clinical outcomes improved with the use of amphotericin B and have improved further with the introduction of mold-active triazole antifungals such as voriconazole, posaconazole, and itraconazole, which are also associated with fewer adverse events than is amphotericin B (7). Resistance to triazoles has been associated with treatment failure and increased mortality, but the prevalence of infection with resistant strains in U.S. hospitals is unknown (1,4). Structurally similar triazoles are used extensively as fungicides in agriculture and other environmental applications. A. fumigatus is not typically a plant pathogen but is common in soil and decaying plant material. Incidental exposure of A. fumigatus to fungicides during agricultural or other environmental applications can select for mutations conferring resistance to triazoles. A. fumigatus spores are known to be carried long distances in the air, putting patients at risk for infection with resistant strains, even in areas without known agricultural fungicide usage.
In Europe, molecular epidemiologic studies have identified two resistant A. fumigatus genotypes associated with environmental triazole exposure (4). These genotypes, TR34/L98H and TR46/Y121F/T289A, confer resistance to triazoles by altering the drug target, Cyp51A, which is involved in fungal cell wall synthesis. Importantly, TR34/L98H confers resistance to all mold-active medical triazoles without incurring a fitness cost or survival disadvantage to the fungus. A. fumigatus strains of this genotype have been isolated from the environment (e.g., compost, seeds, soil, commercial plant bulbs, and patient households) (8). Although these mutations have been detected repeatedly in environmental isolates, they have not been common among isolates from patients treated with long-term triazoles in whom resistance might have been expected to develop. Most (50%–75%) patients with TR34/L98H isolates have not been exposed to triazole therapy, further suggesting environmental acquisition of resistance (3).
Until 2015, no isolates with these genotypes had been reported in the United States; that year, a U.S. fungal reference laboratory reported detecting two TR34/L98H and two TR46/Y121F/T289A A. fumigatus isolates among 220 clinical isolates collected from 2001 to 2014 (5). In 2017, TR34/L98H A. fumigatus isolates were first detected in U.S. environmental samples obtained from a commercial peanut field treated with triazole fungicides (9). Together, these reports demonstrate that triazole-resistant A. fumigatus strains have emerged in the United States in both patients and the environment, likely caused by selection for resistance during environmental triazole use.
In 2011, CDC issued a request for clinical A. fumigatus isolates on the ClinMicroNet e-mail listserv of approximately 800 U.S. clinical microbiology laboratory directors, leading to a U.S. laboratory-based convenience sample of A. fumigatus isolates (systematic public health surveillance for A. fumigatus has not been conducted in the United States). In 2016, CDC received the first TR34/L98H isolate through this passive monitoring system, and an additional four have been identified to date among approximately 2,300 total isolates received (6). Together, these five and the two previously reported isolates (5) represent the first seven TR34/L98H isolates identified in the United States (Table). This report provides epidemiologic and clinical descriptions of the patients associated with these A. fumigatus triazole-resistant isolates.
Pennsylvania, 2010. Following stem cell transplantation for sickle cell anemia, a woman developed graft-versus-host disease and respiratory failure. Resistant A. fumigatus was isolated from sputum. Despite therapy with voriconazole and caspofungin, her respiratory status worsened, and therapy was switched to amphotericin B and caspofungin. She deteriorated further and died of multisystem organ failure 6 months after isolate collection.
Pennsylvania, 2014. A man with A. fumigatus colonization following lung transplantation initially was treated with long-term voriconazole followed by itraconazole. He was hospitalized with bacterial and viral pneumonia, developed clinical invasive pulmonary aspergillosis, and was treated with itraconazole and caspofungin, followed by posaconazole and caspofungin, then inhaled amphotericin B. Resistant A. fumigatus was isolated from a bronchoalveolar lavage. With worsening clinical status and persistently positive A. fumigatus cultures, therapy was switched to liposomal amphotericin B and caspofungin; however, bronchoscopy indicated ongoing fungal infection. He died from multisystem organ failure approximately 2 months after isolate collection.
Pennsylvania, 2016. A woman with sarcoidosis and invasive pulmonary aspergillosis was treated with low-dose voriconazole because of vision-associated side effects at higher doses. Respiratory symptoms had worsened at the time of sputum collection, and when the resistant A. fumigatus isolate was identified, therapy was changed to caspofungin for 12 months. Following therapy, the patient was clinically stable with no radiographic evidence of progression to chronic cavitary pulmonary aspergillosis or aspergilloma.
Pennsylvania, 2017. A resistant A. fumigatus isolate was collected by bronchoalveolar lavage from a woman with chronic obstructive pulmonary disease, interstitial pulmonary fibrosis, and hypersensitivity pneumonitis, while she was hospitalized for hydropneumothorax and bacterial pneumonia secondary to trauma; no antifungal treatment was given. The patient died of complications of her hydropneumothorax thought to be unrelated to A. fumigatus.
Virginia, 2016, case 1. A man who visited Virginia from Guatemala was hospitalized for acute bronchitis 3 weeks after his arrival. Resistant A. fumigatus was isolated from sputum during this hospitalization. No antifungals were administered, and the patient was discharged to primary care.
Virginia, 2016, case 2. A woman with cystic fibrosis had resistant A. fumigatus isolated from sputum at an outpatient visit 2 days before hospital admission for a cystic fibrosis exacerbation. While hospitalized, she received steroids and antibiotics but not antifungals. She was later discharged with oral antibiotics.
California 2017. A woman with a history of chronic obstructive pulmonary disease requiring inhaled corticosteroids, chronic heart failure, and chronic kidney disease was evaluated as an outpatient for a productive cough. Sputum cultures grew A. fumigatus, and IgG antibody to A. fumigatus was twice the normal value. She was not started on antibiotics or antifungals.
A. fumigatus strains with mutations conferring resistance to mold-active triazole agents have been found in clinical and environmental specimens in the United States. In total, 10 U.S. clinical isolates with these genotypes (seven TR34/L98H and three TR46/Y121F/T289A) have been reported (5,10). Together, these reports likely underrepresent the number of U.S. isolates because aspergillosis and A. fumigatus colonization are not reportable in any state and few laboratories perform susceptibility testing for Aspergillus species. Four of the seven patients with TR34/L98H were not treated with antifungal therapy following culture; these four isolates, all from sputum or bronchoalveolar lavage, likely reflected A. fumigatus colonization rather than infection. However, the presence of highly resistant A. fumigatus strains in patient isolates suggests that U.S. clinicians need to be aware of the risk for triazole-resistant aspergillosis. Notably, four patients had no known exposure to antifungal medications before culture of the resistant isolate, supporting possible environmentally acquired resistance.
The five isolates identified at CDC during 2016–2017 were collected from patients who did not share health care facilities, procedures, or county of residence, arguing against shared health care acquisition. Given that A. fumigatus can undergo selection for antifungal resistance during triazole fungicide exposure in the environment, and spores of resistant strains might be transmitted through the air and inhaled, further exploration of triazole fungicide use and presence of triazole-resistant A. fumigatus in these areas is warranted.
The findings in this report are subject to at least two limitations. First, among the seven A. fumigatus isolates with the TR34/L98H mutations identified in the United States to date, four were collected in Pennsylvania, two in Virginia, and one in California. These three states contributed only 28% of all CDC A. fumigatus isolates collected during 2015–2017, raising the possibility of geographic localization. Second, because isolates were collected through passive monitoring and not systematic surveillance, caution must be exercised when interpreting these findings.
With environmentally derived TR34/L98H triazole-resistant A. fumigatus detected in the United States, systematic surveillance, detailed geographic data, and data on triazole fungicide use could be important for assessing the scope of the problem and trends in resistance. Exploration of risk factors for patient acquisition might provide opportunities to prevent exposure and mitigate risk for invasive infection in susceptible populations. Clinicians and microbiologists need to be aware of the possibility of triazole-resistant A. fumigatus infections, even in triazole-naïve patients. Expanded capacity to test for antifungal susceptibility in A. fumigatus could help inform clinical and public health decisions.
Kevin Alby, Ana María Cárdenas, Brian Fisher, Talene Metjian, Christine Murphy, Kumar Nalluswami, Minh-Hong Nguyen, Natalie Nunnally, Anthony Pasculle, David Pegues, Bonnie Van Uitert, Sharon Watkins, Blair Weikert, Nathan Wiederhold.