Trends in the activity of mold-active azole agents against Aspergillus fumigatus clinical isolates with and without cyp51 alterations from Europe and North America (2017–2021)

ABSTRACT

Azole resistance in Aspergillus fumigatus (AFM) is increasing and often associated with cyp51 alterations. We evaluated the activity of isavuconazole and other mold-active azoles against 731 AFM isolates causing invasive aspergillosis collected in Europe (EU; n = 449) and North America (NA; n = 282). Isolates were submitted to CLSI susceptibility testing and epidemiological cutoff value (ECV) criteria. A posaconazole ECV of 0.5 mg/L was used as no CLSI ECV was determined. Azole non-wild-type (NWT) isolates were submitted for cyp51 sequencing by whole genome sequencing. Overall, isavuconazole activity (92.7%/94.0% WT in EU/NA) was comparable to other azoles (WT rate range, 90.9%–96.4%/91.8%–98.6%, respectively), regardless of the region. A total of 79 (10.8%) azole NWT isolates were detected, and similar rates of these isolates were noted in EU (10.7%) and NA (11.0%). Although most AFM were WT to azoles, increasing azole NWT rates were observed in NA (from 6.0% in 2017 to 29.3% in 2021). Azole NWT rates varied from 4.9% (2019) to 20.6% (2018) in EU without an observed trend. cyp51 alterations occurred in 56.3%/54.8% of azole NWT from EU/NA, respectively. The cyp51A TR₃₄/L98H alteration was observed only in EU isolates (72.0% of EU isolates), while cyp51A I242V occurred only in NA isolates (58.3%). Isavuconazole remained active (MIC, ≤1 mg/L) against 18.5/47.1% of azole NWT AFM exhibiting cyp51 alterations in EU/NA, along with voriconazole (29.6/82.4%; MIC, ≤1 mg/L) and posaconazole (48.1/88.2%; MIC, ≤0.5 mg/L). Fourteen different cyp51 alterations were detected in 44 of 79 NWT isolates. The in vitro activity of the azoles varied in AFM that displayed cyp51 alterations.

IMPORTANCE

A few microbiology laboratories perform antifungal susceptibility testing locally for systemically active antifungal agents. The identification of emerging azole-resistant Aspergillus fumigatus is worrisome. As such, there is a critical role for antifungal surveillance in tracking emerging resistance among both common and uncommon opportunistic fungi. Differences in the regional prevalence and antifungal resistance of these fungi render local epidemiological knowledge essential for the care of patients with a suspected invasive fungal infection.

INTRODUCTION

Aspergillus fumigatus is the leading cause of invasive (IA) and other forms of aspergillosis (1). Although IA continues to be associated with considerable morbidity and mortality, the introduction of mold-active triazoles as therapeutic agents of choice for first-line therapy has resulted in substantial improvement in outcomes when compared to earlier therapy with amphotericin B (2, 3). Among the mold-active triazoles, itraconazole has been available since 1997, voriconazole since 2002, posaconazole since 2006, and isavuconazole since 2015 (2, 4). Voriconazole and isavuconazole are recommended as first-line agents for the treatment of IA, itraconazole remains useful for chronic and allergic noninvasive forms of aspergillosis, and posaconazole is effective as prophylaxis in neutropenic patients (2).

Over the past two decades, both the frequency of IA and resistance to azoles have increased throughout the world, with the highest rates of resistance reported from Europe (especially the Netherlands) followed by North America (mostly in the USA) (3, 5 – 8). There are two major routes for the development of azole resistance in A. fumigatus. The first is a patient-exposure route in which an infecting isolate is exposed to prolonged azole therapy (usually itraconazole) in the setting of chronic obstructive pulmonary aspergillosis or aspergilloma (9 – 11). The second is an environmental-exposure route in which infection is due to the inhalation of conidia that already harbor a resistance mechanism secondary to exposure to azole-type fungicides in an agricultural setting (3, 8, 11, 12). These two scenarios generate azole-resistant isolates with one or more alterations in cyp51A or its homologue, cyp51B (3, 8, 11, 12). In the patient-exposure route, the resistant strains of A. fumigatus harbor different single nucleotide polymorphisms (SNPs) alone or in combination, most frequently at positions G54, M220, G138, or G448 in cyp51A. In the environmental-exposure route, SNPs along with tandem repeats (TR) of base pairs in the promoter region of cyp51A [most commonly TR₃₄/L98H (TR₃₄) and TR₄₆/Y121F/T289A (TR₄₆)] lead to increased expression of the altered cyp51A gene (3, 8, 11, 12). Other mutations in cyp51 have been associated with increased azole MIC values in A. fumigatus (8, 13). These mutations have been described as de novo mutations and are the result of extended exposure to azole antifungal agents (8, 13). Notably, between 25% and 50% of isolates of A. fumigatus with an azole-resistant phenotype are found to have no alterations in cyp51, suggesting that other mechanisms of resistance (MOR) are present, many of which are unknown (3). Regardless of the MOR, increased azole MIC values for A. fumigatus correspond with lower azole efficacy (3, 14). Alterations in cyp51 can have variable effects on the in vitro activities of individual azole antifungal agents (3, 8, 11 – 13). These effects are best delineated by testing all four triazoles (3). At present, it is unclear that infection with an isolate of A. fumigatus that is phenotypically resistant to one azole can be successfully managed using another azole agent with a susceptible or wild-type (WT) MIC value (14). Antifungal susceptibility testing (AFST) of all four triazoles is recommended when the prevalence of resistance in A. fumigatus is greater than 5%. Monotherapy with a triazole is not recommended when the prevalence of resistance exceeds 10% (3, 14).

Previous reports from the SENTRY Antifungal Surveillance Program have demonstrated a predominance of azole-resistant A. fumigatus from Europe with a low level of decreased susceptibility to azoles in North America and the rest of the world (13, 15 – 19). These findings confirmed reports by other investigators showing that the frequency of azole resistance in clinical isolates from Europe ranged from 0% to 20% and in the USA from 1.5% to 5.0% (3, 8, 11, 12, 20 – 22). More recently, the US Centers for Disease Control and Prevention (CDC) reported results from passive surveillance of azole resistance in A. fumigatus demonstrating that in 2011–2013, 5.0% of 1,026 US isolates expressed a non-wild-type (NWT) phenotype to itraconazole with no isolates harboring an environmental-type alteration in cyp51A (TR₃₄ or TR₄₆) (22). A repeat survey conducted in 2015–2017 found that only 1.5% of 1,356 isolates were NWT to itraconazole or voriconazole but documented the presence of the environmental alteration TR₃₄ in 5 isolates (0.4% of the total) (21). These results, supported by other recent reports from four different states, indicate that isolates with the environmental-type resistance mutation are present in the USA (21, 23). In addition, recently a report from SENTRY showed that antifungal resistance among isolates of A. fumigatus appears to be increasing in North America, Europe, and the Asia-Pacific region (24).

In the present survey, we assessed the frequency of triazole resistance among isolates of A. fumigatus from Europe and North America for the years 2017–2021. We report the MIC distributions for four mold-active azole antifungal agents (isavuconazole, itraconazole, posaconazole, and voriconazole) and 731 isolates of A. fumigatus that were submitted to the SENTRY Antifungal Surveillance Program (JMI Laboratories, North Liberty, IA, USA) for reference identification and in vitro antifungal susceptibility testing by the CLSI broth microdilution (BMD) method. Isolates submitted for testing were collected in 2017–2021 from clinically significant infections in Europe (449 isolates) and North America (282 isolates) as part of the SENTRY Antifungal Surveillance Program. All isolates were subjected to antifungal susceptibility testing to detect emerging resistance by applying epidemiological cutoff values (ECVs), where available. Isolates with elevated MIC values (greater than the ECV) were subjected to molecular analysis to characterize any alterations in cypP51.

MATERIALS AND METHODS

Organisms

A collection of 731 non-duplicate clinical isolates of Aspergillus fumigatus from the SENTRY Antimicrobial Surveillance Program collected from 2017 to 2021 were included in the study. A single isolate per infection episode determined to be significant by local criteria as the reported probable cause of infection were included in this investigation. A total of 35 medical centers in North America (18 sites; 2 countries; 282 isolates) and Europe (17 sites; 11 countries; 449 isolates) sent isolates to the coordinating laboratory as part of the SENTRY Program.

Identification methods

Isolates were submitted to JMI Laboratories (North Liberty, IA, USA) where species identification was confirmed using DNA sequencing and proteomic methods (13, 25). Mold isolates were subcultured on potato dextrose agar (Remel, Inc.; Lenexa, KS, USA) to assess purity and viability. Isolates confirmed as pure were inoculated into Sabouraud Liquid Broth Modified (Becton, Dickenson and Company; Sparks, MD, USA) and the hyphae harvested and prepared for formic acid extraction. Isolates then were submitted to matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) using the MALDI Biotyper (Bruker Daltronics; Billerica, MA, USA). Isolates that did not score ≥1.8 by spectrometry were identified using sequencing of the 28S ribosomal subunit, followed by analysis of β-tubulin or internal spacer regions (ITS) (13, 18, 25, 26). Nucleotide sequences were analyzed using Lasergene software (DNASTAR; Madison, WI, USA) and compared to sequences using BLAST (https://blast.ncbi.nlm.nih.gov/Blast.cqi).

Susceptibility testing

All isolates of A. fumigatus were submitted to antifungal susceptibility testing by BMD using CLSI M38 methodology (27). Frozen-form microdilution panels using RPMI 1640 broth supplemented with MOPS (morpholinepropane sulfonic acid buffer) and 0.2% glucose were inoculated with 0.4–5.0 × 10⁴ CFU/mL conidial suspensions for a final concentration of 0.2–2.5 × 10⁴ CFU/mL. Minimal inhibitory concentrations (MICs) were visualized after 48 hours. MIC endpoints were read at the lowest concentration producing visually clear wells. Quality control was performed in accordance with CLSI M38 (2017) guidelines using Aspergillus flavus ATCC 204304 and A. fumigatus ATCC MYA-3626. MIC values were within the QC ranges.

Clinical breakpoints (CBPs) have been published by CLSI for A. fumigatus and voriconazole [susceptible (S), ≤0.5 mg/L; intermediate (I), 1 mg/L; resistant (R), ≥2 mg/L] and approved but not yet published for isavuconazole (S, ≤1 mg/L; I, 2 mg/L; R, ≥4 mg/L) (28). However, ECVs have been developed for A. fumigatus and isavuconazole (ECV, 1 mg/L), itraconazole (ECV, 1 mg/L), posaconazole (ECV, 0.5 mg/L), and voriconazole (ECV, 1 mg/L) (29 – 31). Isolates for which azole MIC results exceed the ECV were considered NWT (28, 29). The European Committee on Antimicrobial Susceptibility Testing (EUCAST) (32) has developed both ECVs (based on MIC distribution) and clinical breakpoints (based on MIC distributions) to reflect dosing and pharmacokinetic/pharmacodynamic parameters, on the one hand, and likelihood of clinical success and failure, on the other.

Characterization of mutations in the sterol 14 alpha-demethylase-encoding gene

A. fumigatus isolates displaying isavuconazole, itraconazole, posaconazole, or voriconazole MIC values above the ECV (NWT) were submitted to molecular detection of cyp51A and cyp51B mutations as previously described (13). These sequences were compared to the GenBank sequences available under the accession numbers AAK73659.1 for cyp51A and AAK73660.1 for cyp51B.

RESULTS

The cumulative frequency of MIC distributions for the four mold-active azoles and A. fumigatus isolates from EU and NA is presented in Table 1. Isavuconazole, itraconazole, posaconazole, and voriconazole displayed similar activities (MIC_50/90, 1/1 mg/L, 1/1 mg/L, 0.5/0.5 mg/L, and 0.5/0.5 mg/L, respectively; Table 1) against A. fumigatus isolates from EU and NA. Greater than 90.0% of the isolates tested (EU/NA) were WT to isavuconazole (92.7/94.0% WT), itraconazole (90.9/91.8% WT), posaconazole (96.4/97.5% WT), and voriconazole (95.5/98.6% WT). The overall frequency of NWT strains of A. fumigatus (EU/NA) was 7.3/6.0% for isavuconazole, 9.1/8.2% for itraconazole, 3.6/2.5% for posaconazole, and 4.5/1.4% for voriconazole; 10.7/11.0% of EU/NA isolates were NWT to one or more triazole (Tables 1 and 2). The overall frequency of resistance (R) to isavuconazole and voriconazole was determined by application of the CLSI CBPs: R to isavuconazole was 3.1% overall [4.7/0.7% (EU/NA)] while resistance to voriconazole was 3.3% [4.5/1.4% (EU/NA)].

TABLE 1.

Antimicrobial activity of isavuconazole, itraconazole, voriconazole, and posaconazole tested against A. fumigatus isolates from NA and EU, 2017–2021

Organism/organism group	Continent (no. of isolates)	No. and cumulative % of isolates inhibited at MIC (mg/L) of										MIC50	MIC90
		≤0.03	0.06	0.12	0.25	0.5	1	2	4	8	> a
Aspergillus fumigatus
Isavuconazole	NA (282)			0	18	180	67	15	1	0	1	0.5	1
				0	6.4	70.2	94	99.3	99.6	99.6	100
	EU (449)		0	3	30	265	118	12	13	5	3	0.5	1
			0	0.7	7.3	66.4	92.7	95.3	98.2	99.3	100
Itraconazole	NA (282)			0	11	103	145	19	3	0	1	1	1
				0	3.9	40.4	91.8	98.6	99.6	99.6	100
	EU (449)			0	20	163	225	21	10	4	6	1	1
				0	4.5	40.8	90.9	95.5	97.8	98.7	100
Voriconazole	NA (282)		0	3	116	137	22	3	1			0.5	0.5
			0	1.1	42.2	90.8	98.6	99.6	100
	EU (449)		0	9	185	215	20	15	2	0	3	0.5	0.5
			0	2	43.2	91.1	95.5	98.9	99.3	99.3	100
Posaconazole	NA (282)	0	2	39	128	106	7					0.25	0.5
		0	0.7	14.5	59.9	97.5	100
	EU (449)	0	4	64	241	123	14	1	1			0.25	0.5
		0	0.9	15.2	69	96.4	99.6	99.8	100
Azole-wild-type Aspergillus fumigatus
Isavuconazole	NA (251)			0	18	178	55					0.5	1
				0	7.2	78.1	100
	EU (401)		0	3	30	260	108					0.5	1
			0	0.7	8.2	73.1	100
Itraconazole	NA (251)			0	11	101	139					1	1
				0	4.4	44.6	100
	EU (401)			0	20	161	220					1	1
				0	5	45.1	100
Voriconazole	NA (251)		0	3	116	119	13					0.5	0.5
			0	1.2	47.4	94.8	100
	EU (401)		0	9	185	196	11					0.5	0.5
			0	2.2	48.4	97.3	100
Posaconazole	NA (251)	0	2	39	123	84	3					0.25	0.5
		0	0.8	16.3	65.3	98.8	100
	EU (401)	0	4	64	232	100	1					0.25	0.5
		0	1	17	74.8	99.8	100
Azole non-wild-type Aspergillus fumigatus
Isavuconazole	NA (31)				0	2	12	15	1	0	1	2	2
					0	6.5	45.2	93.5	96.8	96.8	100
	EU (48)				0	5	10	12	13	5	3	2	8
					0	10.4	31.2	56.2	83.3	93.8	100
Itraconazole	NA (31)				0	2	6	19	3	0	1	2	4
					0	6.5	25.8	87.1	96.8	96.8	100
	EU (48)				0	2	5	21	10	4	6	2	>8
					0	4.2	14.6	58.3	79.2	87.5	100
Voriconazole	NA (31)				0	18	9	3	1			0.5	2
					0	58.1	87.1	96.8	100
	EU (48)				0	19	9	15	2	0	3	1	4
					0	39.6	58.3	89.6	93.8	93.8	100
Posaconazole	NA (31)			0	5	22	4					0.5	1
				0	16.1	87.1	100
	EU (47)			0	9	23	13	1	1			0.5	1
				0	19.1	68.1	95.7	97.9	100
Azole non-wild-type with cyp51 alterations Aspergillus fumigatus
Isavuconazole	NA (17)				0	1	7	8	0	0	1	2	2
					0	5.9	47.1	94.1	94.1	94.1	100
	EU (27)				0	1	4	2	12	5	3	4	>8
					0	3.7	18.5	25.9	70.4	88.9	100
Itraconazole	NA (17)					0	3	11	2	0	1	2	4
						0	17.6	82.4	94.1	94.1	100
	EU (27)						0	8	9	4	6	4	>8
							0	29.6	63	77.8	100
Voriconazole	NA (17)				0	7	7	2	1			1	2
					0	41.2	82.4	94.1	100
	EU (27)				0	5	3	14	2	0	3	2	>8
					0	18.5	29.6	81.5	88.9	88.9	100
Posaconazole	NA (17)			0	3	12	2					0.5	1
				0	17.6	88.2	100
	EU (27)				0	13	12	1	1			1	1
					0	48.1	92.6	96.3	100
Azole non-wild-type cyp51 wild-type Aspergillus fumigatus
Isavuconazole	NA (14)				0	1	5	7	1			2	2
					0	7.1	42.9	92.9	100
	EU (21)				0	4	6	10	1			2	2
					0	19	47.6	95.2	100
Itraconazole	NA (14)				0	2	3	8	1			2	2
					0	14.3	35.7	92.9	100
	EU (21)				0	2	5	13	1			2	2
					0	9.5	33.3	95.2	100
Voriconazole	NA (14)				0	11	2	1				0.5	1
					0	78.6	92.9	100
	EU (21)				0	14	6	1				0.5	1
					0	66.7	95.2	100
Posaconazole	NA (14)			0	2	10	2					0.5	1
				0	14.3	85.7	100
	EU (21)			0	9	10	1					0.5	0.5
				0	45	95	100

TABLE 2.

Frequency of triazole NWT isolates of A. fumigatus in NA and EU, 2017–2021

Year	Region (N)	% NWT
		ISC	ITC	PSC	VRC	≥ 1 triazole
2017	NA (50)	4.0	0.0	2.0	0.0	6.0
	EU (60)	10.0	5.0	5.0	5.0	10.0
2018	NA (62)	9.7	6.5	1.6	0.0	14.5
	EU (68)	16.2	14.7	7.4	7.4	20.6
2019	NA (84)	2.4	2.4	0.0	1.2	2.4
	EU (103)	4.9	4.9	0.0	4.9	4.9
2020	NA (45)	4.4	17.8	4.4	2.2	17.8
	EU (93)	8.6	11.8	5.4	5.4	11.8
2021	NA (41)	12.2	22.0	7.3	4.9	29.3
	EU (125)	2.4	9.6	2.4	1.6	10.4
All years	NA (282)	6.0	8.2	2.5	1.4	11.0
	EU (449)	7.3	9.1	3.6	4.5	10.7

The mold-active azoles have been tested against isolates of A. fumigatus in the SENTRY Program since 2001. The data from 2001 to 2009 (17) and 2015 to 2017 (18) have been published previously. These data show that the percentage of isolates for which the MIC was greater than the ECV (e.g., % NWT) for itraconazole increased from 3.0% in 2001–2009 to 4.2% in 2015–2017 (data not shown), indicating a gradual increase in isolates likely to harbor an acquired resistance mechanism. Given this earlier data, it is notable that the % NWT to itraconazole in the present survey (2017–2021; EU and NA data combined) is 8.8% [9.1% (EU) and 8.2% (NA)] (Table 2).

The frequency of azole NWT strains for each of the triazoles for each year from 2017 through 2021 in EU and NA is shown in Table 2. The frequency of isolates that were NWT to at least one of the four triazoles was greater among isolates from EU for the years 2017–2019, but this shifted in the last 2 years of the survey such that the % NWT was greater among isolates from NA compared to those from EU. The % NWT when taking all years (2017–2021) into account was 11.0% for NA isolates and 10.7% for EU isolates. Whereas the % NWT for each of the triazoles tested against isolates from NA increased over time (except for 2019), this was not the case for isolates from EU. The % NWT to at least one of the triazole agents tested against isolates from NA increased steadily from 6.0% in 2017 to 29.3% in 2021, exempting 2019 (Table 2). The % NWT isolates from EU ranged from a low of 4.9% in 2019 to 20.6% in 2018 without a consistent trend up or down over the 5-year period (Table 2).

A total of 79 isolates [48 (EU) and 31 (NA)] were NWT to one or more of the triazoles tested, 44 (55.7%) of which [27 (EU; 56.3%) and 17 (NA; 54.8%)] harbored one or more alterations in the cyp51 genes (Table 3). A total of 35 isolates harbored alterations in cyp51A and 9 had alterations in cyp51B (2 isolates contained alterations in both cyp51A and cyp51B) (Table 3). Among the 44 isolates with cyp51 alterations, 31 [70.5%; 81.5/52.9% (EU/NA)] were NWT to isavuconazole, 41 [93.2%; 100.0/58.8% (EU/NA)] were NWT to itraconazole, 16 [36.4%; 51.9/11.8% (EU/NA)] were NWT to posaconazole, and 22 [50.0%; 70.4/17.6% (EU/NA)] were NWT to voriconazole (Table 3). cyp51 alterations were similarly detected among A. fumigatus isolates from NA (17/282; 6.0%) and EU (27/449; 6.0%).

TABLE 3.

Summary of CYP51 alterations detected among non-wild-type Aspergillus fumigatus isolates ^a

State and/or country	Organism	No. of isolates	MIC range (mg/L)				Amino acid alterations
			ISC	ITR	VRC	PSC	cyp51A	cyp51B
VT, USA	Aspergillus fumigatus	1	2	1	0.5	0.25	A9T	Wild type
Czech Republic	Aspergillus fumigatus	1	1	2	0.5	0.5	D172V	Wild type
Czech Republic	Aspergillus fumigatus	2	1	2	0.5	0.5	F46Y, D172V, E427K	Wild type
VT, USA	Aspergillus fumigatus	1	2	2	1	0.5	F46Y, M172V, E427K	Wild type
Czech Republic	Aspergillus fumigatus	1	2	2	1	0.5	F46Y, M172V, N248T, D255E, E427K	Wild type
VT, USA	Aspergillus fumigatus	1	1	2	0.5	0.5	F46Y, M172V, N248T, D255E, E427K	Wild type
VA, USA	Aspergillus fumigatus	1	>8	>8	4	0.5	G448S	Wild type
France	Aspergillus fumigatus	1	>8	>8	>8	2	H147Y	Wild type
AL, USA	Aspergillus fumigatus	1	1	2	0.5	0.25	I242V	Wild type
Canada	Aspergillus fumigatus	1	1	2	0.5	0.5	I242V	Wild type
IN, USA	Aspergillus fumigatus	1	1	2	1	1	I242V	Wild type
VA, USA	Aspergillus fumigatus	2	1	2	0.5	0.5	I242V	Wild type
IN, USA	Aspergillus fumigatus	2	1–2	1–4	1	0.5	I242V	Q42L, S501Q
France	Aspergillus fumigatus	1	0.5	2	0.5	0.5	K67Q	Wild type
MI, USA	Aspergillus fumigatus	1	2	2	2	1	K67Q	Wild type
Belgium	Aspergillus fumigatus	1	4	4	2	1	L98H,TR34	Wild type
Czech Republic	Aspergillus fumigatus	1	4	>8	2	1	L98H,TR34	Wild type
Germany	Aspergillus fumigatus	1	8	>8	4	1	L98H,TR34	Wild type
Italy	Aspergillus fumigatus	10	2–> 8	2–> 8	1–> 8	0.5–4	L98H,TR34	Wild type
Slovenia	Aspergillus fumigatus	1	4	>8	2	0.5	L98H,TR34	Wild type
UK	Aspergillus fumigatus	4	4–8	4–> 8	2	0.5–1	L98H,TR34	Wild type
France	Aspergillus fumigatus	1	1	2	0.5	0.5	Wild type	F149V
France	Aspergillus fumigatus	1	4	4	1	1	Wild type	Q42L
NJ, USA	Aspergillus fumigatus	5	0.5–2	1–4	0.5–2	0.25–0.5	Wild type	Q42L
Belgium	Aspergillus fumigatus	1	>8	8	>8	0.5	Y121F, M172I, T289A, G448S, TR46	Wild type

^a ISC, isavuconazole; ITR, itraconazole; VRC, voriconazole; PSC, posaconazole.

The most frequent alteration was cyp51A TR₃₄/L98H, carried by 18 isolates from EU (10 from Italy, 4 from the UK, and 1 each from Belgium, Czech Republic, Slovenia, and Germany), all of which were NWT to isavuconazole and itraconazole, 17 were NWT to voriconazole (and resistant applying the CLSI CBPs), and 12 were NWT to posaconazole (Table 3). A single isolate from Belgium harbored the environmental TR₄₆ along alterations Y121F, M172I, T289A, and G448 and was NWT to isavuconazole, itraconazole, and voriconazole (Table 3). None of the isolates from NA harbored the TR₃₄ or TR₄₆ alterations. Single substitutions in cyp51A were detected in 10 of 17 isolates from North America, 7 of which carried the alteration I242V (all WT to voriconazole, 6 of 7 WT to itraconazole, isavuconazole, and posaconazole) (Table 3). One NA isolate carried the cyp51A alteration G448S (NWT to isavuconazole, itraconazole, and voriconazole), one carried K67Q (NWT to all four triazoles), and one carried A9T (NWT to isavuconazole). Single substitutions were seen in 5 of 28 isolates from EU (3 in cyp51A and 2 in cyp51B): Q42L (cyp51B; NWT to isavuconazole, itraconazole, and posaconazole), K67Q (NWT to itraconazole), H147Y (NWT to all 4 triazoles), F149V (CYP51B; NWT to itraconazole), and D172V (NWT to itraconazole) (Table 3).

A series of three (F46Y, M172V, E427K) or five (F46Y, M172V, N248T, D255E, E427K) alterations on cyp51A were detected in three (two from EU and one from NA) and two (one from EU and one from NA) isolates, respectively. All three of the isolates with F46Y, M172V, and E427K substitutions were NWT to itraconazole, and one was NWT to isavuconazole and itraconazole (Table 3). Both isolates with the substitutions F46Y, M172V, N248T, D255E, and E427K were NWT to itraconazole, and one was also NWT to isavuconazole.

Alterations in cyp51B were noted in nine isolates; six of nine carried Q42L, five from NA (four NWT to itraconazole, four NWT to isavuconazole, and one NWT to posaconazole) and one from EU (NWT to isavuconazole, itraconazole, and voriconazole). The remaining three isolates included one isolate from France with a F149V substitution (NWT to itraconazole) and two isolates from NA with cyp51B Q42L coupled with a cyp51A I242V alteration (one isolate NWT to itraconazole and one NWT to isavuconazole) (Table 3).

This collection of 731 isolates of A. fumigatus contained 652 isolates (89.2%) that were WT to all 4 azoles, 35 isolates (4.8%) that were NWT to one or more azole but showed no alterations in cyp51A or cyp51B and 44 isolates (6.0%) that were NWT to one or more azole and harbored alterations in cyp51 (Table 1). Among the 35 NWT isolates without cyp51 alterations, 45.7% were WT to isavuconazole, 34.3% were WT to itraconazole, 91.2% were WT to posaconazole, and 94.3% were WT to voriconazole. By comparison, among the 44 NWT isolates with cyp51 alterations, 29.5% were WT to isavuconazole, 6.8% were WT to itraconazole, 63.6% were WT to posaconazole, and 50.0% were WT to voriconazole.

DISCUSSION

Aspergillus fumigatus is an opportunistic fungal pathogen that is the major cause of IA as well as several different environmentally acquired respiratory infections (33). Triazole-resistant A. fumigatus is a serious threat to human health and has been detected on six continents and in several different environments throughout the world (8, 34 – 36). The frequency of triazole-resistant A. fumigatus as a cause of IA has been increasing over the past two decades, most notably in EU and NA (3, 5 – 8, 20). Resistance rates as high as 26% have been reported in an intensive care unit (ICU) in the Netherlands (37). Despite the detection of triazole-resistant A. fumigatus throughout the world, most clinical laboratories do not perform AFST of filamentous fungi (1, 11, 38 – 41). In addition, the low rate of Aspergillus recovery might also play a role in the lack of susceptibility data, and, as such, the frequency of triazole resistance can be largely underestimated.

Thus far, MOR documented for the triazoles and A. fumigatus has been limited to alterations in cyp51 (3, 8, 11, 12). Whereas triazole-resistant A. fumigatus in EU has been shown to involve single point mutations in cyp51A (e.g., G54W, M220T), the environmental mutations TR₃₄ and TR₄₆ are the most common alterations found in both clinical and environmental samples (3, 8, 11, 12). Isolates of A. fumigatus with these environmental fungicide-associated MOR are especially concerning as they are often resistant/NWT to all 4 triazoles and cause infections in azole-naïve patients (3, 11). In the USA, there has been little study of the association between agricultural triazole fungicide use and human infections (23, 42). Together with comparatively low rates of resistance in clinical isolates (17, 18, 20 – 22), limited surveys in the USA have determined that the most frequent alterations in cyp51A were SNPs M220I and I242V (21, 23). The TR-based MOR in US patient isolates has been identified in four isolates (two with TR₃₄ and two with TR₄₆) from as early as 2008 along with an additional six isolates (five with TR₃₄ and one with TR₄₆) detected through 2018 (21, 23, 42). Most recently, a fatal environmental fungicide-associated triazole-resistant isolate of A. fumigatus (TR₃₄) was reported in a patient from Pennsylvania (24). Given the lack of standardized surveillance and limited clinical testing, it is likely that these isolates reflect only a small proportion of resistant infections (3, 21).

Although triazoles are the most frequently employed antifungal agents in the USA (42, 43), it is notable that triazole use in US hospitals declined by 21% (mostly fluconazole) during 2006–2012 (42, 43), whereas the agricultural use of triazole-type fungicides increased by fourfold in recent (2006–2016) years (42). Given the experience in EU, it is reasonable to assume that this experience can select for A. fumigatus strains harbouring an environmental-associated cyp51A gene mutation such as TR₃₄ or TR₄₆. The SENTRY Antifungal Surveillance Program has monitored infections due to A. fumigatus since 2001 and has performed DNA sequence analysis of cyp51 since 2015. Over this time, environmental alterations have been detected in isolates from EU but not from NA (13, 15 – 19).

There are several findings from this survey that are notable: (i) the vast majority of isolates tested (89.2%) expressed a WT azole phenotype that was similar in isolates from both EU (89.3%) and NA (89.0%) (Table 2); (ii) as in previous surveys (13, 15 – 19), the % NWT for each of the triazoles was greater for isolates from EU versus NA for the years 2017–2019, but this was reversed for years 2020–2021 (Table 2); (iii) as noted in previous studies, alterations in cyp51 were detected in 55.7% of triazole NWT isolates (56.3% in EU and 54.8% in NA); and (iv) there was a consistent trend of increasing % NWT for isolates from NA with a high of 29.3% NWT to ≥1 triazole in 2021 (Table 2). The highest % NWT among isolates from EU was 20.6% in 2018, but there was no consistent trend over time (Table 2); (v) the overall NWT percentage was 11.0% in isolates from NA and 10.7% in isolates from EU (Table 2); and (vi) the environmental alterations, TR₃₄ and TR₄₆, were only detected in isolates from EU despite a comparable frequency of azole NWT isolates in both EU and NA. This observation suggests that different non-environmental exposures account for the emerging resistance observed in isolates from NA. The rate of resistance/NWT was >10% in 3 of 5 study years for NA and in 4 of 5 years for EU. AFST is encouraged in the setting of 5% or greater resistant/NWT isolates, and triazole monotherapy is discouraged when rates of resistance exceed 10% (3). All studies to date indicate that azole resistance is associated with the therapeutic failure of this class of agents for the treatment of IA.

These results show for the first time in the SENTRY Antifungal Surveillance Program that the triazole NWT percentage for isolates from NA exceeded that of isolates from EU (17, 18). In contrast to the isolates from EU, the NWT isolates from NA did not possess any of the environmental alterations TR₃₄/L98H or TR₄₆/Y121F/T289A (Table 3). As such, it would appear that the resistant/NWT strains in NA are the result of drug pressure in the healthcare milieu rather than exposure of A. fumigatus to azole-type fungicides in the environment, despite decreased use of this class of agents in US hospitals. Unfortunately, we have no information concerning azole exposure of patients in this survey, and very little data exist to document the presence of azole-resistant A. fumigatus from the environment as has been recorded elsewhere in the world (42).

There are some limitations in this SENTRY Surveillance Program survey that must be acknowledged. First, we neither collect clinical outcome data nor do we identify those individuals who received an antifungal agent. As such, we are unable to establish any clinical correlation between MIC values and clinical outcomes. Second, we did not identify any mechanisms of resistance beyond alterations in cyp51A/B (e.g., HapE, Hmg1, SrbA). There were several isolates of A. fumigatus that were NWT to an azole but did not possess specific alterations in cyp51. The potential for an efflux mechanism accounting for elevated MIC values was not evaluated. Finally, the SENTRY Program is a sentinel surveillance and does not represent population-based surveillance.

In summary, the data presented in the present study expand upon the azole MIC distributions for A. fumigatus. We note that the frequency of azole NWT strains of A. fumigatus has increased since the survey conducted in 2001–2009 and now approaches 10% overall, a level at which the use of azole monotherapy is questionable (3). Importantly, only approximately half of the azole NWT isolates displayed alteration in the CYP51 sequences. Therefore, other potential mechanisms of resistance not evaluated in this study could play an significant role in A. fumigatus. The azole NWT isolates harbored alterations in cyp51 that included environmental alterations (e.g., TR₃₄/L98H) in isolates from EU and non-synonymous point mutations in isolates from NA. Antifungal resistance among isolates of A. fumigatus appears to be increasing in NA and remains stable in EU. State-of-the-art methods for species identification and antifungal susceptibility testing will be important to further define the impact of azole resistance in both local and regional settings.