Potency of Olorofim (F901318) Compared to Contemporary Antifungal Agents against Clinical Aspergillus fumigatus Isolates and Review of Azole Resistance Phenotype and Genotype Epidemiology in China

Triazole resistance in Aspergillus fumigatus is an increasing worldwide problem that causes major challenges in the management of aspergillosis. New antifungal drugs are needed, with novel targets, that are effective in triazole-resistant infection.

amphotericin B. Currently, triazole antifungals are recommended as the first choice for prophylaxis and treatment of aspergillosis (1). However, since the first report of triazole resistance in 1997 (2), many centers/hospitals around the world have reported resistance. Furthermore, voriconazole-resistant IA was found to be associated with treatment failure and excess mortality, which threatens the current treatment strategy for this pathogen (3,4).
The most common mechanism of triazole resistance is associated with mutations in the cyp51A gene, which encodes the protein targeted by the triazoles (5). Apparently, the mutant allele has spread throughout the A. fumigatus population and, thus, has been reported worldwide from patients as well as from the environment. In addition, several point mutations, such as G54, G138, and M220, intervene with the docking of azole drugs to CYP51A protein and render an azole-resistant phenotype (3). Rates of azole resistance in A. fumigatus vary extensively among countries and centers worldwide (6)(7)(8)(9), and in many countries the presence and frequency of azole resistance remain unknown. Multiple factors contribute to the observed variation in resistance frequency, including sample size, method of resistance detection, and geographical differences (10). The overall azole resistance rates ranged from 0 to 27.8% in different surveys (11)(12)(13). Since the spread of antifungal drug resistance has shown no signs of diminishing and new resistance mechanisms continue to emerge (14), understanding the genetic variability and relationship among resistant isolates from various parts of the world is of major importance. Azole resistance surveillance programs are scarce, and in China data on the prevalence of azole-resistant A. fumigatus are very limited. A few Chinese reports on triazole resistance in A. fumigatus are available, although most reports are from restricted geographic areas and include only a modest number of isolates (7,13,(15)(16)(17)(18)(19). Furthermore, the genetic relationship and variability of azole-resistant isolates of A. fumigatus in China remain unclear.
The clinical development of new antifungal drug classes is critical to overcoming current and future challenges in the management of Aspergillus diseases. Olorofim (formerly F901318), a leading representative of a novel class of drug belonging to orotomides, is an antifungal drug in clinical development that demonstrates excellent potency against a broad range of dimorphic and filamentous fungi, and it targets an important enzyme for pyrimidine biosynthesis, dihydroorotate dehydrogenase (20). The drug has in vitro activity against Aspergillus species and other difficult-to-treat molds, including Scedosporium and Lomentospora species, but lacks activity against Candida, Cryptococcus, and Mucorales species due to differences in drug target affinity (20)(21)(22). For Aspergillus species specifically, Buil et al. demonstrated in vitro activity against azole wild-type (WT) isolates as well as azole-resistant cyp51A mutant A. fumigatus isolates, also including a limited number of other Aspergillus species originating from the Netherlands (20).
We aimed to evaluate the potency of olorofim against a large set of clinical A. fumigatus isolates collected from China and compare the activity with that of contemporary antifungal agents. We further reviewed the prevalence of azole resistance and underlying cyp51A mutations in clinical and environmental A. fumigatus isolates in China.
Among the four TRAF isolates detected in this study, MIC values of olorofim (range, 0.016 to 0.062 mg/liter) were in the same range as those observed for the azole WT isolates. The lowest olorofim MIC was seen in isolate 247-32 with G54V, and the highest MIC was 0.062 mg/liter, for azole-resistant A. fumigatus isolates with the WT cyp51A gene.
Microsatellite typing. The genetic polymorphism of TRAF isolates from China and outside China was studied using short tandem repeat (STR) typing ( Fig. 1). Multiple distinct clusters can be identified based on microsatellite markers. STR typing of 29 Chinese TRAF isolates revealed 21 distinct genotypes distributed among environmental and clinical isolates that represented a major complex of the TRAF isolates disseminating all around the world.
Three microsatellite complexes (MCs) among the 21 cyp51A mutant genotypes of Chinese TRAF were recognizable, representing three distinct complexes of TRAF ( Fig.  1). Seven isolates with TR 34 /L98H in complex 1 were clonal and shared all nine loci except for two isolates, with one difference in one repeat at a single locus (2B) and the other at three loci (2A, 2B, and 3A). Thirteen isolates with mutation TR 34 /L98H/S297T/ F495I in complex 3 were highly polymorphic and different from the isolates with the   same mutation from the Netherlands and Denmark, which clustered in complex 1. Among 13 polymorphic genotypes observed in TR 34 /L98H/S297T/F495I isolates, an identical allelic profile was observed in a clinical (isolate C485) and an environmental isolate (isolate E739). One isolate with TR 46 /Y121F/T289A from Beijing was clustered in a complex group with isolates harboring TR 46 /Y121F/T289A from the Netherlands and Columbia.
The genotypic relationships among Chinese and global isolates were also inferred from the minimum spanning tree (Fig. 2). High genetic variability was observed among A. fumigatus isolates, which was not associated with the country and continent of origin. TRAF in China showed divergence in genetic variability as well.

DISCUSSION
In this study, we show that olorofim exhibits potent in vitro activity against 111 clinical A. fumigatus isolates, including TRAF from China. For the determination of wildtype upper limits (WT-UL) of visual values of A. fumigatus susceptibility to olorofim, we followed the 0.25 mg/liter value, as proposed by Jørgensen et al. (25). Olorofim MICs were low against 111 A. fumigatus isolates (modal MIC, 0.031 mg/liter; MIC range, 0.008 to 0.062 mg/liter), indicating that all MICs were within the range of the WT population. The observed MIC ranges are similar to those reported in previous reports from other geographic areas (19)(20)(21). The potency of olorofim was superior to that of triazoles and amphotericin B and comparable to those of three echinocandins tested. No substantial implications of the specific azole resistance mechanism for the activity of olorofim were demonstrated.
In an itraconazole-resistant A. fumigatus isolate with a G54V mutation, obtained from a patient undergoing high-dose itraconazole therapy, olorofim was 5-to 6-fold more potent than voriconazole and posaconazole, respectively. Furthermore, in an isolate harboring TR 34 /L98H/S297T/F495I, olorofim was 4-, 5-, and 9-fold more potent than voriconazole, posaconazole, and isavuconazole, respectively. Olorofim was also more active than voriconazole and isavuconazole against the two other TRAF isolates with WT cyp51A genes. These findings confirm previous reports (20,22,26) and indicate that triazole resistance does not affect olorofim activity, as olorofim MICs of these isolates are within the olorofim WT population (25).
As shown by microsatellite genotyping, STR typing of the Chinese TRAF isolates demonstrated two major clusters. Seven isolates with the TR 34 /L98H mutant type in China showed no genetic variability, suggesting a single and recent origin for these resistant isolates. Similarly, Abdolrasouli et al. (31) have described a similar structure in the TR-mediated azole-resistant A. fumigatus population in India. However, these observations contrast with the heterogeneity that was observed in environmental and clinical isolates in the Netherlands (32). The total of 13 Chinese isolates with TR 34 /L98H/ S297T/F495I emerged from only one branch, notably an identical allelic profile with TR 34 /L98H/S297T/F495I, present in clinical and environmental A. fumigatus isolates from China, suggesting an environmental origin of this major resistance mechanism. The two groupings suggested that these isolates have different evolutionary sources than the major TR 34 /L98H complex. Our study confirmed that resistance due to TR 34 / L98H mutation among A. fumigatus isolates evolved from separate local isolates (33). Our study was limited by the relatively small number of clinical A. fumigatus isolates included and the uneven geographic distribution in China. There are currently no azole resistance surveillance programs in China and many other countries, which would allow for more systematic collection and analysis of clinical A. fumigatus isolates. Furthermore, routine MIC testing is not performed in most clinical microbiology laboratories, which further complicates setting up such surveillance networks.
In conclusion, olorofim displays potent in vitro activity against A. fumigatus originating from China, including TRAF isolates. Further studies are needed to evaluate the in vivo efficacy of olorofim for the treatment of IA.
The need for novel targets is underscored by the increasing reports of TRAF both in patients and the environment. Despite multiple reports of TRAF in China, there is a need for systematic resistance surveillance to increase our understanding of resistance epidemiology and to guide antifungal treatment recommendations.

MATERIALS AND METHODS
Aspergillus isolates and species identification. A total of 111 clinical A. fumigatus isolates were collected from Huashan Hospital, Fudan University, from 2012 to 2017 in Shanghai, China. The isolates were identified based on morphological features and sequence analysis of the partial b-tubulin gene (benA) sequences (7). The primers used are listed in Table S1 in the supplemental material. Isolate information and GenBank accession numbers for the generated benA sequences are listed in Table S2.
Resistant isolates were defined according to the EUCAST breakpoints (version 10.0). There are no clinical breakpoints available for echinocandins and olorofim. Candida parapsilosis ATCC 22019 and C. krusei ATCC 6258 were used as the quality control strains.
cyp51A gene sequencing. Non-WT A. fumigatus isolates were selected for detection of cyp51A mutations. Genomic DNA was extracted as previously described (34), and full sequences of the cyp51A gene together with the promoter region were amplified and sequenced (35) (the primers used are listed in Table S1). The promoter and full sequence of cyp51A were aligned with the WT A. fumigatus strain (GenBank accession no. AF338659) using MAFFT version 7 (36). Tandem repeats (TR) in the gene promoter and mutations in the open reading frame were characterized after sequence alignment.
Genotyping of A. fumigatus isolates. Four azole-resistant A. fumigatus isolates were subjected to microsatellite typing, as previously described (37). Nine STR loci (STR Af2A, 2B, 2C, 3A, 3B, 3C, 4A, 4B, and 4C) were amplified in three separated multiplex PCRs. Each of the multiplex PCRs contained three different STRs. The fragments obtained were mixed with formamide and analyzed with GeneScan 500 LIZ on a 3730 DNA analyzer (Applied Biosystems, Foster City, CA, USA). The repeat numbers of the nine markers of all isolates were analyzed using Peak Scanner software 2 (Thermo Fisher, CA, USA).
Genetic analysis of microsatellite genotypes. To understand the genetic relationship of the azoleresistant A. fumigatus isolates in China to the global collection, a total of 29 Chinese azole-resistant A. fumigatus isolates (27 clinical and 2 environmental) and 102 azole-resistant A. fumigatus isolates collected globally were included by literature searching in PubMed. The twenty-nine Chinese azole-resistant A. fumigatus isolates included 25 isolates from the literature (7,13,15) and 4 isolates from the current study. The 102 azole-resistant A. fumigatus isolates were selected from the literature (6,19,(38)(39)(40)(41)(42)(43)(44)(45)(46) as representative of different genotypes and geographic areas worldwide. The composite genotype for each of the 131 A. fumigatus isolates was identified based on alleles at all nine microsatellite loci. The genotype markers were then used to identify genetic relationships among isolates. Dendrograms were generated by the unweighted pair group method using average linkages implemented in BioNumerics 7.6 (bioMérieux). A minimum spanning tree was also calculated in BioNumerics 7.6 using advanced cluster analysis. Results of these analyses were used to infer the potential source(s) of the triazole-resistant clinical and environmental A. fumigatus isolates in China.

SUPPLEMENTAL MATERIAL
Supplemental material is available online only. SUPPLEMENTAL FILE 1, PDF file, 0.2 MB.