Analysis of a Candida auris Outbreak Provides New Insights into an Emerging Pathogen

Candida auris is an emerging human fungal pathogen that is being increasingly linked to outbreaks. It is concerning to health care workers because of its high mortality rate, due primarily to its antifungal resistance. Among the tools being developed to study this yeast are large cohorts of regional isolates, which can be useful for studying epidemiology, antifungal susceptibility patterns, and diagnostic methods. In this issue of the Journal of Clinical Microbiology, Y.

ABSTRACT

Candida auris is an emerging human fungal pathogen that is being increasingly linked to outbreaks. It is concerning to health care workers because of its high mortality rate, due primarily to its antifungal resistance. Among the tools being developed to study this yeast are large cohorts of regional isolates, which can be useful for studying epidemiology, antifungal susceptibility patterns, and diagnostic methods. In this issue of the Journal of Clinical Microbiology, Y. Zhu, B. O’Brien, L. Leach, A. Clarke, et al. (J Clin Microbiol 58:e01503-19, 2020, https://doi.org/10.1128/JCM.01503-19) describe the laboratory findings of a collection of isolates from a large outbreak of C. auris obtained from numerous health care facilities in the New York area. Real-time PCR was used as a screening tool with great accuracy, while internal transcribed spacer (ITS) and D1/D2 sequencing were successfully employed for isolate clade assignment. South Asia clade I was identified as the major genotype, while South American clade IV was a minor genotype. Surveillance isolates from patients confirmed axilla/groin and nare colonization; however, results of quantitative analysis of fungal burdens showed that when the nares are colonized, burdens are significantly higher than for axilla/groin colonization. Antifungal susceptibility testing was in agreement with past studies. High levels of fluconazole resistance were detected, while few isolates were resistant to echinocandins. Resistance to multiple antifungals was frequent, and three isolates were recovered that appeared to be pan-resistant. This type of study is yet another useful tool for investigating C. auris, which is becoming an increasingly important human fungal pathogen that should be monitored very closely.

TEXT

Candida auris is an emerging fungal pathogen that is spreading rapidly throughout the world. It has now been detected in 39 countries, 13 states (in the United States), and on all continents except Antarctica (1). The fungus, which grows primarily as a budding yeast, was first described in 2009 (2) but has been detected in culture collections as far back as 1996 (3, 4). The initial isolate was recovered from an ear (2); however, clinical infections are most frequently associated with the bloodstream and occur most often in immunosuppressed patients, particularly in patients in the intensive care unit (ICU) (5, 6). In fact, in some parts of the world, it is endemic in hospitals, which could lead to further infectious risk for ICU patients (7). Through genome sequencing, four clades have been identified (8), with a fifth clade possibly emergent based on an Iranian isolate recently discovered that is significantly different from the four other clades (9). The current four clades include populations grouped into South Asian (I), East Asian (II), African (III), and South American (IV) clades.

C. auris infections have frequently been misdiagnosed, most often as C. haemulonii, and less often as C. famata, C. guilliermondii, C. lusitaniae, C. parapsilosis, C. sake, R. glutinis, C. duobushaemulonii, C catenulate, C. tropicalis, or Saccharomyces cerevisiae, and usually on platforms that utilize biochemical phenotyping (10–12). However, FDA-approved methods, such as matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS), are quickly coming online (13), and a gold standard for identification based on ITS or D1/D2 sequence has arguably been established (14). Although misdiagnosis is potentially problematic, the major cause for concern from C. auris infections is the high mortality rate, which can range from 28% to 78% depending on country (15). Concomitant with this rate is the cost to treat these patients, which is among the highest of all mycoses (16, 17). A major characteristic driving the high mortality rates, besides the tendency to infect already severely debilitated patients, is the high frequency of drug resistance. Most C. auris isolates display elevated resistance to fluconazole, while resistance to other antifungals has been variable (12). However, multidrug resistance has been found to be frequent (8), and panfungal resistance has been described (18). These factors have contributed to C. auris becoming a growing public health problem that has recently triggered a public health alert from the Centers for Disease Control (19). Better management of these infections, including prevention, diagnosis, and treatment, requires multipronged approaches to understand the epidemiology, basic biology, and molecular biology of this fungus. Included in these approaches are large cohort studies that employ collections of isolates from outbreaks, different geographic regions, and longitudinal studies.

C. auris has a number of characteristics that make it an intriguing pathogen. Since its emergence, it has spread quickly around the world and now is being increasingly associated with outbreaks. Its emergence was not from a single point source, instead, multiple clades appeared to have emerged simultaneously throughout the world. Although it is a member of the genus Candida, whose most important pathogenic members have as an ecological niche warm blooded mammals, the primary niche of. C. auris is unknown. Furthermore, in contrast to many pathogenic members of the genus, C. auris appears to survive very well in the environment. While numerous studies of individual isolates are vital and will continue to reveal crucial information about some of the characteristics that make this fungus so clinically important, larger studies involving many isolates have the potential to reveal important clinical information. These studies have been few in number; however, in this issue of the Journal of Clinical Microbiology, Zhu et al. have described a number of important characteristics concerning the largest United States outbreak of C. auris (20). Their study encompassed a 2-year outbreak from 2016 to 2018 in the New York area and included the analysis of more than 500 clinical isolates, more than 11,000 patient surveillance specimens, and more than 3,500 environmental surveillance specimens. The samples were collected from 59 hospitals, 92 nursing homes, 1 long-term acute care hospital, and 2 hospices (151 total facilities). MALDI-TOF MS and Sanger sequencing were used for identification and genotyping, while real-time PCR and culture were used for screening. Antifungal susceptibility testing was conducted on live cultures.

The most common clinical specimen that C. auris was recovered from in this study was the blood, which was also found in previous studies (6, 14). Genotyping results from the study revealed that a single clade predominated, which was clade I, South Asia, while clade II, East Asia, was a minor clade. Neither of the other clades, South Africa clade III or the South America clade IV, were detected, which was surprising given the number of transportation hubs (e.g., airports and bus and train stations) and the population density in the New York metropolitan area. This bias suggests differences in clade exposure; however, perhaps there are underlying factors that need to be investigated more closely, such as clade-specific differences in virulence or infection route (person-person route versus environment-person route). Importantly, clade genotyping of C. auris has typically been conducted by whole-genome sequencing, which is not possible for most laboratories; however, Zhu et al. (20) found that by using a combination of D1/D2 and ITS sequencing, the strains could be accurately identified and genotyped. Identification was found to be more accurate than either MALDI-TOF MS or Vitek MS; consequently, the sequencing approach should be investigated further as a rapid genotyping method, because laboratories that do not sequence, but can perform PCR, can send amplicons out to sequence to a variety of commercial companies at a small cost. Additionally, the ITS and D1/D2 regions were sequenced with common conserved primers (ITS1-ITS4 for the ITS region, NL1-NL4 for the D1/D2 region) as two separate amplicons; however, we have traditionally amplified these regions as a single amplicon using ITS1 and NL4 as the primer pair when we need information from both loci (21), which could make genotyping even faster and less expensive, since the typical amplicon from these two primers (∼1,200 to 1,500 bp) can be spanned in an overlapping forward and reverse sequencing run. Nonetheless, the ability to genotype without whole-genome sequencing is an important and useful observation for C. auris epidemiology. Additionally, real-time PCR assays are being increasingly approved as a diagnostic for infectious agents. In this study, a real-time PCR assay was used effectively with 98.36%, 93.32%, and 98.38% accuracy, sensitivity, and specificity, respectively, which suggests the utility of real-time PCR as a rapid screening method.

Colonization of different body sites, not just the gastrointestinal (GI) tract as is frequent with Candida sp., by C. auris was previously demonstrated, and colonization long term is not uncommon (7). Zhu et al. (20) have confirmed the axilla/groin-nare colonization and extended previous observations by noting that the axilla/groin region seems to be a preferred colonization site for the cohort in this study. However, if the nares are colonized, the burden is roughly 2-log higher than for axilla/groin colonization. The recovery of colonizing organisms from the three sites in multiple studies suggests that moisture or components in nasal secretions may be a factor in site preference. Infection from previously colonized sites would fit the Candida sp. infection paradigm. However, a potentially more troubling issue with regard to transmission and infection control of C. auris may be the resilience of C. auris outside the body, which could be atypical compared to that of other Candida sp. While many characteristics of C. auris infections are typical of Candida infections, Ku et al. have identified three distinct characteristics: frequent misidentification, an unpredictable antifungal resistance profile, and a robust ability to survive in the environment compared to other Candida sp. (15). The resilience of C. auris may be related in part to biology, as the fungus has been shown to form biofilms, grow as aggregated clusters, and to be somewhat resistant to drying (22–25). One or more of these factors, as well as yet to be identified factors, may contribute to the ability of C. auris to survive in the environment. A number of studies on C. auris disinfection have been performed; however, these studies are difficult to standardize. Currently, C. auris is known to be refractory to quaternary ammonium and other cationic disinfectants as well as UV light (26, 27). Zhu et al. (20) confirmed what others have found with regard to the presence on abiotic surfaces. C. auris was recovered in large numbers, in some cases (>10⁵ CFU), with nonporous substrates more heavily colonized than porous substrates. Fomite transmission by medical devices, including blood pressure cuffs and axillary thermometers, have also been reported (28), suggesting infection control of C. auris must include a focus on disinfection.

Finally, the clinical significance of C. auris is unquestionably drug resistance (for reviews see references 29 and 30). Of 277 clinical isolates, Zhu et al. (20) found one that was susceptible to fluconazole, while 81% had elevated MICs to voriconazole. Most were resistant to amphotericin B (61%; MIC > 2.0 μg/ml); however, none were resistant to the echinocandins, as were none of the environmental isolates. Of note, subsequent urine specimens collected from some patients showed echinocandin resistance, suggesting that these isolates can be selected for after exposure to antifungals. Perhaps more importantly, three pan-resistant isolates were identified, with one confirmed by the CDC.

It is clear that C. auris is a growing health care problem with the potential to get much worse. While the response to the emergence of this fungus has been formidable, continued progress needs to be made in the areas of rapid diagnosis, epidemiology, understanding the molecular and genetic basis of antifungal resistance, transmission and persistence in the environment, and virulence. The importance of this fungus may even reenergize antifungal discovery efforts, which may have somewhat slackened with better AIDS management. We have learned an enormous amount about this fungus in only a few years, just from the application of newer technologies, such as whole-genome sequencing. However, a telling statement by Zhu et al. (20) noted that the volume of samples that they are analyzing has now become a limiting factor in what can be investigated and understood about C. auris. Incorporating more molecular methods into clinical laboratories may provide a partial remedy, which may make it possible to glean information in real time, with some of it potentially actionable.

The views expressed in this article do not necessarily reflect the views of the journal or of ASM.