Genomic Characterizations of Clade III Lineage of Candida auris, California, USA

Candida auris is an emerging multidrug-resistant yeast. We describe an ongoing C. auris outbreak that began in October 2019 in Los Angeles, California, USA. We used genomic analysis to determine that isolates from 5 of 6 patients belonged to clade III; 4 isolates were closely related.

C andida auris was isolated from a patient in Tokyo, Japan in 2009 (1), although clinical isolates have been retrospectively identifi ed from as early as 1996 (2). Since then, bloodstream and other invasive infections caused by C. auris have been reported worldwide (3)(4)(5). Many strains of C. auris are multidrug-resistant; some strains require elevated MICs to azoles, echinocandins, and polyenes. In 2019, the US Centers for Disease Control and Prevention (CDC) listed C. auris as an urgent threat to public health (6), highlighting the need for active surveillance and appropriate infection prevention.
Whole-genome sequencing (WGS) and phylogenetic analyses have revealed >4 major clades of C. auris; each clade covers a distinct geographic area, giving C. auris a global distribution (7,8). Researchers have documented several C. auris outbreaks in the United States, mostly caused by strains belonging to clades I and IV (9). We describe several cases of C. auris colonization and infection in patients of long-term acute-care (LTAC) facilities in and around Los Angeles, California, USA.
During September 2019-September 2020, we screened 113 patients using in-house PCR selective for C. auris according to Los Angeles County Public Health and CDC guidelines (Appendix 1). Six patients tested positive for C. auris with cycle threshold (C t ) values of 22.6-39.7 (Table 1). Patient A tested positive in October 2019; patients B-F tested positive during July-September 2020.
The 6 patients were residents of 4 LTAC facilities in Los Angeles County. All 6 had a history of tracheostomy. Patients A and F had prior history of C. auris colonization; patient F had active infection of a bronchopulmonary fi stula. Patient D had C. auris and severe acute respiratory syndrome coronavirus 2 co-infection (Table 1). We cultured C. auris isolates from inguinal and axillary swab samples of patients A, C, D, and E; pleural fl uid of patient F; and tracheal aspirate of patient A. The sample from patient A produced few colonies; we treated the patient for bacterial pneumonia. We were not able to isolate C. auris from patient B (C t = 39.5).
All C. auris isolates were resistant to amphotericin B (MIC = 2 μg/mL) and fl uconazole (MIC >64 μg/ mL) but susceptible to echinocandins ( Table 2). We conducted k-mer analysis using 261 C. auris sequences available on GenBank, most of which were described previously (10) ( Appendix 2 Candida auris is an emerging multidrug-resistant yeast. We describe an ongoing C. auris outbreak that began in October 2019 in Los Angeles, California, USA. We used genomic analysis to determine that isolates from 5 of 6 patients belonged to clade III; 4 isolates were closely related. UCLA isolates belonged to clade III (Appendix 2 Table  1). We conducted a phylogenetic analysis of clade III isolates using k-mers (Appendix 1 Figure).
In the United States, researchers have identified isolates belonging to all 4 clades; although these isolates show geographic relationships (9), clade I is predominant across the country. Clade III isolates have been identified in Indiana, Texas (11), and Florida. We conducted a k-mer-based phylogenetic analysis of C. auris isolates in the United States ( Figure). SNP analysis showed that 5 of the UCLA isolates were closely related (3-12 SNPs); isolate F1 was genetically distinct (77-79 SNPs). All 6 isolates were distinct from isolates from Indiana (65-139 SNPs) and Florida (47-117 SNPs) (Appendix 1 Table 1).
The F126L mutation appears to be exclusive to clade III (10). These findings are consistent with results of antifungal susceptibility testing, which showed that all isolates were resistant to fluconazole ( Table 2). Sequences of fks1 were identical in 5 isolates (A1, A2, C1, D1, E1), with 99.9% pairwise nucleotide identity to the reference (GenBank accession no. CP043531); these isolates had 1 amino acid substitution: I1572L (Appendix 1 Table 2). Isolate F1 had the same substitution in addition to I1095L. All isolates had a wildtype serine at aa 639; mutations at this location are linked to echinocandin resistance in C. auris (13). All isolates were susceptible to caspofungin, micafungin, and anidulafungin.

Conclusions
To identify and prevent the spread of C. auris in this hospital system, we used an in-house PCR to screen patients for this pathogen. WGS of isolates from patients transferred from LTAC facilities revealed that these isolates are closely related, suggesting an ongoing outbreak with community spread in the Los Angeles area. The isolates described here were all resistant to fluconazole and amphotericin B but susceptible to echinocandins according to the CDC tentative breakpoints (https://www.cdc.gov/fungal/candida-auris/c-aurisantifungal.html). In addition, all isolates had an F126L mutation in the erg11 gene, which is unique to clade III strains and associated with fluconazole resistance (10).
Patient D was admitted to an SNF after complications from pneumonia caused by coronavirus disease (COVID-19). Few cases of C. auris and COV-ID-19 co-infection have been reported (14,15). After COVID-19 infection, patient D had multiple complications requiring a tracheostomy and enteral feeding tube; the patient was subsequently transferred to an LTAC for rehabilitation. A substantial portion of adult patients who recover from severe COVID-19 have long-term sequelae and might require admission to SNFs or LTACs. Therefore, the COVID-19 pandemic might lead to increased transmission of C. auris in SNFs because of increased admissions and shortages of personal protective equipment. During critical shortages, CDC guidelines permit extended use of isolation gowns for patients who are known to be infected with the same infectious disease if there are no additional known coinfections transmitted through contact (https://www.cdc. gov/coronavirus/2019-ncov/hcp/non-us-settings/ emergency-considerations-ppe.html#ppe-specificstrategies). To encourage appropriate use of personal protective equipment and prevent transmission, it is essential that facilities screen patients for C. auris.
One limitation of this study is the lack of additional epidemiologic history of the patients, especially in the context of travel-related exposures. The ability to track cases to a location with known outbreaks of clade III C. auris strains is essential to determining the origin of the current outbreak. Further investigation is needed to explain why patient F had a genetically distinct isolate, suggesting a separate introduction.
In conclusion, we identified a unique clade III C. auris strain in an ongoing outbreak in LTAC facilities since 2019. These findings indicate active community spread of multidrug-resistant C. auris in the Los Angeles area.