We complemented qPCR-based wastewater surveillance with culturing and subsequent whole-gene sequencing of recovered isolates. Initial attempts with the centrifugation-based method previously used for southern Nevada wastewater (14) were unsuccessful (Table 1). As such, we explored a filtration-based alternative that was observed to be superior to our original method across 2 split samples from Nevada (Appendix Figure 1). Using that improved method, we were able to recover 15 C. auris isolates from 2 samples consecutively collected on March 28, 2023, and April 4, 2023 (Table 1). All wastewater isolates belonged to clade III and segregated topologically into 2 individual subgroups distinctly separated by collection date (Figure 3). Isolates within the subgroup linked to the March 28 wastewater sample were highly related to the clinical isolate available for patient 2 (0-6 SNPs), who was transferred to St. George on March 23. The isolates within the subgroup linked to the April 4 sample displayed ≈12 SNP differences from the patient 2 isolate (Figure 3). Unfortunately, no clinical whole-genome sequencing (WGS) data were available for patient 3 because the patient's colonization status was not determined in Nevada and the positive clinical sample collected in Utah was not submitted to UPHL. Patient 1 was infected with a clade I strain, but we were unable to culture C. auris in any wastewater samples collected during the time of the patient's stay at SNF A or before March 28, 2023 (Table 1). As such, we were unable to study the contribution of clade I isolates to the overall C. auris signal in the St. George sewershed.
The recent history of C. auris in the United States highlighted the challenges in controlling the pathogen after it becomes established in an area (7). Early detection strategies coupled with aggressive infection control measures could reduce its effect on healthcare facilities and the general population. In that respect, wastewater-based surveillance is a promising tool for detecting pathogens that are circulating in the population at very low prevalence and have not been overtly manifested at the clinical level (12).
The application of wastewater-based surveillance to the C. auris problem is in its infancy (14-16), so fundamental parameters that will guide its use have not yet been studied in detail. For example, levels of C. auris shedding in human excreta and body site densities during colonization have not yet been adequately characterized. C. auris effectively colonizes skin and nares (26,31) but is not typically recovered from the buccal mucosa (31,32). A regular nylon swab can usually recover 10-10 CFU of C. auris from various skin sites (e.g., palms/fingertips, toe web, perianal skin, axilla, inguinal crease, neck) and nares (31). As such, it is conceivable that the organism could be released in great numbers into the sewershed via skin shedding during routine hygiene practices or even when laundering items that have been in contact with a colonized person. Because skin is the primary C. auris colonization site, clinical studies aimed at determining the actual bioburden released through hygiene practices (33) will be essential for assessing quantitative measurements in wastewater. C. auris is also commonly recovered from urine (4,26) of patients with candiduria (34), as well as from asymptomatic persons (35). C. auris is less frequently recovered from fecal samples but has been recovered via rectal swabbing (26,32,36). Of note, a correlation between gut colonization and urinary tract infections has been observed in cohorts of patients affected by C. auris (36).
As has been accomplished for wastewater-based surveillance of SARS-CoV-2, additional modeling and parameterization are needed to fully characterize relationships between incidence/prevalence and expected wastewater concentrations (28,37). Early attempts to establish those correlations for C. auris have been extremely challenging (16). Nevertheless, the qPCR data and the excreta-only model used in our study fit very well with the clinical course of patient 1 (the putative introduction event in St. George) and indicate that detecting 1 C. auris shedder to a community-scale wastewater system of moderate size is plausible (Figure 2). Although our study focused on early detection of pathogen introduction, future studies should consider monitoring wastewater C. auris loads after the population presumably returns to a zero-infection status.
Recovery of C. auris in culture has been instrumental in obtaining isolates for molecular epidemiology analyses by WGS (14). The incorporation of WGS into wastewater-based surveillance systems for C. auris should be universally adopted to understand the origin of introduction events as well as the evolving diversity of contributions to sewersheds. Motivated by the initial inability to recover C. auris isolates in St. George (Table 1), we worked at improving our original culture method by changing the sample concentration step from centrifugation to membrane filtration (Appendix Figure 1), an approach also recently used by Babler et al. (16). Yet, culture from wastewater samples remains a highly variable endeavor, possibly because of variable competition from other species of fungi or fluctuations in environmental factors within the sewer environment affecting the growth of C. auris, such as dissolved oxygen concentration (38,39). In addition, our broth enrichment approach remains unsuitable for isolating fluconazole-susceptible isolates (14).
WGS analysis indicated a close relationship between C. auris wastewater isolates and 1 isolate from patient 2 (Figure 3). When those wastewater samples were collected, both patients 2 and 3 were potentially contributing C. auris to the St. George WWTP (Figure 1, panel C). With the data available, we cannot discriminate whether the genetic diversity of the wastewater isolates collected on 2 separate dates encompasses shedding from patient 3 or other unidentified colonized persons. Moreover, mixed colonization consisting of clones separated by SNP distances greater than those displayed in Figure 3 is not unusual (40,41) and represents another layer of complexity in the interpretation of molecular epidemiology analyses for C. auris (42). However, incorporating WGS analyses into C. auris wastewater surveillance would still be invaluable for detecting contributions from strains belonging to different clades or displaying very large SNP distances. In addition, if C. auris strains can persist in sewer pipes as biofilm (a phenomenon not yet investigated for this organism) (38), WGS could potentially distinguish persistent signals from a new shedding event.
In conclusion, we used a holistic approach to C. auris wastewater-based surveillance that entailed using qPCR as the main testing method as well as culture and WGS to better characterize the source of the molecular signals. After being proven effective, metagenomic approaches could potentially bypass the need for culture (43). We believe that our case study illustrates the potential of wastewater-based surveillance to be a sufficiently sensitive method for discovering C. auris transmission at early stages of introduction into a community.