Prospective study of Candida auris nucleic acids in wastewater solids in 190 wastewater treatment plants in the United States suggests widespread occurrence

ABSTRACT

Candida auris is an emerging, multidrug-resistant fungal pathogen that poses a significant public health threat in healthcare settings. Despite yearly clinical cases rapidly increasing from 77 to 8,131 in the last decade, surveillance data on its distribution and prevalence remain limited. We implemented a novel assay for C. auris detection on a nationwide scale prospectively from September 2023 to March 2024, analyzing a total of 13,842 samples from 190 wastewater treatment plants across 41 U.S. states. Assays were extensively validated through comparison to other known assays and internal controls. Of these 190 wastewater treatment plants, C. auris was detected in the wastewater solids of 65 of them (34.2%) with 1.45% of all samples having detectable levels of C. auris nucleic-acids. Detections varied seasonally, with 2.00% of samples positive in autumn vs 1.01% in winter (P < 0.0001). The frequency of detection in wastewater was significantly associated with states having older populations (P < 0.001), sewersheds containing more hospitals (P < 0.0001), and sewersheds containing more nursing homes (P < 0.001). These associations are in agreement with known C. auris epidemiology. This nationwide study demonstrates the viability of wastewater surveillance for C. auris surveillance and further highlights the value of wastewater surveillance when clinical testing is constrained.

IMPORTANCE

This study highlights the viability of wastewater surveillance when dealing with emerging pathogens. By leveraging an existing framework of wastewater surveillance, we reveal the widespread presence of C. auris in the United States. We further demonstrate that these wastewater detections are consistent with demographic factors relevant to C. auris epidemiology like age and number of hospitals or nursing homes. As C. auris and other pathogens continue to emerge, the low-cost and rapid nature of wastewater surveillance will provide public health officials with the information necessary to enact targeted prevention and control strategies.

INTRODUCTION

Healthcare-associated infections are rapidly emerging as a global public health threat due to increasing levels of antimicrobial resistance and an aging worldwide population (1–3). Among these infections, Candida auris, first identified in 2009 in Japan, is an opportunistic, drug-resistant fungal pathogen with significant levels of associated morbidity and mortality (3–6). It has now spread to over 40 countries and has been designated as a priority pathogen by both international organizations, such as the World Health Organization, and national organizations, such as the Center for Disease Control and Prevention (CDC) in the United States, where it is now a nationally notifiable condition (4, 6, 7). The estimated crude mortality rate within the United States for C. auris infections was 34% between 2017 and 2022 (5).

C. auris prevention and control efforts are exacerbated by the fungus’ hardiness on fomites and high levels of transmissibility (6–8). The fungus has been found to persist for at least 7 days on both moist and dry surfaces, increasing the frequency and chance of transmission events in healthcare settings (9, 10). Of particular concern are high-touch plastic surfaces within healthcare facilities, where C. auris was found to be viable for up to 14 days (10). In addition, C. auris is resistant to many common disinfectants, so proper decontamination protocols are difficult and time intensive in already strained healthcare facilities (11). Further complicating containment efforts are the different presentations in colonized individuals compared to those with acute infections, with most colonized individuals presenting as asymptomatic (4). A lack of screening in high-risk patients and misidentification of specimens as other Candida subspecies has also contributed to the rapid spread of C. auris (5, 12). In the United States, clinical cases rose from 77 between 2013 and 2016 to 8,131 in 2022 and are projected to continue rising (5, 13).

Despite this tremendous increase in cases and the accompanying screening efforts, clinically available data are still sparse, with many institutions not speciating Candida cases resulting in underreporting cases in long-term care facilities and nursing homes (13, 14). Many of these facilities do not have the necessary equipment or human capital to implement speciation testing and screening, which has been shown to be a necessary part of successful containment efforts (12). Alternative approaches to clinical surveillance are therefore necessary to better track both the spread and severity of outbreaks. Wastewater represents a naturally composite sample which is particularly useful for disease surveillance as it contains urine, feces, vomit, saliva, sloughed skin cells, sputum, and other excretions. Wastewater surveillance has been successfully used to track a wide variety of human pathogens including enterovirus, SARS-CoV-2, influenza A virus, and flavivirus (15–17). Like these viruses, C. auris has been detected in urine and stool samples from colonized or infected individuals, though it is typically diagnosed through the presence in the blood or dermis of infected patients (4, 10, 18). Furthermore, recent studies have demonstrated the persistence and subsequent detection of C. auris in wastewater samples in Nevada and Florida (19). Early detection of this emerging pathogen through environmental surveillance may complement clinical surveillance efforts and inform containment and screening procedures (20).

In this paper, we present the results of a United States-wide wastewater monitoring effort for C. auris and compare the data to recent, publicly available clinical case data. We also compare results to available demographic indicators and healthcare facility locations. The detections of C. auris in wastewater demonstrate wide geographical occurrence, and we demonstrate that these wastewater detections appear to be associated with age and the presence of hospitals and nursing homes in adjacent communities. These results demonstrate the advantages of using wastewater surveillance for the detection of emerging pathogens and for supplementing clinical testing efforts.

MATERIALS AND METHODS

Reverse transcription-PCR assays

We designed a novel probe for C. auris to use in conjunction with previously published specific and sensitive forward and reverse primers (21). The assay targets the region of the genome encoding the 5.8S ribosomal RNA, all of ITS2, and a fragment of 28S ribosomal RNA. To design the probe, C. auris genome sequences were downloaded from the National Center for Biotechnology Information (NCBI) and aligned to identify conserved regions located between the primers. The probe was designed in silico using Primer3Plus (https://primer3plus.com/). Parameters used in the design process (e.g., sequence length, GC content, and melt temperatures) are provided in Table S1. The primers and probe were then screened for specificity in silico by blasting the sequences in NCBI and in vitro against fungal panels or cDNA sequences. The in silico analysis was performed by excluding all C. auris genomes from a BLAST search, thereby only giving off-target matches (22).

The primers and probe were tested in vitro for specificity and sensitivity using a fungal panel (NATCTVPOS-BD, Zeptomatrix Buffalo, NY) and C. auris Satoh et Makimura (ATCC MYA-5001) purchased from American Type Culture Collection (ATCC, Manassas, VA). The fungal panel NATCTVPOS-BD includes chemically inactivated Candida albicans, Candida krusei, Candida glabrata, and Trichomoniasis vaginalis. Nucleic acids were extracted from fungi using Chemagic Viral DNA/RNA 300 Kit H96 for Chemagic 360 (PerkinElmer, Waltham, MA).

Nucleic acids were used undiluted as template in digital droplet reverse transcription PCR (ddRT-PCR) singleton assays for sensitivity and specificity testing in single wells. The concentration of targets used in the in vitro specificity testing was between 10³ and 10⁴ copies per well. Negative RT-PCR controls were included on each plate.

We used an RT-PCR assay for the in vitro sensitivity and specificity testing as the goal was to ultimately multiplex the C. auris assay in a panel that includes RNA-genome viruses. The implications of the use of RT-PCR and a comparison to results obtained using PCR are further investigated, as described below. Regardless, this means the assay detects both DNA and RNA targets.

Clinical surveillance data

The National Notifiable Disease Surveillance System (NNDSS) is a nationwide collaboration in which all levels of healthcare providers share health information about nationally notifiable infectious and noninfectious diseases (23). Case reports are compiled on a weekly basis using uniform case definitions from 50 states, the District of Columbia, New York City, and five territories (American Samoa, Guam, Northern Mariana Islands, Puerto Rico, U.S. Virgin Islands) (23). For this study, we used publicly available information on NNDSS for C. auris from 11 September 2023 to 1 March 2024 (https://data.cdc.gov/api/views/x9gk-5huc/rows.csv?accessType=DOWNLOAD). Dates of case determination varied but included the following in order of preference: date of disease start, date of diagnosis, date of laboratory result, date of first report to public health system, or date of state report (24). States that did not report clinical case data were omitted from analyses. It should be noted that this database is known to be incomplete as many local jurisdictions do not submit data; additional data on C. auris cases or Candida cases may be available from other local or regional sources that are not publicly available or currently updated.

Wastewater data: sample collection

Wastewater measurements were made prospectively as part of a wastewater surveillance program. Between 11 September 2023 and 1 March 2024, wastewater samples (either 24-hour composited influent or grab samples from the primary clarifier, Table S1) were collected by wastewater treatment plant (WWTP) staff using sterile containers. Samples were typically obtained approximately three times per week and shipped overnight to the laboratory at 4°C where they were processed immediately with no storage. Samples were collected from 190 distinct WWTPs across a total of 41 states over the course of the study period (Fig. 1). A total of 13,842 samples were collected and analyzed as part of this study.

Fig 1.

Location of WWTPs with C. auris DNA data included in the paper. Each blue dot represents the location of a WWTP.

Wastewater data: pre-analytical processing

C. auris is expected to be associated with the solid phase of wastewater based on previous empirical measurements across liquid and solid phases in wastewater and its size (diameter = 2.5–5 µm), so in this study, we made measurements in the solid phase of wastewater (25, 26). Details of the isolation of solids from the samples are provided in other peer-reviewed publications (27). In short, samples were centrifuged to dewater the solids. One aliquot of dewatered solids was used for nucleic-acid extractions and another was used to determine the dry weight of the solids using an oven (27). Approximately 75 mg of dewatered solids was added per milliliter of DNA/RNA shield (Zymo, Irvine, CA) spiked with bovine coronavirus (BCoV) vaccine as a positive extraction control. This mixture of solids and buffer was previously shown to have minimal inhibition (27). The mixture was homogenized after the addition of grinding balls, and centrifuged; the supernatant was used immediately for nucleic acid extractions. Nucleic acids were extracted from 6 to 10 replicate aliquots of the supernatant using the Chemagic Viral DNA/RNA 300 kit H96 for the Perkin Elmer Chemagic 360 (Perkin Elmer, Waltham, MA) followed by PCR inhibitor removal with the Zymo OneStep-96 PCR Inhibitor Removal kit (Zymo Research, Irvine, CA). Whether six or 10 replicates were used is specified in Table S1. The suspension volume entered into the nucleic-acid extraction process was 300 µL and 50 µL of nucleic acids were retrieved after the inhibitor removal kit. Negative extraction controls consisted of BCoV-vaccine spiked DNA/RNA Shield. Nucleic acids were used immediately as template in PCR as described next, with no storage.

Wastewater data: analytical processing

We used droplet digital reverse transcription PCR (ddRT-PCR) to measure nucleic acids in wastewater. We measured pepper mild mottle virus (PMMoV) as an endogenous positive control as concentrations tend to be extremely elevated in wastewater samples; we measured BCoV as an exogenous, spiked-in control (27–29). Methods applied to wastewater solids to measure PMMoV and BCoV in a duplex reaction are provided in detail elsewhere (27). The C. auris assay (Table 1) was run in multiplex using a probe-mixing approach. The exact assays that C. auris was multiplexed with varied slightly over the duration of the project for some of the plants (information is provided in Table S2). Multiplexing C. auris with these targets did not affect C. auris quantification (see SI Fig. S1). The results from those other assays are not provided herein. The PMMoV/BCoV duplex and C. auris multiplex assays were run on 96-well plates. Each 96-well PCR plate of wastewater samples included PCR-positive controls for each target assayed on the plate in one well, PCR negative no template controls in two wells, and extraction negative controls in two wells. PCR-positive controls consisted of C. auris DNA. Each of the 6 or 10 replicate nucleic-acid extracts were run in their own wells to measure C. auris. Two randomly selected or 10 replicate nucleic-acid extracts were run in their own wells to measure PMMoV and BCoV (Table S1).

TABLE 1.

Primers and probes used in this study for detection of C. auris nucleic acids^a

Primer or probe	Sequence
Forward primer	CGCACATTGCGCCTTGGGGTA
Reverse primer	GTAGTCCTACCTGATTTGAGGCGAC
Probe	CTTCTCACCAATCTTCGCGGT

^a Primers and probes were purchased from Integrated DNA Technologies (Coralville, IA, USA). The probes contained fluorescent molecule HEX and quenchers (5′ HEX/ZEN/3′ IBFQ); HEX, hexachloro-fluorescein; ZEN, a proprietary internal quencher from Integrated DNA Technologies (Coralville, IA, USA); and IBFQ, Iowa Black FQ. Amplicon size is 163 base pairs.

ddRT-PCR was performed on 20-µL samples from a 22 µL-reaction volume, prepared using 5.5 µL of template mixed with 5.5 µL of One-Step RT-ddPCR Advanced Kit for Probes (Bio-Rad 202 1863021), 2.2 µL of 200 U/µL Reverse Transcriptase, 1.1 µL of 300 mM dithiothreitol, and primer and probe mixtures at a final concentration of 900 nM and 250 nM, respectively.

Primer and probes for assays were purchased from Integrated DNA Technologies (IDT, San Diego, CA; Table 1). C. auris was measured in reactions with undiluted template, whereas PMMoV and BCoV were run on template diluted 1:100 in molecular grade water. It is important to note that an RT step was used because during the prospective study, we were measuring a large number of RNA genome viruses. Therefore, the C. auris target quantified should be interpreted as copies of RNA plus DNA.

Droplets were generated using the AutoDG Automated Droplet Generator (Bio-Rad, Hercules, CA). PCR was performed using Mastercycler Pro (Eppendforf, Enfield, CT) with the following cycling conditions: reverse transcription at 50°C for 60 minutes, enzyme activation at 95°C for 5 minutes, 40 cycles of denaturation at 95°C for 30 seconds and annealing and extension at 59°C (for C. auris) or 56°C (for PMMoV and BCoV) for 30 seconds, and enzyme deactivation at 98°C for 10 minutes then an indefinite hold at 4°C. The ramp rate for temperature changes was set to 2°C/second, and the final hold at 4°C was performed for a minimum of 30 minutes to allow the droplets to stabilize. Droplets were analyzed using the QX200 or the QX600 Droplet Reader (Bio-Rad). A well had to have over 10,000 droplets for inclusion in the analysis. All liquid transfers were performed using the Agilent Bravo (Agilent Technologies, Santa Clara, CA).

Thresholding was done using QuantaSoft Analysis Pro Software (Bio-Rad, version 1.0.596) and QX Manager Software (Bio-Rad, version 2.0). Replicate wells were merged for analysis of each sample. In order for a sample to be recorded as positive, it had to have at least three positive droplets. Concentrations of nucleic acid targets were converted to concentrations in units of copies (cp) per gram dry weight using dimensional analysis. The total error is reported as SDs. Three positive droplets across six merged wells correspond to a concentration ~1,000 cp/g; the range in values is a result of the range in the equivalent mass of dry solids added to the wells. This represents the lowest detectable concentration using these analytical and pre-analytical methods and is used as a lower detection limit. Data collected as part of the study are available from the Stanford Digital Repository (https://purl.stanford.edu/hv291tt0888).

Wastewater: confirmatory analyses

To further validate and confirm the performance of our novel C. auris assay, we carried out additional analyses on a subset of the samples. We ran a second, published C. auris PCR assay on a subset of samples, and we sequenced amplicons obtained using our primers. A small subset of samples for this purpose is justified as these ancillary analyses are meant to provide additional validation; further research could expand on this work.

We selected seven samples (Table S3) with relatively high concentrations of C. auris, as measured above, to test another published C. auris assay by Barber et al. (25) for detecting C. auris in wastewater. The assay used by Barber et al. targets the same region of the C. auris genome as the assay in Table 1 but uses different primers and probes (Table S4) (25). Archived nucleic acid extracts, stored at −80°C, from those samples, were thawed and then run as a template using the Barber et al. assay as well as the assay described herein (Table 1). The samples were run in duplex with an assay that targets the N gene of SARS-CoV-2 (27). C. auris was run using a probe labeled with fluorescein amidite. The methods described above for the analytical procedures (number of replicate wells and negative and positive controls) were used.

These same seven samples were assayed with and without the reverse transcription step. Nucleic-acid extracts were thawed and run exactly as described for the prospective study except that the RT, and dithiothreitol were not added to the mastermix.

We selected a subset of four of the seven samples (Table S3) for Sanger sequencing of the PCR amplicon. One microliter of archived (at −80°C) nucleic acid extracts was used as the template in a 50 µL PCR prepared according to the manufacturer’s protocol with Q5 High-Fidelity 2X Master Mix (New England Biolabs, Ipswich, MA) using the primers in Table 1. The thermocycler program was run at an annealing temperature of 67°C for 30 seconds and extension at 72°C for 30 seconds. C. auris DNA was used as a positive control. PCR products were run on a 2% agarose gel stained with SYBR Safe gel stain (Thermofisher, Fremont, CA), and amplicons of the expected size (259 bp) were excised and gel purified using Zymo gel DNA recovery kit (Irvine, CA). The purified DNA was then sent to Molecular Cloning Laboratories (South San Francisco, CA) for Sanger sequencing. Sequences were downloaded and aligned with C. auris genomes using Snapgene software.

Demographic data

C. auris hospitalizations are most common among older adults (5). We used 5-year estimates from the 2022 American Community Survey at the state level to identify the median age of states with participating WWTPs (n = 41 states) (30).

Hospital and nursing home data

C. auris is particularly a concern in hospitals and nursing homes (5, 14). We obtained the locations of hospitals and nursing homes in the United States from the U.S. Department of Homeland Security’s Homeland Infrastructure Foundation-Level Data (HIFLD) database (https://gii.dhs.gov/HIFLD) (31, 32). We selected all types of hospitals and nursing homes with an “open” status and then determined the number of hospitals and nursing homes in each sewershed using the Tabulate Intersection geoprocessing tool in ArcGIS Pro (version 3.1.1). Sewershed boundaries were provided by most WWTPs (n = 115). For WWTPs that did not provide a sewershed (n = 75), we approximated sewershed boundaries based on the zip codes serviced by the WWTP and USA ZIP Code Boundaries geospatial data set published by Esri (Source: TomTom, US Postal Service, Esri) (33).

Data analysis

The use of an RT-PCR assay, allowing the detection of both RNA and DNA, should increase the sensitivity of the assay (by detecting both nucleic acids) yet perhaps complicate the interpretation of measured concentrations. For this reason, we only use the wastewater data in the presence/absence format. As such, the primary outcome of the wastewater analyses is the percentage of samples that are positive for C. auris (hereafter, percent positive detections).

Percentage positive detections of C. auris were calculated by counting all positive observations for a single WWTP and dividing by the total number of observations. Wastewater measurements in each state were aggregated by Morbidity and Mortality Weekly Report (MMWR) week, and we calculated the percentage of wastewater measurements that were positive for C. auris each week.

Spearman’s rank correlation coefficients were used to assess the association between weekly (MMWR weeks) publicly available NNDSS clinical data and wastewater percent positive detections of C. auris for each state (n = 18 states with wastewater and clinical data). Case data from NNDSS, available on a weekly basis for each state, were used after adjusting for population (cases per million people; Fig. 2a).

Fig 2.

Map showing the percentage of positive wastewater samples and population normalized case numbers from NNDSS across the United States. Figure 2a shows the normalized number of cases per million in each state as reported to NNDSS during the study period. Figure. 2b shows the percentage of positive wastewater samples in each of the states surveilled during the study period. Yellow represents low normalized values, while black represents high normalized values. Gray states represent areas without available data.

We tested whether concentrations measured by the assay presented in Table 1 were different from those measured by the Barber et al. assay using RT-ddPCR on nine representative samples (Table S3; Fig. S2) (25). We ran the same samples using RT-ddPCR and ddPCR with no RT step (Table S3; Fig. S2). Differences between these controls were assessed using a paired t test.

We also assessed the relationship between C. auris percent positive detections and several other contextual and demographic factors. We explored seasonality by assessing C. auris percent positive detections in wastewater samples collected in fall (defined as September 1 to November 30) and winter (defined as December 1 to February 29). We then assessed the association between C. auris percent positive detections and age at the state level. We grouped WWTPs by those in states with a median age less than vs greater than or equal to the national median age (39 years) (30). Lastly, we assessed the association between C. auris percent positive detections and hospitals and nursing homes at the sewershed level. We grouped WWTPs by those with a sewershed intersecting (i) less than vs greater than or equal to the median number of hospitals among all sewersheds (median: one hospital) and (ii) less than vs greater than or equal to the median number of nursing homes among all sewersheds (median: seven nursing homes). For seasonality, we compared the percentage of wastewater samples positive for C. auris in each season. For age, hospitals, and nursing homes, we compared the percentage of wastewater samples that were positive for C. auris across WWTPs in the less than median group vs the greater than or equal to median group. The null hypothesis tested for each of these variables was that there were no differences between the two groups.

Statistical significance and 95% confidence intervals between the groups described above were assessed through bootstrapping (34, 35). For each variable and group, a distribution of values was generated through random sampling with replacement 10,000 times. Significance was then calculated using the proportion of values below our calculated statistic (i.e., whether the calculated statistics were above the 95% confidence intervals). This same method was used when assessing differences between assays (Fig S2) and seasonal differences in detection. In those cases, grouping was done based on assay and season, respectively.

RESULTS

Quality assurance/quality control (QA/QC)

The C. auris assay was found to be sensitive and specific. The in silico analysis indicated that no primer or probe sequence was more than an 85% match to off-target matches, and there was no overlap found between combinations of forward primer off-targets, reverse primer off-targets, and probe off-targets. These factors indicate a high level of specificity in the primer-probe combination used. The in vitro testing confirmed the sensitivity and specificity of the C. auris assay as C. auris nucleic acids were positive while the remaining non-target nucleic acids were negative. Tests against the fungal panel NATCTVPOS-BD, which includes chemically inactivated C. albicans, C. krusei, C. glabrata, and T. vaginalis, showed no amplification using the C. auris assay described.

Results are reported following the Environmental Microbiology Minimal Information (EMMI) guidelines (see SI and Fig. S3). (36) All positive and negative controls performed as expected, indicating acceptable assay performance. Figure S1 demonstrates that the assay has no cross-reactivity with the multiplexed targets, as all C. auris negative controls remained negative in the presence of positive controls of the other seven targets. Median (interquartile range [IQR]) BCoV recoveries across all wastewater samples were 1.08 (0.82, 1.43), indicating good recovery across all samples. Recoveries exceeding one are the result of uncertainties in the measurement of BCoV added to the buffer matrix. PMMoV levels were elevated in all samples indicating a lack of gross extraction failures (median = 4.8 × 10⁸ cp/g, min = 5.0 × 10⁵ cp/g, max = 2.02 × 10¹¹ cp/g).

As a further validation step, C. auris was quantified in a subset of samples also using the Barber et al. assay (Fig. S2) (25). Concentrations measured by the two assays were closely matched, with no statistically significant differences (P > 0.05). Comparisons between concentrations measured using RT-ddPCR and ddPCR showed similar results (Fig. S2); no significant differences were found (P > 0.05). Additionally, the PCR amplicons obtained using the forward and reverse primers were sequenced, and the sequences matched the C. auris genome segment targeted (Fig. S4). These ancillary analyses provide further confirmation of the assay’s specificity. Additional comparisons or sequencing may be completed in the future to satisfy other research objectives.

National overview

The study period spanned 11 September 2023 to 1 March 2024. Wastewater data were available from 190 distinct WWTPs in 41 states (Table S1). The number of WWTPs in these states ranged between 1 and 57 WWTPs (median: 2 WWTPs), with population coverage of the sewersheds ranging from 0.13% to 59.5% of the population of each individual state (median: 5.75%). The number of wastewater samples collected and analyzed from individual WWTPs ranged between 23 and 173 samples (median: 72 samples) during the study period. In total, 13,842 wastewater samples were collected and analyzed across all WWTPs during the study period. C. auris concentrations in these samples ranged from below the limit of detection (approximately 1,000 cp/g) to 1,456,751 cp/g, with 1.45% (n = 200 samples) of samples above the detection limit. Sixty-five (34.21%) WWTPs from 26 distinct states had at least one detectable C. auris concentration during the study period, and the percentage of positive samples from these WWTPs ranged between 0.58% and 46.48% (median: 1.49%). Time series heatmaps of C. auris percent positive detections at each of the 190 WWTPs are provided in Fig. S5. Seasonally, 2.00% (n = 6,754) of observations were positive during autumn (defined as September 1 to November 30), and 1.01% (n = 7245) of observations were positive during winter (defined as December 1 to February 29). This difference was statistically significant when assessed by a bootstrapping approach (P < 0.0001).

Clinical data from the NNDSS were downloaded for comparison to wastewater data, and no significant correlations were found between wastewater percent positive detections and population-adjusted clinical case rates (Spearman’s rho between −0.10 and 0.23 for the 18 states, all P > 0.05). Several states had obvious divergence between wastewater positivity rates and population-adjusted NNDSS clinical case rates (Fig. 2). For example, in Maine, 13.3% of wastewater samples were positive for C. auris despite not reporting any clinical cases during the study period to NNDSS (Fig. 2). Twelve states that did not report C. auris cases to the NNDSS were found to have positive detections of C. auris. Figure 3 shows the percentage of positive detections as a function of population-normalized clinical cases and highlights the limited clinical data available.

Fig 3.

Comparison of wastewater percent positive detections and total clinical cases as reported by NNDSS. Only states with clinical data reported to NNDSS are included. Each labeled point represents the percentage of positive wastewater samples in that state and the number of cases reported to NNDSS. The state abbreviation is above each data point.

Population demographics and healthcare facility locations and C. auris

WWTPs were grouped using three separate statistics: the median age of the state, number of hospitals in the sewershed, and number of nursing homes in the sewershed (Fig. 4). The median age in the United States is 39 years as of the latest American Community Survey (ACS) data (30). Seventy-eight WWTPs were located in a state with a median age at or above the national median age; 112 WWTPs were located in a state with a median age below the national median age. As shown in Fig. 4a, among WWTPs in states with a median age at or above the national median age, 2.03% (95% CI: 0.80–3.50) of all wastewater samples were positive for C. auris. Among WWTPs in states with a median age below the national median age, 0.83% (95% CI: 0.34–1.38) of all wastewater samples were positive for C. auris. This difference was statistically significant when assessed by bootstrapping (P < 0.001).

Fig 4.

Comparison of wastewater percent positive detections across different groups. *** indicates P < 0.001 and **** indicates P < 0.0001. Figure 4a shows the percentage of positive samples for states at or above the median age compared to those below. Figure 4b shows the percentage of positive samples when grouped by number of hospitals at the sewershed level. Figure 4c shows the same statistic when grouped by number of nursing homes, again at the sewershed level. Error bars represent 95% confidence intervals derived from the bootstrapping approach.

The median number of hospitals in the sewersheds was 1 hospital (range: 0–74 hospitals). WWTPs were grouped by whether they contained zero hospitals (fewer than the median, n = 54) or one or more hospitals in their sewersheds (equal to or greater than the median, n = 136). We found that 1.78% (95% CI: 0.92–2.90, n = 10,019) of samples from WWTPs with at least one hospital in their sewersheds were positive for C. auris whereas 0.62% (95% CI: 0.34–1.11, n = 3,819) of samples from WWTPs with no hospital in their sewershed were positive for C. auris (Fig. 4b). This difference was significant when assessed by bootstrapping (P < 0.0001).

The median number of nursing homes in the 190 sewersheds was 7 (range: 0–420 nursing homes). WWTPs were grouped by whether they contained 0–6 nursing homes (fewer than the median, n = 91) or seven or more nursing homes in their sewersheds (equal to or greater than the median, n = 99). We found that 1.92% (95% CI: 0.89–3.46, n = 7,544) of samples from WWTPs with seven or more nursing homes in their sewersheds were positive for C. auris, whereas 0.87% (95% CI: 0.47–1.47, n = 6,294) of samples from WWTPs with 0–6 nursing homes in their sewersheds were positive for C. auris (Fig. 4c). This difference was significant when assessed by bootstrapping (P < 0.001).

DISCUSSION

Previous studies at the sewershed scale have shown that the detection of C. auris in wastewater is indicative of ongoing outbreaks in the contributing population (19, 25). To our knowledge, this study represents the first nationwide wastewater monitoring effort of C. auris and compares it to national clinical case data (23). We first demonstrate the sensitivity and specificity of the assay used in this study through comparisons to published assays, sequencing (Fig. S2 and S4), and in silico analyses. The widespread detection of C. auris in wastewater suggests a significant gap in clinical case data reported to the NNDSS. Indeed, it is known that many local jurisdictions do not provide data for inclusion in NNDSS. We found that C. auris detections are more frequent in states with older populations and in sewersheds with higher numbers of nursing homes and hospitals. These findings are consistent with C. auris infections being more common in older, more susceptible populations and further support the assertion that clinical case data may be incomplete.

C. auris has gone from initial detection in the United States in 2013 to being declared an “Urgent” and notifiable pathogen in 2019 by the United States CDC (14, 20, 37). The rate of testing volume and clinical positives has rapidly increased, with 17 states identifying their first C. auris cases between 2019 and 2021 (14). Despite this, clinical testing for C. auris is far from ubiquitous and not conducted uniformly across the United States (14, 38). On-site PCR testing of clinical specimens is rare, and samples often have to be sent to public health or reference laboratories, further increasing the cost of testing and the turnaround time before the data reaches public health officials (38). The detection of C. auris nucleic acids in wastewater collected throughout the United States demonstrates a need for increased clinical testing and reporting capacity. It also highlights the capability of wastewater testing to serve as a sentinel surveillance mechanism for emerging pathogens.

C. auris is an opportunistic pathogen that often infects the elderly and immunocompromised within long-term healthcare facilities (10, 20, 39). Following our detection of C. auris, we investigated potential demographics and healthcare facilities related to its pathology using data from the ACS and U.S. Department of Homeland Security’s HIFLD database (30, 40). We analyzed age, number of hospitals, and number of nursing homes as plausible factors impacting the presence of C. auris within a sewershed or state. We found significant differences between the percentage of positive detections of C. auris in states with a median age less than vs greater than or equal to the national median age. Similarly, we found significant differences between the percentage of positive detections of C. auris in sewersheds with less than vs greater than or equal to the median number of hospitals and nursing homes. States with a population above the national median age were more than three times as likely to detect C. auris in a given wastewater sample. At the sewershed level, locations with one or more hospitals had three times the amount of C. auris detections than locations with no hospitals. Similarly, sewersheds with greater than or equal to the median number of nursing homes had over twice as many detections of C. auris as those with fewer than the median number of nursing homes. These findings are consistent with the known epidemiology of C. auris, as infected patients’ median age is 68, and outbreaks are prevalent in long-term healthcare facilities and nursing homes (5, 14). The presence of an at-risk population appears to be strongly associated with C. auris wastewater detections, suggesting that public health efforts are appropriately focused on these at-risk populations.

Significant steps were taken to ensure the validity of detection results. A similar assay was used by Barber et al. for the detection of C. auris in southern Nevada during an outbreak (25). We tested nine samples from three different states using both the assay presented in this study and the one used by Barber et al. (25). No statistically significant difference was found between the two assays, which both targeted the same region of an internally transcribed spacer region. Our assay was further validated using a subset of four of these nine samples through sequencing. The amplicons generated by our forward and reverse primers independently and accurately matched the reference amplicon, indicating specific amplification (Fig. S4). Together, these results demonstrate that the assay used in this study is highly specific. Finally, we used an RT-PCR assay to detect C. auris which means that the assay should detect both RNA and DNA. We measured some samples using both RT-PCR and PCR, and the results indicated that most of the target nucleic acids in those samples were DNA. This was somewhat surprising since we expect the DNA target (located in the region of the genome that encodes ribosomal RNA) to be transcripted into RNA by C. auris. Future work will need to be done to further investigate the relative proportion of target RNA vs DNA C. auris targets in wastewater. Regardless, the use of an RT-PCR assay, allowing the detection of both RNA and DNA, should increase the sensitivity of the assay (by detecting both nucleic acids) yet perhaps complicate the interpretation of measured concentrations. For this reason, we only use the data in the presence/absence format.

Overall, our results demonstrate that C. auris wastewater surveillance can provide timely information on geographical distribution and help identify at-risk populations; wastewater data can be available within 24–48 hours of sample collection. As an emerging disease, clinical testing for C. auris largely depends on shipping samples to public health laboratories (7, 13, 14). Wastewater surveillance can help fill this gap for municipalities and states without the capital to fund testing systems, providing community-level information as to the prevalence and, more importantly, the spread of C. auris. These data are critical to public health responses, particularly when dealing with pathogens that do not or did not have robust clinical testing systems such as Dengue virus, SARS-CoV-2, and C. auris (14, 16, 17, 38). C. auris wastewater surveillance data can then be used to implement stricter screening protocols, cleaning protocols, and other non-pharmaceutical interventions that can prevent the spread of illness among vulnerable populations (9, 10, 20).

There are limitations associated with this study. Importantly, the availability of clinical data was limited due to the recent classification in 2018 of C. auris as a notifiable disease, making in-depth analyses of associations between wastewater percent positive detections and clinical data on infections unfeasible. In addition, it is well understood that the NNDSS database may not contain all the information on C. auris-positive cases in the United States; local jurisdictions may have local data to inform public health response to infections, and those data were not available for this work. Local jurisdictions are encouraged to use these wastewater data to compare to their local case rates. Furthermore, wastewater sampling was not uniform across the United States or within individual states, which could introduce biases into the analyses. Within the demographic data available at the sewershed level, information was limited as to the size of nursing homes and hospitals, which might be important factors controlling associations with wastewater data on C. auris. Lastly, we were unable to link specific wastewater concentrations to population-level incidence. Further experiments are necessary to understand the shedding patterns of C. auris in human excretions as to provide this direct link to disease occurrence in the contributing population. These experiments are particularly important to the determination of the significance of wastewater detections, as C. auris contributions to wastewater could potentially be from colonized individuals or individuals with acute infection. Finally, an additional limitation is that we cannot rule out the potential for C. auris detections in wastewater to be from zoonotic or environmental sources. Detections have been shown to match underlying expected demographic indicators such as age and the presence of nursing homes, further lending credence to the source of detections being human. While there is no evidence to support the possibility of zoonotic or environmental contributions of C. auris nucleic acids to wastewater, further research is needed to see if such evidence might exist. There have been case reports of C. auris infections in canines, and Candida spp. have been detected using culture and other molecular methods from biofilms in sewer pipes (41–44). It should be noted that wastewater detection of C. auris collected at the wastewater treatment plant scale does not imply community spread of the pathogen but is consistent with its presence in one or more individuals within the sewershed.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/mbio.00908-24.

Supplemental Material. mbio.00908-24-s0001.pdf.

Supplemental figures, tables, and text.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.