6

Wastewater Surveillance for Emerging Pathogen Threats

Broadly, emerging pathogen threats fall into three categories. The first category includes known pathogens that have been identified as causing human disease in other countries but the risk for spread to the United States is unknown and potential for local containment is uncertain (e.g., SARS-CoV-2 at the outset of the global pandemic in early 2020; Mpox in 2023).¹ The second category involves pathogens present only in animals, either in the United States or internationally, with potential for spread into humans. Influenza viruses detected in poultry, swine, cattle, wild bird, or wild mammal populations exemplify this category. The third category consists of previously unknown pathogens with unknown properties that make preparation for an outbreak difficult—the “pathogen X” scenario.

The specific characteristics of wastewater surveillance for emerging pathogens in these three categories differ somewhat from the approaches described for endemic pathogens in Chapters 2–4 because, outside of an outbreak, they are not expected to be detected in U.S. wastewater samples. Surveillance for emerging (or re-emerging) threats presents a much higher level of uncertainty in terms of detection potential in wastewater, analytical sensitivity and specificity, and the distribution of temporal and spatial sampling needed to provide evidence and inform public health action on these rare but high-consequence targets. Surveillance will be triggered by key events, such as an outbreak in another region or detection of cases in the clinical setting (animal or human).

___________________

¹ See https://www.cdc.gov/poxvirus/mpox/outbreak/2023-drc.html.

In this chapter, the committee discusses potential uses of sentinel sites in a national wastewater surveillance system to address emerging and re-emerging pathogens. A sentinel site here refers to a location where enhanced or specific surveillance should take place because it represents a likely “front line” of entry to a larger community. The first section offers a vision for wastewater surveillance for emerging human pathogens that are previously known outside the United States or in animal populations. Sentinel sampling approaches, analytical methods, and specific needs for data analysis, visualization, and interpretation of wastewater surveillance data are discussed for these threats. In the final section of the chapter, the potential use of wastewater surveillance to address a “pathogen X” scenario and research and technology development opportunities associated with nontargeted detection are reviewed.

VISION FOR WASTEWATER SURVEILLANCE FOR KNOWN, EMERGING PATHOGEN THREATS

This section presents a vision for sentinel-site-based wastewater surveillance of emerging pathogen threats that are known to public health agencies. These include emerging human pathogens that may be limited to certain high-risk states (e.g., dengue virus in the southern United States) and those with epidemic or pandemic potential with serious consequences to public health that are known from prior disease outbreaks or detection outside the United States. Also included are diseases known to occur in animals with potential for human transmission.

Pathogens Emerging Outside the United States

The identification and reporting of the Omicron variant of SARS-CoV-2 in South Africa serves as a clear example of how internationally collaborative surveillance aided U.S. and global pandemic preparedness. The variant B.1.1.529 (Omicron) was first identified in South Africa and reported to the World Health Organization on November 24, 2021,² along with evidence of rapid spread within the country, thus providing an early warning to other nations. The first case was subsequently confirmed in the United States on December 1, 2021, and Omicron became the dominant variant by the end of the month (Iuliano et al., 2022). This advance warning allowed the U.S. surveillance systems, including wastewater surveillance, to initiate specific testing for B.1.1.529 and alerted hospital systems to take measures to preserve intensive care unit bed availability (Kirby et al., 2022). In retrospect,

___________________

² See https://www.who.int/news/item/26-11-2021-classification-of-omicron-(b.1.1.529)-sars-cov-2-variant-of-concern.

wastewater surveillance would have been similarly informative for SARSCoV-1, a distinct coronavirus belonging to the same taxonomic grouping (Sarbecovirus) as SARS-CoV-2, which was first identified in 2003 as a cause of human morbidity and mortality outside the United States. The risk for spread of SARS-CoV-1 into the United States was unknown, but only syndromic surveillance was available with no population-based testing that would have been afforded by a national wastewater surveillance system.

Wastewater surveillance at sentinel sites could substantially enhance systematic surveillance for pandemic preparedness, including pathogens emerging outside the United States. Among the promising candidates for sentinel sites for routine surveillance are airports across the country that serve large numbers of international passengers and international mass gatherings and sporting events. Routine ongoing wastewater surveillance at these sites would provide baseline data before outbreaks and enable early detection of pathogen spread, supporting an effective outbreak response. If or when concerns are raised about a potential pandemic threat, this sentinel sampling can be intensified in terms of frequency and granularity, perhaps even targeting aircraft lavatory waste on arrival from specific countries with outbreaks, depending on the specific threat and potential public health actions. For example, wastewater surveillance for human cases of Middle East respiratory syndrome coronavirus (MERS-CoV) (Box 6-1) would focus on flights arriving from Middle Eastern countries where humans frequently come into contact with camelids, as well as other countries experiencing an outbreak of MERS. If airplane-specific data are collected, the results could be used for contact tracing or to encourage or require passengers to undergo follow-up testing or quarantine, depending on the threat. If pooled

data from airplanes are collected, the data could be used to notify local and national agencies and clinicians of a new public health threat.

Similarly, the detection of an outbreak of a known but nonendemic pathogen could trigger localized or regional surveillance based on the understanding of specific risk factors. These pathogens are considered too rare to have value for routine monitoring throughout the National Wastewater Surveillance System (NWSS), but localized community-level wastewater surveillance may have strong utility once triggered by another data source, such as an increase in local community cases, hospitalizations, deaths, or other syndromic surveillance markers. Having a truly national system for wastewater surveillance would allow a rapid response, using the existing system for sample collection, processing, analysis, and reporting. The dengue virus outbreak in 2022–2023 is an example of this regional threat, based on distribution of the mosquito vector responsible for its transmission (Box 6-2). In addition, existing NWSS samples from affected regions could be analyzed for these additional targets until the pathogen outbreak is no longer seen as a significant threat. This approach is currently used informally by many jurisdictions (e.g., in response to local emergence/outbreaks of polio, measles, hepatitis A, or Mpox). This type of flexibility was highlighted as a key characteristic in the NASEM (2023) vision of national wastewater surveillance. The committee notes that this targeted

“emergency” surveillance would likely require additional national, state, or municipal resources, but having a robust functioning national system for endemic pathogens (e.g., SARS-CoV-2) allows rapid deployment in response to an outbreak or high-level risk of such an outbreak, avoiding the time lag required to start a wastewater surveillance program where one does not exist.

Pathogens Emerging from Animal Sources

An estimated 75 percent of all recent emergent human pathogens originated in animal hosts (Taylor et al., 2001). These include those with direct animal-to-human transmission (e.g., rabies virus), transmission via an animal product (e.g., Salmonella), and transmission among humans after initially emerging from an animal host or hosts (e.g., influenza, MERS-CoV [see Box 6-1]). The majority of animal pathogens are most likely species- or genus-restricted and have limited, if any, risk for zoonotic transmission. Consequently, identifying useful targets for routine wastewater surveillance from among pathogens newly emerging from animal sources, in the absence of identified human infections, is challenging. Thus, wastewater surveillance of new or emerging pathogens with the potential for human transmission should be considered as an additional tool when justified based on data on infectious disease outbreaks in animals. National and state laboratories (e.g., the National Animal Health Laboratory Network) routinely investigate outbreaks of infectious diseases in the animal production sector to ensure a safe food supply in the United States and have the analytical capacity to detect pathogens in domesticated animals. Newly recognized or particularly rare animal pathogens are best investigated directly in livestock production surveillance programs, agricultural isolation facilities, and clinical settings where transmission dynamics and zoonotic potential can be investigated and disease reporting to relevant authorities should be routine.

Incorporation of animal pathogens into wastewater surveillance could be triggered by animal disease outbreaks with known or a high likelihood of risk for spread into domestic animal production and human populations in close contact with susceptible animals. As an example, H5N1 avian virus bird and mammal outbreaks represent events that should trigger considerations for wastewater surveillance, given the high risk for H5 influenza viruses to infect humans. The recent spread of H5N1 influenza A into dairy cows with transmission to farm workers reflects the highly adaptive and dynamic nature of influenza viruses and makes a case for ongoing vigilance for zoonotic viruses with high propensity for mutations (CDC, 2023a; Uyeki et al., 2024). The potential for detection of the influenza H5 gene in wastewater was demonstrated at three wastewater treatment plants with agricultural inputs in a state where H5N1 influenza was identified in

dairy cattle (Wolfe et al., 2024b). Other avian influenza viruses with high probability of infecting humans, such as H7 and H9, would also be priority targets for sentinel-site-based wastewater surveillance in the event that outbreaks are observed in animal production facilities.

Identifying sentinel wastewater surveillance sites that could be ramped up in response to outbreaks would enable the development of an integrated animal and human surveillance system that complements clinical case detection. Sentinel sites would ideally be located in communities where high concentrations of people reside who work in animal production or processing. These sentinel sites should be identified in advance, with support from the U.S. Department of Agriculture (USDA), so that wastewater surveillance could be initiated rapidly when a disease with potential zoonotic transmission is detected in animals by animal health surveillance. In these settings, wastewater surveillance could enable early detection, assess rapidly changing outbreak dynamics, and inform effective combined responses that engage both animal and human health sectors.

SAMPLING FOR EMERGING PATHOGEN THREATS

In this section, the committee discusses sampling characteristics for emerging pathogen threats.

Sampling Design

A primary challenge for designing wastewater surveillance for emerging pathogen threats, including those that have been well characterized, is that the initial goal is to determine whether the pathogen is or is not present in the population—a binary outcome. Whereas a single positive test result can trigger additional testing, including repeat sampling and genetic sequencing to confirm the detection, a nondetection at a large wastewater treatment plant could mean the virus is present at a level that is currently beneath the detection limit. The significance of a nondetection is affected by the analytical detection threshold, the prevalence within the sampled population, the size of the population sampled, the probability that infected people will contribute waste to a sample, and the quantity of pathogen (or pathogen nucleic acid) shed into a waste stream Therefore, sentinel-based surveillance will enhance the predictive value of wastewater testing strategies, relative to surveillance at large wastewater treatment plants, by concentrating efforts on smaller at-risk populations where occurrence of emerging infections is more likely to be detected with current technology.

A sampling design for an emerging pathogen threat also needs to be developed with a clear vision for what actions will be taken if detection occurs. For example, if wastewater surveillance is conducted to detect a rare

but high-consequence emerging pathogen from a single region, sampling of wastewater from incoming aircraft from that region could be conducted, but such granular data are most useful if actions will be taken to require quarantine or further testing of the passengers from planes with confirmed positive results. In contrast, if the public health goal is to inform local communities to prepare for a new emerging disease, sampling of triturators that pool waste from many planes may be more appropriate (and would reduce costs). In this chapter, the committee presents its vision for a sentinel surveillance program based on current understanding of emerging public health threats and useful actions that could be taken, but emerging pathogen surveillance ultimately requires flexibility to be successful. Necessary temporal and spatial resolution of sampling sites would be expected to evolve with changing assessments of public health threats, both nationally and internationally. Public health agencies’ views on what actions they need to take regarding emerging pathogens under various scenarios may also evolve over time.

Optimized sampling for pandemic preparedness targets does not require routine surveillance at all locations. Doing so would be inefficient and jeopardize the sustainability of the surveillance system (as seen with ending of Project Biowatch in 2014). The following sections discuss sentinel sampling characteristics for known, emerging human pathogens and known emerging pathogens with zoonotic risk from animal sources.

Sampling Characteristics for Known Emerging Human Pathogens Sentinel Site Sampling Locations

Airplane and airport monitoring.

In the year ending December 2023, 237 million airline passengers transited between the United States and other countries (USDOT, 2024). Airport-based, aircraft-based, and arriving traveler sentinel surveillance have been shown to have the greatest potential to rapidly detect the importation of recognized pathogens that are circulating outside the United States (Agrawal et al., 2022; J. Li et al., 2023; Wegrzyn et al., 2023). Wastewater sampling at airports can be conducted at a range of scales:

• At an airport-based wastewater treatment plant or downstream sewer line to sample the entire airport (including staff, travelers, and airplane waste),
• At an international terminal (including staff and travelers),
• At triturators that combine and process airplane lavatory waste from all incoming aircraft prior to release into the sewer system, and

• From individual planes when lavatory waste is emptied at the end of each route.

To address active known threats emerging from other countries and spread through human-to-human transmission, monitoring would most appropriately focus on aircraft arriving from international locations. Aircraft-based wastewater surveillance offers early detection of pathogen spread into the United States from regions where these pathogens have emerged. For aircraft-based sampling in the earliest phases of epidemics, long-haul flights entering the country are most productive and efficient, given the likelihood of passengers to use the toilet on the aircraft (Jones et al., 2023). Early research showed six of eight aircraft vessels had a positive SARS-CoV-2 detection in wastewater when only a single case on each flight was identified (Bivins et al., 2024). Ahmed et al. (2022) sampled long-haul flights arriving in Australia and reported a positive predictive value of 87.5 percent and a negative predictive value of 76.9 percent based on clinical detections of airline passengers during a subsequent 14-day quarantine. In contrast to other airport sampling, aircraft sampling is more granular and can be targeted toward pathogens of concern with specific countries of origin. (See Box 6-3 for a hypothetical example.) This surveillance approach

may also be useful for detection of pathogens that could be circulating in countries where there is inadequate surveillance and reporting to identify country-specific risk for pathogen importation into the United States. However, sampling of individual planes is staff intensive and costly and should be triggered only when there is a clearly identified threat and a defined public health response action.

Sampling at an airport triturator facility provides a composite sample limited to arriving aircraft (i.e., without addition of nontraveling airport personnel), can be automated, and has the least burden on airline and airport personnel. In a pilot program, the Centers for Disease Control and Prevention (CDC) has implemented surveillance using an automated sampler device at the San Francisco International Airport triturator.

To be able to respond in a timely manner, NWSS will need to pilot sentinel sampling approaches to meet different scenarios, so that the sampling strategies and uncertainties are well understood before the next public health emergency. Baseline data collection at these sentinel sites can be used to evaluate the sensitivity and specificity of this sentinel program in detecting outbreaks that are occurring in international locations. For example, information on SARS-CoV-2 variants through this baseline surveillance and additional pilot sampling efforts for individual long-haul international flights could be compared to data reported globally to evaluate capabilities for rapid and accurate detection of internationally imported cases. CDC is currently piloting wastewater surveillance in airplanes and airports and analyzing the results in comparison to individual voluntary passenger screening through its Traveler-Based Genomic Surveillance program and air sampling at airports (Tanne, 2023).³ Meanwhile, other emerging technologies (e.g., airborne virus sampling; Ramuta et al., 2022) and existing approaches (Appiah et al., 2022) should continue to be assessed for use in international airports, airplanes, and other transportation centers, given their capacity to sample incoming travelers at a large scale and tailor testing to detect emerging pathogens prior to their spread in the community.

International event monitoring.

In addition to airport- and aircraft-based surveillance, sentinel surveillance could include wastewater sampling in municipalities hosting multiday international events (e.g., sporting and entertainment events, conferences) to detect localized infectious disease transmission from an imported case or cases. Wastewater surveillance could be considered as part of the toolkit for early detection and monitoring for disease outbreaks at large events and could potentially complement bioterrorism monitoring (Elachola et al., 2016). Host cities could make wastewater monitoring part of their permitting and funding requirements for large

___________________

³ See https://wwwnc.cdc.gov/travel/page/travel-genomic-surveillance.

events, connecting event planners with individuals and organizations that are already involved in the wastewater surveillance in the region.

Wastewater treatment plant monitoring.

If analytical methods can detect extremely low levels of emerging pathogens or if high shedding rates occur for a target organism, existing wastewater surveillance samples from select cities in the United States could also serve as sentinel surveillance data. This would expand the population covered by sentinel-site monitoring without any additional sampling. However, public health actions facilitated by community-level wastewater surveillance data would differ from information gained from a specific flight. Also, if very few cases exist in a large city, false negatives are more likely to occur. Confidence limits for the presence of an emerging pathogen, if in fact it was present, could be computed to assess the feasibility of using a municipal wastewater treatment plant as a sentinel site for an emerging pathogen (see Box 6-4). Once a pathogen has been detected and is spreading within a community, emergency wastewater surveillance at the wastewater treatment plant scale can be used to monitor trends and potential recurrence in that locality and inform public health actions (Box 6-5).

Sampling Frequency for Emerging Human Pathogens

Sentinel surveillance at airports or other major ports of entry for new pandemic threats will be most useful for the first few days or weeks during the onset of an epidemic or an outbreak emerging in a new country or location, when findings can translate to action. The temporal sampling intensity for an emerging agent in the population should be based on the urgency and scale of the actions that might be taken if the emerging pathogen is detected. For example, monitoring for an emerging deadly virus with human-to-human transmission would merit much more intense sampling (e.g., individual aircraft level) than for a new, but mild illness.

Once a specific emerging pathogen becomes widespread in the population, sentinel monitoring for that pathogen at ports of entry is no longer needed. If the pathogen continues to pose a significant public health risk and wastewater surveillance provides useful information to support public health response, it could be added as an emergency target to existing or expanded wastewater surveillance in affected areas (Box 6-5) or, if useful, added to the NWSS core panel of endemic targets that are monitored across the nation.

Sampling for Pathogens Emerging from Animal Sources

Animal disease outbreak data can guide where to locate sentinel wastewater surveillance sites, although finding appropriate sites may be challenging because wastewater systems tend to underserve the rural areas

where most intensive animal production occurs. Therefore, sewershed maps should be overlaid with locations where cattle, swine, and poultry production and slaughterhouses are located and/or large numbers of animal agricultural workers reside in order to identify potential sentinel sites. Such sites should be identified with assistance from the USDA and assessed in advance for the feasibility of wastewater surveillance so they can be rapidly activated in the event of animal outbreaks with high risk of zoonotic spillover. Although baseline data are less important when looking for emerging pathogens than for other pathogens, adding some such sites to NWSS for routine sampling may be beneficial to ensure that capacity is already built into the wastewater surveillance system for more intensive sentinel-site sampling when needed.

A major limitation of using wastewater surveillance for understanding the transmission of zoonotic pathogens is that current molecular detection methods in wastewater cannot distinguish between animal and human infections if there are sources of animal waste in the wastewater or presence of other materials containing zoonotic pathogens discharged into wastewater, such as milk. Ideally, sentinel sewered sites would be identified that are not likely to be confounded by animal waste inputs to maximize the specificity for detection of cases and outbreaks in humans. Sentinel sites could be tested for baseline levels of animal markers that could determine the extent of contamination at sites. Even for sites in which animal-based inputs are expected to be minimal, detection of animal pathogens in wastewater will require follow-up by testing methods optimized to distinguish animal infections from human infections (Fisher et al., 2015; Noble et al., 2003; Sinton et al., 1998). Triangulation of wastewater data with potential animal-based sources as well as active surveillance to assess the occurrence of clinical and subclinical cases in humans would constitute a coordinated and integrated approach to emerging pathogens from animal sources.

Biosafety Issues

As mentioned in Chapter 2, determination of appropriate biosafety protocols is needed in advance to ensure that the staff responsible for sampling and processing the samples have necessary protections in place to handle potential future targets and that laboratories with the appropriate biosafety level have the analytical capacity. Groups developing biosafety recommendations should include individuals with expertise in wastewater sampling and analysis (as distinct from clinical laboratories) as well as microbial risk. It is important to understand existing biosafety protections

in conventional wastewater handling⁴ and that detectable genetic signatures of emerging pathogens in wastewater do not necessarily represent viable, infectious agents.

ANALYTICAL METHODS FOR KNOWN, EMERGING PATHOGENS

When analytical methods for known, emerging pathogens are applied, first and foremost, it is critical that the assays used to detect the pathogens are sensitive and specific. Pathogen-specific (targeted) analytical techniques for surveillance for known, emerging human and animal pathogens (including pathogens endemic outside of the United States) are used largely at universities and government agencies that have worked to detect pathogens where outbreaks have occurred to date. Reference genomes exist in public databases, and polymerase chain reaction (PCR) assays have been developed for clinical detection of the specific emerging pathogen. Thus, the molecular detection techniques used to identify these pathogens will be similar to that described for endemic pathogens in Chapter 3. However, the sensitivity and specificity of existing PCR assays designed for clinical samples need to be assessed for application to wastewater and the specific sentinel-site conditions. An example of how a clinical PCR assay has been adapted for wastewater monitoring is the use of the non-variola orthopox (Mpox) clinical PCR test, which now has been adapted for monitoring this pathogen in wastewater. CDC should invest in additional methods development to ensure that appropriate assays are available for emerging pathogens of pandemic potential and that these assays are validated in a range of wastewater samples. A formal advisory committee to NWSS could be valuable in shaping analytical method and assay development and providing rapid advice as needed in the realm of emerging pathogens. Furthermore, for emerging animal pathogens with zoonotic potential, collaboration and resource sharing with the USDA and U.S. Agency for International Development would reduce effort duplication and support data sharing.

Among zoonotic pathogens of particular concern for public health, coronaviruses and influenza viruses are known to be readily detected in wastewater (see Chapter 5), but there are fewer data on wastewater detection of zoonotic paramyxoviruses and flaviviruses. Studies on other high-risk pathogens would guide decisions on how best to implement wastewater-based surveillance at sentinel sites and focus efforts for further method development and refinement.

___________________

⁴ See https://www.cdc.gov/global-water-sanitation-hygiene/about/workers_handlingwaste.html#cdc_generic_section_2-personal-protective-equipment-ppe.

Emerging pathogen targets of interest may be present in a sample at very low abundance, especially in wastewater samples with a larger contributing population that dilutes the sample. Therefore, methods with low detection levels and high specificity are particularly valuable to detect emerging targets while minimizing false negatives and false positives. As discussed in Chapter 3, digital PCR has been shown to be more sensitive with lower limits of detection for SARS-CoV-2 and is less likely to be impacted by the presence of inhibitors, which can cause falsely negative results (Ahmed et al., 2022a; Ciesielski et al., 2021). Quantitative PCR requires a standard curve, which can introduce variability if not performed meticulously. However, digital PCR equipment is more expensive. Ultimately, sentinel-site sample processing and analytical methods should be standardized, given the importance of comparable data across the nation when assessing emerging pathogens of pandemic threat.

Unlike with standard NWSS analytical methods for endemic pathogens, positive detections of emerging pathogens require particular scrutiny because it is possible that false positives can occur, and public health agencies will likely need confirmation that the laboratory analytical results are accurate before taking action. Laboratories that process sentinel-site samples should implement strict quality criteria and protocols, as discussed in Chapter 3, including consistent use of spike-in controls that assess nucleic acid extraction efficiency and assay sensitivity. In accordance with best practices, all negative results should be reported in light of the limits of detection for specific pathogens. Positive results should be individually inspected, with re-analysis of the sample and confirmatory sequencing. Quality control protocols for sentinel-site samples should be standardized so that all data are of the same laboratory quality. If something has not been detected before or is rarely detected, the laboratory should flag it for follow-up validation when reporting to the public health agency. In Chapter 3, trade-offs between sample processing and analysis speed, cost, and sensitivity were discussed for endemic pathogens, but for emerging pathogens, accuracy and sensitivity are the highest priorities. Given the extensive amount of work required to follow up on positive results and the potential public health implications of false negatives, investments that reduce detection limits while also improving specificity will save resources in the long run.

Although the planned sentinel surveillance assays are expected to be performed and reported in real time, circumstances may arise where it is advisable to retrospectively analyze archived samples to determine trends of newly detected pathogen signals. Therefore, all processed samples from sentinel sites should be archived for a minimum of 3 months to ensure that samples can be re-analyzed if data quality issues are raised. However, archiving for up to a year would be useful to confirm absence of detection

prior to any pathogen emergence timeline and/or better refine timelines for outbreak emergence.

DATA ANALYSIS, INTERPRETATION, COMMUNICATION, AND INTERVENTION PLANNING FOR EMERGING PATHOGENS

Wastewater surveillance’s key contribution to safeguarding public health is identifying when a new pathogen has entered the community so that public health authorities can quickly intervene. Therefore, the tools and strategies for information sharing, visualization, and data analysis for emerging pathogens in wastewater will need to be designed primarily for public health agencies rather than the general public.

In a national sentinel monitoring program, CDC should bear primary responsibility for standardizing protocols and conducting real-time analysis of the data, while sharing the data and interpretation with local public health agencies on a regular, pre-defined basis, regardless of whether or not a threat is detected. CDC should design tools and strategies for visualization and data analysis to evaluate potential pandemic threats, utilizing information on other detections and clinical data, including data from outside of the United States. As part of ongoing pandemic preparedness and response planning, CDC should prepare response action plans for a range of scenarios based on key factors, including its confidence level that the pathogen is detected, the timeliness of the information, and its understanding of the severity of associated health effects, expected duration of illness, spread mechanism, and potential interventions. These plans should include internal and external communication instructions and indicate which entity (CDC or local agency) is responsible for implementation, tracking, and reporting of intervention plan actions. CDC, potentially through the Centers of Excellence, should practice response action plan implementation with local public health agencies using tabletop exercises to ensure that localities are prepared.

To assist interagency coordination and make the best use of existing resources, wastewater surveillance should be included in the existing notifiable disease reporting system. State government (public health, agriculture agencies), national (CDC, USDA), and international programs (World Health Organization, World Organisation for Animal Health) for surveillance and reporting of notifiable human and animal diseases provide outstanding infrastructure including expertise, resources, and communication support. Including wastewater detections in this infrastructure will maximize public health actionability of this information.

EMERGING TECHNIQUES FOR FUTURE PUBLIC HEALTH MONITORING OF UNRECOGNIZED PATHOGENS

To date, validated laboratory assays that have been developed and used for diagnosis of infections in either humans or animals have been adapted for wastewater surveillance. However, syndromic surveillance can detect outbreaks caused by previously unknown pathogens for which no definitive laboratory or wastewater compatible test exists. These agents fall into two broad groupings: (1) variants of a known pathogen species or a “new” pathogen species or strains but within a known family (e.g., coronaviruses, orthomyxoviruses) and (2) truly novel pathogens. While clusters of human or animal infection due to an unknown pathogen are investigated, untargeted or semi-targeted methods can be used to detect and sequence these types of novel agents. Once “pathogen X” sequences are widely available, pathogen-specific assays (see Chapter 3) can be developed within days to weeks to detect the novel pathogen and shared broadly. Then, wastewater surveillance strategies can be designed to target these now known pathogens using more readily available analytical equipment, as discussed previously in this chapter (known, emerging pathogens). In this section, the state of technology for detection of unknown, emerging pathogens is discussed, followed by a review of the potential of these untargeted methods in the future of wastewater surveillance.

State of Technology for Semi-targeted and Untargeted Analytical Methods

The majority of wastewater surveillance efforts have been focused on a handful of known, high-priority pathogens, but, in the future, it is possible that wastewater surveillance will broaden to monitor the emergence and evolution of unrecognized pathogens and other important epidemiological transitions. The committee envisions that two broad categories of approaches not in active use right now could be used to surveil for these types of agents: a semi-targeted and a fully untargeted approach.

Semi-targeted Methods

A semi-targeted approach can include methods such as amplicon sequencing, microarray, or hybrid capture and sequencing. Amplicon sequencing (Figure 6-1), which is currently among the most widely applied semi-targeted methods, involves PCR amplifying a specific DNA region of a target genome using primers, or short DNA sequences, that are complementary to highly conserved regions of the target DNA. An example of

amplicon sequencing is 16S rRNA gene sequencing, where primers are generated to bind to highly conserved regions of the 16S rRNA gene, which is a “marker gene” in bacteria and other prokaryotes. Once amplicons are generated, the intervening region is amplified and then sequenced. The intervening region is variable among different organisms, and thus the amplified sequence, or amplicon, can be compared to reference genomes which then will provide information about the composition of prokaryotic organisms within the samples. In addition to use for sequencing of prokaryotic communities, other types of amplicons can be generated. For example, these types of approaches have proven quite useful in the identification of

and surveillance for SARS-CoV-2 variants in the past. In this case, variable regions of the SARS-CoV-2 genome are amplified using primers against highly conserved regions of the genome. These types of methods provide insight into emerging variants and are more cost-effective than fully untargeted methods, which are outlined below.

In addition to amplicon sequencing, a similar but technologically distinct approach is to use a hybrid capture (see Figure 6-1). In this approach, a single-stranded nucleic acid sequence that is complementary to a pathogen or pathogens of interest is developed. All nucleic acid sequences that are complementary (i.e., similar) enough to that “bait” sequence are bound by it. This approach allows for imperfect complementarity, and thus novel variants that are somewhat related to known targets can be recovered. The sequences that are “captured” by the bait are then sequenced and analyzed (Nasir et al., 2020). Microarrays are similar to hybrid capture except that hybridization is assessed directly, typically through fluorescent tags; sequencing is not performed. Microarrays provide a fast and inexpensive way to identify differences between closely related genomes (Gresham et al., 2008).

Untargeted Methods

While the semi-targeted approaches outlined above can be useful in identifying novel targets that are related to known targets, they are less useful for totally novel or “unknown unknowns.” Such targets may be viral, prokaryotic, or eukaryotic in nature. To detect such targets, assuming they are organisms with DNA or RNA as their genetic material as opposed to protein-based infectious agents (prions), untargeted approaches can be used. These methods provide information on the totality of DNA and RNA sequences that are within a sample. Such a sample can be a bulk metagenomic sample or viral particle enrichment. Interpretation of these data in complex sample types such as wastewater, in which there are likely thousands or more different microbes in addition to other types of nucleic acid from sources such as foodstuffs, is much less straightforward than for clinical samples.

Untargeted analysis methods for wastewater might fall into one of two major categories. In the first category, efforts might be made to identify the novel target or pathogen of interest in a “reference-free” manner. This type of approach could build on well-established methods that have been pioneered in the study of environmental and host-associated microbiomes (e.g., shotgun whole genome sequencing for DNA or total RNA sequencing; see Figure 6-2), in which the genome is broken up into small pieces, which are then assembled using computational algorithms into larger genomic fragments. This type of approach, which is akin to putting together the pieces

of a jigsaw puzzle from a mixture of puzzle pieces, enables the identification of novel organisms or targets that might be present in a sample (Parks et al., 2017; Tyson et al., 2004). When such a sample is compared to a reference library of known organisms, novel agents could be discovered. Indeed, these types of approaches have enabled the assembly and surveillance of important agricultural pathogens, such as the tomato brown rugose fruit virus, a recently emerged pathogen of nightshade plants (Natarajan et al., 2023; Salem et al., 2015).

In the second category of untargeted analysis, shotgun metagenomic data can be queried in a “reference-based” manner. Here, untargeted metagenomic data are compared to a large, well-curated database of available and relevant reference sequences using read-based or marker-based approaches. Such approaches could identify the presence of a novel variant of a known pathogen, such as mutated versions of the mpox virus. Alternatively, this could serve as a way to query increasing and decreasing abundances of novel, emerging candidate antimicrobial resistance genes that have some level of relatedness to known genes.

Excellent approaches exist for the types of analysis outlined in both reference-free and reference-based strategies, and with modern computing capabilities, many of these analyses can be carried out quickly and with relative ease. All methods require there to be adequate abundance of the organism to be detected within the given sample. The number of sequences per sample (“sequencing depth”) collected determines how sensitive a given assay is, and deeper coverage improves specificity substantially because the ability to distinguish a pathogen from its close relatives becomes greater. Untargeted analysis of wastewater may necessitate very deep sequencing to detect signals from pathogens that may be rare or poorly abundant in the samples. All in all, these data can provide information that can guide the development and execution of more targeted and quantitative assays that follow up on these initial screens and subsequently allow for the development of targeted and cheap detection methods for specific pathogens discovered during this untargeted discovery phase.

Although untargeted sequencing methods can be tuned to be exquisitely sensitive, there can be trade-offs between sensitivity and specificity. This can be particularly challenging when an emerging pathogen or target of interest shares substantial genomic sequence homology with harmless, prevalent, and abundant nonpathogens. Rapid read-based methods for data analysis cannot in principle distinguish closely related organisms. With current technology, to differentiate between two closely related organisms can require additional detailed and sometimes customized data analysis that may involve applying so-called “metagenotyping computational tools”

(Zhao et al., 2022) to deep shotgun sequencing data. These types of processes require knowledgeable scientists and large computer resources to carry out.

These technical and workforce issues currently limit the scalability of assay interpretation from a public health application perspective, but many opportunities exist in the research space. While expensive, and therefore not routine, generation of metagenomic data from wastewater will enable both prospective and retrospective evaluation of known, emerging, and novel pathogens, and this may help inform future public health applications. For example, total RNA or DNA sequencing of historically banked samples might enable identification of emergence or re-emergence of pathogens, such as a bird flu or Salmonella enterica outbreaks. By mining existing data or samples from pre-outbreak time periods, researchers could possibly evaluate the cost-effectiveness of such approaches for sensitively and specifically detecting future outbreaks.

It is possible that other signals in wastewater that are secondary to the pathogen of interest might be altered in the presence of a pathogen, and in these cases, future analytical approaches may focus on the discovery of these secondary signals as opposed to rare pathogen signals. However, development and application of these types of approaches will require additional analytical advancements, likely aided by the rapid emergence of artificial intelligence–based data analysis tools. Overall, while untargeted methods hold great promise, because of the limitations enumerated above, they have yet to enter the mainstream.

LOOKING AHEAD

Although semi- and untargeted methods may not be ready to roll out on a broad and national level at present, as future molecular and analytical approaches evolve and are refined, it is expected that such data will be an important part of future surveillance efforts. Indeed, large-scale efforts to collect longitudinal biological data in a well-curated manner (for example, collection of protein structures in the Protein Data Bank⁵) have enabled great advances in machine learning–based protein structure prediction (Jumper et al., 2021). Few, if any, could have anticipated that having such protein structure data would eventually lead to machine learning–driven computer programs that can accurately and quickly predict the protein structures of nearly any amino acid sequence. Thus, the committee anticipates that continued close communication between wastewater scientists and engineers, microbiologists, health care researchers, epidemiologists, and data scientists will be critical in this next phase of innovation.

___________________

⁵ See https://www.rcsb.org/.

There are several ways in which modern artificial intelligence approaches may enhance the effectiveness and utility of wastewater surveillance for emerging pathogens. First, machine-learning approaches excel at automated pattern recognition and thus may be optimally positioned to identify patterns and anomalies in large datasets. For example, new variants of SARSCoV-2 could be readily identified from large datasets by identifying newly present short RNA sequences, where parts of the sequence have similarities to known SARS-CoV-2 but specific regions are different. Second, once an adequate set of wastewater surveillance data and outcomes is developed, predictive models can be developed, which might provide advance warning of upcoming shifts in disease incidence and prevalence. Finally, with advances in computing capabilities and artificial intelligence algorithms, it is possible to incorporate a variety of highly diverse data types to inform integrated datasets that might further improve predictive models. Although current investment in artificial intelligence related research is primarily through the National Institutes of Health (NIH) and the National Science Foundation (NSF), CDC and its NWSS Centers of Excellence should periodically evaluate applicability of modern machine-learning approaches to wastewater surveillance.

A carefully considered investment in this realm may enable early detection of targets or pathogens that are so-called black swans—events that come as a surprise and have a major effect (Velappan et al., 2022). With advancements in data analytics, wastewater surveillance may allow scientists to follow the totality of the sequence data (as a combined and very rich source of information on both pathogens and the microbiome response to pathogens) over time. Changes in the patterns of the composition of the sequence data that are beyond typical variance could be determined by advanced machine learning methods. These deviations from patterns could then be triangulated with other available data, such as clinical syndrome information or climate information, to identify possible evidence of emerging epidemiological shifts.

All in all, while trade-offs exist in making investments in these types of future-focused efforts versus more immediate or near-term surveillance efforts, an arm of wastewater-based surveillance research and development that looks to the future and adapts to the strengths of emerging technologies and analytical approaches is sage. In the near term, the types of efforts described in this section may be information-gathering exercises and are unlikely to trigger specific public health actions, but with the learning that is gained over time, they may transform into leading indicators of emerging pathogens, and thus may help prepare our public health system for specific actions.

CONCLUSIONS AND RECOMMENDATIONS

The committee’s Phase 1 report (NASEM, 2023) recommended strategic incorporation of sentinel sites in the NWSS to allow early detection of emerging pathogen threats. Ideally, a wastewater surveillance system for emerging threats could identify pathogens as they enter the country. Importantly, the routine surveillance provided by the NWSS allows rapid emergency response to a disease outbreak, which may be national (e.g., an emerging SARS-CoV-2 variant), regional (e.g., dengue virus), or limited to high-risk settings (e.g. Influenza H5N1 in dairy communities). These detections would enable timely public health actions, such as targeted therapies or recommendations regarding behaviors (e.g., limiting contact with others) that could mitigate the spread of and adverse outcomes associated with the disease. This chapter discusses specific characteristics of wastewater surveillance at sentinel sites for emerging pathogens and for emergency outbreak response.

International traveler–based wastewater surveillance at airports serving large numbers of arriving international passengers would provide critical sentinel surveillance sites for emerging pathogens. With current technology, sample collection at triturators would allow the most focused and cost-effective sampling of incoming aircraft without additional burden on the airlines and airline personnel. Several (<10) of the largest international airports in the United States should be selected to provide a geographically diverse set of sentinel sites. Sentinel-site samples should be processed and analyzed in a rapid, reliable, and consistent manner to ease data comparison and interpretation and support a timely response. Such sentinel sites would enhance potential local and national response strategies, ranging from early warning to additional testing and quarantine of passengers, depending on the significance of the threat and resources invested. CDC can leverage resources developed for traveler-based programs. CDC should ensure that this sentinel monitoring strategy has clear links to public health action, with plans and protocols for local and national response based on the pathogen detected. CDC should also periodically review the program so that the sentinel strategy stays abreast of technology development, including ongoing assessment of alternative strategies such as airborne pathogen monitoring.

CDC should identify additional, uniquely targeted sentinel sites to provide surveillance for human infections emerging from animal sources. As zoonotic pathogens can emerge from intensified livestock production, CDC should ensure coverage of select representative at-risk communities working in intensive animal production and slaughtering, particularly avian species and swine due to their role in influenza A evolution and transmission. Additional testing may be necessary to distinguish animal infections from human infections.

Temporary community-level wastewater surveillance could provide valuable information on international introductions associated with mass gathering events and localized epidemics and inform public health response. Large-scale public events (e.g., Olympic games, World Cup) where tens or hundreds of thousands of travelers are expected to congregate pose risk for accidental introduction and spread of pathogens, especially if drawing in crowds from international locations for several days. Incorporation of wastewater surveillance in advance event planning and permitting would help ensure readiness for epidemic and pandemic monitoring as needed. Likewise, highly localized outbreaks may merit temporary infectious disease surveillance at the community scale in affected areas.

CDC should bear the primary responsibility for conducting real-time analysis of data from NWSS sentinel sites, rapidly sharing verified data with local, state, and tribal public health agencies and developing a rapid-response plan. Real-time analysis and interpretation of emerging pathogen data from sentinel monitoring sites requires context and situation updates from coordinated international, national, and state public health (and potentially animal health) authorities, which is beyond the reach of most local public health agencies. CDC should develop laboratory and agency guidance and protocols to ensure rapid re-analysis and verification of emerging pathogen detections and ancillary data. Additionally, it should develop guidance for local, state, and tribal agencies on how to respond to positive detections, including additional data collection, interventions, and communication.

Wastewater surveillance should be linked with notifiable disease reporting systems to enhance existing infrastructure for interagency communication. To achieve this, pathogens classified as notifiable diseases that are detected in wastewater should be reported within existing local, national, and international disease reporting systems. These programs for surveillance and reporting of notifiable human and animal diseases provide resources including expertise and communication support. Data on detection of notifiable pathogens in wastewater should be seamlessly shared between animal and human health agencies to help local, state, national, and international health agencies track and respond to emerging outbreaks.

Research and technical development in partnership with academic labs that specialize in pathogen detection will help augment and advance wastewater surveillance efforts for use in monitoring emerging pathogens. Specifically, the following research needs are high priority:

• Ongoing assessment of the suitability of wastewater surveillance for newly emerging pathogens. Research that assesses whether an emerging pathogen is shed into wastewater by infected persons and can be detected in wastewater, and whether such data add value beyond existing testing and

• surveillance methods, will help identify pathogens for further method development.
• Continued development and improvement of analytical methods for suspected emerging agents of concern in wastewater, including enrichment approaches, detection thresholds and detection confirmation criteria. Many of the known emerging pathogen threats have available analytical assays, but these analytical methods have not been optimized for wastewater. Additional work on suspected agents of concern is needed to ensure that appropriate wastewater analysis methods are available when needed and their sensitivity and specificity understood across a range of wastewater matrix characteristics.
• Development of systems for rapid sentinel-site data analysis and integration with other disease reporting systems.
• Continued refinement and optimization of the siting of emerging pathogen surveillance. As technologies and analytical approaches improve, leading to increased sensitivity of detection, surveillance of rare and emerging threats may become feasible at the wastewater treatment plant level. Ultimately, envisioned public health response strategies will dictate the best scale for sentinel-site sampling.

Research to adapt and apply nontargeted sequencing methods along with advanced statistical and machine learning–based analysis approaches to wastewater surveillance are expected to be fruitful for expanding future public health benefits. Target-agnostic methods, such as high-throughput short- and long-read DNA sequencing, are rapidly advancing but the approach in wastewater surveillance is currently costly and not routinely applied. Target-agnostic methods, when paired with creative analytical approaches and artificial intelligence, have the potential to provide information on population-level shifts and detect changes in targets of interest, including emerging and unknown pathogens and antimicrobial resistance genes. Such data, if collected in a structured manner and alongside complementary types of data, such as public health and vaccination data, might facilitate the use of machine learning–based approaches to develop predictive models based on these rich data types. This remains an exploratory research space with investment primarily from NIH and NSF, but given the rapid advances in artificial intelligence and machine learning–based analytics, it is expected to be a promising and potentially useful avenue for future investment and CDC should periodically review application to wastewater surveillance.

This page intentionally left blank.