Do Pharmaceuticals in the Environment Pose a Risk to Wildlife?

The vast majority of knowledge related to the question “To what extent do pharmaceuticals in the environment pose a risk to wildlife?” stems from the Asian vulture crisis (>99% decline of some species of Old World vultures on the Indian subcontinent related to the veterinary use of the nonsteroidal anti‐inflammatory drug [NSAID] diclofenac). The hazard of diclofenac and other NSAIDs (carprofen, flunixin, ketoprofen, nimesulide, phenylbutazone) to vultures and other avian species has since been demonstrated; indeed, only meloxicam and tolfenamic acid have been found to be vulture‐safe. Since diclofenac was approved for veterinary use in Spain and Italy in 2013 (home to ~95% of vultures in Europe), the risk of NSAIDs to vultures in these countries has become one of the principal concerns related to pharmaceuticals and wildlife. Many of the other bodies of work on pharmaceutical exposure, hazard and risk to wildlife also relate to adverse effects in birds (e.g., poisoning of scavenging birds in North America and Europe from animal carcasses containing pentobarbital, secondary and even tertiary poisoning of birds exposed to pesticides used in veterinary medicine as cattle dips, migratory birds as a vector for the transfer of antimicrobial and antifungal resistance). Although there is some research related to endocrine disruption in reptiles and potential exposure of aerial insectivores, there remain numerous knowledge gaps for risk posed by pharmaceuticals to amphibians, reptiles, and mammals. Developing noninvasive sampling techniques and new approach methodologies (e.g., genomic, in vitro, in silico, in ovo) is important if we are to bridge the current knowledge gaps without extensive vertebrate testing. Environ Toxicol Chem 2024;43:595–610. © 2022 The Authors. Environmental Toxicology and Chemistry published by Wiley Periodicals LLC on behalf of SETAC. This article has been contributed to by U.S. Government employees and their work is in the public domain in the USA.


INTRODUCTION
With rising living standards increasing the demand for livestock production and the growing and aging human population reliant on healthcare systems, pharmaceutical contamination of the environment is increasingly of concern (see Boxall et al., 2012;Wilkinson et al., 2022).Although environmental contaminants such as pharmaceuticals are just one of the many stressors faced by free-ranging species (e.g., along with habitat destruction, climate change, and disease), with 41% of amphibian species, 21% of reptiles, 13% of birds, and 27% of mammals listed as threatened with extinction by the International Union for Conservation of Nature (IUCN, 2022), it is important that we understand the potential of pharmaceuticals to affect populations.
Pharmaceuticals encompass a broad range of substances of synthetic or biological origin used to diagnose, treat, mitigate, or prevent disease or to promote well-being (also see Supporting Information, 1).They are characterized as substances that have the ability to stimulate, depress, or replace physiological functioning in a biological system.This implies that the underlying mechanism by which the drug functions needs to be present for a physiological effect to occur.Some compounds that are primarily thought of and used as pesticides (e.g., organophosphorus insecticides [OPs]) are also registered and used in veterinary medicine as livestock dips to parasites, thus expanding the definition of what is traditionally thought of as a pharmaceutical.This definition of "pharmaceuticals" generally implies that most drugs have similar effects in whatever biological system they have been placed, based on their mode of action.However, for some species the effects may be exaggerated or unexpected as a result of interspecific physiological differences, including differences in pharmacokinetics and/or pharmacodynamics (Toutain et al., 2010).
The principal source of pharmaceuticals in the environment is believed to be usage in human patients, livestock, and companion animals, which results in excretion of active pharmaceutical ingredients (APIs) and metabolites (Daughton & Ternes, 1999).Manufacturing and inappropriate disposal of pharmaceuticals and the presence of veterinary drug residues in carcasses of livestock represent additional exposure pathways that can lead to exposure of wildlife (see Supporting Information, 2, and Figure S2a-d).To date, research on pharmaceuticals and wildlife has largely focused on effects of nonsteroidal antiinflammatory drugs (NSAIDs) on avian scavengers, with some work also on exposure, hazard, and risk of euthanasia drugs, antidepressants, and synthetic hormones (Table 1), and the role of wildlife in the transfer of antimicrobial resistance (AMR; Arnold et al., 2014;Bean & Rattner, 2018;Shore et al., 2014).Geographically, the focus has been on areas of Europe, North America, South Asia, and South Africa (Kookana et al., 2014; see Supporting Information, 3, for further discussion).
Depending on the environmental fate and mechanism of action of pharmaceuticals, wildlife can be used as sentinels for environmental contamination, to monitor bioaccumulation processes and to act as bioindicators of potential adverse effects.Wildlife are an integral part of complex food webs including humans and livestock, and there is growing recognition for the One Health perspective, indicating a need to give greater consideration to pharmaceutical impacts on wildlife.Examples of direct relevance to human health include the spread of AMR by migratory birds (Blanco et al., 2020;Loucif et al., 2022;Navedo et al., 2021) and the negative impact of some veterinary drugs on avian scavengers (Cuthbert et al., 2014;Plaza et al., 2022) and coprophagous insects (Tonelli et al., 2020), all of which can have consequences on the health of humans and domestic animals, for example, transfer of AMR organisms.
Given the potential for pharmaceuticals to have therapeutic effects, side effects, or unexpected toxicity in nontarget wildlife, it is noteworthy just how many drugs are licensed for use and how little we know about hazard to wildlife.In the United States alone, the US Food and Drug Administration (USFDA) has approved over 1600 animal drug products and 20,000 prescription drug products for human use, which include one or more of the approximately 4000 different APIs (USFDA, 2021).A little over half (51%) of the drugs approved for use in veterinary medicine by the USFDA are also approved for use in humans (Scott et al., 2020).Notably there are no routine regulatory requirements for industry to perform tests in wildlife species for human (European Medicines Agency [EMA], 2006[EMA], , 2016;;USFDA, 1998aUSFDA, , 1998b) ) or veterinary (Veterinary International Conference on Harmonization, 2000Harmonization, , 2006) ) medicines.Typically, a phased approach (see EMA, 2006;USFDA, 1998aUSFDA, , 1998b) ) is used in risk assessment for pharmaceuticals (see Figure 1 for a simplified schematic of regulations and Supporting Information, 4, for a more detailed discussion), with the aim of the initial screening phase to approve those drugs used in very low volumes, while later phases use data from acute and chronic tests in aquatic vertebrates, aquatic invertebrates, algae, terrestrial invertebrates, TABLE 1: Known and potential pharmaceutical classes and individual drugs of concern for wildlife from the perspective of exposure (high volume of use, persistence/pseudopersistence), hazard (highly potent), and risk (known exposure pathway combined with exposure at a sufficient level to cause adverse effects) microbes, and plants.The need for tests in birds and mammals is considered on a case-by-case basis, based on considerations of potential exposure and acute toxicity.Registration of the NSAID flunixin by the USFDA as a transdermal formulation for cattle in the United States is one example of when testing in wildlife species was required because of concern around the toxicity of NSAIDs to avian scavengers (USFDA, 2017).
In the present review, we discuss current knowledge and research priorities involved with answering some of the key questions of the risk that pharmaceuticals in the environment pose to "wildlife."Although there are many varied definitions of "wildlife," we limited our scope to free-ranging amphibians, reptiles, birds, and mammals, as in Rattner (2009; although the definition varies among sources; see Supporting Information, 5).We focus on livestock dips, NSAIDs, euthanasia drugs, AMR, and antifungal resistance (AFR), drug effects on the microbiome of wildlife, the significance of exposure via wastewater, exposure hazard and risk for species other than birds, and 21stcentury approaches for assessment of exposure and hazard.

PESTICIDES AND OTHER PHARMACEUTICALS USED AS DIPS FOR LIVESTOCK
What is our current understanding of the topic?
For over a century, a variety of chemicals applied in the form of a topical dip or pour-on have been used for control of parasites or disease treatment in livestock; in some instances, they have posed a significant secondary exposure hazard to wild birds.For example, in South Africa, arsenic-based compounds were introduced in the 1890s for livestock dipping, and frequent dipping of cattle often became compulsory in an effort to control the tick vector of the protozoan parasite that causes East Coast fever (Ramudzuli & Horn, 2014).South African ornithological records indicate that the yellow-billed oxpecker (Buphagus africanus), heavily dependent on tick prey, failed to breed in that country between 1907 and 1941 (Stutterheim & Brooke, 1981).This observation, in combination with lethality data from an experimental trial in which red-billed oxpeckers (Buphagus erythrorhynchus) were fed ticks dipped in arsenic trioxide (Bezuidenhout & Stutterheim, 1980), suggests that the collapse of the yellow-billed oxpecker population in this region was likely due to the reductions in tick prey coupled with consumption of arsenic-poisoned prey (Stutterheim & Brooke, 1981).
In the United States, avian poisoning events related to topically applied OPs have been documented and studied in far greater detail.A 1982 field study demonstrated that black-billed magpies (Pica pica) and red-tailed hawks (Buteo jamaicensis) were killed by famphur used as a pour-on to control warbles (Hypoderm sp.) on cattle in the states of Washington and Oregon (Henny et al., 1985).Gizzard contents of many dead magpies included cattle hair containing famphur or its activated metabolite, famphur oxon, and brain acetylcholinesterase (AChE) activity of the dead magpies was markedly depressed.During peak magpie mortality, some 10 days after cattle had been treated with famphur, a dead red-tailed hawk was found nearby with brain AChE activity depressed by 87% and magpie remains with 21 µg/g famphur detected in its crop.Subsequently, five cases involving bald eagles (Haliaeetus leucocephalus), red-tailed hawks, and a great horned owl (Bubo virginianus) collected in Oregon, California, Iowa, and Colorado were also linked to topically applied OPs, with remains of dead livestock found in proximity to dead birds (Henny et al., 1987).Stomach or crop contents of birds contained hair, flesh, and hide of livestock that tested positive for famphur and, in one case, fenthion; brain AChE activity was markedly depressed in the raptors.The exposure route was complex, including exposure via cow hair to magpies as an intermediary (sometimes cowbirds [Molothrus ater] or starlings [Sturnus vulgaris]) and finally to raptors.
In a large-scale study of avian scavenger poisoning between 2004 and 2013 in Spain, four cases involving bearded vultures (Gypaetus barbatus) tested positive for antiparasitics (three with diazinon, one with permethrin; Mateo et al., 2015), likely FIGURE 1: Simplified schematic of risk assessment for human and veterinary pharmaceuticals highlighting that specific testing with wildlife species is not routinely required but may be needed on a case-by-case basis.PEC = predicted environmental concentration; OECD = Organisation for Economic Co-operation and Development.
representing legal use of these veterinary pharmaceuticals.A large number of lamb feet collected from abattoirs and supplemental vulture feeding stations had detectable quantities of OP or pyrethroid insecticides.A relatively simple washing procedure greatly reduced residues and was proposed as a risk-mitigation measure to protect birds.There are other groups of pharmaceuticals used as pour-on formulations including some NSAIDs, such as flunixin (e.g., Banamine ® transdermal) used for control of pain and pyrexia associated with bovine respiratory disease (USFDA, 2017).Differential sensitivity among avian species is the hallmark of some NSAIDs (e.g., diclofenac [Rattner et al., 2008]), and flunixin may cause renal lesions in some avian species at a therapeutic dose extrapolated from that used in mammals (i.e., 1 mg/kg/day; Klein et al., 1994).However, a detailed environmental assessment of Banamine transdermal examining potential exposure pathways with calculation of risk quotients for nontarget wildlife (red-tailed hawk, coyote [Canis latrans]) concluded no significant impact from its proposed use (USFDA, 2017).
What are the future research priorities related to livestock dips and wildlife?
1. Evaluate the extent to which topical use of flunixin (and other NSAIDs if registered for that use) are hazardous to insectivorous birds.2. Determine the extent to which active ingredients, in some cases thought of as plant protection products, that are also used as antiparasitics in veterinary medicine are a risk to wildlife.

NSAIDS AND SCAVENGERS
What is our current understanding of the topic?
One of the most recent instances of population-level effects in wildlife due to an environmental contaminant occurred in Old World vultures on the Asian subcontinent following their exposure to residues of diclofenac in the carcasses they fed on (see Figure 2).Oaks et al. (2004) was the first published study to make the link between diclofenac residues in liver and kidney of dead and dying vultures in Pakistan, with visceral gout, tubular necrosis, and renal failure which led to death and the observed population-level declines.Although diclofenac is now recognized as the cause, it took many years to reach this conclusion (Green et al., 2004;Oaks et al., 2004;Shultz et al., 2004;Swan, Cuthbert, et al., 2006).The exposure scenario was unpredictable and linked to the large number of cattle in the area due to the religious significance of cattle in Hinduism.Cattle are not slaughtered as in typical production systems and when ill were treated palliatively with NSAIDs, commonly diclofenac (a cyclooxygenase-2 and prostaglandin synthetase inhibitor).The toxicity of diclofenac to Gyps vultures was also unexpected, with it being extremely nephrotoxic (Meteyer et al., 2005).This resulted in estimated population declines for some species of 99% over a 15-year period from levels in the early to mid-1990s.In India, according to the IUCN (2022), white-rumped vulture (Gyps bengalensis), slender-billed vulture (Gyps tenuirostris), Indian vulture (Gyps indicus), and red-headed vulture (Sarcogyps calvus) are critically endangered, while Egyptian vulture (Neophron percnopterus) is listed as endangered, cinereous vulture (Aegypius monachus) is classed as near threatened, and bearded vulture (Gypaetus barbatus), griffon vulture (Gyps fulvus), and Himalayan vulture (Gyps himalayensis) are listed as of least concern.
Following the elucidation of the toxicity of diclofenac, meloxicam was introduced as a safe alternative, with countries on the subcontinent banning the manufacture and importation of diclofenac for use in veterinary medicine but not its use (Swan, Naidoo, et al., 2006).As a result, initial usage of stockpiles caused ongoing losses of birds (Taggart et al., 2009).As meloxicam use became more prominent and diclofenac was phased out (see Cuthbert et al., 2014), certain areas on the Asian subcontinent have seen a degree of vulture population recovery, albeit not a return to their previous numbers (see Supporting Information, 6.1, for more detail on vulture population status and the Indian government's response).In areas where signs of total recovery were evident, the change has been attributed to migration and not true recovery (Galligan et al., 2014(Galligan et al., , 2020;;Paudel et al., 2016).Indeed, captive breeding programs successes have mainly been restricted to providing a sanctuary to prevent extinctions; only relatively small numbers of endangered vultures have been bred, and habituation of captive bred chicks limits success of releases (V.Naidoo, personal communication, June 21, 2021).Nonetheless, despite the positive benefits achieved with the removal of diclofenac, the safety of vultures was not fully protected; and other NSAIDs have since been identified as being toxic to birds, such as aceclofenac and nimesulide (Galligan et al., 2016(Galligan et al., , 2022)).In areas where these drugs are used, toxicity is being reported with continued loss of birds.A summary of the toxicity of diclofenac and other NSAIDs to other avian species is provided in Table 2, with further discussion of differential metabolism in Supporting Information, 6.2.There are a couple of notable instances of insensitive species; for example, New World vultures are apparently tolerant of diclofenac (Rattner et al., 2008; Table 2).Perhaps the most significant development on this topic in the last decade came in 2013, when diclofenac was approved for veterinary use in Spain and Italy, countries holding approximately 95% of all European vultures.In this instance, the potential for exposure (Herrero-Villar et al., 2020) combined with the known hazard could translate into risk to individuals and even populations.In fact, the first case of diclofenac poisoning in a wild vulture in Europe was detected in 2020 in a cinereous vulture in Spain (Herrero-Villar, Delepoulle, et al., 2021).
What are the future research priorities related to NSAIDs and hazard to birds?
1. Continue to monitor the use of diclofenac and other NSAIDs for treatment of livestock in the Old World and monitor the status of critically endangered vultures.2. Gain further understanding of exposure, hazard, and risk of NSAIDs for other avian species.3. Determine exposure (e.g., carcass surveys) and risk posed to vultures in Europe and Africa as a result of NSAID use in veterinary medicine.

EUTHANASIA DRUGS AND SCAVENGERS
What is our current understanding of the topic?
What are the future research priorities related to the risk posed by euthanasia drugs in the environment to wildlife?
1. Understand the extent to which accidental and intentional barbiturate intoxication occurs in geographic regions beyond Spain and North America using existing incident reporting databases, and if such schemes are not currently monitoring for barbiturates, start including them in the list of analytes.

AMR, AFR, AND THE MICROBIOME
What is our current understanding of the topic?
AMR.The increased use of antibiotics and antifungals in human and veterinary medicine has led to frequent detection of antimicrobials (and, to a lesser extent, antifungals) in the environment (Wilkinson et al., 2022).One of the most disconcerting effects of their presence is the development of resistant (AMR and AFR) organisms.In this context, scavenging birds have been identified as potential dispersants of AMR because of their likely exposure via domestic animal carcasses potentially treated with veterinary drugs and their capacity to travel long distances (Blanco et al., 2020).Some of the AMR organisms are zoonotic pathogens (i.e., Salmonella) that are a particular concern from a human health and food safety standpoint, and some of these organisms have been reported in vultures and wild ungulates (Blanco, 2018;Marin et al., 2018;Ramos et al., 2022).Other avian species have been described as being potential dispersants and reservoirs of AMR, such as shorebirds, with compound exposure linked to use in aquaculture (Carroll et al., 2014;Navedo et al., 2021).Antimicrobial resistance has also been reported in wild ungulates, where the pathway of transfer is via direct contact with domestic livestock as a consequence of human disturbance  and habitat conversion to agricultural systems (Espunyes et al., 2021;Ramos et al., 2022).Some of these wild ungulates play an important role in transmission between environmental compartments, such as wild boar, with interactions between urban, agricultural, and natural environments (Torres et al., 2020).This has also been reported in wild rodents with close contact with farms (Arnold et al., 2016).This may indicate a potential for the spread of AMR organisms globally, with subsequent implications from a "One Health" perspective.
What is our current understanding of the topic?
AFR. Antifungal resistance needs to be studied further before the role of wildlife in the dispersal of resistant organisms can be understood.There are only a few reports that describe resistant fungi isolated from European mammals such as wild boar and hedgehog that are closely linked to urbanized habitats (Gnat et al., 2021;Rhimi et al., 2022).Triazoles are also used as pesticides, and, as such, development of cross-resistance to triazoles used in the treatment of human fungal diseases (Bowyer & Denning, 2014;Snelders et al., 2012) should be considered.
What is our current understanding of the topic?
Microbiome.The role of the microbiome in human health is an emerging topic (Valles-Colomer et al., 2019;Zheng et al., 2020) that may also be an important aspect of wildlife ecotoxicology.
Studies in laboratory animals and humans have shown changes in the enteric microbiome that are caused by antibiotics (Cho et al., 2012;Dethlefsen & Relman, 2011), and there is evidence that the enteric microbiome can also be affected by nonantibiotic pharmaceuticals (Maier et al., 2018).In captivity, the treatment of koalas for chlamydia with antibiotics caused changes in their intestinal microbiome which affected plant tannin degradation (Dahlhausen et al., 2018).In a hypothesis-driven research study, Thomason et al. (2017) demonstrated that ocular microbiomes of house finches were altered after antibiotic treatment; the subjects exhibited more severe Mycoplasma-induced conjunctival inflammation than untreated finches.To date, however, although the presence of AMR bacteria has been widely described across taxa throughout the world (see previous discussion), few studies have explored the effects of pharmaceuticals on the microbiome of wildlife.Exceptions include impacts of antimicrobials on resistant bacteria in the intestinal microbiota of mallard ducks (Anas platyrhynchos; Atterby et al., 2021), andPitarch et al. (2017) observed oral mycoses in avian scavengers exposed to antibiotics that may alter the host's normal microbiota composition and that can facilitate opportunistic pathogenic yeast growth to cause disease.
What are the future research priorities related to AMR/ AFR and the microbiome?
1. Gain a greater understanding of the role of wildlife in transferring AMR organisms.2. Develop an understanding of the significance of wildlife in transferring AFR. 3. Conduct controlled experiments to understand the enteric microbiome in different wildlife species and their implications for immune function.

IMPORTANCE OF WASTEWATER AS A PHARMACEUTICAL EXPOSURE PATHWAY FOR WILDLIFE
What is our current understanding of the topic?
Foraging directly on wastewater-treatment plants.Controlled studies designed to simulate exposure of European starlings foraging on invertebrates at wastewater-treatment plant (WWTP) trickling filter beds to the antidepressant fluoxetine have been conducted in the United Kingdom (Bean et al., 2014(Bean et al., , 2017;;Whitlock et al., 2018Whitlock et al., , 2019)).These studies found indications that environmentally realistic concentrations administered via invertebrates injected with fluoxetine for approximately 6 months may cause subtle effects on foraging (e.g., timing and frequency of foraging bouts [Bean et al., 2014]) and courtship behavior (Whitlock et al., 2019).At present, the importance of this exposure pathway (i.e., birds eating fluoxetine-contaminated invertebrates from WWTPs) and the biological significance of the effects of fluoxetine remains a knowledge gaps (i.e., do they translate from the laboratory to the field, behavior as a relevant apical endpoint; see Supporting Information, 7).Only the experiment of Whitlock et al. (2019) detected fluoxetine residues in free-ranging starlings.However, the residues were detected in feathers of starlings that had been grown in the wild (21 of 25 birds, up to 27 ng/g dry wt) but sampled after the birds had been brought into the laboratory and dosed with fluoxetine.As there was also fluoxetine contamination of feathers and the liver of one control bird, it was not clear whether the fluoxetine in these feathers was transferred during captivity (e.g., via contact with excreta or during preening) or was the result of exposure prior to capture.The importance of this exposure route and the risk posed remain to be determined.
Water-fish-osprey food webs.To date, two studies have examined exposure, potential effects, and trophic transfer of APIs and metabolites in the osprey food web.Between 2012 and 2015, water samples, blood samples from various species of fish commonly consumed by osprey, and blood samples from 40-to 45-day-old osprey nestlings were collected in Chesapeake and Delaware Bays and associated tributaries in the United States (Bean et al., 2018;Lazarus et al., 2015).Water and blood plasma samples were analyzed for more than 20 APIs or metabolites by liquid chromatography-tandem mass spectrometry.The antihypertensive diltiazem consistently exceeded detection limits in osprey nestling plasma samples from the Chesapeake region.In the Delaware region, the analgesic acetaminophen was detected in 75% of the osprey nestling plasma samples, and the NSAID diclofenac was detected in only 7% of the nestling plasma samples.Although the effect thresholds of these three APIs are unknown for ospreys, observed concentrations were well below the human plasma therapeutic concentration (HTC; 28% of the HTC for diltiazem and two to three orders of magnitude lower for acetaminophen and diclofenac).Overall, these data and predictions may indicate that the risk of therapeutic or toxicological effects associated with trophic transfer of APIs and metabolites to osprey nestlings in the Chesapeake and Delaware regions is low (see Supporting Information, 8, for further discussion of kinetics).
What are the future research priorities related to evaluating the significance of exposure to pharmaceuticals from wastewater?
1. Fill knowledge gaps around the significance of exposure to pharmaceuticals at or near WWTPs in regions other than Europe and North America.2. Understand the exposure and hazard for aerial insectivores foraging on insects that emerge from WWTP filter beds.3. Periodic monitoring/decadal reevaluation of pharmac eutical contamination of the environment using noninvasive methods in sentinel species.

EFFECTS ON WILDLIFE OTHER THAN BIRDS
Research related to pharmaceuticals in the environment has focused on exposure and effects in birds, leaving knowledge gaps on the risk of pharmaceuticals to amphibians, reptiles, and mammals.
What is our current understanding of the topic?Amphibians and reptiles.The detection of many classes of pharmaceuticals in wastewater effluent, groundwater, untreated drinking water, and runoff from concentrated animal feed operations (see Barnes et al., 2008;Bartelt-Hunt et al., 2011;Focazio et al., 2008;Roberts & Thomas, 2006) has led to extensive laboratory exposure studies, field monitoring, and modeling efforts to assess the risk to aquatic species, including reptiles and amphibians.Notably, direct evidence of pharmaceutical exposure through detection of parent compound or metabolites is generally lacking for free-ranging amphibians and reptiles.However, in laboratory studies the pharmaceutical 17α-ethinylestradiol (EE2) and several progestogens have been shown to evoke reproductive toxicity in Xenopus tropicalis and Xenopus laevis at environmentally relevant concentrations (Orlando & Ellestad, 2014;Safholm et al., 2014).The highly publicized findings of feminization and endocrine disruption in American alligators (Alligator mississippiensis) at Lake Apopka, Florida, USA, were principally attributed to chlorinated hydrocarbon pesticides and not pharmaceuticals (see Guillette et al., 2000); however, this triggered many other such investigations in reptiles and other wildlife, with the potential for such effects from pharmaceuticals like EE2 and diethylstilbesterol often mentioned (see Guillette & Edwards, 2008).Nile crocodiles (Crocodylus niloticus) were studied at a commercial crocodile farm downstream from a sewage-treatment plant in Brits, South Africa (Arukwe et al., 2015(Arukwe et al., , 2016)).Carbamazepine, EE2, galaxolide, and tonalide were detected in water at various locations in proximity to the farm using passive samplers, and more commonly studied contaminants (aliphatic hydrocarbons, aromatic hydrocarbons, metals, halogenated pesticides) were present in the liver of crocodiles, with correlative evidence potentially indicating effects on biotransformation and oxidative stress endpoints, and reproductive and endocrine pathways.
Aerial insectivores.Bats can also be exposed to pharmaceuticals when they forage on insects that emerge from sewage filter beds (Park & Cristinacce, 2006).Park et al. (2009) determined that concentrations of the synthetic estrogen EE2 were greater in insects collected around trickling filers than at sites over 2 km away but never actually demonstrated exposure of bats to pharmaceuticals.In highincome countries at least, activated sludge has been replacing trickling filter beds in recent years; and thus, the relevance and importance of this exposure route for aerial insectivores remains unknown.
Marine mammals.A recent review of pharmaceuticals and personal care products and their toxicity to aquatic organisms (Srain et al., 2021) describes effects on a range of aquatic invertebrates and fish but reports evidence from only one nonfish vertebrate, citing an in vitro study which suggests potential impacts on immune function in harbor seals (Phoca vitulina; Kleinert et al., 2018).
Otters feed almost exclusively in aquatic systems, which may be contaminated by both effluent discharge and runoff.Fish are the predominant prey of most of the 13 otter species worldwide, although for some species aquatic crustaceans are commonly consumed (Kruuk, 2006).Trophic transfer of pharmaceuticals to otters seems likely in view of evidence of contamination of freshwater invertebrates and fish (see Cerveny et al., 2021;Miller et al., 2019).As yet, there is no published evidence of internal pharmaceutical exposure of otters, but recent data from the LIFE Apex Project (Gkotsis et al., 2022) indicates that a wide range of antidepressants, drugs of abuse, and stimulants are present in apex predators, and their prey, including a metabolite of the analgesic metamizole in an otter from Germany, and screening of hair samples identifies NSAIDS (ibuprofen and diclofenac) in Eurasian otter (Lutra lutra) from the United Kingdom (Richards et al., 2011).
What are the future research priorities related to studying exposure and effects of pharmaceuticals in the environment to reptiles, amphibians, and mammals?
1. Despite a clear potential for exposure via trophic transfer from fish and invertebrates, research has yet to document the occurrence, accumulation, or risk from pharmaceuticals in aquatic and, particularly, marine mammals (Kleinert et al., 2018).2. There are still many knowledge gaps around the exposure and hazard of pharmaceuticals to reptiles and amphibians.
Evaluating the ability to use read-across data from fish and endocrine disruption screening studies (e.g., amphibian metamorphosis assay) would be helpful for risk assessment.

NONINVASIVE METHODS FOR EVALUATING PHARMACEUTICAL EXPOSURE OF WILDLIFE
What is our current understanding of the topic?
Opportunistic sampling of animals found dead or carcasses provided by hunters or animal control programs represents one method for evaluating exposure without invasive sampling.As previously documented, wildlife poisoning by pharmaceuticals has focused on avian scavengers exposed to highly toxic compounds such as NSAIDs and barbiturates (Pain et al., 2008;Russell & Franson, 2014;Wells et al., 2020).The catastrophic situation in Asia with diclofenac could potentially have been ameliorated sooner, preventing the vulture population crash.For example, the WILDCOMS network (www.wildcoms.org.uk) in the United Kingdom brings together a number of surveillance schemes which monitor disease and contaminants in vertebrates found dead, including predatory birds, otters, and cetaceans.Collection is necessarily ad hoc, and careful consideration must be given to potential biases (e.g., unequal probability of sampling certain demographics or locations) and variables which might confound interpretation (e.g., spatial distributions shift over time).
Other noninvasive sampling matrices such as feathers or hair and sampling food items such as invertebrates or ungulate carcasses represent options that adhere to the principles of the 3Rs (reduce, refine, replace).Supporting Information, Table S9, contains examples of some of the studies and schemes in place for contaminant exposure assessment in wildlife.
What are the future research priorities for exposure assessment of pharmaceuticals and wildlife?
1. Further investigate the link between internal exposure and residues in feathers and hair as a noninvasive matrices.2. Routinely monitor livestock carcasses left for scavengers and determine the awareness of stakeholders responsible for the disposal of medicated carcasses (i.e., veterinarians and farmers) about the impacts of drug residues for wildlife.3.There is a need to implement a global wildlife monitoring system that enables the correlation of nontarget wildlife intoxications with residues of emerging contaminants such as pharmaceuticals.

3Rs APPROACHES THAT COULD BE USED FOR EFFECTS ASSESSMENT
As outlined in Figure 1, industry is not routinely required to conduct Organisation for Economic Co-operation and Development guideline tests to assess the safety of pharmaceuticals for wildlife species.With thousands of active ingredients, conducting in vivo safety tests would be extremely costly and entail use of large numbers of vertebrates.For effects assessment, there are options with minimal vertebrate usage (e.g., Swan, Cuthbert, et al. [2006] demonstrated that treating only two birds provides sufficient replication to determine whether the toxicity of diclofenac was similar to that of the white-rumped vulture) of nonreleasable animals at wildlife rehabilitation centers (e.g., Galligan et al., 2020;Rattner et al., 2008).Figure 3 outlines some of the approaches that could be used to fill these data gaps (e.g., quantitative structure-activity relationships [QSARs], approaches looking at effects on the genome or transcriptome, indirect effects, in ovo and in vitro methods, evaluation of sublethal effects, identification of biomarkers that could be used in pathway-based approaches to assist read-across from preclinical mammalian data), while further discussion of priorities related to effects assessment is provided in Supporting Information, 10.
What are the future research priorities for effects assessment and wildlife?
1. Validation and eventual utilization of new approach methodologies to complement or replace safety data in mammals generated in preclinical trials.

CONCLUSIONS
Pharmaceuticals in the environment have been found to cause individual lethality and even population-level effects in wildlife as a result of unique exposure pathways and unexpected sensitivity.Much of our knowledge on the topic has focused on NSAIDs and birds as a result of the Asian vulture crisis.With thousands of drugs licensed for use and no regulatory requirement to conduct in vivo safety tests on a routine basis, it is important to employ noninvasive methods in the field and laboratory to prevent another such crisis involving wildlife.The key questions related to the topic are identified as follows (Textboxes 1-5).TEXTBOX 1: NSAIDs and wildlife, diclofenac and vultures, current status of population recovery, exposure and risk Almost two decades after the work of Lindsay Oaks and coworkers (2004), the risk posed by NSAIDs to avian scavengers remains a key research priority.Specifically, 1) documenting the current status of vulture populations in Asia, 2) determining the extent of illegal use of diclofenac in cattle on the Indian subcontinent (Galligan et al., 2021), 3) documenting mortality incidents in avian scavengers associated with diclofenac and other NSAIDs in other geographic regions, 4) determining the relevance of human use of NSAIDs as an exposure route for wildlife, 5) characterizing the hazard of other NSAIDs to other avian species, and 6) elucidating the mechanism of toxicity of NSAIDs in Old World vultures.

TEXTBOX 2: Antimicrobials and effects on microbiota
Antimicrobial resistance is a major challenge in human medicine but may be less relevant for wildlife because they are not intentionally treated with antimicrobials unless admitted to a wildlife sanctuary, which is a rare circumstance.Nonetheless, gaining a greater understanding of the role of wildlife in transferring AMR and AFR organisms is a data gap and research need.In addition, further research is needed to determine whether exposure to antimicrobials and antifungals at much lower doses (i.e., environmental concentrations or residues in treated food rather than therapeutic doses), together with the presence of resistant microorganisms, can affect the microbiome in wildlife.
TEXTBOX 3: The importance of trophic transfer from WWTPs Insectivorous and omnivorous wildlife (e.g., birds, bats) could theoretically be exposed to pharmaceuticals through direct dietary or dermal routes at or near urban WWTPs (particularly in developing countries with poor secondary or tertiary treatment works) and in rural settings where wastewater or biosolids are used as fertilizer to amend soils, fields, and woodlands (reviewed in Bean & Rattner, 2018).Although modeling efforts have identified physiochemical and pharmacokinetic properties (e.g., environmental persistence, log octanol-water partition coefficient, leachability, half-life) for which pharmaceutical exposure of wildlife has greatest likelihood, robust data (i.e., parent compound or metabolite detected in tissue or excreta) documenting such exposure are lacking.
TEXTBOX 4: To what extent are pharmaceuticals in the environment affecting populations and the diversity of reptiles, amphibians, and mammals?
At present, much of the research on wildlife and pharmaceuticals has been on birds.Global biodiversity is changing rapidly, and initiatives to reduce or halt losses have thus far had limited success (Jetz et al., 2019).Causative agents are likely to be multiple, simultaneous, and potentially synergistic, making direct links between population-level change and specific toxicological threats difficult to decipher.Amphibians are likely to be particularly at risk because of their permeable skin and their reliance on both aquatic and terrestrial habitats at different life stages.Reptiles are the least studied group of vertebrates with regard to environmental contaminant exposure (Hopkins, 2000).Pharmaceutical risk to this vertebrate class remains largely unexplored (although see Mesak et al., 2019).Despite potential trophic transfer of pharmaceuticals by fish and invertebrates, very little research has yet focused on the occurrence, accumulation, or risk from pharmaceuticals in marine mammals (Kleinert et al., 2018).
TEXTBOX 5: Development of specific noninvasive tools for exposure and effects assessment Development of tools that enable exposure assessment (e.g., feather and hair, expansion of wildlife incident reporting schemes) and the use of new approach methodologies for effects assessment (e.g., genomic and transcriptome analysis, development of wildlife-specific QSARs) would be highly beneficial to minimize the likelihood of another Asian vulture crisis.Ideally, such approaches would involve noninvasive or alternative (to vertebrate) methods.
Supporting information-The Supporting Informationis available on the Wiley Online Library at https://doi.org/10.1002/etc.5528.

FIGURE 2 :
FIGURE 2: Timeline of Asian vulture crisis and research in the years that followed identification of diclofenac as the cause.LD50 = median lethal dose; bw = body weight.

FIGURE 3 :
FIGURE 3: In vitro, in vivo, and modeling approaches for filling data gaps for pharmaceutical effects assessment in wildlife.QSAR = quantitative structure-activity relationship.

TABLE 2 :
Collated evidence from the literature of exposure, hazard, and risk of nonsteroidal anti-inflammatory drugs to birds