Licensing and Approval of Antimicrobial Agents for Use in Animals

ABSTRACT

The importance of antimicrobial resistance and the urgent need to combat it has increased the already existent complexity of licensing and approval of antimicrobial agents for use in animals due to its possible impact on animal and public health. VICH—the International Cooperation on Harmonisation of Technical Requirements for Registration of Veterinary Medicinal Products—is the trilateral (European Union-Japan-United States) program that has the goal of harmonizing technical requirements for veterinary product registration. This article aims to describe the data requirements and testing necessary to build a registration file to obtain marketing authorization for a new antimicrobial agent for use in animals. This information is needed in the context of the risk assessment framework currently used in the approval of veterinary medicinal products containing antimicrobial substances. This framework considers the consequences of the uncontrolled quality of the antimicrobial product, the direct exposure of people to the antimicrobial product (human occupational safety and consumer safety), inadvertent exposure of organisms to the antimicrobial product (environmental safety), the antimicrobial product causing harm in the treated animals (target animal safety), and failure to achieve claims (efficacy). Approved veterinary medicines need to have a clear positive benefit associated with their use because of the risk to public health, animal health, and the environment. However, the presence of antimicrobials in the environment exerts a selective pressure for resistance genes in bacteria, and there is growing worldwide concern about the role of polluted soil and water environments in spreading antimicrobial resistance and the role of the contaminant resistome due to food-producing animal antimicrobial treatment. Additionally, the international developments regarding the categorization of critically important antimicrobials with the possible restrictions of use and the monitoring and surveillance of antimicrobial resistance in animals are reviewed.

INTRODUCTION

Veterinary medicines and vaccines are indispensable for the treatment and prevention of farm animal and pet diseases around the world. To ensure that these medications are high quality and appropriately produced, countries require that animal health medicines are manufactured to specific standards of quality, with proven safety and efficacy. The responsible authority in a given country must authorize that a veterinary medicine can be manufactured, sold, and used. The marketing authorization, also known as “registration” or “license,” implies that the responsible authority has approved not only the product to be marketed, but also the conditions that will characterize the use of the product. These conditions become part of the labelling, packaging, and information leaflets of the product and include (i) the characteristics of the active substance, its purity and concentration, and the complete composition of the medicinal product; (ii) the pharmaceutical form in which the medicine will be delivered (e.g., tablet, powder, cream, solution for injection), and the way it will be administered to the animal (e.g., injection, by mouth, in feed or water, or by topical application); (iii) the animals for which it is intended to be used, including specific ages and weights where relevant; (iv) what diseases or conditions it can be used to prevent, treat, or control, also known as the “indications”; (v) the dosages for each indication, the duration of treatment, and the withdrawal period, which means the number of days the medication must be withheld from farm animals prior to their produce entering the food system; and (vi) other circumstances regarding its use, including storage, shelf life, safety warnings, disposal instructions, and possible contra-indications. The sponsor develops this data package (or “application” or “dossier”) according to the testing requirements and standards of the authorizing legislation in place in the jurisdiction. Authorised veterinary medicines have a clear positive benefit associated with their use while considering the risk to public health, animal health, and the environment. The authorization and production of veterinary medicinal products also require manufacturing controls for the active substance(s) and the final product and sampling and testing of products. Regular manufacturing site inspections and pharmacovigilance ensure continued monitoring once the medicine has been authorised and is being marketed. Additionally, many countries have monitoring systems in place to ensure compliance with maximum residue limits (MRLs). This means that any residues of the medicine in food from its use in food-producing animals must remain below the established safe levels for the consumer.

International food standards are developed by the Codex Alimentarius, which was established in 1962 by the Food and Agriculture Organization of the United Nations and the World Health Organization (WHO) (1). The Codex international food standards, guidelines, and codes of practice are voluntary for countries to reference or implement as part of their national regulations. While the Codex Alimentarius Commission is responsible for food safety from primary processing through to consumption, the World Organisation for Animal Health (OIE) is responsible for setting standards in the domains of animal health and veterinary public health, including animal production food safety, to manage risks arising from the farm through to primary processing. The OIE and Codex collaborate closely in the development of standards relevant to the whole food production continuum. As a consequence of world trade globalization, a trend toward harmonization of the data requirements for authorization of veterinary medicines has occurred because approved veterinary medicines must satisfy the concerns about public health, animal health, and the environment. VICH—the International Cooperation on Harmonisation of Technical Requirements for Registration of Veterinary Medicinal Products—is the trilateral (European Union-Japan-United States) program that aims at harmonizing technical requirements for veterinary product registration (http://www.vichsec.org). VICH was officially launched in April 1996 by the animal health industry and regulators from the European Union, Japan, and the United States, with Australia, New Zealand and Canada joining as observers. It has the support and network of the OIE, and countries that are not part of VICH are invited to provide comments on the guidelines. It was based on a comparable international effort to harmonize technical requirements for human medicines called the International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use. The main objectives of VICH are (i) to harmonize regulatory requirements in the VICH regions to ensure high quality, safety, and efficacy standards; (ii) to provide a basis for broader international harmonization of registration requirements; (iii) to monitor and maintain existing VICH guidelines, and (iv) to encourage constructive technical dialogue between regulators and industry (http://www.vichsec.org). VICH does not usually address issues regarding the assessment of data. That important task is kept for the regulatory authorities in each of the VICH countries.

The approval of veterinary antimicrobials is based on the principles of risk analysis. The risks considered in the approval of veterinary antimicrobials are (i) the consequences of the uncontrolled quality of the antimicrobial product, (ii) the direct exposure of people to the antimicrobial product (human occupational safety and consumer safety), (iii) the inadvertent exposure of organisms to the antimicrobial product (environmental safety), (iv) the antimicrobial product causing harm in the treated animals (target animal safety), and (v) failure to achieve claims (efficacy). Antimicrobial resistance constitutes another hazard emerging from the use of antimicrobial veterinary products. Adequate preapproval risk assessment must include the potential increase in the number of resistant bacteria in the gastrointestinal tract of exposed animals due to the use of the antimicrobial product and the probability that humans will be exposed to the resistant bacteria or resistant determinants and that this exposure will result in adverse human health consequences.

This article will focus on the licensing and approval of antimicrobial agents for use in animals including current initiatives on antimicrobial resistance.

PREAPPROVAL INFORMATION FOR REGISTERING NEW VETERINARY ANTIMICROBIAL AGENTS FOR FOOD-PRODUCING ANIMALS WITH RESPECT TO ANTIMICROBIAL RESISTANCE

In 2003, harmonized technical guidance was agreed upon in the European Union, Japan, and the United States for registration of antimicrobial veterinary medicinal products intended for use in food-producing animals with regard to characterization of the potential for a given antimicrobial agent to select for resistant bacteria of human health concern (2). This guidance outlines the types of studies and data recommended to characterize the potential resistance development in the food-producing animal that may arise from the use of an antimicrobial agent. The necessary information includes the attributes of the drug substance, the drug product, the nature of the resistance, and the potential exposure of the gut microbiota (flora) in the target animal species. The VICH GL27 guideline recommends that sponsors provide basic information regarding (i) antimicrobial class (common name, chemical name, chemical abstract services registry number, chemical structure, and manufacturer’s code number and/or synonyms), (ii) mechanism and type of antimicrobial action, (iii) antimicrobial spectrum of activity, (iv) antimicrobial resistance (AMR) mechanisms and genetics, (v) occurrence and rate of transfer of AMR genes, (vi) occurrence of cross-resistance and coresistance, and (vii) pharmacokinetic data (2). The sponsor must provide information on the antimicrobial spectrum of activity, including data from MIC susceptibility testing against a wide variety of microorganisms or studies from the literature regarding (i) the target animal pathogens (as per product label claim) and (ii) foodborne pathogens (Salmonella enterica and Campylobacter spp.) and commensal organisms (Escherichia coli and Enterococcus spp.) (2).

The sponsor may present additional information regarding in vitro mutation frequency studies, antimicrobial agent activity in the intestinal tract, and other animal studies aiming to characterize the rate and extent (coselection due to linked resistance genes) of resistance development associated with the proposed use of the antimicrobial product (2). The overall submitted information should characterize the potential for the use of the veterinary medicinal product to select for antimicrobial-resistant bacteria of human health concern in terms of the exposure of foodborne pathogens and commensal organisms to the microbiologically active substance in the target animal after administration of the veterinary medicinal product under the proposed conditions of use (2).

SAFETY REQUIREMENTS FOR VETERINARY MEDICINAL PRODUCTS CONTAINING ANTIMICROBIAL SUBSTANCES

Target Animal Safety

International standards for studies for the demonstration of target animal safety for pharmaceutical veterinary medicines (for terrestrial animals) have been agreed on by VICH in their guideline 43 (3). The aim of the studies is to provide information on the margin of safety of the medicine and to identify adverse effects associated with an overdose or treatment beyond the recommended duration. The designs of the studies are based on the known pharmacological and toxicological properties of the active pharmaceutical ingredient, based on data from preliminary studies in target and nontarget laboratory animals, and the proposed conditions of use of the medicine (formulation, dosing regimen, target species, and production type) (3). Target animal safety studies are conducted on relatively small numbers of experimental animals in observance of the 3Rs principle of animals in research (replacement: replace with nonanimal system or with phylogenetically lower species; refinement: lessen or eliminate pain or distress in animals; reduction: lower the number of test animals needed) (3). Specific reproductive safety studies are required for medicines intended for use in breeding animals and for products to be administered locally into the mammary gland (3–5).

In addition to laboratory safety studies, safety data are also collected from field studies intended to evaluate the effectiveness of the medicine under conditions of field use—that is, in a typical population with or at risk of the disease of interest and under normal conditions of care. These studies collect data from a much larger number of animals and may therefore detect adverse events occurring at lower frequency than could be observed in the target animal safety study or that are related to particular animal characteristics (3).

Similar principles apply for demonstration of safety in aquatic species, although with flexibility in study design and interpretation considering differences in practicality of clinical sampling and the possible need to extrapolate between different species and varying environmental conditions (6).

Environmental Safety

International harmonized guidance for conducting environmental impact assessments (EIAs) for veterinary drugs consists of a two-phase procedure stated in Ecotoxicity Phase I VICH Harmonized Tripartide Guideline GL6 and Ecotoxicity Phase II VICH Harmonized Tripartide Guideline GL38 (7, 8). In phase I, the intended use of the drug determines the potential for environmental exposure. It is anticipated that drugs with limited use and limited environmental exposure will have limited environmental effects. Consequently, they stop at phase I (7). Phase I also recognizes veterinary drugs that require a more extensive EIA under phase II. The phase I EIA for a veterinary drug uses a decision tree via which an applicant determines that their veterinary drug needs or does not need an EIA (7). High potential exposure is defined by the initial route by which the veterinary drug enters the environment. Two main ecosystems are targeted: veterinary drugs intended for treatment of species reared in the aquatic environment and for treatment of species reared in the terrestrial environment. Veterinary drugs introduced directly into the aquatic environment or used in animals raised on pasture have a greater potential to contaminate aquatic habitats and the terrestrial environment, respectively. If the veterinary drug is an ecto- and/or endoparasiticide, its ecotoxicological potential needs to be assessed by conducting aquatic and terrestrial effects tests, respectively, in phase II. The exposure level of both ecosystems are also addressed in the guidelines (7, 8). A veterinary drug released from aquaculture facilities with an environmental introduction concentration of less than 1 μg/liter (the level shown to have adverse effects in aquatic ecotoxicity studies with human drugs) may stop at phase I (9). If the predicted environmental concentration of the veterinary drug in soil is less than 100 μg/kg, then the EIA for the veterinary drug may stop in phase I (9, 10). For a veterinary drug that has no possible mitigation measures aimed at reducing its aquatic environmental introduction concentration or predicted environmental concentration in soil, phase II guidance provides recommendations for standard datasets and conditions for determining whether more information should be generated (8). The phase II guidance contains sections for each of the major branches: (i) aquaculture, (ii) intensively reared terrestrial animals, and (iii) pasture animals, each containing decision trees. Figure 1 shows the decision tree/flow diagram for veterinary drugs used for aquaculture according to ecotoxicity phase II VICH Harmonized Tripartide Guideline GL38 (8).

FIGURE 1.

Partial decision tree/flow diagram for veterinary medicinal products used for aquaculture according to the Ecotoxicity Phase II VICH Harmonized Tripartide Guideline GL38 (8).

The presence of antimicrobials in the environment exerts a selective pressure for resistance genes in bacteria, and thus the importance of the environment as a reservoir for AMR genes is now widely recognized (11). The cycling of resistance genes between the different ecosystems deserves great attention; in particular, further consideration should be given to the contribution of veterinary antimicrobial use to the environmental resistome. Assessing the risk of AMR transmission from the environment to humans is also difficult, given that a nondirect proportionality between abundance and risk may exist (11). There are currently no legal requirements to monitor or prevent possible effects of antibiotic residues on the development or spread of AMR in the environment. Yet worldwide concern is growing about the role of polluted soil and water environments in the spread of AMR, and frameworks that regulate environmental risk assessment are considering the feasibility of including a risk assessment of AMR in the environment in their assessment schemes and thus of establishing the role of the contaminant resistome due to food-producing animal antimicrobial treatment.

Human Food Safety

The safety of food containing residues of veterinary drugs is generally assessed in laboratory animals (12). The requirements for toxicological testing of veterinary drugs are based on the toxicological tests for human medicines, food additives, and pesticides. International guidelines indicate the tests particularly relevant to the identification of a no-observed-adverse-effect level for veterinary drugs. These are (i) basic tests required for all new drugs used in food-producing animals to assess the safety of drug residues in human food (repeat-dose toxicity testing, reproduction toxicity testing, developmental toxicity testing, and genotoxicity testing), (ii) additional tests that may be required depending on specific toxicological concerns such as those associated with the structure, class, and mode of action of the drug (testing for effects on the human intestinal microbiota, pharmacological effects testing, immunotoxicity testing, neurotoxicity testing, and carcinogenicity testing), and (iii) special tests which might assist in the interpretation of data obtained in the basic or additional tests (12).

Repeat-dose (90-day) toxicity testing is performed in a rodent and a nonrodent species to assess the toxic effects on target organs based on repeated and/or cumulative exposures to the compound and/or its metabolites, the incidence and severity of the effect in relation to dose and/or duration of exposure, the doses associated with toxic and biological responses, and the no-observed-adverse-effect level (13, 14). Additionally, reproduction toxicity testing is performed to detect any effect on mammalian reproduction (male and female fertility, mating, conception, implantation, ability to maintain pregnancy to term, parturition, lactation, survival, growth and development of the offspring from birth through to weaning, sexual maturity and the subsequent reproductive function of the offspring as adults) (15). Further basic testing involves developmental toxicity and genotoxicity testing (16, 17).

Among additional testing for specific toxicological properties of a veterinary drug, testing for effects on the human intestinal microbiota of antimicrobial compounds involves the determination of the effects of residues of the drug on the human intestinal microbiota (17). The ecology of the intestinal microbiota may be potentially altered by the ingestion of an antimicrobial drug. A drug may reach the colon either due to incomplete absorption or by the enterohepatic cycle. VICH guideline 36 outlines the steps for determining the need for establishing a microbiological acceptable daily intake (ADI) and how to derive those recommends test systems and methods for determining no-observable-adverse-effect concentrations (NOAECs) and (no-observed-adverse-effect levels) for the endpoints of health concern (18).

A harmonized approach to determine the threshold dose that might adversely disturb the human intestinal microbiota has not been established. International regulatory bodies have used a formula-based approach that takes into consideration relevant data including MIC data against human intestinal bacteria for determining microbiological ADIs for antimicrobial drugs (18). If the endpoint of concern is the disruption of the colonization barrier, the ADI may be derived from MIC data, fecal slurries, and semicontinuous, continuous, and fed-batch culture test systems as shown in Table 1. If the endpoint of concern is an increase in the population(s) of resistant bacteria to establish a microbiological ADI, NOAECs derived from semicontinuous, continuous, and fed-batch culture test systems may be used as in the previous case (18).

TABLE 1.

Derivation of ADI from in vitro data^a

ADI derived from MIC data	ADI derived from other in vitro test systems
ADI = MICcalcb × mass of colon content (220 g/day)d/ (fraction of oral dose available to microorganisms)e × (60 kg person)	ADI = NOAECc × mass of colon content (220 g/day)d/ (fraction of oral dose available to microorganisms)e × (60 kg person)

^a Source: reference 18.

^b The MICcalc is derived from the lower 90% confidence limit for the mean MIC₅₀ of the relevant genera for which the drug is active.

^c The NOAEC derived from the lower 90% confidence limit for the mean NOAEC from in vitro systems should be used to account for the variability of the data. Therefore, in this formula uncertainty factors are not generally needed to determine the microbiological ADI.

^d The 220-g value is based on the colon content measured from accident victims.

^e The fraction of an oral dose available for colonic microorganisms should be based on in vivo measurements for the drug administered orally.

Residues of veterinary drugs, namely antimicrobial agents, in food are routinely evaluated for effects following chronic exposures, and a corresponding ADI is established, as described previously. Yet sometimes there is a potential for veterinary drug residues to cause adverse effects in humans following only a single meal. The ADI in such cases may not be the most appropriate value for quantifying the level above which a single exposure (after a single meal or during 1 day) can produce adverse effects (19). The acute reference dose is the appropriate approach. The acute reference dose is the estimate of the amount of a substance in food or drinking water, expressed in milligrams of the chemical per kilogram of body weight, which can be ingested in a period of 24 h or less without appreciable health risk to the consumer (19, 20). Regarding acute effects of antimicrobial agents on the human intestinal microbiota, the most relevant microbiological endpoint for acute exposure would be disruption of the colonization barrier. It is considered that a single exposure is unlikely to provide the selective pressure necessary to change the susceptibility of the bacterial population within the microbiome (20).

Maximum Residue Limits

The various toxicological tests previously described provide data for the establishment of a toxicological and microbiological ADI. The overall ADI (pharmacological, toxicological, and microbiological), generally expressed as microgram (μg) or milligram (mg)/kg body weight per day, is the amount of antimicrobial residues that can be consumed by an adult daily for a lifetime without appreciable risk to human health (21). The overall ADI provides the basis for determining the MRL of an antimicrobiological agent in a treated animal intended for human consumption. The MRL is the maximum concentration of residue accepted in a food product obtained from an animal that has received a veterinary medicine or that has been exposed to a biocidal product for use in animal husbandry (21). In the European Union/European Economic Area, the assessment of the safety of residues is carried out by the European Medicines Agency (EMA) Committee for Medicinal Products for Veterinary Use (CVMP) and in the United States by the Food and Drug Administration (FDA). The European Union requires by law that foodstuffs, such as meat, milk, and eggs obtained from animals treated with veterinary medicines or exposed to biocidal products used in animal husbandry must not contain any residue that might represent a hazard to the health of the consumer. Once the substances have been assessed and following the adoption of a commission regulation confirming the classification of the substances, the substances that may be used are listed in Table 1 of the annex to Commission Regulation (European Union) no. 37/2010. An external database (https://www.globalmrl.com/home) contains maximum acceptable levels of pesticides and veterinary drugs in food and agricultural products in the United States, as well as 70 other countries, the European Union, and the Codex Alimentarius Commission (https://www.fas.usda.gov/maximum-residue-limits-mrl-database).

EFFICACY REQUIREMENTS FOR VETERINARY MEDICINAL PRODUCTS CONTAINING ANTIMICROBIAL SUBSTANCES

Among the larger jurisdictions, only the European Union provides a specific guidance for industry on the design of pharmacological and clinical studies to support demonstration of the efficacy of antimicrobial veterinary medicines (22). An applicant for a marketing authorization for such a product must provide these data to support a specified “claim”/“indication.” The indication concerns a clinical disease associated with named bacterial pathogens in a particular animal species. In the European Union, indications are usually related to “treatment” alone or in association with “metaphylaxis” (22). Treatment relates to administration of a product to animals that are showing clinical signs of disease, whereas metaphylaxis is administration to animals in close contact and presumably infected but not yet showing signs of disease (22). Prevention claims (administration to healthy animals to prevent infection) are nowadays rare and only considered when the risk for infection is high and consequences are severe. The definitions and terminology differ between different areas of the world.

In accordance with European Union guidance, preclinical studies are required. Pharmacodynamic and pharmacokinetic data that are used to support the indications and dosing regimen for the product (22). The pharmacodynamics data should include information on the mechanism of action of the antimicrobial and MIC data determined using an accepted standardized methodology. The derived MIC distribution should include a sufficient number of isolates of target pathogen(s) obtained from recent European Union clinical cases, and ideally the epidemiological cut-off value should be available (22). Pharmacokinetics studies should provide data on the bioavailability of the product according to its route of administration and the concentration of the antimicrobial in the plasma and, if feasible, at the site of infection over time. These data can be used to explore a pharmacokinetics/pharmacodynamics relationship, which can be used to support the dosing regimen (22).

Dose determination studies investigate the dose level, dosing interval, and number of administrations of the product required. They are usually conducted in a small number of experimental animals that have been challenged with a strain of the target pathogen known to elicit a predictable level of disease. The efficacy of the selected dosing regimen is then corroborated in dose-confirmation studies, preferably using naturally infected animals, but conducted under well-controlled conditions.

Finally, the effectiveness and safety of the proposed antimicrobial medicine are investigated in clinical field trials. These involve a much larger number of animals, representative of the target population, under conditions of a natural disease outbreak. Clinical field trials should be multicentric randomized controlled studies. In the European Union they may use a negative or placebo control group, but more often, for animal welfare reasons, the control product is an authorized medicine with the same indication. The presence of the target pathogens and their antimicrobial susceptibility should be determined prior to treatment in diagnostic samples from a representative number of animals. The response to treatment is determined based primarily on the clinical response rate, which is evaluated on clinically relevant endpoints for the disease (e.g., pyrexia, respiratory and depression scores for swine respiratory disease). The microbiological cure rate is also an important (or often coprimary) endpoint. The outcomes of field trials must be statistically robust, and therefore it is important that studies are suitably designed and powered. Guidance is further provided on the statistical design and evaluation (23).

In addition to the above-mentioned guidance for antimicrobials, in the European Union and other VICH regions, clinical studies of all types of products should be conducted in accordance with VICH Harmonized Tripartide Guideline GL9 on good clinical practice (24). The purpose of the guideline is to establish guidance for the conduct of clinical studies that ensures the accuracy, integrity, and correctness of data, ensuring the welfare of the study animals and the effects on the environment and on residues in the edible products derived from food-producing study animals. Public access to information on clinical trial databases could be generalized; see as an example the European Union Clinical Trials Register (https://www.clinicaltrialsregister.eu/), which does not, however, include veterinary medicinal products.

Several jurisdictions (4, 5, 25) provide more detailed guidance on the data requirements for the demonstration of effectiveness of antimicrobial medicines intended for treatment and prevention of mastitis in cattle. In general, for subclinical mastitis, effectiveness is based on elimination of the udder pathogen and normalization of quarter somatic cell counts. Additionally, for clinical mastitis, resolution of local clinical signs of the udder and the quality of milk should be demonstrated. Prevention claims relating to protection from the establishment of new intramammary infections during the dry period can also be considered for products intended for dry cows.

RISK ANALYSIS FRAMEWORK IN THE APPROVAL OF VETERINARY MEDICINAL PRODUCTS CONTAINING ANTIMICROBIAL SUBSTANCES

The risk to public health from the development, emergence, and spread of resistance consequent to the use of antimicrobials in veterinary medicine is dependent on multiple risk factors (26, 27). Figure 2 summarises the chain of events that may lead from use of antimicrobials in animals to a compromised antimicrobial treatment in humans. The OIE and the Codex Alimentarius Commission have developed risk analysis guidance. VICH has developed guidance the data required to assess the potential for the use of new antimicrobial products to select for resistant bacteria of human health concern, as stated regarding the preapproval requirements (2).

FIGURE 2.

Chain of events that may lead from use of antimicrobials in animals to a compromised antimicrobial treatment in humans.

The OIE indicates that authorities regulating antimicrobials should have in place a policy framework for monitoring, measuring, assessing, and managing risk involved with the use of antimicrobials in food-producing animals. To assist risk assessors and risk managers, the OIE has provided guidance on the principles for conducting a transparent and objective risk analysis (28, 29). The OIE risk analysis process is divided into four components: hazard identification, risk assessment, risk management, and risk communication, based on the terminology of the Covello-Merkhofer system. The hazard is defined as “the resistant microorganism or resistance determinant that emerges as a result of the use of a specific antimicrobial in animals.” The resistant organism may itself be pathogenic or pass its resistance determinant to other organisms that are pathogenic. The risk assessment is based on scientific data, and the process is divided into four steps (Table 2). The OIE recommends that a qualitative risk assessment is always conducted as a preliminary evaluation. This will help to identify risk pathways which are feasible and those that can be discounted. The model should only be as complex as necessary to evaluate the risk management options available. It is acknowledged that a lack of appropriate data often prevents a complete quantitative risk assessment; alternatively, a semiquantitative approach can be adopted in which estimates of the probability and size of potential consequences are assigned to well-defined categories which can be combined into a severity score for the risk. This allows risks to be compared systematically and a threshold to be set for unacceptable risks.

TABLE 2.

Risk analysis frameworks in the approval of veterinary medicinal products containing antimicrobial substances^a

OIE hazard identification and risk assessment process	Codex Alimentarius Commission foodborne AMR risk assessment	OIE risk management process	Codex Alimentarius Commission foodborne AMR risk management process
Hazard identification: Resistant microorganism or resistance determinant that emerges as a result of the use of a specific antimicrobial in animals.	Hazard identification: Identification of the foodborne antimicrobial-resistant microorganisms or determinants of concern in the food commodity.	Risk evaluation: The process of comparing the risk estimated in the risk assessment with the reduction in the risk expected from the proposed risk management measures.	NA
Release assessment: Biological pathways that may lead to release of resistant microorganisms or resistance determinants into a particular environment due to the use of a specific antimicrobial agent in animals.Exposure assessment: Biological pathways necessary for exposure of animals and/or humans to the hazards released from a given source and the probability of the exposure occurring.	Exposure assessment: Detailing of the exposure pathways both preharvest, in the environment, and postharvest. The objective is to provide an estimate of the probability and level of contamination of the food product with AMR microorganisms at the time of food consumption.	Option evaluation: The evaluation of a range of risk management measures, which may include regulatory and non regulatory measures, such as development of codes of practice and disease control measures.	RMOs, pre- and postharvest risk factors, and approaches already adopted under good agricultural practice, good veterinary practices, good hygiene practices, and hazard analysis and critical control pointRegulatory controls on the use of veterinary antimicrobial medicines (limiting marketing authorizations), non regulatory controls (e.g., development of treatment and responsible use guidelines), and checking compliance of food products with microbiological criteria.
Consequence assessment: Describes the potential adverse health consequences, which may in turn lead to socioeconomic consequences, due to the specified exposures of humans or animals to resistant microorganisms. The probability of the potential consequences should also be estimated.	Hazard characterization: Considers the characteristics of the hazard, food commodity, and host to determine the probability of disease in humans and to estimate the additional frequency and severity of disease due to resistant pathogens, including the risk of treatment failure.	Development of an implementation plan.	Development of an implementation plan.
Risk estimation: Integrates the results from the release assessment, exposure assessment, and consequence assessment to provide an overall measure of the risks associated with the identified hazard(s).	Risk characterization: Considers the key findings from the first three steps to estimate the risk according to the risk manager’s needs, e.g., increased rates of hospitalization and mortality due to resistant infections, risks to sensitive subgroups, existence of alternative treatments.	Monitoring and review of the effectiveness of risk management measures.	Monitoring and review of the effectiveness of risk management measures, including evaluation against specific food safety metrics such as those used for national surveillance programs.

^a NA, not applicable; RMO, risk management option.

Although there should be communication between risk assessors and risk managers, the responsibilities should be kept separate to ensure independence of the evaluation of the risk and decision-making. Risk managers should have a policy framework that explains the risk management options that are available under the legislative and regulatory framework of the country and the level of risk deemed acceptable. The final step of the risk analysis is risk communication, defined as “the interactive transmission and exchange of information and opinion throughout the risk analysis process concerning risk, risk-related factors and risk perceptions among risk assessors, risk managers, risk communicators, the general public and interested parties.” It is noted that communication is essential to avoid failure of the risk analysis process (28, 29).

As previously mentioned, the Codex Alimentarius Commission develops international standards and guidelines to protect the health of consumers and to ensure fair practices in the food trade. In 2005, the Codex Alimentarius Commission produced its code of practice to minimize and contain AMR, which provides guidance for the responsible and prudent use of antimicrobials in food-producing animals (30). The code aims to minimize the potential adverse impact on public health resulting from the use of antimicrobial agents in food-producing animals. The code addresses in a summary manner the assessment of the efficacy of antimicrobials, the potential of veterinary antimicrobial drugs to select for resistant microorganisms, the establishment of ADIs and MRLs, and withdrawal periods for those products. The code of practice will be revised to address risk mitigation measures and other factors such as strategies that prevent or reduce the need to use antimicrobial agents, the use of lists of critically important antimicrobials, and the use of antimicrobials as growth promoters. Furthermore, the Codex Alimentarius Commission has developed a guideline providing a risk analysis framework to address the risks to human health associated specifically with foodborne AMR linked to nonhuman use of antimicrobial agents (30). The guideline address the risks associated with veterinary applications, plant protection, and food processing. The initial step of the framework is a scoping exercise in which an AMR food safety issue is identified by the risk manager followed by the development of an AMR risk profile to consider the context of the problem. This includes consideration of the food production chain, information on adverse public health effects, and available risk management options and leads to the establishment of a risk assessment policy.

The foodborne AMR risk assessment uses a science-based approach to identify the frequency and amount of AMR microorganisms to which humans are exposed through the consumption of food and the resulting magnitude and severity of the adverse health effects. To achieve this, the risk assessment is performed in four steps (Table 2). Risk management options should be evaluated with regard to their capacity to achieve an “appropriate level of protection” (WTO Agreement on the Application of Sanitary and Phytosanitary Measures) or other public health criterion (Table 2) (31). In keeping with the OIE guidance, the need for early communication between risk managers and risk assessors, and with consumer and industry representatives (producers, food processors, pharma), is recommended.

The Codex guidelines additionally highlight the importance of surveillance programs for the use of antimicrobials and prevalence of foodborne AMR in providing data for use throughout the risk analysis process.

Regional risk analysis frameworks for veterinary medicinal products also exist in the United States (32), the European Union (33), Canada (34), Australia (35), and Japan. (In Japan, the risk assessment for AMR arising from the use of antimicrobials in animals is performed by the Food Safety Commission of Japan at http://fsc.go.jp/english/index.html.)

The FDA’s methodology includes a qualitative approach to the risk assessment and provides guidance on the ranking of certain risk factors as high, medium, or low. A matrix is used to integrate the qualitative outcomes of the release, exposure, and consequence assessments into an overall risk estimation for the antimicrobial as having either low, medium, or high risk potential for human health due to selection of resistant foodborne bacteria associated with the use of the drug in food-producing animals (32). Whether a new antimicrobial animal drug is considered approvable is dependent on whether the FDA can conclude that “there is a reasonable certainty of no harm to human health when the drug is approved under specific use restrictions.”

The legislative framework for veterinary medicinal products in the European Union is currently under review. Key among the objectives of the review is to address the public health risk of AMR and to strengthen the benefit-risk assessment for antimicrobial veterinary medicinal products. In this respect, the CVMP, in collaboration with its Antimicrobials Working Party, have prepared draft guidance on the assessment of the risk to public health from AMR due to the use of antimicrobial veterinary medicinal products in food-producing animals (36). The guidance predominantly relates to the foodborne route of exposure, although transmission of AMR through direct contact by handling animals or animal produce should also be considered. The EMA’s risk assessment methodology is adapted closely from the OIE framework but also takes account of that provided by Codex and other regulatory jurisdictions. The CVMP Antimicrobials Working Party has also provided over the past decade a series of reflection papers addressing the use of certain antimicrobial classes in food-producing animals in the European Union and the development of resistance and its impact on human and animal health (37–39). Based on these papers, the CVMP has made recommendations which have been followed up, according to priority, by “class referral” procedures aimed at amending the summary of product characteristics (SPCs) of groups of related veterinary medicinal products to ensure that they are in line with the CVMP’s risk profiling and responsible use principles (40, 41). Following a referral procedure, a decision is issued by the European Commission (EC), requiring member states to implement the CVMP’s recommendations. In 2015 the EMA/CVMP published a reflection paper on the risk of AMR transfer from companion animals (42, 43). Although it was recognized that the use in companion animals of antimicrobials that are critically important for human health and the close contact between humans and pets increases the risks for transfer of important resistances, owing to the extensive knowledge gaps at the time, it was concluded that currently only an abbreviated risk assessment would be possible when approving veterinary antimicrobials for companion animals.

COMBATING ANTIMICROBIAL RESISTANCE

In addition to the risk analysis framework for veterinary medicinal products, many authorities provide general guidance on how to use antimicrobials to minimize the risks related to AMR. Novel main areas of awareness include the categorization of antimicrobials, responsible use guidance, and monitoring and surveillance strategies.

Critically Important Antimicrobial Categorization and Restrictions on Use

Lists categorizing antimicrobials have become popular, because they allow for targeting of risk management measures and provide overall recommendations on the prudent use of antimicrobials. Those recommendations take the lists as one of the factors, or the basis, for the recommendations on antimicrobial use and adapt them to the local situation. Lists of critically important antimicrobials (CIAs) provide a ranking of the antimicrobials currently used in medicine for humans or animals. For most risk assessors and regulators the most important criteria for the preparation of such lists is the impact of the use of those antimicrobials on public health, followed by animal health. Different institutions, such as the WHO, OIE, EMA, and FDA, have considered the impact of some of those factors and created lists ranking antimicrobials for use in animals; those lists might vary depending on the above-listed objectives and other factors such as the availability of medicinal products, including the pharmaceutical form in which they are available, or the area/region for which the lists are produced. In most cases the lists are based on the WHO list of CIAs for human medicine, highlighting the importance of the WHO list.

WHO list of CIAs

The WHO follows a One Health approach, so the list of Critically Important Antimicrobials is of use for public and animal health. According to the WHO, the list is intended to be used by “authorities, practicing physicians and veterinarians, and other interested stakeholders involved in managing antimicrobial resistance.” Most of the lists that have been produced, including the WHO list, aim to help prioritize risk assessment and risk management. Importantly, the WHO indicates that the list should not be “considered as the sole source of information to guide a risk management approach” (44).

Criteria for the ranking

According to the WHO list, the criteria for the ranking of antimicrobials are the following:

• Criterion 1: The antimicrobial class is the sole or one of limited available therapies to treat serious bacterial infections in people.

• Criterion 2: The antimicrobial class is used to treat infections in people caused by either (i) bacteria that may be transmitted to humans from nonhuman sources or (ii) bacteria that may acquire resistance genes from nonhuman sources.

In addition, the criteria for prioritizing the antimicrobials of the critically important category are as follows:

• Prioritization criterion 1: A high absolute number of people are affected by diseases for which the antimicrobial class is the sole or one of few alternatives to treat serious infections in humans.

• Prioritization criterion 2: There is a high frequency of use of the antimicrobial class for any indication in human medicine, since use may favor selection of resistance.

• Prioritization criterion 3: The antimicrobial class is used to treat infections in people for which there is evidence of transmission of resistant bacteria (e.g., nontyphoidal Salmonella and Campylobacter spp.) or resistance genes (high for E. coli and Enterococcus spp.) from nonhuman sources.

Categories of antimicrobials

The WHO divides antimicrobials into three categories: (i) critically important—antimicrobial classes which meet the first and second criteria, (ii) highly important—antimicrobials that meet one of the two criteria, and (iii) important—antimicrobials that do not meet either of the two criteria.

The WHO also includes a group of substances categorized as being highest-priority CIAs. These are those CIAs that meet all three prioritization criteria listed above.

As a summary, the classes of antimicrobials include:

• Critically important antimicrobials: Aminoglycosides, ansamycins, carbapenems (and other penems), cephalosporins (third and fourth generation), phosphonic acid derivatives, glycopeptides, glycylcyclines, lipopeptides, macrolides and ketolides, monobactams, oxazolidinones, penicillins (natural, aminopenicillins, and antipseudomonal), polymyxins, quinolones, and drugs used solely to treat tuberculosis or other mycobacterial diseases

• Highly important antimicrobials: Amidinopenicillins, amphenicols, cephalosporins (first and second generation) and cephamycins, lincosamides, penicillins (antistaphylococcal), pleuromutilins, pseudomonic acids, riminofenazines, steroid antibacterials, streptogramins, sulfonamides, dihydrofolate reductase inhibitors and combinations, sulfones, and tetracyclines

• Important antimicrobials: Aminocyclitols, cyclic polypeptides, nitrofurantoins and nitroimidazoles

• Highest-priority CIAs: Quinolones; third-, fourth-, and fifth-generation cephalosporins; polymyxins; glycopeptides; macrolides; and ketolides

WHO guidelines on use of medically important antimicrobials in food-producing animals

The WHO has recently published guidelines on the use of medically important antimicrobials in food-producing animals (45). According to the WHO, the guidelines are “evidence-based recommendations and best practice statements on use of medically important antimicrobials in food-producing animals, based on the WHO CIA List.”

The WHO recommendations are based on different themes and ranked according to their type of recommendation (e.g., strong) and the quality of the evidence. A summary of WHO recommendations is presented in Table 3.

TABLE 3.

Summary of WHO recommendations on use of medically important antimicrobials in food-producing animals^a

Theme (subject)	Recommendation	Type and quality of evidence
Overall antimicrobial use	We recommend an overall reduction in the use of all classes of medically important antimicrobials in food-producing animals.	Strong recommendation, low-quality evidence
Growth promotion use	We recommend complete restriction of use of all classes of medically important antimicrobials in food-producing animals for growth promotion.	Strong recommendation, low-quality evidence
Preventive use (in the absence of disease)	We recommend complete restriction of use of all classes of medically important antimicrobials in food- producing animals for prevention of infectious diseases that have not yet been clinically diagnosed.	Strong recommendation, low-quality evidence
Control and treatment use (in the presence of disease)b	We suggest that antimicrobials classified as critically important for human medicine should not be used for control of the dissemination of a clinically diagnosed infectious disease identified within a group of food-producing animals.	Conditional recommendation, very low-quality evidence
Control and treatment use (in the presence of disease)b	We suggest that antimicrobials classified as highest priority critically important for human medicine should not be used for treatment of food- producing animals with a clinically diagnosed infectious disease.	Conditional recommendation, very low-quality evidence

^a Source: reference 45.

^b To prevent harm to animal health and welfare, exceptions can be made when veterinary professionals judge that culture and sensitivity tests demonstrate that the selected drug is the only treatment option.

WHO best practices statements

In addition to the above recommendations, the WHO has produced two best practice statements. Of relevance for this article is the second one, “Medically important antimicrobials that are not currently used in food production should not be used in the future in food production including in food-producing animals or plants” (45).

OIE list of antimicrobials of veterinary importance

The OIE has produced a list of antimicrobials of veterinary importance (46). The list is the result of a questionnaire that was sent to the OIE member states and other institutions. The main difference between the WHO list and the OIE list is that the OIE list aims to establish the degree of importance for classes of veterinary antimicrobials.

Criteria for the ranking

The first criterion for the OIE list is the response rate to the questionnaire that was sent. The second criterion refers to the treatment of serious animal disease and the availability of alternative antimicrobials.

In line with the WHO list, the OIE list is divided between veterinary CIAs, veterinary highly important antimicrobials, and veterinary important antimicrobials. Taking into account the two above-listed criteria, the following categories were established:

• Veterinary CIAs: Those that meet both criteria 1 and 2

• Veterinary highly important antimicrobials: Those that meet criteria 1 or 2

• Veterinary important antimicrobials: Those that meet neither criteria 1 nor 2

The three categories of antimicrobials include the following classes of antimicrobials:

• Veterinary CIAs: Aminoglycosides, cephalosporins (all generations), macrolides, penicillins, phenicols, quinolones, sulfonamides (including trimethoprim), and tetracyclines

• Veterinary highly important antimicrobials: Ansamycin/rifamycins, fosfomycin, ionophores, lincosamides, pleuromutilins, and polypeptides

• Veterinary important antimicrobials: Bicyclomycin, fusidic acid, novobiocin, orthosomycins, quinoxalines, and streptogramins

The EMA categorization of antimicrobials

The EMA, following a request from the EC, has produced a categorization of antimicrobials for use in food-producing animals (47). The categorization was part of the answer to a request from the EC concerning the impact of the use of antibiotics in animals on public and animal health and measures to manage the possible risk to humans. For the categorization, the EMA assembled a group of experts, including experts on the use of antibiotics and resistance in humans. The group was named the Antimicrobial Advice ad hoc Expert Group. The group’s opinions, including the categorization, were adopted by the CVMP and the Committee for Medicinal Products for Human Use. Specifically, the answer to the question of categorization was adopted in December 2014. Two main factors were taken into account for the categorization: the need for antimicrobials in human medicine and the risk for spread of resistance from animals to humans. One of the main intentions of the ranking was to take into account the use of veterinary medicinal antimicrobials in the European Union and to adapt the recommendations to the specific conditions of the European Union.

Criteria for the ranking

As indicated above, the two main factors for the ranking were (i) the need for a specific class of antimicrobials in human medicine and (ii) the risk for spread of resistance from animals to humans. These two criteria were addressed as follows: hazard of zoonotic relevance (e.g., Campylobacter spp., Salmonella spp.), probability of resistance transfer (e.g., low or high), use in veterinary medicine (indicating if the substance is approved for use in the European Union and if it is authorized for group treatment), and information from member states’ marketing authorizations. For each antimicrobial class it was considered which are the bacterial targets in human medicine in the European Union for which the availability of a class/substance is critically important because there are few alternatives. For the classification of antimicrobial classes according to their probability of transfer of resistance genes and resistant bacteria, the following parameters were considered: vertical transmission of resistance genes, mobile genetic element-mediated transfer of resistance, coselection of resistance, potential for transmission of resistance through zoonotic and commensal foodborne bacteria, and evidence of similarity of resistance (genes, mobile genetic elements, and resistant bacteria). In addition, the recommendations were product independent and apply across the whole of the European Union independently of the animal health situation and of the availability of antimicrobial products for animals in individual member states.

Categories of antimicrobials

The EMA/Antimicrobial Advice ad hoc Expert Group categorization was divided into three categories:

• Category 1: Antimicrobials used in veterinary medicine where the risk for public health is currently estimated as low or limited

• Category 2: Antimicrobials used in veterinary medicine where the risk for public health is currently estimated as higher

• Category 3: Antimicrobials currently not approved for use in veterinary medicine

Category 1 includes substances which are considered the first choice in treatment guidelines and for which no specific associated hazards were identified to which people could be exposed from use in animals in the European Union. Category 2 includes substances that should be reserved for the treatment of clinical conditions which have responded poorly, or are expected to respond poorly, to other antimicrobials. The recommendations also indicate that “these reserved antimicrobials should be included in treatment guidelines only when there are no alternatives that could be used.” The recommendations also indicate that all efforts should be made to reduce the need for the use of category substances and to convince companies to seek marketing authorizations for alternative substances presenting less risk for public health. Category 3 includes antimicrobials currently not approved for use in veterinary medicine (but used for human medicine in the European Union). According to their legal status these substances may only be used by way of exception and only in companion animals (non-food producing species).

Lists of antimicrobials by category

Category 1 includes some classes of antimicrobial that have widespread use in veterinary medicine (48) and substances which are regarded as the first choice in animal treatments. These are certain penicillins, tetracyclines, and macrolides (polymyxins were initially included but later were moved to category 2). In addition, there is some limited use of rifampicin (a rifamycin) in veterinary medicine. Penicillins with a narrow spectrum of activity (e.g., penicillin G and penicillin V) belong with tetracyclines in a category in which the risk to public health is estimated as low.

In human medicine, certain macrolides (e.g., azithromycin) are being increasingly used in developing countries to treat invasive Salmonella spp. and Shigella spp. infections in humans, such as those caused by typhoidal salmonellae (e.g., Salmonella enterica serovar Typhi) or by Shigella dysenteriae type 1 (Shiga’s bacillus), when patients fail to respond to treatment with more conventional antimicrobials such as the fluoroquinolones. So far, use of these antimicrobials is limited in the European Union, and S. Typhi, S. enterica serovar Paratyphi, and S. dysenteriae 1 are not zoonotic hazards, but there is a need for awareness because in the future, macrolide-resistant Salmonella spp. other than typhoidal serovars may become a concern. For more information on the most extensively used polymyxin in veterinary medicine, i.e., colistin, see the response to the first request from the EC (49, 50). Currently, there are no recommendations to avoid the use of category 1 substances beyond what is stated by general responsible use principles. Nevertheless, these antimicrobials are not devoid of negative impact on resistance development and spread, and even if extensive use in veterinary medicine is to be expected, it is also important to ensure that any use is responsible. Category 1 substances might be of concern, e.g., if they facilitate spread of multidrug-resistant strains due to coresistance. This is a known problem for, e.g., livestock-associated methicillin-resistant Staphylococcus aureus where many antimicrobials, and in particular tetracyclines, (51), might collaborate to resistance selection. Likewise, resistance by coselection through the use of macrolides has also been involved in the persistence of vancomycin-resistant enterococci in livestock (39).

Category 2 includes antimicrobials used in veterinary medicine where the risk for public health is currently estimated as higher than the risk for category 1. Fluoroquinolones and third- and fourth-generation cephalosporins are of special concern. These antimicrobials have been used in some countries as the first-line treatment for a variety of infections in veterinary medicine. The EMA/CVMP Scientific Advisory Group on Antimicrobials (SAGAM) has provided risk profiles for fluoroquinolones and third- and fourth-generation cephalosporins (37, 38), and the CVMP concluded, among other recommendations, that an appropriate level of risk mitigation would be to reserve these agents for the treatment of clinical conditions which have responded poorly, or are expected to respond poorly, to other antimicrobials. These reserved antimicrobials should be included in treatment guidelines only when there are no alternatives that could be used. In some member states these category 2 substances are the only available choices approved for certain species and infections. In such cases, all efforts should be made to reduce the need for their use and to convince companies to seek marketing authorizations for alternative substances (including nonantimicrobial agents) presenting less risk for public health. The recommendations for these category 2 substances as reserved antimicrobials have been implemented in all summaries of product characteristics for veterinary medicinal products for food-producing species. For fluoroquinolones, regulatory actions have been taken (by the EMA), as they have for systemically active (parenteral and oral) third- and fourth-generation cephalosporins (40, 41). These actions have resulted in partial harmonization of relevant parts of the scientific literature concerning those medicinal products.

Aminoglycosides and some penicillins are classes of antimicrobials for which no risk profiling has yet been done by the EMA/CVMP. These classes have been added to category 2 based on the information available on criticality of use in human medicine and probability of spread of resistance. The EMA/CVMP/Committee for Medicinal Products for Human Use/Antimicrobial Advice ad hoc Expert Group recommends profiling the risk to public health related to the use of these classes in veterinary medicine. Future assessments could result in a change of the categorization.

Aminoglycosides are used extensively in veterinary medicine and are also given as oral group/flock medication; no use restrictions apply for this class. Because they may be effective against multidrug-resistant Enterobacteriaceae in humans and because the risk for spread of resistance from animals to humans is ranked as high, there might be a concern with the use of this class which is currently not addressed. To further elaborate on possible risks from aminoglycoside use in animals, a more detailed risk profile would be needed.

Penicillins are a diverse class that includes substances such as penicillin G and V, which have no activity against Enterobacteriacea, and substances that have an extended spectrum. Those with an extended spectrum could be a concern if their ability to facilitate spread of extended-spectrum beta-lactamases is similar to that of third- and fourth-generation cephalosporins. Therefore, a more detailed risk profile on penicillins with activity against Enterobacteriaceae is recommended. It is recommended to consider the diversity of the penicillin class when discussing the risk to public health from a veterinary treatment guideline perspective.

A number of the classes/substances listed are not currently approved for use in veterinary medicine, and these are classified as category 3. The extent of use of these classes would be low provided the restrictions detailed in articles 10 and 11 of Directive 2001/82/EC. According to these restrictions, they may only be used by way of exception and only in companion animals (non-food producing species), as MRLs have not been established to allow their use in food-producing animals.

Categorization may be considered one element when deciding on when/whether to use a certain class/substance in veterinary medicine, but it may not be used as the sole basis when creating treatment guidelines or when deciding on risk mitigation activities. It should not be interpreted as a recommendation for treatment guidelines. The categorization could also be taken into account when considering hazard characterization for the risk assessment in applications for marketing authorizations for veterinary medicinal products. Development and implementation of evidence-based national and regional treatment guidelines is encouraged.

FDA list of medically important antimicrobials

In 2003, the FDA published the Guidance for Industry number 152, “Evaluating the Safety of Antimicrobial New Animal Drugs with Regard to their Microbiological Effects on Bacteria of Human Health Concern,” for the evaluation of antimicrobial substances for food-producing species (33). Annex A of the this guidance provides the categorization of antimicrobials according to their importance for antimicrobial use.

Criteria for the ranking

The annex of the criteria for the ranking defines the criteria as follows:

• Antimicrobial drugs used to treat enteric pathogens that cause foodborne disease: The Infectious Disease Society of America guidelines on the treatment of diarrhea and other sources such as the Sanford Guide provide the drugs typically used in the treatment of foodborne diseases.

• Sole therapy or one of few alternatives to treat serious human disease or the drug is an essential component among many antimicrobials in treatment of human disease:

  Includes antimicrobials such as vancomycin and linezolid for methicillin-resistant S. aureus infections. Although they are not the “sole” therapy, they are one of only a few alternatives.

  This would also include a drug like polymyxin, which is one of few alternatives for multidrug-resistant Pseudomonas aeruginosa infections.

  Rifampin is not only a drug used to treat tuberculosis, but is also an essential part of the treatment regimen, because the cure rate is lower without it.

  Serious diseases are defined as those with high morbidity or mortality without proper treatment regardless of the relationship of animal transmission to humans.

• Antimicrobials used to treat enteric pathogens in non-foodborne disease: Enteric pathogens may cause disease other than foodborne illness.

• No cross-resistance within the drug class and absence of linked resistance with other drug classes:

  Absence of resistance linked to other antimicrobials makes an antimicrobial more valuable.

  Cross-resistance within antimicrobial classes and absence of linked resistance may change over time and will need to be updated periodically.

  In this context, “cross-resistance” refers to the transmission of resistant determinants between bacterial species or genera and does not refer to transmission of resistant organisms between animals and humans.

• Difficulty in transmitting resistance elements within or across genera and species of organisms:

  Antimicrobials to which organisms have chromosomal resistance would be more valuable than antimicrobials whose resistance mechanisms are present on plasmids and transposons.

  This does not refer to “ease of transmissibility” from animals to humans of the resistant pathogen because this is addressed elsewhere in the guidance in the release assessment.

Categories of antimicrobials

The annex of the FDA guidance classifies the antimicrobials (Table 4).

TABLE 4.

Categories of antimicrobials according to the annex of the FDA guidance^a

Category	Criteria	Classes of antimicrobials
Critically important	Antimicrobial drugs which meet both criteria 1 and 2	Third-generation cephalosporins, fluoroquinolones, macrolides, trimethoprim/sufamethoxazol
Highly important	Antimicrobial drugs which meet either criterium 1 or 2	Natural penicillins, penase resistant penicillins, antipseudomonal penicillins, aminopenicillins, fourth-generation cephalosporins, carbapenems, aminoglycosides, clindamycin, tetracyclines, glycopeptides, streptogramins, oxazolidones, pyrazinamide, rifamycins, chloramphenicol, metronidazole and polymyxin B
Important	Antimicrobial drugs which meet criterion 3 and/or 4 and/or 5	First- and second-generation cephalosporins, cephamycins, monobactams, quinolones

^a Source: reference 33.

Responsible Use

Guidelines for responsible antimicrobial use provide a set of recommendations for antimicrobial prescribers, and users, to optimize the use of antimicrobials in animals and minimize the risk of AMR from the use of antimicrobials in animals spreading to humans; such guidelines may, or may not, be driven and endorsed by national or international competent authorities or organizations. Recommendations are usually provided for animal species or for production groups (e.g., veal calves). Responsible use guidelines have been changing ever since the first findings of acquired AMR. Earlier recommendations such as “broadspectrum, combination of compounds and prolonged (oral) therapy to avoid secondary infections and relapses” are, following new insights, no longer justified. Randomized controlled trials, reviews, and meta-analysis have led to new responsible use recommendations. Currently, a narrow-spectrum, single compound with high loading dose and a minimum of therapy duration is estimated to be appropriate to maintain clinical efficacy while minimizing the selection and spread of antibiotic resistance (52). Curative treatment and (single-dose) surgical prophylaxis should be the standard, whereas purely preventive treatments should be avoided. Metaphylaxis is only appropriate in clearly defined circumstances when there is a potential for high morbidity due to rapidly spreading disease, as e.g., in current aquaculture and poultry production systems. Even in countries where antimicrobial growth promoters are banned, e.g., in the European Union as of 2006, prophylaxis via large oral group treatments remains routine practice in livestock (52). Whenever possible, individual treatment should be preferred to group or mass treatment, and this also applies for metaphylactic purposes (53). In addition, the use of antimicrobials as feed additives, still allowed in many European countries, causes cross-contamination at different levels and should be discouraged as much as possible (54). To justify appropriate off-label use and assess the public health risks involved (26), the mandatory reporting of off-label use in the future is inherent to responsible use of medication. Mandatory recording is already in place in many European Union member states. Oral group treatments are frequently under- and sometimes overdosed (26, 55), and underlying misevaluation of body weight or dosing instructions should be avoided.

Veterinarians require good advisory skills to address farmers’ questions, concerns, and needs related to antimicrobials (51). Adequate sampling and antimicrobial susceptibility testing of organisms at the site of infection is encouraged when symptoms are not indicative of a specific pathogen.

A delay in laboratory results has historically led to empiric broad-spectrum approaches, followed by de-escalation or switching agents. Such additional selection pressure might even aggravate the disorder due to unforeseen multiplication of pathogens due to inactivity and will extend the abundance of resistance determinants present. The development of rapid and reliable patient-side tests (56) should be encouraged. Some mild clinical infections, if supported by adequate diagnostic procedures, can even be cured without antimicrobials, as described for Rhodococcus equi infections in foals (57). At the herd level, sample size guidelines should be developed and training given, ideally by syndrome or disease complex. Independent and clinical examination of all animals involved remains the cornerstone of good and responsible veterinary practice and, thus, responsible use of antimicrobial drugs. In countries where monitoring and benchmarking of antimicrobial prescribing has been installed, a substantial decrease in consumption has been documented (58). Illegal use of antimicrobial compounds should be penalized.

Monitoring and Surveillance

For as long as patient-side and on-farm diagnostics do not identify pathogens and susceptibilities during clinical examination, antimicrobial therapy will be guided by the experience of the veterinarian and the history of the animals involved. Regular consultation of monitored resistance trends can further justify the administration of certain antimicrobial classes or compounds. This should not be limited to commensal bacteria (E. coli, enterococci) due to substantial differences between trends and resistance rates across production systems and organ systems and depending on whether pathogens are tested (55). In countries where antimicrobial consumption has decreased, monitoring programs made it possible to find some significant relationships with decreasing resistance (58, 59).

Reference centers can validate the susceptibility result in the function of the clinical breakpoint (MIC that bridges susceptible versus resistant isolates). Reference centers also should give guidance on the involvement of pathogens that cannot readily be cultured under routine laboratory conditions (e.g., Lawsonia intracellularis, Brachyspira spp., Mycoplasma spp., and Ornithobacterium rhinotracheale).

Marketing authorization holders should provide guidance on identification and susceptibility testing. Multiresistance monitoring should also be integrated in programs to adjust for linked resistance genes (plasmids, transposons, integrons) and horizontal gene transfer. These pathways can easily increase the spread resistance, but if the presence of linked genes and horizontal gene transfer is declining, they also can slow down and even lower the AMR reservoir (59). Defined targets for the reduction of antimicrobial use and benchmarking of farms should be aligned with surveillance of AMR at the appropriate level (farm, veterinary practice, production type, region, country). Zoonotic organisms (e.g., livestock-associated methicillin-resistant S. aureus) should be included (60–62). The role of the food industry and that of consumer organizations will likely increase with regard to surveillance and monitoring, including off-label use. Surveillance data of all kinds should be readily available for time-series research purposes, ideally in a global context (31).

CONCLUSION

The challenge of combating AMR has increased the already existent complexity of licensing and approval of antimicrobial agents for use in animals. This article has summarized the evolving data requirements and processes necessary to obtain the marketing authorization for a new antimicrobial agent for use in animals. Due to the possible impact of antimicrobial resistance on animal and public health, a risk assessment framework is currently used in the approval of veterinary medicinal products containing antimicrobial substances.