NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
National Research Council (US) Committee on Drug Use in Food Animals. The Use of Drugs in Food Animals: Benefits and Risks. Washington (DC): National Academies Press (US); 1999.
Today, Americans don't have to worry about safety or effectiveness when they buy [drugs and medical devices] from cough syrups to the latest antibiotics and pacemakers. The Food and Drug Administration has made American drugs and medical devices the envy of the world and in demand all over the world. And we are going to stick with the standards we have, the highest in the world. But strong standards do not mean business as usual.
President Clinton, 1995
National Performance Review
The history of federal government involvement in controlling, regulating, and assuring the quality of therapeutic drugs in the United States dates back to the mid-nineteenth century and the congressional enactment of the Drug Importation Act (to stop entry of adulterated foreign drugs into the United States). Subsequent to this, President Lincoln (1862) appointed a scientist to the U.S. Department of Agriculture (USDA) to begin the Bureau of Chemistry, which ultimately would evolve into the modern Food and Drug Administration (FDA). President Theodore Roosevelt signed into law the Pure Food and Drug Act, further defining aspects of adulteration and healthfulness of food and drug preparations. In 1927, an enforcement agency was authorized, the Food, Drug, and Insecticide Administration, which, in 1930, was renamed the Food and Drug Administration. A significant act of Congress was passed in 1938, the Federal Food, Drug, and Cosmetic Act (FDCA), which set standards for safety, efficacy, prevention of adulteration, tolerances, factory inspections, penalties, and seizures associated with drugs, cosmetics, and medical devices. FDA was transferred from USDA to the Federal Security Agency, and a final reorganization in 1968 placed it in the Public Health Service of the U.S. Department of Health and Human Services (DHHS).
The federal structure that oversees animal drug use is complex, extending through the Centers for Disease Control and Prevention (within DHHS), the Center for Veterinary Medicine (CVM, within FDA), the Food Safety and Inspection Service (within USDA), and the U.S. Environmental Protection Agency. Other organizations, such as the Agricultural Research Service (within USDA), work in support and they generate the necessary research data to answer questions of scientific importance in agriculture and human nutrition.
It is not main charge of this report to review the responsibilities of each federal agency that influences animal drug use, but the principal organization, CVM, is highlighted here because of its work in the approval and monitoring activities that govern animal drug use in the United States. Specific issues of monitoring drug use and drug residues are discussed in Chapter 5.
CVM has the important tasks of protecting society from harmful animal drugs and maintaining public confidence in the drugs that are in use. These objectives are achieved by ensuring that new drugs pass a rigorous approval process. CVM has pledged to be more active in its efforts to increase the availability and diversity of safe and effective animal drugs (CVM 1997b).
Under the larger structure of FDA, CVM regulates the manufacture and distribution of drugs and feed-additive drugs intended for food animals and companion animals. The structure of CVM has evolved to its present state after a series of reorganizations initiated with the change from the Bureau of Veterinary Medicine to CVM in the early 1980s. The current structure reflects the larger drug approval and monitoring process. In essence, the process through which a sponsor (a party interested in developing a drug to market) develops, manufactures, and markets a drug product is divided into two main categories: preapproval and after-market monitoring. The preapproval process is overseen by a section of CVM called the Office of New Animal Drug Evaluation, and the monitoring activities of CVM are overseen by the Office of Surveillance and Compliance. The CVM structure is shown in Figure 4–1.
CVM Organizational Structure. Source: htpp://www.fda.gov/dvm/fda/aboutdvm.html; May, 1998.
The regulations that govern all aspects of the drug approval, marketing, and monitoring process are detailed in the Code of Federal Regulations (CFR): Title 21, Parts 500 to 599 govern animal drugs, animal feeds, and associated products; Parts 200 to 299 govern registration, labeling, and good manufacturing practices; good laboratory practice (GLP) regulations are contained in Part 58. The GLP regulations are an essential part of the regulatory process. They contain the standards of uniformity for study conduct, and they influence the uniformity and validity of data needed to sustain the review process. Some aspects of the requirements for environmental assessment are found in Part 25. Updates to these procedures are printed in the weekly editions of the Federal Register.
To obtain approval to manufacture and sell a product, developers must contact CVM with an investigational new animal drug (INAD) application and, ultimately, submit a New Animal Drug Application (NADA). Because of the reading and interpretation of the regulations, an animal drug is technically “unsafe” without an approved NADA on file at CVM, and the use or sale of an unsafe compound is illegal. The director of the office of New Animal Drug Evaluation, within CVM, is responsible for evaluating data and information submitted by the sponsor in one of several submission modes to satisfy the requirements for intended-use effectiveness, animal safety, human drug-residue-consumption safety, environmental impact, manufacturing processes, and NADA completeness.
Armed with preliminary data to justify further processing, a sponsor seeks initial permission from CVM to conduct animal studies. This step constitutes an INAD application and provides CVM with the data and information necessary to evaluate the stated claims of safety and efficacy for the compound with respect to proposed studies. The INAD is critical because it permits unapproved drugs to be transported to sites where they can be used legally in animal evaluation studies and contains the information needed to obtain an investigation withdrawal time and permission to slaughter test animals. FDA is notified when the animals are slaughtered, and the animals must be slaughtered at a federally inspected facility. A USDA inspector is assigned to the facility. 1 Tissues from test animals may not be used for food. The protocol for clinical study or trial dictates that animals be slaughtered after allowing a withdrawal time for drug depletion.
A newly evolving feature of the approval process added flexibility to relationships between CVM and sponsor companies. The levels and timing of dialogue, communication, and the structure of the review are more flexible. CVM is developing new strategies to be more responsive to sponsors' schedules within the approval process by allowing sponsors to establish the nature of the communications and review process through which approval is sought. For example, initial discussions between CVM and the sponsor will establish early in the development and approval process the expectations for data submission (phased review of data and studies at critical development points versus review of the total data package) and field testing.
One fear in the U.S. animal and drug production industries and allied industries is that agriculture in the United States is in danger of losing its competitive edge to foreign interests because of the unusually long approval process. Drug products often are available in other countries before they are on the market here. However, it is an equally valid fear of regulatory agencies and the medical community that rushing the review process might jeopardize human health by allowing critical information to be overlooked and, thus, potentially introducing new problems to the nation. This is the price paid for ensuring a high level of confidence in the “unadulterated” condition of the food supply—a condition sometimes unattainable in other parts of the world. In recent years, the CVM's practices and procedures for reviewing submissions and approving products have been substantially revised but have maintained the goal of ensuring the human health. Sometimes problems that develop with the use of a given drug do not become apparent until the product has been on the market for a significant period. Oftentimes, neither reviewers, developers, nor manufacturers can determine whether an adverse reaction will develop in animals or humans. Therefore, it is critical that, after a drug has been approved and marketed, additional tracking information is collected on that product to ensure that new problems are detected quickly.
In the recent past, drug sponsors devoted 10 to 11 years in developing and obtaining FDA approval to market a drug. A substantial portion of that time was spent fulfilling CVM requirements and waiting for documents and responses to be evaluated and returned. As a major contribution to the “drug crisis,” the federal approval process was considered ill-defined and slow by sponsors and animal producers and added years of additional work and expense to the frustrating experience of seeking approval (AHI 1982, 1992). Of particular concern was the burdensome requirements to continuously revalidate safety and efficacy results for compounds and drug combinations already established for other intended uses and to conduct costly multiple field trials in several locations. In part associated with constituent pressure and in part associated with the Clinton Administration's larger goal of “reinventing government,” CVM initiated newly streamlined processes to decrease some of the cumbersome paperwork that slowed the approval process. The most recent contributions to streamlining are in the areas of drug availability and “extra-label usage,” which allows veterinarians to exercise their judgment to recommend uses for drugs beyond those specified on labels or package insert.
A comparison of the old and new approval processes, including schedules, is presented in Figure 4–2. The “traditional“ approval process often was cumbersome and unresponsive to sponsors. During a 6 to 8 year period, developers and manufacturers would initiate and conduct the experiments needed to generate the data for efficacy, target animal safety, toxicology, residue chemistry, manufacturing chemistry, and environmental impact. After these data were collected, the sponsor would enter into iterative negotiations with CVM, refining the requirements of data to support the approval. Often, the sponsor would need to undertake additional studies to satisfy the review process, and this added significantly— up to 4 years (AHI 1993)—to the time required to obtain approval. The new process is considerably more interactive with the sponsor, and it is more dynamic. The new process incorporates the iterative discussion phase early in the data-gathering and study-design phases and gives the sponsor immediate feedback on the requirements sought by the review. In the traditional system, the combined length of time of INAD plus NADA stages was as much as 12 years, with 3 to 4 years in the iterative NADA stage alone. In the re-engineered process, the initial INAD stage is comparable to that in the old process, but the NADA stage is reduced significantly to 90 days.
Comparison of the Traditional and Re-Engineered Approval Processes. Source: Adapted from a statement by Michael A. Friedman, M.D., lead deputy commissioner, FDA, DHHS, before the Subcommittee on Agriculture, Rural Development, and Related Agencies, Committee (more. )
Dispute settlement in the past often was time-consuming, because there was no formal process, and adversarial attitudes often added needlessly to the time required to reach a decision.
Revisions occurring within CVM are reducing the time a drug application stands in review before approval. During the past few years, the trend has been to shorten of the average time to NADA approval from 47.7 months in 1992 to 39.1 months in 1995 (Figure 4–3). Although FDCA calls for FDA to make a decision on NADA approval within 180 days, even with the improvements in review time, it now takes 6 to 7 times longer than that (AHI 1994). Data were unavailable to determine whether the shortening of time to approval between 1991 and 1995 resulted from the increased efficiency of the review process or from decreased filings. The adoption of the revised policies was phased in beginning 1993 and continues. Overall, there is a trend toward increased approval by CVM since 1991, and many of approvals are for new chemical entities or new species uses (Figure 4–4). Still another factor that influences the approval turnaround time is the overall increased quality of sponsor applications.
Effect of Re-Engineering the Approval Process on the Time to Approve New Animal Drug Applications. Source: CVM Summary of NADA Approvals, 1996.
Trends in Animal Drug Approvals since 1990. Source: DHHS, FDA, CVM Office of New Animal Drug Evaluation, 1996.
The aquaculture industry is one area of food-animal production that is truly hampered by a shortage of medications to treat diseases. Most compounds available are not antibiotics in the true sense but rather chemicals with limited specificity that are applied to the water. They include acetic acid, calcium salts, some vitamin preparations, herbals such as whole onion, formalin, oxytetracycline, and sulfa compounds. Treatments with nonapproved materials can be permitted for use by CVM through the filing of a “compassionate INAD,” which permits the use of a nonapproved drug when the only recourse would be death or euthanasia. The requirements of the compassionate INAD are that some measure of public safety be demonstrated for the application and that data are derived from the process to support a formal NADA. The compassionate INAD is issued for a specified number of animals and is valid for 1 year with refiling necessary for extension. CVM considers the compassionate INAD a sound measure for offsetting the shortage of approved drugs in aquaculture and in its recommendations for these uses weigh the potential for excessive use against the safety of the intended use.
Important new changes in many aspects of the approval process are aimed at clarifying the expectations of submitting and approving parties alike. Multiple presubmission conferences iron out the details of required documentation of efficacy and safety. New regulations on multiple and repeated field trials are being implemented to eliminate the redundancy of information processing, thus reducing the time and cost to conduct, summarize, and justify such trials by the sponsor and reducing CVM's need to review such documents.
In addition to the time required for approval, criticisms of the CVM policies also have focused on the cost and time associated with redirecting the use of a drug already approved for other purposes or species and the perception that finding flaws in approval applications is more important than is facilitating the approval process. There also are concerns that some of the requirements for accountability in manufacturing are too stringent and that the tissues used as drug-residue sentinel sites do not adequately reflect the risk of carryover into the food chain. Stringent drug approval requirements and processes that are questionably rooted in scientific data have been blamed for contributing to the shortage of available animal drugs and slow decision making that forces manufacturers to avoid research and development of new antibiotics and applying for approval for use in food animals (AHI 1992, Feedstuffs 1996). Substantial redundancy appears to exist in the CVM regulatory process. Animal-drug approval decisions that address potential human health concerns are handled conservatively, and progress on approving new drugs is hindered. The clash in views between manufacturers and regulators is affected by the lack of rigorous data upon which to base decisions and questions regarding exactly whose responsibility it is to provide data beyond a reasonable set of criteria or concerns. For example, the argument can be made that concern for the consequences of antibiotic use in animals on human health is valid even though few data address that concern. In reviewing the issues presented in commissioned papers and invited workshop presentations, the committee identified 6 points for which control over animal drug approval might be too stringent:
FDA technical and regulatory requirements for manufacturing animal drugs are nearly identical to those for human drugs.
Extensive, rigid, and statistically bound efficacy requirements for animal drug applications are equivalent to those for human drug applications; statistically significant dose titration studies for each claim of effectiveness and pharmacokinetics support studies for each species are required.
Extensive target-animal safety studies are required for each species and claim. All are done under the rigorous GLP regulations. Many of the terms used in the GLP regulations are vague and poorly defined. Acute laboratory studies must be done for the Poison Control Center. It is often necessary to clarify the relationship between regulations and guidelines. In this instance, when a law or act is written, regulations are written to clarify and interpret the law. Guidelines (as might be presented by CVM) offer ways to meet and satisfy the terms and criteria in the regulations. This does not mean that the only way to meet the regulations is by way of the guidelines, but they do offer one or more ways to fulfill obligations.
For food animals, a rigorous, rigid, repetitive, expensive, and time-consuming food-safety research program is required for each species. This requirement alone takes 3 to 6 years and between $5 million and $8 million to complete. The food safety research and development program represents 50 to 70 percent of the research expenditures for a new chemical entity; those funds are committed early in the development procedure. The food safety requirements include extensive toxicological testing to calculate acceptable daily intake of residues of the drug. Then, elaborate metabolism and withdrawal studies are required to determine withdrawal times.
The environmental concerns for drugs developed for use in companion animals are equivalent to those for drug use in humans; a major concern is in manufacturing. Food-animal use requires an extensive and expensive research package for evaluation of environmental fate. That area continues to grow. A minimum projected cost of $300,000 and 18 months of testing environmental impact are required for the simplest nontoxic compounds.
A factor of concern in the United States and Europe is the emergence of socioeconomic and political pressure to shape the approval process. These sources of input can be driven by political and special interests and beliefs, generally with little consideration of scientific data.
The implication of the points is that, even with re-engineering, in many cases, there could be significant opportunities to further shorten the approval process. Those opportunities may need to be reviewed.
In 1968, when the congressional animal drug amendments to the FDCA created the Bureau of Veterinary Medicine (changed to CVM in the early 1980s), Congress prescribed an animal drug regulatory system no less strict than that for human drugs. During the ensuing years, new animal drugs were approved on the basis of increasingly stringent safety and efficacy regulations. One important requirement for the safe and efficacious use of an animal drug relates to how and under what circumstances a drug can be used. This information must be contained on package instructions (inserts or directly on the label) and is called labeling. A drug can be used only for the specific purposes stated on the label; any departure from that use is called extra-label use. However, enforcement was minimal, and approved uses and dosages were widely ignored, because they were considered ineffective or inapplicable for animal health needs.
In its policy statement on extra-label drug use, CVM (1984) recognized that strict enforcement of product labeling would be detrimental to livestock producers, veterinarians, the pharmaceutical industry, the consuming public, and the animals themselves. Strict enforcement would result in unnecessary animal pain and suffering, increased animal losses, decreased use of animal drugs, and higher prices for animal products. To practice a high standard of medicine, veterinarians must resort to extra-label drug use when, in their professional judgment, the clinical situation demands it for the well-being of animals.
Extra-label use may be classified in three broad categories: (1) drugs approved for human use that are used in animals, (2) approved animal drugs used in nonapproved species, and (3) approved animal drugs used in the approved species but for a nonlabeled purpose or dosage.
The provisions of the CVM (1984) extra-label drug use policy were as follows:
A careful diagnosis is made by a veterinarian who has a professional knowledge of the animal's health.
No other drugs are specifically labeled to treat the condition diagnosed, or the dosages recommended on the labels of available drugs are ineffective.
Thorough treatment records are kept on the animals, and the treated animals are identified.A withdrawal period between drug treatment and marketing of the animals is carefully observed and extended if necessary to ensure that the meat, milk, or eggs are free from illegal residues.
The extra-label drug use policy provided the means for CVM to address public health concerns by taking enforcement action against those who placed animals or public health at risk. This policy made the attending veterinarians responsible for drugs prescribed for or administered to food animals. Professional decisions on withdrawal times from all extra-label drug use were made by attending veterinarians, and the government continued to monitor animal products for violative residues.
The authority granted by CVM for extra-label drug use did not include lay persons. Extra-label use did not extend to medicated feeds and, because of human food safety concerns, some drugs were not allowed to be used in food animals under any circumstances. These include diethylstilbestrol (DES), chloramphenicol, ipronidazole, dimetridazole, and the nitrofurans.
On October 7, 1994, legislation legalizing discretionary extra-label drug use by veterinarians within the framework of a valid veterinarian–client–patient relationship (VCPR) was passed by the 103rd Congress. On October 22, 1994, President Clinton signed the Animal Medicinal Drug Use Clarification Act of 1994 (AMDUCA) into law. FDA published final regulations in the Federal Register (21 CFR Part 530) November 7, 1996 (Federal Register 1996). Some of ADMUCA's key provisions are as follows:
Extra-label use of FDA-approved animal drugs or human drugs is permitted under the following conditions:
—by the lawful written or oral order of a licensed veterinarian —within the context of a VCPR, and —in compliance with regulations promulgated by the secretary of Health and Human Services. Extra-label use of animal drugs in or on animal feed is not permitted.Extra-label use of an animal drug is not permitted if another animal drug contains the same active ingredient, can be administered in the same form and concentration, and lists the intended use in its specifications.
The secretary of Health and Human Services may prohibit particular uses of an animal drug.Use of an animal drug that results in residues, exceeding established safe concentrations is considered an unsafe use.
The secretary may provide access to the records of veterinarians to ascertain any use or intended use that the secretary has determined might present a risk to the public health.
The secretary may, after allowing opportunity for public comment, prohibit an extra-label use of an animal drug if it presents a risk to the public health or if an analytical method has not been developed for testing residue concentrations.
If the secretary finds that an extra-label use of an animal drug might present a risk to the public health, the secretary may establish, either by regulation or order, a safe concentration for residues of that animal drug and require development of a practical analytical method to detect unsafe concentrations of residues.
Even though extra-label drug use has been legalized, it is less than ideal, because a veterinarian's involvement might be peripheral or intermittent and, from an enforcement perspective, difficult to assess. An important question is how to establish “an appropriate withdrawal time” for a compound in a species for which hard data are lacking. One asset to AMDUCA is the Food Animal Residual Avoidance Databank (FARAD), a nationally sponsored project of USDA's Cooperative State Research, Education and Extension Service funded through the Food Safety and Quality national initiative. FARAD developers largely are associated with CVM, and are located in Florida, North Carolina, and California. Each location has separate but overlapping responsibilities. For example, the University of California at Davis is responsible for helping veterinarians establish safe and appropriate withdrawal times for extra-label drug formulations. The University of Florida serves as a repository for information on FDA-approved animal and veterinary products in a database that is readily accessible through the Internet as well as by phone and fax.
Under AMDUCA, a veterinarian who works in a valid VCPR (which involves total accountability for the use of the veterinarian-recommended extra-label product) can access the database or be referred to a professional pharmacologist to obtain information on the use, dose, and suggested withdrawal interval for an extra-label product. A distinction is made between a withdrawal interval and a withdrawal time. Only FDA can establish a withdrawal time, as the term is used in its legal application. FARAD establishes a conservative withdrawal interval or withdrawal time period, to provide a large margin of safety in eliminating residues from treated animals. In addition, FARAD makes no claims as to the efficacy of the intended treatment and responds only to inquiries from licensed veterinarians with suggestions for withdrawal intervals and periods. The use, dose, and withdrawal specifications listed in FARAD are obtained from several sources, including international compendiums of the same or similar products and formulations that are already approved abroad, and from published literature citations and data extrapolations based on referenceable data that can be used for the specific applications (Craigmill, A. 1998. FARAD, personal communication).
It would be desirable to improve the process of reviewing and approving animal drugs at CVM to enhance the availability of safe and efficacious drugs for use in food animals and companion animals under CVM-established conditions of use. Drug manufacturers and FDA should set boundaries for safe and effective use of drugs in animals as they do in humans. Approvals are needed for the dosage ranges (rather than for specific dosages) found to be minimally effective and maximally nontoxic, and withdrawal times for maximum dosages should be established. Local evaluations and decisions by veterinarians with appropriate training in drug prescription are needed to optimize effectiveness and ensure the safety of animals and consumers. However, if the issue of animal drug availability is not addressed aggressively, the legal extra-label drug use actually could be a deterrent to the animal drug industry in its attempts to discover and develop new pharmaceutical products.
Given current drug efficacy requirements, relatively few food-animal species and a small number of diseases or production improvement uses are seen by the animal health industry to warrant the risk and capital investment now required for successful drug development. If those drugs or other new drugs are to be made available to a broad range of species, efficacy requirements must be interpreted to maximize rather than limit the potential of a drug for approval.
Clinical trials need not be the sole or predominant measure of a product's effectiveness. The realities of livestock production and drug use are that, in the case of food animals raised in flocks or herds, any drug product will be tested quickly in the marketplace, and those found uneconomical or ineffective will be eliminated. To remedy many of these problems and facilitate the approval of drugs, CVM undertook a major investigation of how it conducted its affairs. The second piece of legislation awaiting final rule is the Animal Drug Availability Act (ADAA), signed by President Clinton in October 1996, which introduced many new ways for the regulatory process to approve animal drugs and medicated feeds more rapidly. ADAA's overall intent is to lessen the burden placed on the animal health industry to follow the approval process while maintaining the protection of the public, summarized as follows:
ADAA eliminates the strict requirement for field studies (except as requested and justified by CVM) by redefining and broadening the interpretation of what constitutes “substantial evidence” of efficacy. However, CVM still maintains strict requirements for proof of efficacy; where valid alternatives do not exist, the field study is the recommended and required evidence.
ADAA provides for greater interaction between the sponsor and CVM; studies can be identified that must be conducted to provide needed data.
ADAA creates a new category of animal drugs: veterinary feed directive drugs.ADAA supports flexible labeling to permit a range of recommended and acceptable dosages within a given species.
ADAA defines adequate and well controlled more explicitly with regard to the conduct of field trials.
Collectively, the act reflects a significant improvement in partnership and in interactive and constructive discourse between CVM and other interested parties.
Risk depends not only on the nature and severity of a hazard but also on the probability of its occurrence. The probability of an adverse health effect occurring depends on the exposure of consumers to a compound. Thus, exposure assessment has as much of an influence on overall risk characterization as does toxicity assessment. In the extreme, there is no risk from even the most hazardous compound if no one is exposed to it. A more relevant risk assessment approach could be designed by identifying the residue of toxicological concern for a particular compound. The portion of the residue that is still potentially bioactive and bioavailable to the consumer could be identified. An assessment could be performed to determine the exposure of consumers to residues in their diets. The information would then be used to determine the most relevant approach for determining the risk posed by a compound (Mulligan 1995).
This type of assessment would require more expertise and evaluation by CVM and drug sponsors. It would require use of all available scientific information and extensive communication between CVM and sponsors rather than reliance on guidelines that spell out requirements. However, this approach would enable manufacturers to spend time and resources on those tests that would be most beneficial in identifying relevant toxic end points for the consumer and still protect public health (Mulligan 1995). The costs and delays associated with preparing inordinately comprehensive environmental-assessment reports have hampered research and development of new products. A modified, simplified environmental assessment, perhaps based on standard and uniform tests, could be implemented to circumvent an aberrant oversight. The ecological interactions that shape the evolution of pathogens (Pfisteria, for example) are only beginning to be understood and could substantively affect decision making in the future for farm animals, drugs, and the environment. By adding to the already escalating costs of drug development and by increasing the quantity of data needed for approval, this area of the approval process might have contributed to industry reluctance for developing new drugs for animal use, and it might have produced a shortage of new drugs to treat diseases with new resistance patterns. The effect of the costs and delays has been seen especially in the development of drugs for minor species and minor-use claims.
To better explain the position of animal producers, animal health professionals, and the animal health industry and how availability of drug choices affects their work, it is useful to view the problem in terms of the size of the related industries. As shown in Table 4–1, the sheer number of food animals raised in the United States is staggering. There are approximately 10 million dairy cows and more than 7 billion poultry. Actually, the efficiency of production (remembering that a substantial portion of increased efficiency relates to animal health and animal drug use) serves to hold the numbers of production animals down.
Food-Animal Populations in the United States.
Table 4–2 shows that, as of 1996, the total comparative values of the animal drug industry, encompassing prescription drugs, over-the-counter preparations, and feed efficiency drugs, amounts to $3.2 billion. That is approximately 6 percent of the value of human prescription drugs.
Comparative Value of FDA-Regulated Industries.
Finally, as seen in Table 4–3, 87 percent of all animal drugs have annual individual product sales of less than $1 million, and only 5 percent of the available compounds generate sales of more than $5 million.
Annual Sales of Animal Drugs.
Animal pharmaceutical companies and animal health divisions of large parent corporations are expected to be financially independent and profitable. The number of corporations with animal drug development programs is declining through mergers, sales, or downsizing in relation to profitability and competitiveness. The number of CVM animal drug approvals had declined up to 1991. Some questions have arisen regarding the rate at which new approvals were authorized. In particular, few new drugs were approved for use in production of veal calves, sheep, goats, and fish. Reassessment of human risk by FDA actually resulted in the removal of some animal drugs from use in production, including the nitroimidazoles, nitrofurans, and DES, which have been identified as carcinogens and which, by the Delaney Clause 2 of the FDCA, were prohibited as “additives” to the food supply.
For the animal production and health industries, the issues regarding antibiotic development, approval, and use are in some ways more complex than are those for human health. The human health industry focuses on approval processes for a single species, even though within that species, drug use applications are further subdivided and classified as to route of administration (because the safety profile differs by local and systemic toxicity as do the pharmacokinetics) and disease (because some are more severe and more risk might need to be tolerated). These drug criteria also are stratified by age, health status, sex of the patient, and so forth. Animal drugs traditionally were approved for each species and each application within a given species, and manufacturers were required to validate the claims of efficacy and safety for each use. For example, drugs for bovine use need separate government approval for applications in milk production, meat production, reproduction, and juvenile uses; poultry drugs are approved separately for laying hens, broilers, and turkeys. The reason for the separate approvals for each use is similar to that for humans: Relative local and systemic toxicities vary with the pharmacokinetics and these are affected by age, health, disease virulence, and sex of the animal. But also separate approvals are needed because, for residue regulatory actions, husbandry practices differ for the uses of drugs and thus the potential to affect human health varies. The redundancy in expected paperwork to substantiate an application submission for government approval and the response time on the part of the authorizing federal agencies were considered by the animal industry to be major impediments to the process and progress of drug development. Historically, the authorizing federal agencies have held firm that the health of the human and animal populations was of paramount importance in the approval process and that the integrity of the process would not be violated. Ultimately, the preservation of human health was the standard by which all drug-related decisions were made.
Data packages are increasingly comprehensive, and multinational development of products is becoming more common. Harmonization of U.S. and foreign approval standards should be a major goal supported by CVM and the animal drug industry. In reviewing an application for a drug that has already been approved elsewhere, regulatory officials should take advantage of the valuable resources of other countries whenever possible. For example, approval of a product in the United States that has already been registered by the European Economic Community (EEC) should be a straightforward, speedy process. If a product is approved outside of the United States in a country with a comparable approval process, the process in the United States could be expedited.
The areas of human food safety, target-animal efficacy and safety, and environmental fate and worker safety are the major areas of data required in all countries for approval of a veterinary compound. Harmonized review requirements could be envisioned for these data packages. Data would include the following:
Human food safety. This area includes toxicology, metabolism, biological effects, residue profiles, and consumption calculations.
Target animal safety and efficacy. This area includes basic principles and studies on the use of the drug, such as its mechanisms, toxicities, interactions, and limits. These data would be needed to show that the product works and to define the limits. Exact dose forms, local clinical trials, and support use studies would be done for each country.
Environmental safety. This area includes all the basic transformation studies, fate studies, environmental-toxicity studies, and worker safety studies.
Harmonization of testing procedures and standards for the approval of human drugs as part of the International Conference on Harmonization is continuing among the United States, EEC, and Japan. Many of the guidelines generated for human drugs also could be applied to animal drugs. Every effort should be made to harmonize such requirements as toxicity testing of human drugs, animal drugs, and pesticides.
Under the United States and Canada Free Trade Agreement, scientists from CVM and the Canadian Bureau of Veterinary Drugs (BVD) have harmonized human food safety requirements for approval of drugs used in food animals. The United States and Canada are to use a 6-step procedure for human food safety evaluation of new animal drugs. Using these harmonized standards, both countries have agreed on identical tolerances for 37 animal drugs (Brynes and Yong 1993).
Two important lessons can be learned from the efforts of CVM and the Canadian BVD under the free trade agreement. The first is the importance of involving working scientists in the harmonization efforts. Having scientists who used testing protocols and standards as part of their everyday review work helped to ensure that harmonization was achieved. The scientists were familiar with the background and rationale for each requirement and could easily determine which areas were open to compromise and which were not.
The second and perhaps more important lesson is that harmonized standards do not automatically result in harmonized acceptable residue tolerance levels. Setting standard concentrations is the ultimate goal of most current harmonization efforts. For example, differences between the U.S. and Canadian tolerance levels resulted from different but equally valid conclusions made by different scientists about the same data. Harmonized standards and requirements would not have changed the outcome to any significant degree.
Although harmonizing testing protocols and standards is a worthwhile goal, until countries conduct joint reviews that lead to a single tolerance level, more emphasis must be placed on delineating guidelines for determining the equivalence of different tolerance levels for the same compound. One mechanism is to use dietary exposure estimates to determine the equivalence of tolerance levels, which normally are calculated from the acceptable daily intake (ADI) determined for a compound for human consumption. ADI is the amount of residue of a compound that can be ingested daily over a consumer's lifetime without appreciable health risk. Therefore, ADI can be considered the safety standard for a compound. If use of one country's tolerance level does not result in residues above another's ADI, then the tolerance level should be considered equivalent for purposes of consumer safety, trade, and, perhaps, regulatory decisions (Fitzpatrick et al. 1995). Again, agreement on the definition of terms affects how harmonization processes can proceed. The United States and the EEC have similar requirements for animal drug approval in all of the major data areas, and decisions regarding ADI revolve around a “no observable effect” definition. Japan differs in its approach to evaluating the human food safety aspects of an animal drug. Rather than calculate an ADI for a compound, Japan bases its regulations on a “no-residue” standard. On the basis of the most sensitive analytical method available, no residues of a compound can be found in edible animal products. Essentially, one definition constitutes a form of bioassay where the other is pure analytical chemistry.
The drug review process also could be harmonized by using expert panels to report their recommendations to regulatory groups for action. The panels should be international, so that their findings would be accepted by all regulatory groups as definitive summaries of scientific evidence. The panels would be charged with reviewing documentation by using the harmonized approach described above. The panels would evaluate the data and compile expert summaries. If the panels found the databases satisfactory, they could then recommend approval of drugs. If the databases were inadequate, the panels would report deficiencies and recommend studies needed for complete evaluation.
Two examples of the panel approach currently exist in the animal drug area. The panels are involved only in the review of human food safety data for an animal drug. The first is the EEC Committee on Veterinary Medicinal Products (CVMP). The initial step in the approval of an animal drug in one EEC country is the calculation of an EEC tolerance level for the compound by the CVMP. The CVMP consists of animal drug regulatory officials from the various EEC nations. The CVMP reviews all available data from toxicology studies and residue metabolism and depletion studies, and establishes a tolerance level for residues of that compound in edible animal products. That tolerance level is then adopted by all EEC countries that approve the compound for use.
A second example of an international panel is the WHO and Food and Agricultural Organization Joint Expert Committee on Food Additives (JECFA), which evaluates human food safety data on selected animal drugs for the Codex Committee on Residues of Veterinary Drugs in Foods (CC/RVDF). JECFA is an ad hoc committee of animal drug experts from the codex committee countries. The committee evaluates toxicological and residue data on priority animal drugs submitted by the CC/RVDF. These animal drugs are already approved in at least one codex member country. JECFA establishes tolerance levels for the animal drugs; those levels are then sent through CC/RVDF for acceptance by other member countries. The United States currently does not accept a codex tolerance level for an animal drug if it differs from the level established here.
The issue of harmonization, however, can not be considered purely in black and white. Sometimes, harmonization of drug regulations is not feasible or practical. Although it is beyond the scope of this report to go into detail, it must be mentioned that some countries have regulations significantly different from those implemented in the United States. In part, the decision could be shaped by socioeconomic factors that make the greater risk in the use of a drug to resolve a rampant health problem more acceptable then the potential risk of the drug's side effects.
There are several points in the process of drug development and federal approval at which the added cost and time to acquire data and review them impede drug approval. Although historically the development of antibiotics was more-or-less a slow process of trial and error, the evolution of newer biochemical and molecular biological techniques has changed that situation. Newer methods provide tools that allow scientists to predict quickly how chemical modifications of basic parent antibiotic compounds can keep pace with the natural microbial changes that help populations of bacteria to develop resistance. The economics of drug development, however, make antibiotic discovery a matter of industrial priority setting. In the process of discovery and development, antibiotics are more readily prepared for human clinical use than for animal use. The high cost of new drugs makes them impractical for widespread use in agriculture, especially when a potential use is for disease prevention at subtherapeutic concentrations. In addition, in the past, approval of animal drugs in the regulatory process had the added burden of needing to show human food safety as well, thus adding costly and time-consuming projection studies to the food-animal drug development process. Recent developments at CVM, in part facilitated by the 1995 reforms to streamline the federal government, have shifted the regulatory process for animal drug approvals to a more interactive, quicker process. Examples of laws that have added and promise to add increased efficiency to the food-animal drug approval process are AMDUCA and ADAA. Additional considerations limit the availability of some newer antibiotics for animal use, as exemplified by the prohibition on the extra-label use of fluoroquinolone antibiotics for subtherapeutic uses in food animals.
Veterinarians and animal producers are concerned that the recent increase in emergence of antibiotic-resistant strains of pathogens jeopardizes the future use of the sparse number of available antibiotics. The use of large amounts of antibiotics in food animals has been justified by the suggested benefits to human and animal health (that is, drug use ensures the healthfulness of animal-derived foods). That view might need to be reassessed. A growing concern is that the occurrence of disease and drug-resistant microorganisms in food animals as well as development of multidrug resistance in human pathogens poses a threat to human health. The development of a sound database needs to continue and expand rapidly to assess the relationship between the use of antibiotics in the United States in food-animal production and the impact on human health. The relationship will need to be reassessed continuously, and new procedures will need to evolve, just as microorganisms evolve.
The committee concludes that the pursuit of increased drug development and approval efficiency should be continued in a formalized reiterative process that integrates human and animal health needs with continuously updated data on patterns of antibiotic resistance, efficacy, and usefulness. With sound judgments based on data and emergence projection models, the availability of drugs for human and animal applications can be better coordinated. Decisions to approve or restrict the use of antibiotics must be based on rational and valid data.
The committee recommends that CVM continue procedural reform to expedite the drug approval review process and to broaden its perspective on efficacy and risk assessment to encompass data review on products already approved and used elsewhere in the world. Particular emphasis should be placed on adverse reactions, residue carryover into food, and antibiotic-resistance-emergence patterns. Efforts need to continue to further streamline the iterative INAD period, and, based on the time required for target animal safety trials and efficacy studies, the following are reasonable areas to streamline more:
• An arbitration procedure should be developed to expedite the regulatory approval process.
A formal written procedure needs to be established for resolving scientific and regulatory issues between the sponsor and CVM in a timely manner. CVM and the industry could compile a list of experts in the various disciplines willing to serve as consultants to CVM. In the event of an impasse, the consultants would be asked to provide a written opinion on the dispute within a certain period.
• CVM should eliminate the guideline that all studies be conducted in multiple locations.
The number and location of the studies should be determined for the specific drug, claim, and species. CVM could save additional resources by placing more emphasis on data from other countries that have previously demonstrated an ability to provide reliable data. The original policy that suggested the need for three locations of study was a CVM guideline and sometimes has been misinterpreted as a requirement.
• CVM should review the requirement that all studies provide the same quantity of evidence to establish efficacy for supplemental applications as for original applications.
The quantity of evidence required to establish efficacy should depend on scientific data supporting the relationship between the existing claim and the proposed one. If a supplemental claim is closely related to the one approved, fewer additional studies should be required.
• More flexibility in CVM's evaluation of manufacturing requirements is needed (Stribling 1992).
The issue is not whether an animal drug should meet the same standards of safety, effectiveness, potency, quality, and purity as a human drug. The issue is the quantity of the data required to demonstrate that the animal drug meets the standards. In every instance, the amount of data required should be assessed individually with a scientific proposal submitted by the sponsor and the obligation residing with the sponsor to substantiate the case.
• More realistic estimates of human dietary exposure should be made when residue tolerance levels are developed.
The requirements for human food safety testing for drug-related residues in meat, milk, and eggs are complex, and demanding toxicological tests of the parent drug and any potentially toxic metabolites, residue identification and quantitation, and method development for quantification of residues in edible animal products are required. In evaluating residues and contaminants, CVM assumes that all residues present in food have the same toxicity as the parent drug based on enterohepatic recirculation and hydrolysis of metabolites to the parent drug. It also assumes that residue is present in a food commodity at its highest permitted daily concentration over the lifetime of the consumer, based on a worst-case scenario. Those assumptions do not account for what the consumer is actually exposed to in the daily diet. Often, the residue remaining in food is no longer bioactive or bioavailable to the consumer. The food commodities in which residue is present might not be part of the daily diet of the consumer, or the residue might not be present in the edible portion of the commodity (Farber 1995).
• To improve drug availability, worldwide harmonization of requirements for drug development and review should be considered and further enhanced within the federal agencies responsible for ensuring the safety of the food supply.
Data and criteria for review should be standardized among countries, with final approval remaining with each country. Such harmonization could lead to direct savings in costs of drug development and even greater savings in time and return on investment for the sponsor and the animal producer. The harmonization process needs to be coordinated with drug-resistance-emergence surveys, so that the trends and patterns of antibiotic resistance in other regions of the world—developing countries in particular—can be modeled and structured into the drug development and approval process. Ultimately, increased use of international harmonization agreements will allow FDA to make more efficient use of its resources. Initiating the process of harmonization reform could prove slow and cumbersome, but diligence in this effort should produce a more efficient and responsive collective review and monitoring process. Desirable advances in the regulatory process would be to establish drug use guidelines based on maximum safe regimens for the target food animals, to set drug withdrawal times accordingly, and to develop tests for use on farms to certify the absence of violative residues of toxicologically active drugs or their metabolites.
For additional information on the INAD and NADA process, see CFR Title 21, Parts 511 and 514; CVM Staff Manual Guides 1240.300, 1240,3030 and 1240.31000; and FDCA, Section 512.
The Delaney Clause, which was included in the 1958 Food Additives Amendment to the FDCA, directs that “no additive shall be deemed to be safe if it is found to induce cancer when ingested by man or animal, or if it is found, after tests which are appropriate for the evaluation of the safety of food additives, to induce cancer in man or animal.”
