Establishing Efficacy of Human Products Using Animals

Abstract

In 2002, the US Food and Drug Administration issued regulations to allow the approval of human drugs and biological products based on animal efficacy studies when human efficacy studies would be unethical or not feasible. These regulations are intended to assist in the approval process for products aimed at preventing or treating human diseases caused by nuclear, radiological, biological, and chemical agents that have the potential to harm a significant percentage of the US population. This article discusses the criteria that must be met to use the Animal Rule to demonstrate efficacy in place of human clinical trials.

The US Food and Drug Administration’s (FDA) Center for Biologics Evaluation and Research and Center for Drug Evaluation and Research are responsible for ensuring the safety and efficacy of vaccines and drugs, respectively, prior to releasing those products for public use. Historically, the safety and efficacy of a drug or biologic have been demonstrated through nonclinical testing followed by carefully designed and closely monitored human clinical trials. This method has proven successful in demonstrating that a drug or biologic can treat or prevent a condition or disease without causing undue serious adverse effects to the patient.

In the last 2 decades, the emerging threat of domestic terrorism, and concern regarding the agents that could accompany that threat, have led to increased interest in developing new biologics and drugs to prevent and treat the serious clinical events caused by those agents. These threats include nuclear, radiological, biological, and chemical agents with the potential to harm large numbers of individuals, either on the battlefield or among the US population. Specific biological agents, such as those that cause anthrax, smallpox, botulism, plague, and tularemia, have received much attention regarding their potential use as agents of bioterrorism. ⁶ Countermeasures against these diseases, including novel drug therapies and preventive vaccines, would constitute a crucial element of the nation’s emergency preparedness. However, because exposure to any one of these agents is rare and carries dire consequences for individuals, it would be neither ethical nor feasible to perform human clinical trials to establish efficacy for proposed prophylactic or therapeutic products. Nonetheless, a scientifically sound method is needed for demonstrating that a proposed biologic or drug product will provide the intended protection against these threats, before such a product is released for use by the US government or the general public.

Origins of the Rule

To address the need to demonstrate efficacy, the FDA, in response to the Department of Defense’s desire to protect US military personnel from biological attacks and in recognition of the need to protect the general public from such threats, first issued a proposed rule in 1999. This new rule proposed to permit the demonstration of efficacy using animal studies in lieu of human clinical trials, in those instances where human trials would be unethical because of the serious clinical effects of the agent or where such trials would be unfeasible because of the infrequency of human exposure to the agent. After a period for public comment and further modifications by the FDA, the rule, titled “Approval of Biological Products (or New Drugs) When Human Efficacy Trials Are Not Ethical or Feasible,” was published in final form on May 31, 2002 (http://www.gpoaccess.gov/cfr/index.html). Although there are separate Code of Federal Regulations (CFR) citations for drug products ¹³ and biological products, ¹² their language is nearly identical. The combined CFR citations are commonly referred to as the Animal Rule or the Animal Efficacy Rule.

The term Animal Rule, both inside and outside the FDA, has become shorthand to refer to this substitute for human clinical trials presumably because of the extraordinary emphasis on the use of animals in place of human subjects to establish the effectiveness of products against deadly agents. Up to this point, 2 products have been approved using the Animal Rule. One is for a countermeasure to exposure to the cholinesterase inhibitor, Soman, and the other is a treatment for cyanide toxicity. Other potential applications of the Animal Rule are for vaccinations or postexposure treatments against anthrax spores, plague bacteria, tularemia bacteria, smallpox virus, botulism toxin, and viral encephalitis viruses. Postexposure therapies for acute radiation exposure could also apply the Animal Rule.

Comparison of Animal Efficacy Trials to Clinical Trials

In essence, the Animal Rule states that under specific circumstances, the FDA may grant marketing approval to new drugs or biological products following efficacy trials in adequate and well-controlled animal studies. These studies must demonstrate that the product can reasonably be expected to provide similar protection for humans as it does for animals upon exposure to the potentially deadly agent. Under the Animal Rule, animal studies must answer the same efficacy questions as human clinical trials while providing even more detail on disease pathogenesis and on the mechanism by which the product prevents or treats the disease. This additional information is intended to strengthen the correlation of the animal study data to the human experience. For instance, the pathogenesis of Yersinia pestis–induced plague in humans is well known, and a well-characterized animal model has been established for this disease. Current literature indicates that similarities exist between human and murine plague lesions ⁴ and that some critical elements of murine immunity are the same ones that are important in human immunity. ¹ This knowledge could be the starting point to establish that the murine model could fulfill the requirements of the Animal Rule for plague.

Whereas much of the preclinical toxicity and efficacy studies for Animal Rule products are the same as for clinical trial products, there is one key difference. In conventional drug efficacy studies, the therapy only needs to be shown to be effective. When investigators apply the Animal Rule, however, the animal models must also establish how the drug works. This need for additional data does not apply to human clinical trials, so it could be argued that the Animal Rule sets a higher standard for proof of efficacy than currently exists in conventional clinical trials. Therefore, the Animal Rule should not be considered a more expedient route to product approval.

Use of the Animal Rule

There are several caveats to consider when using the Animal Rule to request marketing approval for a product. First, the sponsor must establish that the disease in question meets the criteria of “serious or life-threatening conditions caused by exposure to lethal or permanently disabling substances.” The FDA defines a “lethal substance” as one that is likely to kill at least some humans, and a “permanently disabling substance” is defined as one that will cause a permanent physical or mental impairment that substantially limits one or more of the major life activities in at least some humans. Institutional review boards, which are required to review all proposed human clinical trials, could provide valuable input to this assessment by confirming that traditional clinical trials would be unethical in a given instance.

Second, conventional safety evaluation of the product must still be conducted in human clinical trials. In other words, if a potential tularemia vaccine is shown to be efficacious in animals, the vaccine would still need to be administered to humans in conventional clinical trials to demonstrate that the proposed dosage is safe. The FDA recognizes that some safety data might not always be available, such as data on possible adverse interactions between the challenge agent and the proposed product. This situation should not prevent the agency from making an adequate safety evaluation. Such data could be obtained as part of the postmarket safety and activity evaluation, which is required for all products approved under the Animal Rule.

Third, the Animal Rule does not preclude the need for preclinical animal toxicity studies, which include standard in-life data collection and postmortem tissue analysis. These preclinical animal toxicity studies provide the basis for dose selection and provide early data for potential toxicity of the products to enable clinical safety trials using escalating doses when appropriate.

Finally, the Animal Rule does not apply to products that can be approved based on standards described elsewhere within the FDA’s regulations. For example, if a product can be approved under the accelerated approval process—in which approval is based on human clinical trials using surrogate markers or clinical end points rather than survival or irreversible morbidity—then that product will not qualify for approval under the Animal Rule.

Requirements of the Animal Rule

To be considered for marketing approval under the Animal Rule, a product must meet the following 4 requirements:

A detailed understanding exists regarding the pathophysiological mechanism for toxicity of the agent in humans and also regarding the mechanism by which the proposed product reduces or prevents adverse effects in humans.

The prophylactic or therapeutic effect of the product is demonstrated in more than 1 animal species.

The animal study end points correlate to increased survivability or prevention of significant morbidity in humans in the face of lethal challenge.

Pharmacokinetic and pharmacodynamic data for the proposed product are available and can be used to select an effective dose in humans.

To further clarify the critical aspects of the animal models used under the Animal Rule, the FDA published a Draft Guidance for Industry in January 2009. ¹⁰ This draft guidance, titled “Animal Models—Essential Elements to Address Efficacy Under the Animal Rule,” focuses on the critical animal study elements that address the Animal Rule’s 4 criteria. Although this draft guidance outlines the FDA’s current approach to the development of animal models to study efficacy, it is not binding on sponsors. Alternative approaches to answering key questions of challenge agent pathogenesis, and the drug or biologic’s mechanism of action, can be considered provided that a strong scientific justification exists.

It is the intent of this article to provide conceptual and practical guidance in applying the Animal Rule in lieu of human efficacy studies. The following discussion expands on the 4 criteria to be met when requesting product approval under the Animal Rule.

Mechanism of Toxicity and Protection

The pathophysiological mechanism of toxicity for the specific chemical, biological, nuclear, or radiological agent should be understood in detail. This mechanism may be established using previously published clinical data or using data derived from animal models. For example, it has been shown that once anthrax spores enter the body, they are taken up by macrophages. The spores then germinate into vegetative bacteria with an antiphagocytic capsule that provides protection from the host’s immune response. The bacteria’s virulence results from 3 toxin proteins: protective antigen (PA), edema factor (EF), and lethal factor (LF). To cause disease, PA must bind to cellular receptors of the host and then with either EF or LF. ³ This type of knowledge can demonstrate that humans and animals share the same mechanisms of disease and could assist in fulfilling this requirement of the Animal Rule.

There should be similarities between the human clinical data and the animal model regarding pathogenic determinants resulting from the challenge agent. For example, in human inhalational anthrax infection, lesions are reported to predominantly involve the mediastinum, the lymphatic system, and the respiratory system. ⁷ The rhesus macaque has been shown to share the mediastinal, lymphatic, and pulmonary lesions with humans, ² providing strong justification that this species may be an acceptable animal model to meet the Animal Rule requirements. Rabbits exhibit the mediastinal and lymphatic lesions seen in humans, ¹⁴ and thus rabbits could serve as a second surrogate species in lieu of human clinical subjects.

In most animal models, the challenge agent should be identical to the etiologic agent that causes human disease; however, this may not always be possible. Smallpox, a highly lethal orthopoxvirus, has been identified as a potential bioterrorism agent, and renewed interest exists in developing new vaccines and antiviral drugs to combat it. ⁵ However, it has proven difficult to identify an animal species in which the Variola virus, which causes smallpox in humans, replicates the fatal disease. In these instances, the FDA may allow the use of closely related orthopoxviruses to demonstrate efficacy of a proposed product in animals. Several related viruses cause a similar disease course in various animal species, such as monkeypox in nonhuman primates, ectromelia or vaccinia in mice, and rabbitpox in rabbits. ⁸ Any of these virus–host combinations may be satisfactory as a model for potential smallpox countermeasures in humans. The selection of such a model would require FDA approval prior to designing pivotal studies to meet the requirements of the Animal Rule.

Other details concerning the pathogenesis of the challenge agent and the animal model’s response to challenge should be as analogous to the human experience as possible. The animal model and the human experience should be as similar as possible in terms of the time course of the disease, the organ systems affected by the challenge dose, and the route of challenge agent exposure. A challenge dose in the animal model should be established that reliably reproduces the disease seen in humans. The timing of treatment intervention should be the same in the animal trial as its intended use in humans; that is, whether preexposure, postexposure, or in conjunction with other therapies, all conditions of projected human use of the product should be replicated as closely as possible in the animal model. If multiple doses are anticipated in humans, this should be reproduced in the animal study. Any deviation between the animal model and intended human clinical use should be scientifically justified and discussed with the FDA prior to the animal study.

Demonstration of Efficacy in More Than One Species

In most cases, efficacy data from at least 2 species are required. Prospective species may be chosen based on published literature. In the absence of published data, studies may be required to demonstrate that the proposed species' response to challenge is similar to the human response. It is unlikely that any single species will completely replicate the human response to challenge or the method of protection afforded by the product; therefore, a second species may be able to fill in the gaps in knowledge or confirm the findings of the first. In either case, demonstrating efficacy in the 2 species increases the confidence that the animal models are predictive of the human response to challenge and treatment.

The Animal Rule does allow the use of a single species to establish efficacy if that species represents a well-characterized animal model for predicting the human response. This may require that the single species responds nearly identically to the challenge agent, as established either in the literature or in preclinical studies, and that the product’s mechanism of prevention or treatment in the animal model is identical to that mechanism in humans. If the developer of a new product believes that a single species meets these criteria, the developer should present supporting data and enter into discussions with the FDA early in the product development process. The FDA may require that 2 separate efficacy studies be performed in the single species to demonstrate reproducibility and to increase the confidence in that model.

Finally, it should not be a foregone conclusion that nonhuman primates must be one of the species used for efficacy trials in fulfillment of Animal Rule criteria. If the scientific data demonstrate that other species are predictive of the human experience, then these species should be chosen as the model.

Animal Study End Points

These end points are the most relevant because they correlate most closely with the desired results in humans, that is, a reduction in instances of death or serious clinical disease resulting from the challenge agent. These end points are preferable to in-life parameters, such as liver enzyme panels in the case of hepatotoxic agents, because they provide greater confidence in translating animal efficacy data to projected efficacy in humans. Specific end points should be discussed with the FDA well in advance of beginning pivotal efficacy studies.

Because the Animal Rule allows animals to serve as surrogates for human subjects in challenges with deadly agents, these animals will likely be subjected to painful or lethal outcomes. Therefore, their welfare as research subjects must be addressed. In accordance with the requirements of the Animal Welfare Act, ⁹ the chosen end points must be strongly justified to institutional animal care and use committees as part of all proposed efficacy studies. Although the Animal Rule specifically states that the desired end points should be enhanced survival or prevention of serious morbidity, the rule does not prohibit the establishment of earlier surrogate end points if they can be shown to still be predictive of the challenge agent’s toxicity and do not interfere with evaluation of the product’s effectiveness. In survivability studies, efforts should be made to validate end points that indicate impending death. Validation of these early end point data may occur during preclinical studies to assist in addressing animal welfare concerns.

The challenge agent should be standardized for identity, strength, purity, and quality. The route of exposure by the challenge agent should mimic projected human exposure, and a dose of the challenge agent should be established that reliably reproduces the human disease. The delivery of the therapeutic or prophylactic product in the animal trials should also match the proposed human treatment. Group sizes in the pivotal efficacy study should be large enough to power adequate statistical analysis for confirming the desired effect. The potential for bias in making clinical evaluations should be minimized by blinding the treatment groups in the study. If it is anticipated that humans exposed to the challenge agent will receive supportive care, the same care should be administered in the animal model. The indications for this supportive care should be clearly defined in the study protocol. This will also require a placebo study group that receives the same supportive care without the countermeasure in order to demonstrate the increased benefit of the product over supportive care alone.

The pivotal efficacy studies must be conducted in accordance with the requirements of Good Laboratory Practices (GLP). ¹¹ This requirement also applies to any animal studies referred to in the product labeling, including package inserts. However, early pilot studies to determine proof of concept would not need to be conducted under provisions of GLP. In the pivotal efficacy trial, all assays used to monitor the animal model’s response to the challenge agent should be validated to demonstrate that they are predictive and reproducible.

Data to Support an Effective Dose in Humans

To determine the effective dose for a drug product, there should be sufficient pharmacokinetic (PK) and pharmacodynamic (PD) data to establish the dosing regimen for the disease in question. Dosage data from the animal model should elucidate a dosing schedule in humans that will give the equivalent protection. There is less emphasis on PK and PD data for vaccine products. Vaccine trials in the animal model should establish an adequate protective immune response that can reasonably be expected to be reproduced in humans. For drugs, conventional nonclinical safety testing would be conducted to identify potential adverse effects of the product prior to the clinical safety trials still required when using the Animal Rule. In certain cases, studies performed in animal models of disease may be used as an acceptable alternative to toxicity studies in normal animals. The scientific justification for use of these animal models of disease to support safety should be provided.

To summarize, the indication being sought for the proposed product should drive the study design in developing animal models that will adequately demonstrate product efficacy. Well-characterized animal models may not exist for many bioterrorism agents; therefore, the animal models may need to be established and validated as part of the research and development process. In addition, the optimal effective dose of the drug or vaccine in humans must be determined. A well-organized, detailed plan derived at the beginning of product development is critical to providing the necessary data to fulfill the Animal Rule requirements. This plan should be discussed with the FDA early in the process to ensure that scientific and regulatory objectives are met as product development proceeds. These discussions should be held with the appropriate product division within the relevant center for the intended countermeasure. The Animal Rule represents a significant departure from conventional FDA pathways for demonstrating efficacy of new products to prevent or treat human diseases. Its effective use by sponsors will require frequent communication with the FDA to determine the most efficient way forward in this newly opened pathway to product approval.