Under fire on several fronts in recent months, the beleaguered FDA is facing particular scrutiny regarding the safety of drugs approved for human use. Recent high-profile drug safety failures such as Trasylol, Pergolide, Zelnorm, Ketek, and Avandia add to a litany of dangerous drugs, many still prominent in the public’s memory, including Vioxx, FenPhen, Propulsid, Rezulin, Seldane, and Accutane.
While most analyses and proposals have focused on improving postmarketing drug surveillance, the inability of preclinical drug testing including animal toxicity testing to predict human drug responses and toxicities has also garnered attention. According to the FDA’s own data, only 8% of all drugs progressing to human trials after demonstration of safety in animal studies will gain approval.
Remarkably, this approval percentage is down from 14% in 1985 despite two decades of efforts to improve the accuracy of preclinical drug testing—efforts that notably have not addressed the inherent and seemingly immutable inaccuracy of animal toxicity testing. About half of clinical-trial drug failures are due to undetected toxicities, and about half of the small percentage of sanctioned drugs will later be withdrawn or receive black-box warnings for the same reason.
After all this attrition, even marketed drugs do not work for most patients. More than 90% of these therapeutics only work in a minority of patients, and efficacy rates are as low as 25% for oncology treatments and 30% for Alzheimer’s disease products. Something is clearly amiss with preclinical drug testing when only 4 in 100 drugs entering clinical trials are truly safe and effective and that effectiveness means that only a minority of patients will benefit.
The seriousness of the drug-safety problem and the role of unreliable animal toxicity testing, is underscored by several recent reports including well-publicized critiques from the Institute of Medicine (IOM), the Science Board of the FDA, and the National Research Council (NRC) of the National Academies.
Reports Suggest Answers
In its response to a September 2006 IOM report “The Future of Drug Safety: Promoting and Protecting the Health of the Public,” the FDA acknowledged the inadequacy of animal toxicity testing and outlined its approach to the problem: “The FDA is involved in an ongoing scientific collaboration intended to yield more sensitive, specific, and informative tests for drug organ toxicity than the toxicology-screening techniques currently in use.”
A July 2007 NRC report, “Toxicity Testing in the 21st Century: A Vision and a Strategy” was prepared at the request of the EPA. The report makes sweeping recommendations regarding the replacement of animal testing with human-based toxicity measures, and many of its recommendations are equally applicable for chemical and drug toxicity testing.
The report states that advances in toxicogenomics, bioinformatics, systems biology, epigenetics, and computational toxicology could transform toxicity testing from a system based on whole-animal testing to one founded primarily on in vitro methods that evaluate changes in biologic processes using cells, cell lines, or cellular components, preferably of human origin.
The November 2007 Science Board report, “FDA Science and Mission at Risk”, is critical of several aspects of FDA performance and specifically charges that the agency lacks the scientific expertise to fulfill its regulatory responsibilities.
The Science Board concluded that science at the FDA is in a precarious position; the agency suffers from serious scientific deficiencies and is not positioned to meet current or emerging regulatory responsibilities. Among the Science Board’s conclusions is the following: “Today, not only can the agency not lead, it can not even keep up with the advances in science.”
Consequently, the FDA continues to rely on inaccurate animal testing data for IND applications and drug approvals, contributing to the extraordinarily high clinical trial-drug attrition rate and the ongoing approval of dangerous drugs.
As FDA toxicology reviewer Anita O’Connor stated in 2004, “Most of the animal tests we accept have never been validated. They evolved over the past 20 years and the FDA is comfortable with them.” The hazards of such inertia are now widely recognized and reported. The FDA must no longer be comfortable with a failed drug-testing paradigm that puts the public health at risk.
Because the FDA has not made substantial efforts to adopt validated, accurate alternatives for animal testing of drugs or to investigate and approve potential alternatives through its National Center for Toxicological Research, a coalition of organizations submitted a formal petition requesting targeted action by the FDA. The Mandatory Alternatives Petition (MAP) was presented to the FDA in November 2007 and signed by more than 120 physicians, scientists, and other professionals.
The MAP reviews problems that make animal drug testing unreliable, such as species differences in anatomy, organ structure and function, drug absorption and metabolism, toxicokinetics, drug efficacy, and toxicity endpoints. Many examples are given of favorable animal-toxicity testing results followed by dangerous or lethal human results and conversely of long-standing beneficial drugs that display lethal animal toxicities.
The MAP also describes fields of drug development that have a 100% failure rate in human trials despite decades of successful animal testing. Prominent examples include stroke treatments (more than 150 failures), HIV/AIDS vaccines (more than 80 failures), and the entire field of cancer therapeutic vaccine research. Such disparate results can only be attributed to fundamental species differences that assure animal-testing results will be unreliable for humans.
The history, development, and advantages of scientifically sound replacements for animal-testing are presented in the MAP, including a variety of human cell and tissue methods, drug toxicity databases, computer modeling and simulations, microfluidics and tissue engineering methods, various omics techniques, scanning technologies, and microdosing.
Specific and definitive action is asked of the FDA in the MAP. “We request the FDA to promulgate a regulation to mandate that an animal experiment should not be performed if another scientifically satisfactory method for obtaining the result(s), not involving the use of animals, is available.”
This action would bring the FDA in line with European Council Directive 86/609/ EEC, which governs drug and chemical testing in the EU and would contribute to international harmonization for the validation, adoption, and implementation of alternatives. Several steps are recommended to facilitate the regulation.
The MAP comes at an opportune time, when there is unprecedented concordance on problems, consequences, and solutions regarding preclinical drug testing. Whereas postmarketing safety strategies can only improve identification of dangerous drugs after they are in widespread use, better preclinical drug testing can prevent unsafe drugs from reaching the public. For an FDA squeezed between good science and limited resources, the MAP provides a straightforward approach to issues surrounding animal testing of drugs that is scientifically sound and consistent with international best practices.