Increasingly, scientists and animal advocates are questioning the scientific validity of animal experimentation. Some of the main limitations of animal research are discussed in detail below:
Animal Studies Do Not Reliably Predict Human Outcomes
Both obvious and subtle differences between humans and animals, in terms of our physiology, anatomy, and metabolism, make it difficult to apply data derived from animal studies to human conditions.
Acetaminophen, for example, is poisonous to cats, but is therapeutic in humans; penicillin is toxic in guinea pigs, but has been an invaluable tool in human medicine; morphine causes hyper-excitement in cats, but has a calming effect in human patients; and oral contraceptives prolong blood-clotting times in dogs, but increase a human’s risk of developing blood clots. Many more such examples exist. Even within the same species, similar disparities can be found among different sexes, breeds, age and weight ranges, and ethnic backgrounds.
Furthermore, animal ‘models’ are seldom subject to the same causes, symptoms, or biological mechanisms as their purported human analogues. Indeed, many health problems currently afflicting humans, such as psychopathology, cancer, drug addiction, Alzheimer’s, and AIDS, are species-specific.
As a result, accurately translating information from animal studies to human patients can be an exercise in speculation. “Patients and physicians should remain cautious about extrapolating the findings of prominent animal research to the care of human disease.”  Even high-quality animal studies will replicate poorly in human clinical research.
Nine Out of Ten Drugs That Appear Promising in Animal Studies Go on to Fail in Human Clinical Trials
Due to the inherent differences between animals and humans, drugs and procedures that work in animals often end up failing in humans. According to Health and Human Services Secretary, Mike Leavitt, “nine out of ten experimental drugs fail in clinical studies because we cannot accurately predict how they will behave in people based on laboratory and animal studies.” 
A significant amount of time and money, not to mention animal lives, is squandered in the process. Pfizer, for example, reported in 2004 that it had wasted more than $2 billion over the past decade on drugs that “failed in advanced human testing or, in a few instances, were forced off the market because of liver toxicity problems.”
There have been numerous reports of approved drugs causing serious and unexpected health problems, leading the Food and Drug Administration (FDA) to remove the products from the market, or require black box warnings on their labels. The FDA has reported that “adverse events associated with drugs are the single leading contributor to preventable patient injury, may take the lives of up to 100,000 Americans, account for more than 3 million hospital admissions, and increase the nation’s hospitalization bill by up to $17 billion each year.” The agency estimates that drug-related injuries outside the hospital add $76.6 billion to health care costs.
Reliance On Animal Experimentation Can Impede and Delay Discovery
Drugs and procedures that could be effective in humans may never be developed because they fail in animal studies. It is difficult to know how frequently this occurs, since drugs that fail in animals are rarely tested in humans. There have, however, been some notable cases.
Although Lipitor, Pfizer’s blockbuster drug for reducing cholesterol, did not seem promising in early animal experiments, a research scientist requested that the drug be tested in a small group of healthy human volunteers. It was only then that its effectiveness was demonstrated. 
In many instances, medical discoveries are delayed as researchers waste time, money, effort, and animal lives trying to create an animal model of a human disease. A classic example is the discovery that smoking significantly increases the risk of lung cancer. The finding was first reported in 1954 on the basis of an epidemiological study. The report was dismissed, however, because lung cancer due to inhalation of cigarette smoke could not be induced in animal models. It wasn’t until 30 years later that the U.S. Surgeon General finally issued the warning on cigarettes.
Another noteworthy example concerns the development of the polio vaccine. Researchers spent decades infecting non-human primates with the disease, and conducting other animal experiments, but failed to produce a vaccine. The key event, which led directly to the vaccine and a Nobel Prize, occurred when researchers grew the virus in human cell cultures in vitro.
Animal Studies are Flawed by Design
We know that animals make poor surrogates for humans. On top of this, the design of animal experiments is often inherently flawed, making it that much less likely that results obtained from such studies will be useful.
Researchers from the Vanderbilt University Medical Center described some of the problems with animal ‘models’ in their 2004 article: “…[T]he design of animal studies automatically controls many variables that can confound human studies”; “…[T]he phenotypes studied in animals are not truly identical to human disease but are limited representations of them”; and “In most cases, animal studies do not assess the role of naturally occurring variation and its effects on phenotypes.” 
Furthermore, in their effort to secure research funds, expand the territorial boundaries and influence of their laboratories, or simply maintain their employment, it is common practice for biomedical researchers to generate an endless series of experiments. They do so by devising minor variations on a common theme, redefining previous work, subdividing one problem into multiple parts, or manipulating new technology and equipment to answer old or irrelevant questions. Such practices are endemic in such fields as experimental psychology, substance abuse/addiction, and most of the neuroscience and transplantation protocols. Yet, by their very design, these experiments do little to improve human or animal lives.
 Hackam, D.G. and Redelmeier, D.A. (2006). Translation of Research Evidence From Animals to Humans. Journal of the American Medical Association, 296(14): 1731-1732.
 Food and Drug Administration (2006, Jan. 12). FDA Issues Advice to Make Earliest Stages of Clinical Dug Development More Efficient. Press Release. Retrieved March 2008, from www.fda.gov/bbs/topics/news/2006/NEW01296.html.
 Rotman, D. (2004). Can Pfizer Deliver? Technology Review. Retrieved March 2008, from www.technologyreview.com/Biotech/13462/page1/.
 Food and Drug Administration (2004). Innovation or Stagnation: Challenge and Opportunity on the Critical Path to New Medical Products. Retrieved March 2008, from www.fda.gov/oc/initiatives/criticalpath/whitepaper.pdf.
 Agres, T. (2006, Mar. 9). FDA Input Aids Early Trials. Drug Discovery and Development. Retrieved March 2008, from www.dddmag.com/fda-input-aids-early-trials.aspx
 Williams, S.M., Haines, J.L., and Moore, J.H. (2004). The use of animal models in the study of complex disease: all else is never equal or why do so many human studies fail to replicate animal findings? Bioessays, 26(2): 170-179.