Contemporary Animal Experimentation


A. Selected Diseases

1. Cancer

In 1971 the National Cancer Act initiated a "War on Cancer" that many sponsors predicted would cure cancer by 1976. Instead, this multibillion dollar research program has proven to be a failure. The age-adjusted total cancer mortality rate climbed steadily for decades until the early 1990s,35,36 when this rate started to fall slowly, due largely to reduced smoking.37

In order to encourage continued support for cancer research – now exceeding two billion dollars annually in the U.S. alone – researchers and administrators have misled the public. In 1987 the U.S. General Accounting Office (GAO) found that the statistics of the National Cancer Institute (NCI) "artificially inflate the amount of 'true' progress", concluding that even simple five-year survival statistics were manipulated.38 For one thing, the NCI termed five-year survival a "cure" even if the patient died of the cancer after the five-year period. Also, by ignoring well known statistical biases, the NCI falsely suggested advances had been made in the therapy of certain cancers.38

Commenting on the research program's discouraging results after 15 years, epidemiologist and program administrator John C. Bailar III stated in 1986: "[We] are losing the war against cancer. A shift in research emphasis, from research on treatment to research on prevention, seems necessary if substantial progress against cancer is to be forthcoming."39 In a review of cancer mortality more than a decade later, Bailar reiterated in 1997: "The more promising areas are in cancer prevention."35

Why hasn't progress against cancer been commensurate with the effort (and money) invested? One explanation is the unwarranted preoccupation with animal research. Crucial genetic,40 molecular,41 immunologic42 and cellular43 differences between humans and other animals have prevented animal models from serving as effective means by which to seek a cancer cure. Mice are most commonly used, even though the industry’s own Lab Animal magazine admits: "Mice are actually poor models of the majority of human cancers."44 Leading cancer researcher Robert Weinberg has commented: "The preclinical [animal] models of human cancer, in large part, stink… Hundreds of millions of dollars are being wasted every year by drug companies using these models."45 According to Clifton Leaf, a cancer survivor himself: "If you want to understand where the War on Cancer has gone wrong, the mouse is a pretty good place to start."45


Despite their extensive use since the early 1980s, animal models have not contributed significantly to AIDS research. While mice, rabbits and monkeys born with severe combined immunodeficiency can be infected with the AIDS virus (HIV), none develops the human AIDS syndrome.46 Of over 150 chimpanzees infected with HIV since 1984, only one allegedly developed symptoms resembling those of AIDS.47,48 Even AIDS researchers acknowledge that chimpanzees, as members of an endangered species who rarely develop an AIDS-like syndrome, are unlikely to prove useful as animal models for understanding the mechanism of infection or means of treatment.49

Other virus-induced immunodeficiency syndromes in nonhuman animals have been touted as valuable models of AIDS, but they differ markedly from AIDS in viral structure, disease symptoms and disease progression.50 Animal experimenter Michael Wyand, discussing anti-AIDS therapy, has acknowledged: “Candidate antivirals have been screened using in vitro systems and those with acceptable safety profiles have gone directly into humans with little supportive efficacy data in any in vivo [animal] system. The reasons for this are complex but certainly include … the persistent view held by many that there is no predictive animal model for HIV infection in humans.”51

AIDS researcher Margaret Johnston has concurred: "HIV/AIDS [animal] models have not yielded a clear correlate of immunity nor given consistent results on the potential efficacy of various vaccine approaches."52 Indeed, since the first HIV vaccine clinical trial in humans in 1987, more than 100 clinical trials have been funded by the U.S. National Institute of Allergy and Infectious Diseases through mid-2006. Yet every one of the more than 50 preventive vaccines and more than 30 therapeutic vaccines that were successful against HIV/AIDS in primate studies has failed in human clinical trials.53

Human clinical investigation has isolated HIV, defined the disease's natural course and identified risk factors.54 In vitro (cell and tissue culture) research using human white blood cells has identified both the efficacy and toxicity of anti-AIDS medicines, including AZT,55 3TC56  and protease inhibitors.57 Federal law, however, still mandates misleading and unreliable animal toxicity testing.

3. Psychology and Drug Abuse

Animal "models" in experimental psychology, which researchers traditionally subject to painful stimuli in order to study their behavior, have been strongly criticized in part because human psychological problems reflect familial, social and cultural factors that cannot be modeled in nonhumans.58-63 Indeed, most psychologists disapprove of psychological animal experiments which cause animal suffering.64

Harry Harlow's "maternal deprivation" experiments in the 1950s and 1960s involved separating infant monkeys from their mothers at birth and rearing them in total isolation or with "surrogate" mothers made of wire and cloth. Their terror and subsequent psychopathology, Harlow claimed, demonstrated the importance of maternal contact. However, this had been shown conclusively in previous human studies.65-68 

Despite their conceptual shallowness, numerous maternal deprivation studies continue, claiming relevance to human developmental psychology, psychopathology and even immune and hormone function.67-69

Experimental psychology continues to rely on painful research on animals, despite clinical psychologists' disregard for animal research literature. A review of two clinical psychology journals revealed that only 33 out of 4,425 citations (0.75%) referred to animal-research studies.70

Animal models of alcohol and other drug addictions are similarly ill-conceived, failing to reflect crucial social, hereditary and mental factors. Pharmacologist Vincent Dole has acknowledged: "Some 60 years of offering alcohol to animals has produced no fundamental insights into the causes of this self-destructive behavior or even a convincing analogue of pathological drinking."71

4. Genetic Diseases

Scientists have located the genetic defects of many inherited diseases, including cystic fibrosis and familial breast cancer. Trying to "model" these diseases in animals, researchers widely use animals – mostly mice – with spontaneous or laboratory-induced genetic defects. However, genetic diseases reflect interactions between the defective gene and other genes and the environment. Consequently, nearly all such models have failed to reproduce the essential features of the analogous human conditions.72 For example, transgenic mice carrying the same defective gene as people with cystic fibrosis do not show the pancreatic blockages or lung infections that plague humans with the disease,72 because mice and humans have different metabolic pathways.73

B. Toxicity Tests

Numerous standard animal toxicity tests have been widely criticized by clinicians and toxicologists. The lethal dose 50 (LD50) test – which determines how much of a drug, chemical or household product is needed to kill 50% of a group of test animals – requires 60 to 100 animals (usually rats and mice), most of whom endure great suffering. Because of difficulties extrapolating the results to humans, the test is highly unreliable.74 Also, since such variables as an animal's age, sex, weight and strain can have a substantial effect on the results, laboratories often obtain widely disparate data with the same test substances.75,76 In vitro tests have been validated to replace the LD50 test,76-78 which was deleted from the test guidelines of the Organisation for Economic Cooperation and Development (OECD) in 2002.79

The Draize eye irritancy test, in which unanesthetized rabbits have irritant substances applied to their eyes, yields results that are inherently unreliable in predicting human toxicity.80 Humans and rabbits differ in the structure of their eyelids and corneas, as well as in their ability to produce tears. Indeed, when comparing rabbit to human data on duration of eye inflammation after exposure to 14 household products, they differed by a factor of 18 to 250.81 A battery of in vitro tests would be less expensive and likely far more accurate than the Draize test.75,82

Animal tests for cancer-causing substances, generally involving rodents, are also notoriously unreliable. When applied to human cancer causation, Lester Lave et al. found the false positive rate of rodent testing to be as high as 95%.83 The authors stated: "Tests for human carcinogens using lifetime rodent bioassays are expensive, time-consuming and give uncertain results." The tremendous economic costs of such research have recently been reported in a study which examined over 500 rodent carcinogenicity studies and concluded that rodent cancer assays are scientifically invalid and fiscally indefensible.84

A combination of in vitro tests provides data that compares favorably with existing carcinogenicity databases and costs far less than animal tests.85 In the late 1980s, the U.S. National Cancer Institute (NCI) developed a panel of 59 human cancer cell lines to screen compounds for anti-cancer activity, due to its "dissatisfaction with the performance of prior in vivo primary screens [animal cancer assays]."86 This panel replaced animal testing at the NCI in 1990, by which time the agency had also adopted a panel of about 100 human cell lines to screen compounds for carcinogenicity.87

Animal tests for teratogens (drugs and chemicals that cause birth defects) are equally misleading and unreliable. Jarrod Bailey et al. conducted a comprehensive review of animal tests of 1,396 different substances and found that of those substances known to cause birth defects in humans, animal tests indicated that almost half were safe. Conversely, of those substances known to be safe in humans, animal tests indicated that almost half were dangerous. And almost one-third of all substances tested yielded varying results, depending on the species used.88 In pregnant animals, differences in the physiological structure, function and biochemistry of the placenta aggravate the usual differences in the absorption, distribution, metabolism and excretion of drugs and chemicals that exist between species, thus making reliable predictions in pregnant women impossible.88

In vitro tests, such as the embryonic stem-cell test, the whole embryo culture, and the micromass test, provide data that are considerably more reliable and predictive and far less costly than animal teratogenicity tests. While such in vitro tests currently utilize cells and embryos derived from animals (thus rendering their extrapolation to humans difficult), advances in human cell culture technology should, in the future, permit a much closer in vitro approximation of teratogenesis in humans.88

C. Medical Education

Animal laboratories are not necessary for teaching biological and medical principles and skills to medical students, and 85% of U.S. and Canadian medical schools have eliminated animal labs from their educational curricula.89 Effective alternative teaching methods include lectures and written course materials, videos and interactive virtual reality programs, mentored patient care encounters and surgery participation, and lifelike programmable interactive patient simulators. Comparative studies of simulation technologies for many aspects of medical education (e.g. anatomy, physiology, pharmacology, surgical skills, trauma management and invasive procedures) have repeatedly demonstrated superior training outcomes, fewer patient complications, greater trainee acceptance, and more efficient use of educational time and resources.90-99

Further evidence of the emerging primacy of simulation-based medical education is the American College of Surgeons’ (ACS) endorsement and implementation of the TraumaMan® simulator to replace the use of animals and human cadavers for its Advanced Trauma Life Support (ATLS) program. Furthermore, in 2006 the ACS implemented a sweeping educational reform that incorporated a wide variety of simulators to eliminate animal use in its own conferences and educational programs, in addition to establishing the Accredited Education Institutes program to achieve the same goal in surgery training programs.100

Scientific Limitations of Animal Models

Animal studies can neither confirm nor refute hypotheses about human physiology or pathology; human clinical investigation is the only way such hypotheses can be tested.  At best, animal experiments can suggest new hypotheses that might be relevant to humans.101,102 However, there are countless other, far superior ways to derive new hypotheses.2,101

How valuable is animal experimentation? The Medical Research Modernization Committee's review of ten randomly chosen animal models of human diseases did not reveal any important contributions to human health.103 Although the artificially induced conditions in animals were given names analogous to the human diseases they were intended to simulate, they differed substantially from their human "counterparts" in both cause and clinical course. Also, the study found that treatments effective in animals tended to have poor efficacy or excessive side effects in human patients.103 Indeed, when MRMC physicians evaluate specific animal-research projects, they consistently find them to be of little, if any, relevance to the understanding or treatment of human diseases.104-110

MRMC's reviews have revealed that, because animal models differ from human diseases, researchers tend to investigate those aspects of the animal's condition that resemble features of the human disease, generally ignoring or discounting fundamental anatomical, physiological and pathological differences.  Because most disease processes have system-wide effects and involve many interacting factors, focusing on only one aspect of a disease belies the actual complexity of biological organisms.

In contrast to human clinical investigation, animal experimentation involves manipulations of artificially induced conditions. Furthermore, the highly unnatural laboratory environment invariably stresses the animals, and stress affects the entire organism by altering pulse, blood pressure, hormone levels, immunological activities and a myriad of other functions.111,112 Indeed, many laboratory "discoveries" reflect mere laboratory artifact.10,113-119

For example, artifact from unnaturally induced strokes in animals has repeatedly misled researchers.117,120 Macleod et al. reported on over 4,000 studies demonstrating efficacy for more than 700 drugs in animal models of stroke.121 About 150 drugs subsequently tested in human clinical trials failed to show any benefit.122 Only recombinant human tissue plasminogen activator (rt-PA) administered within three hours of stroke onset has proven beneficial in reducing symptoms, but it was associated with ten times as many intracerebral hemorrhages and did not increase survival.123 David Wiebers et al. have concluded: "Ultimately, the answers to many of our questions regarding the underlying pathophysiology and treatment of stroke do not lie with continued attempts to model the human situation more perfectly in animals, but rather with the development of techniques to enable the study of more basic metabolism, pathophysiology and anatomical imaging detail in living humans."117

Since 1990, several hundred gene therapies that were successful in animal studies have been tested on thousands of patients worldwide. Yet only one gene therapy, for children with the severe immune system disorder X-SCID, appears to have succeeded. Of the ten successfully treated children, however, three developed leukemia and one of them died of it – a side effect that animal experiments failed to predict and that prompted the U.S. Food and Drug Administration (FDA) to halt several gene therapy trials in 2005.124,125 Similarly, a highly touted gene therapy that cured dogs of hemophilia was discontinued in 2004 due to "afety problems … in the human trial that weren’t predicted in animal studies", including liver damage.126,127

Animal tests are frequently misleading.128 Milrinone increased survival of rats with artificially induced heart failure, but humans taking this drug experienced a 30% increase in mortality.129 Fialuridine appeared safe in animal tests, but it caused liver failure in 7 out of 15 humans taking the drug, five of whom died and two of whom required a liver transplantation.130 Animal studies failed to predict the dangerous heart valve abnormalities in humans caused by the diet drugs fenfluramine and dexfenfluramine.131

Hormone replacement therapy increased women’s risk of heart disease, breast cancer and stroke, but experiments with mice, rabbits, pigs and monkeys had predicted the opposite effect.132 The widely prescribed arthritis painkiller Vioxx appeared safe and even beneficial to the heart in animal tests, but was withdrawn from the global market in 2004 after causing an estimated 320,000 heart attacks, strokes and cases of heart failure worldwide – 140,000 of them fatal.133 David Graham, the Associate Director for Science and Medicine in the Office of Drug Safety at the FDA, described Vioxx as the "single greatest drug safety catastrophe in the history of this country or the history of the world".134 Animal tests also failed to predict the cases of partial or total blindness suffered by some men taking the popular impotence drug Viagra.135,136 Despite mandatory, extensive animal testing, adverse drug reactions remain the fifth leading cause of mortality in the United States, accounting for more than 100,000 deaths per year.137

In London in March 2006, a new anti-inflammatory drug called TGN1412 caused devastating reactions including multiple organ failure in all six volunteers in phase 1 clinical trials, despite "proof of safety" established by tests on monkeys who were given 500 times the human dose. Many commentators noted that the animal tests provided a false sense of security. The incident prompted calls for an overhaul of drug safety testing requirements and clinical trial design.138

In animal tests to evaluate the carcinogenicity of the artificial sweetener saccharin, the weight-adjusted daily saccharin dose given to rats was equivalent to a human consuming about 1,100 cans of soda containing saccharin. Such massive dosing alone can result in cancers, regardless of a compound's actual carcinogenicity at typical human exposure levels.116 Extrapolating such data to humans is further complicated by the observation that saccharin-induced bladder cancers occurred only in male rats. It was later found that male rats possess a protein in greater quantity than female rats (and lacking in humans) that interacted with saccharin to form irritating crystals in the male rats' bladders, causing cancer. The fact that some rats developed cancers did not (and cannot) clarify whether or not saccharin causes cancer in humans.139

Similarly, despite almost 40 years of human consumption, its use in more than 9,000 food and beverage products worldwide, and the irrelevance of animal tests to humans, the artificial sweetener aspartame is still being tested on animals, and regulatory authorities continue to evaluate the results of such studies. Most recently, an Italian study carried out in 2005 on 1,800 rats demonstrated an increased risk for lymphomas and leukemias in rats fed aspartame – but only in females.140 A subsequent NCI epidemiological study involving 340,045 men and 226,945 women and reported on at the 2006 meeting of the American Association for Cancer Research refuted the findings in rats.141 So, despite male rats getting bladder cancers from saccharin and female rats getting lymphomas and leukemias from aspartame, no cancer risk from either sweetener has been found for humans of either sex.

Scientists recognize that, even between humans, gender, ethnicity, age and health can profoundly influence drug effects.142,143 Perhaps the most striking example of the specificity of drug effects comes from the demonstration that even human monozygotic twins display different drug responses and that these become more disparate as the twins age.144 Obviously, extrapolating data between species is much more hazardous than within a species. Indeed, according to the FDA, a staggering 92% of all drugs found safe and therapeutically effective in animal tests fail during human clinical trials due to their toxicity and/or inefficacy, and are therefore not approved.145-147 Furthermore, over half of the mere 8% of drugs which do gain FDA approval must later be withdrawn or relabeled due to severe, unexpected side effects.148

Risks of Animal Experimentation

In addition to squandering scarce resources and providing misleading results, animal experimentation poses real risks to humans. The mind-set that scientific knowledge justifies and requires harming innocent individuals endangers all who are vulnerable. Even after Nazi and Japanese experiments on prisoners horrified the world, American researchers denied African-American men syphilis treatment in order to assess the disease's natural progression,149 they deliberately exposed students and minorities to toxic chemicals in order to determine safe levels of exposure to pesticides,150 they intentionally exposed thousands of unsuspecting civilians to lethal bacteria in order to test biological warfare,151 they injected cancer cells into nursing home patients,149 subjected unwitting patients to dangerous radiation experiments,152 and, despite no chance of success, transplanted nonhuman primate and pig organs into children, as well as chronically ill and impoverished people.153 Psychiatrist Robert Jay Lifton argues that this "science at any cost" mentality may have provided medical justification for the Holocaust.154

Furthermore, through animal research, humans have been exposed to a wide variety of deadly nonhuman primate viruses. About 16 laboratory workers have been killed by the Marburg virus and other monkey viruses, and two outbreaks of Ebola have occurred in American monkey colonies.155-157 Polio vaccines grown on monkey kidney cells exposed millions of Americans to the simian virus 40, which causes human cells to undergo malignant transformation in vitro and has been found in several human cancers.158 Ignoring the obvious public health hazards, researchers transplanted baboon bone marrow cells into an AIDS patient. The experiment was unsuccessful;159 moreover, a large number of baboon viruses, which the patient could have spread to other people, may have accompanied the bone marrow. Indeed, animal experimentation may have started the AIDS epidemic. HIV-1, the principal AIDS virus, differs markedly from all other viruses found in nature, and there is evidence that it originated either through polio vaccine production using monkey tissues160,161 or through manufacture in American laboratories, where HIV-like viruses were being produced by cancer and biological weapons researchers in the early 1970s.162

Failing to learn from the AIDS epidemic, many policy makers and industrial interest groups support animal-to-human organ transplants (from pigs and primates) known as xenotransplants. These have failed in the past and will most likely continue to fail because of tissue rejection, the impossibility of testing animal tissues for unknown pathogens, and the prohibitive expense.163-165

Similarly, the rapidly expanding field of genetic engineering includes adding genetic material to animals’ cells to change the animals’ growth patterns or induce the animals to produce human proteins in their milk, meat or urine. Harvesting such proteins poses serious human health risks, such as exposure to pathogens (viruses, prions and other microorganisms)166,167 or the development of malignancies,168,169 allergic reactions170 or antibiotic resistance.171 These concerns contributed to the European Union’s ban on rBGH, a genetically engineered bovine growth hormone that increases cows’ milk production.172

The Importance of Clinical Research

Typically, medical discovery begins with a clinical observation,9,10 which animal experimenters then try to mimic with artificially induced conditions in laboratory animals.7 These researchers tend to highlight animal data that agrees with the previous clinical finding, while discounting or ignoring conflicting animal data (which is usually voluminous). Although animal experimentation advocates routinely take credit for discoveries that actually occurred in a clinical context,7 many clinicians have recognized the primary role of human-based clinical research. Reviewing the history of hepatitis, physician Paul Beeson concluded: “Progress in the understanding and management of human disease must begin, and end, with studies of man… Hepatitis, although an almost 'pure' example of progress by the study of man, is by no means unusual; in fact, it is more nearly the rule. To cite other examples: appendicitis, rheumatic fever, typhoid fever, ulcerative colitis and hyperparathyroidism.”11

Similarly, key discoveries in immunology,12 anesthesiology,13 first aid,173 alcoholism71,174 and psychopharmacology175,176 were based primarily on human clinical research and investigation. Furthermore, clinical research is the only means by which effective public health education and prevention programs can be developed and evaluated.

Return to main menu

Table of Contents

Next page -->

Home  Info  Join