Animal Testing Experiments Misleading Results
By Doctor Robert Sharpe, Scientific Director, International Association Against Painful Experiments on Animals
Scientists at the University of Oklahoma described the effects of LSD on a male elephant called Tusko. They were attempting to induce “musth”, a natural condition in which elephants periodically become violent and uncontrollable. The dose of LSD was calculated from the amount that puts cats into a rage, adjustments being made for the elephant’s greater body weight. At 8.00am, Tusko received his injection but the dose proved too large. Tusko immediately began trumpeting and rushing around the pen, then he stopped and swayed, finding it increasingly difficult to stay upright. His mate Judy tried to support him but Tusko collapsed, defecated and went into convulsions. Within 2 hours he was dead. The scientists concluded that elephants are particularly sensitive to LSD.
It may be hard to believe that scientists carry out such experiments but the result, that one animal is more susceptible than another, would come as no surprise to veterinarians. Knowledge of differences between the species is crucial in veterinary practice since animals often vary in their response to drugs, poisons and disease.
There are many examples. Although dichlorvos is safely used to treat intestinal worms in the horse, it can poison poultry. In contrast, sodium monensin controls parasites in broiler chickens but is lethal to horses. Aspirin poisons cats, as do phenol-based disinfectants; rabbits can eat freely the leaves of the deadly nightshade; horses suffer brain disease if they eat yellow star thistle, a weed found in the Southern United States, although other animals are not affected; rats, rabbits and mice are physiologically unable to vomit; the intestinal treatment haloxon is 100 times more toxic to geese than to hens; rats are hardly affected by TB; and unlike guinea pigs, cats and dogs do not need vitamin C in their diet. As one veterinarian explains, “it is unwise to extrapolate information concerning drugs from one species to another.”
It might be thought that similar, common-sense conclusions would apply to human medicine. After all, knowledge can be gained through human population studies, clinical observation of patients who are ill or who have died, work with healthy volunteers, test-tube experiments with human tissues, and computer simulations programmed with data from patients, all of which are directly relevant to people. In fact millions of animals suffer and die as human surrogates, and America’s National Institutes of Health spends more on research with animals than on investigations of our own species! It is as if the problems caused by species variation have conveniently disappeared, since they are rarely mentioned by animal research advocates in their propaganda.
It is easy to understand why the issue is ignored or avoided. In the heated debate surrounding animal experiments, an admission that results may only be relevant to the species under test would destroy the pretence that vivisection is vital to our health, or that all medical advances require animals. Examples described here show that, on the contrary, animal research can be dangerously misleading.
It might still be argued that these are only isolated cases and that most animal experiments correctly predict human responses. Perhaps surprisingly, few major comparative studies have been carried out but an analysis of drug side-effects by Ralph Heywood, former Director of Britain’s Huntingdon Research Centre, found only a 5-25% correlation between harmful effects in people and the results of animal experiments. This confirms the view of Dr Kenneth Melmon of the Cardiovascular Research Institute, University of California, that “In most cases, the animal tests cannot predict what will happen when the drug is given to man.”
Some side-effects are better predicted than others. A review of 45 drugs by Britain’s Committee on Safety of Medicines found that animal experiments were most likely to predict vomiting and gastrointestinal disturbances. Overall, however, the survey found that, at best, just 25% of the toxic effects observed in animal tests actually occurred in people.
An especially controversial area is the use of animals to identify cancer-causing chemicals. In 1983 the pharmaceutical company Pfizer carried out a special study to test the efficiency of animal tests. The results would be vitally important because despite costing millions of dollars, no one really knew whether they provided adequate protection against hazardous substances. Human findings were compared with experimental data from rats and mice for all chemicals known to cause cancer in people. The outcome was disturbing: in most cases animal tests had given the wrong answer. The report concluded that we would have been better off to toss a coin!
Subsequent reports, however, suggest that when tested “adequately”, nearly all human carcinogens have, eventually, been shown to cause cancer in some species of animal. But this is misleading: if substances like asbestos, tobacco, arsenic, benzene, alcohol and soot were not already suspected of causing cancer in people, scientists would never have persisted with attempts to induce the disease in animals. Reliance would have been placed on one or two routine feeding or inhalation tests to which new chemicals are now submitted. Since all the first animal tests for the above-named substances proved negative, some of the most dangerous human carcinogens would have been deemed safe.
Misgivings over the validity of animal research also apply to the study of disease. After analysing 10 randomly chosen “animal models” of human illness, the Medical Research Modernization Committee found little, if any, contribution towards the treatment of patients. They discovered, for instance, that chemically-induced cancer of the colon in rats was not relevant to our understanding, diagnosis and treatment of human colon cancer.
Obviously there are times when people and animals respond in a similar way, although this is never known until after human studies have taken place. The problem is that artificially induced disease in animals is never identical to the naturally arising disorder in patients, making animal research a logically flawed process. Furthermore, the differences in body chemistry between species can be so great that one well-known drug researcher states “…it is often a matter of pure luck that animal experiments lead to clinically useful drugs.”
In the scientific journals at least, some animal researchers recognise that species differences limit the value of their experiments. David Galloway of Glasgow’s Western Infirmary concedes that “the ultimate dilemma with any animal model of human disease is that it can never reflect the human situation with complete accuracy.”17 Nevertheless, experimenters insist they are still able to generate insights and, focussing on cancer of the colon, Galloway argues that “The lack of a single animal model which closely parallels the human disease need not be regarded as a major disadvantage.”
Tragically, vivisection is so ingrained that entirely new strains or species are being developed in an elusive search for the ideal animal model. Using the techniques of genetic engineering, scientists can alter an animal’s genetic make-up, producing creatures that are designed to suffer and die in medical research. The unfortunate “oncomouse”, produced by inserting human cancer genes into the embryos of mice, quickly develops cancer and dies within 90 days. In other cases researchers have introduced similar genetic defects to those found in sick people.
Although experimenters hope that genetically engineered animals will more accurately mimic human disease, “…quite often the model does not resemble the corresponding human condition as closely as might be expected.” In the case of Waadernburg’s syndrome type 1, a genetic disorder in which loss of hearing is a common feature, the genetically engineered mice do not become deaf. Another example is “cystic fibrosis mice”.
Discovery of the cystic fibrosis gene in human sufferers led vivisectors to disrupt a similar gene in mice. The animals become ill and die within 40 days but although hailed as a major breakthrough, there are differences from the disease in people: most importantly the animal’s lungs do not become infected or blocked with mucus as they do in human patients yet it is lung infections that kill 95% of people with cystic fibrosis.
A further case is the use of transgenic mice in AIDS research. In an attempt to make mice susceptible to the disease, the genetic material which produces HIV was inserted into their tissues. Again, the animals do become ill but their immune system is not suppressed as it is in people with AIDS.
Whatever animal research may have contributed in the past, this is no reason for uncritical acceptance of the method for all time. Moral values change and the mediaeval idea that lives can only be saved by sacrificing animals is no longer acceptable. These ethical objections are strongly reinforced by scientific arguments since the physiological and biochemical differences between people and animals stress the urgent need for a more reliable approach: it is in all our interests to challenge animal experiments.
If there is to be a more enlightened future, pressure for reform must be combined with a new generation of scientists who no longer regard animals as the disposable tools of research.