Contact : 25 rue Jacques Callot - 66000 Perpignan - France
Tel : 33 (0)4 68 80 53 32
Lundi et jeudi de 10h30 à 18h / Monday and Thursday from 10h30 to 18h
Learn more
Join us

Why animals are not good models for the study of human disease

What is a model?

In order for an experimental object to be considered as a suitable study model, it must respond in an identical manner to the original, or else, provide a significantly similar result, within the framework of clearly defined laws and principles.

As an example of this, a small scale model of a submarine, built using the very same materials as the real thing, could be used to study its hydrodynamic properties, with a view to improving the performance of full- sized submarines.

When it comes to animal experiments, it seems fairly obvious at the outset, that there are no two species of animal sufficiently alike as to provide a suitable model for the other. Indeed, each species occupies a particular ecological niche, which is evident in terms of their individual eating habits, activity patterns, reproductive cycles and so on. All of these unique attributes are governed by physiological mechanisms, which are species-specific.

If we expose two different animal species to the same set of circumstances, such as a stress situation or exposure to a particular chemical, we would be unable to predict whether the outcome would be the same, or different. Whereas two centuries ago, one could have attempted to predict an outcome based on superficial anatomical or physiological similarities, modern-day knowledge has made us cognisant of the biological uniqueness of each and every species.

Indeed, ever since the beginning of animal experimentation in ancient times, budding physiologists recognised the inadequacy of their animal models and turned instead, to human experimentation. Under the patronage of the Ptolemaic dynasty in Alexandria, Herophilus and Erasistratus - two of the founding fathers of experimental physiology - were allowed and even encouraged to experiment on living humans. In the 19th century, the physiologist Claude Bernard who popularised the practice of vivisection using animals, stated that human studies logically would provide the best evidence. However, as he explained in his discourse 'An Introduction to the Study of Experimental Medicine', the prevailing laws and ethics precluded any experimentation on humans.

By law, all pharmaceutical drugs will have been tested on animals before being allowed onto the market. Yet despite these animal experiments, adverse drug reactions rank as one of the leading causes of death in developed nations. Bernard Kouchner, a former French health minister, put the annual figure for those who died from medical drugs at almost 20,000 and 1,3 million the number of cases requiring hospitalisation as a result of adverse reactions (Le Monde, 13 November 1997). If animals are indeed good models of humans, why is there no decrease in the number of these lethal adverse drug reactions? By the same token, why have we not yet found a cure for cancer, considering the many decades and the huge amounts of money spent looking for such a cure through animal research.

The following explanation by Antidote Europe illustrates why no animal species is a reliable model for another.

First and foremost, all species can be defined in terms of their reproductive isolation. That is to say, two different species cannot interbreed, with very rare exceptions, which invariably produce sterile progeny. This is due to the fact that, in order to produce a fertilised egg, the genetic material contained in the ovum and sperm that are to unite, must complement each other. This can only occur if the female and male animal are both of the same species. It cannot happen where the individuals are of different species, as the genetic material would be incompatible.

The second point of note is that all biological functions are determined by the genes of the individual. Overall, these biological functions are controlled chiefly by proteins, including enzymes, which help with functions such as the digestion of food, muscle contraction, the transport of oxygen in the blood, and so on. All of these proteins are unique and are geared to perform specific tasks. Their unique structure is determined, in turn, by the genes responsible for their synthesis. Two almost identical genes may produce two completely different proteins.

To sum up, every species of animal has a unique genetic code. The genetic code determines the structure of the proteins that ensure the biological activities associated with that species. Two different species will thus produce different proteins, which will be evident in terms of their biological activity.


A good example to illustrate this principle of species differences is the body's reaction to a chemical. Once a substance enters the bloodstream, it will reach the liver, where it will be metabolised. The way in which the substance is metabolised, or altered, will depend on the types of liver enzymes possessed by the individual. Thus, some compounds will be toxic for one individual, but not for another, depending on their liver enzymes. Since this principle applies to different individuals even within the same species, it makes an even more compelling case against trying to compare liver function between different animal species.

The poison arsenic is far more toxic to humans than to sheep, while formaldehyde is more carcinogenic to rats than it is to mice, and so on. However, even though we know this as a result of empirical data, we still would not know what would happen if we administered these same chemicals to a previously untested animal species. It turns out that animals are not the best models for the study of human disease or for finding new therapies. Human cancers and malignancies do not have much in common with artificially induced cancers in animals. Since we now know that cancers have their origins at the cellular and molecular levels, it makes sense to study these biological processes and phenomena in human cells.

Another example of species differences can be found in the field of infectious diseases. The chimpanzee is essentially immune to AIDS, and only mildly affected by the hepatitis B virus, whereas both of these diseases are severe and often deadly in humans. In contrast, the Ebola virus is lethal to humans and chimpanzees. Once again, these outcomes were discovered only after the fact, that is, on the basis of empirical observation in chimpanzees and in humans. By the same token, a therapy that is successful in chimpanzees would not necessarily be useful in humans.

All of these arguments and observations serve to confirm the logical conclusion that no species can reliably be used as a model for another species. Many researchers and medical scientists share this perspective and have expressed this view through various organisations, including AFMA (Americans For Medical Advancement), DLRM (Doctors and Lawyers for Responsible Medicine), EMP (Europeans for Medical Progress), PCRM (Physicians Committee for Responsible Medicine)and others. In a recent survey of 500 GPs conducted by EMP in the UK, more than 80% of respondents indicated that they did not trust data obtained from animal tests and that they would support an independent scientific evaluation of the whole subject.

Animal experimentation continues to provide academic careers for some, as well as providing industry with a false sense of security, by equating animal research with human safety standards. However, an awareness is growing that new methodologies are required to gain insight into important questions of human health and disease. The time for change is now.