To predict the effect of a chemical on people, scientists have traditionally looked at that chemical's effects on animals. Classical toxicology works on the premise of a similarity between animals and humans. This empiricism lies behind most tests of a new drug that are required by law.
Yet this classical approach has sometimes led to dramatic failures, as in the case of the drug thalidomide, which passed animal tests but distorted the limbs of babies born to some women who had taken the drug. The truth is that animals do not always react to a drug in the same way as humans.
The two-year studies in rodents used to test whether a chemical can cause cancer are associated with difficulties. The pharmaceuticals industry and the regulatory authorities acknowledge that experiments on rodents are of limited relevance.
Alternatives to testing on animals, using human cell and tissue cultures, are available. It is, of course, essential that these alternative methods be fully validated before they are accepted. But their performance is assessed against unfair criteria. Very often, the classic animal study is used as the gold standard, and the results of an alternative method must be shown to correlate to the results of the original animal test.
For example, in the Draize test, in which chemicals are dropped into rabbits' eyes, the scoring system is subjective and the results are variable. The inherent variability of Draize data makes it impossible for any alternative method to achieve the required degree of correlation. There is a fundamental illogicality in requiring the results of an alternative system, which may be based on human cells, to correlate with data derived from animals.
Regulators are justifiably conservative. They are responsible for the safety of millions of people. However, it is a sad fact that their practice is based on non-validated animal studies. While they may nod through proposals for animal studies, they put impossible obstacles in the face of alternative methods. Regulators may feel comfortable with animal studies; in my view it is high time this feeling of comfort is challenged.
Krys Bottrill works for the Fund for the Replacement of Animals in Medical Experiments. Last week she addressed the Royal Society of Chemistry's annual conference
* Interview by Helen Hague