Many animal tests are badly flawed, say scientists

http://www.scenta.co.uk/scenta/features.cfm?cit_id=1365095&FAArea1=customWidgets.content_view_1

Many animal tests are badly flawed, say scientists
Source: Guardian Unlimited

The real value of animal experiments is questioned today by a team of senior scientists who found that many are flawed and do not predict how well a prototype medicine will work in humans.

The new paper, published by the British Medical Journal, is likely to be seized on by the animal rights lobby as substantiation for their case to stop all experiments. Their case was bolstered by the disaster of the Northwick Park clinical trial, where a drug that had been safe in animals had catastrophic side-effects in the human volunteers.

But the BMJ authors say the jury is out - it is not yet possible to determine how useful animal trials are because at present the methodological standards are poor.

"Sadly you can't say anything in this area without it being used politically by one or other interest group," said Professor Ian Roberts of the London School of Hygiene and Tropical Medicine, one of the authors. "But this paper is neither saying they are good nor that they are bad." That judgement could not be made, because of the poor standard of much of the work.
The British public has consistently said in polling that it supports animal tests if they help human healthcare, there is no alternative and no unnecessary suffering to the animals, he said.

The first question was one for scientists to answer - and this was the first attempt to take a rational look at the results of animal trials to see whether they predicted how well different drugs would work in humans.

The team looked at six drugs - two for stroke, and one each for head injury, haemorrhage, neonatal distress and brittle bones. They looked at results of the human clinical trials and then at the animal trials that had been carried out first.

They found that the results of the animal and the human trials were often different. Animals given corticosteroids after a head injury appeared to benefit, but when the drug was tried in humans, it did not help their recovery. Professor Roberts pointed to a major methodological flaw, however - rats were given corticosteroids just five minutes after their head injury.

The animal results for a drug called tirilazad for stroke were positive, but when the drug went into human trials, doctors found it actually increased the numbers who became dependent or who died. The paper, by Pablo Perel and colleagues, all from the London school, says of the 18 animal trials that "the quality of the experiments was poor".

Some of the animal studies were inconclusive, but in two cases, in osteoporosis (brittle bones) and neonatal respiratory distress, where babies struggle to breathe, they came up with the same result as the human trials.

The authors call for more systematic reviews like this one, where the results of a number of animal trials are pooled so that researchers can get a clearer idea of the effects of drugs.