The British Parliament, for all its efficacy at passing thousands of pieces of legislation each year and civilising debate, nevertheless still contains one strange confection – the Early Day Motion (EDM).
The name makes it sound as if these statements, which Members of Parliament can put their names to in support, lie at the foothills of some sort of legislative process, sparking debates and justifying Parliamentary scrutiny. In fact, the process begins and ends with MPs sending a letter to the Vote Office, where their name will be added to a list. They are essentially meaningless.
They are beloved, however, by lobbyists who can claim to their clients that they have gained enormous influence over the policymaking process and journalists who can report that a “Parliamentary Motion” has been put forward, garnished by quotes for and against whatever it is the EDM is promoting.
EDMs, then, are naturally attractive to animal rights activists, who also like to look busy. Like politicians, they can claim that they are doing something and, as lobbyists, can claim to be influencing Parliament. They can even give the journalist their quote.
Animal rights activists also have a rich heritage of making spurious claims and peddling pseudoscience. So what do we get when EDMs meet pseudoscience? One splendid recent example is EDM 263, which asserts:
“That this House notes the new campaign For Life On Earth which is critical of avoidable experiments on animals; is alarmed that all studies measuring the claimed ability of animals to predict human responses expose a low success rate in the region of 31 per cent; further notes that a success rate in the region of 90 per cent is required by medical practice; further notes that the National Cancer Institute has said that cures for cancer have been lost because studies in rodents have been believed; and calls for properly moderated scientific public debates on the misleading results and bad science of animal experiments.”
Regular readers will be familiar with For Life On Earth (FLOE), which draws heavily on the work of Ray Greek and Niall Shanks, from an earlier SR investigation into its activities. So what on Earth have these 22 MPs put their names to?
I suppose the first thing to note is that the UK legislative environment is entirely different to that in the US but, putting that aside, every aspect of the EDM is a fraud. The reference to the National Cancer Institute is a case in point. Far from criticising animal models, the NCI have a website devoted to demonstrating why animals are essential to researching cancer. The source of the EDM’s claim comes from a 1997 news article in Science:
“But not only have very few of the drugs that showed anticancer activity in xenografts made it into the clinic, a recent study conducted at the National Cancer Institute (NCI) also suggests that the xenograft models miss effective drugs. The animals apparently do not handle the drugs exactly the way the human body does. And attempts to use human cells in culture don’t seem to be faring any better, partly because cell culture provides no information about whether a drug will make it to the tumor sites.”
So the article, referring only to xenografts, discussed why there was a lack of new cancer drugs (in 1997), partly due to no methods – animal or non-animal – being effective enough at identifying working drugs. Certain animal methods had missed cancer treatments, but so had, for example, tissue cultures and clonogenic assays. The article does discuss the promise of the then new hollow fibre models, which rely on the use of mice, and have shown positive results in recent years. The cherry picking of quotes from the news section of scientific journals gives a misleading impression regarding the efficacy of animal experiments. Since then other improved mouse cancer models of cancer have been widely adopted, including genetically modified mice and “tumourgrafts”, leading to the development of new therapies.
Elsewhere, we encounter the assertion that “a success rate in the region of 90 per cent is required by medical practice” which, apart from being not supported by its references, is complete nonsense, particularly in a country where homeopathy is available on the NHS (I know!). It is also inapplicable to some 90% of research, and rather neglects to mention that animal research does not claim to achieve prediction rates of 90% – that’s why there are three stages of human trials before drugs are used in a clinical setting. Likewise with the reference to “avoidable” experiments – what is an “avoidable” experiment, particularly since it is the case that each UK experiment must specifically be licensed as unavoidable by the Home Office?
The idea of a debate is interesting, and indeed many of us do participate in such debates already – on campuses, in the press, on television and online. However, it is a rotten way to make policy, and the opposite of the scientific method. Protesters who are confident in their speaking ability love a debate – the science is too complex to get across to a lay audience, but narratives can be adequately developed and the “Gish Gallop” rhetorical trick can be employed as it has been in the text of the EDM. If they were as confident in their science they would be using their time to write papers for scientific journals, since the probable truth is established via scientific method, scepticism and challenge, not a bun fight in a church hall followed by a show of hands.
However it is the claim of the ‘low success rate of 31%’ that really showcases pseudoscience at its most insidious.
Putting aside the odd “success rate” term, the reference given by FLOE appears in numerous papers by Ray Greek, which is presumably where FLOE found it. It references an out-of-print book from 1990 called Animal Toxicity Studies: Their Relevance for Man, published by Quay. I managed to find a second-hand copy in Maryland and had it shipped to the UK.
In it, Heywood makes a claim that the correlation (not prediction) of toxicology effects seen between humans and rats is “about 30%”. Other studies from 1990 pegged it at closer to 70% given their understanding at the time. Heywood’s claim is not substantiated in the book, but the reader is referred to one of his own papers from 1981: Target organ toxicity in Toxicology Letters, 8 : 349 – 358. Reference 2 of this paper accompanies this passage:
“Fletcher [2] reported on drug safety tests and subsequent clinical experience with 45 major new drugs. Some effects were seen only in animals, while others were observed only in man. The survey established that 25% of toxic effects observed in animals might be expected to occur as adverse reactions in man.”
Once again, we have no evidence, but we do have a further reference: A.P. Fletcher. Drug safety tests and subsequent clinical experience, J. R. Sot. Med., 71 (1978) 693-696.
Journeying further into the rabbit-hole, we discover that it’s a paper in the Journal of the Royal Society of Medicine, Volume 71, September 1978, called ‘Drug safety tests and subsequent clinical experience’ by one A P Fletcher MB PhD of the former Department of Health and Social Security. Hardly “all studies”.
AP Fletcher had been looking at “45 major new drugs that have been considered by the Committee on Safety of Medicines during the past eight or nine months.”
And on the basis of this small sample size and 1970s toxicology science concluded that:
“It can be said with certainty that correlations between animal toxicity and adverse side effects in man do exist and that they are considerably more frequent than discrepancies. As a very approximate estimate, for any individual drug, up to 25% of the toxic effects observed in animal studies might be expected to occur as adverse reactions in man.”
Why would only 25% of the toxic effects seen in preclinical animal studies be observed as adverse reactions in humans? One frequent reason was dosage; many of the adverse effects were seen at high doses – including acute toxicity tests- in animals, doses that were a lot higher than those used in subsequent human trials. It’s hardly surprising that many adverse effects are not seen in the clinical trials. it’s what you would expect to see if there is concordance between the animal and human studies. The animal studies also included tests of reproductive toxicity that were not assessed in humans, since pregnant women were excluded from clinical trials.
Our ‘region of 31%’, then, is based on a “very approximate estimate” of one small aspect of animal research, by a Department of Health doctor in 1978 based on 45 drugs that happened to have been licenced in the last year. Not only does the paper conclude that animal experiments correlate to human reactions, in terms of prediction it was a generation ago, it was a different regulatory environment, it was based on a 35-year old understanding of toxicology, it had a sample size of 45 and it was a rough analysis. The claim that this paper shows that animal studies have a “low success rate of 31%’” is simply pseudoscience.
As well-intentioned as I’m sure FLOE is, this jolly but dishonest organisation has misconstrued toxicology experiments as all “animal experiments”, using a cherry-picked quote and a guessed percentage based on an incorrect interpretation of a paper from 35 years ago wrongly extrapolated not just to all animal experiments, but all efficacy metastudies, to support an EDM that has fooled 22 of the UK’s lawmakers. The further risk is that the MPs will be taken in again, perhaps repeating similar myths in a debate, or legislating on the basis of bunkum to the detriment of man, animals and the environment.
It is at times like this that we feel furthest away from a sensible dialogue on the purpose and ethics of animal research. Researchers want to use the most predictive models, and if the debate was about how to refine statistical and biological models, as is it within the scientific community, we would all be happy. However, when gullible MPs are being tricked into flinging mud at medical, veterinary and scientific researchers on the basis of a paper-thin claim about the utility of that research, it seems clear that we still have a long way to go.
Chris
Speaking of Research 