Quotes from Doctors & Researchers:

It is often claimed that scientists and doctors are united in their belief in the value of, and necessity for, animal experiments to protect human health. This stream of quotes from scientists and doctors, stretching from as far back as the early 20th Century right up to today, shows that there has been a long tradition of scepticism about this issue. Many of the quotes are from scientists who support or conduct animal research; far from diminishing the impact of their words, this fact ought to give them extra weight.

Note: FDA stands for Food and Drug Administration, the US drug regulatory body.

Cancer | Neurological diseases/ conditions | Teratogenicity and thalidomide | Immunology & AIDS & TGN1412 | Miscellaneous

Miscellaneous

2009

It started out not wanting to hurt bunnies, but we've found that using [human equivalent] tissue is better science, more cost effective, and much more accurate than doing the same tests on animals. Dave Ingalls, spokesman for MatTek. Quoted by Colin Nickerson, the Boston Globe, 30^th March.

Over the last twenty years e-learning has transformed the way in which university students are taught and supported in their learning. For many students it is even possible to replace laboratory-based practical classes in disciplines such as pharmacology and physiology by computer-based learning programs which have been shown to offer effective alternatives to traditional teaching methods - similar learning objectives are achieved, costs are often lower, students are happy and unnecessary use of animals is reduced. David Dewhurst, Professor of E-Learning at Edinburgh University http://www.politics.co.uk/opinion-formers/press-releases/animal-welfare/adi-meps-buckle-under-vivisection-industry-pressure-$1293593$464772.htm

The following quotes are all from Dr Kelly BéruBé, Cardiff University, speaking on the Today Programme, BBC Radio 4, 4^th June

Now we can do our experiments in petri dishes with functional human tissue. Now you have human data, so no need for the rat. I say, why use a rat when a human lung will do?

I don’t see any reason why we can’t use human tissue. It’s the best way to go. You get human end point data, you don’t have to worry about saying, well, this happened in the rat, this might happen in man. So you can do it. It’s just a matter of having an application, a tissue supply & of course funding from agencies such as the NC3Rs.

There are people using neuronal tissue in tissue engineering,

They (in silico modellers) are basically taking data from animal experiments & seeing if they can run them on computer models & get the same results. So I’d like them (in silico modellers) to start working on modelling the results that we get from our tissue models & then see if they marry up. I think that will bring us more advance than using rodents. I’m talking about lung biology here & they use the rats traditionally & the data never matches up.

2008

The animal models are pretty useless, to be honest. Dr Clive Svendsen, a cell biologist from the Waisman Center at the University of Wisconsin, Madison, speaking about models for Spinal Muscular Atrophy in Nature, 22^nd December

The acid test is what happens in the human. Professor Ian Chopra, a biofilm specialist from Leeds University http://news.bbc.co.uk/go/em/fr/-/1/hi/health/7599818.stm

It's slow. It's expensive. We are not rats and we are not even other primates … After all, ultimately what you are looking for is, does this compound do damage to cells? Can we, instead of looking at a whole animal, look at cells from different organs?
Dr. Francis Collins, Director, National Human Genome Research Institute, speaking to reporters at an American Association for the Advancement of Science conference.

Traditional animal testing is expensive, time-consuming, uses a lot of animals and from a scientific perspective the results do not necessarily translate to humans. It’s a bold, ambitious thing to try to do but our goal is to eliminate animal use in toxicology in ten years. Dr Christopher Austin, Director, National Institutes of Health Chemical Genomics Center, speaking to reporters at an American Association for the Advancement of Science conference.

(Supporters of animal research are) also too unwilling to admit the inadequacies of some aspects of animal research - the benefits are simplistically exaggerated in many cases. Prof Colin Blakemore, quoted by Zoe Corbyn, Times Higher Education, 28 February.

The reason we use animal tests is because we have a comfort level with the process . . . not because it is the correct process, not because it gives us any real new information we need to make decisions. Animal tests are no longer the gold standard, it is a marvelously new world. Melvin E. Andersen, director of the division of computational systems biology at the Hamner Institutes for Health Sciences, USA. Quoted by Gaul, Washington Post, Saturday, April 12, 2008; Page A01

Some animal tests haven't changed in 60 years. The tests are frozen in time. This is not science. Science is always moving ahead. Thomas Hartung, head of the European Centre for the Validation of Alternative Methods. Quoted by Gaul, Washington Post, Saturday, April 12, 2008; Page A01

With more than 250 rat strains and 330 mouse strains to choose from, it is not surprising that the health authorities have been confused for the past 25 years and unbelievably slow to recognise the dangers to human health of BPA (bisphenol A, a component of many plastic products). That is why we used human cells in our analysis. Dr Claude Reiss, president of Antidote Europe (http://www.antidote-europe.org/)

The modern tools of molecular biology now permit the human to be the best model organism to study biology. Dr Arnold J. Levine, Simons Center for Systems Biology, Institute for Advanced Study, Princeton, USA. Journal of Virology, 1^st October.

One of the greatest disappointments for the pharmaceutical industry is having experimental compounds that show early promise in animals that later fail miserably when tested in humans. In building our initial model of a 'virtual mouse', scientific teams at Entelos discovered subtle but important differences between species that can have a big impact on interpreting results of animal studies and on designing clinical trials. Our models and predictive platforms help to better translate results and design experiments to make animal studies more meaningful and relevant for finding effective medicines for patients. We plan to increase access to Entelos' powerful simulation technology to enhance researchers' capabilities to rapidly test 'virtual mice' and 'virtual patients' and ultimately use fewer animals, run more decisive clinical trials, and save time and money. James Karis, president and CEO of Entelos http://www.reuters.com/article/pressRelease/idUS56996+08-Apr-2008+BW20080408

The following quotes are all taken from: Prof. Kenneth R. Chien of Massachusetts General Hospital and Harvard University Medical School, Regenerative medicine and human models of human disease, Nature, Vol. 453, 15th May 2008:

- Clearly, a better understanding of disease mechanisms is required for regenerative medicine to succeed. But most of what is known about the molecular pathways that lead to human diseases has come from studying animal models, particularly genetically engineered mouse models, which do not necessarily mimic human physiology or precisely recapitulate human disease. The cardiovascular systems of mice and humans, for example, diverge with respect to serum lipid profiles, the way that abnormalities in heart rhythm develop and the phenotype that results from certain single-gene disorders.

- These technological advances [using human stem cells] provide the opportunity to set up human models of human disease, which should markedly improve the understanding of human disease mechanisms.

- Progress in stem cell technology is beginning to offer the unprecedented possibility of using humans as model organisms.

- In terms of disease, these systems [using human stem cells] could allow the study, for example, of the complex common genetic variants that are now being uncovered by genomewide association studies of patients with specific forms of heart disease, variants that would be extremely difficult to mimic in mice.

- This convergence of human genetics and human stem-cell models could be especially powerful for studying rare chromosomal disorders, such as Down’s syndrome, that are usually associated with numerous physiological changes across multiple organ systems. The chromosomal translocations, deletions and duplications seen in many of these genetic disorders in which several organs are affected have been difficult, if not impossible, to study in model organisms.

- Perhaps we are about to enter a new age of enlightenment in regenerative medicine, with human models of human disease leading the transition from bedside to bench.

The following quotes are all taken from: Heidi Ledford, Nature 452, 510-511 (1st April 2008) (On the use of antiarrhythmic drugs to prevent heart attacks (which actually doubled the risk)):

- Drugs play a prominent part in the management of cardiorespiratory arrest, although rigid clinical evidence to demonstrate their value is sparse. Their use depends principally on tradition, on an unproven expectation that antiarrhythmic effects are likely to be beneficial for potentially lethal arrhythmias as well as for less malignant conditions, and on extrapolation from animal experiments. Chamberlain, Heart, 80: 408-411 (1998).

- It is not easy to think of a greater medical error, since the practice of therapeutic bleeding, than the use of antiarrhythmic drugs in patients after myocardial infarction. Robert Temple, director of the Office of Medical Policy at the US FDA, Clinical Measurement in Drug Evaluation (eds Nimmo, W. S. & Tucker, G. T.) (publ. Wiley) (1995)- cited in the above Nature article.

2007

For quotes from the US National Research Council Report for the US Environmental Protection Agency published this year, see http://curedisease.net/reports/toxicitytesting.shtml .

There is currently no good way of predicting whether a compound is toxic in humans, "Testing in animals is never going to be able to predict all human toxicity. Steven Tannenbaum, professor of toxicology and chemistry at Massachusetts Institute of Technology. Quoted on 29th November in an article by Katherine Bourzac writing in the Technology Review.

We know animal models do not translate to human responses...Animal models of treatments for HIV to psoriasis and flu are not representative of human responses. William Warren, president, CEO and co-founder of VaxDesign, a company that has developed an in vitro human immune system, quoted by Allan Maurer in: VaxDesign developing a clinical trial in a test tube, 2nd October, TechJournal South.

The safest thing to say is that about 70 percent of drug failures occur due to toxicity...There's always a question you have to ask, and that is whether testing on an animal is predictive of how a human will react. Prof Jonathan Dordick, co-founder of Solidus (a company that has helped to develop a new Biochip that uses 3D human tissue cultures to assess the affects of chemicals), quoted by Larry Greenemeier, Scientific American, 21st December 2007.

Today, we can be confident that an approved drug will be effective for its labeled uses. The same cannot be said of the drug’s safety. William B. Schultz, New England Journal of Medicine, 357; 22, 2217-2219.

There is an increased interest & effort from researchers in developing new alternative methods by using human material, since these models will improve the relevance of the results produced. Bremer & colleagues, Alternatives to Laboratory Animals, 35, 421-439.

We believe that experimental findings in animals cannot be extrapolated to
observations in human beings… Leung, Bjarnason and Chan, the Lancet, 369:1691.

According to experts, the liver appears to be the most common target organ for toxicity during the course of drug development, and animal studies appear to be limited at best in predicting human hepatotoxicity. James Karis, CEO of Entelos, quoted in: Entelos to develop liver injury model for the FDA, DrugResearcher.com, 6^th August.

In conclusion therefore while some surrogate in vivo models may inform on the mechanisms of human as well as animal disease many others are potentially a menace and may actually slow our progress. Janusz A Jankowski, Professor in GI Oncology & Lea-Anne Harrison, Department of Clinical Pharmacology, University of Oxford, Radcliffe Infirmary, rapid response to: Comparison of treatment effects between animal experiments and clinical trials: systematic review, British Medical Journal, 334: 197.

It is common (some might say 'fashionable') to conduct studies in KO [Knock Out: where a gene has been turned off] mice, but do such results reveal important aspects of normal physiology and pharmacology or instead, are the findings primarily a description as to how a particular mouse strain compensates (as in the current study) for a total body knockout of a protein essential for caveolae formation and that is absent from embryonic life onward? Given such extensive losses of caveolin-1 and caveolae and the length of time of these losses, are results in these knockout mice physiologically relevant or only pharmacological curiosities? Insel & Patel, British Journal of Pharmacology, 150, 251–254.

Medical science has long been concerned about adverse occurrences in human trials that aren't predicted in traditional animal studies. Gary Dreher, CEO of AMDL. www.drugresearcher.com , 21^st May.

Since the natural mouse immune system differs from that in humans, experiments in mice are not always predictive of how therapeutic compounds will work in humans - both in terms of efficacy and side-effects. Gary Dreher, CEO of AMDL. www.drugresearcher.com , 21^st May.

It is strange that pharma companies accept to make decisions based on animal data, where the correlation with therapeutic data is highly variable - correlation between PK in rats and PK in humans is between 60 and 70 per cent - but find it difficult to accept data based on human testing. It clearly shows the conservative nature of the industry. Professor Colin Garner, CEO Xceleron Ltd. www.drugresearcher.com, 5^th June.

“Professor John Warner, the head of the department of paediatrics at Imperial College, London, said that clinical research should be carried out on children because doctors were currently obliged to scale down treatments designed for adults… he said that about 40 per cent of medicines prescribed to children had never been tested on
them… Professor Warner said that therapies should be designed specifically for children as they have different metabolisms, immature organs and diseases that can behave unexpectedly. He said: ‘We have a desperate need to understand precisely how children’s bodies work so that we can custom-design therapies for them.’” The Times, 9 July.

Unfortunately, novelty does not easily penetrate into the pharmaceutical environment and, in particular, the very conservative world of safety sciences, which all too readily responds with the often heard boring refrain of 'the regulators won't accept these new approaches'! The safety science community should be reminded that regulators expect them to take the initiative to improve their sometimes catastrophic record in predicting clinical drug safety as a result of their investigational efforts. Dr Icilio Cavero, reporting on a Safety Pharmacology Society workshop held at Pfizer, Kent, 13th February 2007. Expert Opinion Drug Safety, 6(4):465-471.

The following quotes are all taken from: Rowell, Journal of Applied Physiology, 102: 837–840.

Understanding of human physiology indeed rests in part on what we have learned from other species, but a great body of knowledge comes from studies performed on human subjects that entailed measurements as accurate and quantitative as those possible in work with animals.

Extrapolation as a means of learning about one species from another has its limits (and dangers)—we would have little luck in understanding the problems for cardiovascular regulation in the giraffe from studies on dogs. No amount of extrapolation would have revealed particular features of human physiology that set this species apart.

Two major disputes concerning Starling's Law of the Heart and active redistribution of blood flow pitted thinking of human physiologists and cardiologists against that of many (not all) animal physiologists. But the findings of both groups were valid: extrapolation of findings from dogs to humans caused major misunderstandings.

It suffices to say here that temperature regulation in humans during heat stress at rest or during exercise is so fundamentally different from that of other species that extrapolation from animals has no meaning.

2006

Animals are very, very poor predictors of human metabolism. These drugs are not being developed for the veterinary market. They're been developed for us. That's why it's (microdosing) hot and promising. Eventually, all first in human tests are going to be in microdoses. Vitalea co-founder & nutritionist, Dr Stephen Dueker, quoted in: Bio-IT World. 15th February.

Relative lack of severe toxicity in animal models should never be construed as a guarantee of safety in man, as the story of thalidomide taught us. Prof Goodyear, British Medical Journal, 332:677-678.

It has been estimated that ~ 50% of the new drugs that produce hepatotoxicity in human clinical studies do not demonstrate any concordance with animal toxicity studies. Caldwell and Yan, Screening for reactive intermediates and toxicity assessment in drug discovery, Current Opinion in Drug Discovery and Development, 9(1):47-60, referring to Olson and colleagues, Concordance of the toxicity of pharmaceuticals in humans and in animals. Regulatory Toxicology and Pharmacology, (2000) 32(1):56-67).

Animal studies suggest that early visual deprivation can cause permanent functional blindness. However, few human data on this issue exist. Given enough time for recovery, can a person gain visual skills after several years of congenital blindness? In India, we recently had an unusual opportunity to work with an individual whose case history sheds light on this question…These results suggest that the human brain retains an impressive capacity for visual learning well into late childhood. They have implications for current conceptions of cortical plasticity and provide an argument for treating congenital blindness even in older children. Ostrovsky and colleagues, Psychological Science, December 2006.

In many cases, we found that, other than having aberrations in insulin-signalling and glucose levels, there was no similarity between the animal model and the human disease condition…If you're developing a drug in that animal model, it's clearly not going to work in humans because they have a different disease. Dr Elliston, CEO, Genstruct. Chemical & Engineering News, Volume 84, Number 31, pp. 17-26.

Conventional animal studies have value, because they are required by law. Professor Johannes Doehmer, CEO GenPharmTox, FRAME lecture, London, November 2006.

Our major finding is that human pancreatic islets have a unique architecture, and work differently than rodent islets. We can no longer rely on studies in mice and rats. It is now imperative that we focus on human islets. At the end of the day, it is the only way to understand how they function. Proceedings of the National Academy of Sciences, 14th February 2006; 103(7): 2334-2339.

Poor replication of even high-quality animal studies should be expected by those who conduct clinical research. Drs Hackam & Redelmeier, Journal of the American Medical Association, October 11, 2006, Vol 296, No. 14 1731-1732.

There's no scientific basis for it. Sometimes it works and sometimes it doesn't. (Commenting on the FDA's historic requirement to test drugs on animals before humans) Dr Carl Peck, director of the University of California, San Francisco's Center for Drug Development Science and former head of the FDA's Center for Drug Evaluation and Research, San Francisco Business Times, 20^th January.

Many resources are invested in, and thus wasted on, [animal studies of] candidate products that subsequently are found to have unacceptable profiles when evaluated in humans. US FDA guidance document, Guidance for Industry, Investigators, and Reviewers, Exploratory IND Studies, January.

The following quotes are all taken from: Dr. Bob Coleman, Asterand Plc. Next Gen Pharmaceutical, www.ngpharma.com/pastissue/article.asp?art=268318&issue=170.

- PhaseZEROTM [Asterand's human tissue tests] was established to address what was perceived to be a serious problem for the pharmaceutical industry, namely its staggeringly poor performance in translating good ideas into safe effective medicines. This is no reflection on the quality of the scientists involved, but of the tools available to identify and validate new therapeutic approaches with any certainty. Although the target for such new medicines is man, the methods employed to identify and validate have historically been almost entirely non-human, and their predictivity for man consequently unreliable.

- In many cases however, it may prove that for establishing proof of concept for efficacy, there is no useful animal model. A good example is cystic fibrosis, a disease that in its severest form is invariably fatal... Cystic fibrosis appears to be a uniquely human disorder, and attempts to develop a useful animal model have failed. Indeed, even [cystic fibrosis receptor] knock-out mice fail to display the broncho-pulmonary symptoms that are characteristic of human cystic fibrosis.

- It would be a mistake to consider that only [animal] data have any real relevance in clinical go/no go decisions, and in cases such as cystic fibrosis, where there is little point in utilising animal models, efforts would be better directed towards [Asterand's human tissue tests] studies on human lung epithelial cells in vitro. The results of such studies would provide a much sounder basis for the decision as to whether or not a clinical study is warranted.

- To illustrate, I will use the mouse, for which there has been considerable enthusiasm as a human surrogate, based ostensibly on the much vaunted genetic similarity between the two species. In truth, this enthusiasm is as much a function of the fact that mice are cheap and easy to use, and that there is a vast range of established models available to the experimenter.

- An enthusiasm for established, available animal models in the face of evidence of their unsuitability for purpose is commonplace.

- Early [Asterand's human tissue tests] profiling of troglitazone [a diabetes drug that was withdrawn due to liver toxicity] would have identified a high risk of liver toxicity, and provided a means of identifying a safer compound of comparable efficacy.

This article is timely as it highlights the unsatisfactory surrogate of many animal model systems for human disease…while some surrogate in vivo [animal] models may inform on the mechanisms of human as well as animal disease many others are potentially a menace and may actually slow our progress. Janusz A Jankowski, Professor in GI Oncology & Lea-Anne Harrison, Department of Clinical Pharmacology, University of Oxford, Radcliffe Infirmary, rapid response to: Comparison of treatment effects between animal experiments and clinical trials: systematic review, British Medical Journal, 334: 197.

"In the animal models, the drug worked beautifully. We had all the evidence of efficacy, the drug was hitting the receptor, but we went into humans and the drug failed miserably. . . Dr Dixit, senior director of toxicology, drug evaluation at Johnson & Johnson Pharmaceutical Research and Development, Drug Discovery and Development, 7th April.

It is undoubtedly the case that all animal models are limited in their predictability for humans. Weatherall, The use of non-human primates in research., Academy of Medical Science.

2005

It has also become clear that available animal models of human disease are often inadequate, necessitating even more research on human populations and biologic samples. Dr Elias Zerhouni, director of the US National Institutes of Health, New England Journal of Medicine, Volume 353:1621-1623.

I don’t think we can continue to do what we’re doing now- pouring money into old ways of doing things when it comes to developing drugs, ways that are costly and not as efficient and effective as we can be. In too many places, discoveries are made using advanced sciences, but when they are brought up to the preclinical stage we start applying technologies that have been used for 50 or 60 years, such as animal toxicology. This is a completely empirical approach, even though we have technologies such as predictive assays that in selected cases can be more predictive and less costly to acquire than animal toxicology data. Scott Gottlieb, MD, Deputy Commissioner for Medical and Scientific Affairs, US FDA. Speech at PhRMA Scientific Regualtory Meeting, 29^th November 2005.

Instead, in this and other areas of drug development, we should be adopting approaches that increase the use of mechanistic data in preclinical and clinical research to actually intelligently target new medicines to patients who are likely to experience more of the benefits and fewer of the side effects of a new drug. Scott Gottlieb, MD, Deputy Commissioner for Medical and Scientific Affairs, US FDA. Speech at PhRMA Scientific Regualtory Meeting, 29^th November 2005.

In short, drug development has to be moved from empirical process to a more scientific process, just as the discovery and early development phase of drug creation has moved to be less trial-and-error and more deliberate and scientific. Scott Gottlieb, MD, Deputy Commissioner for Medical and Scientific Affairs, US FDA. Speech at PhRMA Scientific Regualtory Meeting, 29^th November 2005.

Most medical researchers rarely, if ever, see patients. Most who argue for the necessity of the pyramid of discovery, with biochemistry and genetics at the base and clinical research at the apex, work at the base themselves [2]. However, many of the major discoveries that have had a direct impact on clinical practice arose from clinical disciplines rather than from generic biomedical approaches— consider hip replacements, cataract surgery, the importance of Helicobacter pylori, phototherapy, in vitro fertilisation, and minimally invasive surgery [4]. In the biomedical model, these successes are brushed aside as being of historical interest only. From finding genes to gene therapy and stem cell therapy, both the public and research community itself are fed a biased view of medical advances. Cancer cured (in mice) again. Jonathan Rees, Public Library of Science Medicine, April, Volume 2, Issue 4, e111.

The following quotes are all taken from: Keith Matthews & colleagues, Animal models of depression : navigating through the clinical fog, Neuroscience and biobehavioral reviews, Animals models of depression and antidepressant activity, vol. 29, n^o 4-5 (1 p.1/4), pp. 503-513.

- Modelling human mental disorders in experimental animals is fraught with difficulties. Depression models generally lack both clinical and scientific credibility and have, thus far, failed to inform treatment strategies previously acquired through serendipity.

- In light of these findings, we therefore pose two questions: if depression is infrequently (or even rarely) a consequence of sustained hypercortisolism, what is the biology of the relationship between psychosocial adversity and depression? Also, what is the validity and relevance of the wealth of studies modelling such processes in animals where the focus is almost exclusively on corticosteroid function and regulation?

- It is, therefore, an article of faith that the neural substrates underpinning certain behavioural and cognitive phenomena in experimental animals are the same as in humans.

- Many of the core features of depression may be ‘unmodellable’ in experimental animals.

- As one enters the ‘real world’, it becomes apparent that it is almost impossible to control not only the variables manipulated in the lab, but also the variables that cannot be modelled in animals—cognition, emotion, social behaviour, relationships, etc.

- Brain imaging consistently draws our attention to subregions of the prefrontal cortex as key structures whose altered function is associated with depression. How do we model such function in the rat? Although the rat may have a prefrontal cortex, the homology is highly controversial.

- There seems little prospect of using animals to model those features, which rely on a verbal description of a subjective experience (for example, suicidal ideation, guilt, lack of self-confidence, and low self-esteem).

2004

In many cases, developers have no choice but to use the tools and concepts of the last century to assess this century’s candidates. As a result, the vast majority of investigational products that enter clinical trials fail. Often, product development programs must be abandoned after extensive investment of time and resources…Often, developers are forced to use the tools of the last century to evaluate this century’s advances. US FDA, Innovation or Stagnation: challenge and opportunity on the critical path to new medical products.

The contribution of animal studies to clinical medicine requires urgent formal evaluation. Pound et al, Where is the evidence that animal research benefits humans? British Medical Journal, 328:514-517.

2003

Does the use of animal models of disease take us any closer to understanding human disease? With rare exceptions, the answer to this is likely to be negative. David F Horrobin, Nature Reviews Drug Discovery 2, 151 - 154.

Animal models - represent nothing more than an extraordinary, and in most cases irrational, leap of faith. David F Horrobin, Nature Reviews Drug Discovery 2, 151 - 154 (01 Feb 2003)

One must conclude that most predictions of near term human benefit are not only overblown but are actually fraudulent. David F Horrobin, Nature Reviews Drug Discovery 2, 151 - 154.

If we devoted as much effort to the human disease as we do to unvalidated [animal] models, then we might be much further forward in understanding. David F Horrobin, Nature Reviews Drug Discovery 2, 151 - 154.

Perhaps the single greatest limitation of modern toxicological practice has been the uncertainty of quantitative extrapolation from laboratory models to the human. James T. MacGregor, The Future of Regulatory Toxicology: Impact of the Biotechnology Revolution, Toxicological Sciences 75, 236-248.

Professor Sir Michael Rawlins, the chairman of NICE, told representatives from the pharmaceutical industry, regulators, academics, and patient groups that the animal study regime, which could take up to six years, was "utterly futile.” Roger Highfield, Science Editor, the Telegraph, 17^th September.

A major contributor to the rate of attrition is the failure of preclinical [animal] models to predict these behaviours in human subjects. The villain in this story is the inherent lack of predictability of our available models for complex biological processes. Geoffrey Duyk, chief scientific officer of Exelixis Inc, writing in Science, 302:603-605.

As Dr. Francesco M. Marincola, Editor-in-Chief of the Journal of Translational Medicine, points out, "These models do not represent the basic essence of human diseases. Prestigious journals, however, appear more fascinated with the modern mythology of transgenic and knock-out mice than the humble reality of human disease." Other authors in the journal explain; "The pathology of humans, in contrast to that of inbred laboratory animals faces the challenge of diversity addressed in genetic terms as polymorphism. Thus, unsurprisingly, treatment modalities that successfully can be applied to carefully-selected pre-clinical models only sporadically succeed in the clinical arena. Indeed, pre-fabricated experimental models purposefully avoid the basic essence of human pathology: the uncontrollable complexity of disease heterogeneity and the intrinsic diversity of human beings." Jin and Wang, Journal of Translational Medicine, 1:8.

As immunologist Ralph Steinman of Rockefeller University, New York, observed astutely; "Patients have been too patient with basic research. Most of our best people work in lab animals, not people. But this has not resulted in cures or even significantly helped most patients." Sharon Begley, Wall Street Journal, 25^th April.

The commission (Royal Commission on Environmental Pollution) recommended substituting hazardous chemicals with safer, "greener" alternatives. Drivers for this could include the introduction of a banded charge for the use of chemicals of concern, greater product liability, and faster "smarter" methods of screening and assessing the risk posed by chemicals. Slow and inefficient toxicity tests in animals should be abandoned. Chemicals should be assessed using new computational technologies, already widely used by the pharmaceutical industry. This would enable 90% of the 30000 chemicals to be screened within three years (rather than three decades) and would avoid tests on up to 12 million animals. Smallwood & Richards, British Medical Journal, 327:10 (5 July).

It is a remarkable fact that at least 14 new drugs have been developed during the past 10 years as potential neuroprotectants in stroke, all showing excellent neuroprotective properties in laboratory animals, & all failing to show efficacy in Phase III clinical trials. Stephen Curry, Why have so many drugs with stellar results in laboratory stroke models failed in clinical trials? A theory based on allometric relationships, Annals of the New York Academy of Sciences, 993:69-74.

It appears, in fact, to be quite easy to show neuroprotection in animal models, even with compounds whose physical properties do not support belief in their claimed mode of action, or in passage across the blood-brain barrier. Patients are clearly another matter. Stephen Curry, Why have so many drugs with stellar results in laboratory stroke models failed in clinical trials? A theory based on allometric relationships, Annals of the New York Academy of Sciences, 993:69-74.

Chlormethiazole is a very interesting case in point. This compound has a mechanism of action that forecasts neuroprotective properties, it is highly active as a neuroprotectant in gerbils, rats, & marmosets, & it has a sufficient margin of safety to permit dosing at levels likely to induce the desired concentrations of the drug in the brain, albeit with the necessity of clinical monitoring for excessive sedation...A 2nd trial was therefore organised..., the report of which was published this year, failed to show any useful effect. Stephen Curry, Why have so many drugs with stellar results in laboratory stroke models failed in clinical trials? A theory based on allometric relationships, Annals of the New York Academy of Sciences, 993:69-74.

Assuming that the human stroke victim has the potential to benefit from neuroprotection, the issue appears to be 1 of extrapolation from animals to humans. This is no easy task, as issues of anatomy, biochemistry, pathology, treatment receptivity, pharmacokinetics, to name but a few, all need to be taken into account. Stephen Curry, Why have so many drugs with stellar results in laboratory stroke models failed in clinical trials? A theory based on allometric relationships, Annals of the New York Academy of Sciences, 993:69-74.

2002

The Toxicology Working Group at the conference held by the House of Lords Select Committee on Animals in Scientific Procedures concluded that "the effectiveness and reliability of animal tests is unproven.” It recommended that “the reliability and relevance of all existing animal tests should be reviewed as a matter of urgency." It commented that "the scientific basis for the use of two species was questionable; tests could be conducted in any number of species and the relevance of the findings for man would be equally uncertain for all the species used…the formulaic use of two species in safety testing is not a scientifically justifiable practice, but rather an acknowledgement of the problem of species differences in extrapolating the results of animal tests to predict effects in humans…. some animal testing was done for 'administrative', rather than scientific reasons. House of Lords Select Committee on Animals in Scientific Procedures Report, Volume 1 The Stationery Office Ltd. HL Paper 150-1 p70-72.

There isn’t a single genetically manipulated mouse that has been used yet to produce a drug that cures a disease,’ says [Kathleen] Murray of Charles River Laboratories. the Scientist, February 4, 2002 p22.

One of the major challenges facing the drug discovery community is the limitation and poor predictability of animal-based strategies. Over the last decade, drug discovery has largely been based on finding targets in animal models and then identifying the human homologue...many drugs have failed in later stages of development because the animal data were poor predictors of efficacy in the human subject… One of the overriding interests of the pharmaceutical and biotechnologies industry is to…create alternative development strategies that are less reliant on poor animal predictor models of human disease…Although the species [chimpanzees] share more than 98.9% gene identity [with humans], the expression of genes in the brain was more than five-fold greater in humans than in the chimpanzees....Differences from mice were even greater. These differences reinforce the importance of using human disease models in drug discovery as a real predictor of human efficacy. Drs Palfreyman, Charles and Blander, The importance of using human-based models in gene and drug discovery, Drug Discovery World, Autumn 2002, p.33-40

''Our needs move in tandem with the industry,'' said Jim Foster, the company's chief executive. ''Companies need to get drugs to market faster. To do that, they need more nonanimal testing technologies. I don't want to sit here and say, `Hey, there goes our animal business.'' In the past five years, the lab animal portion of Charles River's business has gone from 80 percent to 40 percent. Earlier this month, the firm bought a lab test, called DakDak, that allows researchers to measure how effectively sunscreens prevent skin damage. The test does in days what would take months in animal studies…. Studying them all in animals is simply an economic impossibility. Animal tests can take months, even years, and quickly run into the hundreds of thousands of dollars. Charles River estimates that DakDak can test five or six products for less than half what it would cost to study one product in animals…. ''It is driven by pure necessity and economics,'' said Melvin Balk, a veterinarian by training and president of the nonprofit foundation associated with Charles River Laboratories. Boston Globe, 27^th February 2002.

I agree with Schechter and Rettig that most basic research does not find an application in clinical medicine; that is exactly why we have to do so much of it. David G. Nathan, MD, Journal of the American Medical Aassociation, letters, Vol. 288 No. 7, August 21, 2002.

It generally has been assumed that advances in basic research propel applied or clinical accomplishments through the intermediate stage of translational research. This conceptual model implicitly assumes a unidirectional flow of innovation and is used to justify ongoing expansion of basic and translational research, sometimes at the expense of patient-oriented research, without real criteria to judge its clinical importance or relevance.2 Even if translational animal research can generate useful information about gene function or biochemical pathways in nonhuman species, such knowledge is often of uncertain clinical relevance. At the same time, however, the sequencing of the human genome should open up enormous opportunities for patient-oriented research, especially by physician-scientists. Schechter & Rettig, letters, Journal of the American Medical Aassociation, Vol. 288 No. 7, August 21, 2002.

Although the biomedical research community may increasingly evaluate itself in terms of dollars spent or researchers trained, recent concerns over the shrinking pipeline of new drugs3 and the still very infrequent genuine clinical (as distinct from scientific) breakthroughs suggest that those may not be the crucial metrics in the goal of improving human health. Schechter & Rettig, letters, Journal of the American Medical Aassociation, Vol. 288 No. 7, August 21, 2002.

2001

There is no way to judge the efficacy of a new approach without testing it on human beings. Dr Jerome Groopman, Chief of Experimental Medicine, Harvard University Medical Centre, 2001.

2000

Irrespective of weak or strong relationship between body weight & absolute bioavailability, unreliable prediction of absolute bioavailability was obtained using the allometric approach. The error between predicted & observed bioavailability ranged from 0% to 265%. The overall results of the study indicate that none of the methods used in this study to predict bioavailability is reliable. Iftekhar Mahmood, Drug Metabolism & Drug Interactions, 2000, vol. 16, pp14-155.

1999

A small fraction of the failed therapies are withdrawn due to commercial considerations of the sponsors, however, the rest of the research drugs are withdrawn due to unfavorable benefit/toxicity ratio or simply because they are not efficacious. This means that the predictive value of animal studies has serious limitations when applied to the human organism. Consequently, even if we use the best scientists and physicians to predict the efficacy of the proposed drug our knowledge is very limited and the value of prediction is probably hardly greater than a simple gut-feeling. Dr lmre Szebik, clinical trials research group, McGill University, Canada; letter to the FDA, December 1999.

Whether diseased human arteries are subject to the same deep stent-induced injury as animals is debatable… it is difficult if not impossible to extrapolate directly from animal data to human responsiveness…It is not possible in an animal either to reproduce the multifactorial processes that enable complex human lesions to evolve over decades or to mimic precisely the acute, high-intensity injuries of vascular interventions. Elazer R. Edelman, MD, PhD; Campbell Rogers, MD, Circulation. 1999;100:896-898.

1998

The only universal model for a human - that is, one which would best predict what would happen at a given endpoint across the full range of chemical structures, concentrations, etc. - is other humans. Gad and Chengelis, Acute Toxicity Testing, (publ. Academic Press) p4.

Identifying improved analogs of nicotinic acid without common side effects has been hampered in part by the lack of suitable animal models. Krause, Atherosclerosis, Vol. 140, (1) p15-24.

The metabolism of low density lipoproteins in hamsters is widely thought to be similar to that in humans, yet neither statins or fibrates lower plasma lipids in these species. Krause, Atherosclerosis, Vol. 140, (1) p15-24.

There is no one perfect animal model that commonly replicates the stages of human atherosclerosis. T. M. A. Bocan, Parke-Davis Pharmaceuticals Current Pharmaceutical Design, 4:37-52.

1997

No one had initially thought to examine patients' hearts because animal studies had never revealed heart abnormalities and heart valve defects are not normally associated with drug use. Dr. Michael A. Friedman, the acting commissioner of the F.D.A., quoted by Gina Kolata, The New York Times, 16^th September.

1995

The occurrence of major qualitative and quantitative differences in the metabolism of drugs between species is probably the single greatest complicating factor in the use of animals as predictors of drug toxicity and kinetics in patients. Dr J Paxton, Clinical & Experimental Pharmacology & Physiology, vol 22, p 851-854, 1995.

1994

It is impossible to give reliable general rules for the validity of extrapolation from one species to another. [This] can often only be verified after the first trials in the target species [humans]. Extrapolation from animal models. . . will always remain a matter of hindsight. The Handbook of Laboratory Animal Science Vol. II, p6.

Because of the increasingly observed species differences in metabolism, in pharmacokinetics and in mechanisms of toxicity, a variety of animals had to be exposed, including rodents, rabbits, dogs, monkeys and baboons, to warrant that no aspect of toxicity was missed... Yet this extensive screening in various experimental animals for potential toxicity in humans still threw up numerous false negatives, such as the drugs perhexiline, benoxaprofen and tienilic acid, which subsequently caused human death s...So, there are indeed more appropriate alternatives to experimental animal studies and, for the safety evaluation of new drugs, these comprise short-term in vitro tests with micro-organisms, cells and tissues, followed by sophisticated pharmacokinetic studies in human volunteers and patients. Dr Dennis Parke, Alternatives To Laboratory Animals, vol 22, p 207-209.

My former teacher, Sir Alexander Fleming, in his late years, chided me, saying ‘How fortunate we didn't have these animal tests in the 1940s, for penicillin would probably never been granted a license, and possibly the whole field of antibiotics might never have been realized.’ Professor Dennis Parke as quoted in Alternatives To Laboratory Animals, 22:207-209.

1993

There are substantial species differences in the toxicity of many chemicals. Such variability reflects differences in metabolism e.g. the nature and kinetics of metabolism; and the disposition and elimination of relevant metabolic products, or in tissue, organ or cellular susceptibility. Kai Savolainen, Head of Department of Toxicology of the National Public Health Institute in Finland, and Secretary-General of The International Union of Toxicology, IUTOX Newsletter, p15, 1992/1993.

1992

Today the subject and practice of toxicology has become exalted to the eminence and influence of a religion. It is, moreover, an established form of worship, actively supported by the State. It has its creeds and its commandments, and its hierarchy of high-priests, worshippers, adherents and novitiates. Again, like a religion, it relies rather more on faith than reason. Botham and Purchase in New Scientist 2nd May.

One fundamental deficiency of animal tissues is that they contain animal receptors - a boon in the development of drugs for rats, cats, and dogs but of dubious value in human healthcare. John Hodgson, Bio/Technology, vol 10, p 974.

Most animal research into asthma is of "overwhelming unimportance". Professor Andrew Wardlaw, Director, Institute for Lung Health. British Medical Journal, 18^th July, p191.

1991

Medicine is, essentially, a science of observation, in which experimentation is only a minor part of the medical investigation. But the worst of it is that this minor part has been contaminated by a gross error – that of having taken animals as models for humans. Prof Pietro Croce, Speech delivered at the International Scientific Congress, London, April.

1990

There is no doubt that the best test species for Man is Man. This is based on the fact that it is not possible to directly extrapolate animal data to Man, due to inter species variation in anatomy, physiology and biochemistry. Drs MacLennan and Amos of Clinical Science Research Ltd, UK, in Cosmetics and Toiletries, Manufacturers and Suppliers, vol XVII, p24.

Another study revealed that in only 4 of 24 cases did animals predict human toxicity, and in yet another, 6 out of 114 times. Spriet-Pourra, and Auriche (Eds) 1994 SCRIP Reports (publ. PJB), and Animal Toxicity Studies: Their Relevance for Man p57-67 (publ. Quay).

1989

The following quotes are all taken from a book chapter written by Ralph Heywood, former Director of contract research company Huntingdon Life Sciences. Animal Toxicity Studies: Their Relevance for Man, Chapter 7, Clinical Toxicity- could it have been predicted? Post-marketing experience.

- Many adverse reactions in man, in particular immunotoxicity, allergy, hypersensitivity & effects on bone marrow, are unpredictable in animal models.

- There is little evidence that toxicological studies have been performed with the route, dose & frequency of administration selected with due regard to the dynamics of action & target receptor sites, & with a kinetic profile that is relevant to the in-use situation. Therefore most toxicological data cannot be interpreted.

- The published data base is inadequate to make proper judgements, & the best guess for the correlation of adverse reactions in man & animal toxicity data is somewhere between 5% & 25%.

- Some analyses have attempted to assess the value of chronic animal toxicology studies. The correlations between target systems toxicity in the rat & a non-rodent species have been shown to be about 30%; it is unrealistic to expect correlations between target organ toxicity in laboratory animals & adverse reactions in man to be any better, In the absence of more substantial data, one can only conclude that there is no reliable way of predicting what type of toxicity will develop in different species in response to the same compound.

- Toxicology is a science without a scientific underpinning. Animal studies are based on 2 assumptions; that there are appropriate animal models & that a dose-response relationship can be demonstrated, the practice of toxicology requires the administration of high (toxic) doses to animals. What hypothesis is being tested when we give to laboratory animals multiples of human doses on the basis of mg/kg body weight? There is little evidence that toxic effects observed in laboratory animals can be extrapolated through to adverse reactions to man, & “no effect” doses in animals cannot be interpreted as meaning safety.

1987

Many species differences in chemical toxicity, attributable to differences in metabolism, are known. For example, the mouse and hamster are much more susceptible to the hepatotoxic action of paracetamol than is the rat... Similarly, TCDD is much less toxic to the hamster than it is to the rat or guinea pig, and the hamster but not the rat is able to metabolise this 'non-metabolisable' chemical, albeit very slowly. Prof Dennis Parke, The Future of Predictive Safety Evaluation, (publ. MTP Press).

1986

Dioxin is extremely dangerous to some animals, but not to others. The amount of dioxin necessary to kill one hamster, for example, will kill 5000 guinea pigs. This quantity is unlikely to kill a human being. If humans had reacted like guinea pigs, then Seveso would have been the chemical industry's Hiroshima. The fears of a decade ago were not realised. Dr Fred Pearce, New Scientist, p 60, 5^th June.

The problem of drug interactions is not one that is related only to antibiotics; it is a problem with all new drugs that may be prescribed with other marketed drugs. Quite frankly, in this area animal models are of limited value... We could be doing a lot of wasted experiments in animals. Dr Flack, toxicologist, in a discussion reported in the book Prediction and Assessment of Antibiotic Clinical Efficacy (publ. Academic Press).

1985

…any in vitro method using human tissue gives a degree of reassurance not provided…by animal experiments. Alternatives To Laboratory Animals, 1985, vol.13, 38-47.

The following quotes all come from Silvio Garattini, Toxic Effects of Chemicals: Difficulties in extrapolating data from animals to man; Critical Reviews in Toxicology, vol 16, issue 1, p1-29.

- …it is evident that the results in animals depend so much on the selected set of experimental conditions as to make impossible any meaningful extrapolation to man.

- … different animal species dispose of chemicals in different ways… equal concentrations of chemicals (or of their metabolites) do not mean equal effects across animal species, because the sensitivity of organs, cells, enzymes, or receptors may be different in different species… the extrapolation of toxicological data from animals to man cannot be based solely on the execution of a number, however large, of routine tests…

…similar doses may produce different concentrations in different animal species or even in the same animal species under different experimental conditions.

… there is no one animal species that resembles another…

… The formation of metabolites may be quite different in different cells of the same animal species.

- ... kinetics & metabolism of chemicals may be quite different in various animal species & under various experimental conditions even in the same animal species. If this creates problems in the extrapolation of toxicological data from animals to man, even greater difficulties arise from the differences between animal species as regards the biological target. The possibility of chemicals or their metabolites expressing their toxicity depends on their presence in a given tissue but also on the levels & characteristics of the target, which may be an enzyme, a receptor, a transport mechanism etc.

- If it were possible to determine the concentrations of a chemical (& of its active metabolites) in animals & in man in the target tissue for a certain toxicological effect, could it be said that equal concentrations determine equal toxicological effects? Unfortunately the answer is no, because in several cases equal concentrations of chemicals elicit quite different effects in various species, strains or even individuals of the same species.

- There can be no doubt that the process of extrapolating data from animals to man is very complex. So many factors affect kinetics, metabolism, & the tissue constituents necessary for eliciting toxic effects that it becomes difficult to decide even which sets of animal data should be selected for comparison with which subgroups of human subjects.

- It is evident that the results in animals depend so much on the selected set of experimental conditions as to make impossible any meaningful extrapolation to man.

- There is evidence that metabolism is quantitatively and/ or qualitatively different in various animal species.

1984

The following quotes are all taken from Professors Lawrence, McLean and Weatherall, Safety Testing of New Drugs-Laboratory Predictions and Clinical Performance (publ. Academic Press).

- The procedures intended to confer greater safety have become enshrined in the legal requirements of many countries, although their efficacy is questionable. (p1)

- The methods of assessing toxicity in animals are largely empirical & unvalidated. (p2)

- It is urgently necessary to know whether the tests as in fact conducted have sufficient predictive value to be justifiable, or whether they are a colossal waste of resources to no good purpose. (p2)

- The problems of predicting toxicity in humans from experiments in animals have been known for a long time & remain unresolved. (p164)

- Differences in tolerance between species, & between strains within species, differences in metabolism between species & between strains, differences in life span, & in duration of dosage can all be argued to make a given result crucial, or negligible, according with the preferred point of view. In such circumstances the comfort given by decisions based on an extended range of [animal] tests is largely an illusion. (p165)

- The experiments are done to conform with rules & with no preconception about their purpose except to see whether anything turns up which perhaps ought to be extrapolated to man. Constant critical review of such tests is therefore necessary, both to see whether the experimental results are reproducible & to judge whether interpretations put on them are reasonable.

- One way of assessing the value of toxicity testing to contemporary standards is by historical comparison…Penicillin was evidently used immediately for life saving purposes, because the quantity originally available was minute (Abrahams, Chain, Fletcher, Florey, Gardner, Heatley & Jennings, 1041, The Lancet, ii, 177-189). It is interesting to note that much testing was done on cell cultures in vitro as a means of assessing safety swiftly & with a minimum of material. When manufacturing was established in USA, further experiments were not so prolonged as to delay clinical availability later than 1944. One may wonder whether the present regulatory demands would crumble if a situation were to recur in which such urgency of therapeutic need was unmistakable. (p166)

- If the continued increase in regulatory requirements since the 1950s has served a useful purpose, alarming incidents should now be less common…But the general trend does not suggest to us a notable reduction of adverse reactions since preclinical testing on a large scale became mandatory, & confirms Barnes’s (1963) prediction that “the hazards from new drugs remain as much a problem in the future as it is today”. (full Barnes quote given earlier: Testing procedures outlined by authority whether national or international will either be so vague as to be not worth disseminating or become by necessity more precise. The recommended tests would then be carried out by scores of technicians who will supply a mass of data eventually to be pushed under official noses. The scientific study of toxicology will atrophy & the hazards from new drugs remain as much of a problem in the future as it is today.) (p166)

- Whatever tests are done, it is as true now as then that: “In the last analysis, it is only clinical trials which can reveal whether an effect observed in animals is also relevant to man, & it is only from extensive clinical use of a new compound that valid conclusions can be drawn as to its safety when employed for a specific therapeutic purpose.” Bein, 1963, Proc. Eur. Soc. Study Drug Tox. 2, 57-66.

1983

When new drugs are introduced, however, detailed information on human toxicity is seldom available and existing animal toxicity data may not be relevant to the effects of the drug in humans. Dr Frank Dennison in MIMS Magazine, p 20, 15^th January.

Animal studies frequently fail to predict human toxicity, and clinical trials are generally too small to detect rare but, nevertheless, important reactions. Dr William Inman, University of Southampton Drug Surveillance Unit, quoted in MIMS Magazine, p 10, 15^th January.

Unfortunately, this use of animal models for predicting risks for man is beset with difficulties, it is rarely possible to be sure that the animal model properly represents the relationship in man... Even if it were possible to improve the accuracy of the present-day test procedures so that the risk to the test animals were known with greater precision, this would not necessarily bring about a corresponding improvement in the assessment of potential risk to man because of the uncertain regarding the relevance of the animal data for man. Risk Assessment Report of a Royal Society Study Group, p72-73.

1982

As long as social & legal sanctions are applied if a drug turns out to have an adverse effect, while no penalties are imposed for failure to discover new remedies, it is inevitable that tests will proliferate, if only to show that both developers & regulators did all they could to minimize hazards. Regulatory safety tests consist of experiments “which have neither an observation as starting point, nor are designed to test a scientific hypothesis”. Zbinden, 1982, Naturwissenschaften, 69, 255-259.

Every species has its own metabolic pattern, and no two species are likely to metabolise a drug identically. Small differences in the rate of conversion of drug to inactive, or toxic, metabolite can have large effects on the concentration of active substances at the point of action…differences in diet, gut physiology, rate of passage and liver enzymes raise serious questions about the relevance of findings in rats or mice to man. Compounds that are not absorbed in laboratory animals are not, with minor exceptions, ever tested in man. Nobody knows how many drugs, which would be useful in man, may have been lost in this way. Similarly compounds toxic in laboratory animals at doses near the predicted therapeutic level do not receive trial in man, so it is never revealed whether they would actually have been harmful in man. Thus we lack the evidence of the false positive element in animal toxicology studies, so it is easy to give more weight to such studies than is justifiable.” Dr Miles Weatherall, former director of Establishment, Wellcome Research Laboratories, Nature, April 1, p 387-390.

Whilst the data from the animal studies required by regulatory bodies provide some basic information of the mechanism of toxicity and relative toxicity, it cannot be assumed that this information will be entirely relevant for man. Furthermore, whilst these studies may give indications as to the appropriate treatment for acute overdosage, they are unlikely to indicate the efficacy of treatment. Experience gained from a careful assessment of patients suffering from acute overdosage of drugs is potentially much more useful than that obtained from animal tests. Dr Roy Goulding and Dr Glyn Volans, of the Guy's Hospital Poisons Information Centre, quoted in the Bulletin of the Lord Dowding Fund, no: 10.

1981

The following quotes are from Bernard L. Oser, Journal of Toxicology & Environmental Health, vol. 8 pp521-642.

- …the critical investigator cannot ignore the degree to which qualitative & quantitative differences affect interspecies extrapolation.

- There are no specific ages at which animals of either sex within a given species attain maturity, nor precise ages of physiological equivalence between species. Comparison of the growth rates of various species of mammals, including the rat, reveals that humans are unique in their relatively slow rate of growth to maturity. (referring to Brody, 1945, Bioenergetics and Growth, Reinhold, reprinted 1964 by Hafner)

- When nutrient requirements are compared on a body weight basis, the species differences become apparent. For example, the rat requires many times more fat-soluble & water-soluble vitamins per unit body weight than the human. Omitting the extremes of the rat: human ratios (0.009 versus 46.3 for folacin & cobalamin respectively), they vary over a 6-fold range.

- …it is evident that the degree of similarity to man exhibited by a given species varies from 1 test substance to another.

Difficulties of interpretation are compounded because the species routinely used in toxicological studies are chosen not on consideration of their phylogenetic relationship to man but on practical grounds of cost, breeding rate, litter size, ease of handling, resistance to intercurrent infection, and laboratory tradition. Highly inbred strains of animal are used to achieve greater consistency of response, but this in no way overcomes the fundamental problem of interspecies variation, and there is no assurance that consistency increases the relevance of a response to man. Dr J F Dunne, Textbook of Adverse Drug Reactions, p 37 , ed. D Davies.

Although it cannot be denied that toxicity testing is necessary and that, on occasions, it can identify a potential problem it can have no absolute predictive value for unwanted effects in man. For example it has been impossible to reproduce the oculomucocutaneous syndrome associated with practolol use in any animal species. The only way to find out whether a new drug is safe and efficacious in man is to give it to him. Dr C F George, Clinical Science, vol 60, p 247-250.

First of all, the science of toxicology is terribly imprecise in relation to drug development. Drug testing toxicology seems to have developed haphazardly and many of the guidelines that both the CSM [Committee on Safety of Medicines] and the industry keep to do not appear to have a very firm base. Prof D Grahame-Smith, Department of Clinical Pharmacology, Oxford University, in Risk-Benefit Analysis in Drug Research. ed. Cavalla.

...for the great majority of disease entities, the animal models either do not exist or are really very poor. The chance is of overlooking useful drugs because they do not give a response in the animal models commonly used. Prof D Grahame-Smith, Department of Clinical Pharmacology, Oxford University, in Risk-Benefit Analysis in Drug Research. ed. Cavalla.

The weakness and intellectual poverty of a naive trust in animal tests may be shown in several ways, e.g. the humiliatingly large number of medicines discovered only by serendipitous observation in man (ranging from diuretics to antidepressants), or by astute analysis of deliberate or accidental poisoning, the notorious examples of valuable medicines which have seemingly 'unacceptable' toxicity in animals, eg. griseofulvin producing tumours and furosemide causing hepatic necrosis in mice, the stimulant action of morphine in cats, and such instances of unpredicted toxicity in man such as the production of pulmonary hypertension by Aminorex and SMON. The rapidly increasing interest in clinical pharmacology, and the drive to better means of measurements in man, also reflect the uncertainty of animal experimentation and the realisation that the study of man alone can ever prove entirely valid for other men. Dr Anthony Dayan, Wellcome Research Laboratories, in Risk-Benefit Analysis in Drug Research, p97.

I share Dr Dayan's scepticism [see quote above] about the value of much of the animal toxicology we do now in predicting toxicity in man. Sir Douglas Black, Royal College of Physicians, in Risk-Benefit Analysis in Drug Research, p187.

What I am trying to get at is a situation where we do not automatically accept the traditional toxicology... I think I am saying that the present tests are well known to us but that does not make them good. There may be better tests around, but we have no incentive whatsoever to look for them at the moment. In fact, quite the reverse. [because of the fear that regulatory agencies will require the new tests in addition to the current ones]. Dr R Brimblecombe, Vice-President of Research and Development, Smith, Kline & French Laboratories, in Risk-Benefit Analysis in Drug Research, p153.

Pritchard’s tenacious studies on the hypotensive action of propranolol eventually paved the way for the extensive use of beta blockers in hypertension even though this therapeutic application was not predicted from animal studies.
Recent Developments in Cardiovascular Drugs, Coltart and Jewitt eds. p13 (publ. Churchill Livingstone).

1977

It has to be admitted that the antihypertensive effect of some drugs such as the diuretics, clonidine or the Beta-adrenergic blockers were first observed in man, and only later were studied in animal experiments with respect to their blood pressure-lowering activity. Gross, Anti-hypertensive Agents, p7 (publ. Springer Verlag).

1970

Much of the experimental animal work on atheroma has held back our progress rather than advancing it. Medical News Tribune London, 18^th September.

1964

The idea, as I understand it, is that fundamental truths are revealed in laboratory experimentation on lower animals and are then applied to the problems of the sick patient. Having been myself trained as a physiologist, I feel in a way competent to assess such a claim. It is plain nonsense. Sir George Pickering, Regius Professor of Medicine at the University of Oxford, British Medical Journal, 26^th December.

Animal studies are done for legal reasons and not for scientific reasons. Dr. James Gallagher, Director of Medical Research, Lederle Laboratories, Journal of the American Medical Association, 14 March.

1953

Cawadias (1953) has said that "the history of medicine has shown that, whenever medicine has strayed from clinical observation, the result has been chaos, stagnation, and disaster." British Medical Journal, Oct 8th, 1955, p.867.

1933

My own conviction is that the study of human physiology by way of experiments on animals is the most grotesque and fantastic error ever committed in the whole range of human intellectual activity. Dr G F Walker, Medical World, 8^th December.