Where is the evidence that animal research benefits humans?
BMJ 2004; 328 doi: https://doi.org/10.1136/bmj.328.7438.514 (Published 26 February 2004) Cite this as: BMJ 2004;328:514All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Much animal research is wasted1, because the results are unwelcome.
Beatson demonstrated in 1898 that ovariectomy ameliorated the
clinical course of breast cancer in women.2 In 1919 Loeb confirmed this
observation in a rodent model.3 In 1932 Lacassagne showed a large
continuous dose of oestrogen induced mammary tumours in mice.4
Since then a multiplicity of evidence from animal research has
accumulated proving that exogenous sex hormones - oestrogen, progesterone
and testosterone, cause cancers of the breast, ovary, endometrium, uterine
cervix, testis, pituitary, thyroid, kidney, liver and lymphoid tissue.5
This evidence has been widely disregarded as not being applicable to
humans. The result has been huge increases in preventable cancers caused
by those with hormone-pushing agendas.6
The original animal research predicted more accurately the outcome of
hormone misuse than the results of many expensive epidemiological trials
which have conspired to mislead doctors for decades.
1 Pound P, Ebrahim S, Sandercock P, Braken MB, Roberts I. Reviewing
Animal Trials Systematically (RATS) Group. Where is the evidence that
animal research benefits humans? BMJ 2004;328:514-517 (28 February),
doi:10.1136/bmj.328.7438.514
2 Beatson GT. On the treatment of inoperable cases of carcinoma of
the mamma. Lancet 1898; 2:104-107.
3 Loeb L. Further investigation on the origin of tumors in mice. J
Med Res 1919; 40: 477-479.
4 Lacassagne A. Apparition de cancers de la mamelle chez la souris
male soumise a des injections de folleculine. Cr Acad Sci (Paris) 1932;
195:603-32.
5 Li JJ. Perspectives in hormonal carcinogenesis: animal models to
human disease. Cellular and molecular mechanisms of hormonal
carcinogenesis. Eds J Huff, J Boyd, J Carl Barrett. 1996 Wiley-Liss, Inc
pp 447-454.
6 Grant ECG. Increases in breast cancer incidence
http://bmj.com/cgi/eletters/328/7445/921#55298, 1 Apr 2004
Competing interests:
None declared
Competing interests: No competing interests
Dear Sir
Not in the good name of medical research?
In their article on animal research, Pound et al referred to the
public accepting such research “only on the assumption it benefits
humans.”1
Last year I received a photograph from Naturewatch, which shows a
distressed monkey, immobilised upright in a transparent container, with
two implants in its skull, from which blood has spattered the container
walls. It is one of several images, apparently obtained undercover by
“N”, from a laboratory in Israel, reported in the Tel-Aviv Times2 and
reproduced on the internet.3 (www.aesop-project.org)
The photograph, sent to 60,000 people,3 will have caused distress,
presented medical researchers as cruel and diminished the standing of
medicine as a caring profession. According to the newspaper report: “The
research was trying to investigate which specific brain cells are active
while using the visual memory.” (translation from the Hebrew) Whatever
its purpose, since no human being can be subjected to such experimental
conditions, the results cannot be extrapolated to humans.
Lack of direct clinical applicability is a feature of animal
research: the British Medical Journal hardly ever publishes such work “not
because we are against animal research, but because we favour research
that may have results that are directly applicable to clinicians and those
making public policy.”4 Nevertheless, much more research funding is spent
on basic animal research than clinical research and this may underlie the
failure of research to develop effective preventive strategies in response
to the rising prevalence of chronic diseases such as asthma, dementia,
cancer and, particularly, diabetes.5
Growing public anxiety concerning the realities of primate research
and the deterioration of health, particularly children’s health, should be
addressed before public goodwill for medical research is lost. A radical
shift of resources away from animal research into preventive research is
required.
Yours faithfully
Edward Moore OstJ, MBChB, MFCM
Retired Consultant in Public Health Medicine
References:
1. Pound, P., Edbrahim, S., Sandercook, P., Bracken, M.B. and
Roberts, I. Where is the evidence that animal research benefits humans?
BMJ 2004; 328 : 514-7
2. Lahav, S. Horrific primate experiments exposed in Israel - an eye-
witness account. Tel-Aviv Times, 14 December 2001
3. Tweedy, R. Personal communication 7 March 2005. Naturewatch,
Cheltenham GL 52 6AA
4. Smith, R. Animal research: the need for middle ground. BMJ 2001;
322 : 248-9
5. Dyer, O. First cases of type 2 diabetes found in white UK
teenagers. BMJ 2002; 324 : 506
Competing interests:
None declared
Competing interests: No competing interests
John Mercer appears unable to distinguish between ethical and
scientific arguments against vivisection. The ethical argument is based
on the premise that regardless of whether research on animals or using
animals has benefit, that it is unethical to do so. There have been
hundreds of books and scholarly articles advocating this stance, from the
point of view of utilitatianism, rights theory and virtue ethics. I would
go as far as to suggest that the philosophical case against vivisection,
as well as the use of animals for food and entertainment has largely been
won, though the political battle is far from over.
The scientific argument is that animals cannot be used to model human
disease or reaction to drugs or toxins because it is not possible to
extrapolate from one complex system to another on the cellular and
biochemical level. This argument only holds for experiments where animals
are used as models; it is not valid for agricultural experiments, or for
the use of animals for cell culture or monoclonal antibodies such as
Mercer describes. It also does not hold true for basic research, some of
which has proved useful.
All these uses can however be refuted on ethical grounds. It is also
possible to refute them on the basis that opportunity costs have not been
considered when making cost-benefit analyses. If this were done, it is
likely that spending scarce resources on non-animal alternaitve, including
preventative medicine and public health education, would prove to be a
better investment in the nation's health and wellbeing than expensive
animal experimentation.
Mercer is correct that I stated I had no ethical problems with cell
culture as a non-animal alternative. This is not however because of any
hypocrisy or inconsistency in my ethical stance. It is simply because I
was unaware that cell cultures required the regular killing of fresh mice.
Now that Mercer has made me aware of my error, I oppose the use of mouse
cell cultures on ethical grounds.
Mercer may also be interested in the two references below that make
specific reference to experiments using animals for monoclonal antibody
production.
http://homepages.ihug.co.nz/~nezumi1/possums.html
http://homepages.ihug.co.nz/~nezumi1/ANZCCART.html
Competing interests:
My opposition to animal experimentation is on primarily ethical grounds
Competing interests: No competing interests
The paper by Pound et al. should never have been published in a peer-
reviewed journal. Others have pointed out the methodological problems and
that that the studies cited showed concordance between animal and human
trials. I will amplify on the authors’ use of incredibly deceptive polemic
rhetoric introduced by Dr Blakemore above. The very subject of the paper
is
boldly misrepresented in the title, because the authors never bothered to
look
at anything approaching the breadth of “animal research.” Second, the
authors compound their misrepresentation with additional
misrepresentations and false dichotomies, including “Much animal
research,”
“basic animal research” (never addressed), “clinical (as distinct from
scientific),” and “basic animal research”/”clinical research,” when all of
the
studies cited were clinical applications!
Dr Blakemore pointed out the most dishonest part of the paper: the
fact that
the authors “did not even consider the important research involving
animals”
before arriving at a categorical conclusion. Now, for any lay people
reading
this, stop and take a deep breath before proceeding to the next sentence.
Careful readers will note that Blakemore cited research INVOLVING animals,
not research ON animals, which is but a minority of research involving
animals. Pound, in her response, could not address this important point
because doing so would reveal the fundamental dishonesty of the leaders of
allegedly antivivisection organizations.
Before assessing their claims about the ethics of experiments ON
(while they
conveniently ignore the much larger set of experiments USING) animals and
the alleged inability to extrapolate from nonhumans to humans, one needs
to
know the true breadth of animal usage in science, both basic and applied.
The use of animals is far more widespread than Pound, Morris, Nobis,
Greek,
Shanks, and Lafollette will ever admit, because doing so would demolish
both
their ethical and “scientific” positions.
For example, Morris makes the incredible claim that “advocates of the
anti-
vivisection position are well aware of the distinction between animals as
"models" for human systems, and the use of animals to foster general
understanding of scientific processes.” I thank him for clearing this up
for
me, because if I accept Morris's claim as true, I can then safely conclude
that
all antivivisectionists are simply lying when they promote “non-animal”
and/
or “alternative” methods in research and testing. This conclusion is
inescapable because virtually all of the methods they actively promote
(epidemiology, in vitro methods, cell culture) still use animals as tools,
not
merely subjects, and many of them still require functional
interchangeability
of animal and human mechanisms, which is much more stringent than the
analogies they claim are so weak.
To illustrate this hypocrisy, I offer a highly representative case.
Close
examination of cell culture (neither non-animal nor an alternative)
methods
shows the moral and intellectual bankruptcy of claiming that animals have
rights, because virtually all cell culture exploits live animals as tools
in the
form of calf and fetal calf serum ("serum-free" culture usually uses
growth
factors purified from serum). Morris, in email correspondence, insisted
that
he has no ethical problems with cell culture and that it is “nonanimal”
because the entire animal isn’t used! In addition to this false moral
dichotomy, cell culture perfectly illustrates the scientific hypocrisy of
claiming
that animals are not analogous to humans while promoting a method that
involves growing human cells in an entirely bovine growth-factor
environment--in other words, using bovine serum as a model for the human
environment. Why do Morris et al. hide animal exploitation and
extrapolation
to present patently false dichotomies to the public?
Competing interests:
For 25 years, I have used both
whole
animals and "non-animal
alternatives," but they have
never been alternatives to each
other. Currently, I use hundreds
of mice annually in experiments
that are promoted by animal-
rights polemicists as "non-
animal," because the
experiments are performed on
cells in vitro. I also use bovine
serum (obtained by cardiac
puncture without anesthesia) to
culture human and mouse cells,
and antibodies made in animals
to localize proteins within those
cells. None of these are
opposed by the animal-rights
industry; however, if animals
have rights, all of these things
are wrong.
Competing interests: No competing interests
For an answer to the question asked in the title, the authors need to
look no further than their own paper. Their own cited references confirm
consistency of results obtained with animal and human studies. Horn et
al. (Nimodipine) conclude that "There were no differences between the
results of the animal experiments and clinical studies", Lucas C et al.
(laser therapy)state "In fact, there were no differences between the
results of these experiments (animals) and clinical trials". Even the
fluid rescuscitation experiments were generally inconclusive. The fact
that a number of responders appear to interpret this paper by Pound et al.
as more proof of the invalidity of using animals in research, makes me
think that some did not process the actual information contained in this
paper. This may have been due to the catchy title. The fact that data
obtained from animal studies were ignored in the human trials, does not
invalidate the use of animals nor the results obtained.
It is unfortunate that from this already very small section of research
reviewed, most papers deal with drug testing. Only one deals in some way
with basic research, the other major area in which animals are used. This
study has found that contrary to humans, dominant nonhuman primates are
reported to be more prone to heart disase. The actual results of research
are not as simplistic as that. Kaplan et al. found that dominant nonhuman
primate males are more prone to develop coronary artery atherosclerosis
only if they are constantly exposed to an unstable social environment.
The message of the paper by Pound et al. is that scientists and clinicians
must not conduct animal and human trials concurrently, and that human
trials must not proceed despite inconclusive or negative animal data. The
title of this paper is therefore inconsistent with the content.
As far as flaws in methodology is concerned, these are difficult to
comment on without seeing the papers. A major criticism appears to be the
use of too few animals resulting in insufficient statistical power.
However, even one author cited by Pound et al. as reference (Roberts et
al. 2002)alludes to the dilemma of balancing reduction (3Rs), as well as
legislative requirements of using the minimum number of animals, with the
need for enough statistical power.
Roberts et al: Does animal experimentation inform human healthcare?
Observations from a systematic review of international animal experiments
on fluid resuscitation. BMJ 2002, 324: 474-476.
Kaplan et al.: Social behaviour and gender in biomedical
investigation using monkeys: studies in atherogenesis. Lab An Sci 1991,
41: 334 - 343
PS: Above represents my own opinion and not necessarily that of my
employer
Competing interests:
None declared
Competing interests: No competing interests
This contentious topic is dramatically illustrated by an event
related in BMJ No7149(25April 1998 p1296 second paragraph)
Although animal studies have undoubtedly benefited us in many
ways,the animal studies related in BMJ No 7140p1296 had
tragic consequences.The whole rationale for animal studies, if based upon
the assumption that we share a common evolutionary ancestry with other
animals(particularly those subjected to animal studies)is irrational.Let
me explain.
Animal survival depends upon reproduction.In sexually reproducing
animals reproduction requires the genesis of gametes which requires
meiotic cell division.Meiosis requires the synapsis of homologous
chromosomes,which requires a certain degree of structural and genetic
homology
between each of each of the pairs of such chromosomes in the sexually
reproducing animal.
Each animal species has a uniquely different chromosome number(apart
from those which incidentally share a similar chromosome number).Please
visit www.kean.edu/~breid/chrom2.htm
The requirements of synapsis(and therefore of meiosis) have consequently
"fixed"the chromosome numbers of species,which means there is no way we
could share a common ancestry with other animals.Unless multiple
generations of our forebears had accidentally spawned almost identical new
chromosomes in the gametes of each of a mating couple in the same place at
the same time ,and unless such new chromosomes were compatible with life
and were inherited by succeeding generations.
Does this sound feasible?I think not.
This is the fundamental flaw in the rationale for animal experiments,since
we cannot be phylogenetically related to other animals,and therefore the
only similarities in our respective reactions to experiments are the
result of accident or design,and whether by accident or whether by
design,we cannot know them,except by experimentation,and not by deduction.
Competing interests:
None declared
Competing interests: No competing interests
Sir,
We read with interest the manuscript written by Pound et al.(1) and agree
with the authors on the need of systematic review of animal trials. We
also congratulate them for initiating the Reviewing Animal Trials
Systematically (RATS) Group in order to assure that experimental animal
studies are designed to provide sufficient support for future human
studies. However, the authors quoted the study by Horn et al. on
nimodipine in experimental animal models(2), published in 2001, as the
first systematic review of animal research. In 2000, we published a meta-
analysis of animal studies investigating the neuroprotective effects of
drugs inhibiting glutamate release (3). This study was incepted in Mexico
while we were discussing the possibility of testing lamotrigine as a
neuroprotective agent in cardiac surgical patients. The meta-analysis,
conducted and analyzed according with the method described by D’Agostino
and Weintraub,4 revealed that the administration of drugs inhibiting
glutamate release produce a “meta-analytic” reduction of 0.5 mm3 (95%CI -
0.35 to -0.65 mm3) of infarct volume. We discussed that although the
difference was statistically significant, the extrapolation to humans was
difficult because the clinical translation of such a reduction was
unknown. This information appeared summarized in the abstract of our meta-
analysis. Although Pound et al. cited our manuscript, they did not
recognize it as the first meta-analysis on animal studies. This omission
would not be important in a typical review paper, however, in a document
written by a group that is proposing to systematically review the studies
performed in experimental animal models, this is a significant omission
that demonstrates a lack of a thorough review of the literature by them.
The scientific literature is not only abundant in poorly conducted animal
studies but also in reviews performed with imprecision and bias.
References:
1. Pound P, Ebrahim S, Sandercock P, Bracken MB, Roberts I. Where is the
evidence that animal research benefits humans? BMJ 2004; 328: 514-7.
2. Horn J, de Haan RJ, Vermeulen M, Luiten PGM, Limburg M. Nimodipine in
animal model experiments of focal cerebral ischemia:a systematic review.
Stroke 2001; 32: 2433-8.
3. Nava-Ocampo AA, Reyes-Perez H, Bello-Ramirez AM, Mansilla-Olivares A,
Ponce-Monter H. For ischemic brain damage, is preclinical evidence of
neuroprotection by presynaptic blockade of glutamate release enough? Med.
Hypotheses 2000; 54: 77-9.
Competing interests:
None declared
Competing interests: No competing interests
Dear Editor - Please see the commentary by two of the authors of the
BMJ paper, Peter Sandercock and Ian Roberts, in The Lancet of August 24,
2002. It is entitled "Systematic reviews of animal experiments". Many of
the same arguments are made there as here in the BMJ. The nimodipine
example is duplicated. The Lancet reference does not appear in the BMJ
reference list. It should, shouldn't it?
Competing interests:
None declared
Competing interests: No competing interests
That animals contributed to medical discoveries hundreds of years ago
is debateable. Covering over thirty years research, in his thoroughly
referenced book, 'Who Goes First', senior medical correspondent to The New
York Times, Lawrence Altman MD, has put the history together and uncovered
a solid but previously unrecognised part of the underpinning of modern
medical science. It should be required reading for anyone interested in
medical history and contains documented evidence of pioneers of past
centuries who used themselves as guinea-pigs to advance our knowledge of
medicine, and the discoveries that have been mistakenly attributed to
animal experiments.
A hundred years ago childbirth and infectious diseases were the main
causes of death due to poor living conditions, but as living conditions
improved infant mortality declined and life expectancy soared. Many of
todays diseases didn't exist, were rare or unheard of, and infectious
diseases had declined to almost zero, before the introduction of vaccines.
However short memories and vested interests have turned medical history
upside down.
Today despite billions of animal experiments, millions spent on
research, hi-tech equipment, screening devices, more hospitals, doctors
and drugs, we have more sickness and disease than at any time in history,
new diseases unheard of before, and drug related diseases have reached
epidemic proportions. An appalling testimony to animal rsearch. The
overuse of drugs and vaccines have created new diseases and caused viruses
and germs to mutate creating more virulent strains that are drug
resistant. Medicines answer? Stronger drugs and vaccines. Chemicals-
developed in animals-are linked to neurological disorders, birth defects
and genetic diseases, all of which are a bonanza for the pharmaceutical
industry to develop more drugs.
The bulk of medical research today commissioned by the pharmaceutical
industry, motivated by profit. Animals are used by researchers to try out
their ideas and theories in the hope of developing drugs for their funding
source. However an artifically induced disease in an animal cannot
accurately mimic a naturally ocurring human diseas. Scientists admit the
difficulty of extrapolating results from animals to humans, and are unable
to unravel the thousands of chemical and biologial reactions a drug
triggers in the human body. (systems biology). Drug reactions differ
between different strains of the same species. Even age, sex or race can
make a difference. Only after a drug is marketed can they really be
evaluted. In other words, after trial and error in human patients - the
real guinea-pigs. Many important discoveries have been accidental. The
fact that drugs are a major cause of death is ignored, as is the means by
which they were developed!
If, as Mr. Blakemore claims animal research is successful, shouldn't
we be seeing a decline sickness and disease? In fact the opposite is true.
In a recent interview Mr. Blakemore says that future medicine will be in
treating the patient rather than the disease, embracing diet and lifestyle
factors - something the alternative professions have always known and
practised! This is indeed a step in the right direction. Tailoring the
treatment to suit the patient cannot be studied in any other species, and
will hopefully consign animal experiments to history.
Competing interests:
EFMA - Europeans for Medical Advancement
Competing interests: No competing interests
The evidence is right in front of you
It is a given that every department of science, however humble or
auspicious, be it medical or non-medical, animal or non-animal, clinical
or pre-clinical, should be subject to critical scrutiny and comparison by
means of peer reviews, systematic reviews and meta-analyses.
The authors of the study “Where is the evidence that animal research
benefits humans?” conclude that animal research should be subject to
systematic review. Is there nothing more interesting to be gleaned from
this study than a statement of the obvious? Of course all areas of science
should be constantly reviewed, but that is just a means not an end in
itself. The end is a better understanding of the evidence.
Before drawing attention to a more useful conclusion which may be
drawn from this study I will make one criticism of its method. The study
succeeds in identifying a total of 25 systematic reviews from a search of
Medline and promptly discards 19 of them for reasons that appear
altogether arbitrary given the description of their nature. They sound
highly pertinent to the topic in question. It is the judgement of the
authors that the six selected reviews “shed the most light on the
contribution that animal research makes to clinical medicine.”
25 reviews must shed a lot more light than just half a dozen. We
begin to approach the meta-analysis in scale. And it is difficult to
believe that the clinical relevance of one of the six reviews – by
Petticrew and Davey Smith – is worth including if these others are not. It
is a great shame that these reviews were identified and then excluded.
None of the five systematic reviews included in this study which
feature pharmaceutical or physical interventions demonstrate a concurrence
of beneficial outcome in both the clinical and animal case; nor do any of
them provide an example where the animal model shows a benefit from
intervention whilst clinical results demonstrate harmful effects.
All five cases of intervention fail to produce any clinical benefits
and they all fail to show any definite evidence of benefit on animal
models. In fact all five cases are exactly concurrent in terms of the
effects of intervention on humans and animals. In three cases the
interventions in question are ineffective for both humans and animal
models, and in the other two cases there is evidence of harmful effects
for animals and humans. This exact concurrence might just be chance – or
it might not.
This last result constitutes the real importance of this study. The
question which the study poses is whether animal research can be a benefit
for humans and the more interesting conclusion than the one which the
authors give is the tentative answer – yes it can.
This study provides some reasonably working evidence that animal
research is able to indicate when medical interventions are going to be
ineffective or harmful on humans – when that evidence is carefully
assessed. The benefits of this are twofold.
We can protect people from exposure to unnecessary risk and at the
same time we can save time and money by avoiding unnecessary clinical
trials. If careful study shows that a proposed new medical treatment
either does not work for, or harms animals, then we had better not go
there!
Competing interests:
None declared
Competing interests: No competing interests