Summary of Council decision:
Three issues were investigated, all of which were Upheld.
A leaflet for the National Anti-Vivisection Society featured an image of a monkey in a laboratory cage on the front page and large text, which stated "April 24 World Day for Laboratory Animals" and "HELP STOP Britain's shocking monkey business". Text on the back page included "The UK uses more monkeys than any other EU country", "Animals react differently to substances, including drugs, so animal safety tests cannot accurately predict what will happen in humans. Test drug TGN1412 caused near-fatal reactions in human volunteers, yet doses given to monkeys 500 times higher did not produce such side effects" and "The TGN 1412 drug trial disaster could have been avoided using a technique known as 'micro-dosing' with spectrometry analysis".
Understanding Animal Research challenged whether the following claims were misleading and could be substantiated:
1. "The UK uses more monkeys than any other EU country";
2. "animal safety tests cannot accurately predict what will happen in humans"; and
3. "The TGN 1412 drug trial disaster could have been avoided using a technique known as 'micro-dosing' with spectrometry analysis".
1. The National Anti-Vivisection Society Ltd (NAVS) said the claim "The UK uses more monkeys than any other EU country" was supported by the officially reported EU statistics, which are published every three years. They said they had used the statistics from the last three reports as a sample, and to avoid the picture being distorted by one particular report. They provided copies of the relevant reports from 2005, 2008 and 2011 as well as other older reports. These demonstrated that the UK used an average of 2643 New and Old World monkeys combined, compared to 2323 in France and 1826 in Germany. They said these figures did not include prosimians, such as lemurs, which were not monkeys. They also said the 2012 figures available individually for Germany and the UK, and the latest figures available for France (which were figures from 2010 used in the 2011 EU report), showed that, ahead of the publication of the next set of official EU statistics, the UK currently used the highest number of primates.
2. NAVS said that animal tests could not accurately predict what would happen in humans because of species differences, and that each species reacted differently to drugs and other products. The problem with animal experimentation was therefore that researchers tried to extrapolate results from one species to another in spite of known and unknown differences. They said that many scientists and other academics globally agreed with their view.
They said a book published in 2007 by the United States National Research Council on the future of toxicity testing, which they provided an extract of, stated that the relevance of animal studies for toxicity testing to heterogeneous human populations which would be exposed at much lower concentrations had been questioned. It further stated that those tests provided little information on modes and mechanisms of action, which were critical for understanding interspecies differences in toxicity, and little or no information for assessing variability in human susceptibility.
They provided a 2008 article about the limitations of animal tests by the then head of the European Centre for the Validation of Alternative Methods (ECVAM), part of the European Commission's Joint Research Centre (JRC), published in a peer-reviewed journal which focused on research about development and promotion of alternatives to animal experiments. They said the article stated that most animal test protocols were not validated, that a systematic assessment of performance characteristics was lacking and that there was an unwillingness to critically reassess those older methods. They also referred to a statement on the JRC website which said that traditional risk assessment approaches are insufficient to adequately predict the potential risk associated with a substance. They provided four further journal articles, which concluded that the animal model in toxicity testing was unreliable. They also referred to testimony given to a House of Lords Select Committee on Animals in Scientific Procedures in 2002 by a scientist, in which they referred to the fact that animal experiments on tamoxifen (a widely prescribed drug) showed that it caused cancers in rats and mice, but that those results were dismissed as irrelevant to human beings.
They said that toxicity could and did arise even after safety testing in animals appeared to indicate there would not be problems. They provided two published journal articles which referred to the TGN1412 incident. A further journal article examined whether animal models of disease could inform human studies, and stated that clinical trials were essential because animal studies did not predict with sufficient certainty what would happen in humans. NAVS said the legal requirement to conduct safety testing in two species of mammal (a rodent and larger non-rodent) before any human testing, was not in itself an indication that animal testing was reliable. They provided a 2014 article published in a reputable medical journal which stated that poorly designed studies and a lack of methodological rigour in pre-clinical research could result in fruitless clinical trials which needlessly exposed humans to potentially harmful drugs.
They provided an information paper from an animal welfare charity on the use of animals in toxicity testing, which set out their belief that the use of animals to test toxicity was a flawed approach, and that alternatives should be developed. They also provided a published article on lay and expert judgements of chemical risks. They said the article found a division of opinion among experts on the validity of toxicological testing in animals. They also provided a peer-reviewed and published article on what clinical studies revealed about the value of preclinical work, including animal testing. The article referred to adverse reactions and stated that the rate of these demonstrated that despite the preclinical toxicological testing, drugs were not very safe. They also provided a published debate article on whether or not animal tests were inherently valid. The article argued that they were not, and also referred to a general assumption that animal tests could be considered relevant and reliable for predicting toxicity to humans.
They also provided a 2006 research letter published in a medical journal which systematically reviewed highly cited animal research and looked at whether the findings translated to human randomised trials. It concluded that only about a third of highly cited animal research translated into positive outcomes at the level of human randomised trials and that poor replication of even high-quality animal studies should be expected by those who conduct clinical research. A further peer-reviewed and published article was provided which analysed published systematic reviews of the human clinical or toxicological utility of animal tests. The authors concluded that most systematic reviews demonstrated that animal experiments were insufficiently predictive of human outcomes to provide substantial benefits in deriving human toxicity assessments. They stated that despite this, most animal models were (incorrectly) assumed to be predictive of human outcomes. They also provided a published article which examined the future of research into abnormal prenatal development and congenital malformations caused by external chemical or physical agents (teratology). The authors concluded that animal testing methods constituted questionable science and that the future of such research lay in other non-animal methods.
NAVS provided a number of articles which they said demonstrated their views regarding animal testing were widely held. An open letter to the UK Prime Minister and Health Secretary, published in a medical journal, referred to an over reliance on animal testing to predict drug behaviour in humans and that many studies had shown they were no more predictive than the toss of a coin. An article from a scientific magazine also referred to the unreliability of animal testing for toxicity. A 2008 article published in a veterinary journal examined the validity of animal models for predicting results in humans and concluded that they failed to predict human responses. An article published in a medical philosophy and ethics journal came to the same conclusion. An article published in a nature journal called for industry to get behind an alternative to animal testing and stated that the data available gave grounds for concern, and a longer news feature expanded on this and stated that animal tests were poorly predictive. A thesis examined the scientific value of non-clinical animal studies in drug development and stated that animal studies did not always predict pharmacological effects or safety risks for humans. An ad for a talk by an MEP also stated that toxicity testing was not delivering the required safety standards.
They said there were numerous examples where the effects of substances seen in animals were not highly predictive of the effects seen in humans. For example, the potentially fatal toxicity of chocolate to dogs, and the TGN1412 incident in which the test subjects in a clinical trial had a life threatening reaction to the test drug.
3. NAVS said that in a published letter, which they provided, to the Expert Working Group on the TGN1412 incident (from a group that represented several organisations which funded the development of replacements for laboratory animals in research) stated that microdosing in combination with Accelerator Mass Spectrometry (AMS), during which very low doses were administered to volunteers, was increasingly accepted as a technique which could reduce animal testing of some compounds. The letter went on to explain how AMS could be used and to refer to a specific manufacturer of AMS technology. NAVS said it would not make sense for the letter to refer to this specific manufacturer if spectrometry could not be used for TGN1412. They also provided an article from a national newspaper in which a scientist was quoted as saying that the use of microdosing could have avoided the TGN1412 incident. They referred to a news article on a medical technology website, which they provided, which stated that many agreed the incident could have been avoided using AMS. They provided an opinion piece on microdosing published in a journal which stated that a microdose study with TGN1412 could have provided data relevant to the humans. They also provided a brief document from the European Union Microdose AMS Partnership Programme which explained the relationship between microdosing and AMS, and that microdosing studies were dependent on ultra-sensitive techniques such as AMS.
They also said the claim stated "The TGN 1412 drug trial disaster could have been avoided using a technique known as 'micro-dosing' with spectrometry analysis" rather than that it "would" have, and that it was important to take into account the different definitions of "could" and "would"; dictionary definitions were supplied.
The ASA considered that in the absence of qualification, readers would expect the claim "The UK uses more monkeys than any other EU country" to be based on the most up to date comparative figures available. We understood that the EU collected data on the numbers of animals used in testing and that those figures were published every three years, with the most recent report relating to data from 2011 (although the figures quoted for France in the report related to 2010). The report showed that in those years France and Germany had used more monkeys in research than the UK. However, NAVS had based the claim on the average of figures from the reports relating to 2005, 2008 and 2011. We considered that it was reasonable to base the claim on an average from the three most recent reports, but that the claim should have been qualified to make this clear. We acknowledged that figures for 2012 indicated that the UK currently used a greater number of primates than Germany, but noted that the most up to date figures for France related to 2010 and that at that time the UK's use had not been the highest. We therefore considered that NAVS had not demonstrated that at the time the ad was published the UK used more monkeys in research than any other EU country. Because the claim was not qualified, we considered it implied that it related to the most up-to-date comparative figures only, and because that was not the case, we concluded it was misleading.
On this point the ad breached CAP Code (Edition 12) rules 3.1 3.1 Marketing communications must not materially mislead or be likely to do so. (Misleading advertising), 3.7 3.7 Before distributing or submitting a marketing communication for publication, marketers must hold documentary evidence to prove claims that consumers are likely to regard as objective and that are capable of objective substantiation. The ASA may regard claims as misleading in the absence of adequate substantiation. (Substantiation) and 3.9 3.9 Marketing communications must state significant limitations and qualifications. Qualifications may clarify but must not contradict the claims that they qualify. (Qualification).
We considered that, in the context of the ad, readers would understand from the claim "animal safety tests cannot accurately predict what will happen in humans" that animal testing for safety had little predictive value for clinical testing in humans. We considered the claim would be perceived as going further than a statement that animal safety testing would not always accurately predict the outcome in humans. We also considered that the claim was presented as factual and universally accepted.
A number of the articles provided by NAVS focused on the reliability of animal testing in predicting efficacy in clinical trials and, because the claim focused on safety, we therefore considered those articles to be of limited relevance.
We considered the articles which focused on the reliability of animal testing in predicting human toxicology. Two of the articles which focused on safety had the same authors and had been published in a peer-reviewed journal. The research had been funded by organisations opposed to animal testing, although they had not been involved in the research or reviewed the articles. The articles concluded that toxicity studies involving dogs were highly inconsistent predictors of toxic responses in humans, and that the same was true of toxicity studies of animals in general. A further article published in the same journal examined systematic reviews and concluded that the majority demonstrated animal tests were not sufficiently predictive of toxicity in humans to provide substantial benefits. Further articles which they had provided also came to similar conclusions. We noted that a number of the articles had authors in common, and that most had been published in reputable peer-reviewed journals that focused on alternatives to animal research. A number of articles had also been published in journals that did not focus on alternative methods, and we therefore considered that such views were held by a variety of scientists across the informed community. We therefore considered that the evidence provided by NAVS demonstrated that toxicological testing in animals was not 100% predictive of that outcome in humans, which we understood was generally accepted, and that some sections of the scientific community had questioned whether animal models were as useful in determining the efficacy and safety of new substances as generally believed.
However, we understood that it was generally accepted within the scientific community that there was value in toxicity testing on animals, and also that it was a legal requirement in the UK that new medicines were tested on animals before human testing could proceed to ensure their safety. We also noted that many of the articles provided by NAVS specifically referred to the fact that it was generally thought that toxicity testing on animals had value, albeit that the authors generally did not agree with that position or the strength of that view. We did not consider that the evidence provided by NAVS demonstrated that animal testing for safety had little predictive value for clinical testing. We considered the claim "animal safety tests cannot accurately predict what will happen in humans" had not been substantiated and that the ad did not make clear that it related to the view of one section of the scientific community only. We therefore concluded that the claim was misleading.
On this point the ad breached CAP Code (Edition 12) rules 3.1 3.1 Marketing communications must not materially mislead or be likely to do so. (Misleading advertising), 3.7 3.7 Before distributing or submitting a marketing communication for publication, marketers must hold documentary evidence to prove claims that consumers are likely to regard as objective and that are capable of objective substantiation. The ASA may regard claims as misleading in the absence of adequate substantiation. (Substantiation) and 3.13 3.13 Marketing communications must not suggest that their claims are universally accepted if a significant division of informed or scientific opinion exists. (Exaggeration).
We considered that readers would understand from the claim "The TGN 1412 drug trial disaster could have been avoided using a technique known as 'micro-dosing' with spectrometry analysis" that if microdosing with spectrometry analysis had been used prior to the TGN1412 drug trial the outcome of the trial, which involved serious adverse reactions of those tested, would have been avoided. We considered the claim implied there was a high degree of certainty that the use of microdosing would have avoided that outcome. The evidence demonstrated that some scientists had suggested the use of microdosing might have detected the toxicity to humans of TGN1412, but did not demonstrate that was highly likely to be the case. We also noted that the final report of the Expert Scientific Group on the TGN1412 incident did not state that it could have been avoided by the use of microdosing with spectrometry analysis or recommend this as a course of action in future, although it did state that care should be taken in selecting the minimum possible dose for 'first-in-man' trials. We did not consider that the evidence demonstrated that "The TGN 1412 drug trial disaster could have been avoided using a technique known as 'micro-dosing' with spectrometry analysis" and therefore concluded the claim had not been substantiated.
On this point the ad breached CAP Code (Edition 12) rules 3.1 3.1 Marketing communications must not materially mislead or be likely to do so. (Misleading advertising) and 3.7 3.7 Before distributing or submitting a marketing communication for publication, marketers must hold documentary evidence to prove claims that consumers are likely to regard as objective and that are capable of objective substantiation. The ASA may regard claims as misleading in the absence of adequate substantiation. (Substantiation).
The ad must not appear again in its current form. We told the National Anti-Vivisection Society Ltd to ensure that claims were appropriately qualified and not to repeat the claims "animal safety tests cannot accurately predict what will happen in humans" and "The TGN 1412 drug trial disaster could have been avoided using a technique known as 'micro-dosing' with spectrometry analysis" in future ads. We told them not to imply claims were universally accepted if that was not the case.