A response to Diabetes UK

What follows is some information that may help supporters who wish to reply to Ms Young.

1. Ms Young states that ‘all our grant applications go through a strict peer review system when deciding which to fund… When our researchers do use animals, all work is carried out in line with strict Home Office guidelines. When assessing applications, our Committee members are asked to comment on the appropriateness of animal usage’.

There is evidence that the peer review process for animal experimentation is flawed, secretive and biased at all stages (please see page 39 of Victims of Charity). Furthermore, as noted by the British Union for the Abolition of Vivisection, the regulatory framework for animal experiments is ‘permissive, cloaked in secrecy and offers little meaningful protection to animals’.

It is absurd to claim that a regulatory framework that licenses an increasing number of animal experiments is ‘strict’. 3.79 million experiments were started last year, an increase of 68,000 on the number conducted in 2010. This is the highest level in 24 years. Furthermore, Diabetes UK (via its membership of the Association of Medical Research Charities) has recently supported the weakening of many elements of the UK’s current laws on animal experiments.

As to the ‘appropriateness of animal usage’, it would be instructive if Diabetes UK could publish its Committee’s comments on this example of their currently funded experiments. A researcher at Southampton University is being paid to make pregnant mice obese (to later kill them and their offspring), supposedly to ‘study the effects of the anti-diabetes drug metformin on obese pregnant mothers and their children’. The research apparently ‘could provide the foundation for a treatment strategy for obese pregnant mothers to improve their health and the future health of their children’. Firstly, it is obvious that mouse experiments are not studying effects on people. Furthermore, it is already known from human studies that children born to obese mothers have an increased risk of diabetes in later life, and that metformin is reportedly both safe and effective during pregnancy. The ‘treatment strategy’ being researched is already in widespread use in people.

2. Ms Young claims that ‘Completely replacing all animals in research is not yet possible. There is no alternative method that can reproduce the complex biological processes that lead to diabetes and its complications.’

There are a great many alternative approaches to disease research that do not involve animals. Diabetes UK itself funds cell cultures, computer models, and human studies, and references to such techniques can be widely found. A Manchester professor is researching diabetes using human stem cells, and believes that ‘these cell lines will make a significant contribution to reducing and replacing the numbers of animals used in this field and will provide significant advantages over existing model systems’.

A letter signed by 16 eminent researchers, and published in The Guardian in July, states:

‘A host of powerful human-biology-based cellular, genomic and computational tools are available that can often better predict people’s real-world reactions to drugs and chemicals than conventional animal tests. For many of us, replacing, reducing and refining animal experiments is driven by a desire to develop better approaches to researching human illness. Research innovation can bring huge societal benefits by improving the speed, reliability and human-relevance of the tools we use to answer biomedical questions.’

The researchers call for ‘substantial, dedicated funding’ for humane alternatives. Part of the explanation for the continued rise in animal experiments is that alternatives have been consistently underfunded and marginalised. The endlessly repeated mantra of ‘no alternative’ is still loudly shouted by animal experimenters, and receives disproportionate attention from government.

The claim that animals reproduce ‘the complex biological processes that lead to diabetes and its complications’ in humans is an unsubstantiated, subjective assertion, not a scientific fact. A recent systematic study entitled ‘The Costs and Benefits of Animal Experiments’ (Knight, 2011) demonstrated that ‘there is a growing body of empirical evidence that casts doubt on the scientific utility of animals as experimental models of humans’.

People are not giant rodents, and experiments on the latter are unlikely to be valid for the former. Diabetic ‘mouse models’ have led to the development of literally hundreds of ‘curative’ therapies for Type 1 diabetes that have failed in people – some of these ‘false dawns’ are highlighted below. This led two researchers to conclude in 2004:

‘We would indeed maintain that the Type 1 diabetes research community has developed a “selective blindness” as evidenced by its failure to recognise a number of shortcomings associated with animal models of the disease.’

One of these researchers subsequently wrote in 2007: ‘Despite decades of research using various animal models for type 1 diabetes, we are still struggling to define the initiating autoantigens, the precise mechanisms of β cell destruction, and suitable immune-based interventions to prevent or treat human diabetes…It is therefore time to ask ourselves whether we are making major strategic mistakes when employing ‘rodent models’ for the study of type 1 diabetes.’

These researchers are not alone – another review from 2009 points out that ‘clinical translation, in this case mostly from mouse to man, has been slow and fraught with difficulties. It is therefore important to take a critical look at the present diabetes mouse models and their utility for creating scenarios that might or might not apply to the pathogenesis of human T1D’.

Once again, animal experimenters are trying to mimic in mice what they do not understand in humans. This back to front process epitomises bad science – how can an animal model based on erroneous assumptions even contribute to medical knowledge, let alone deliver treatments for people?

3. Ms Young claims that ‘insulin therapies were discovered using animals in research as were the most common drugs used today to treat Type 2 diabetes’.

The history of the discovery of insulin is actually far more complex than Banting and Best’s horribly cruel dog experiments, in which researchers removed or surgically tied off their pancreas glands. A thorough examination of the historical facts is presented by Greek in his book ‘Sacred Cows and Golden Geese – the Human Costs of Experiments on Animals’. He calls insulin ‘the animal experimenters’ poster drug’, and points out that ‘there are profound holes in the assumption that animal experimentation was necessary’.

Human autopsy studies by Cawley, Bright and Bouchardat in the 18th and 19th centuries first revealed the importance of pancreatic damage in diabetes. Human studies by Paul Langerhans in 1869 led to the discovery of insulin-producing islet cells in the pancreas. The link between this organ and diabetes was firmly established before the twentieth century.

The name insulin was not coined by Banting and Best, but by Jean de Meyer in 1909. Sir Edward Albert Sharpey-Schäfer also suggested the name in 1913, when he described the actions of the hormone he believed was produced by islet tissue in the pancreas.

Another vital piece of observational medicine, which acted as an incentive for Banting to cut up dogs, was provided by Moses Barron. While doing routine autopsies, he had come across a diabetic patient who had suffered a rare case of a pancreatic stone. The stone had completely blocked the main pancreatic duct, destroying the cells that produce digestive enzymes but not the insulin-producing islets. Banting theorised that the same damage could be provoked in dogs by tying off their pancreatic ducts, and then extracting insulin from the islets that remained. He hypothesised – incorrectly – that pancreatic enzymes were destroying the insulin that others had previously tried to isolate, and so thought that his surgery on dogs was essential.

During the experiments, many dogs suffered a prolonged and miserable death from surgically induced illnesses that included not just diabetes, but also infection and trauma. A 2002 historical article details how Banting and Best may well have purchased companion animals due to the high death rate of their victims:

‘It was a laborious task for someone with no experience in animal work, and it did not go well at first as Banting struggled to improve his surgical technique. By the end of the second week, 7 of their 10 dogs had died. To resupply the animal cages, they resorted to buying dogs on the streets of Toronto for $1.00 to $3.00 with no questions asked of the suppliers.’

Many other dogs died after the brutal surgery to which they were subjected. Later in the experiments, the duo (having run out of duct-ligated dogs) made extracts of whole fresh dog pancreas, which also lowered blood sugar in animal subjects. Their previous surgical procedures were shown to have been unnecessary.

It should be clear from the facts that the dog experiments were only ‘necessary’ due to the primacy afforded to vivisection as scientific method. The convention at that time (and still to a large extent today) was to denigrate observational studies in humans unless they could be ‘proved’ in animal experiments. The 1920s dog vivisection simply acted as a spur to the interest of other workers, who used chemistry to isolate and purify Banting’s crude extracts. Plentiful human tissue was available with which to have performed the same chemistry.

Type 2 diabetes has been treated for many years with only a few key drugs, mainly metformin and the sulphonylurea drugs. The real potential of metformin to lower blood glucose stemmed from observations made in humans, when the drug was used to treat cerebral influenza. The sulphonylureas were likewise developed following serendipitous discoveries in human patients. Even as far back as 1964, the US pharmacologist F. Gilbert McMahon commented with reference to these drugs: ‘Many major therapeutic advances have been made by serendipitous observations in man, rather than animal screening.’

The development of new treatments using rodent models of diabetes has led to the marketing (and subsequent withdrawal) of some truly dangerous drugs. Rosiglitazone, sold as Avandia, was withdrawn from clinical use in Europe in 2010, because of concerns that it increased cardiovascular risk – patients taking the drug were more likely to suffer a heart attack. Multiple cruel studies on diabetic rats showed a completely contrary result – the drug protected their hearts from experimentally-induced damage. Troglitazone was withdrawn in 2000, as it caused numerous cases of liver failure in humans. Tests in mice, rats and monkeys failed to show this risk, except in rats given the drug in doses toxic to their hearts.

It is interesting to contrast Ms Young’s robust defence of animal experiments with the more qualified remarks of Dr Iain Frame, Director of Research at Diabetes UK. Dr Frame, when commenting on a cell transplant therapy that worked in mice in 2009 (it has not reached trials in humans) said: ‘We have to bear in mind that these are very early experiments in mice. Transferring this technology to provide a real and lasting treatment for diabetes in humans will not be straightforward and may well take a long time’. When commenting on (yet another) therapy for diabetes that works in mice, Dr Frame said: ‘The research is at a very early stage and has shown some benefit in female mice with diabetes and less benefit in male mice. Whilst promising, it would take an enormous leap of faith to assume a new pill will soon be on the market for people with, or at high risk of, Type 2 diabetes.’

Most medical charities issue strong and unequivocal defences of animal experiments in response to public campaigns that could affect their funding. Researchers, however, are often much more measured when addressing their own communities, as they are well aware of the repeatedly dashed hopes of patients, due to animal research hype.

4. Ms Young states that ‘a vaccine to prevent Type 1 diabetes is on the horizon…but this research needs to be developed and tested before it can be used successfully in humans’.

A ‘vaccine’ for type 1 diabetes has been ‘on the horizon’ for some time, with ‘development’ ongoing in animal models. As with the other therapies listed above, there are many compounds that cure the disease in rodent models but, thus far, none has made it past a Phase III trial in humans. Focus on animal models has, true to form, wasted time, money and resources.

The concept of a vaccine for type 1 diabetes stems from the fact that the disease is autoimmune – the body basically attacks itself, destroying the cells in the pancreas that produce insulin. Approaches to a cure, which include vaccines, are often designed to prevent or lessen this deranged immune response. Such therapies are generally termed immunomodulatory.

So far, immunomodulatory therapies as a whole have been a crushing disappointment for patients. This is despite resounding success in animal models, especially the non-obese diabetic (NOD) mouse. Some recent examples are:

• The failure of monoclonal antibody therapies. One of the first of these antibodies, OKT3, was obtained from mouse tissues and caused immediate severe toxic side effects in patients (chills, fever, low blood pressure, and breathlessness). Phase III trials of ‘humanised’ versions were suspended in 2010 and 2011 due to lack of efficacy.
• The failure of interleukin-1 blockers to reverse or prevent type 1 diabetes in people, despite success in rats and mice.
• The failure of glutamic acid decarboxylase (GAD) vaccines to preserve pancreas insulin-making function in children (Phase III trial failed in 2011). GAD vaccines had shown spectacular results in female prediabetic mice
• The failure of various vaccine strategies, using insulin, to work in people. An example is the recent lack of efficacy of the IBC-VS01 vaccine in a small clinical trial, despite its preventing or delaying diabetes in NOD mice.

The situation to date is summed up by the following quote from a 2011 paper: ‘A major hurdle in the design of human vaccines for T1D is the translation from animals to humans…. a treatment that proves efficacious in mice may … only prevent disease in a subset of human patients whereas a vaccine that fails to prevent disease in mice, arguably because of a specific genetic feature in the NOD mouse, would not be considered for further evaluation, without knowing if it would probably be efficacious in a larger cohort of humans.’

In other words, using mouse models can result in useful treatments for people being scrapped at the outset, whilst (at best) furthering the development of minimally effective therapies. In summary, contrary to the assertions of Diabetes UK, animal models for diabetes have more often been misleading and misinforming in the search for human treatments. The mocking up of a poorly understood human disease in animals, in order to have a ‘living system’ to study, is both bad science and bad ethics. It is not just a betrayal of the animals, but of patients who are rightly sick of endless false dawns.

(References available on request)