Of mice and men: debating animal research in drug development

Animal research is a central pillar of drug development, but in medical terms, do the ends justify the means? The consensus in the research community is that it is a vital aspect of building scientific knowledge and improving clinical trials, while critics argue that the practice is unnecessary and ineffective. Dr Jarrod Bailey of Cruelty Free International and Chris Magee of Understanding Animal Research make the case for both sides of the debate.

The debate about animal research and testing has no shortage of emotion. The testing of cosmetics and other consumer products on animals is already banned in Europe and several other countries, and the practice is contested with a justifiable ferocity in the US and other markets where it is still legal.

While the days of legal animal testing for consumer research are hopefully numbered, in the medical context the picture is more complex. Popular opinion tends to be more accepting of animal research on an ethical level when it is used as a means of improving medical, scientific or veterinary knowledge. In the UK, for example, a 2014 survey found that 68% of respondents accepted the practice for medical research if there were no other alternatives.

How vital is animal research?

However, beyond the ethical debate over the use of animals for medical research, just how scientifically useful is data from animals in the development of effective drugs for humans? The general consensus in the scientific community is that animal research has been a cornerstone of medical research for countless decades, and has played a vital role both as a means of pre-clinical safety and efficacy testing before starting human trials, and as a method of building the fundamental knowledge that feeds into new treatments and medical innovations.

"All but a handful of Nobel Prize winners for physiology and medicine have been animal researchers," says Chris Magee, head of policy and media at Understanding Animal Research, a UK-based advocacy group campaigning to build understanding and acceptance of animal testing in biomedical research. "As well as discovering things like the role of insulin in diabetes using dogs, we've been able to develop and test medicines for many conditions, elucidate biological mechanisms and base treatments on animal physiology."

Critics from the anti-vivisection movement, meanwhile, contend that data from animal models is of dubious value to human medicine and that more reliable, ethically-sound and human-focused alternatives are available.

"Genetically, biochemically and physiologically, animals are not humans," says Dr Jarrod Bailey, a geneticist and senior research scientist at anti-vivisection group Cruelty Free International. "Mice aren't rats; cats aren't dogs, and monkeys aren't humans. Not only is there no robust, published evidence to support claims that animal testing of drugs is predictive of human responses - something bizarre after so many decades of the practice - but there are several reports suggesting a poor predictive nature of animal tests for humans; that toxicity in animals is not a good predictor of toxicity in people."

The case against animal research

Bailey argues that the role of animal research in the medical breakthroughs of previous decades has been overplayed by proponents of the practice.

"The past contribution of animal research to medicine is wildly exaggerated, and any sort of involvement of animals in any area of research, however tenuous or even confounding, is translated to a necessary and helpful involvement," Bailey says. "Many of these claims don't stand up to scrutiny: there are formidable arguments supporting the crucial and pivotal roles of non-animal approaches to salient medical breakthroughs throughout history, rather than of animal experiments. This includes painkillers, antibiotics, chemotherapy drugs, anaesthetics, AIDS drugs, and more - all of which are used by pro-animal research lobby groups to defend [animal testing]."

A major role played by animal research is in the pre-clinical testing phase for new drugs, to gather data on the safety and efficacy of a treatment before human subjects are exposed to its effects. But do the myriad genetic differences between species make it impossible to dependably extrapolate data from animal testing into a human model?

"The animal testing paradigm, on which drug development is still largely based, is failing."

Bailey says research conducted by himself and others has shown that inter-species genetic variability means that making meaningful inferences between animal and human reactions is not feasible; the genetic differences are just too high. He cites dozens of HIV/AIDS vaccines and hundreds of stroke treatments that have demonstrated efficacy in animal models - including monkeys and chimpanzees - but which have yielded no human HIV/AIDS vaccines and precious few stroke treatments to date.

"Even after relatively few years of study, these differences are myriad and of great consequence," says Bailey. "Many of them affect immune function, which explains the failure of many disease models - and of course if understanding of a disease is flawed, the search for effective treatment is going to be - and many affect the metabolism of drugs directly. So we don't just know empirically that the paradigm is failing - we are beginning to understand why."

This variability has also raised questions about the usefulness of animal testing to improve the safety of Phase I clinical trials in humans. There are well-known instances in which animal testing failed to predict sometimes deadly health risks for human patients, including clinical trials of monoclonal antibody TGN1412 at Northwick Park Hospital in London in 2006, during which the experimental drug caused catastrophic systemic organ failure in six healthy participants, despite demonstrating safety and efficacy in a wide range of animals.

A statistic often used by opponents of biomedical animal research is that, according to data from the US Food and Drug Administration (FDA), 92% of drugs that are successful in animal testing subsequently fail in human trials. Bailey notes that failure rates are getting worse and are indicative of the systemic weakness of animal testing.

"This tells us that the animal testing paradigm, on which drug development is still largely based, is failing," he says. "The scientific literature is replete with concerns over this, as well as calls to transition to more predictive and more human relevant approaches as a matter of urgency. These calls are getting more desperate as the industry struggles."

In defence of animal research

Magee is quick to offer a rebuttal to the failure rate figures, noting that these statistics refer to all pre-clinical research, the majority of which, he says, isn't animal testing.

"The failure of a candidate compound to make it all the way through pre-clinical testing and clinical trials is indeed this high, which isn't surprising at all and merely points to the fact that it's hard to find a compound that can become a drug whatever method you're using," Magee says. "In fact, some 86% of compounds that pass a stage one clinical trial in human volunteers 'fail' by stage three for reasons as diverse as efficacy and economic viability, none of which implicates the animal model."

The contribution of animal research to the safety of human trials, Magee adds, can be measured in the fact that adverse events in clinical trials are the dramatic exception in an otherwise excellent record.

"The numbers rather speak for themselves in that sense," Magee notes. "In the UK each year there are around 600,000 people volunteering for clinical trials, yet there have been no fatalities and only one serious incident in over 30 years of phase one clinical trials. This simply could not be the case if animal models weren't successfully screening out harmful compounds."

As for the TGN1412 incident at Northwick Park Hospital, Magee attributes the safety failure to a lax trial approach, which treated volunteers with the varied doses of the drug at the same time, rather than staggering the treatments one at a time.

"More often, animal research provides the basic knowledge upon which treatments can be based."

When it comes to extrapolating efficacy data between species, Magee acknowledges that this requires "more caution with regard to the species used and the target system", and notes that "humanising" test subjects using genetic manipulation shows potential as a means to reduce the problem of genetic variation while still having the benefit of working within a living system, which is one of the disadvantages of some alternatives to animal testing.

While Magee staunchly defends the utility of animal research to improve clinical trials, he argues that the practice plays a larger role in the underlying scientific experimentation that can lead to important innovations.

"As useful as it is, toxicological testing of potential new drugs makes up only around 15% of animal research," he says. "More often, animal research provides the basic knowledge upon which treatments can be based and we can trial experimental techniques like new surgical approaches to childhood heart conditions or curing paralysis. In general we are interested in both the similarities and the differences between species. Breast cancer drug Herceptin is based on a mouse antibody for instance, and we've candidate treatments for Ebola which were developed in a similar way."

Scientific discoveries can be decades apart from the medical innovations they inform, as is the case with Sir John Gurdon's pioneering frog-cloning experiments in the 1960s, for which he developed the nuclear transfer techniques that have facilitated modern stem cell research, including Gurdon and Shinya Yamanaka's Nobel Prize-winning discovery that mature cells can be reprogrammed to become pluripotent.

Encouraging the use of alternatives

One of the most important questions when considering animal research and its usefulness in medicine is not whether it has made a positive impact - which it clearly has done at many points, to some degree or another - but whether it is irreplaceable. In Europe, after all, it is illegal to conduct animal testing if an alternative is available, although there is no such law in the US and many other regions.

Naturally, Bailey argues that animal research is eminently replaceable, and somewhat surprisingly, Magee generally concurs. Technologies such as stem cell platforms, 3D tissue and organ cultures, and computer simulation are opening up new options both as complements to animal research and as potential replacements for it.

Bailey is particularly excited about the use of 3D tissue and organ cultures, "in which mini human organs made up of different cell types all function individually as well as together, and also 'body on a chip' approaches that permit the study of several human tissues/organs at once, with circulatory systems, metabolism, and so on. Would you rather take a drug for the first time that had been tested on rats and dogs, or that had been tested on a suite of alternatives that are human-relevant?"

"We are in the beginning of a golden age of alternatives."

Magee, meanwhile, notes that "we are in the beginning of a golden age of alternatives, with many more becoming available, and many parts of the pharmaceutical industry are busy assessing the value of various non-animal assays and approaches. Our organisation is increasingly focused on identifying and removing roadblocks to the creation and adoption of alternatives - if there are cheaper or better models, there is no advantage to our members in using an animal study."

This gradual process of reducing reliance on animal models is supported by legislation such as Europe's decision to outlaw animal research when alternatives are available, and by organisations like the UK's National Centre for the 3Rs (Replacement, Refinement and Reduction of Animals in Research), which Magee says "spends millions of pounds every year funding research into alternative technologies".

Where Bailey and Magee's views diverge is the speed at which the transition should be made. Bailey doesn't claim that alternatives aren't being used, rather that they are "not used nearly enough, and still take a back seat to animal use". Cruelty Free International, along with other organisations, would support an immediate ban on animal research, and believes that pharma companies and research organisations would be able to adapt to using alternative testing methods in the time it would take for a ban to come into force.

On top of legislative change and funding for alternatives, pressure applied by organisations like Cruelty Free International represents an important means of helping to ensure that companies and laboratories don't simply turn to animal testing as a default, and that alternatives are employed whenever they are feasible. There may well come a time when technological advances turn animal research into a relic of the past, but the current best practice employed in Europe seems a sensible compromise until that time comes, and other regions would do well to follow suit.

"In many ways the animal research 'debate' is a little bogus," says Magee. "It doesn't genuinely have two sides because nobody is suggesting we should always use animals, whereas there are those who say we should never use animals. Instead, researchers are in the middle ground, saying that only if an experiment has passed two tiers of ethical review is it ethical. The 'debate' asks us 'animal research: yes or no?', when we've only ever said 'maybe, if it's worthwhile, ethical and there's nothing better available.'"