This is a story of mice and men. It is not a happy story – not for us, and not for them. It is estimated that about 115 million animals are used annually in biomedical research and drug testing around the world, costing hundreds of thousands of dollars for every substance tested. Most of these millions are rodents – rats and mice. Rabbits and guinea pigs are also commonly used, and smaller amounts of dogs (mostly beagles) and non-human primates.
Medical research is often considered to be somewhat of a grey area when it comes to animal ethics. If a few animals have to die, but we end up finding a wonder drug that saves millions of human lives, then surely the benefit outweighs the cost, right?
The problem is it is more than just a few animals – it is millions of them – and the number of effective drugs actually being approved each year to treat human diseases is slim compared to the numbers of drug candidates initially tested in animals. The chances of animal experiments actually benefiting humans of often wildly overestimated.
In 2018, 59 new drugs were approved by the Food and Drug Administration (FDA), but scientists say the commercial potential of these new drugs is “lacklustre” (1). Although hundreds of drug candidates progress through animal studies to enter clinical trials in humans each year, only about 10% ever penetrate the market, and truly impactful drugs that save human lives are depressingly rare.
About 25% of these drugs approved in 2018 were for cancer treatment. We are nowhere near curing cancer but we have an army of drugs to treat it. The majority of these drugs improve patient survival by maybe two or three months, only benefit a small minority of patients at an exorbitant cost, and do little to improve quality of life. Our obsession with progress in drug development is fuelled by a culture of medical excess. What makes us so eager to sacrifice years of labour, billions of dollars and thousands of animal lives to achieve even the faintest glimmer of clinical promise?
Much of what is called scientific or medical “progress” may not in reality improve life but rather delay death. To me, delaying death in one species by sentencing death on another, does not seem to add up. Are such drugs truly worthy of the thousands of animals sentenced to death for their development?
What about the thousands of drugs which fail during the transition from bench to bedside? Are they worth the millions of lives sacrificed for them?
In cancer research, the average rate of successful translation from animal models to cancer trials in humans is less than 8%. Of the drugs that make it into clinical trials, an even smaller fraction will make it through to achieve marketing approval (2).
The idea that animal experimentation is “taking one like to save another” is extremely flawed. In animal research we are rarely, if ever, presented with the stark situation in which we can save the life of a human by taking the life of an animal (3).
Why do so many drugs fail in Clinical trials?
The main reasons for failure in clinical trials are lack or efficacy and lack of safety – either the drug doesn’t work well enough, or it causes some harmful side-effect that detracts from any clinical benefit it may possess.
90% of drugs entering clinical trials in humans fail to make it through, and many so-called “silent failures” happen before human testing even begins. This is the pre-clinical stage, where animals come into the picture. Before we can give a drug to a human – which will begin with a small group of healthy volunteers – scientists are required to first prove its safety in both rodent and non-rodent animal models.
All major reasons for failure in clinical trials can be attributed to the use of animal models, which due to inherent genetic differences, do not accurately reflect safety or efficacy endpoints in humans.
Do they suffer?
We know that animals feel pain. In fact it is quite likely that their experience of it is stronger than ours. Like us, animals have evolved the capacity to feel pain as a warning when an action is putting us in danger. “Don’t do that again” it says. Richard Dawkins – the great evolutionary biologist – suggests that those animals which have lower capacity for memory might actually require a more intense warning signal to deter the action.
“We have erected a moral wall around our human species”, Dawkins says. How is it that a human embryo, even in its very earliest stage, even before it has developed a nervous system, is somehow more worthy of our moral consideration than an adult chimpanzee?
We humans like to think of ourselves as the most intelligent beings, often using this point as justification for the desperate measures we take towards human betterment. But consider this: the only reason humans are the most intelligent creature is because the concept of “intelligence” was created by humans. We picked out a few things our brains happen to do better than other animals, bundled them all together and called it “intelligence” and told ourselves that was what made our race great. Intelligence does not equal superiority. Our special blend of cognitive function does not make us any more worthy of life. It is difficult to argue that the human race, which in recent decades has inflicted irreversible harm on our planet, is still somehow more worthy to live here than any more peaceful branch of the evolutionary tree.
Maybe other animals are “less intelligent”; maybe they are “not quite like us”. The notion that animals are not like us is true – and this should be used as an argument against animal research – not supporting it. The quest to understand and treat human diseases by experimenting on animals which are not like us and which don’t even naturally contract most human diseases creates misleading and erroneous results which do more harm than help to human healthcare.
Some of the same arguments we make in justification of animal experiments could similarly be used to justify the practice in the severely disabled, the mentally ill, children, prisoners and people of colour.
Today we look back at our slave-owning ancestors with horror. It is absurd to us to consider such an immoral system, where treatment of other humans is based solely on the colour of their skin. But what will future generations think of us?
Ethics committees, which decide whether an animal experiment is justified, may consist of “intelligent humans” but they do not consist solely of vegans. I can’t imaging that anyone who thinks it is somehow morally justifiable to end an animals life simply for the pleasure of taste, can possibly evaluate whether the outcome of an experiment truly justifies the means.
The Three R’s: Replace, Reduce, Refine
The use of animals in scientific research is guided by three basic principles:
- Replacement of animals by alternative tools where available
- Reducing the number of animals used to a minimum
- Refining the conditions of use to improve animal welfare and reduce suffering as much as possible
The Three R’s have been embedded in international legislation and regulations for more than 50 years, but they are loosely defined and loosely enforced. Today, with promising new tools like organ-on-chip models burgeoning in this space, replacement needs to be at the top of the agenda.
Should we be getting to human trials sooner?
There is a concept called “microdosing” of “Phase 0 trials” which has recently been implemented in some drug discovery projects to reduce the high rate of failure and associated high costs. Phase 0 involves testing the drug in humans at a very tiny dose early on in the drug development pipeline (4) .
At the small doses used, avoiding risks of toxicity and sparing costs of large scale production, microdosing would give us an idea of how the human body processes the drug and whether it is effectively hitting its target in the right place in the body. Scientists can monitor the tiny dose as it passes through the body using highly sensitive imaging techniques, to see for example whether the chemical passes across the gut wall, how long it lasts in the body and which organs it distributes to, how it is broken down, and how it is excreted.
Collectively these processes make up what is called the drug’s pharmacokinetic properties. Normally, pharmacokinetics is first assessed in animals, but animal models are often poor predictors of pharmacokinetics in humans, and can obscure interpretations of how effective a drug will actually be.
Lost in Translation
Let’s for a moment imagine a world where animals don’t suffer, where they are merely dolls with robot hearts and unfeeling brains. Separate from any moral predicament, the use of animals in research is still, I believe, unjustified. Not only is it extremely costly to breed and home these genetically manipulated beings, but they don’t actually work very well. The ability of animal models to predict human drug responses is tenuous at best.
Findings which appear highly promising in animal studies most often fail to translate to human trials and rarely make a dent in human clinical practice. On the other hand, we can assume that many drugs which are abandoned after fruitless studies in animals, may actually have held some clinical benefit. This is true across a wide range of disease areas (5).
A systematic review published in the British Medical Journal in 2007, compared all animal data and human data collected for six different treatments across a broad spectrum of diseases (6). For 50% of treatments, the authors reported major discordance between the results form animal and human experiments. In other words, the animal data was no more likely to predict whether these treatments would be safe and effective in humans than a toss of a coin.
It is important to understand that this is not solely an ethical debate for animal welfare, but for human welfare too.
Mouse and man share only about 80% of genes. It is no wonder that they replicate complex biological endpoints with such limited accuracy.
The paradox is that the more similar an animal is to humans in its physiology, the more useful a model it is to address biomedical problems, but also the more likely it is to experience similar mental states.
“The more justified the use of a species is on scientific grounds, the less justified it is on moral grounds”, says Hal Herzog, author of the book “Some we love, Some we hate, Some we eat”. Herzog, an anthrozoologist and psychologist, has written extensively about the complex psychology of human animal interactions, including the moral disconnectedness in our attitude towards animal research.
“But remember that one time when something good came out of it?”
Considering the sheer quantity of animals being used, animal research will inevitably stumble upon a few handsome successes simply by chance. Communities involved in animal research continue to back claims for its indispensable value with the same handful of selected instances, where use of an animal model once upon a time produced some significant clinical benefit. Here is a table provided by the National Institute of Health (7). These seven feeble examples are feats we accomplished decades ago.
I would love to see some examples from the last decade, during which about a quarter of a billion rodents, and lesser millions of other animals have been created, used and killed for medical progresses that were never fulfilled. It more current examples were out there, they would be the ones being advertised.
This report attempts to justify animal experimentation with cringeworthy arguments like this one:
“Scientists study animals because they are a lot like people when it comes to basic body functions like breathing, eating, hearing, and seeing.”
Oh look! Here’s a thing that breathes! That must mean surely that it is similar to a human in every meaningful way – surely every molecular pathway and enzymatic reaction going on inside its body is carried out exactly like ours. What’s more, it even eats, sees and hears! Indeed, we are essentially twins.
These instances of successful translation are few and far between. Meaningful genetic and epigenetic differences between man and rodent, and even non-human primates, gives rise to interspecies variation in molecular pathways and metabolic processes which control a body’s response to a drug.
How we process drugs largely comes down to enzymes – the biological catalysts which break drugs down, inactivate them, and in some cases activate them. Often, mouse enzymes differ from ours in small but important ways. Rodents express more of particular enzymes and less of others. They differ in their rates of activity and how strongly they bind to drugs and mediate their effects.
This is not “animal testing in the name of scientific progress”. It may even be “animal testing in the name of scientific detriment”.
In some cases, animals die as direct result of the experiment. For example in drug lethality testing, substances are forcibly fed to animals at increasing doses, for the purpose of calculating the amount which will kill 50% of animals, called the LD50.
However, most experiments do not require the animals to be killed. Many experiments do not in themselves involve pain, and lab animals are usually housed and treated with much higher standards of care than animals raised for food. But even if no suffering occurs during the experiment itself, what happens afterwards?
Once all of the desired data has been collected, most animals will meet the inevitable fate of euthanasia, usually using Carbon Dioxide (CO2).
The familiar oxymoron “humane killing” is a pervasive term in these circles. Even if a “kind murder” did exist, CO2 would not fit the definition. Studies have found that animals exposed to even low concentrations of CO2 release high levels of the stress response hormones adrenaline and noradrenaline.
Moreover, CO2 causes what is known as “conditioned place aversion” in rodents – they learn to avoid places where they have been previously exposed to CO2. They will avoid CO2 even if it means missing out on a highly desired pleasurable food reward or exposing themselves to other painful stimuli in non-CO2 environments (8). This indicates a strong negative emotional response, and coupled with elevated stress hormones, is clear evidence that this method of euthanasia is for animals a distressing death.
Giving Animals Human Diseases
Many of the diseases which are studied using animal models do not even naturally occur in animals. Naturally, rodents rarely develop most common human diseases like cancer, heart disease and diabetes. To study such diseases in them, and to investigate the effects of drugs, the disease first needs to be created in the model.
In cancer research this is done by genetic manipulation; for heart diseases, diabetes and obesity, a mixture of genetic manipulation and dietary intervention is used. These methods do not adequately reflect the complexity of these diseases in humans, in whom the process develops naturally and over a much longer timeframe.
Forcing the most dreaded of human diseases like cancer on innocent animals from birth to death, and pumping their bodies with massive doses of experimental drugs, is not something any sentient being would consent to if he had a voice.
What do the World’s Leading Academics have to say?
In 2015, Oxford University’s Centre for Animal Ethics, in partnership with Cruelty Free International, commissioned a report on the ethics of animal use in research entitled “Normalising the Unthinkable”. An international group of academic experts, ethicists and scientists came together to discuss their views on the subject. The report highlights the need to denormalise and deinstitutionalise animal research (9).
“The deliberate and routine abuse of innocent, sentient animals involving harm, pain, suffering, stressful confinement, manipulation, trade, and death should be unthinkable. Yet animal experimentation is just that: the ‘normalisation of the unthinkable’”.
Doing harm to animals is especially difficult because animals cannot give or withdraw their consent. They cannot vocalise or express their own interests. They are vulnerable and relatively defenceless.
For a summary of thoughts from the Oxford Centre, have a look at this short film.
Emeritus Professor of Veterinary Medicine at University of California, Davis, Dr Nedim C. Buyukmihci, expresses his views at the Oxford Summer School for Animal Ethics:
“We do to other animals things that we would find abhorrent if we did them to each other. We have to ask ourselves: “Why is it that we do not do these things to each other?” Because from a scientific perspective, we could learn a lot more, a lot more quickly, and with far greater benefits for ourselves, if we used people. All the experiments we do now in other animals – we don’t do these things in people because we say they have certain values and certain traits that we want to protect. My argument is that other animals – particularly mammals, for example nonhuman primates – have exactly the same traits and we cannot do things to them that we’re not willing to do to ourselves.”
Professor Stephen F. Eisenman from North Western University highlights the personal and psychological problems which permeate these settings. He says scientists are guilty of what he calls “splitting” – animals seen as pets to be loved and cherished are psychologically split into a separate category from those seen as tools to be experimented on in whatever way humans desire.
“Getting scientists to recognise that the dog you have in your home is no different from the dog you’re doing research on in the laboratory, is a very hard task”.
Why is it that we react with visceral disgust when we see a man kicking a dog in the street, but not when we see a man in a clean white coat injecting a toxic substance into a mouse, or even a beagle, in a clean white laboratory?
It seems the human race is suffering from a case of Moral Schizophrenia – a term coined by American legal Scholar, Gary Francione, to describe humanity’s ambivalent relationship with other animals. The term describes the conflict between the stated desire not to cause animals unnecessary pain and suffering, and the continued support of practices that cause pain and suffering that cannot be regarded as “necessary” in any normal sense of the word.
New ways of doing things
Mouse may not be quite like man, but he is much more like man than a layer of cells in a petri dish. Mouse has a mouth, a digestive system and blood vessels. He has a liver that breaks down drugs, and two kidneys that flush drugs out. He is almost a perfect match, but not quite.
Thankfully, many in the scientific community are increasingly recognising the need to replace animal models with alternative tools – not only for ethical reasons but in order for science and medicine to flourish. Classical cell culture will not replace animals models, but it is no longer the only alternative we have.
During the past two decades, a number of impressive technological advancements have emerged which promise to replace failing animal models. These novel systems include human organs on chips, miniature organoids created using human stem cells, and computer models of disease. As these new methods become more physiologically relevant, we might envision a future of research free from animal experimentation.
The scientific community needs to shift from “progress-oriented” to “success-oriented” research. Success-oriented research requires human-relevant models that animal experimentation can never provide.
Animals are not instruments, objects, commodities, machines or tools. They have not been placed on Earth for us to breed, use and discard. They have complex interior lives, experiences and emotions that deserve respect.