Before I start, I have to declare that I am a scientist and my view on this matter is based on evidence. Some would consider that I am biased, but for those who are willing to keep reading, I hope I have managed to remain as neutral as possible.
The debate over whether animals should be used for drug testing and for the progression of science has lasted for many many years. So, have we really reached a point with our current technology to advance forward without using animals?
It is often claimed that non-animal models are available for the testing of new drugs. This statement is true, but only to a certain extent. Many drugs that are used on animals are often tested in a number of non-animal systems beforehand. These include human cells in culture, computational analyses and predictions, microbial cultures and artificial matrix that somewhat mimic specific tissue types. The problem is that these non-animal systems are even more different to human than the animal models.
Let’s look at a couple of examples in a little more detail. One might think human cells in culture must be similar to cells that are in humans, after all, they came from the same species. However, this is not true. Cells have to be cultured in a variety of culture media that are quite different to their native surroundings to keep them alive outside of the body. These change the characteristics of the cells and the fastest growing cell will expand and take over the culture, essentially annihilating the slower growing ones. In the normal human body, the rate of cell turnover can be quite low depending on the organ and some cells hardly grow at all inside the body. Furthermore, one key component has been difficult to establish in culture. The blood flow. This is responsible for the transport of a drug to its target organ or its removal and it is well known that it has a significant effect on the dosage and effectiveness of a drug.
What about computational analyses? Can we predict if a drug is going to work by using softwares? This is a difficult question to answer. Whilst it is certainly possible to analyse if the chemical structure of a drug has the potential to bind to a certain target and whether the drug has off targets, it is difficult to know whether the drug will actually work until you test it. Many biological factors come into play to determine if a drug will work. First of all, is whatever the proposed drug targeting actually a good target? In theory, a number of proteins activated in cancer cells can be targeted, but one must not forget that these same proteins are also in normal cells and the drug itself will not “know” if it is attacking the correct tissue. Secondly, how fast is the drug removed? If it is degraded too quickly by the body, then it will require a higher dose, but that can become extremely toxic. It is a fine balance that one must test in order to find out.
Scientists are fully aware that animal testing is not flawless either. Rodent response can be very different to humans. Take diabetes for example. 11mM blood glucose is considered indicative of diabetes in humans, but that is actually the normal level of glucose in mice and that a much higher level of 30-40mM glucose is required to determine if a mouse is diabetic. Consequently, how the cells handle glucose can be very different between these concentrations. Many disease models are simply not very good representations of human diseases and many cancer drugs fail because for many years a reliable cancer model was not developed. For example, implantation of human cancer cells onto the back of immunodeficient mice was a popular model, but it is not a good representation of human cancer (despite a number of anti-cancer drugs have been developed this way). However, in recent years, the orthotopic models have become more popular due to the fact that these are more closely associated to the human cancers.
It is frequently quoted that “90% drugs tested on animals fail” (British Union for the Abolition of Vivisection and “92% of drugs fail in clinical trials, having successfully passed through animal studies” (Safer Medicines Trust). The claim is that better alternative non-animal testing methods are available. However, 94% of drugs that pass animal AND non-animal preclinical tests will fail in human tests. But what the animal testing is very good at doing is to identify if a new drug entity has the therapeutic potential and also safe enough to proceed forward to clinical trials. Considering that there are more than 100 phase 1 clinical trials that take place in the UK every year, each involving multiple individuals, there has not been a major incident for the past 30 years (the Northwick Park trial incident is obviously an exception to what I just said). Of course, without a controlled experiment, we will not know if many more trials like Northwick Park would have occurred without animal testing. But it is probably reasonable to speculate that drugs with more severe side effects are less likely to be eliminated if tested directly on humans with only the non-animal testing beforehand.
Of course, some disagree with animal research purely because of ethical reasons. Animal cruelty is often an issue. This is what ventures into personal believes, subjectiveness and there is no right or wrong answer to this. A common argument put forward is that no living beings should be bred and used solely for experimentation. This is not something I can argue against, as I would be trampling on the feeling of those who feel compassionate towards animals. Though I work to pursue knowledge and advancement in our understanding of science. As already discussed, it is impossible at this moment in time to understand something completely using only cells in a dish and computational analyses. So, the options are limited to scientists. The vast majority of biological and medical science researchers use a number of tools combining computational, cells in culture as well as animal models to answer a specific question. Anti-vivisecitonists portray researchers as these cruel human beings who kill without a second thought. This is simply not true. Whilst research was done on the animals and that a degree of suffering has been inflicted, appropriate steps are in place to ensure the well being of these animals is maintained wherever possible. If an animal under my care is sick (with or without experimentation), I will contact the vets, go in at whatever time called (evenings, late nights, weekends) make sure the animals are fine just like you would with a pet. Animals are also looked after by a team of fantastic technicians who work throughout the week (yes, including weekends) to make sure they are well fed, the cages are clean and make sure nothing goes wrong.
What about using humans for experimentation? So, only those who are aware of the situation and given written consent will participate in an experiment. This certainly does sound more ethical. But only to a certain extent as well. Whilst some tests can arguable be done on humans, e.g. imaging using X-ray and MRI or minimally invasive procedures such as blood and urine sampling for certain diseases can be done on humans, some are simply not possible. Diseases caused by genetic variations and mutations, how does one start looking for this and recruit volunteers without first identifying the entire population and make a record of every individual’s genetic make up? For example, if 1% of a population carries a mutation of interest, then in theory, you will have to recruit 100 people to get 1 suitable individual. Often computational or experimental genetic analyses require hundreds of samples when using human patients, so 100,000 individuals will have to volunteer and screened before researchers can narrow down who is suitable for the study. For ethical and health reasons, many strict criteria are in place, so some would not be eligible to even be considered for certain studies. These include pregnancy, gender, age and known underlying diseases which may affect the study. So it is quite clear why it would be difficult to start a human volunteer recruitment without any knowledge as to why a genetic mutation should even be looked at in the first place. This is another reason why animal testing is important, because mice with specific mutations can be generated and a much lower number is required (between 5 and 12) to identify an effect of a mutation, thereby justifying whether some of these ought to be tested on humans.
Is it ethical that we administer a drug with little information on its side effects directly to humans even if consent was given? Clinically, A much lower dose than the anticipated therapeutic dose is first given to volunteers to identify if a drug can be tolerated without severe toxic effects. But essentially, where to start on this “lower” dose will be difficult to determine without pre-clinical testing using both cells in culture and on animals. Hypothetically, a drug shows a good safety profile on cells in culture and skip the animal testing stage. It is tested directly on humans where consent is given, but causes organ damage, internal bleeding, severe pain and in the worst case scenario, death, what do we do then? Do we then turn around and say it is simply an unfortunate event? Or is it because not enough precautions were taken and tested for before humans are used?
I do no intend to persuade others that animal testing is good and there is nothing wrong with it. But I do believe some of the statistics and argument presented by certain individuals are misrepresented and taken out of context. In reality, it is not so black and white. If this post makes you think of some of these things a little bit more, consider the pros and the cons of animal testing, then this post has achieved its purpose.