My father suffers from diabetic peripheral neuropathy. His diabetes led to nerve damage that causes him severe, constant pain. I want the best medical treatments possible for him and, as a neurologist, I am always on the lookout for good, new drugs, but none of them have effectively slowed down his diabetes and nerve damage. As long as experimenters continue to try to recreate diabetes in animals, instead of studying human diabetes, I have little hope that my father's pain will end.
Although numerous drugs are available, diabetes remains among the top killers in the U.S. and worldwide. The newest drugs are generally no more effective than the older drugs or are much more harmful. Just recently, two new diabetic drugs, Onglyza and aleglitazar, failed clinical trials after testing in animals.
At first glance, it might seem that if we can recreate diabetes in dogs or mice, we would better understand diabetes. But here's the problem: we end up better understanding animal diabetes-- in dogs and mice-- but not necessarily human diabetes.
In this article in my medical research series, I discuss the second major reason (click here for the first reason) why animal experimentation is unreliable for understanding human health and disease.
Animals Don't Get Human Diseases
Not only do animal models fail to help us better understand human diseases, they often lead us down the wrong path of investigation. In 2006, the Diabetes Research Institute announced that after over thirty years of experiments on mice and rats, researchers discovered that the internal structure and function of the human pancreatic islet cell, which is central to the development of diabetes, are dramatically different from that in the "well-studied rodent". As one of the researchers stated:
We can no longer rely on studies on mice and rats. It is now imperative that we focus on human islets. At the end of the day, it is the only way to understand how they function.
The inability to recreate human diseases accurately in other animals is an inherent and fundamental flaw in the use of animal experiments.
Let's take a look at stroke experiments in animals to examine how they strike out. In humans, stroke is usually caused by the gradual narrowing of a blood vessel to the brain by atherosclerosis or by a blood clot that developed in another part of the body. Animals in labs don't naturally get strokes. Experimenters artificially induce strokes by methods such as clamping off major blood vessels in animals' brains or artificially inserting clots into their vessels. Here are the problems with this:
Strike 1: Artificially inducing stroke in animals does not recreate the complex physiology that causes the natural disease in humans, which may develop over decades.
Diseases are diseases in context. In humans, stroke is usually linked to pre-disposing conditions such atherosclerosis, high blood pressure, and diabetes.
Strike 2: Animal stroke models don't usually include the underlying conditions, which contribute to human stroke.
Experimenters try to recreate the underlying human conditions such as diabetes in animals. However, these underlying conditions are usually also artificially induced in animals, and as we saw with diabetes, are often wrong anyway.
Strike 3: Artificially inducing in animals the underlying conditions that lead to human stroke does not replicate the processes that occur in humans.
Recognition of each difference between animal models and human diseases leads to renewed efforts to eliminate these differences. But in trying to recreate the complex physiology behind the human diseases, experimenters try to reproduce the complex physiology of the underlying conditions, which are just as difficult to accomplish. Thus animal experimenters are continuously going around in circles.
Stroke is probably one of the easiest human diseases to try to recreate in animals. Yet, over 150 stroke drugs found effective in animal stroke models failed in humans (1).
Even when human genes are inserted into animals, the diseases that develop are still notably different from human diseases. This is because those human genes will be affected by all the genetic determinants and physiological mechanisms that are unique to those species. Animal experiments in Alzheimer's disease, multiple sclerosis, cystic fibrosis, and Parkinson's disease are perfect examples of this.
A study published in Science found that a crucial protein that controls blood sugar in humans is missing in mice. Even when the human gene that makes this protein was expressed in genetically altered mice, it behaved differently. In fact, it had the opposite effect from that in humans -- it caused loss of blood sugar control in mice.
The use of animal models not only confounds us with irrelevant data, they actually divert us away from unraveling the causes behind human diseases and finding treatments that work. As Susan Fitzpatrick, former Associate Executive Director of the Miami Project to Cure Paralysis said:
Even if we know all about the animal model, we don't necessarily know about the disease..."The model becomes what we study, not the human disease".
In the past, it may have been the best we could do to use another animal to understand basic human physiology and anatomy. But we are well beyond that now. Medicine now deals with the subtle nuances of biology and disease. Yet, experimenters continue to rely on faith that animal models accurately replicate human diseases. That's not good enough. Our loved ones who are suffering from illnesses deserve science, not faith.
Stay tuned. In the next article, I will discuss the third major reason why animal experimentation doesn't work: animals aren't little humans.
1. Macleod M. What can systematic review and meta-analysis tell us about the experimental data supporting stroke drug development? Int J Neuroprot Neuroregener 2005; 1: 201