Some animal activists argue human disease cannot be modeled in animals. They think physiological differences between species imply that treatments developed by means of animal research will not translate to humans.
Prediction through the development of models is no doubt a goal of scientific work. Predictions are the fruits of theories that can be tested experimentally. If a prediction is false so is the theory, and a new one must be generated based on prior knowledge and the specific way in which the data falsified the theory.
Unfortunately, those that claim animal models are not predictive of human response take some literary license in restating the above along the following lines:
Predictions, generated from hypotheses, are not always correct. But if a modality or test or method is said to be predictive then it should get the right answer a very high percentage of the time […]
If a modality consistently fails to make accurate predictions then the modality cannot be said to be predictive simply because it occasionally forecasts a correct answer. The above separates the scientific use of the word predict from the layperson’s use of the word, which moreclosely resembles the words forecast, guess, conjecture, project and so forth. […]
Many philosophers of science think a theory (and we add, a modality) could be confirmed or denied by testing the predictions it made.
This language delicately nudges one to equate different concepts, namely theory, hypothesis, modality and method. In this deceptively innocuous equation, resulting from either an honest misunderstanding or mischievous intent, lies the foundation to a seriously flawed argument.
Consider the domain of physics. Here, physicists put forward mathematical theories of some natural phenomenon which, in turn, generate testable predictions. If a prediction is falsified, so is the theory. When this occurs, scientists seek to understand how the data depart from the prediction and use prior knowledge and intuition to develop a new working hypothesis, which is embedded in a new theory.
Mathematics is the language of physics — its methodology. Obviously, by using mathematics one can create many different theories. The overwhelming majority of them will be false. Science is difficult because most of the time our ideas turn out to be wrong.
But one’s ability to conjure up large numbers of incorrect theories does not invalidate mathematics as a method in the physical sciences. Mathematics can in fact be used to arrive at accurate descriptions of how matter behaves. It makes no sense to describe this state of affairs by stating that mathematics (the modality) gets it right occasionally.
A similar situation arises in the domain of biomedical research. Researchers create models of disease in animals by trying to replicate what they believe are the essential components at play. These animal models can then be used to generate predictions for therapeutic interventions, which can then be tested in human clinical trials. If a prediction is falsified, so is that specific animal model of the disease.
When this happens, scientists seek to understand how the data depart from the prediction, what factors were ignored that might play a role, and use prior knowledge and intuition to develop a better, improved model. In the course of developing and refining such a model, scientists will go through many such cycles. A model is expected to be valid if and only if it captures all the key ingredients of the human condition.
The fact that one can postulate inaccurate animal models of human disease does not invalidate the whole methodology of animal research, it merely shows the work is difficult. But animal models can in fact be successful.
One of the proponents of the idea that animal research cannot be used to predict human response to disease is Dr. Ray Greek who was recently interviewed by Steven Novella for the Skeptics Guide to the Universe (as it turns out, Dr. Greek won a bid to appear in the podcast).
There is an interesting part of the exchange where Dr. Novella attempts to explain hat some models have indeed been extremely predictive of human response. Starting at 15:45min into the program he gives Dr. Greek the example of a how SOD1 mutant mice have helped in the treatment of ALS. The model “is a home run for humans with SOD1 mutation”, he said.
Dr. Greek’s reply was simply “Well, let’s face it. If you study 10,000 genetically modified mice there is bound to be one that you are going to hit a home run with.”
In the eyes of Dr. Greek and the animal rights activists that adhere to his views, the type hard scientific work that leads to the development of a predictive model of human disease boils down to a mere chance discovery.
Dr. Novella tries insists that such a characterization of animal research as not predictive is meaningless — it is as if one were to ask “Does surgery work?”. The answer, he says, is “of course, some surgeries work and some don’t, and you have to ask which ones work and for what […] You [Dr. Greek] want to make a final pronouncement for surgery as a medical intervention.”
But there is little hope of getting the message across.
Dr. Greek retreats to discussing toxicology testing and declares disease research to be, well… “more complicated.”
Dr. Novella appears to politely give up in frustration and rapidly moves on with the rest of his show.
Indeed, genetically modified mice have been and continue to be a very useful tool to dissect the roots of human disease and develop new treatments. This includes the study of type II diabetes using mice with mutations in the glucokinase gene, the shaker1 mouse as a model of human genetic deafness, the role of genes in inherited psychiatric disorders, in cancer research in general and for the development of successful new therapeutics for breast cancer in particular, in the advance of new treatments for lupus, and Duchenne muscular dystrophy, and so on.
The Nuffield Council on Bioethics has a full chapter dedicated to how genetically modified animals are used in the study of human disease.
It is absurd for anyone to claim such advances are the product of chance. They are the product of the hard work of dedicated individuals who spend countless hours in laboratories around the world with the goal of advancing the well-being of those affected by disease. They are the product of those that go to bed thinking about how a protein may work, why muscles may weaken, how a tumor spreads, or why memory fails, in the hope of waking up the next day with some new ideas. They are the product of those that are determined to solve some of the most complex puzzles of biology that afflict human kind. They are the product of talented students, staff and scientists that together work to rid the world of disease. They are the product of science.