This blog is a little sciency, but give it a try anyway! It’s less than 1000 words.
There are three areas of science one can examine that cast strong doubt on the concept that animal models can predict human response: evolutionary biology, complex systems, and the empirical evidence from the practice of human medicine. I will very briefly outline what can be learned from these three areas. For an in-depth analysis you will have to read Animal Models in Light of Evolution. (Sorry that it always boils down to that but judging scientific concepts does require an appreciation of the facts.)
In this book we will thus discuss the following propositions:
1. When the animal model community discusses the use of animals in research they give the definite impression that such results have been and will be translated directly to humans. (The community here includes those who use animals as models for humans, their employers, those in the press who support their activities, and so on.) Some theorists in these debates acknowledge the difference between basic and applied research but even these commentators often encourage belief in the predictive utility of animal models with respect to translational research. Interestingly enough, animal welfare activists often buy into these claims about the predictive utility of animal-based research, hoping (with varying degrees of disingenuity) that animal-based research can be replaced by non-animal methods that work just as well. We will argue here that they should be careful what they wish for given the actual predictive track record of animal-based research.
2. Animal models are not predictive for humans, indeed even different humans respond differently to drugs and disease, for many reasons. We will discuss some of these reasons and we will examine the meaning of the word predict—a matter that calls for attention if only because it has acquired a semantic shiftiness that makes its usage highly susceptible to equivocation in public discussions of these matters.
3. Animals can be used in science in many endeavors that have little or nothing to do with prediction. Animals can be used as bioreactors, for the study of other animals of the same species or strain, as an aid in learning and so forth. Clearly, one can obtain much important basic scientific knowledge that may or may not go on to be important in the study of human disease. It is here however that we again criticize the animal model community. There are indeed important connections between basic biological research on animals on the one hand, and human medicine on the other, but these connections are typically much more distant and indirect and suggestive than those engaged in predictive animal modeling tell the public and their policy makers.
a. Organisms belonging to different species or even different strains of the same species may manifest different responses to the same stimuli due to:
i. differences with respect to genes present, and also with respect to the versions (alleles) of genes present;
ii. differences with respect to mutations in the same gene (where one species has an ortholog of a gene found in another);
iii. differences with respect to proteins and protein activity;
iv. differences with respect to gene regulation;
v. differences in gene expression;
vi. differences in protein-protein interactions;
vii. differences in genetic networks (robustness, pleiotropy etc);
viii. differences with respect to organismal organization (humans and rats may be intact systems, but may be differently intact);
ix. differences in environmental exposures; and last but not least
x. differences with respect to evolutionary histories.
These are some of the important reasons why there are species differences with respect to the response to drugs and toxins, and why different species (and strains of a given species) experience different disease states.
b. Even nearly identical organisms (e.g., chimpanzees and humans in some debates, or monozygotic twins in other contexts) may respond differently to drugs and experience different diseases. These observations serve to sharpen the prediction problem with which we are concerned.
c. Current biomedical research is studying disease and drug response in ways that have uncovered reasons, even at the molecular and cellular levels of description where very small differences between organisms (be they members of two different species or members of different strains of a given species) become highly significant in the generation of responses to the same stimuli. Hence, by the standards of our current best biomedical sciences, using animals (e.g. mice) as predictive models for human disease and drug testing is highly questionable from a scientific point of view.
d. Immense empirical evidence supports this position.
To put the above another way:
Because all vertebrates are examples of robust complex systems (on many levels) which have emergent and nonlinear properties, and because of differences in genes, proteins, gene regulation, gene expression, protein-protein interactions, gene networks (vis-à-vis evolutionary and molecular biology), and environmental exposures, we now understand why even two nearly identical complex systems (say a chimpanzee and a human, or even monozygotic twins) may respond differently to the same stimuli, and hence why one complex system/species, cannot predict responses for a different complex system/species. Current biomedical research is studying disease and drug response at the level where the differences between complex systems (be they 2 different species or 2 different humans) manifest, hence using animals (e.g. vertebrates) as predictive models for human disease and drug testing is a scientifically invalid paradigm.