The reasons I focus on drug development when assessing the predictive value of animal models include the fact that drug development offers direct animal to human comparisons, there is ample published data, there are many scientists involved in drug development that are saying one of their biggest problems is the lack of predictive value of animal models, and most medical research is ultimately aimed at intervening in a disease via a drug. Research designed to lead to new drugs is a very large percentage of animal-based research. If animal-based research is not working for drug development, the paradigm should be questioned. When the paradigm is analyzed, we find that it violates basic scientific principles from complexity and evolutionary biology. Under normal circumstances, this would be enough to mandate a change in scientific disciplines that use animal models.
A press release issued by the American Chemical Society (ACS) addresses a related issue: “How many breakthrough new drugs never reach patients because tests in clinical trials suggested a high risk of liver damage when the drug actually was quite safe?” This is an aspect of animal modeling that I have brought up before. It’s not just that animal models fail to identify harmful drugs, they actually prevent good drugs from coming to the market. The ACS national meeting addressed the issue of liver toxicity. If animal models do not reveal liver toxicity, a potentially harmful drugs may go to market. On the other hand, if animal models reveal liver toxicity in a drug that would be safe in humans, society has lost the benefit of that drug. Needless to say, animal models offer no predictive value for assessing liver toxicity.
It turns out, the historically relied on blood tests for human liver toxicity may also be misleading. Physicians have employed blood tests that measured levels of aspartate aminotransferase (AST) and alanine aminotransferase (ALT). These chemicals are released into the blood when liver cells die. However, recent research revealed that sometimes the AST and ALT could be elevated without permanent damage being done to the liver. Animal models are unable to guide drug development in terms of toxicity or efficacy. But as I have stated many times, Pharma needs predictive tests in many areas. It appears that even tests that were historically relied on can be misleading. This was learned from studying humans.
Along similar lines, a recent article in Salon addressed research on cancer. “Cancer research in crisis: Are the drugs we count on based on bad science? A paper finding that important experiments can't be reproduced suggests new drugs could be based on bad science,” was written by Jalees Rehman, MD. Rehman refers to an article by Begley and Ellis  that revealed scientists at Amgen “were unable to replicate the vast majority of published pre-clinical research studies. Only 6 out of 53 landmark cancer studies could be replicated, a dismal success rate of 11%!” Rehman then correctly states, “Reproducibility of research findings is the cornerstone of science,” and continues:
The lack of reproducibility in pre-clinical cancer research has a significance that reaches far beyond just cancer research. Similar or comparable molecular and cellular experimental methods are also used in other areas of biological research, such as stem cell biology, neurobiology or cardiovascular biology. If only 11% of published landmark papers in cancer research are reproducible, it raises questions about how published papers in other areas of biological research fare.
This lack of reproducibility of animal studies is being addressed by a coalition of scientists. See our paper Systematic Reviews Of Animal Models: Methodology Versus Epistemology for more on this attempt and why it is destined to fail.
On March 11, 2013 in Madison, WI, Robert Streiffer, a UW professor, and Rick Marolt discussed the use of animals in research. Streiffer responded to criticisms of his presentation and included the following:
The first key point I was making is that there is a distinction between knowledge and benefit per se and that research, in the first instance, produces knowledge, not benefit. The phrases “in the first instance” and “per se” are important here. While knowledge often leads to benefits, research can produce knowledge without producing any benefit at all. However, this conceptual point does not imply that research that does not produce benefits is not valuable, as there is also a distinction between the concept of a benefit, which I take to be an improvement in individual well-being, and the more general concept of value. It is thus perfectly consistent to say that something is not itself a benefit and that it does not lead to benefits while maintaining that it is nonetheless valuable. Indeed, I explicitly stated that there are significant kinds of knowledge worth spending a fair bit of money on even if they do not actually improve anyone’s well-being. . . . However, the second key point I made is that, when one is evaluating research that harms and kills animals, the moral threshold is higher than it is with other academic pursuits: that kind of research cannot be justified merely on the grounds that it produces knowledge if that knowledge is “totally unrelated to anything practical” (by which I meant “totally unrelated to improving individual well-being”).
On September 20, 2013, Erika Trent wrote an article in MIT’s newspaper, The Tech, titled: Scientific discoveries have compensated for the pain we impose upon research animals. I do not know whether Trent wrote the title, some papers write the title themselves instead of the author writing it, but the title is telling. Scientific discoveries are different from medical relevant discoveries. Trent begins by recounting classroom dissection. I have noted many times that dissecting animals can provide the student with a general idea of the normal anatomy and physiology of the species in question as well as the general contours of vertebrate or mammalian anatomy.
But Trent then changes topics yet again (from scientific discoveries in the title to classroom dissection to medical research). Trent: “In fact, the majority of major medical breakthroughs over the last century wouldn’t have happened if not for these animals.” Trent counts among these advances the vaccines against smallpox and tetanus, “the development of antibiotics and insulin and the advancement of organ development technology.” And that is only scratching the tip of the massive, and continually expanding, iceberg.” Wow! That is apparently a big iceberg. But Trent supplies absolutely no evidence for these positions. This is common in such essays whether students or professional vested interest groups or lobbyists write them. Finally, Trent states: “Laboratory animals offer us the gift of knowledge.”
Make no mistake, animal modeling, dissection, and other animal-based endeavors generate knowledge. But if society ever learns how little benefit there is in animal modeling, it will end.
Image courtesy of aopsan / FreeDigitalPhotos.net.
Last weekend I lectured at Vancouver, BC and the University of British Columbia. I want to thank all the people that helped out but unfortunately cannot remember every name. I can remember, and want to thank Anne, Bette, Caroline, Darren, Karl, Laura-Leah, Lisa, Lois, Stephen, and Wilson. I had a wonderful time and met a lot of great people.
1. Begley, C.G. and L.M. Ellis, Drug development: Raise standards for preclinical cancer research. Nature, 2012. 483(7391): p. 531-533. http://dx.doi.org/10.1038/483531a