A press release from Kansas State University (KSU) announces: “A 'game-changer': By studying animal health, researchers find improved ways for developing, testing cancer therapies.” Researchers Rowland and Troyer discovered a group of pigs with naturally occurring severe combined immunodeficiency syndrome (SCID). They then noticed that the pigs were prone to certain cancers such as melanomas and cancer of the pancreas. The press release continues: “The researchers call it a scientific achievement with huge potential to improve surgeries and drug development involved with cancer. ‘This could be a game-changer,’ Troyer said.” The press release then quotes Rowland as saying: “Pigs are closely related to humans anatomically and physiologically. This can have huge potential for human cancer research” including the development of drugs.
The press release continues:
While similar research has been performed with SCID mice, it has not adapted well to human cancer research. Rowland said there is about a 90 percent failure rate in translation of results from mice to humans. Research involving SCID pigs may be more beneficial to human cancer research because pigs are closer anatomically to humans. The research opens a variety of doors for both animal and human health research. It may now be easier for scientists to improve strategies for bone marrow transplants. They also have a better way to detect cancer drug side effects and test surgical interventions, Troyer said. “The potential is a little daunting because it is as if there is no horizon limiting possible ways to utilize this model,” Troyer said. “It is an opportunity for Kansas State University to be a leader in the field and to become a center for large animal biomedical research.”
The press release continues with the same hyperbole stating that the new model will also be valuable for other many areas of research.
The article describing the research is in the journal BioResearch Open Access. The following is from the article:
Animal models for cancer therapy are invaluable for preclinical testing of potential cancer treatments; however, therapies tested in such models often fail to translate into clinical settings. Therefore, a better preclinical model for cancer treatment testing is needed. . . . Immunodeficient pigs, which do not reject xenografted human tumors, have the potential to become an extremely useful animal model for cancer therapy because of their similarity in size, anatomy, and physiology to humans.(1)
I must admit the hyperbole of the press release and the article surprised even me and I read these things every day. I could write blogs for a month just based on what the article and press release say, but will try to limit myself to this one essay.
To begin with, the press release justifies all the mania because: “Pigs are closely related to humans anatomically and physiologically.” That is a meaningless statement. On some level of examination humans are closely related to every mammal. For that matter, we are related to stars and we are composed of their dust. No two people are more closely related than monozygotic twins yet even they do not always react the same to drugs and disease. Even a superficial understanding of evolution should cause the author to question such a statement. (Of course, considering the location of the university, maybe the researchers are creationists.)
Consider the following from Hicks of UC-Irvine. (Dr Hicks is a supporter of animal-based research.)
I think he [Ray Greek] makes a good point about CAMs [causal analogical models]. . . . Studying a rat will not tell you anything of specific information about a disease in a baby, specific information about the disease in a baby. So if you want to study babies, you study babies. Obviously. . . . But causal analogical models, Dr. Greek's completely right, it dominated physiology and biochemistry for 20 -- for the -- most of the twentieth century, most of the nineteenth century. It was the result of linear thinking, that if I study a dog, it's directly applicable to the human, if I study a rat it's directly applicable to the a human, if I study a frog it's directly applicable to a human. . . . Later on physiologists also began to learn that there was a problem in the '70s and '80s. You couldn't make one-to-one relationships. It wasn't quite possible. There are species differences.
I agree that physiologists did in fact learn that one to one relationships were not possible. They just didn’t care. You can read literature from the 1960s and find claims for animal models just like the one that came from KSU. Moreover, such claims can be read daily from animal-based researchers in the form of press releases or articles in the peer-reviewed, scientific literature. I have said many times that animal modelers are not stupid. But, I have never accused them of having any character.
Also note that the article simultaneously states, “Animal models for cancer therapy are invaluable for preclinical testing of potential cancer treatments” AND “however, therapies tested in such models often fail to translate into clinical settings.” In fairness to the authors, this is standard fare for researchers writing about animal models, but at least most are self-conscious enough to put a few pages between the outlandish claim and the denial of the outlandish claim. In this article, both are in the same sentence. In light of the fact that cancer treatments fail in preclinical trials more than almost all other drugs, it takes a lot of impudence to claim that the research that led up to the failure is “invaluable.”
Drugs in general entering Phase I trials have only about a 9% chance of eventually making it to the market.(2-5) Of the drugs that advance to Phase III, less than 50% are marketed.(6) The failure rate for oncology drugs is even higher.(6-9) Only 5% of cancer drugs that have an Investigational New Drug Application (IND) eventually go to market.10 Lack of safety or efficacy accounts for approximately 90% of drug failures during clinical trials (6, 11) and animal models are the basis for establishing both safety and efficacy. Not a modality that I would say is invaluable in light of the scientific facts. So why all the hyperbole?
The press release ends with the following:
The research also improves the study of zoonotic diseases, which are diseases -- like swine influenza -- that can be transmitted between animals and humans. By developing vaccines for diseases in SCID pigs, scientists can gain insight into human vaccine development. The university's Biosecurity Research Institute provides the ideal location for developing these vaccines, Rowland said. The scientists have performed research on a small scale and now want to build it up to a larger scale. They see possibilities for new research with the Kansas State University Johnson Cancer Research Center as well as cancer research partnerships and collaborations with the University of Kansas Cancer Center, especially with its recent National Cancer Institute designation. "Agriculture benefits the people of the state in so many ways," Rowland said. "While it includes jobs and exports, there is a human element that we sometimes forget. This is a good example of how we can take animal health research and all of a sudden it has the potential to help cure human cancer."
Once again we see researchers ignoring evolutionary biology and the fact that animals and humans are complex systems in order to panhandle for money. The fact that a pig reproduces some signs and symptoms of a human disease is not sufficient to expect the model to predict human response to drugs and disease. I have written books and articles explaining why this is the case but to put it succinctly, animals and humans are evolved systems that are differently complex. This fact, if one understands the science behind it, invalidates the use of animals as predictive models for humans.
Yes, animals can be used as a heuristic but then so can sailboats. Animals can be successfully used in many ways in science but those ways are not how the modality is sold to society. The use of animal models is accepted by society because society has been told exactly what the press release and article say: animal experimentation is THE way to find cures. It is a necessary evil.
In reality, it pays the bills.
The mindset of the biomedical research community in the US was aptly illustrated by an editorial in Science where Bourne of the University of California at San Francisco and Lively of Wake Forest School of Medicine compared a cut in NIH to the sinking of the Titanic.(12) No mention was made of the poor return on the funding dollar just the prediction that the world would end if funding were cut. Also mentioned was the idea that universities that employ scientists should pay them more in direct salaries and rely less on grants to pay salaries (indirect salary support). In the very next issue of Science, NIH director, Francis Collins pledged continued support of basic research, which he defined as: “systematic study directed toward fuller knowledge or understanding of the fundamental aspects of phenomena and of observable facts without specific applications in mind.”(13) Collins goes on to say that: “Since 2003, the proportion of NIH funds spent on basic research, defined in this way, has ranged from 53 to 57%, standing at 54% for fiscal year (FY) 2012.”
That claim is suspect as what a researcher claims is basic research on one day may be claimed as applied research or translational research on a different day, depending upon what circumstances dictate. The grant application may claim the research is applied or translational but when nothing of note comes from it, then the scientist can say that it was basic research all along so no one should be upset, as basic research is performed “without specific applications in mind.” (For more on this, see: Is the use of sentient animals in basic research justifiable?)
Collins is correct in stating that “most of the 135 Nobel Prizes won by NIH-supported scientists” were in the category of basic research. This raises four questions however. 1. Did the research eventually result in treatments for human disease? (See (14)) 2. In terms of model organisms, if the research used sentient animals was such use necessary? 3. What is the track record of model organism in general? 4. Are the Nobels that are awarded in Medicine or Physiology awarded based on interesting science or advances that actually benefit patients? (The answer is interesting science. Many of the advances that actually are responsible for how a physician practices medicine today were ignored by the Nobel committee.) One cannot cherry-pick the data and conclude that animal models are a good use of grant dollars.
The above must be contrasted with basic research in the physical sciences. The UK is currently setting limits on basic science research in the physical sciences (15) and in my view, that is a bad idea. Many important advances in science that did lead to medical advances came from the physical sciences. The physical sciences are not misleading like animal models are, they are less expensive to fund, and the quality of the research is better in my opinion (see here and here and here). If the UK wants to cut back on blue-sky research, animal models are the place to start.
1. M. T. Basel et al., Human Xenografts Are Not Rejected in a Naturally Occurring Immunodeficient Porcine Line: A Human Tumor Model in Pigs. BioResearch Open Access.1, 63 (2012).
2. FDA. (2004).
3. S. K. Sarkar, Molecular imaging approaches. Drug Discovery World, 33 (2009).
4. Editorial, Same old story? Nat Rev Drug Discov6, 97 (Feb, 2007).
5. S. M. Paul et al., How to improve R&D productivity: the pharmaceutical industry's grand challenge. Nat Rev Drug Discov9, 203 (Mar, 2010).
6. J. Arrowsmith, Trial watch: Phase III and submission failures: 2008-2010. Nat Rev Drug Discov10, 87 (2011).
7. Editorial, A stronger role for science. Nature reviews. Drug discovery10, 159 (Mar, 2011).
8. G. Caponigro, W. R. Sellers, Advances in the preclinical testing of cancer therapeutic hypotheses. Nat Rev Drug Discov10, 179 (Mar, 2011).
9. C. G. Begley, L. M. Ellis, Drug development: Raise standards for preclinical cancer research. Nature483, 531 (2012).
10. S. Kummar et al., Compressing drug development timelines in oncology using phase '0' trials. Nature reviews. Cancer7, 131 (Feb, 2007).
11. I. Kola, J. Landis, Can the pharmaceutical industry reduce attrition rates? Nat Rev Drug Discov3, 711 (Aug, 2004).
12. H. R. Bourne, M. O. Lively, Iceberg Alert for NIH. Science337, 390 (July 27, 2012, 2012).
13. F. S. Collins, NIH Basics. Science337, 503 (August 3, 2012, 2012).
14. W. F. Crowley, Jr., Translation of basic research into useful treatments: how often does it occur? Am J Med114, 503 (Apr 15, 2003).
15. A. Bhattacharya, Science funding: Duel to the death. Nature488, 20 (Aug 2, 2012).