The results of an experiment at UC-Irvine (in association with UC-San Diego and Harvard University) are being widely reported in the press and are representative of my accusation that animal-based breakthroughs are reported and sold to society as predictive for humans. From the Orange County Register
A research team including UC Irvine made nerve connections regrow in mice with spinal cord injuries -- a scientific first that could lead to restoring function to paralyzed limbs in humans.
Need I say more? The only way such a claim could be made by reasonable people is if there is in fact a history of results from animal models predicting human response. If there is not such a history, then the above claim is nothing more than the equivalent of saying: “Gosh wow Goober, we really, really hope this results in better treatment for humans!” Well, don’t we all? I also hope that when a fortuneteller says a person is going to win $1 million in the lottery that he does in fact win $1 million. Of course the likelihood of that (we are back to those pesky statistics that so plagued a previous blogger-critic) is so low that no reasonable person would put any stock in what the fortuneteller said. (See Animal Models in Light of Evolution for more on the science of why animal models are not predictive for drug and disease response.)
Lest the reader think this particular research might be different from all the other uses of animals to predict human response, lets examine the research in more depth.
1. The scientists deleted a gene in mice that allowed the nerves in the spinal cord to regenerate. What does past experience tell us about gene manipulation in mice predicting human response?
Oti and Brunner Clin Genet 2007:
This situation is further complicated by the fact that there is not a one-to-one relationship between syndromes and genes. Typically, multiple syndromes can be caused by mutations in the same gene, and a single disorder can be caused by mutations in different genes (5). The existence of phenotypic variability is not surprising given that genes work in concert to form and maintain the human body. Different alleles of different genes in different individuals integrate differently with each other to create different final phenotypes. This is the basis of human phenotypic diversity and an important factor contributing to the fact that no two individuals are identical . . . (Oti and Brunner 2007)
Van Zutphen states that animals can be used to find out which gene causes a disease in humans but even he goes on to say:
However, results to-date suggest that the predictive value of a candidate gene, established in such an animal model, is rather low: thus far only few genes have been allocated as causative factors for corresponding disorders in man. In fact, it can be questioned whether the use of animal models is the most effective way to detect candidate genes for complex human disorders. Due to the complexity of the genotype-environment interactions, the pathways that lead to an aberrant phenotype often differ between man and animal. (van Zutphen 2000) (Emphasis added.)
The importance of context is also illustrated by studies of the effects of “knockouts” of specific genes in mice, a method that completely eliminates the function of a gene’s product. For example, knockout of a retinoblastoma-related gene causes severe abnormalities and embryonic death in one strain of mice, but the same mutation in another strain has no effect. The mutant mice are viable and become fertile adults, as shown by Michael Rudnicki (LeCouter et al. 1998) and his colleagues at McMaster University. (Nijhout 2003)
Shapiro Eur Respir J 2007:
Gene-targeting studies into a1-antitrypsin (a1-AT) and emphysema in mice have demonstrated that the genetic locus for a1-AT in mice is very complex and that the loss of one gene is lethal in embryo lung development. This underlines the differences between mice and humans that limit the ability to translate between systems in some instances. Gene targeting has also highlighted complex roles for transforming growth factor-b in COPD and has been used to determine important molecules and pathways in COPD. (Shapiro 2007) (Emphasis added.)
Dr M. G. Palfreyman, Dr V. Charles and J. Blander 2002:
Mice and humans have more than 95% of their genes in common, yet mice are not men, or women . . . Although cell-based and animal models of disease have been the cornerstone of drug discovery, it is increasingly apparent that they are of limited predictive value for complex disorders…One of the major challenges facing the drug discovery community is the limitation and poor predictability of animal-based strategies. Over the last decade, drug discovery has largely been based on finding targets in animal models and then identifying the human homologue . . . many drugs have failed in later stages of development because the animal data were poor predictors of efficacy in the human subject . . . One of the overriding interests of the pharmaceutical and biotechnologies industry is to create alternative development strategies that are less reliant on poor animal predictor models of human disease . . . Although the species [chimpanzees] share more than 98.9% gene identity [with humans], the expression of genes in the brain was more than five-fold greater in humans than in the chimpanzees.... Differences from mice were even greater. These differences reinforce the importance of using human disease models in drug discovery as a real predictor of human efficacy . . . Discovery of drugs that act on the human central nervous system, are best studied in human-cell based systems. (Palfreyman, Charles, and Blander 2002) (Emphasis added.)
New Scientist 21 April 2007:
One puzzling discovery is that several mutations that cause genetic diseases in humans - such as phenylketonuria and Sanfilippo syndrome, which lead to mental retardation - are the normal form in macaques and, presumably, our own ancestors. "How can genes that seem to be fine in one species give disease in another closely related one?" asks Richard Gibbs, a geneticist at Baylor College of Medicine in Houston, Texas, who led the consortium. (Holmes 2007)
It was Terry Magnuson of the University of North Carolina at Chapel Hill who opened many mouse geneticists’ eyes to the influence of the rest of the genome on knockout experiments. In 1995, his team disabled the gene for the epidermal growth-factor receptor. In one mouse strain, CF-1, the knockout embryos perished at around the time of implantation in the uterus. But in the CD-1 strain, they survived for up to three weeks after birth (Threadgill et al. 1995). “From that time on, everyone started paying much more attention to the genetic background,” says Magnuson. (Pearson 2002) (Emphasis added.)
However, there is probably a more fundamental reason for these failures: namely, that most of these approaches have been guided by unmitigated reductionism. As a result, the complexity of biological systems, whole organisms and patients tends to be underrated (Horrobin 2001). Most human diseases result from the interaction of many gene products, and we rarely know all of the genes and gene products that are involved in a particular biological function. Nevertheless, to achieve an understanding of complex genetic networks, biologists tend to rely on experiments that involve single gene deletions. Knockout experiments in mice, in which a gene that is considered to be essential is inactivated or removed, are widely used to infer the role of individual genes. In many such experiments, the knockout is found to have no effect whatsoever, despite the fact that the gene encodes a protein that is believed to be essential. In other cases, the knockout has a completely unexpected effect (Morange 2001). Furthermore, disruption of the same gene can have diverse effects in different strains of mice (Pearson 2002). Such findings question the wisdom of extrapolating data that are obtained in mice to other species. In fact, there is little reason to assume that experiments with genetically modified mice will necessarily provide insights into the complex gene interactions that occur in humans (Horrobin, 2003).
The disappointing results of knockout experiments are partly caused by gene redundancy and pleiotropy, and the fact that gene products are components of pathways and networks in which genes acting in parallel systems can compensate for missing ones (Morange 2001). As many factors simultaneously influence the behaviour of a system, one part might function only in the presence of other components. The essential contribution of other genes in achieving a particular function will therefore be missed, which will further encourage the reductionist view that a single gene has adequate explanatory power (Regenmortel 2004) It remains true that human disease is best studied in human subjects (Horrobin, 2003). (Van Regenmortel 2004) (Emphasis added.)
2. There have also been numerous such past breakthroughs (getting paralyzed rodents to walk again), none of which were effective in humans. Of 22 drugs tested on animals and shown to be therapeutic in spinal cord injury by 1988, none were effective in humans (American Paraplegia Society 1988). Sharon Begley wrote of this in the Wall Street Journal on January 26, 2007:
Science has made paralyzed rats walk, cured mice of cancer and eliminated Alzheimer's in more lab rodents than you can count. Human patients? Not so much.
In a previous article Begley stated:
From 1998 to this year, the budget of the National Institutes of Health doubled. The 2004 budget request is $27.9 billion. Millions more in private money gushes into biomedical research. Despite those billions, it's the paralyzed rats that walk again. Solutions, anyone? (Begley 2003)
I especially like it when the monkey researchers point out how worthless the rodent studies on spinal cord injury are. Courtine et al., Nature Medicine 2007:
Progress continues in the development of reparative interventions to enhance recovery after experimental spinal cord injury (SCI). Here we discuss to what extent rodent models of SCI have limitations for ensuring the efficacy and safety of treatments for humans, and under what circumstances it would be advantageous or necessary to test treatments in nonhuman primates before clinical trials. We discuss crucial differences in the organization of the motor systems and behaviors among rodents, nonhuman primates and humans, and argue that studies in nonhuman primates are critical for the translation of some potential interventions to treat SCI in humans. . .
Differences between rodents and primates in the pattern of CST terminations are qualitative and quantitative. In rodents, the CST projects mainly to dorsal horn neurons and premotor spinal circuits. In many nonhuman primates, such as the rhesus monkey, the projection pattern of the CST is much more complex: a significant proportion of CST fibers projects to the ventral horn, and some axons synapse directly on motoneurons8, in particular those innervating hand muscles. In humans, this trend is even more marked10. Stimulation of CST neurons in the motor cortex evokes motor responses that markedly differ in primates compared to rodents11, as well as between different primate species8. For example, there is a strong correlation between the number of direct connections between cortex and motor neurons and the level of manual dexterity of nonhuman primate species.
There also are substantial differences between rodent and most primates in the distances over which neural fibers might be required to regenerate after injury. This difference could limit the inferences that can be made from regeneration studies between rodents and primates. This is relevant for injury to the cervical spinal cord, but may be even more problematic for reinnervation of the lumbar regions, owing to the long distance that fibers may need to travel to reach locomotor circuits in human . . . (Courtine et al. 2007)
The Orange County Register article continues:
Any injury that involves disruption of nerve connections -- brain trauma from a bomb blast in Iraq or Afghanistan, or from a car accident -- could potentially be reversed using the enzyme-deletion process, he said.
WOW! They are not just curing spinal cord injury here, they are curing neurological injuries in general. Sounds like homeopathy or magic bracelets; no matter the problem this will cure you. Such claims are almost always to good to be true and this one is no exception. But when large amounts of money are involved claims such as this persist.
American Paraplegia Society. 1988. Symposium on spinal cord injury models. Presented at the 33rd annual meeting of the American Paraplegia Society. September 1987. J Am Paraplegia Soc 11 (2):23-58.
Begley, Sharon. 2003. Physician-Researchers Needed To Get Cures Out of Rat's Cage. Wall Street Journal, April 25.
Courtine, G., M. B. Bunge, J. W. Fawcett, R. G. Grossman, J. H. Kaas, R. Lemon, I. Maier, J. Martin, R. J. Nudo, A. Ramon-Cueto, E. M. Rouiller, L. Schnell, T. Wannier, M. E. Schwab, and V. R. Edgerton. 2007. Can experiments in nonhuman primates expedite the translation of treatments for spinal cord injury in humans? Nat Med 13 (5):561-6.
Holmes, Bob. 2007. Monkey genome springs surprise for human origins. New Scientist.
Horrobin, David F. 2001. Realism in drug discovery—could Cassandra be right? Nat Biotech 19 (12):1099-1100.
LeCouter, J. E., B. Kablar, P. F. Whyte, C. Ying, and M. A. Rudnicki. 1998. Strain-dependent embryonic lethality in mice lacking the retinoblastoma-related p130 gene. Development 125 (23):4669-4679.
Morange, M. 2001. The misunderstood gene. Cambridge: Harvard University Press.
———. 2001. A successful form for reductionism. The Biochemist 23:37-39.
Nijhout, H Frederik. 2003. The Importance of Context in Genetics. American Scientist 91 (5):416-23.
Oti, M., and H. G. Brunner. 2007. The modular nature of genetic diseases. Clin Genet 71 (1):1-11.
Palfreyman, M G, V Charles, and J Blander. 2002. The importance of using human-based models in gene and drug discovery. Drug Discovery World Fall:33-40.
Pearson, H. 2002. Surviving a knockout blow. Nature 415 (6867):8-9.
Regenmortel, M. H. V. Van. 2004. Biological complexity emerges from the ashes of genetic reductionism. Journal of Molecular Recognition 17 (3):145-148.
Shapiro, S. D. 2007. Transgenic and gene-targeted mice as models for chronic obstructive pulmonary disease. Eur Respir J 29 (2):375-8.
Threadgill, D. W., A. A. Dlugosz, L. A. Hansen, T. Tennenbaum, U. Lichti, D. Yee, C. LaMantia, T. Mourton, K. Herrup, R. C. Harris, and et al. 1995. Targeted disruption of mouse EGF receptor: effect of genetic background on mutant phenotype. Science 269 (5221):230-4.
Van Regenmortel, M. H. 2004. Reductionism and complexity in molecular biology. Scientists now have the tools to unravel biological and overcome the limitations of reductionism. EMBO Rep 5 (11):1016-20.
van Zutphen, L. F. 2000. Is there a need for animal models of human genetic disorders in the post-genome era? Comp Med 50 (1):10-1.