Scientists respond to New York Times article on mice research
As physician scientists, we are deeply concerned that the Feb. 12, 2013, New York Times article "Mice Fall Short as Test Subjects for Humans’ Deadly Ills" is misleading. The sweeping generalization of the headline is contradicted by the story’s fourth sentence: "The study’s findings do not mean that mice are useless models for all human diseases." But the headline’s impression remains: using mice for research has not translated well to human medicine. That message is simply not true.
The article cites a study, published in the Proceedings of the National Academy of Sciences (PNAS), that shows the mouse genomic response to artificial models of human sepsis, burns and trauma, as measured in blood cells, had very little correlation with responses measured in humans. From this niche piece of research arose the implication that all mouse models are suspect in their contribution to medical science. But the story failed to disclose that the researchers used the mouse equivalent of one individual human—one genetically identical inbred strain (C57BL/6)—to study the correlation. This approach would be akin to prescribing a single treatment for all human beings after testing that treatment on only one person from an ethnically isolated population. So, it would be far more accurate to say "A Single Mouse Falls Short…" rather than "Mice Fall Short...."
In addition, experimental protocols for shock from sepsis, burns and trauma in mice are difficult to match with humans who are receiving all the therapeutic modalities that modern medicine has to offer. Therefore, many in the scientific community would seriously question if the results reported in this study are the definitive test of the value of mouse models for understanding acute inflammatory responses in humans.
Although mice are clearly different than humans, they are by far the best mammalian model for the study of human disease. Their genome is well known with unambiguous correlations of their genes to more than 99 percent of human genes, with the added benefit that the mouse genome can be engineered to produce genetically identical models of human disease. Moreover, mouse models are cost-effective experimental tools and allow unlimited access to disease-affected tissues.
Much of what we learned about human biochemistry and physiology today used mice in the discovery process. Research leading to nearly all Nobel Prizes in Medicine or Physiology awarded since 1901 was dependent on data from animal models, with mouse models predominating since the 1980s. Scientists vote through their publications, and in the definitive database of publications in biomedical research, PubMed, 35,361 papers were found contain the terms "human disease" and "mouse models." Mouse models played pivotal roles in the development of treatments for all forms of human cancer, retinal and neuronal degeneration, aortic aneurysms and vascular disorders, muscular dystrophies, multiple sclerosis, and more. Thus, the major reason why the mouse continues to be an important contributor to scientific advances is that it works very well. The key, however, is to construct the appropriate mouse models and design the experimental conditions that mirror the human situation.
Recently, researchers have found new ways to increase dramatically the utility and impact of the mouse in biomedical research. One approach is to create colonies with more genetic diversity than in the entire human population instead of a single genetic configuration. Genetically diverse outbred colonies provide powerful tools to pinpoint the genes leading to adverse drug reactions.
Another approach involves the development of mouse models that can host human cells and tissues, reducing the gap between mice and the human condition. Engrafting human immune tissue into immunodeficient mice potentially provides a powerful platform for research with medical significance. Studies of other tissues – human cancers – in these immunodeficient mice are at the forefront of current efforts to generate "personalized" treatments for disease (as reported by Andrew Pollack, "Seeking Cures, Patients Enlist Mice Stand-Ins," Sept. 25, 2012). Mouse models can now be "fine tuned" for the medical conditions under study.
Medical progress is hard-won. Improvement is usually gained in small increments over long periods. But experimental research using mice is critical to building knowledge and to translating this knowledge into medical practice in a manner that minimizes risk to human beings. We sincerely hope that the ensuing discourse on the use of laboratory mice will be one informed by its exemplary history of successful contributions.
David Valle, M.D.
Henry J. Knott Professor and Director,
McKusick-Nathans Institute of Genetic Medicine
The Johns Hopkins School of Medicine
Robert Darnell, M.D., Ph.D.
Heilbrunn Professor and Senior Physician
The Rockefeller University
Investigator, Howard Hughes Medical Institute
President and Scientific Director, New York Genome Center
Edison Liu, M.D.
President and CEO, The Jackson Laboratory
President, Human Genome Organization