New York Times Article Misleads on the Value of Mouse Models

In February 2013, the New York Times¹ covered a study published in a prestigious research journal, the Proceedings of the National Academy of Sciences (PNAS).² The sensational headline was “Mice Fall Short as Test Subjects for Humans’ Deadly Ills.” The study compared gene expression data in white blood cells from both humans and mice under conditions of trauma, burn, or sepsis.

A detailed response to the Times article was posted by Mark Wanner³ on a blog at the Jackson Laboratory, a non-profit organization that conducts mammalian genetics research to advance human health. His rebuttal highlights the main problem with the PNAS study - it compares heterogenous human data to a single mouse strain, which is the equivalent of one mouse, given that animals within an inbred strain are genetically identical. Furthermore, the strain used has long been known to be resistant to bacterial infections.

The general public may be left with the impression that research using mice has been a waste of years of research and billions of dollars of funding. Headline scanners and casual readers may miss a key sentence: “The study’s findings do not mean that mice are useless models for all human diseases.”¹

The hyperbole in the Times’ article underscores the importance in distinguishing the various opinions of scientists from conclusions that are supported by actual data. In the original PNAS article, the study conclusions are more conservative, suggesting that genomic studies in the dawn of this new era of molecular medicine could complement or decrease the need for mouse models in drug discovery and development. That is a reasonable statement. The authors then claim their data support prioritization of genomic studies such as theirs over mouse studies. They set up a “straw mouse,” using a model that one would not expect to recapitulate human sepsis - an infection-resistant strain - and then knock it down, citing the lack of correlation of its gene expression data compared to humans.

Though the media overstated the study’s findings, it revitalized an important conversation about the strengths and limitations of mouse models. For understanding the role of genes in whole animals, the mouse is the best model. The Times article ultimately conceded the potential contributions of mice, stating “Researchers said that if they could figure out why mice were so resistant, they might be able to use that discovery to find something to make people resistant.”¹ Funding agencies clearly recognize that medical progress will most likely occur through diverse funding portfolios that include both basic research using animal models and clinical research in human patients.

Biography

Catherine E. Hagan, DVM, PhD, is an Assistant Professor, Department of Veterinary Pathobiology, University of Missouri.

Contact: haganc@missouri.edu