Content Header Mask
Science Blog

The Science Blog

27/03/2013

Mouse models found to poorly mimic human disease

For decades mouse ‘models’ have been used to identify potential drugs for human trials, not because they are particularly relevant to human research, but because mice are cheap, small and quick to breed. Despite shockingly high drug failure rates, scientists remain reluctant to evaluate how well these models are actually able to mimic human conditions – until now. 

A paper published in PNAS (1) and reported last month in the New York Times (2) describes how, in a project that started 10 years ago, a group of 39 researchers from across the USA teamed up to find out which genes are involved in the human body’s response to deadly inflammatory diseases. They began by collecting white blood cells from over 400 patients with severe burns, trauma or sepsis to discover what genes were being used to generate the inflammatory response. 

The researchers discovered significant patterns that seemed to be able to predict which patients would survive the infections. Shockingly, when they tried to publish their findings, the study was rejected by several journals because they did not demonstrate the same gene response in mice. Dr. Ronald W. Davis, a genomics expert at Stanford University and a lead researcher in the project said, “They were so used to doing mouse studies that they thought that was how you validate things. They are so ingrained in trying to cure mice that they forget we are trying to cure humans”.

After the rejection of their human data set, the scientists then decided to investigate if the responses were indeed similar in mice. When they compared gene responses in mouse models of inflammatory disease with the patterns found in humans, they were unable to find a single similarity; in fact the responses in mice were close to random in matching the human data. In some cases, genes that seemed to play a part in the mouse form of the condition were not even involved in the human condition. Dr. Mitchell Fink, a sepsis expert at the University of California told the New York Times; “When I read the paper, I was stunned by just how bad the mouse data are. It’s really amazing — no correlation at all. These data are so persuasive and so robust that I think funding agencies are going to take note.”

The paper claims that so far, out of 150 clinical trials aimed at testing potential drugs to block the inflammatory response in critically ill patients, not one has proven successful. The findings from this project not only explain the high failure rate of inflammatory drugs but also raise questions over the use of mice to study other diseases where the immune system plays a huge part, such as cancer and heart disease. “To understand sepsis, you have to go to the patients”, said Dr. Richard Hotchkiss, a sepsis researcher at Washington University in response to the study. Clearly, it is time to move on from fruitless animal models and focus on more human-relevant approaches, which could save millions of both human and animal lives.

Sources:

1. Genomic responses in mouse models poorly mimic human inflammatory diseases, PNAS early edition found at here.

2. New York Times

ARCHIVE