Data-Mining for New Treatments

A new database helps scientist predict new uses for existing drugs.

Aug 18, 2011
Jef Akst


Repurposing existing drugs for new uses is not a new thing, but when it happens, it’s usually little more than a happy accident—a lucky observation or a fortuitous mistake. Now, scientists are taking the chance out of the equation (see The Scientist’s recent feature on this topic). The latest in a string of efforts to streamline repurposing efforts, bioinformatician Atul Butte of Stanford University School of Medicine in California and his colleagues compiled a database that collates information on gene activity profiles for the 25,000+ human genes, and how they're  affected by drugs and disease.

"This promises new uses for drugs that have already been tested for their safety and offers a faster and cheaper way to new medicine," Butte told ScienceNOW.

So far, the team has examined data for 100 diseases and 164 drug molecules, and found drug candidates for 53 different diseases, some of which had already been discovered, but some of which were entirely new. For example, one over-the-counter drug known as cimetidine, an acid inhibitor that is used to treat heartburn, appeared to interact with pathways involved in lung adenocarcinoma, a type of non-small-cell lung cancer. When they tested it in a mouse model of lung cancer, the researchers found that cimetidine slowed the growth of human lung cancer cells. A second prediction—that the epilepsy drug topiramate could help fight inflammatory bowel diseases such as Crohn's disease—also proved true in a rodent model. The researchers published their results yesterday (August 17) in Science Translational Medicine.

Though these two drugs are not necessarily likely to be approved— topiramate, for example, has side effects such as behavioral changes and cognitive problems—researchers are hopeful that the approach will lead to more fruitful findings in the future. "This is a really important concept," John Overington, a computational chemical biologist at the European Bioinformatics Institute in Hinxton, UK, told ScienceNOW. “It is almost like they are looking for an antidote to a disease.”