×îÐÂÂ鶹ÊÓƵ

New Uses for Existing Medicines

In a novel approach, researchers used computers and genomic data to find new applications for existing FDA-approved drugs. The accomplishment represents a major step forward in drug discovery.

Drug approval takes many years of research, development and safety testing. When drugs that have already been approved are used for other purposes, it can avoid a great deal of time and investment. However, it’s difficult to figure out what other uses a given drug could serve. A team of researchers at Stanford University led by Dr. Atul J. Butte hypothesized that effective drugs might induce gene expression profiles that are opposite to the profiles caused by the condition they could treat. For example, if a disease increases the activity of certain genes, drugs that decrease the activity of those genes might be used to treat the disease.

To test this idea, the researchers drew on data from the NIH ×îÐÂÂ鶹ÊÓƵ Center for Biotechnology Information Gene Expression Omnibus, a publicly available database that contains the results of thousands of genomic studies submitted by researchers across the globe. This resource catalogs changes in gene activity under various conditions, such as in diseased tissues or in response to medications.

The research team created a computer program to compare the expression profiles of about 164 drugs and 100 diseases. The program searched through the thousands of possible drug-disease combinations to find drugs and diseases whose gene expression patterns essentially cancelled each other out. The work was supported by NIH’s ×îÐÂÂ鶹ÊÓƵ Institute of General Medical Sciences (NIGMS), ×îÐÂÂ鶹ÊÓƵ Cancer Institute (NCI) and ×îÐÂÂ鶹ÊÓƵ Library of Medicine (NLM). The results appeared in 2 articles on August 17, 2011, in Science Translational Medicine.

The approach pointed to potential drug-disease relationships for 53 of the 100 diseases examined. The program associated each of the 164 drugs with at least one disease. It predicted drug-disease pairs that are already in the market, validating the approach. For example, it matched prednisolone, a well known corticosteroid, with Crohn’s disease, for which it is already a standard therapy.

The researchers selected 2 candidate drug-disease pairs for further testing. They found that cimetidine, which is prescribed for heartburn, successfully inhibited the growth of human lung tumors both in the laboratory and when implanted in mouse. They also found that topiramate, an epilepsy drug, effectively decreased symptoms in rodents with inflammatory bowel disease.

In addition to identifying new potential drug-disease relationships, this approach may provide more basic insights. The program clustered drugs based on their mode of action. By studying unexpected members of these clusters, scientists could learn more about how certain diseases progress and how drugs work at the molecular level.

"This work is still in an early stage, but it is promising proofs of principle for a creative, fast and affordable approach to discovering new uses for drugs we already have in our therapeutic arsenal," says Dr. Rochelle M. Long, Director of the NIH Pharmacogenomics Research Network.