Two GlaxoSmithKline investigators have published a method that combines side-effect information from drug labels with drug-disease relationships from publicly available resources to predict possible new uses for these treatments.
The method, dubbed Drug Repositioning based on the Side-Effectome, or DRoSEf, was published in a recent edition of PLoS One.
The basic hypothesis underlying the method is that "if the [side effects] associated with a drug D are also induced by many of the drugs treating disease X, then drug D should be evaluated as a candidate for treating disease X," the authors explain in the paper.
While it's true that side effects are typically associated with unpleasant responses to drugs, DRoSEf's premise is that both side effects and disease indications are "measurable behavioral or physiological changes in response to the treatment," they wrote.
As a result, "if drugs treating a disease share the same [side effect], there might be some underlying mechanism of action linking this disease and the side effect," the researchers wrote.
For example, "if we find [a] drug that cause[s] hypoglycemia ... we possibly could reposition this drug to [treat] diabetes ... if we carefully control the dose and the formulation," Lun Yang, one of the study authors, told BioInform.
DRoSEf is comprised of two basic steps, the paper explains. First, the researchers combined 584 side effects from 888 drugs contained in the Side Effect Resource database with 303 drugs and 145 disease indications from the Pharmacogenomics Knowledegebase. These were used to create a list of 3,175 disease/side-effect associations that could suggest possible new drug indications.
For example, DRoSEf suggested that treatments that list porphyria — a hereditary disorder that affects hemoglobin — as a side effect may be used as antidiabetics, the paper states. Additionally, the tool indicated that drugs that list delusions as side effects could be effective treatments for depression.
The team also showed that in cases where side-effect information is lacking, DRoSEf can make "acceptable" predictions based on compound structures, Yang said. In the paper, the researchers used the method to predict new indications for 4,200 drug molecules from GeneGo's Metabase that had no listed side-effect information.
The next step would be to test DRoSEf's predictions in animal models, Yang said, explaining that selection would be based on the strongest and best-understood relationships between candidate drugs and disease or treatments that target unmet health needs.
Drug Recycling is in
DRoSEf is one of a number of commercial and open-source efforts aimed at recycling drugs.
For example, Stanford University spun off NuMedii last year to commercialize its drug repurposing technology (BI 8/19/2011) called New Indications Discovery, which uses a set of integration and inference algorithms to match drugs and diseases based on gene-expression signatures (BI 12/2/2011).
Separately, at the annual meeting of the American Association for Cancer Research, Life Biosystems' researchers outlined their company's computational platform, called Molecular Analysis of Side Effect information, or MASE, which uses data from the US Food and Drug Administration's Adverse Event Reporting System and curated information about biological mechanisms that underlie those side effects to predict the outcomes of specific combinations of drugs, targets, and metabolizing enzymes (BI 4/15/2011).
Prior to joining GSK, Yang was part of an international research team that developed the Drug Repositioning Potential and Adverse Drug Reaction via the Chemical-Protein Interactome, or DRAR-CPI, web server, which uses drug-drug associations and protein interactions to suggest new drug targets and identify unexpected adverse reactions (BI 6/3/2011).
The developers have submitted a patent application related to the DRoSEf method. Yang could not disclose whether GSK plans to use the approach internally.
Have topics you'd like to see covered in BioInform? Contact the editor at uthomas [at] genomeweb [.] com.