A team of researchers at MIT, Brigham and Women’s Hospital, and Duke University have developed a method to identify the transport proteins that specific drugs use to leave the digestive tract. This is particularly important because drugs that use the same transport protein can interfere with each other’s function.
Understand the workings of these transporters could improve patient care by avoiding these interference and even help drug developers in creating medications with better absorbability. They’ve developed a two-pronged strategy that uses both tissue models and machine-learning algorithms to gather these insights.
The team has already identified an interaction between a common antibiotic and a blood-thinner that could potentially interfere with each other. The study, published in Nature Biomedical Engineering, aims to use these insights to make drugs safer, more effective, and predict potential toxicities that might be otherwise difficult to foresee.
The method they used involved creating models from pig intestinal tissue grown in a laboratory. Through the use of short RNA strands, they then manipulated the expression levels of individual transporters in different sections of the tissue. This allowed the researchers to understand the specific role each transporter played with various drugs.
The team extended the findings by training a machine-learning model to predict drug-transporter interactions based on the chemical structure of the drugs. This model was then applied to a set of 28 currently used drugs and 1,595 experimental drugs, producing nearly 2 million predictions of potential drug interactions.
One significant prediction identified was an interaction between doxycycline, an antibiotic, with warfarin, a widely used blood-thinning agent. When data was tested from around 50 patients using these drugs, it was found that when doxycycline was added to patients already taking warfarin, the level of warfarin in the bloodstream increased, then decreased again once the doxycycline was discontinued.
In addition to predicting drug interactions in currently used drugs, the method holds promise for drugs in development. Drug developers could apply this technology to design new medication molecules that do not interact with other drugs or have superior absorbability. Vivtex, a biotech company co-founded by the researchers, is utilizing this technology as a part of their drug development process.
The researchers suggest that a more comprehensive understanding of drug transporter interactions could lead to safer and more effective medications, potentially revolutionizing drug delivery systems. The research is currently funded by the U.S. National Institutes of Health, the Department of Mechanical Engineering at MIT, and the Division of Gastroenterology at Brigham and Women’s Hospital.
