MIT News
March 25, 2021
A new screening platform combines machine learning with high-throughput experimentation to identify self-assembling nanoparticles for drug delivery. Nanoparticles, usually made from lipids, polymers or both, can improve a drug’s pharmacokinetics. However, nanoparticle production can be complex and their drug payload small. In a study published in Nature Nanotechnology, researchers from the Langer and Traverso Labs screened 2.1 million pairings of small molecule drugs and inactive drug ingredients, identifying 100 new nanoparticle formulations that are simple to create and shuttle larger drug cargoes. One of those nanoparticles, combining the cancer medicine sorafenib with glycyrrhizin (the primary flavoring of licorice), proved more effective than than sorafenib alone in both cell culture and a genetic mouse model of liver cancer.