Back

A machine learning liver-on-a-chip system for safer drug formulation

Shi, Y.; Lin, C.-H.; Reker, D.; Steiger, C.; Hess, K.; Collins, J. E.; Tamang, S.; Ishida, K.; Lopes, A.; Wainer, J.; Hayward, A. M.; Walesky, C.; Goessling, W.; Traverso, G.

2022-09-06 bioengineering
10.1101/2022.09.05.506668 bioRxiv
Show abstract

Drug metabolism leads to biotransformations of pharmaceutical substances that alter drug efficacy, toxicity, as well as drug interactions. Modeling these processes ex vivo stands to greatly accelerate our capacity to develop safe and efficacious drugs and formulations. Recognizing the liver as the primary site of drug metabolism, here we report a novel whole-tissue ex vivo liver screening platform that enabled modeling of hepatic metabolism and tracking of hepatotoxic drug metabolites. We applied the system for the characterization of acetaminophen (APAP) metabolism and identified interactions that can mitigate the generation of toxic metabolites ex vivo. Combining our experimental platform with state-of-the-art machine learning, we validated two novel functional excipients that can prevent APAP hepatotoxicity in vivo in mice. To assess translational potential, we prototyped a novel solid dosage form with controlled release of both APAP and our functional excipients. Our this platform provides innovative potential access to actionable data on drug metabolism to support the development of new therapeutic approaches.

Matching journals

The top 12 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.