Research published in Nature Biomedical Engineering highlights a groundbreaking method for screening oral drugs to determine their interactions with the intestinal transportome using porcine tissue explants and machine learning. By manipulating transporter expression in intact porcine tissue explants through the delivery of small interfering RNAs using ultrasound, researchers achieved a comprehensive understanding of drug-transporter relationships.
The study demonstrated the effectiveness of a random forest model trained on drug-transporter interactions, achieving a perfect match for 24 drugs with known interactions. In analyzing 28 clinical drugs and 22 investigational drugs, the model successfully identified 58 previously unknown drug-transporter interactions. Subsequent testing on mice showed that 7 out of 8 predicted interactions aligned with drug-pharmacokinetic measurements.
Furthermore, the model’s predictions for interactions involving doxycycline and four other drugs (warfarin, tacrolimus, digoxin, and levetiracetam) were validated through an ex vivo perfusion assay and patient pharmacologic data analysis. This innovative approach of screening drugs for their interactions with the intestinal transportome shows promise in expediting drug development and enhancing drug safety evaluations.
The use of cutting-edge techniques like porcine tissue explants and machine learning represents a significant step forward in the quest for oral drugs with improved bioavailability. By bridging the gap between in vitro and in vivo conditions in the gastrointestinal tract, this research paves the way for more efficient drug development processes and a deeper understanding of drug-transporter interactions.
