Static and PBPK Models Give Different Estimates of Cannabidiol Drug Interactions
A stepwise modeling study compared in vitro enzyme-inhibition signals with clinical pharmacokinetic observations. PBPK predictions were generally lower than mechanistic static estimates, with CYP2C19 emerging as the main interaction concern.
Testing Three Levels of Prediction
A University of Florida-led team compared basic, mechanistic static and physiologically based pharmacokinetic (PBPK) models for cannabidiol-mediated drug interactions. The inputs included reversible and time-dependent cytochrome P450 inhibition by cannabidiol and its active metabolite, 7-hydroxycannabidiol.
The researchers first used the static approaches to identify potentially sensitive metabolic pathways. They developed a PBPK model for the parent drug and metabolite, compared it with clinical pharmacokinetic data, and simulated interactions. Where needed, inhibition parameters were refined against clinical observations and then carried back into the static models for comparison.
A Smaller Interaction Signal after Dynamic Modeling
The mechanistic static model predicted 3.6-, 4.1- and 2.2-fold increases in exposure for sensitive CYP1A2, CYP2C9 and CYP3A4 substrates, respectively, together with a strong CYP2C19 interaction. PBPK modeling gave lower interaction estimates that aligned more closely with the available clinical data.
In the PBPK assessment, CYP2C19 inhibition remained the most clinically relevant pathway, with a moderate interaction predicted for sensitive substrates; CYP1A2 inhibition approximately doubled sensitive-substrate exposure. The authors also used the model to examine special populations and commonly co-prescribed drugs.