Predicting human tissue exposures to xenobiotics using a bottom-up physiologically-based biokinetic model

Area of Science:

• Pharmacokinetics and Drug Metabolism
• Biophysical Modeling
• Translational Pharmacology

Background:

• Physiologically-based biokinetic (PBK) modeling, in vitro-to-in vivo extrapolation (IVIVE), and permeability-limited models enable drug concentration predictions without human data.
• A critical gap exists in in vivo human tissue concentration data for validating these predictive models.

Purpose of the Study:

• To validate a previously developed bottom-up rosuvastatin (RSV) PBK model using clinical data.
• To assess the model's accuracy in predicting hepatic RSV concentrations during drug-drug interactions (DDIs) with cyclosporine A (CsA).

Main Methods:

• Utilized positron emission tomography (PET) imaging data to quantify hepatic [11C]RSV concentrations.
• Compared simulated plasma and hepatic RSV concentrations (AUC, Cmax) with observed clinical data.
• Modified PBK model inputs to incorporate RSV within bile canaliculi post-biliary excretion.

Main Results:

• Simulated plasma RSV concentrations (AUC0h-t, Cmax) were within 1.5-fold of observed data before and after DDI.
• Simulated DDI plasma concentration ratios (AUC0-30min, Cmax) closely matched observed ratios (within 1.1-fold).
• The modified PBK model accurately recapitulated observed hepatic RSV concentrations and the DDI-induced decrease, with predictions within 1.5-fold of PET data.

Conclusions:

• Validated the predictive capability of bottom-up PBK modeling for hepatic RSV concentrations.
• Demonstrated the importance of including drug distribution in bile canaliculi for accurate hepatic tissue concentration predictions.
• Confirmed the utility of PBK models for predicting drug behavior in human tissues, especially during DDIs.