Physiologically based pharmacokinetic model predictions of natural product-drug interactions between goldenseal, berberine, imatinib and bosutinib

Purpose This study implements a physiologically based pharmacokinetic (PBPK) modelling approach to predict the effect of hydrastine and berberine, two major alkaloids present in goldenseal extract, on pharmacokinetics of imatinib and bosutinib. Methods PBPK models of hydrastine and berberine were developed in the Simcyp Simulator (version 17), integrating prior in vitro knowledge and published clinical pharmacokinetic data. The models account for reversible and irreversible (mechanism-based) inhibition of CYP3A enzymes as well as inhibition of the P-glycoprotein transporter. Inhibitory potencies of hydrastine and berberine on imatinib and bosutinib were estimated based on in vitro inhibition of metabolite formation. Results The PBPK models provided reliable estimates on the magnitude of interactions due to co-administration of goldenseal extract or high-dose berberine on substrates of CYP3A enzymes (midazolam, indinavir and cyclosporine) and P-glycoprotein (digoxin). PBPK simulations predicted a moderate twofold increase (5 th to 95 th percentiles of prediction of 1.4–3.1) in systemic exposure (AUC) of bosutinib when co-administered with clinically relevant doses of goldenseal extract. A high dose of berberine (300 mg thrice daily) was also expected to affect bosutinib exposure, albeit to a lesser extent than that predicted with goldenseal (AUC ratio of 1.3, 5 th to 95 th percentile: 1.1–1.6). Conversely, the corresponding effects on imatinib exposure are unlikely to be of clinical importance (predicted AUC ratios of 1.0–1.2). Conclusion PBPK model-based predictions highlighted potential clinically significant interactions between goldenseal extract and bosutinib, but not imatinib. Dose adjustment may need to be considered if co-administration is desirable. These findings should be confirmed with optimally designed controlled drug interaction studies.