Development and Verification of a Physiologically Based Pharmacokinetic Model of Furmonertinib and Its Main Metabolite for Drug–Drug Interaction Predictions

Abstract

Furmonertinib demonstrated potent efficacy as a newly developed tyrosine kinase inhibitor for the treatment of patients with epidermal growth factor receptor (EGFR) mutation-positive non-small cell lung cancer. In vitro research showed that fur monertinib is metabolized to its active metabolite AST5902 via the cytochrome P450 (CYP) enzyme CYP3A4. Furmonertinib is a strong CYP3A4 inducer, while the metabolite is a weaker CYP3A4 inducer. In clinical studies, nonlinear pharmacoki netics were observed during chronic dosing. The apparent clearance showed time- and dose-dependent increases. In this evaluation, a combination of in vitro data using radiolabeled compounds, clinical pharmacokinetic data, and drug–drug interaction (DDI) data of furmonertinib in oncology patients and/or in healthy subjects was used to develop a physiologically based pharmacokinetic (PBPK) model. The model was built in PK-Sim Version 11 using a total of 44 concentration-time profiles of furmonertinib and its metabolite AST5902. Suitability of the predictive model performance was demonstrated by both goodness-of-fit plots and statistical evaluation. The model predicted the observed monotherapy concentration pro files of furmonertinib well, with 32/32 predicted AUClast (area under the curve until the last concentration measurement) values and 32/32 maximum plasma concentration (Cmax) ratios being within twofold of the respective observed values. In addition, 8/8 predicted DDI AUClast and Cmax ratios with furmonertinib as a victim of CYP3A4 inhibition or induction were within twofold of their respective observed values. Potential applications of the final model include the prediction of DDIs for chronic administration of CYP3A4 perpetrators along with furmonertinib, considering auto-induction of furmonertinib and its metabolite AST5902.