Creation of Novel Sensitive Probe Substrate and Moderate Inhibitor Models for a Comprehensive Prediction of CYP2C8 Interactions for Tucatinib

Abstract

A physiologically‐based pharmacokinetic (PBPK) model was developed to simulate plasma concentrations of tucatinib (TUKYSA®) after single‐dose or multiple‐dose administration of 300 mg b.i.d. orally. This PBPK model was subsequently applied to support evaluation of drug–drug interaction (DDI) risk as a perpetrator resulting from tucatinib inhibition of CYP3A4, CYP2C8, CYP2C9, P‐gp, or MATE1/2‐K. The PBPK model was also applied to support evaluation of DDI risk as a victim resulting from co‐administration with CYP3A4 or CYP2C8 inhibitors, or a CYP3A4 inducer. After refinement with clinical DDI data, the final PBPK model was able to recover the clinically observed single and multiple‐dose plasma concentrations for tucatinib when tucatinib was administered as a single agent in healthy subjects. In addition, the final model was able to recover clinically observed plasma concentrations of tucatinib when administered in combination with itraconazole, rifampin, or gemfibrozil as well as clinically observed plasma concentrations of probe substrates of CYP3A4, CYP2C8, CYP2C9, P‐gp, or MATE1/2‐K. The PBPK model was then applied to prospectively predict the potential perpetrator or victim DDIs with other substrates, inducers, or inhibitors. To simulate a potential interaction with a moderate CYP2C8 inhibitor, two novel PBPK models representing a moderate CYP2C8 inhibitor and a sensitive CYP2C8 substrate were developed based on the existing PBPK models for gemfibrozil and rosiglitazone, respectively. The simulated population geometric mean area under the curve ratio of tucatinib with a moderate CYP2C8 inhibitor ranged from 1.98‐ to 3.08‐fold, and based on these results, no dose modifications were proposed for moderate CYP2C8 inhibitors for the tucatinib label.