PBPK modeling predicts rilzabrutinib dose adjustments using BTK occupancy for CYP3A4 interactions and hepatic...

This study aimed to develop and validate a population-based physiologically-based pharmacokinetic (PBPK) model for the Bruton's tyrosine kinase inhibitor rilzabrutinib (RIL) to predict its pharmacokinetic (PK) and BTK occupancy (BO) profiles under various scenarios, including drug-drug interactions (DDIs) with CYP3A4 modulators and in patients with hepatic impairment. A PBPK model for RIL was constructed using PK data, BO parameters, and interaction parameters. The model was validated against clinical PK parameters (AUC and C_max), BO level, DDI ratios, and ratios in patients with hepatic impairment. The PBPK model accurately predicted RIL plasma concentrations and BO time-course profiles, with AUC and C_max ratios within the acceptable range of 1.5-fold error and 15% difference between observed and predicted BO data. The predicted ratios for DDIs and hepatic impairment scenarios demonstrated good agreement with clinical observations. Based on BO as the efficacy criterion, DDI simulations predicted significant interactions with strong CYP3A4 inhibitors, necessitating dose reductions for RIL when co-administered with ritonavir (RIT; 25 mg BID or 50 mg OD), itraconazole (ITR; 50 mg BID), or clarithromycin (CLA; 75 mg BID). For moderate inhibitors like fluconazole (FLUC) or fluvoxamine (FLUV), dose adjustments to 250 mg BID and 300 mg BID, respectively, were required to maintain efficacy. The model advised against co-administration with strong inducers like rifampicin (RIF) or moderate inducers like efavirenz (EFA) and rifabutin (RIFA), though increasing the dose to 800 mg BID with RIFA could overcome the inductive effect. Hepatic impairment simulations predicted increased RIL exposure, requiring dose reductions to 300 mg BID, 200 mg BID, and 75 mg BID for mild, moderate, and severe impairment, respectively, to achieve target BO levels. The PBPK model for RIL demonstrated predictive capabilities for PK and BO profiles under various scenarios. This model can guide clinical decision-making to optimize RIL dosing in patients with and without DDIs and hepatic impairment.