Although various lipophilic drugs are bound to lipoproteins, lipoprotein binding in plasma is not usually considered in current physiologically-based pharmacokinetic (PBPK) models. Amiodarone is extensively bound to serum triglyceride-rich lipoproteins. Total plasma amiodarone concentration, which is the sum of both unbound and bound concentrations, increases with increasing serum triglyceride levels. We investigated the impact of lipoprotein binding on amiodarone pharmacokinetics using PBPK modeling and simulations. An amiodarone PBPK model that incorporates plasma lipoprotein binding (LPP model) was developed based on the correlation between serum triglyceride levels and lipoprotein-bound amiodarone. The predicted unbound fraction of amiodarone in plasma and systemic clearance in the LPP and base models (with albumin binding only) were similar, but the coefficients of variation for the LPP model were greater than those for the base model and were closer to the observed data. The total plasma amiodarone concentration predicted using the LPP model increased with higher levels of plasma lipoprotein binding and serum albumin. In contrast, changes in plasma lipoprotein binding and serum albumin levels did not influence the predicted unbound plasma amiodarone concentration at steady-state. This study demonstrates that incorporating plasma lipoprotein binding into a PBPK model improves the accuracy of predicting interindividual variabilities in amiodarone clearance by more reliably predicting the interindividual variability in the plasma unbound fraction of amiodarone. Plasma lipoprotein binding should be considered in PBPK modeling and simulations for lipoprotein-associated drugs if there is available information on the relationship between plasma lipoprotein binding and hyperlipidemia.
Read full abstract