Abstracts

Rationale: Valproic acid (VPA, brand name: Depakote) is a long-standing first-line antiepileptic drug used for treating many types of seizures and other various neurological disorders. It is almost entirely metabolized by the liver through three major pathways including: glucuronidation (~ 50%), ߭oxidation within mitochondria (~ 40%), and cytochrome P450-mediated oxidation (~ 10%). Several formulations of VPA are available on the market for treatment including immediate release (IR) and extended-release (ER). This study presents a new physiologically based pharmacokinetic (PBPK) model that considers UGT enzyme kinetics, ontogeny for VPA, and utilizes an Advanced Dissolution, Absorption and Metabolism (ADAM) model for ER formulation. The objective was to predict VPA disposition within adult and pediatric populations. Methods: PBPK models for VPA IR and ER formulations were constructed using Simcyp Simulator (Version 15), which utilized the ADAM model for ER and a Pediatric population simulator. The observed data were obtained from literature. Concentration-time profiles for adults were simulated for single and steady-state IR doses ranging from 250 mg to 1000 mg. Similarly, profiles were also simulated for ER single and steady-state doses of 500 mg and 1000 mg, respectively. Additionally, the model simulated observed concentration-time profile data for IR and ER formulations administered within pediatric patients ranging from infants to adolescents in several body-weight normalized doses. Results: The relevant observed in vivo data for VPA was obtained from nineteen clinical studies reported, where there is considerable variability among findings. Our developed VPA PBPK model was able to adequately predict PK concentration-time profiles for both IR and ER formulations administered in adult and pediatric populations within the 95th and 5th percentile confidence intervals. Predicted AUC and Cmax results for IR single doses of 250 mg, 300 mg, 400 mg, 500 mg, 800 mg, 900 mg, and 1000 mg were within 40% and 25% of observed data, respectively; and for IR steady-state dosing of 200 mg bid, 500 mg bid, 625 mg bid, 900 mg bid, and 1000 mg bid were within 40% and 30% of observed data, respectively. The ADAM model, which incorporated a controlled-release profile, was used for ER formulation where the simulated concentration-time profiles were in good agreement with observed data. The predicted AUC and Cmax values for ER single dose of 500 mg was within 20% and 30% of observed data, respectively; ER steady-state dosing for 1000 mg qd was within 40% and 25% of observed data, respectively. Conclusions: The PBPK model was able to predict the concentration-time disposition of VPA for ER and IR formulation profiles within adult and pediatric populations. This validated model will be helpful to design dosing regimens for VAP dosing within adults and children. Funding: Simcyp provided the software 'Simcyp Simulator-Version 15' for this work.