PIII‐58

AIM Model the distinct pharmacokinetic (PK) profiles of 5 oral formulations of valproic acid (VPA) that are commonly used for treatment of epilepsy & bipolar disorder, & for migraine prophylaxis. METHODS Plasma VPA concentration‐time profiles, following single oral dose administration of 5 VPA formulations under fasting conditions, from 4 PK studies in healthy subjects (N=10–15) were compared: VPA syrup & capsule, divalproex sodium sprinkles capsule, delayed‐release (DR) tablet & extended‐release (ER) tablet. Mammillary compartmental disposition models (CM) coupled with either 1st‐ or 0‐order, or multiphasic absorption models, were fit to the observed data using WINNONLIN. RESULTS The optimal models & mean absorption parameters were: (1) syrup: 2CM, k0=1 g/h; (2) capsule: 1CM, ka=1.7 1/h; (3) sprinkles capsule: 1CM, ka=0.99 1/h, Tlag=0.89 h; (4) DR tablet: 1CM, ka=0.79 1/h, Tlag=1.4 h; & (5) ER tablet: 1CM, multiphasic absorption characterized by a peak rate of 0.12 mg/h/mg‐dose that occurs immediately after dosing followed by a constant rate of 0.027 mg/h/mg‐dose over 22h. Mean CL/F & V/F values were consistent across formulations & ranged over 0.52–0.65 L/h & 9.9–11 L, respectively. CONCLUSIONS The 5 VPA formulations demonstrated distinct absorption characteristics. The rate of absorption may be rank‐ordered as VPA syrup > VPA capsule > divalproex sprinkles capsule ≅ divalproex‐DR tablet > divalproex‐ER tablet. ER is the only formulation exhibiting true sustained‐release characteristics. Clinical Pharmacology & Therapeutics (2005) 79 , P74–P74; doi: 10.1016/j.clpt.2005.12.266