Pharmacokinetic simulations of topiramate plasma concentrations following dosing irregularities with extended-release vs. immediate-release formulations

BACKGROUND Once-daily extended-release (XR) antiepileptic drugs (AEDs) offer potential adherence and tolerability advantages over their BID immediate-release (IR) counterparts. However, patients with epilepsy will inevitably be at least occasionally nonadherent with a prescribed dosing regimen, regardless of formulation. Although perturbations in plasma concentrations due to dosing irregularities may have clinical consequences for AEDs with concentration-response relationships, clinical studies that deliberately expose patients to specific dosing irregularities in order to assess the effect on plasma concentrations and determine appropriate corrective actions would be unethical. METHODS Computer simulation was used to assess the impact of irregular dosing on topiramate (TPM) concentrations in noninduced (monotherapy/neutral cotherapy) and induced (adjunctive therapy with enzyme-inducing AEDs) states using a population pharmacokinetic (PK) model developed to predict steady-state plasma concentration-time profiles produced by once-daily Trokendi XR (extended-release topiramate capsules, Supernus Pharmaceuticals) and BID TPM-IR. RESULTS Computer simulations predicted that, relative to adherent dosing, delaying a dose 4 to 24h in noninduced patients would decrease trough (Cmin) levels 9% to 31% in the case of TPM-IR and 6% to 27% with Trokendi XR; a single omitted dose would reduce Cmin by 21% (TPM-IR) and 27% (Trokendi XR). After dose recovery to correct for a delayed or omitted dose, simulated peak concentration (Cmax) was higher than steady-state Cmax, regardless of formulation, although the magnitude of "overshoot" was consistently lower with Trokendi XR vs. TPM-IR. Doubling of a dose would increase Cmax by 26% and 28%, respectively. Predicted changes for nonadherent vs. adherent dosing were greater in the induced vs. noninduced state but were generally comparable for the two TPM formulations. Because the long half-life of TPM has been cited as a justification for QD dosing of TPM-IR, simulations also compared steady-state PK profiles of once-daily Trokendi XR and QD TPM-IR. Predicted TPM plasma concentration-time profiles were markedly different, as demonstrated by peak-trough fluctuation (QD TPM-IR, 64%; QD Trokendi XR, 18%) and 34% lower Cmin with QD TPM-IR. CONCLUSIONS Based on these simulations, dosing irregularities with once-daily Trokendi XR should pose no greater risk than with BID TPM-IR. In the event of a delayed or omitted Trokendi XR dose, TPM concentrations can be restored in noninduced and induced states by administering the delayed/omitted dose at any time during the next dosing interval or by adding the missed dose to the next scheduled dose.