Abstracts

Rationale: ZX008, an oral solution of fenfluramine, is under development as a low-dose adjunctive treatment of seizures in patients with Dravet syndrome (DS) and other pediatric-onset epilepsies. Treatment of DS often involves multiple antiepileptic drugs (AEDs), some of which have the potential to interact with fenfluramine, increasing exposure. One such common treatment regimen includes stiripentol, clobazam, and valproate. Determining the optimal dose of ZX008 to use with this regimen will guide clinicians in selecting a dose that will provide good tolerability when initiating or titrating ZX008. Our objective was to construct a system of physiologically-based pharmacokinetic (PBPK) models to describe the time course of fenfluramine concentrations and quantify the potential impact of concomitant medications on that time course. This system was used to facilitate dose selection for a Phase 3 study designed to assess the efficacy and safety of ZX008 when added to a stiripentol-containing regimen in pediatric patients with DS. Methods: The PBPK model system was developed for fenfluramine and a concomitant stiripentol regimen (stiripentol, clobazam, and valproate). The models for the concomitant medications were developed by replicating published PBPK models from the literature; the model for fenfluramine was developed de novo using basic properties of the molecule. Model qualification was based on a Phase 1 drug-drug interaction (DDI) study in adults and a single-dose PK cohort from a Phase 3 study of ZX008 in pediatric patients with DS. Results: The PBPK model system reliably predicted the extent of the DDI in adults (n=26). After scaling to body size and accounting for maturational trends in children, the PBPK model system also adequately predicted fenfluramine and metabolite (norfenfluramine) concentrations in DS patients from the single-dose PK cohort (n=13). Model-based simulations were then used to identify the necessary ZX008 dose reduction (from 0.8 to 0.5 mg/kg/day; maximum from 30 to 20 mg/day) to ensure ZX008 exposure when given with stiripentol, clobazam, and valproate across the range of ages studied, was consistent with that expected for 0.8 mg/kg/day without the stiripentol-containing regimen. In two ongoing Phase 3 trials where stiripentol is excluded, 0.8 mg/kg/day is the maximum dose being investigated. Conclusions: Use of a pharmacometric-based strategy allowed for prediction of potential DDI in patients and dose selection for a Phase 3 trial of ZX008 when used as an adjunctive treatment to a stiripentol treatment regimen. These data will help clinicians select the safest dose of ZX008 that can be added to current regimens for treatment of DS and other pediatric-onset epilepsies. Funding: This study was funded by Zogenix, Inc.