Abstracts

Rationale: To evaluate the long-term effectiveness of pharmaceutical formulation of highly purified cannabidiol (Epidiolex) in children with intractable epilepsy, using treatment persistence (retention rate) as the surrogate outcome measure, and to assess the impact of concomitant AEDs on retention rate. Methods: The data are from a prospective multisite expanded access protocol open label study in children ages 1-18 years. Kaplan-Meier survival curves were generated for time to CBD discontinuation, and separately for time to CBD failure, defined as discontinuation of CBD or addition of another antiepileptic therapy. The impact of concomitant use of clobazam (CLBZ) or valproate VPA), and patient and epilepsy characteristics were evaluated using Cox regression. Results: Seventy six patients were enrolled at 3 sites. Excluding patients with < 1 month follow up, the analysis cohort included 70 patients (30 female). Median follow time was 6.5 months (IQR 3-43), and 26 patients were followed for more than 24 months. Median age was 10 years (range 1 to 20 years). Median CBD dose was 25 mg/kg/day (IQR 20-25). Median number of concomitant AEDs was 3 (range 1-6). Concomitant AEDs at CBD initiation included VPA in 16 (23%) and CLBZ in 42 (59%). Probability of remaining on CBD was 97% at 12 months and 93% at 24 months. LGS diagnosis was not associated with a difference in CBD failure (HR 0.60, 95%CI 0.26-1.37). Probability of remaining on CBD without addition of another therapy, or increase in concomitant AED dose was 54% at 12 months and 24% at 24 months. Use of VPA at the time of CBD initiation had no effect on time to CBD failure (HR 0.85, 95%CI 0.37-1.98). Use of CLBZ at CBD initiation was associated with lower rate of CBD failure (HR 0.27, 95%CI 0.12-0.62, p = 0.002). Likelihood of remaining of CBD without increase/addition of other therapy was 72% in patients on CLBZ at CBD start, and 37% in patients with no concomitant CLBZ. Conclusions: This study provides observational evidence for long-term treatment persistence of CBD in children with intractable epilepsy, using time to CBD discontinuation and time to treatment failure (CBD discontinuation, addition or increase in other therapy) as the surrogate outcome measures. In this cohort, etiology did not effect treatment persistence. Concomitant use of CLBZ was associated with a lower rate of CBD failure. Funding: Commonwealth of Pennsylvania