Efficacy and Safety of Epidiolex (Cannabidiol) in Children and Young Adults with Treatment-Resistant Epilepsy: Update from the Expanded Access ProgramAuthors: Orrin Devinsky, Elizabeth Thiele, Linda Laux, Daniel Friedman, Anup Patel, Judith Bluvstein, Michael Chez, Charuta Joshi, Roberta Cilio, Francis Filloux, Evan Fertig, Angus Wilfong, Paul D. Lyons, Yong Park, Robert Flamini, Matthew Wong, Ian Miller, Eric Marsh.
Rationale: Cannabidiol (CBD) is the most abundant non-psychoactive cannabinoid in the cannabis plant. Animal studies demonstrate anticonvulsant efficacy in multiple species and models. Anecdotal reports suggest efficacy in children with treatment-resistant epilepsies (TRE), including Dravet Syndrome (DS) and Lennox-Gastaut Syndrome (LGS). We report current results in our expanded access treatment program.
Methods: Children and young adults with TRE in an expanded access compassionate use program for CBD were enrolled in a prospective observational study. During the 4 week baseline, parents/caregivers kept prospective seizure diaries of all countable motor seizure types. Patients received a highly standardized pharmaceutical plant-derived, purified CBD. (Epidiolex: GW Pharma), at a gradually increasing dose from 2-5 mg/kg/day until intolerance occurred or a maximum dose of 25 mg/kg/day was achieved. Patients were seen at regular intervals of 2-4 weeks during the initial 12 weeks of therapy. Testing for hematologic, liver, kidney function and AED levels was performed at baseline, and after 4, 8 and 12 weeks of CBD therapy.
Results: 261 patients received at least 3 months of treatment and had available data at last group data collection (136 (52%) were male; average age 11.8 years, range 4 months-41 years; average weight 38 kg; range 6.4-127). The most common diagnoses were DS (44; 17%) and LGS (40; 15%). The average # of concomitant AEDs was 3.0. After 3 months of therapy, the median overall seizure frequency reduction was 45.1% in all patients and 62.7% in DS patients. For LGS patients, the median reduction of atonic seizures from baseline was 71.1%. Among all patients, 47% had a ≥50% reduction in seizures. Seizure-freedom at 3 months occurred in 9% of patients and 13% of DS patients. Clobazam co-therapy was associated with a higher rate of treatment response (≥50% convulsive seizure reduction): 57% v. 39%; this may reflect elevations in the desmethyl clobazam metabolite. Safety data from 313 patients representing 180 patient years was available at 16 sites. Adverse events in ≥10% of patients included somnolence (23%), diarrhea (23%), fatigue (17%), decreased appetite (17%), convulsions (17%) and vomiting (10%). 14 patients (4%) had an adverse event leading to discontinuation of CBD. 36 patients (12%) withdrew primarily due to lack of efficacy. Serious Adverse Events (SAEs) were reported in 106 patients (34%), including 7 deaths, none of which were considered treatment-related. 16 patients (5%) had SAEs that were considered treatment-related, including altered liver enzymes (4 pts; all were also on valproate and clobazam), status epilepticus/convulsion (4), diarrhea (4), decreased weight (3), thrombocytopenia (1), and others.
Conclusions: These results from an uncontrolled study support the animal studies and prior reports showing that CBD may be a promising treatment for TRE and it is generally well-tolerated in doses up to 25mg/kg/day. Epidiolex is now being investigated in randomized controlled studies in DS and LGS.