Abstracts

Rationale: There has been an increased interest in the use of artisanal cannabinoids in pediatric epilepsy but there is little prospective data assessing efficacy.  We aimed to prospectively characterize the use and efficacy of oral cannabis extracts (OCE) in a refractory pediatric epilepsy population. Methods: Children who were considering use of an OCE were enrolled in a prospective observational study. Children must have at least 2 seizures per week, failed at least 2 medications with stable dosage for 4 weeks prior to enrollment and not have a rapidly degenerative disorder.  Variables collected include demographics, medications, seizure types, baseline EEG, and characteristic of OCE.  Baseline seizure frequency was assessed over a period of 4 weeks.  Once OCE were initiated, study measures (including seizure frequency, CBD and THC levels and information regarding the product used) were taken every 4 weeks.  EEG was reassessed 12 weeks after use of OCE started.  Demographics were analyzed using descriptive statistics (medians and interquartile ranges for continuous variables, counts and proportions for categorical variables).  The parameter of interest was proportion of subjects responding to treatment, with response defined as at least a 50% reduction in seizure frequency over the final 8 weeks of the study relative to baseline.  For patients with missing seizure diary data, chart review was performed to determine response status. Logistic regression was used to assess the association between blood levels of THC and CBD and response. Results: Consent was obtained in 32 children; 11 were excluded from analysis (3 failed to complete baseline data, 3 started OCE before completing baseline period and 5 did not start OCE) and 21 were included in subsequent analyses.  Median age was 10.3 years (IQR 6.8 – 12.6), 13 (62%) were male.  Participants used 1-9 anti-seizure medications prior to participation and were still taking 1-5 anti-seizure medications during the study period.  There was a median seizure frequency of 2.7 seizures/day during baseline period.  Several CBD products were used, with 33% and 19% of subjects primarily using the two most common products, respectively.   Of the 21 subjects who were included in the analysis, 5 (24%) were responders.  OCE use was stopped early in 14% due to a perceived increase in seizures.  THC and CBD blood levels did not have a significant association with response status (p=0.58 CBD, p=0.30 THC). Conclusions: Families using OCE who completed the study report a 24% responder rate and 14% stopped use of OCE due to a perceived increase in seizures.  The response rate is lower and the withdrawal rate is greater than prior data obtained with retrospective methods as well as prospective randomized trials of pharmaceutical grade products.  These results are limited due to our relatively small sample size and possibly informatively missing data.  Further prospective studies of OCE need to be performed to determine efficacy and safety.  Funding: Colorado Department of Public Health and Environment