Cannabidiol Cuts Seizures in Resistant Pediatric Epilepsy

Damian McNamara

December 17, 2018

An FDA-approved, purified form of cannabidiol (Epidiolex, GW Pharmaceuticals) reduces seizure burden in children with drug-resistant epilepsy (DRE), a new review confirms.

In a new systematic review of the four randomized controlled trials, few children treated with the cannabis compound experienced complete freedom from seizures or a significant improvement in quality of life.

Nonetheless, "we found an increase in the number of children who experienced at least a 50% reduction in seizures, as well as a significant reduction in the frequency of seizures — both of which can be considered a measure of seizure burden," principal investigator Jesse Elliott, PhD candidate at the School of Epidemiology and Public Health, University of Ottawa, Canada, told Medscape Medical News.

"This reduction of seizures is an important finding because any reduction in seizures can have a striking impact on the lives of these children and their families," Elliott added.

The systematic review, which also included 17 nonrandomized studies, was published online December 4 in Epilepsia.

Considerable Burden

About one third of the 50 million patients with epilepsy worldwide live with a drug-resistant form of the condition. This means they failed to respond adequately to two or more antiepileptic-drug regimens.

In addition, "consequences of [DRE] during childhood are catastrophic, with frequent seizures impairing neurological and cognitive development, impacting quality of life, and contributing high costs to the health care system," the researchers note.

A systematic review involving children and adults with epilepsy published in March also demonstrated that cannabidiol was more effective than placebo for improving treatment response and reducing seizure frequency.

However, those researchers also reported that cannabidiol was associated with greater seizure freedom and improved quality of life.

"These apparent discrepancies are likely related to methodological differences," the current investigators write.

Publication of the current research follows the FDA approval this past June of Epidiolex, the first cannabis-based product cleared by the agency, to treat Dravet and Lennox-Gastaut syndromes.

The four included randomized controlled trials included a total of 550 patients. Participants had either Dravet or Lennox-Gastaut syndrome and received placebo or oral cannabidiol (5 to 20 mg/kg/day) for up to 14 weeks. The mean age in each study was between 7 and 16 years.

The 17 nonrandomized studies included 1115 total participants. These patients presented with variable baseline seizure frequency and treatment durations. The mean age range of participants, when available, was 7 to 14 years.

Although the types of cannabis-based interventions varied across the nonrandomized studies, more than two thirds (68%) involved cannabidiol. However, not all interventions were well described — and some included 'artisanal' products, products from illicit cannabis suppliers, or homemade extracts.

Bias Considered

Investigators drew a distinction between the randomized and nonrandomized research based on likelihood of bias.  

"In our review, the randomized controlled trials were all considered to be at overall low risk of bias, meaning that we can have confidence in the results from these studies. With other study designs, we need to be more cautious in interpreting their findings," Elliott said.

"In particular, it is important to consider that many of the nonrandomized studies did not include a concurrent control group of patients who did not receive a cannabis-based product, and so it is difficult to separate out the effects of the treatment from a potential placebo response, or a response that is not due to the medication itself," she added.

The primary outcome was seizure freedom. Secondary outcomes included seizure frequency, quality of life, sleep, status epilepticus, death (all-cause, as well as sudden unexpected death in epilepsy), and gastrointestinal adverse events.

Fifteen of the 21 included studies assessed seizure freedom. For example, one randomized controlled trial of children with Dravet syndrome who received cannabidiol revealed 3 (5%) of 61 patients achieved freedom from seizures. In contrast, no children treated with a placebo were seizure-free during the 14-week study.

In the nonrandomized studies, which ranged from 8 weeks to 33 months, seizure freedom estimates varied from 1% to 20%.

Nine of the systematic review studies assessed seizure frequency. The pooled median difference in monthly seizure frequency in the randomized controlled trials between cannabidiol and placebo was −19.8% (95% confidence interval [CI], −27.0% to −12.6%). The investigators assigned "moderate certainty" to this finding.

Among the nonrandomized studies, the reduction in total seizures was 30% to 90%, but investigators assigned this finding "very low certainty."

Thirteen studies assessed treatment response, defined as a 50% or greater reduction in seizure frequency compared to baseline. The one randomized clinical trial with this endpoint found 32 (37%) of 86 children with Lennox-Gastaut syndrome achieved this response at 14 weeks compared with 18 (21%) of 85 children randomly assigned to received placebo. The researchers assigned moderate certainty to this finding.

The 12 nonrandomized studies showed treatment responses ranging from 24% to 100% over 8 weeks to 75 months. Again, the authors gave this finding a very low certainty.

No Impact on QoL

Three randomized controlled trials assessed quality of life but showed no statistically significant difference between cannabidiol and placebo. In contrast, two nonrandomized studies reported significant improvements relative to baseline associated with cannabidiol therapy.

Researchers in 12 studies looked for changes in sleep. Three randomized studies showed no significant differences in sleep disruption between treatment and placebo. The nonrandomized research was split between reports of improved and impaired sleep.

Three randomized and three nonrandomized studies assessed status epilepticus. Among the randomized trials, the relative risk (RR) for the condition was 1.39 (95% CI, 0.55 ‐ 3.47). This finding was assigned low certainty. The pooled prevalence of status epilepticus was 4% among the nonrandomized reports and was assigned very low certainty.

Six studies reported participant deaths. In one randomized controlled trial, investigators attributed a death to respiratory failure unrelated to treatment. Five nonrandomized trials reported six deaths from various causes, including two reports of sudden unexpected death in epilepsy.

Researchers reported gastrointestinal (GI) events in both types of studies. In the randomized controlled trials, the pooled RR was 1.54 for a GI adverse event, 1.00 for vomiting, and 2.25 for diarrhea. However, this finding was assigned low certainty.

Among the nonrandomized reports, pooled data showed 8% experienced at least one episode of vomiting or diarrhea, with the outcome assigned very low certainty.

"The existing evidence from high-quality randomized controlled trials suggests that cannabidiol probably reduces seizures among children with Dravet and Lennox-Gastaut Syndrome, two severe forms of pediatric drug-resistant epilepsy," Elliott said.

However, she added a caveat. "The randomized controlled trials that have been published to date all involve cannabidiol and the findings should not be extended to other cannabis-based products," she said.

First-Line Therapy?

In addition, the majority of the studies included in the review added cannabidiol to established regimens.

"As such, whether cannabis-based products are effective as first-line therapy for newly diagnosed epilepsy is unknown," the researchers write.

For example, in the randomized studies, 66% of participants received clobazam (Onfi, Lundbeck) in addition to cannabidiol. A higher proportion of children taking both agents experienced seizure freedom (44%) compared with those who took clobazam monotherapy (38%) or cannabidiol alone (33%).

Concomitant therapy raises another caveat. The CYP pathway metabolizes many commonly used antiepileptic medications as well as cannabinoids, Elliott noted.

"This means that when cannabinoids are added on to existing medication regimens, there may be interactions," she said.

For example, cannabinoids may increase the potency and accentuate somnolence with clobazam. With another commonly prescribed antiepileptic, valproate, addition of cannabinoid may lead to elevated valproate levels and clinical signs and laboratory changes indicative of valproate toxicity, she added.

The researchers note past reports of similar cannabidiol effects on other antiepileptic medications as well.

"Monitoring is essential for patients taking cannabinoids in addition to other antiepileptic medications," Elliott said. Many clinicians check patients' liver and kidney function and obtain a full blood count before starting patients on cannabinoid therapy, she said. In addition, repeated testing may be warranted for dose adjustments or emergence of any adverse events.

In terms of a potential limitation, the primary outcome of seizure freedom may not have been the optimal choice, the researchers note.

"Although seizure freedom…was identified by family members and a practicing neurologist as being of utmost importance, it may not be an attainable outcome for children with DRE," they write.

"In this population, a reduced burden of seizures and the avoidance of status epilepticus may be more realistic goals, with a 50% reduction in seizures being a desirable outcome."

Unlike traditional systematic reviews with a cut-off date, Elliott and colleagues consider their current work a 'living systematic review' and are planning to provide regular updates.

"Living systematic reviews are best suited for use in areas where there is ongoing research, such that there is frequently new evidence becoming available, and for topic areas that are of importance for clinical and policy decision-making. The use of cannabis-based products to treat epilepsy in children is one such area," Elliott said.

When she and her colleagues first looked at cannabis-based products for epilepsy in 2015, there were four published studies. The number jumped to 23 studies as of April, "and we know of at least an additional 33 studies registered in," she said.

Most of these ongoing studies involve cannabidiol and children with a wider range of epilepsy syndromes.

This living systematic review will be updated at 6-month intervals, and is part of a larger project about the use of cannabis-based products for the treatment of epilepsy in children. 

"We are currently working on studies of the experiences of parents whose children have taken cannabis for treatment of drug-resistant epilepsy, as well as the perspectives of neurologists who provide care for these children," Elliott said.

"Not a Magic Bullet"

Commenting on the findings for Medscape Medical News, Ley Sander, MD, PhD, medical director at the Epilepsy Society and professor of neurology at University College London, United Kingdom, confirms that "medicinal cannabis is not a magic bullet but, like the choice of antiepileptic medications currently available, could be helpful for some but not for others with drug-resistant epilepsy.

"Of course, the Holy Grail is to achieve seizure freedom, but this is only rarely seen in children with Dravet and Lennox-Gastaut syndromes. However, reducing the frequency of seizures for these children could also be life enhancing," he added.

"I hope that a change in regulations in the UK around cannabis-based products for medicinal use will open the door for more research providing robust evidence for the use of these products in others with drug-resistant epilepsy, too," Sander said.

"What is desperately clear, as always, is the need for more treatments."

Dr Elliott has disclosed no relevant financial relationships. Dr Sander reported being principal investigator for a CBD trial as part of a deal between University College London and GW Pharmaceuticals but has no shares in any company in this area.

Epilepsia. Published online December 4, 2018. Full text

Follow Damian McNamara on Twitter: @MedReporter. For more Medscape Neurology news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.