Subscribe to eTOC

Cannabidiol for Intractable Epilepsy
Promising Open-label Results Increase Interest in Forthcoming Blinded Trials



DR. ORRIN DEVINSKY: “These were very, very refractory patients, and the fact that some became seizure-free was pretty impressive. But we did not have a placebo arm, and the data have to be considered quite preliminary. We are therefore limited in our ability to say what the true safety and efficacy of this drug is.”

Cannabidiol, one of the two major compounds found in marijuana, reduced seizures by one-third in patients with intractable epilepsy, according to a new report of a 12-week open-label trial.

Twelve weeks of treatment with the purified marijuana extract cannabidiol (CBD) reduced seizures by one-third in patients with intractable epilepsy, according to the largest trial of its kind published to date in the December 23, 2015, edition of Lancet Neurology.

The study authors and outside experts called the results from the open-label trial highly encouraging, but stressed that an understanding of the true efficacy and safety of the drug will have to await the completion of double-blind trials, currently under way.

CBD is one of the two major compounds found in marijuana. The other, tetrahydrocannabinol (THC), is responsible for the psychoactive effects of the drug through its interaction with the brain's cannabinoid receptors. In contrast, CBD acts on different receptors, and is not a psychoactive compound, said the principal investigator of the new study, Orrin Devinsky, MD, FAAN, a professor of neurology, neurosurgery, and psychiatry at the New York University School of Medicine.

Dr. Devinsky's interest in cannabinoids for epilepsy was sparked by early observations by Drs. Russell Reynolds and William Gowers, who in the late 1800s reported on the potential benefit of cannabis on seizures; by patients who confided in him that smoking marijuana helped control their epilepsy; and by preliminary evidence from animal studies and small clinical trials suggesting that CBD in particular might be beneficial.

The opportunity to conduct a clinical trial of CBD arose when GW Pharmaceuticals, the manufacturers of Sativex, offered to make a pharmaceutical grade CBD product available for a large-scale trial. Sativex is a 50-50 mixture of THC and CBD, and is approved in Europe and Canada for treatment of spasms in multiple sclerosis. The CBD product, whose trade name is Epidiolex, is 99 percent pure, and dissolved in sesame oil. It was granted Orphan Drug status by the Food and Drug Administration in 2014.


Dr. Devinsky and colleagues at 10 other centers across the United States enrolled 214 patients, ages 1 to 30 years, with severe childhood-onset epilepsy, whose seizures were not controlled despite multiple medications. Patients received oral CBD at an initial dose of 2 to 5 mg/kg daily, which was titrated up to either 25 or 50 mg/kg/day, depending on the center. Most patients remained on stable doses of their other antiseizure medications; they were taking a median number of three antiseizure drugs at the time of enrollment, and as many as seven. In some cases, sedation led to a reduction in doses of other drugs.


CANNABIDIOL interacts with the brains cannabinoid receptors, but is not a psychoactive compound.

“Many of these kids are not even deemed eligible for most other drug studies, because of their high rate of non-response,” said the study co-author Elizabeth Thiele, MD, PhD, a professor of neurology at Harvard Medical School in Boston.

The primary efficacy outcome was reduction in number of motor seizures per month at week 12. Motor seizures were chosen, Dr. Devinsky explained, because they can be more objectively determined by caregivers than impaired awareness or myoclonic jerks.

Of the 214 enrolled patients, 52 did not have 12 weeks of follow-up, leaving 162 for the safety analysis. Of these, 20 percent had Dravet syndrome, and 19 percent had Lennox-Gastaut syndrome; the rest included a variety of etiologies.

Somnolence, decreased appetite, diarrhea, and fatigue were common adverse events, which were likely due to treatment. Somnolence and fatigue may have been worsened by an increase in a metabolite of clobazam, Dr. Devinsky said, which occurs when CBD inhibits a liver enzyme in genetically susceptible individuals.

Gastrointestinal disturbance may have been caused by the sesame oil in which the CBD was dissolved. Five patients withdrew because of adverse events. Status epilepticus (SE) occurred in 12 percent of patients, which some investigators deemed possibly related to CBD. Dr. Devinsky noted, however, that some of these patients experienced SE before treatment as well. “I never thought the seizures were a consequence of the treatment, and similarly animal studies have not shown that CBD exacerbates seizures, or lowers the seizure threshold. But the double-blind studies will be very helpful in exploring this possibility.”

Twenty-five patients were excluded from the efficacy analysis, most because they had no motor seizures. In the remaining 137 patients, the median monthly seizure frequency fell from 30 at baseline to 15.8 at the end of the study. The median reduction over the course of the study was 36.5 percent. Thirty-nine percent of patients had a reduction of 50 percent or more in motor seizures, and 21 percent had a reduction of 70 percent or more. Patients with Dravet syndrome appeared to do somewhat better than average, while those with Lennox-Gastaut syndrome responded right in the mid-range.

“The fact that we saw the kind of response we did means to me that this is a very effective medication,” Dr. Thiele said. “It clearly doesn't work for everyone, but for a subgroup of kids, it's an effective medication, and extremely well tolerated.”

“These were very, very refractory patients,” Dr. Devinsky said, “and the fact that some became seizure-free was pretty impressive. But we did not have a placebo arm, and the data have to be considered quite preliminary. We are therefore limited in our ability to say what the true safety and efficacy of this drug is.”


Michael Privitera, MD, a professor of neurology at the University of Cincinnati, commented that both Dravet and Lennox-Gastaut syndromes are difficult to treat, “so any new drug that works in this population is welcome. The numbers in this open-label study are promising,” he said, “but need to be looked at with a grain of salt, since we always see higher numbers in such trials than in subsequent double-blind trials. These results are roughly in line with other new antiepileptic drugs, but this study was done in a highly medication-resistant group.”

Amy R. Brooks-Kayal, MD, a professor of pediatrics, neurology and pharmaceutical sciences at the University of Colorado in Denver, added that, based on these initial findings, CBD “may become a nice addition to our treatment options for intractable epilepsy.” But, she cautioned, this will depend on the size of the placebo effect in the trial. She noted that 19 percent is a typical placebo response rate in pediatric epilepsy trials. Given the median reduction in this trial of 36.5 percent, she said, “if almost 20 percent of that was placebo response rate, then that may put the true efficacy rate down closer to 17 percent That could still be important in this population.”

Dr. Devinsky made a similar point, noting, “If the placebo rate was 10 percent, we have a phenomenal effect. If it was 20 percent, we're OK. If it was above 30 percent, it's problematic.”

More information about the role of CBD in epilepsy treatment should begin to emerge later this year, as the first of four ongoing double-blind trials conclude and their results become public.

“My guess is that in the next five years a first-in-class cannabinoid drug will be approved in epilepsy, and cannabinoids will be under study for modifications that will improve their efficacy,” Dr. Privitera said.

Physicians familiar with CBD all stress that medical marijuana, even the so-called “artisanal” high-CBD preparations, are no substitute for the pure pharmaceutical. No preparation is free of THC, Dr. Thiele noted, “and there are good data that early exposure to THC is not good for the developing brain,” a problem CBD is not believed to mimic. Furthermore, she said, “the ratios of CBD and THC in these preparations are inconsistent. I tell families I can't even begin to know how to dose those preparations.”


•. Devinsky O, Marsh E, Friedman D, et al. Cannabidiol in patients with treatment-resistant epilepsy: an open-label interventional trial Lancet Neurol 2015 pii: S1474–4422(15)00379-8.
    •. Vogelstein F. One man's desperate quest to cure his son's epilepsy—with weed. Wired 2015: