"The gap between patient beliefs and available scientific evidence highlights a set of factors that confound cannabinoid research and therapy, including the naturalistic fallacy (the belief that nature's products are safe), the conversion of anecdotes and strong beliefs into facts, failure to appreciate the difference between research and treatment, and a desire to control one's care, including access to therapies of perceived benefit."1
— Cannabinoids in the Treatment of Epilepsy, Daniel Friedman and Orrin Devinsky
Cannabis for Kids?
If you believe the “pro-marijuana” lobby, cannabis is nothing short of nature’s wonder drug, virtually side-effect free and with astounding health benefits ranging from preventing Alzheimer’s disease to curing cancer. In 2013, such views gained a foothold in the mainstream when celebrity media doctor Sanjay Gupta famously declared that he had “changed [his previously skeptical] mind about weed” after witnessing the seemingly miraculous effects of marijuana on Charlotte Figi, a 3-year-old girl with intractable seizures despite taking 7 different medications. “Medical marijuana,” Gupta wrote, “calmed her brain,” bringing her seizure count down from 300 per week to just two to three per month.
As a result of a handful of anecdotal accounts like Charlotte’s, the possibility that marijuana might have therapeutic potential has been a hot topic in the media as well as in medical research. While that research is in its infancy, existing enthusiasm and hope has led several states to legalize the use of cannabinoids in children with refractory epilepsy. In turn, parents of children with epilepsy have flocked to Colorado in order to obtain easy access to cannabinoids reputed to be effective for childhood seizures. The most well-known product, called Charlotte’s Web, is manufactured in Colorado, but is now widely available and because of its classification as a dietary supplement can be purchased online and transported legally across state borders. This and other similar cannabis products are typically sold in liquid form (described as “cannabis oil”) or as tablets. Children using them for seizures are therefore not typically smoking marijuana, but ingesting an extracted product in the form of a liquid or pill.
With cannabinoids available that have bypassed the traditional process of Federal Drug Administration (FDA) approval for a medical treatment and the companies that manufacture them profiting from worried patients desperate for a cure, where’s the evidence to support treating childhood seizures with cannabis?
In order to answer that, it’s important to first understand that the anecdotal cases of childhood epilepsy responding to cannabinoids have mostly been reported in kids with extremely rare and difficult to treat seizure disorders such as Dravet syndrome or Lennox-Gastaut syndrome. These conditions are often associated with uncontrollable seizures that begin in infancy and continue recurrently throughout the day unchecked by existing FDA-approved anticonvulsants, with seizures that sometimes result in death. No wonder then that the parents of children suffering from such conditions are beside themselves, feeling let down by what doctors have to offer them and eager for the promise of something new.
Cannabis and Cannabinoids
Next, we have to understand the difference between cannabis and cannabinoids. Cannabis, also known as marijuana, is a plant that contains hundreds of cannabis-specific chemicals called “phytocannabinoids.” The two best-characterized cannabinoids are delta-9-tetrahydrocannabinol (THC) and cannabidiol (CBD). Different breeds or strains of the cannabis plant contain variable amounts of THC and CBD, but the overall trend over the past few decades has been to breed cannabis with increasing amounts of THC and negligible amounts of CBD. This trend has been a consumer driven-change, with high THC and low CBD generally contributing to a more potent “high.”
However, it’s generally thought that the therapeutic potential of cannabis for seizures lies with CBD, not THC. Accordingly, Charlotte’s Web, the cannabis strain developed by the company CW Hemp and marketed for “daily health and wellness,” contains less than 0.3 percent THC. This is a tiny amount compared to the typical cannabis product found in medical marijuana dispensaries that often contain as much as 20-30 percent THC. As such, it is not associated with any euphoric “high” effects and can be legally classified as “hemp,” which allows it to be treated as a dietary supplement rather than a medication. In contrast, the CBD content of Charlotte’s Web is described as “high” although the exact amount does not appear on the product labeling.
While we’re on the subject of cannabis product ingredients, one area of concern is that the purification standards for dietary supplements falls far short of FDA requirements for medications. And most other high-CBD products (there are many others besides Charlotte’s Web) sold in medical marijuana dispensaries aren’t even subject to dietary supplement regulations. The actual composition or cannabinoid content of cannabis products is therefore unreliable at best. A 2015 study of edible cannabis products sold in medical marijuana dispensaries found that less than 20 percent of products were accurately labeled with respect to THC and CBD composition.2 Likewise, the FDA has found that product labeling of high-CBD products is often inaccurate, with some products not containing CBD at all.
What’s the Evidence?
Now that we understand those specifics, what’s the evidence that Charlotte’s Web or other high-CBD products sold in medical marijuana dispensaries are effective for childhood seizures? Beyond scattered anecdotal case reports like Charlotte Figi’s, there is no such evidence. Rigorous testing – for example, through randomized controlled trials like those required for FDA approval of a medication – is not necessary for dietary supplements like Charlotte’s Web or high-CBD products available in medical marijuana dispensaries to be sold (although a study involving Charlotte’s Web is reportedly in the works).
In addition to retrospective case reports, a few surveys of parents with children with refractory seizure disorders have been published that describe a reduction in seizure frequency with CBD treatment.3 But caregiver reports of seizures are poor measures of outcome, with a previous study finding that family reports of seizure have not been well-correlated with electroencephalogram (EEG) recordings of seizures.4 Therefore, subjective reports by parents may carry a significant potential for bias, with perceived effects of CBD representing more of a placebo response than an actual response (see my most recent blogpost on the placebo effect). While a few small placebo-controlled trials involving CBD for seizure disorders were conducted in the 1970s and 80s, these studies involved adults rather than children, enrolled too few subjects to draw firm conclusions (a total of only 30 subjects received CBD across 4 studies), and two of them found no clear benefit for CBD.3
Without controlled clinical trials to support the efficacy of high-CBD products in children, what hope is there for parents looking for more solid evidence to inform their decision of whether CBD might be worth trying? Enter Epidiolex, a pure CBD drug (containing 99 percent CBD and less than 0.1 percent THC) manufactured by GW Pharmaceuticals that was granted orphan drug status (a special status that allows a drug to be used to treat a rare disease even though it has not yet been fully tested and approved for the condition) by the FDA in 2013. Since then, this new medication has been made available to qualifying patients within “expanded access programs” located in within epilepsy treatment centers across the country.
In 2016, the first clinical trial of Epidiolex for treatment-resistant epilepsy was published in Lancet Neurology – an “open-label” study (meaning that it wasn’t controlled or blinded – see my earlier blogpost about the importance of blinding in controlled trials) involving 214 patients with intractable childhood epilepsy.5 Enrolled subjects were 1 to 30 years old (average 10.5 years) and had to have 4 or more countable seizures per day despite being on stable doses of antiepileptic medications in order to participate. Nearly half of the subjects had either Lennox-Gastaut syndrome or Dravet syndrome. Subjects treated with Epidiolex were followed over a 12-week period with the major outcome being seizures reported by families and caregivers. The main study finding was that the number of reported seizures decreased from an average of 30/month to 16/month over the 3-month study, representing a reduction in seizure frequency of about 35 percent.
These results were encouraging to say the least, but still represented caregiver reports without a placebo comparator such that it’s impossible to say with confidence that the reported improvements were reliable or actually occurred as a result of treatment with Epidiolex. Also, while five subjects (4 percent) had no seizures at all during treatment with Epidiolex, just as many had to stop treatment due to an adverse event including some subjects who experienced an increase in seizures. In fact, 9 subjects (6 percent) developed a condition called status epilepticus (a sustained form of unremitting seizure activity) that was rated by investigators as “possibly due to CBD use.”
Just this past month however, a second study was published in the New England Journal of Medicine that significantly adds to the case for Epidiolex in refractory childhood epilepsy.6 Performed by some of the same investigators, this clinical trial involved 120 children and young adults (up to age 18 years) with Dravet syndrome and included a placebo control. Subjects were followed for 14 weeks, with the main outcome again limited to caregiver reports of seizures. Epidiolex-treated subjects experienced a reduction in seizure frequency from 12 to 6 per month on average (a median improvement of about 40 percent), while placebo-treated subjects had only modest negligible improvement, a statistically-significant difference. Overall improvement was rated in 62 percent of Epidiolex-treated subjects and 34 percent of placebo-treated subjects. Although no patients developed status epilepticus in this study, more Epidiolex-treated than placebo-treated subjects dropped out of the study due to adverse events (eight vs. one), with somnolence, diarrhea, loss of appetite, and seizures (“convulsions”) reported more frequently as adverse events in the Epidiolex group.
Overall then, by addressing the potential reporting bias and limited ability to infer causality in the first open label study, this second placebo-controlled trial of Epidiolex does provide promising evidence to support a therapeutic role of cannabidiol in children with refractory seizures associated with Dravet syndrome.
What does this mean for parents hoping for an upgrade in available treatments for their children with refractory epilepsy? Here are some take-home points based on the existing research to date:
► The best data on Epidiolex comes from children with Dravet syndrome, which is a distinct seizure disorder. The evidence from controlled trials that CBD is effective for other types of seizures or for seizures in adults in minimal at this time. Without question, the limited available data argues for additional controlled trials of CBD in other seizure disorders.
► While CBD may be effective for refractory childhood seizure disorders like Dravet syndrome, evidence to support the use of marijuana for seizures in lacking. Although some surveys report benefits from users, evidence from animal studies suggests that THC has may have pro-convulsant properties.7 Seizures have been reported adverse event within published cases of synthetic cannabinoid toxicity (see my previous blogpost on synthetic cannabinoids).8 Accordingly, there is not yet any convincing evidence that cannabis or marijuana is an effective treatment for seizures and some limited evidence suggests that using marijuana could worsen seizures.
► Like any medication, CBD is not free of side effects — some patients don't tolerate treatment with CBD.
► As a highly purified CBD medication manufactured by a drug company, Epidiolex’s clinicial trial benefits cannot be assumed to hold true for other high-CBD products like Charlotte’s Web. Although the currently limited availability of Epidiolex might incentivize parents to seek out other CBD products from other sources, the uncertain composition of CBD products found in medical marijuana dispensaries (including those containing no CBD at all) is good cause for concern. As marijuana is increasingly legalized and as medical research with cannabinoid pharmaceuticals expands, expect the battle over profits between marijuana growers and pharmaceutical companies to heat up.
► In the only two clinical trials of Epidiolex published to date, the percentage of subjects who had no seizures during active treatment was four percent in the open-label study and less than one percent in the placebo controlled study. Epidiolex’s potential to reduce seizure frequency is an important advance in the treatment of children with life-threatening refractory epilepsy, but it’s not a miracle cure.
Disclosure: The author reports no conflicts of interest with regard to this blogpost, including no financial affiliation with GW Pharmaceuticals, CW Hemp, or any other purveyor of “medical marijuana” products.
1. Friedman D, Devinsky O. Cannabinoids in the treatment of epilepsy. New England Journal of Medicine 2015; 373:1048-1058.
2. Vandrey R, Raber JC, Raber ME, et al. Cannabinoid dose and label accuracy in edible medical cannabis products. Journal of the American Medical Association 2015;313:2491-2493.
3. Leo A, Russo E, Elia M. Cannabidiol and epilepsy: Rationale and therapeutic potential. Pharmacologic Research 2016; 107:85-92.
4. Press CA, Krupp KG, Chapman KE. Parental reporting of response to oral cannabis extracts for treatment of refractory epilepsy. Epilepsy & Behavior 2015; 45:49-52.
5. Devinsky O, Marsh E, Friedman D, et al. Cannabidiol in patients with treatment-resistant epilepsy: an open-label interventional trial. Lancet Neurology 2016; 15:270-278.
6. Devinsky O, Cross H, Laux L, et al. Trial of cannabidiol for drug-resistant seizures in the Dravet syndrome. New England Journal of Medicine 2017; 376:2011-2020.
7. Karler R, Calder LD, Turkanis SA. Prolonged CNS hyperexcitability in mice after a single exposure to delta-9-tetrahydrocannbinol. Neuropharmacology 1986; 25:441-446.
8. Courts J, Maskill V, Gray A, et al. Signs and symptoms associated with synthetic cannabinoid toxicity: systematic review. Australasian Psychiatry 2016; 24:598-601.