brands of cbd oil for kids with epilepsy

Brands of cbd oil for kids with epilepsy

When you’re driving across the country with a stash of marijuana in your trunk, you follow the speed limit. You signal when changing lanes. You might even pick a route that skips Colorado, because ever since recreational pot was legalized there, police just over state lines have been on the lookout for anyone ferrying the drug from the area.

But the Colorado-free route from California, where you bought your marijuana, to the Northeast, where you live, presents a curveball. Cruising along I-40 in Arizona, you encounter a border patrol checkpoint. “Good evening, officer,” you say. A German shepherd approaches your vehicle and somehow doesn’t detect the marijuana that’s under a pile of ice in a cooler. As you’re sent on your way, adrenaline pulses through your body. Tears pool in your eyes.

You are, after all, committing at least several state and federal crimes, but when you get home a few days later, it’s business as usual. Three times a day, you or your wife will squirt 2.5 milliliters of marijuana oil into the mouth of your severely epileptic daughter. As a baby, the girl was diagnosed with infantile spasms, a neurological disorder that causes frequent seizures and long-term damage to brain development. Multiple times a day, she’s “like a robot whose plug has been pulled,” says her father. None of the drugs she’s been prescribed have helped and although she can walk, the girl, now 3, does not speak. She goes to a school for kids with special needs.

Watch TIME’s feature documentary on children and medical marijuana

The girl was having up to 200 seizures per day when her parents heard anecdotal reports that marijuana could reduce or even stop seizures. Since beginning her marijuana oil regimen earlier this year, the girl’s episodes are down to 30 per day at most, according to her father.

These parents are in good company. About 100,000 U.S. children have intractable epilepsy—a treatment-resistant category of the disease characterized by uncontrolled seizures—and for some of their parents, medical marijuana has gained a reputation as a wonder drug. Fueled by success stories on Facebook and family blogs, these parents are acquiring marijuana through quasi-legal and illegal means, giving their children a derivative oil low in tetrahydrocannabinol (THC), which causes psychoactive effects, but high in cannabidiol (CBD), one of the hundreds of other compounds found in the plant.

And yet there is little science about the safety or efficacy of treating children with CBD. Medical marijuana is legal in 23 states, but federal law still classifies the plant as a Schedule 1 drug, “with no currently accepted medical use and a high potential for abuse,” according to the Drug Enforcement Agency. (Heroin and LSD are also Schedule 1 drugs.) This means that, even though doctors and researchers have suspected for decades that some compounds in marijuana might effectively treat epilepsy and other hard-to-cure conditions, independent clinical trials that include human subjects—let alone minors—are scant and require approval from three separate federal agencies.

Since 1999, these agencies have approved just 16 independent studies of medical marijuana on humans. The lack of research presents a kind of catch-22: Without much scientific study of marijuana, the government has been unwilling to recategorize the drug. But the categorization as a Schedule 1 drug makes testing it on humans extremely difficult. This also presents parents with little expert advice. “The only guidance we were getting was the community on Facebook,” says Lori King, the mother of an epileptic 10-year-old boy whom she’s treated with marijuana oil since 2013.

The little research that does exist raises questions about the effectiveness of CBD therapy for kids. A recent study, for instance, found that epileptic seizures were significantly reduced in just a third of children studied. Experts are also concerned that in a largely unregulated business, contaminants like pesticide residues and molds could lead to adulterated versions of an otherwise potentially low-risk drug.

“All of a sudden you have these parent groups, which traditionally you’d think would be in opposition to any kind of liberalization of marijuana policies, and they have become the biggest advocates in favor of access,” says Amanda Reiman, manager of the marijuana law and policy unit for the Drug Policy Alliance.

Some of these parents are hoping pot can help where mainstream medicine has failed. Epilepsy costs individuals and institutions $15 billion a year. It is far more common than autism, multiple sclerosis or a host of other neurological disorders. And it kills more Americans every year than breast cancer—and yet the disease receives just 20 percent as much research funding from the National Institutes of Health. What’s more, a third of people with epilepsy have an intractable drug-resistant type.

“We’ve introduced a dozen new drugs in the past 20 years, but it’s not clear we’ve made a significant advance in the treatment of drug-resistant epilepsy,” says Dr. Orrin Devinsky, head of the Comprehensive Epilepsy Center at New York University. “We have failed as a scientific and medical community.”

In the meantime, parents admit they are, in effect, experimenting on their children. Some of them do so with a doctor’s help, tracking how the compound interacts with other drugs and measuring progress over time. But just as often, it seems, parents are going it alone.

The parking lot of Noho’s Finest, a medical marijuana dispensary in a gritty industrial area of the San Fernando Valley, is packed most days. From open to close, customers, most of them men, parade into the emporium where everything from bud to THC-laced cookies and pipes are for sale.

At the edge of the busy parking lot, up a plywood ramp and behind a black metal door, there’s an office belonging to Realm of Caring’s Ray Mirzabegian, the go-to guy in Southern California for medical marijuana as a treatment for epilepsy.

Inside, Mirzabegian, a former optician, plays New Age music on a loop and the iPhone on his desk buzzes every few minutes with text messages. He arrived at his current profession after personal experience convinced him marijuana could better curb seizures than anything in mainstream medicine.

Mirzabegian’s daughter Emily had her first seizure when she was five months old. By the time she was four, she was having hundreds of seizures a day. “Our lives were 60 percent in the hospital for years,” says Mirzabegian. Each seizure a child has can cause damage to the developing brain. Emily had been fluent in English and Armenian, her parent’s native language, until one day, around age 6, “It was like somebody pressed the mute button,” says Mirzabegian.

Over the years, Emily has been prescribed 12 separate medications—none seemed to help and most caused side effects, including depression, vision loss, insomnia and lethargy that left her “sitting on the couch in a vegetative state,” says Mirzabegian. The family tried a ketogenic diet, which consists largely of fat, as well as acupuncture and Chinese medicine. They sent Emily’s medical records to doctors in France and Iran, and even flew to the Dominican Republic where a doctor injected Emily with a pink liquid he claimed was filled with stem cells that would cure her epilepsy. The cost: $30,000 in traveler’s checks. “We were so desperate we even went back for a second round,” says Mirzabegian.

Meanwhile, the family met with eight neurologists, eventually finding Raman Sankar, chief of pediatric neurology at the University of California, Los Angeles, who diagnosed Emily with Dravet Syndrome, a type of epilepsy that, with related causes, kills up to one-fifth of sufferers before age 20.

One night, on a Discovery Channel show, of all places, Mizabegian learned of a father giving medical marijuana to his epileptic son. He was intrigued. Mirzabegian went to a doctor, claimed he had chronic headaches and got a medical marijuana recommendation, allowing him to buy the drug in California.

The first strains he tried on Emily didn’t work, but he later learned, via a documentary on CNN, about a family in Colorado called the Stanleys who were growing Charlotte’s Web, a kind of marijuana they said was effective against epilepsy. Mirzabegian decided to give pot another shot, and with the Colorado marijuana, he says, the results were astounding.

Nearly two years after she began a three-times-a-day treatment plan, Emily is down to four seizures per month and is off nearly all the prescription drugs. She loves to swim and go to Chuck E. Cheese. She’s still developmentally behind her peers with the verbal skills of a two year old, but she is waking up to life for the first time since she was an infant, Mirzabegian says.

He was so impressed with the results he decided to go into business with the Stanleys. Mirzabegian became their first licensee outside of Colorado.

Mirzabegian now has his own facility, and the oil he gives Emily is from the plants he grows himself; he sells the rest to parents who visit his office. Demand for Mirzabegian’s product is fierce, with a waiting list of 1,000 families. Although other dispensaries sell CBD-rich oil and charge prices that can run into the thousands of dollars per month, Mizabegian charges 5 cents per milligram of CBD, with dosages that vary based on the child’s weight. He says he is not yet turning a profit.

Among the patients who came to Mirzabegian’s office on the days I visited was the mother of a severely developmentally delayed 5-year-old boy who weighs 20 pounds. She paid $65 for a month’s supply of Charlotte’s Web. There was also a single mother who lives in fear that her son’s doctor, who opposes the use of medical marijuana in place of pharmaceutical drugs, might report her to child protective services. The boy has intractable epilepsy and severe cognitive and physical delays. “There’s no quality of life. This is like living death,” the mother says later. “He’s so beautiful and sweet, but it’s just heartbreaking.”

One by one, Mirzabegian interviewed each parent, jotting down notes about their kid’s medications and writing up dosage charts. During appointments, he also talked about Emily, who all the parents knew of through the various Facebook support groups Mirzabegian runs. “I’m there with you,” he tells them. “We’re all in the same boat.” Mirzabegian strongly recommends his customers undertake a CBD treatment plan in cooperation with a neurologist.

During one visit, Mirzabegian recommended increasing the CBD dose for Lori King’s son Emeric, who had his first seizure at age 4. Even with his current regimen, which costs $147 per month, King says Emeric is “more present and totally more aware. It’s like he’s waking up from a groggy sleep he’s been in for years.”

Despite the change, King says she’s still not completely at ease giving her son CBD oil. “It’s just this gamble based on what everybody else is doing,” she says. “It is an unknown, but there’s that glimmer of hope that this might just be the right ticket.”

“We don’t actually smoke weed,” says Joel Stanley, the eldest of six brothers working in the family’s medical marijuana business in Colorado. With backgrounds in engineering, business and forestry, the Stanleys started growing medical marijuana in 2009 and opened two dispensaries in Colorado Springs the following year, motivated, they say, by a desire to help sick people. The business opportunity was also alluring. Colorado’s current medical marijuana industry brings in some $30 million per month in sales.

Shortly after they got into the business, the Stanleys stumbled upon a strain of marijuana with a high concentration of CBD but very little THC—the compound that gets you high. They called the plant Hippie’s Disappointment and didn’t bother selling it. But in 2012, the mother of a girl with intractable epilepsy approached them, asking for CBD-rich oil. For the girl, Charlotte Figi, the results were remarkable, according to her parents. Charlotte went from hundreds of seizures per day to almost none. Once a writhing, immobile, non-verbal child, she suddenly began walking and talking. The Stanleys renamed the strain Charlotte’s Web.

After the CNN documentary they were featured in—which is controversial among pediatric neurologists—the brothers were inundated with requests for the drug, which at the time was only available in Colorado, where medical marijuana has been legal since 2000 and where recreational marijuana was legalized in 2013. The brothers, who sell Charlotte’s Web at the same price as Mirzabegian—5 cents per milligram—say they have a waiting list of more than 12,000 families, with many relocating to the state to access the product. Five cents a milligram may sound like nothing, but the 17 acres of the strain they harvested in September are poised to produce $23 million worth of oil. (Joel Stanley says the brothers have incurred large debts on lab equipment needed to make their oil, among other infrastructure, and they plan to reinvest the windfall into further expansion rather than pocket the money.)

Meanwhile, to eliminate a chunk of their waiting list and lower costs, the Stanley brothers have decided to test the limits of existing drug laws. They still sell THC-rich pot through their medical marijuana dispensaries, but are now calling Charlotte’s Web something else: “hemp.” The plant is less than 0.3 percent THC, which meets the federal legal definition of hemp and mirrors concentrations of hemp oil already available in U.S. grocery stores that is imported from abroad. Calling the plant “hemp” means the Stanleys can grow far more of it and, they think, legally ship Charlotte’s Web across state lines, with the first bottles of oil scheduled to leave Colorado in November. Under a federal law proposed in June by Republican Rep. Scott Perry of Pennsylvania, making and shipping CBD-rich oil within the U.S. would be explicitly legal. For now, however, shipping domestically produced hemp oil from state to state is, at best, a legal gray area.

“If we can operate in that gray area to help a bunch of kids who are out of options, so be it,” says Joel. The worst case scenario: their entire supply of Charlotte’s Web is seized, the thousands now receiving the medication are cut off and the Stanleys go to jail. “The first few years in this industry, I had nightmares every night of my kids watching me dragged out of the house in handcuffs,” says Joel. “But after we saw Charlotte and all these patients’ successes, it just became very real that you are living a life worth living and you are doing the right thing.”

Beyond the legal perils, many doctors across the U.S. say there’s another hurdle the Stanleys and other growers must clear before CBD should be widely available: proof that it works and that it’s safe.

If there’s a ground zero for the collision of kids on pot and the medical establishment, it’s Children’s Hospital Colorado in Denver. The facility sees more children treated with medical marijuana, including Charlotte’s Web, than any place in the country, according to Dr. Amy Brooks-Kayal, the hospital’s chief of pediatric neurology and a professor at the University of Colorado.

Brooks-Kayal, who will become president of the American Epilepsy Society in December, is among those worried about the unintended consequences of giving CBD oil to kids. “Some of the anecdotal reports are quite exciting,” says Brooks-Kayal, “but really there is no good clinical information on the effectiveness of any form of marijuana for the treatment of epilepsy.” It’s not just the lack of reliable science on how medical marijuana works that troubles Brooks-Kayal. New unpublished research conducted at Children’s and provided, in summary only, to TIME suggests that many of the epileptic children receiving medical marijuana are not benefiting from the drug at all.

In a small study of 75 children receiving CBD-rich oil and being cared for at Children’s, researchers report in the study summary that about one-third had a reduction in seizures exceeding 50 percent, the standard threshold for treatment efficacy in epilepsy. Among those whose seizures were reduced by CBD, researchers reported that the children’s electroencephalographies (EEGs), which measure electrical activity in the brain, remained the same—which is not uncommon but concerns Brooks-Kayal all the same. The study found that families who moved to Colorado for access to medical marijuana were three times as likely to report at least a 50 percent reduction in seizures, suggesting placebo could be a factor.

The Children’s study did not note the kind marijuana oil children were being given, which makes replicating the findings in future research—the ideal with any scientific research—impossible. This brings up another conundrum: the popularity of Charlotte’s Web has some parents scrambling for any medical marijuana touted as having high levels of CBD, even if it’s less or more potent, or includes contaminants that could be dangerous for children to ingest.

“Some days I think it completely helped and other days I feel like it’s no different,” says Carrie Baum, a Los Angeles-area mother who tried CBD-rich oil on her 6-year-old son after an anti-epilepsy drug turned him into “a small zombie child,” as she puts it. “We’re committed to it enough to keep trying and keep working through it. What else is there?”

A soon-to-be-published online poll conducted by the journal Epilepsia shows that among neurologists, including those who specialize in epilepsy, 34 percent said there was sufficient safety data on medical marijuana. Among non-doctors, the figure was 96 percent.

Some science may soon be forthcoming. Earlier this year, British company GW Pharmaceutical launched an FDA-approved a trial of a marijuana-derived CBD drug called Epidiolex. The trial is being conducted at NYU, Massachusetts General Hospital, the University of California, San Francisco (UCSF), and soon, UCLA.

Researchers at NYU, led by Devinksy, and UCSF reported in June that of 27 patients treated with Epidiolex for three months, 13 experienced at least a 50 percent reduction in seizures. Four were seizure-free after 12 weeks. These results, while exciting, are lower than the 70 to 80 percent efficacy rate the Stanleys and Mizabegian have reported among the pediatric epilepsy patients taking Charlotte’s Web—albeit without the kind of scientific rigor that pharmaceutical drugs need to go through. The Epidiolex trial results were also muddied by the fact that patients in the trial also take other prescribed pharmaceuticals. Under GW Pharmaceutical’s timeline, FDA approval could come in late 2016 or early 2017.

One of the Epidiolex study subjects turned up at Mirzabegian’s office on one of the days when I visited. A mother of a 5-year-old epileptic girl brought a bottle of Epidiolex with her and said the drug had not been effective. Unlike Mirzabegian’s thick black oil, the Epidiolex looked like water and contained additives that made it smell like strawberries. Mirzabegian was fascinated. “How do they get it so clear?” he wondered, turning the bottle around in his hands.

Mirzabegian welcomes the research—but it won’t change what he does in the meantime. “You can bring studies that were done in universities—which is great, we need that to happen,” says Mirzabegian, “but I would never discount the research that comes from thousands of parents who have the same exact problem who come together and share their stories.” Doctors and scientists, he says, “have no right—especially when they know everything in their toolbox has failed—to come and point fingers and say there is no evidence, there is no study, no this or no that. That’s their job.”

There’s a race afoot among those trying to bring non-FDA-approved CBD-rich oil to mass market and the clinical drug trials now underway. In addition to the Stanleys, companies peddling natural supplements are also jumping in, along with new marijuana-centered businesses betting that domestically produced hemp and CBD will soon be a multi-billion dollar industry. In 2013, Josh Stanley, the former spokesman for Charlotte’s Web, left the family business to start a non-profit organization and a for-profit company planning to sell CBD-rich oil as a nutraceutical, a product category that falls outside the regulatory rules of the FDA.

Such an approach troubles UCLA’s Sankar, who has worked as a paid consultant to a pharmaceutical company working to develop a synthetic form of CBD. Sankar treats scores of children taking CBD, Emily Mirzabegian among them. “If I’m writing a prescription, I’m going to write it for something that’s approved by the FDA,” he says. Taking a drug out of the FDA approval process because it’s easier to bring to market through a loophole means far less testing is required to measure purity and effectiveness. Mirzabegian says this is a double standard, pointing out that of the 12 drugs prescribed to Emily, eight were not approved by the FDA for use in children.

Although many of the parents who spoke to TIME insisted on anonymity, none refused to be interviewed. In the absence of medical guidance or research, these parents swap tales and experiences on Facebook pages and online discussion boards, eager to share miraculous stories of how their children and families have been transformed by CBD.

The Northeast father who drove marijuana oil across the country says his daughter is now playing and interacting with others for the first time in her life. She is learning to pull up her own pants, take off her own shoes and speak. On a recent trip to a Disney store, she accidentally knocked an object off a shelf and her parents were shocked that she stopped and tried to put it back in place. Even typical toddler temper tantrums have been a welcome sight. “She used to be on so many different drugs that she would just be dragged along wherever we went,” says her father.

The girl is on the waiting list for Charlotte’s Web and her parents are hoping the Stanleys’ plan to ship their product from Colorado to other states pans out. But if it doesn’t, they’re planning another cross-country car trip to stock up on the oil when their supply runs out.

“When it comes to medical marijuana, nobody knows what happens long term,” says the girl’s father. “But since she was nine months old, we’ve had to make choices about her life that we haven’t wanted to make. We put her on a steroid, which bloated her to no end. She was cranky and it delayed her. We put her on anti-seizure drugs that had their own side effects.” These included potential loss of eyesight, liver damage and, in the case of one drug, risk of death.

“We’re doing the best we can with all these decisions,” says the father, “but all these decisions stink.”

Cannabinoids for pediatric epilepsy? Up in smoke or real science?

Public interest in the use of “medical marijuana” for the treatment of childhood epilepsy has burgeoned in the last few years. This has occurred in parallel with a growing interest in “medical marijuana” in general. Physicians and pediatricians must balance their patients’ desire for immediate access to these products with the tenets of evidence-based medicine. This review discusses the biochemistry of cannabis products (the phytocannabinoids) setting this in the context of the endogenous endocannabinoid system. The differing and potentially modulating effects of delta-9-tetrahydrocannabinol (THC) and cannabidiol (CBD) are reviewed. The evidence-base supporting or not the use of cannabis products for the treatment of neurological disease and specifically epilepsy is explored. The potential for adverse effects and particularly of neurotoxicity is addressed. Finally, public health and sociocultural implications are touched upon. Specific recommendations for interested physicians are provided including advocacy for patients and for a change in the “scheduling” of cannabis in order to better foster much-needed high-quality scientific research in this important area.


In just the past 5 years, the remarkable immediacy and reach of modern social media has resulted in a nearly unparalleled expansion of interest in the use of “medical marijuana” products for the treatment of pediatric epilepsy. Until recently this therapeutic option had been largely the fancy of a few cannabis enthusiasts and some dedicated scientists. However, the experiences of two young children with Dravet syndrome who reportedly ceased having seizures and experienced “neurological awakening” after taking cannabidiol (CBD)-rich medical marijuana preparations, led to a public interest in these products that has spread like wildfire (1-3). This phenomenon has rapidly resulted in high-visibility media productions (1), remarkable shifts in public policy (4), legislation of specific medical marijuana laws in multiple states (5), and a high level of interest in these products among physicians, medicinal chemists, and pharmaceutical companies.

These remarkable anecdotal experiences, fueled by an impassioned furor among patient families and advocates, have led to a very high level of expectation regarding the therapeutic potential of cannabinoids for the treatment of epilepsy (3). Families have petitioned their legislators for access to artisanal (vernacular) marijuana products (5), and some have even uprooted their entire family, moving to states with more liberal marijuana policies (6), in order to gain access to these products for their affected children. Meanwhile, several companies are actively developing pharmaceutical products based on medicinal cannabinoids. Physicians are often caught in a quandary complicated by insufficient scientific data. Whereas their patients and families are often demanding access to these products, neurologists and pediatric neurologists may not have the knowledge base or resources to properly address the patient’s concerns nor to advise them in an informed manner (7-9).

This review will seek to provide neurologists and pediatric neurologists a basis to better address these rapidly evolving questions. More than likely, cannabinoids and/or their synthetic derivatives will become a persistent component of our therapeutic arsenal. Hopefully, a solid understanding of cannabinoid chemistry, the endocannabinoid system and the medical evidence surrounding use of cannabis products for the treatment of epilepsy, ultimately can help guide physicians in caring for their patients.

Cannabinoids and chemistry

The plant, Cannabis sativa, often referred to as hemp or marijuana, has been used for its medicinal properties for millennia (10,11). Besides it psychogenic properties, it has been purported to be beneficial for the treatment of a broad range of medical ailments (10-12). The plant contains more than 60 distinct biochemicals which share a common structure most of which presumably have particular bioactive properties (10,11,13,14). Collectively, these are known as the phytocannabinoids. These are terpeno-phenolic compounds, based on their chemical structure (11,13) ( Figure 1 ). The two major phytocannabinoids ( Figure 1 ) are delta-9-tetrahydrocannabinol (THC), the main psychoactive constituent of the marijuana plant, and CBD, a phytocannabinoid that is believed to have no psychoactive properties and is of increasing interest with respect to its therapeutic potential (11-14). The numerous other phytocannabinoids are lesser constituents and are less well studied. However, some may also have medicinal attributes of interest (such as cannabidivarin, which may have antiepileptic properties of its own (15,16).

Terpene phenolic heterocyclic structures of delta-9-tetrahydrocannabinol (THC) and cannabidiol (CBD). Red portions identify basic terpene (left) and phenol (right) backbones.

There is considerable confusion concerning the Cannabis plant and its phytochemicals. According to most authorities, “hemp” and “marijuana” refer to different varieties of the same species (C. sativa) (10). Horticultural practices and breeding strategies have led growers to develop varieties that differ tremendously in the relative and absolute content of various cannabinoids and specifically of THC and CBD. In general, plants grown with less light in more tightly grouped plantations are taller, have a higher fiber content and are bred to have a low (less than 0.3%) THC content (10,11,13,14). These are referred to as “hemp.” Other varietals particularly those grown with greater space between plantings and with greater light exposure can result in plants that produce much higher THC concentrations (colloquially referred to as “marijuana”) (10,11,13,14). The portions of the female flowering plants known as trichomes provide the highest yield of phytochemicals (13). These along with leaves, flower and oil are the primary source for “recreational” and medicinal preparations. Consequently, as will be further elaborated upon below, “medical marijuana” is not “one thing.” Rather, there is enormous variability among various preparations, and even among the different pharmaceutical products derived from natural or synthetic cannabinoids. Hence, the ease with which some individuals and organizations tout the benefits and safety of “medical marijuana” either naively ignores or knowingly disregards the actual complexity of the field (17). In this paper, cannabis products that are derived from marijuana or hemp plants and are not subjected to rigorous pharmaceutical grade purification and quality control will be collectively referred to as “artisanal” or “vernacular” preparations.

Because of the high use and abuse of cannabis products, a great deal is already known about the pharmacological properties of these substances (11,13,14). The cannabinoids all share the heterocyclic terpeno-phenolic structure. As large heterocyclic structures they are very lipophilic. Thus they cross the blood brain barrier readily and distribute easily to lipid laden tissues including brain parenchyma and neuronal cell membranes specifically. They may remain in such lipid laden tissues (presumably including the brain) for weeks and from these are released only gradually into the blood stream. CBD and THC, when taken orally, undergo first pass metabolism, thereby affecting bioavailability and dosage (11). Due to this phenomenon, different routes of administration (oral vs. mucosal vs. inhalational) may result in substantially different biological levels of these agents. Furthermore, in the cannabis plant these substances naturally occur in the relatively inactive carboxylated state. Smoking and heating decarboxylates the molecules thus conferring greater bioactivity (11).

Finally, the cannabinoids and CBD in particular are primarily metabolized by the hepatic cytochrome P450 enzyme system, and in turn CBD appears to be an inhibitor of several of the microsomal hepatic metabolic enzymes and in particular of CYP2C19 (11,18). Thus, it is possible that CBD in significant concentrations may increase levels of concomitantly administered drugs metabolized by CYP2C19. This applies to the benzodiazepines and in particular to clobazam whose major active metabolite N-desmethylclobazam is primarily metabolized by CYP2C19 (19,20). Thus seizure control and/or toxicity may result from pharmacokinetic interactions as well as potentially from the direct CNS effect as such of the cannabinoid. Whether or not this degree of enzyme inhibition will result in clinically significant drug interactions is nevertheless still uncertain.

The endocannabinoid system

The very existence of natural substances (the cannabinoids) that have such remarkable and broad effects on human behavior and function presupposes a set of target receptors or endogenous physiological processes upon which these chemicals act. Indeed, the search for the biological targets of THC ultimately lead to the discovery of specific cannabinoid receptors and then to the identification of endogenous ligands for those receptors. Further study of these has greatly enhanced our understanding of the extraordinarily complex and elaborate endogenous system whereby cannabinoid receptors and other targets respond to the endogenous “endocannabinoids” and to the exogenous substances elaborated by the cannabis plant. This complex physiological system is referred to as the endocannabinoid system (21,22).

The two most prevalent cannabinoid receptors are both G-protein-coupled receptors that exert their physiological effects through the adenylate cyclase second messenger system. The CB1 receptor is widely distributed in brain and largely modulates endocannabinoid effects in the CNS (21-23). These receptors are primarily situated pre-synaptically on axon terminals, with their highest density being in the perisynaptic region ( Figure 2 ) (21,22). Activation of the CB1 receptor results primarily in the inhibition of neurotransmitter release (21,22). While CB1 receptors are known to regulate both GABA and glutamate release, a greater density of these receptors exists on inhibitory versus excitatory synapses in most brain regions (21,22).

Schematic of a GABAergic synapse modulated by CB1 receptors. The perisynaptic location of CB1 receptors (A) is depicted. GABA (E) stimulates post-synatpic GABA receptors (F). Post synaptic changes induce metabolism of membrane derived phospholipids (G) leading to formation of endocannabinoid (H). The latter diffuses back stimulating the perisynaptic CB1 receptor (A). This in turn modulates neurotransmitter release from presynaptic vesicles (D) via G-protein coupled (B) influence on Ca++ channels (C).

CB2 receptors on the other hand are particularly prevalent on lymphocytes, neoplastic cells and other systemic target tissues (12,21,22,24). Presumably, many of the systemic effects of cannabinoids result from binding to CB2 receptors (22,24). The physiological role of CB2 receptor activation in the periphery is suspected to be similar to the neural modulatory role described above for CB1 receptors though details are not as clearly worked out. In addition to these two best characterized cannabinoid receptors, it is believed that the endocannabinoids and phytocannabinoids may act at a number of other receptor or target sites including GPR55 receptors and TRPV type 1 channels (11,12,25). THC appears to act at CB1 and CB2 receptors as a partial agonist (21-23), whereas CBD appears to have a very low affinity for both of the major cannabinoid receptors (23; see below).

This receptor distribution is ideally situated to serve in the capacity of neuromodulation (21,22,26). Indeed a number of complex neuromodulatory processes have been linked to endocannabinoid processing including presynaptic modulation, retrograde neuromodulation and multiple physiologically defined forms of synaptic plasticity including depolarization-induced suppression of inhibition (DISI), depolarization-induced suppression of excitation (DSE) and long-term depression (LTD) (21,22,26). The idea is that synaptic depolarization or hyperpolarization of a post-synaptic membrane results in the induction of post-synaptic metabolic processes which increase the formation and/or release of endocannabinoids. These in turn diffuse back to the presynaptic terminal activating perisynaptic presynaptic CB1 receptors, the activation of which in turn inhibits release of additional neurotransmitter on variable time scales (thereby constituting a form of feedback inhibition) (26). This feedback process may be important during periods of intense synaptic activity.

The two major endocannabinoids are metabolic products of membrane phospholipid metabolism (i.e., of arachidonic acid pathways) (11,13,21-25). These are N-arachidonoyl ethanolamide (AEA; also known as Anandamide, reportedly meaning “inner bliss” in Sanskrit) and 2-arachydonoylglycerol, also referred to as 2AG ( Figure 3 ). 2AG is a full agonist at both CB1 and CB2 receptors and is found at substantially higher levels than AEA, which is a partial agonist at both CB receptor types and at TPVR1 receptors (11,13,21-25). CBD may act by increasing levels of one or more of the endocannabinoids (presumably by interfering with their metabolism or by inhibiting re-uptake; 23). In any case, one can conceptualize these two endogenous cannabinoids as another set of second messenger systems derived from phospholipid metabolism all of which play key roles in neuromodulation.

Chemical structures of the two major endocannabinoids: N-arachidonyl ethanolamide (AEA; “anandamide”) and 2 arachidonoylglycerol (2AG). Similarity to arachidonic acid and arachidonic acid metabolites is apparent.

Therefore, overall the evidence suggests that endocannabinoid signaling serves to decrease synaptic transmission during periods of intense cellular activity. The key features of this system include: (I) neuromodulation; (II) widespread CNS effects via CB1 receptor stimulation; (III) widespread systemic effects including anti-inflammatory and immune mediated effects modulated via CB2 receptors; (IV) neuromodulation via inhibitory presynaptic effects with greater influence over inhibitory than excitatory neurotransmission; (V) a pivotal role in retrograde neurotransmission and resultant presynaptic neuromodulation; and finally, (VI) a biphasic effect of the endocannabinoids in numerous physiological systems (21,22,26).

Cannabinoids for neurological disease: what is the medical evidence?

Given the widespread distribution of cannabinoid receptors in brain and body and the key role these receptors play in the modulation of physiological functions, one might anticipate a very broad range of physiological effects resulting from endocannabinoid signaling (12,27). Indeed these range from roles in pain and sensory modulation to vegetative functions, endocrine regulation, and neurophysiological and psychological functions ranging from motor control to mood and behavioral regulation ( Table 1 ). Likewise, it is easy to see why the exogenously administered cannabinoids might have far ranging effects and hence potential broadly distributed therapeutic potential ( Table 2 ). Consequently, the putative therapeutic benefits of the cannabinoids range from treatment of nausea and vomiting, cancer therapeutics to the modulation of neurological and psychiatric disease (12) ( Table 2 ). It is important to recognize, however, that at this stage these extensive putative therapeutic benefits touted by the proponents of medical marijuana are largely unproven (8,9). At present, high quality medical evidence supporting the use of these agents in most of these conditions remains modest at best.

Table 1

Pain/sensory modulation
Cognitive/memory processing
Mood and behavior
Motor control/coordination
Endocrine functions
Vegetative functions
   (I) Appetite
   (II) Temperature control
   (III) Heart rate regulation
   (IV) Nausea/vomiting
Intraocular pressure
Immune regulation/recognition

Table 2

   Cancer chemotherapeutic agents
   Inflammatory bowel disease, rheumatic diseases, etc.
Neoplastic disease
   Chemotherapeutic agents?, symptomatic (pain, nausea)
Psychiatric/neurological disease
   MS—spasticity (indication in many countries, not US)
   Pain, addiction Rx, anxiety, depression, etc.

Likewise, the widespread physiological effects of the cannabinoids also presuppose the potential for a broad range of toxicities. Thus enthusiasm for therapeutic benefits needs to be tempered by the realistic appreciation of potential adverse effects ( Table 3 ) (4,28,29). Toxicities can be related to direct pharmacologic toxicity of the cannabinoids, including systemic as well as central nervous system toxicity, as well as to coincident harm resulting from associated products inadvertently co-administered with various preparations of “medical marijuana” (8,28-30) ( Table 3 ). The latter could include inhalational injury from smoking, inhalation of associated microorganisms or fungi, inadvertent intake of co-administered pesticides or other byproducts, and so forth.

Table 3

General toxicities
   Pulmonary toxicity (inhaled formulation)
   Pulmonary infection
    Inhalation of associated microorganisms/fungi
   Immune suppression
    Suppresses macrophages, T-lymphocyte
    Increased interleukin I release
   Associated intake of pesticides/byproducts
Neurologic and neuropsychiatric toxicities
   Memory impairment
    Short and long term
   Impaired executive functions
    ↓ concentration, judgment, attention span, motivation, problem solving, reaction time
   Impaired motor function
    ↓ coordination
   Impairment of neural plasticity
    Anxiety, panic attacks
      ? With or without predisposition
    Mania, manic episodes
      ↑ cycling in bipolar
      “gateway drug”

It is important to recall, again, that the cannabis plant produces a remarkably broad array of phytochemicals. It would be anticipated that these could have varying degrees of efficacy as well as toxicity. Specifically, there is considerable evidence that suggests that of the phytocannabinoids, delta-9-THC is likely to be the substance largely responsible for most of the systemic and neurotoxic effects of cannabis preparations (13,14,27). Meanwhile, there is growing evidence that CBD may actually inhibit, reduce or moderate some of these adverse effects (27,31-33). The principal that one endocannabinoid may modulate or act synergistically with another is referred to as the “entourage effect” (13,27). This is the principal that leads some medical marijuana advocates to insist that the “natural product” is preferable to any purified cannabinoid based on the strongly held belief that the synergistic effects of the constituents will be favorable in comparison to the effects of any isolated phytochemical. Obviously, from a scientific standpoint, it becomes virtually impossible to properly conduct carefully controlled studies with vernacular cannabis preparations given that these products likely contain widely disparate relative quantities of the various constituent cannabinoids.

Finally, among the potential toxicities of the cannabinoids, those of greatest concern to neurologists in particular are neurological and neuropsychiatric toxicities ( Table 3 ). Unfortunately, there is considerable high-quality data coming from various sources that indicates that long-term exposure to THC can have serious deleterious effects on neurological functioning (34-37). In particular, there is strong evidence that progressive memory impairment as well as impaired executive functions results (34,37,38). There is additional concern about a deleterious effect on neural plasticity, particularly with regard to the developing brain (39-42). There is strong evidence that cannabis use results in an increased risk of psychosis in predisposed individuals (43,44). Acute and chronic cannabis use have also been linked to aggravation of anxiety, mania and depression. Consequently any future studies of cannabinoid therapy in epilepsy, and in particular in childhood epilepsy, must very carefully assess acute and long-term neurotoxicity.

Cannabinoids in epilepsy: a rapidly evolving field

Despite the extraordinary current enthusiasm for the use of “medical marijuana” in the treatment of epilepsy and specifically for access to CBD among parents of children with intractable epilepsy (2-4,6), the medical evidence supporting the use of “medical marijuana” for the treatment of neurological disease is weak at best (45). A very thorough review of the use of “medical marijuana” in neurological disease was recently published (45). Authors focused on the following neurological conditions: spasticity in patients with MS, central pain and painful spasms in MS, bladder dysfunction in MS, involuntary movements in MS, movement disorders, epilepsy. The overarching conclusion of this systematic review was that there is minimal high quality data to support the use of “medical marijuana” in any neurological condition other than for the treatment of spasticity in multiple sclerosis (45). It is enlightening to briefly review the data that supports this therapy and compare it with what exists with respect to the treatment of epilepsy with cannabis products.

Controlled trials of various pharmaceutical cannabis products for the treatment of multiple sclerosis (specifically the painful spasms of MS) began in the early 2000s (45). Initial studies did use a variety of preparations ( Table 4 ). This culminated in a series of high quality placebo-controlled trials with Sativex (a 50:50 mixture of THC and CBD) from 2006 to 2011 (46,47). These trials demonstrated a statistically significant benefit of Sativex in the management of painful MS spasms, resulting in the marketing of this agent in England and many other countries (not including the United States). However, quick perusal of Table 4 demonstrates that well over 2000 patients were studied in this fashion. In addition, published post-marketing data regarding well over 10,000 patient-years of experience with Sativex demonstrates relatively low toxicity and statistical absence of serious adverse events (46-49). Since Sativex is a 50-50 mixture of THC and CBD, these observations do provide some reassurance that CBD itself may have minimal toxicity and a low propensity to result in dangerous or serious adverse effects.

Table 4

Study Product Design No. of Pts “Result” Other comment
Killestein, 2002 THC vs. C. sativa extract 20 wks, R, DB, Placebo 16 No change AS Worsening in MSFC
CAMS, Zajicek, 2003 Marinol, Cannador 15 wks, R, Placebo 667 No signif change in AS ↓ 10 m walking time; subj imp. Spasticity, pain
Vaney, 2004 THC 2.5 mg CBD 0.9 mg Pro, R, DB, placebo, X-over 57 No signif diff trend in favor of ↓ spasm freq; imp. sleep, mobility
CAMS-ext Zajicek, 2005 Marinol, Cannador Up to 12 mos 502 (80% of CAMS) Small imp AS, Marinol + Cannador
Wade, 2006 Sativex Open label,

CBD, cannabidiol; AS, Ashworth scale (“objective” spasticity scale); MSFC, MS Functional Composite; VAS, visual analogue scale rating spasticity subjectively; NRS, subjective numerical rating scale for spasticity; wks, weeks; mos, months; pts, patients; signif, significant; diff, difference; freq, frequency; subj, subject; imp, important; neg, negative. Marinol, synthetic THC; Cannador, oral C. Sativa extract.

However contrast this with the published data regarding the use of medical marijuana for the treatment of epilepsy ( Table 5 ) (50-53). Until the last few years, the published data was minimal ( Table 5 ) and included less than 70 subjects. Very few of these were children. Furthermore, none of these studies would meet criteria as Class I-III clinical trials (50-53). However this state of affairs is rapidly changing given the current climate. In 2013, Porter and Jacobson (54) published the self-reported experience of 19 patients whose families had given their children some form of high-CBD medical marijuana product for severe intractable epilepsy. In this group, the reported “doses” of CBD administered ranged from <0.5 mg/kg/day to

29 mg/kg/day. The majority of families reported improvement: Ten reported > 80% improvement, while two patients reported complete cessation of seizures. Others reported an ability to discontinue previously administered medications (54). Though clearly this data would not be considered high quality medical evidence, the information at the time was tantalizing.

Table 5

Study No. of patients Dose/duration Results
Mechoulam, Carlini 1978 9 (R, 4 CBD) epilepsy 200 mg/day 3 months 2/4 CBD: seizure free; 0/5 P
Cunha et al., 1980 Phase 1: 16 healthy; R, 8 CBD 200-300 mg/day Phase 1: 2/8 CBD-somnolence
Phase 2: R, 15; 2e Gen Epi with TL Focus; 8 CBD Phase 2: 4/8 CBD-almost sz free
3/8 partial improvement
1/7 P seizure free
Ames, Cridland, 1986 12; MR; Intr Epi institutionalized 200 mg/day (open label) somnolence
Trembly 1990 (abstract only) 12 pts; placebo phase, then R x-over Placebo 6 mos 300 mg/day “some benefit”

CBD, cannabidiol; pts, patients.

In addition, considerable preclinical evidence regarding the potential efficacy of cannabinoids for the treatment of epilepsy does exist. Some of these studies began as early as the 1970s. Phytocannabinoids (particularly CBD) have been studied in a wide array of animal models of epilepsy (55-59). For the most part, these have demonstrated substantial efficacy. There is some evidence that THC itself can be pro- convulsant in some animal models (57). More recently, efficacy in animal models of temporal lobe epilepsy and partial seizures has been demonstrated (59). In addition there is some evidence that tolerance to the anticonvulsant effects of CBD is not a prominent feature in animal models of epilepsy (55). Thus based on these preclinical studies, one would be excited about the potential therapeutic potential of the cannabinoids. However, it is undeniable that the complex regulation that surrounds these schedule I substances has impeded scientific investigation of their therapeutic potential.

Spurred by the widespread interest in the therapeutic potential of CBD for the treatment of intractable childhood epilepsies, GW pharmaceuticals (the makers of Sativex) developed a pure CBD product known as Epidiolex. Epidiolex has been granted orphan drug status through the FDA for Dravet syndrome and Lennox-Gastaut syndrome. More than 20 expanded access IND’s were granted for the use of Epidiolex in up to 25 or more children per site for the treatment of severe intractable epilepsy. Initial reports of the experience with these children (again uncontrolled) were recently presented (60). 137 patients had received epidiolex for >3 months. Overall seizure frequency was reduced by 54% in all patients and by 63% in Dravet syndrome patients. At 3 months 9% of patients and 16% of Dravet patients were seizure free. Adverse effects were modest (somnolence, diarrhea, fatigue and decreased appetite). Though 22 severe adverse effects were deemed “possibly related” to study medication, only 14 of a total of 213 initially treated patients withdrew due to lack of efficacy or side effects (60). While the results remain promising, outcome data nevertheless is based on self-reported seizure frequencies, is uncontrolled and may suffer from the same methodological problems to some degree as did the Porter and Jacobson report. However, placebo controlled trials of Epidiolex for Dravet and Lennox-Gastaut syndrome are now in process. Consequently higher quality medical evidence surrounding the use of CBD for the treatment of pediatric epilepsy syndromes will be forthcoming. Meanwhile Insys Therapeutics, Inc. has developed a synthetic form of CBD and clinical trials have been initiated (61).

In parallel, families throughout the United States are gaining access to various vernacular “hemp oil” preparations with high CBD, low THC content. Here again the analysis of outcomes is complicated by the extraordinary variability between these products, relative lack of consistency in dosing, variable quality control, and uncertainty with respect to the presence or absence of other potentially bioactive constituents within these products (i.e., could there be an “entourage effect” in some instances?). A recent report from Colorado is intriguing in this respect (6). The authors indicate that while 57% of families reported positive results, there was no evidence of improvement in electroencephalogram patterns in 8 of the responders for whom data was available. In addition, there were significant adverse effects including increased seizures in 13%, status epilepticus and even death. Finally, it is intriguing to consider that the highest reported rate of benefit came from families who had specifically moved to Colorado in order to gain access to “hemp oil” products (47%) versus those who already lived in the state (22%) (6). This could strongly suggest a significant placebo effect in the self-reported seizure outcomes.

Clearly, this conundrum with respect to the outcomes of treatment of epileptic children with “medical marijuana” illustrates how challenging this field is given the “dizzying array” of preparations (6,45). These range from natural products to synthetic agents, substances delivered by inhalation versus those that are swallowed or even delivered by an oral-mucosal spray. Again, this makes comparison of outcomes among various studies nearly impossible. In general, as shown in Table 6 , there are clearly distinct advantages to the use of a highly purified, pharmaceutical product. However, advocates of “medical marijuana” might argue that this approach ignores the potentially beneficial impact of the “entourage effect” (4,13).

Table 6

Opposed to Mj use in children and adolescents
Opposes MM use (outside of FDA etc…)
   Recognizes option in desperate situations
Opposes legalization due to perceived risks
In states with legalized Mj, favors strict regulation of access
Supports R+D of Cannabinoid products
   Recommends change to schedule II
In states with legalized Mj, we should advocate for controls similar to alcohol and tobacco
   Revenue from regulation should go to research
Childproof packaging and related precautions
Supports decriminalization of possession
Strongly opposes smoked Mj
Discourages use of Mj by adults in presence of minors

Mj, Marijuana; MM, medical marijuana; R-D, research and development.

Sociocultural considerations

Clearly, it is very difficult for the pediatric neurologist to navigate the complex legal, medical, psychological and sociocultural complexities surrounding the use of “medical marijuana” in children with epilepsy. Challenges exist in general when it comes to balancing the increasing interest espoused by families in complementary or alternative therapeutic strategies with the trend in allopathic medicine toward increasing reliance on “evidence-based medicine.” This becomes even more complicated when the substance in question is controlled by rigid and intimidating federal regulation, while simultaneously being subject to highly variable, contradictory and rapidly changing state regulations (4,5,8). Specifically according to the Controlled Substance Act of 1970, the federal government has placed marijuana and THC under Schedule I of controlled substances. This designation indicates that marijuana and THC are considered to have: (I) high potential for abuse; (II) no currently accepted medical use in the United States; and (III) lack of accepted safety. In addition, this federal regulation specifies that all species of plants from which controlled substances in Schedule I derive are similarly controlled and subject to the same penalties. By virtue of this extended concept of what constitutes a marijuana product, the DEA and FDA have made it clear that CBD falls under this same rubric (Schedule I). This is despite the fact that most experts and considerable evidence now suggest that this particular phytochemical in fact does not have abuse potential and is clearly of substantial medical interest at this time (13,25,27).

In the meanwhile, state regulation of marijuana ranges from complete proscription to full legalization. Some states have enacted CBD-specific laws (Utah, Georgia, others) which allow families to administer CBD-rich, THC-poor marijuana products to their children with intractable epilepsy under very specific circumstances in part regulated by the respective state health departments (5). However, it is left up to families themselves, with or without the tacit assistance of physicians, to determine how to obtain these substances, judge the quality of the available products, and determine what amount to administer (2,6,54). The role of the pediatric neurologist in this setting is complicated by the fact that it is technically illegal for a physician to “prescribe” these substances under federal law. Furthermore, institutions such as children’s hospitals may decide that administration of these substances within the confines of the hospital may put their staff or even their credentialing at risk. Consequently, there is the uncomfortable potential that a child benefiting from a vernacular “hemp-oil” product would not be permitted to receive it in the hospital, conceivably putting the child at risk in the event the vernacular substance is in fact functioning as an effective anticonvulsant.

There is also considerable concern that the “legalization” of marijuana or the liberalization of medical marijuana laws within states could have a variety of deleterious public health effects (4,8,62-65). Concerns have been raised with respect to increase in violent crime or delinquent behavior (66), potential for increase in driving related accidents or deaths (8), potential for neurotoxicity due to increased use of recreational marijuana products (62), unintended neurotoxicity in children (65), and the theoretical (largely unproven) proposition that increased access to cannabis serves as a gateway to more serious drug abuse behaviors (28,29). With respect to these various public health concerns, actual data is in fact conflicting (28,66) and to some degree data available so far is less alarming than might be imagined by those who strongly oppose the liberalization or legalization of medicinal marijuana products. On the other hand, the American Academy of pediatrics has recently published a position paper/consensus statement with respect to this (28,29) ( Figure 4 ) which clearly states the Academy’s position against the legalization of marijuana but in favor of changing the schedule status of marijuana from Schedule I to Schedule II to facilitate quality scientific research in this area.

Comparison of Pharmaceutical Grade Cannabidiol (CBD) vs. Vernacular Preparations (“hemp-oil”). THC, tetrahydrocannabinol.

Recommendations for the future

It is an exciting time with respect to the study of phytochemicals in their application to the treatment of epilepsy and in particular intractable pediatric epilepsies. It is particularly satisfying to see that the grassroots experiences of families who have children suffering from severe intractable epilepsy have been able to move the field forward so rapidly. The upwelling of interest has already had a remarkable impact and in the space of a few years a large body of medical evidence of increasing quality has been accumulated. Ongoing and anticipated double-blind placebo controlled trials promise the availability of high quality medical evidence in the near future. Hopefully, CBD and possibly other phytocannabinoids or combinations thereof will prove to be beneficial for at least a subset of epileptic children.

In order to foster this progress it is suggested that pediatric neurologists may wish to advocate for the following:

Improved public understanding of the complexity of “medical marijuana” and of the value of high quality medical evidence in guiding therapeutic decisions;

Improved physician understanding of “medical marijuana”, the broad range of preparations subsumed under this misleadingly simple term, and the potential risks and benefits deriving from the disparate chemical substances and products falling under this rubric;

A change in federal regulations that would facilitate carefully conducted, scientifically driven, basic, preclinical and clinical studies of phytocannabinoids in the treatment of various neurological diseases including epilepsy.

Furthermore, pediatric neurologists are encouraged to inform themselves on the specifics of federal and local state regulations so as to be able to best inform and advocate for their patients.


The author thanks Claire Filloux, Ph.D., for preparation of Figures 1,3, and Matthew Sweney, MS, MD for careful review of the manuscript.


Conflicts of Interest: The author has no conflicts of interest to declare.