Cannabinoid therapy shows benefit in pediatric epilepsy

SEATTLE – Cannabinoid therapy holds promise for the treatment of pediatric epilepsy, but more research is needed to rigorously assess its safety and efficacy, according to experts in the field.

Cannabis contains numerous cannabinoids, including the nonpsychoactive cannabidiol (CBD) and the psychoactive tetrahydrocannabinol (THC).

Two CBD products – an oil-based extract (Epidiolex) and an oral solution (Insys Therapeutics) – have been granted orphan drug status by the Food and Drug Administration for treatment of some forms of epilepsy. In addition, families can obtain artisanal preparations, which may contain a variety of cannabinoids as well as other chemical compounds.

However, use and study of cannabinoids in the United States has been complicated by the federal classification of marijuana as a schedule I controlled substance and the patchwork legal status of medical and recreational use across states.

Investigators reported findings of several clinical studies of cannabinoid therapy in pediatric epilepsy at the annual meeting of the American Epilepsy Society and discussed implications for clinical practice and future research, as well as legal and regulatory status of this therapy.

Benefit may be greatest in Dravet syndrome

Dr. Orrin Devinsky, director of the Comprehensive Epilepsy Center at the NYU Langone Medical Center, and his colleagues undertook a prospective, open-label, multicenter, observational study among pediatric patients with treatment-resistant epilepsy.

In the study, supported in part by GW Pharmaceuticals, 151 patients were treated with Epidiolex, added to their baseline regimen of up to three antiepileptic drugs (AEDs).

Main efficacy analyses based on the 58 patients treated for at least 3 months showed that the median reduction in seizure frequency was about 35% overall and 55% for patients with Dravet syndrome.

The 3-month rate of seizure freedom was 10% overall and 22% in the subset with Dravet syndrome. “These were much higher rates than we would expect, for example, from a placebo response alone,” Dr. Devinsky said in a press briefing.

Data from the 40 patients having longer follow-up, at least 16 weeks, showed that benefit was sustained, indicating that patients did not develop tolerance.

Safety results for all 151 patients showed the most common adverse events were somnolence, seen in 19% of patients, and fatigue, seen in 11%. A single serious adverse event, status epilepticus, was considered possibly related to the CBD. Laboratory measures did not show any noteworthy changes.

“Based on the experience at these five sites, in an open-label, nonrandomized controlled trial ... there is good evidence that the Epidiolex 98% product of CBD is an effective drug for many children with treatment-resistant epilepsy, that the patients with Dravet syndrome appear to be the most responsive ones (but again, this is a relatively small population right now), that the drug overall is quite well tolerated, and most of the side effects when they occur are relatively mild and oftentimes limited,” Dr. Devinsky said.

“The critical take-home message is that these are promising early results, but we need controlled trials to know how to move forward, he said, noting that randomized phase III trials are planned in both patients with Dravet syndrome and patients with Lennox-Gastaut syndrome.

Classifying CBD as schedule I – reserved for drugs having a high danger of abuse, lack of acceptable safety data, and no potential therapeutic benefit – presents hurdles for researchers and doesn’t make sense, according to Dr. Devinsky, who disclosed no relevant conflicts of interest.

“Cannabidiol is approved as part of Sativex, for example, in more than 25 European countries and Canada [for treatment of spasticity due to multiple sclerosis]. So we know it’s got benefit in double-blind studies, so that part ... is wrong. And cannabidiol has been tested extensively in humans. We have adult safety data; this drug’s been on the market,” he explained. So we know it’s wrong to schedule it as schedule I. And it’s not at all clear to many of us why the [Drug Enforcement Agency] is persisting in this scheduling classification.”

Impact on levels of AEDs is unpredictable

Investigators led by Dr. Daniel Friedman, director of both the ICU Epilepsy Monitoring Service and the Epilepsy Consult Service at NYU Langone Medical Center, assessed levels of coadministered AEDs in 57 patients from the above study. He noted that such research is important as preclinical data suggest that CBD inhibits enzymes that metabolize commonly used AEDs.

Results showed that over the first 2 months of CBD therapy, 33% of patients on valproate and 24% of patients on clobazam (Onfi) had their doses reduced because they experienced elevated drug levels or sedation attributed to the drugs, Dr. Friedman reported.

But none of the AEDs showed a consistent pattern of change in levels with the addition of CBD. “Many patients experienced decreases; some experienced increases. But especially for the most common drugs, we found that the ranges were pretty wide,” he said.

Data from five patients showed an average threefold increase in the level of the active metabolite of clobazam, N-desmethylclobazam, without any change in levels of the drug itself, suggesting that the CBD was interfering with clearance of the metabolite.

The findings are not definitive, given that the patients were on multiple medications that can influence levels of each other, stressed Dr. Friedman, who disclosed no relevant conflicts of interest.

“Until formal drug-drug interaction studies are done, we refrain from any specific advice, except for, make sure if your patients are on CBD, whether they are getting it through artisanal preparations or as part of a compassionate use study, that background AEDs are monitored, and if they are on clobazam, make sure to check the N-desmethyl metabolite level as well,” he recommended.

Expectations may influence perceived benefit

Dr. Kevin E. Chapman, associate professor at Children’s Hospital Colorado, Aurora, and his colleagues retrospectively reviewed data of 75 patients who reported using oral cannabis extract.

In Colorado, cannabis is legal for both medical and recreational use. However, patients must become residents, obtain a prescription from two specialized physicians, get a medical marijuana ID card (“red card”), and purchase the product from a dispensary, paying out of pocket, as it is not yet covered by insurance.

“We have had a fairly large influx of patients from out of state ever since medical marijuana became legal in our state,” Dr. Chapman noted in a press briefing. However, neurologists are often reluctant to raise the topic with patients “because of the concerns about the fact that it’s still illegal at the federal level. We worry about our own personal DEA licenses if we were to prescribe the medication.”

But when they learn patients are starting CBD therapy, “our practice is to encourage families not to make dramatic changes to seizure medications because for us, then it becomes hard to know how many of the problems that they have are from the CBD or how much of the benefit is from the CBD.”

On average, patients used their extract for 5.6 months. Overall, 33% had a response, defined as a decrease of more than half in seizure frequency as reported by parents (a rate similar to that seen with placebo). The rate was 0% among those with Doose syndrome, 25% among those with Dravet syndrome, and 89% among those with Lennox-Gastaut syndrome.

But analyses identified only a single predictor of response: moving to the state to obtain the extract. “I think it’s understandable that families have high hopes when they move to Colorado that it’s going to be effective … I don’t know if there is a bit of bias by families really wanting for it to work and that’s part of the reason we saw that,” said Dr. Chapman, who disclosed no relevant conflicts of interest.

A total of 10% of patients who had electroencephalograms (EEGs) both before and after starting extract had improvements in their interictal EEGs. However, none of them were among the group who had a response.

Use of extract was associated with some other benefits: Thirty-three percent of patients reported better alertness and/or behavior, 11% reported better language and motor skills, and 7% reported better sleep.

The most common adverse event was a worsening of seizures, seen in 13% of children, followed by somnolence, seen in 12%, and gastrointestinal symptoms, seen in 11%.

“What our study really sort of shows for us was that we really need a better study, something a little bit more controlled, because there are a lot of different strains that are being used out there, and we were not able to tease out what strain of cannabis worked best in our patient population. People are using CBD strains, THC strains, a combination of things, whereas I think we really need more of a uniform strain or preparation, and then more of a placebo-controlled trial,” Dr. Chapman said.

“At this point I think it’s a bit early to be selling the farm, so to speak, and moving to Colorado,” he concluded. “Right now there’s quite a bit of uncertainty about how effective the medication is. Also, we still have a lot of unknowns about drug interactions, adverse effects of the medication. We have no long-term data because ours is just short-term follow-up data.”

Open communication is critical

Dr. Jeffrey Gold, a pediatric neurologist at Rady Children’s Hospital in San Diego, and his colleagues reported the case of a boy with Doose syndrome whose seizures resolved after initiation of cannabinoid therapy.

“We face somewhat different challenges in California, where recreational marijuana is not legal. Medical marijuana has some acceptance, but it exists in a nebulous legal area,” he noted in the press briefing.

The boy’s seizures were refractory to medical therapy and video EEG features suggested a high risk of progression. His parents opted to try CBD instead of the ketogenic diet, the standard of care in such cases.

“They obtained the product from the local family group, which is challenging for us because we don’t really know what’s in it. We don’t know what the composition is. The family doesn’t know what the composition of the product is,” Dr. Gold said. However, the boy’s mother obtained independent testing, which showed that it was sometimes high in CBD, other times high in THC.

After starting therapy, the boy had apparent worsening of seizures. Testing revealed that his valproic acid level had risen from 115 to 178 mcg/mL, suggesting the CBD was interfering with the drug’s clearance.

The valproic acid dose was lowered, and over a 4-month period, the patient became seizure free. “We obtained another EEG which was completely normal, honestly to our surprise,” reported Dr. Gold, who disclosed no relevant conflicts of interest. “And I don’t know why that was, if that was because he got some extra valproic acid and that tripped a mechanism that set off improvement, or if that’s because of the CBD. Resolution of seizures and normalization of the EEG has been reported before in Doose syndrome and can happen spontaneously or as a result of medication.

“We wanted to get this report out to the community so that they would know, if families choose to start CBD therapy, we strongly advocate that they share that information with their physicians and that physicians manage the other antiseizure treatments the child is receiving,” Dr. Gold said. “We have heard from parents and from other providers that once they told their provider that they were doing CBD, they were told to not come back to the office because they were doing a nonapproved therapy, and we don’t support that. We want families to feel like they can share their treatments with their physicians and receive the help they need because we don’t know what these medications do to other treatments that the patient might be receiving, and it’s imperative that families stay in contact with their physicians.”

Families often tinker with artisanal cannabis products themselves, hoping to find just the right combination, Dr. Gold noted. “It emphasizes both the difficulties of doing the artisanal preparations but more specifically, the difficulties of not doing it scientifically. Science is hard, it’s very difficult, it’s expensive, and it’s time consuming. But it’s the right way to do it. And studies ... that are coming up are the right way to do it,” he maintained.

“So we encourage families to pursue conventional treatment, enroll in clinical trials when they are offered, to get involved with the scientific community as you can, and we will get an answer to these questions. But it is going to take time. We need to do it the right way,” he concluded.

SEATTLE – Cannabinoid therapy holds promise for the treatment of pediatric epilepsy, but more research is needed to rigorously assess its safety and efficacy, according to experts in the field.

Cannabis contains numerous cannabinoids, including the nonpsychoactive cannabidiol (CBD) and the psychoactive tetrahydrocannabinol (THC).

Two CBD products – an oil-based extract (Epidiolex) and an oral solution (Insys Therapeutics) – have been granted orphan drug status by the Food and Drug Administration for treatment of some forms of epilepsy. In addition, families can obtain artisanal preparations, which may contain a variety of cannabinoids as well as other chemical compounds.

However, use and study of cannabinoids in the United States has been complicated by the federal classification of marijuana as a schedule I controlled substance and the patchwork legal status of medical and recreational use across states.

Investigators reported findings of several clinical studies of cannabinoid therapy in pediatric epilepsy at the annual meeting of the American Epilepsy Society and discussed implications for clinical practice and future research, as well as legal and regulatory status of this therapy.

Benefit may be greatest in Dravet syndrome

Dr. Orrin Devinsky, director of the Comprehensive Epilepsy Center at the NYU Langone Medical Center, and his colleagues undertook a prospective, open-label, multicenter, observational study among pediatric patients with treatment-resistant epilepsy.

In the study, supported in part by GW Pharmaceuticals, 151 patients were treated with Epidiolex, added to their baseline regimen of up to three antiepileptic drugs (AEDs).

Main efficacy analyses based on the 58 patients treated for at least 3 months showed that the median reduction in seizure frequency was about 35% overall and 55% for patients with Dravet syndrome.

The 3-month rate of seizure freedom was 10% overall and 22% in the subset with Dravet syndrome. “These were much higher rates than we would expect, for example, from a placebo response alone,” Dr. Devinsky said in a press briefing.

Data from the 40 patients having longer follow-up, at least 16 weeks, showed that benefit was sustained, indicating that patients did not develop tolerance.

Safety results for all 151 patients showed the most common adverse events were somnolence, seen in 19% of patients, and fatigue, seen in 11%. A single serious adverse event, status epilepticus, was considered possibly related to the CBD. Laboratory measures did not show any noteworthy changes.

“Based on the experience at these five sites, in an open-label, nonrandomized controlled trial ... there is good evidence that the Epidiolex 98% product of CBD is an effective drug for many children with treatment-resistant epilepsy, that the patients with Dravet syndrome appear to be the most responsive ones (but again, this is a relatively small population right now), that the drug overall is quite well tolerated, and most of the side effects when they occur are relatively mild and oftentimes limited,” Dr. Devinsky said.

“The critical take-home message is that these are promising early results, but we need controlled trials to know how to move forward, he said, noting that randomized phase III trials are planned in both patients with Dravet syndrome and patients with Lennox-Gastaut syndrome.

Classifying CBD as schedule I – reserved for drugs having a high danger of abuse, lack of acceptable safety data, and no potential therapeutic benefit – presents hurdles for researchers and doesn’t make sense, according to Dr. Devinsky, who disclosed no relevant conflicts of interest.

“Cannabidiol is approved as part of Sativex, for example, in more than 25 European countries and Canada [for treatment of spasticity due to multiple sclerosis]. So we know it’s got benefit in double-blind studies, so that part ... is wrong. And cannabidiol has been tested extensively in humans. We have adult safety data; this drug’s been on the market,” he explained. So we know it’s wrong to schedule it as schedule I. And it’s not at all clear to many of us why the [Drug Enforcement Agency] is persisting in this scheduling classification.”

Impact on levels of AEDs is unpredictable

Investigators led by Dr. Daniel Friedman, director of both the ICU Epilepsy Monitoring Service and the Epilepsy Consult Service at NYU Langone Medical Center, assessed levels of coadministered AEDs in 57 patients from the above study. He noted that such research is important as preclinical data suggest that CBD inhibits enzymes that metabolize commonly used AEDs.

Results showed that over the first 2 months of CBD therapy, 33% of patients on valproate and 24% of patients on clobazam (Onfi) had their doses reduced because they experienced elevated drug levels or sedation attributed to the drugs, Dr. Friedman reported.

But none of the AEDs showed a consistent pattern of change in levels with the addition of CBD. “Many patients experienced decreases; some experienced increases. But especially for the most common drugs, we found that the ranges were pretty wide,” he said.

Data from five patients showed an average threefold increase in the level of the active metabolite of clobazam, N-desmethylclobazam, without any change in levels of the drug itself, suggesting that the CBD was interfering with clearance of the metabolite.

The findings are not definitive, given that the patients were on multiple medications that can influence levels of each other, stressed Dr. Friedman, who disclosed no relevant conflicts of interest.

“Until formal drug-drug interaction studies are done, we refrain from any specific advice, except for, make sure if your patients are on CBD, whether they are getting it through artisanal preparations or as part of a compassionate use study, that background AEDs are monitored, and if they are on clobazam, make sure to check the N-desmethyl metabolite level as well,” he recommended.

Expectations may influence perceived benefit

Dr. Kevin E. Chapman, associate professor at Children’s Hospital Colorado, Aurora, and his colleagues retrospectively reviewed data of 75 patients who reported using oral cannabis extract.

In Colorado, cannabis is legal for both medical and recreational use. However, patients must become residents, obtain a prescription from two specialized physicians, get a medical marijuana ID card (“red card”), and purchase the product from a dispensary, paying out of pocket, as it is not yet covered by insurance.

“We have had a fairly large influx of patients from out of state ever since medical marijuana became legal in our state,” Dr. Chapman noted in a press briefing. However, neurologists are often reluctant to raise the topic with patients “because of the concerns about the fact that it’s still illegal at the federal level. We worry about our own personal DEA licenses if we were to prescribe the medication.”

But when they learn patients are starting CBD therapy, “our practice is to encourage families not to make dramatic changes to seizure medications because for us, then it becomes hard to know how many of the problems that they have are from the CBD or how much of the benefit is from the CBD.”

On average, patients used their extract for 5.6 months. Overall, 33% had a response, defined as a decrease of more than half in seizure frequency as reported by parents (a rate similar to that seen with placebo). The rate was 0% among those with Doose syndrome, 25% among those with Dravet syndrome, and 89% among those with Lennox-Gastaut syndrome.

But analyses identified only a single predictor of response: moving to the state to obtain the extract. “I think it’s understandable that families have high hopes when they move to Colorado that it’s going to be effective … I don’t know if there is a bit of bias by families really wanting for it to work and that’s part of the reason we saw that,” said Dr. Chapman, who disclosed no relevant conflicts of interest.

A total of 10% of patients who had electroencephalograms (EEGs) both before and after starting extract had improvements in their interictal EEGs. However, none of them were among the group who had a response.

Use of extract was associated with some other benefits: Thirty-three percent of patients reported better alertness and/or behavior, 11% reported better language and motor skills, and 7% reported better sleep.

The most common adverse event was a worsening of seizures, seen in 13% of children, followed by somnolence, seen in 12%, and gastrointestinal symptoms, seen in 11%.

“What our study really sort of shows for us was that we really need a better study, something a little bit more controlled, because there are a lot of different strains that are being used out there, and we were not able to tease out what strain of cannabis worked best in our patient population. People are using CBD strains, THC strains, a combination of things, whereas I think we really need more of a uniform strain or preparation, and then more of a placebo-controlled trial,” Dr. Chapman said.

“At this point I think it’s a bit early to be selling the farm, so to speak, and moving to Colorado,” he concluded. “Right now there’s quite a bit of uncertainty about how effective the medication is. Also, we still have a lot of unknowns about drug interactions, adverse effects of the medication. We have no long-term data because ours is just short-term follow-up data.”

Open communication is critical

Dr. Jeffrey Gold, a pediatric neurologist at Rady Children’s Hospital in San Diego, and his colleagues reported the case of a boy with Doose syndrome whose seizures resolved after initiation of cannabinoid therapy.

“We face somewhat different challenges in California, where recreational marijuana is not legal. Medical marijuana has some acceptance, but it exists in a nebulous legal area,” he noted in the press briefing.

The boy’s seizures were refractory to medical therapy and video EEG features suggested a high risk of progression. His parents opted to try CBD instead of the ketogenic diet, the standard of care in such cases.

“They obtained the product from the local family group, which is challenging for us because we don’t really know what’s in it. We don’t know what the composition is. The family doesn’t know what the composition of the product is,” Dr. Gold said. However, the boy’s mother obtained independent testing, which showed that it was sometimes high in CBD, other times high in THC.

After starting therapy, the boy had apparent worsening of seizures. Testing revealed that his valproic acid level had risen from 115 to 178 mcg/mL, suggesting the CBD was interfering with the drug’s clearance.

The valproic acid dose was lowered, and over a 4-month period, the patient became seizure free. “We obtained another EEG which was completely normal, honestly to our surprise,” reported Dr. Gold, who disclosed no relevant conflicts of interest. “And I don’t know why that was, if that was because he got some extra valproic acid and that tripped a mechanism that set off improvement, or if that’s because of the CBD. Resolution of seizures and normalization of the EEG has been reported before in Doose syndrome and can happen spontaneously or as a result of medication.

“We wanted to get this report out to the community so that they would know, if families choose to start CBD therapy, we strongly advocate that they share that information with their physicians and that physicians manage the other antiseizure treatments the child is receiving,” Dr. Gold said. “We have heard from parents and from other providers that once they told their provider that they were doing CBD, they were told to not come back to the office because they were doing a nonapproved therapy, and we don’t support that. We want families to feel like they can share their treatments with their physicians and receive the help they need because we don’t know what these medications do to other treatments that the patient might be receiving, and it’s imperative that families stay in contact with their physicians.”

Families often tinker with artisanal cannabis products themselves, hoping to find just the right combination, Dr. Gold noted. “It emphasizes both the difficulties of doing the artisanal preparations but more specifically, the difficulties of not doing it scientifically. Science is hard, it’s very difficult, it’s expensive, and it’s time consuming. But it’s the right way to do it. And studies ... that are coming up are the right way to do it,” he maintained.

“So we encourage families to pursue conventional treatment, enroll in clinical trials when they are offered, to get involved with the scientific community as you can, and we will get an answer to these questions. But it is going to take time. We need to do it the right way,” he concluded.

SEATTLE – Cannabinoid therapy holds promise for the treatment of pediatric epilepsy, but more research is needed to rigorously assess its safety and efficacy, according to experts in the field.

Cannabis contains numerous cannabinoids, including the nonpsychoactive cannabidiol (CBD) and the psychoactive tetrahydrocannabinol (THC).

Two CBD products – an oil-based extract (Epidiolex) and an oral solution (Insys Therapeutics) – have been granted orphan drug status by the Food and Drug Administration for treatment of some forms of epilepsy. In addition, families can obtain artisanal preparations, which may contain a variety of cannabinoids as well as other chemical compounds.

However, use and study of cannabinoids in the United States has been complicated by the federal classification of marijuana as a schedule I controlled substance and the patchwork legal status of medical and recreational use across states.

Investigators reported findings of several clinical studies of cannabinoid therapy in pediatric epilepsy at the annual meeting of the American Epilepsy Society and discussed implications for clinical practice and future research, as well as legal and regulatory status of this therapy.

Benefit may be greatest in Dravet syndrome

Dr. Orrin Devinsky, director of the Comprehensive Epilepsy Center at the NYU Langone Medical Center, and his colleagues undertook a prospective, open-label, multicenter, observational study among pediatric patients with treatment-resistant epilepsy.

In the study, supported in part by GW Pharmaceuticals, 151 patients were treated with Epidiolex, added to their baseline regimen of up to three antiepileptic drugs (AEDs).

Main efficacy analyses based on the 58 patients treated for at least 3 months showed that the median reduction in seizure frequency was about 35% overall and 55% for patients with Dravet syndrome.

The 3-month rate of seizure freedom was 10% overall and 22% in the subset with Dravet syndrome. “These were much higher rates than we would expect, for example, from a placebo response alone,” Dr. Devinsky said in a press briefing.

Data from the 40 patients having longer follow-up, at least 16 weeks, showed that benefit was sustained, indicating that patients did not develop tolerance.

Safety results for all 151 patients showed the most common adverse events were somnolence, seen in 19% of patients, and fatigue, seen in 11%. A single serious adverse event, status epilepticus, was considered possibly related to the CBD. Laboratory measures did not show any noteworthy changes.

“Based on the experience at these five sites, in an open-label, nonrandomized controlled trial ... there is good evidence that the Epidiolex 98% product of CBD is an effective drug for many children with treatment-resistant epilepsy, that the patients with Dravet syndrome appear to be the most responsive ones (but again, this is a relatively small population right now), that the drug overall is quite well tolerated, and most of the side effects when they occur are relatively mild and oftentimes limited,” Dr. Devinsky said.

“The critical take-home message is that these are promising early results, but we need controlled trials to know how to move forward, he said, noting that randomized phase III trials are planned in both patients with Dravet syndrome and patients with Lennox-Gastaut syndrome.

Classifying CBD as schedule I – reserved for drugs having a high danger of abuse, lack of acceptable safety data, and no potential therapeutic benefit – presents hurdles for researchers and doesn’t make sense, according to Dr. Devinsky, who disclosed no relevant conflicts of interest.

“Cannabidiol is approved as part of Sativex, for example, in more than 25 European countries and Canada [for treatment of spasticity due to multiple sclerosis]. So we know it’s got benefit in double-blind studies, so that part ... is wrong. And cannabidiol has been tested extensively in humans. We have adult safety data; this drug’s been on the market,” he explained. So we know it’s wrong to schedule it as schedule I. And it’s not at all clear to many of us why the [Drug Enforcement Agency] is persisting in this scheduling classification.”

Impact on levels of AEDs is unpredictable

Investigators led by Dr. Daniel Friedman, director of both the ICU Epilepsy Monitoring Service and the Epilepsy Consult Service at NYU Langone Medical Center, assessed levels of coadministered AEDs in 57 patients from the above study. He noted that such research is important as preclinical data suggest that CBD inhibits enzymes that metabolize commonly used AEDs.

Results showed that over the first 2 months of CBD therapy, 33% of patients on valproate and 24% of patients on clobazam (Onfi) had their doses reduced because they experienced elevated drug levels or sedation attributed to the drugs, Dr. Friedman reported.

But none of the AEDs showed a consistent pattern of change in levels with the addition of CBD. “Many patients experienced decreases; some experienced increases. But especially for the most common drugs, we found that the ranges were pretty wide,” he said.

Data from five patients showed an average threefold increase in the level of the active metabolite of clobazam, N-desmethylclobazam, without any change in levels of the drug itself, suggesting that the CBD was interfering with clearance of the metabolite.

The findings are not definitive, given that the patients were on multiple medications that can influence levels of each other, stressed Dr. Friedman, who disclosed no relevant conflicts of interest.

“Until formal drug-drug interaction studies are done, we refrain from any specific advice, except for, make sure if your patients are on CBD, whether they are getting it through artisanal preparations or as part of a compassionate use study, that background AEDs are monitored, and if they are on clobazam, make sure to check the N-desmethyl metabolite level as well,” he recommended.

Expectations may influence perceived benefit

Dr. Kevin E. Chapman, associate professor at Children’s Hospital Colorado, Aurora, and his colleagues retrospectively reviewed data of 75 patients who reported using oral cannabis extract.

In Colorado, cannabis is legal for both medical and recreational use. However, patients must become residents, obtain a prescription from two specialized physicians, get a medical marijuana ID card (“red card”), and purchase the product from a dispensary, paying out of pocket, as it is not yet covered by insurance.

“We have had a fairly large influx of patients from out of state ever since medical marijuana became legal in our state,” Dr. Chapman noted in a press briefing. However, neurologists are often reluctant to raise the topic with patients “because of the concerns about the fact that it’s still illegal at the federal level. We worry about our own personal DEA licenses if we were to prescribe the medication.”

But when they learn patients are starting CBD therapy, “our practice is to encourage families not to make dramatic changes to seizure medications because for us, then it becomes hard to know how many of the problems that they have are from the CBD or how much of the benefit is from the CBD.”

On average, patients used their extract for 5.6 months. Overall, 33% had a response, defined as a decrease of more than half in seizure frequency as reported by parents (a rate similar to that seen with placebo). The rate was 0% among those with Doose syndrome, 25% among those with Dravet syndrome, and 89% among those with Lennox-Gastaut syndrome.

But analyses identified only a single predictor of response: moving to the state to obtain the extract. “I think it’s understandable that families have high hopes when they move to Colorado that it’s going to be effective … I don’t know if there is a bit of bias by families really wanting for it to work and that’s part of the reason we saw that,” said Dr. Chapman, who disclosed no relevant conflicts of interest.

A total of 10% of patients who had electroencephalograms (EEGs) both before and after starting extract had improvements in their interictal EEGs. However, none of them were among the group who had a response.

Use of extract was associated with some other benefits: Thirty-three percent of patients reported better alertness and/or behavior, 11% reported better language and motor skills, and 7% reported better sleep.

The most common adverse event was a worsening of seizures, seen in 13% of children, followed by somnolence, seen in 12%, and gastrointestinal symptoms, seen in 11%.

“What our study really sort of shows for us was that we really need a better study, something a little bit more controlled, because there are a lot of different strains that are being used out there, and we were not able to tease out what strain of cannabis worked best in our patient population. People are using CBD strains, THC strains, a combination of things, whereas I think we really need more of a uniform strain or preparation, and then more of a placebo-controlled trial,” Dr. Chapman said.

“At this point I think it’s a bit early to be selling the farm, so to speak, and moving to Colorado,” he concluded. “Right now there’s quite a bit of uncertainty about how effective the medication is. Also, we still have a lot of unknowns about drug interactions, adverse effects of the medication. We have no long-term data because ours is just short-term follow-up data.”

Open communication is critical

Dr. Jeffrey Gold, a pediatric neurologist at Rady Children’s Hospital in San Diego, and his colleagues reported the case of a boy with Doose syndrome whose seizures resolved after initiation of cannabinoid therapy.

“We face somewhat different challenges in California, where recreational marijuana is not legal. Medical marijuana has some acceptance, but it exists in a nebulous legal area,” he noted in the press briefing.

The boy’s seizures were refractory to medical therapy and video EEG features suggested a high risk of progression. His parents opted to try CBD instead of the ketogenic diet, the standard of care in such cases.

“They obtained the product from the local family group, which is challenging for us because we don’t really know what’s in it. We don’t know what the composition is. The family doesn’t know what the composition of the product is,” Dr. Gold said. However, the boy’s mother obtained independent testing, which showed that it was sometimes high in CBD, other times high in THC.

After starting therapy, the boy had apparent worsening of seizures. Testing revealed that his valproic acid level had risen from 115 to 178 mcg/mL, suggesting the CBD was interfering with the drug’s clearance.

The valproic acid dose was lowered, and over a 4-month period, the patient became seizure free. “We obtained another EEG which was completely normal, honestly to our surprise,” reported Dr. Gold, who disclosed no relevant conflicts of interest. “And I don’t know why that was, if that was because he got some extra valproic acid and that tripped a mechanism that set off improvement, or if that’s because of the CBD. Resolution of seizures and normalization of the EEG has been reported before in Doose syndrome and can happen spontaneously or as a result of medication.

“We wanted to get this report out to the community so that they would know, if families choose to start CBD therapy, we strongly advocate that they share that information with their physicians and that physicians manage the other antiseizure treatments the child is receiving,” Dr. Gold said. “We have heard from parents and from other providers that once they told their provider that they were doing CBD, they were told to not come back to the office because they were doing a nonapproved therapy, and we don’t support that. We want families to feel like they can share their treatments with their physicians and receive the help they need because we don’t know what these medications do to other treatments that the patient might be receiving, and it’s imperative that families stay in contact with their physicians.”

Families often tinker with artisanal cannabis products themselves, hoping to find just the right combination, Dr. Gold noted. “It emphasizes both the difficulties of doing the artisanal preparations but more specifically, the difficulties of not doing it scientifically. Science is hard, it’s very difficult, it’s expensive, and it’s time consuming. But it’s the right way to do it. And studies ... that are coming up are the right way to do it,” he maintained.

“So we encourage families to pursue conventional treatment, enroll in clinical trials when they are offered, to get involved with the scientific community as you can, and we will get an answer to these questions. But it is going to take time. We need to do it the right way,” he concluded.