Cannabidiol for Absence Epilepsy: More to Learn

Andrew N. Wilner, MD; Selim Benbadis, MD


February 28, 2023

This transcript has been edited for clarity.

Andrew N. Wilner, MD: Welcome to Medscape. I'm your host, Dr Andrew Wilner, reporting on the annual American Epilepsy Society meeting. With me today is my old friend and colleague, Dr Selim Benbadis, professor of neurology and director of the Comprehensive Epilepsy Center at the University of South Florida in Tampa, Florida.

Welcome, Selim.

Selim Benbadis, MD: Thank you, Andrew. Thanks for having me.

Wilner: Selim, your group had a very interesting poster on cannabidiol (CBD) for absence epilepsy, isn't that right?

Benbadis: Yes.

Wilner: Before you delve into the specifics, tell me the background. Why did you do it?

Benbadis: We did it because, as you know, for typical absence seizures, there are very few options to treat. The only two medications over decades that we know work for typical absence seizures are valproate and ethosuximide. If those do not work, the options are completely off-label.

We try all kinds of other things, including safe medications that we have, but none of them have evidence to show that they work in typical absence seizures. That's one part. As you also know, CBD is approved as a pharmaceutical to treat certain types of seizures and epilepsy but not typical absence seizures.

Our question was whether it's a newer drug, it's approved, and it works for some types of seizures; might it work for typical absence seizures?

Wilner: When you say typical absence seizures, are you talking about children, adults, or both?

Benbadis: Either/or, but that means typical absence seizures in the context of an idiopathic generalized epilepsy, so the typical clear-cut, 3-Hz spike-wave complexes on EEG type of absence seizures. As you know, they usually begin in children. It's true. We did enroll adults as long as they continued to have those type of seizures documented by EEG. This isn't a subjective diagnosis.

Wilner: When I think of absence seizures, I think of these brief staring spells. These patients can also have convulsions. Was that part of the study?

Benbadis: You are correct. Patients with idiopathic generalized epilepsy can also have myoclonic seizures and generalized tonoclonic seizures. The purpose of this study was to only determine if they work in typical absence seizures.

Wilner: Tell me what you did.

Benbadis: Typical absence seizures are very brief, can be subtle, and are easily missed, so we cannot rely on mom saying, "Yes, John had three absence seizures yesterday." It doesn't work, which, by the way, is partially why we don't have a lot of studies on typical absence seizures.

We did it in what I think is an elegant and objective way, which is by EEG. We performed a 24-hour home EEG. Patients had to have a certain number of spike-wave complexes because we needed something to measure. We measured objectively the spike-wave burden. How many seconds in 24 hours does this patient spend in spike-wave complexes?

As you know, that's not a subtle EEG finding. Anybody can see it. We measured it and quantified it. How many seconds in 24 hours? Then we started the patient on CBD (Epidiolex, the pharmaceutical CBD) at a dose similar to what is approved on the market for other types of epilepsies.

Then, after 90 days, we conducted a second home EEG. Again, this is an objective measure, not what the parents think. We measured how many seconds in those 24 hours the patient was in spike-wave complexes, and we compared it with the baseline EEG.

Wilner: Before you tell me the results, as an EEG-er myself, I'm interested in the protocol. Looking at 24 hours of EEG can be a little mind numbing — counting all these little bursts. Did you use any software to help you?

Benbadis: No. We used experienced EEG technologists, who are better than software. Again, those are not subtle findings. In fact, on the poster, we had a sample so people can understand that anybody can see this.

It's just a question of counting them, as you said, but 24 hours is not that long, so three bursts of 2 seconds, that's 6 seconds. Five bursts of 1 second, that's 5 seconds. Then we calculated the total at the end. It's a little time consuming, but easy.

Wilner: Every page, they would make a comment every 10 seconds that there was something here, nothing there, and then just go through it. Human beings who knew what they were doing reviewed the EEGs.

Benbadis: Yes. We had a number, such as EEG 1, and noted that the patient spent 422 seconds in spike-wave complexes.

Wilner: How long were participants on the drug?

Benbadis: I believe it was 90 days. We titrated slowly. We used it in the same way that it's used on-label, which, as you know, is mainly for Lennox-Gastaut type of epilepsy.

Wilner: You just added it to whatever drug they were already taking?

Benbadis: Yes. That's a good question. Some individuals were not taking medications. We had a couple of children who were not treated because the family didn't want to treat them. We had a few who were on one or two seizure medications. We added CBD to what they were currently taking because we're going to compare pre-CBD and post-CBD, as long as nothing else was changed.

Wilner: This was 90 days?

Benbadis: Yes.

Wilner: You redid the study 90 days later. You sent them home with the electrodes in a little box, they came back, and the tech looked at it. What did you find?

Benbadis: At the time of the presentation, we had nine patients. We have a few more now, and we're aiming for 15. Only two improved. In other words, the spike-wave burden decreased. In seven of them, not only did it not improve but it increased.

This isn't what we were hoping to find, but it's very important. This is preliminary and it's a pilot study. There are flaws to this and we could do it better.

It would seem to indicate that CBD falls under the category of many other seizure medications, which is that they work for many types of motor seizures, but for typical absence seizures — which is a different bird, really — it may not be effective and possibly could even exacerbate it, which, as you know, other old drugs do.

Wilner: I think that's very important. When the FDA approved CBD for Dravet syndrome, Lennox-Gastaut, and seizures associated with tuberous sclerosis, plus the large amount of anecdotal evidence that somehow it helps for seizures, there's frequent off-label use. Here, you have an example where off-label use in a controlled setting actually was not helpful at all.

Benbadis: Right. Again, it's a small number. You could argue that the 24-hour EEG might be too short, that EEGs fluctuate, maybe it was bad luck, or the child was stressed or didn't sleep well so there was an increase. All of that is possible.

It's motivating us to do another study, maybe with a longer EEG, maybe a 3-day EEG instead of 1, maybe two follow-up EEGs, not just one. Things like that. Certainly, it gives a signal that it's not the panacea. That we can say.

To be fair, like I said, other than ethosuximide and valproate, when people use levetiracetam or lamotrigine for typical absence seizures, there's also zero evidence that it works.

Wilner: One question that comes to mind — and I don't know if you looked at this or not — can CBD affect the metabolism of ethosuximide or valproate? It might have changed the baseline level of the drugs they were taking.

Benbadis: That's a good question. We know of at least one pharmacokinetic interaction that is important for CBD, and that is with clobazam, where they potentiate each other.

We are not aware that it either inhibits or induces the metabolism of other medications. That's, again, a good reason to do another cleaner, more controlled study. This was really a pilot.

We did not expect to find that, I must admit. I thought we were going to find some efficacy.

Wilner: In the next study, you'll do baseline drug levels and you'll do them at the end of the study also. That'll be easy.

Benbadis: Yes.

Wilner: What about clinical seizures? After all, the patients are pretty interested in their clinical seizures.

Benbadis: Of course.

Wilner: Do you think it's worthwhile trying to count them?

Benbadis: We did, as a secondary measure as well as quality of life. We asked the parents how they felt the child was doing, but that's so subjective, especially for absence seizures. We did make a note of it. Sometimes it agreed with the objective measure and sometimes it didn't.

The spike-wave burden might be reduced but they said the child was worse and vice versa. I really feel that the strength of this was that it's really an objective and quantitative measure of the spike-wave absence burden.

Wilner: I think this is very important and very exciting. It's going to lead to another presentation next year, right?

Benbadis: I hope so.

Wilner: You'll have it done by next year. Are you recruiting patients now?

Benbadis: We still are. The goal was 15, so we have a few more to go, and then, hopefully, we'll publish this. Nobody loves negative results, but it's important, as you said.

Wilner: Before we wrap up, Selim, is there anything you'd like to add?

Benbadis: No. It's good to have new drugs. As you said, there is a large amount of off-label use. I think this is an important piece of information for clinicians.

Wilner: Dr Selim Benbadis, thank you very much for joining us on Medscape.

Benbadis: Thank you for having me.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


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