Episode 31: Psychedelics for the Treatment of Cluster and Migraine





Welcome to Spotlight on Migraine, a podcast series hosted by the Association of Migraine Disorders. Through personal stories and interviews with experts, we expose the true scope of migraine by exploring symptoms, treatments, research topics, and more. This episode is brought to you in part by our generous sponsors, Amgen, Novartis, and Alder BioPharmaceuticals.

In this episode, Dr. Christopher Gottschalk gives a presentation explaining research currently being done at Yale University on the use of psychedelic mushrooms in the treatment of migraine and cluster headache. Additionally, he explains the derivatives and characteristics of drugs currently being used for migraine and cluster treatment and how they compare to psychedelics.

Since 2015, Amgen and Novartis have been working together to develop pioneering therapies in Alzheimer’s disease and migraine. Together, Amgen and Novartis share in a mission to fight migraine and the stereotypes and misconceptions surrounding this debilitating disease.

Dr. Christopher Gottschalk: It’s a great pleasure to be here, and I found those talks that you’ve already heard very, very exciting. The best thing about being a headache physician right now and in the last 10 or 20 years is that this is an extraordinary period of revolutionary ideas and applications, devices and medications for this group of people that have been standing in the dark for way too long. And speaking of that, I’d like to speak a little bit about some of the research that we’re currently doing at Yale and the use of psychedelics in the treatment of headache disorders.

So we’re here talking about a slightly different but related phenomenon. You’ve been hearing about mostly migraine so far and, in some cases, things that could be applied to both, but just a quick review. Cluster headache is a distinct form of what we call a primary headache disorder, meaning it is attacks of pain with other symptoms, like migraine. But as opposed to migraine, it is part of what we call the trigeminal autonomic cephalalgias, which simply means it is diseases that have unilateral pain attacks associated with weird changes in your eye or your nose or your ear on the same side, indicating to us that there is something that is affecting the what we call autonomic nerves on the same side of the head as your trigeminal nerve where you’re getting the pain sensation.

So there’s a list here of diseases that fall under that category, and by far, the first member of that group, cluster headache, is the most common type. The rest — paroxysmal hemicranias, hemicrania continua, SUNCT, and SUNA — are very rare disorders, although very interesting, and actually, as a group, the latter ones tend to respond to one specific medication called indomethacin, which I think we don’t understand well enough. But one thing you can say about that chemical is that it was derived from serotonin, and as far as anybody can tell, that’s as good a reason as any why it is effective in those diseases.

Cluster headache, the one that we’ll talk about right now, has a prevalence that is far lower than migraine, somewhere between 1 in 1,000 and 1 in 5,000, and as opposed to what I was taught when I trained in neurology — that the prevalence was 8 times more common in men — it’s probably a lot closer to 3 or 4 times more common in men, which is another example of the kind of biased thinking that all of us have been guilty of for a long time, basically things like, “Well, if you’re a guy, you must have cluster attacks, and if you’re a woman, you couldn’t possibly. You must have migraine.” So that thinking, fortunately, is evolving.

One of the characteristics of cluster is that it is described as intensely painful attacks. Well, up until recently, no one has taken a new approach to try to understand exactly what that means by “intensely painful.” Pain is one of those things that we measure based on subjective report, so some very clever researchers — Drs. Schor, Burish, and Pearson — have created an online cluster headache questionnaire in which part of what they ask is, “Well, how does that pain compare to other pain that you have had?”
So at the time that these slides were prepared, they’d already had over 2,000 people respond, about two-thirds of them being men, which fits with our perception of what the prevalence rates are. The average age of onset for people in this group was about 27, whereas it took at least 6 years for the average person to get diagnosed — another feature of cluster headache, which is that it tends to be hidden in some kind of misdiagnosis of a sinus disorder or a variation of migraine, etc.

In that survey, one of the things they do is present a pain scale, or a set of sliding pain scales, asking people to rate cluster and other pain that they have experienced. When you look at the results, one of the things they show is that the most common location of pain is right around the eye, which is not surprising because that’s the most common location, although it’s not limited to that. But here are some of these initial results: on a scale of 1 to 10, your typical subjective pain scale, people who’d had shingles rate that at about just under 50%. People who’ve had migraine rate that at just over 50%. People who’ve had kidney stones, often described as one of the most painful things on the planet, rate that at about two-thirds. Childbirth, above that. Pancreatitis, also described as intensely disabling pain which almost always leads to hospitalization. And no surprise to people who have cluster headache or know people who do, that is about as high as it gets.

So this is a disorder that is intensely painful, and that probably has something to do with the fact that the suicide rate in this cohort of people is at least 10 times the normal population rate. This is a horrendously disabling disorder, and given the fact that attacks typically occur multiple times a day and often in the middle of the night, I can’t really imagine what it’s like to go on living, thinking, “This is probably going to happen once I fall asleep and over and over again, and why can’t anybody do anything about it?” And that seems to be part of the problem.

So the current definitions of cluster attacks are severe unilateral, usually around the eye or what we call orbital pain, stabbing in nature, associated with those autonomic features, meaning changes in the tone of autonomic nerves around the eye and nose and ear and in the sinuses, and which is typically associated with restlessness. People, during acute attacks, are often moving around trying to distract themselves, as opposed to migraine, which tends to be worse with activity and so has people lying quietly. Unlike migraine, these attacks characteristically are very short — thank goodness, given what we’ve just seen — but could last as long as 3 hours, but migraine is typically defined as at least 4 hours — I cannot literally imagine what it would be like to have something like that for 3 hours, but so it goes — and can occur, characteristically, up to 8 times a day, sometimes more. There are people out there who, unfortunately, are afflicted with chronic cluster headache, meaning never get a break of more than a month, and the majority of these patients have periods of remission lasting weeks to months, sometimes years. So there is a diurnal variation and a circannual variation, changes in frequency over the course of a year.

Here’s the current state of cluster headache treatments. We have, in the upper left, agents that are used to stop individual attacks. Inhaled pure oxygen, amazingly enough, is a phenomenally effective treatment. Ashley Hattle, sitting in the front row here, I’m assuming, is going to tell us a little bit about that. But the bottom line there is if it is prescribed properly and used properly, can be an extraordinarily effective intervention. Go figure why it is that the Centers for Medicaid Services in this country refuses to acknowledge oxygen as a valid treatment for cluster headache despite our years of effort, and that, hopefully, will change in the next year or two. The other primary option for acute attacks is taking an injection of sumatriptan, which fortunately works quite well, and we’ll talk more about that in a minute.

For people who are having lots of attacks or for where their cluster periods are going on for a while, there’s a range of medications that have been identified: high doses of verapamil; lithium, oddly enough, not because it has anything to do with being bipolar, but simply because it is an effective ion in this condition; seizure drugs sometimes of the same type that we use in migraine, Topamax and Depakote; steroids; and another that could be on that list is methylergonovine, the currently existing oral ergot derivative that can be very effective.

Some other options include, probably, these new class of drugs — the CGRP antagonist that we’ll hear more about later — and there is now in the upper right a picture of a version of a vagus nerve stimulator, which has been approved for both stopping acute attacks by tickling the nerve in your neck and even preventing attacks by using it on a daily basis. Sometimes, people will have a nerve block. There is an implantable device, which has been approved in Europe and will probably be approved here this year, which stimulates the SPG ganglion, that we’ve just heard about, on an as-needed basis using an electrical external device to tickle that thing that’s been implanted under your gum — fairly dramatic, invasive procedure, but fortunately effective for some. And there have even been trials of sticking a battery wire all the way deep into your brain to stimulate the brain — the hypothalamus of the brain and stop chronic cluster, and it is effective, although that’s about as invasive and dramatic as it can get.

So here is a map — a diary of a typical cluster headache patient from 15 years ago or more showing, for each day, the time and the severity of a group of attacks. So the closer to red or black those little bars are, the more intense the pain, and on a given day, the number of attacks you can see stacking up. This is somebody who, a few years later, based on some Internet chatter that was going on at the time, discovered that there is the possibility that taking an extract of mushroom, psilocybin, could reduce attacks. And so there you see a first pass, a small dose of half a gram, which dramatically reduced the number of attacks, which then built up again over a few days, and then a larger dose was taken, and the attacks virtually disappeared. A couple of years later, the same patient produced a map showing that if they took a dose of this every few months, they could virtually eliminate the possibility of cluster attacks. Remarkable.

What is that about? Well, there have been a series of papers in various ways that have looked at patient self-reports about the efficacy of these drugs, the first of which was published by Andrew Sewell, who was a Yale researcher and, unfortunately, died suddenly and unexpectedly a few years ago. But in that first paper, they noted that a high percentage of people who said that they had used either psilocybin or a related hallucinogen — LSD — found success with both treating acute attacks. So you see in that first arrow, psilocybin 85% effective, LSD 50% effective. Given the fact that the effects of LSD last so long, that’s kind of a tough bargain to strike. But as a preventive agent, single doses produced dramatic benefits, even more so for LSD. And for extending the period of remission from attacks, similarly both very, very effective.

We published a paper just a few years ago looking in more detail at an online cluster headache survey asking people to report the benefits of various interventions — on the left, for abortive treatments; on the right, for preventive — and the arrows are pointing to the columns that refer to psilocybin or LSD or related agents, showing that the larger the column is black, the greater the percentage of people who say these are effective interventions.

So we have a large amount of citizen science telling us that there is something important here. Here is a controlled trial which used a related non-hallucinogenic agent, something called bromo-lysergic acid diethylamide, showing that those lines — or simply showing that the number of attacks dramatically decreased after being exposed to one or more doses of that drug. Part of my point today is — and part of the point of this program is to make the point, the distinction, that although these drugs — psilocybin and LSD — are referred to as psychedelics because they are capable of producing hallucinations, that is not the reason for studying them and that is not the goal of treatment. So although these terms exist, the point here is we are not exploring empathogens or dissociatives or related compounds like cannabinoids. We are saying that pharmacologically, these are drugs that should be studied because of their potential benefits.

What are psychedelics? Well, they encompass many classes. In terms of organic chemistry, the indoleamine group is broken down into tryptamines and ergolines, ergolines being more familiar to some people as ergots. Tryptamines are by and large synthetic agents and also psilocybin. On the right, there is a picture of Albert Hofmann, the man who, in the early 20th century, did a lot of chemistry to form derivatives of ergots, which are extracts from plants, to find agents that could be useful, since those plant extracts had already been found to be very effective in treating headaches, and produced, among other things, LSD, which is one of those derivatives, and then some other compounds.

Then the other main class is the phenethylamines, of which mescaline may be familiar to some, as well as some related synthetic agents. So here’s a picture of wheat with mold growing on it that is typically black, and it is the extract of that mold which produces extracts called ergot, including the one on the right, which is dihydroergotamine, a semi-synthetic derivative. Outbreaks of craziness and gangrene in the 16th, 17th, 18th, and other centuries were referred to as St. Anthony’s fire, and those outbreaks turned out to be due to grain stores that were infected with this mold. And enterprising, I guess, alchemists in those days took some of the extracts of those plants and found that they had really phenomenally potent effects, like getting rid of headache. But it was Dr. Hofmann who produced derivatives like DHE and then LSD and, because LSD was so potently hallucinogenic, tried to make less hallucinogenic or non-hallucinogenic derivatives, like an agent called methysergide. Fifty-six years ago, methysergide was approved as the first agent for preventing migraine on the market ever, and it was around for about 40 years until it was pulled because of some concerns about causing fibrosis, which may or may not be true.

On the left-hand side, on the other hand, is a drug familiar to most headache patients, sumatriptan, which was derived based on the structure and chemistry of a drug like DHE and which is very closely related, if you look at it, to melatonin, something we’re all familiar with. And there’s psilocybin, almost indistinguishable in its structure from sumatriptan. So my point here is that hallucination and euphoria are intrinsic properties of the brain, and just because a drug can cause those does not mean that it should be forbidden or avoided. If you take enough of almost any drug, alcohol, anticholinergic agents like belladonna, which were identified hundreds of years ago as being capable of causing hallucinations — which, by the way, is where we get the image of witches riding broomsticks, an image I’ll never get out of my mind, having read that book. But the point is, any drug, Parkinson’s drug, can cause this at a given dose. That does not mean that they are terrible and should be forbidden. It means they should be studied and understood, and a drug like psilocybin clearly, in my opinion, fits that criterion.

Here is just a quick map of a neuron synapse, so two neurons communicating with each other with all of — or some number of the transmitters that we understand something about and the receptors that take them up or respond to them, and that the site of action of some of these, like the so-called psychedelics, is specifically at the 5-HT_2A receptor. We understand something about how they work and where.

Here’s a brief point related to the last one, that if you think about drugs and their toxicity based on how much it takes to kill you as opposed to how much it takes to have an effect, then LSD and psilocybin are right at the top of the list of “Boy, do you have to really, really work hard to get into trouble,” as opposed to something like nutmeg, which can kill you if you overdo it, or a drug that is used every single day by thousands of — millions of people and is available over the counter, dextromethorphan or Robitussin. It’s only in the last decade, of course, that those have been placed behind the counter at CVS because of the number of college and high school students who figured out that if you drank enough Robitussin and alcohol at the same time, you can get wildly high. So again, a point about dose. So based on that chemistry and the knowledge that we have about what cluster headaches report in terms of the benefits, a group of investigators at Yale, starting with Dr. Sewell, wanted to do controlled trials looking at that. So along with, now, Dr. D’Souza, who’s been doing studies of psychotropic drugs at Yale for years; Emmanuelle Schindler, who is spearheading this study and who came to Yale to work with Dr. Sewell; and some funding, thank goodness, from the Heffter Research Institute, we are now looking at studies of both cluster headache and migraine headache or post-traumatic headache.

So we are enrolling people who are willing to be in a study where they might get an extract of psilocybin or placebo. We are looking at people who, if they have cluster headache, have at least one attack a day, who are episodic at the beginning of a cycle — if you have migraine, having approximately two attacks a week. There are certain exclusion criteria, like psychotic or manic disorders, and you can’t be taking an SSRI — that’s probably the most common exclusion reason — and you have to limit the amount of triptans.

What I can say about these studies right now is that we’ve had about 20 patients enrolled and 14 completed and that the side effect profile, which you see here, is very, very favorable. So out of all those patients, we have a few with nausea, a few with transient anxiety, light-headedness, agitation, etc.; in the migraine studies, a few who reported headache, which is always seen in migraine studies; and no serious adverse events. All of the subjects and all of the investigators in these studies are still blinded, so we don’t know who got what, and that’s why we can’t report anything about the results. And based on these, we are continuing to recruit subjects, and we will look at optimizing the dose, looking at which groups are the most responsive, and possibly some other non-hallucinogenic congeners.

Thank you for your time.

Thank you for tuning in to Spotlight on Migraine. For more information on migraine disease, please visit

Related Portfolio Items