This is not based on an anecdote. I've been researching cluster headaches for a year+ before this independent anecdote occured. I would have said the same thing before I saw it work first hand. An anecdote should not be evidence by itself, but I hope we can be charitable and recognize that when offered in additon to many other forms of evidence that it is not reason to disagree by itself.

If I can sum up the evidence:

1. The only FDA approved acute cluster headache treatment is an analogue (!) of DMT, with the same underlying mechanism  of 5-HT1B/1D agonism

2. Many published studies point towards multiple classes of serontonergic psychedelics being very effective for preventing and aborting cluster headaches

3. The leading advocacy org ClusterBusters has been advocating for the use of DMT for 20+ years

4. The effect is immediate and massive to a scale not seen in any studied placebo effect and widely recognized among people who try it

May we examine the whole body of research at once and recognize that yes, any lone anecdote or piece of the puzzle would not be sufficient to have this level of confidence. And how blessed we are to find ourselves with many independent sources of confidence (mechanistic rationale, drug class success, broad advocacy agreement) that we may rely on rather than a lone anecdote.

It's also quite unlikely that the 3% placebo figure was complete pain abortion in seconds - as it states this is after 10 minutes and cluster headaches are not known to suddenly end on their own as we see with DMT. Within 10 minutes I'd actually expect 3%+ to naturally come to an end.

I'd love to see evidence like this for similarily terrible diseases. Placebo effects are much more common for lower intensities of pain than shockingly painful ailments like Trigeminal Neuralgia, Kidney Stones and Appendicitis.

I am happy to say that an RCT would provide a standard of evidence not currently available - but to think that we should not offer this as an option to any cluster headache patient who wants to try would be quite a tragedy. 

Thank you both for the care to offer your time and responses on this matter. You're right that anecdotes are often unreliable and that rigor matters, but when someone mid-attack goes from fetal position to conversational in 30 seconds—and this happens repeatedly across independent patients who've tried dozens of ineffective treatments—the usual confounds don't apply.

To go from a feeling worse than torture, to totally okay in a few moments is not any standard placebo effect. Such immediate effect sizes of this magnitude are not something that placebos do. What do placebo responses actually look like for cluster headaches? From the sumatriptan RCT: 3% were pain-free at 10 minutes after placebo. The reported DMT effect (complete resolution in seconds, repeatedly, across independent patients) is orders of magnitude outside that envelope. Cluster attacks last 15–180 minutes. When someone experiences complete resolution in seconds neither placebo nor regression to the mean explains it.

The only FDA-approved acute treatment for cluster headache (subcutaneous sumatriptan) is a sulfonated DMT derivative that works via serotonin receptor agonism. DMT's efficacy is not a speculative hypothesis but a pharmacological expectation.

The evidence is clear that serotonergic psychedelics are extremely effective at the prevention and abortion of cluster headaches. We see this in Psilocybin (an RCT), in LSD, in BOL-148, and in 5-MEO-DALT. Even with zero specific evidence of anyone trying DMT, the prior should actually be quite high that DMT would have an effect for cluster headaches. But we do have evidence. 30 years of people ending their torture with a fast and reliable treatment. We have spent many hours asking about many kinds of treatments with many patients. No class of treatment compares to serotonergic psychedelics. 

As Bob Wold, who founded ClusterBusters in 2002 says:

"One inhalation [of DMT] will end the attack for most people. Everybody is reporting the exact same thing. […] It could end that attack in less than a minute. […] You can take one inhalation, you can wait 30 seconds, and if that cluster is not gone completely, then you know it's time to take another inhalation. You don't have to wait 2h into a psilocybin trip."

I want to mention again that DMT is a Schedule I substance. Schedule I classification requires difficult regulatory approval and specialized DEA licenses, and DMT's unpatentable status means no company has financial incentive to fund trials. This is why we see investment only when the patent barrier is solved—investors committed $26 million this year to develop a psilocin analog (Conjugated Psilocin) specifically for chronic cluster headache, betting on the same serotonergic tryptamine mechanism.

Regarding psychosis risk: DMT has already been administered to 100+ participants across multiple Phase I/II depression trials, with systematic reviews finding no serious adverse events and no prolonged psychotic reactions in controlled settings—the risk at therapeutic doses with psychiatric screening appears very low, not speculative.

Solutions of this magnitude deserve to be legal for compassionate use. The worst case of permission to try is a 15-minute experience they chose to risk. The worst case of prohibition is years of agony they had no choice in. For this body of evidence, I am a staunch supporter of letting people try what has become known in the cluster headache community as a miracle treatment.

The effect size is incredible and the percentage of people for whom it's effective for is very large. 

Yes, we agree that it wouldn't be hard to show in a clinical trial. The reasons why it hasn't been taken through trials are a massive failure of incentives - how many millions would be expected to take a (Schedule I) substance through clinical trials, all for a drug that can't be patented? (Though I do believe a solid return on investment is possible, especially with orphan drug designation and the uniquely high safety and efficacy profile that low dose DMT has for cluster headaches)

For the first time this year, a company will conduct clinical trials on a psychedelic preventative for cluster headaches, for which they've raised 6 million. I expect it to be very effective. Sadly, the uptake time makes it not possible to hit generally accepted headache abortive endpoints. A significant reason for them being able to raise this money is their already developed unique analogue, allowing them to have a composition of matter patent - something not possible for plain DMT. 

To go from Hell, back to baseline in a few moments would be an incredible placebo. And for people who have tried hundreds of treatments with no such placebo, it becomes apparent how real such an effect is.

I have seen a friend in the middle of a cluster headache try DMT for the first time and go from 7/10 pain (on an exponential scale - curled up in a ball in pain) to a 2/10 pain and back to talking normally in just a few seconds. They had never experienced such a quick resolution of pain in any of their ~100 headaches, which normally drag on at some level for hours. They now keep this DMT pen on them at all times, calling it their headache "epi-pen".

This is not a placebo effect.