Research journey
Ayahuasca: 17 trials, 255 papers, one canvas
Two hundred and fifty papers against sixteen trials. A living religious and traditional practice, a large observational literature, one landmark randomised trial, and two drug candidates, neither of which is the brew.
Milestones are curated from the Blossom research database. Registry completion dates are shown as planned; probability-weighted forecasts are part of Blossom Enterprise.
Hover any marker to preview its callout on the canvas, then click to pin it. Pinned is what stays open, fills this panel with the full context and links, and is what an export captures. Landmarks stay labelled on the canvas so the exported image reads on its own.
Download the artifact
Exports rasterise the live SVG, so whatever you have toggled is what you get.
Practice, and the record of it
1851 to 1929Long-standing Amazonian use, from the point Europe began writing it down.1851
Ayahuasca enters the European record
The botanist Richard Spruce documents caapi on the Rio Uaupés and sends specimens back to Britain. The practice he was describing was already old.
The date on this marker is the date Europe started writing things down, not the date the practice began. Ayahuasca is a decoction of the Banisteriopsis caapi vine with a plant containing N,N-dimethyltryptamine, usually the leaves of Psychotria viridis, and preparations of it are used across a wide part of the western Amazon by peoples including the Shipibo-Konibo, the Shuar, the Tukano and many others. The names differ, the plant admixtures differ and the practices differ; treating them as one thing is a convenience of the literature rather than a fact about the Amazon.
Placing any of this at the far left of a research timeline risks implying it was a rough draft of medicine. It was not, and it still is not. These are healing, divinatory and social practices embedded in cosmologies that do not divide the world the way biomedicine does, and they have their own criteria for whether something worked. They were not waiting for a randomised trial to tell them.
What the Western record added, over the following century, was a chemistry: two classes of alkaloid, an explanation for why an orally inactive tryptamine becomes active when combined with a monoamine oxidase inhibitor, and eventually a way to measure a dose. What it did not add, and mostly still has not, is the part the traditions supply around the substance. Almost every question further right on this canvas about set, setting, preparation and integration is really a question about how much of that can be rebuilt in a clinic.
The churches and the law
1930 to 1992Syncretic congregations give the brew an organised, and eventually legal, constituency.1930
Santo Daime is founded in Acre
Raimundo Irineu Serra founds the first of the Brazilian syncretic ayahuasca religions. Barquinha follows in 1945, União do Vegetal in 1961.
This is the event that makes the rest of the canvas possible. The Brazilian churches took a forest practice and gave it an urban, organised, documented constituency: congregations with membership rolls, hymnals, doctrine, minors and elders, and a stated position on what the drink is for. União do Vegetal in particular kept records of the kind researchers can work with.
Two consequences follow. The first is legal. A church can be argued about in a courtroom in a way that a diffuse practice cannot, and both the Brazilian authorisation of 1987 and the US Supreme Court ruling of 2006 turned on religious freedom rather than on drug policy. The second is scientific. Every long-term human safety dataset on the left half of this canvas exists because a church agreed to be studied.
It is worth being precise about what these organisations are. They are religions, not treatment providers, and their members are congregants, not patients. Research conducted with them borrows a population that assembled itself for entirely different reasons, which is a strength for external validity and a weakness for causal inference.
1984-04
Journal of Ethnopharmacology
The pharmacology of the combination is set out
A survey of tryptamine and beta-carboline constituents explains why the brew works: the vine inhibits the enzyme that would otherwise destroy the DMT in the gut.
DMT taken by mouth on its own does almost nothing, because monoamine oxidase in the gut wall and liver breaks it down before it reaches the brain. The beta-carbolines in Banisteriopsis caapi, principally harmine, harmaline and tetrahydroharmine, inhibit that enzyme. Combine the two and an orally inactive compound becomes orally active for several hours.
That is the whole pharmacological trick, and it was worked out in the Amazon long before it was described in a journal. The chemistry papers explain the mechanism; they do not explain how anyone found the combination among the thousands of plants available, and no honest account of that question exists.
This is also the point where the timeline forks. Once you know the mechanism, you can synthesise it: isolated DMT plus an isolated MAO inhibitor, in known milligrams. That preparation is usually called pharmahuasca, and it is pharmacologically related to the brew and culturally unrelated to it. Everything in the pharmahuasca lane at the right of this canvas descends from this insight.
1986
A US patent is granted on a Banisteriopsis caapi cultivar
A plant patent on a caapi variety collected in Ecuador is granted in the United States, and an Amazonian indigenous coalition spends the next decade contesting it.
The patent covered a cultivar, not the brew and not the alkaloids, and its commercial value was close to nothing. Its significance is entirely symbolic, and the symbolism was not lost on anyone: a plant central to the religious life of several hundred communities became, on paper, the property of someone from outside them. A coalition of Amazonian indigenous organisations challenged it, the patent office rejected it in 1999, the rejection was reversed in 2001, and the patent ran out on its own.
Keep this marker in view when reading the money lane, which is otherwise almost empty. Ayahuasca is a plant preparation. Plant preparations of variable composition are close to unpatentable in the way that matters to a pharmaceutical company, which is a claim on a specific molecule at a specific dose for a specific indication. Without that claim there is no way to earn back the two to three hundred million dollars a full development programme costs.
So the absence of drug programmes on this canvas is not evidence that the compound is uninteresting. It is what the incentive structure produces. The only ayahuasca-adjacent work that has attracted commercial capital is the synthesised formulation at the far right, and that is precisely because it can be owned.
1987
Brazil authorises religious use
After a government commission visits the churches and reviews their practice, Brazil permits ritual ayahuasca use. It has never been prohibited there since.
Brazil suspended religious ayahuasca use in 1985 and then, unusually, sent a commission to look at the churches before deciding. The commission found organised communities with low rates of the harms the prohibition was meant to prevent, and the suspension was lifted in 1987. A further resolution in 2010 set out conditions, including a prohibition on commercial sale.
This is the most consequential policy event on the canvas and it is easy to miss, because nothing dramatic happened. A country with the largest ayahuasca-drinking population in the world decided the practice was lawful, which is why the entire clinical programme further right is Brazilian. Researchers in São Paulo and Natal could obtain the brew from a church, verify its alkaloid content and dose people with it. Researchers almost anywhere else could not.
It also set the terms. Brazil legalised a religious practice, not a medicine, and the distinction still constrains what can be studied and how. The brew used in the Brazilian trials was prepared by a church, which is a strength for ecological validity and a complication for the pharmaceutical standard of a known, reproducible product.
Studied without being trialled
1993 to 2011The Hoasca Project, then two decades of observational and naturalistic work.1993
Hoasca Project
The Hoasca Project studies long-term ritual users
An international team assesses União do Vegetal members in Manaus, the first serious biomedical study of people who had been drinking ayahuasca for years.
The Hoasca Project did something no clinical trial can do. It examined people with a decade or more of regular, supervised exposure, and compared them with matched non-drinking controls on psychiatric, neuropsychological and physiological measures. The findings were broadly reassuring: no evidence of cognitive deterioration or psychiatric deterioration, and in several domains the drinkers scored better than the controls.
Be precise about what that can and cannot establish. It is a cross-sectional comparison of a self-selected group. People who stay in a demanding religious community for ten years are not a random sample of the people who joined it, and anyone harmed by the practice is disproportionately likely to have left before the researchers arrived. Survivorship works in exactly the direction that would produce this result even if the drink did nothing.
What it does establish is the absence of gross harm in a population where gross harm would have been visible, which is genuinely useful and was not previously known. The project also produced the first human pharmacokinetics of the brew, and a follow-up assessment of regular users two decades later. This is the shape the ayahuasca literature takes: careful observation of people who were going to drink anyway.
2001-02
Universitat Autònoma de Barcelona
Ayahuasca reaches the laboratory
A Barcelona group runs the first controlled human administration studies of encapsulated freeze-dried ayahuasca, establishing tolerability, dose ranges and cardiovascular effects.
These studies did the unglamorous work that everything downstream depends on. They gave measured doses of a standardised, alkaloid-quantified preparation to healthy volunteers under laboratory conditions, and reported what happened to blood pressure, heart rate, subjective state, hormones and the EEG. Nausea and vomiting were common; nothing dangerous appeared at the doses used.
They were not treatment studies and made no therapeutic claim. Their value is that they turned ayahuasca into something with a dose-response curve, which is the minimum requirement for anyone who wants to test it in patients. A companion paper from the same period set out the regulatory obstacles to doing exactly that, and those obstacles are why the first patient trial is still eleven years to the right of here.
Note what has been lost and gained in the move indoors. A capsule of freeze-dried brew taken in a quiet room with a blood pressure cuff is chemically the same substance and experientially a very different event from a night in a ceremony. Whether the difference matters to the outcome is one of the central unresolved questions on this canvas.
2005-06
Adolescents in the churches are assessed
A set of studies examines teenagers raised in ayahuasca-using congregations, on psychiatric, neuropsychological and drug-use measures.
The churches administer ayahuasca to minors, which is the single most uncomfortable fact in this whole area and the one most often skirted. Rather than skirt it, a group of researchers went and measured: psychiatric assessment, neuropsychological testing, and self-reported use of alcohol and other drugs, against matched controls.
They found no excess psychopathology and no cognitive deficit, and lower reported alcohol use among the church adolescents than among the comparison group. The same caveats as the Hoasca Project apply with more force, since the comparison is between religious and non-religious adolescents and religion itself predicts lower drinking.
The honest summary is that the evidence available does not show harm, and that the evidence available is not the kind that would reliably detect it. Anyone using these papers to argue that the practice is safe for children is over-reading them, and anyone dismissing them because the topic is uncomfortable is under-reading them.
2006-02
The US Supreme Court protects UDV religious use
A unanimous ruling under the Religious Freedom Restoration Act allows the União do Vegetal to import and drink ayahuasca in the United States.
The government argued that the Controlled Substances Act admitted no exceptions. The court disagreed, holding that the state had not shown a compelling interest in barring a small religious group from a sacrament it had used for decades without demonstrable harm. The evidence about that harm came substantially from the observational literature on the left of this canvas.
This is a rare case of a research finding changing a legal outcome for a psychedelic, and it is worth noticing which kind of research did it. Not a randomised trial. Cohort studies of people who were already drinking, conducted by researchers who went to where the practice was rather than importing it into a clinic.
The ruling is narrow. It protects specific religious use by specific congregations, and it is not a general permission, not an approval, and not a finding that ayahuasca is safe. Outside that carve-out and outside Brazil, ayahuasca remains prohibited in most jurisdictions because the DMT it contains is controlled under the 1971 Convention, even though the plants themselves are not.
2010-12
The practice spreads, and the questions follow it
A review of ayahuasca outside the Amazon sets out what happens when a place-bound practice becomes a global one, including who is left behind by it.
By this point ayahuasca was being drunk in European cities, at retreat centres in Peru and Costa Rica built for foreign visitors, and in loosely organised ceremonies with no lineage at all. The demand created an economy: retreat tourism, guide training, vine harvesting at a scale the forest had not previously been asked for.
The problems that come with that are practical, not abstract. Banisteriopsis caapi is slow-growing and wild-harvested in many places, and demand pressure on it is documented. Facilitators without accountable communities behind them have been credibly reported for sexual and financial exploitation of participants. And the money moves overwhelmingly outward: the communities whose knowledge is being drawn on are rarely the ones capturing the value.
Blossom shows a compound page with 250 papers on it. Almost none of that literature was designed by, or reports back to, the peoples whose practice it studies. That is a fact about the evidence base and not a moral aside, because it shapes which questions got asked and which did not.
A small clinical record
2012 onwardsBrazil runs the trials; the literature stays far larger than the evidence base.2012-01
University of São Paulo, Ribeirão Preto
The first clinical trial opens in Brazil
An open-label study gives a single dose of church-prepared ayahuasca to patients with recurrent depression, with SPECT imaging before and after.
The formal clinical record starts here, 161 years after the first European note and roughly two decades after the Hoasca Project. It is a small, unregistered, single-arm study: seventeen patients, one dose, symptom scales out to three weeks, and brain perfusion imaging to see whether anything changed physically.
Depression scores fell quickly and stayed down over the follow-up, and perfusion increased in regions associated with emotional processing and interoception. In an open-label design with no control group, none of that separates drug effect from expectation, and the authors said so. What the study established was feasibility: that this could be done at all, in a hospital, with a preparation obtained from a church and its alkaloid content verified.
The preliminary report appeared in 2015 and the imaging paper in 2016. Both are among the most cited ayahuasca papers in Blossom, which is itself a comment on how thin the interventional literature is.
2012-08
Ritual users followed over a year
A longitudinal study of regular ayahuasca users and controls reports no psychopathology and better performance on several neuropsychological measures.
This is the observational literature at its most careful: the same people measured twice a year apart, with a matched comparison group, on personality, psychopathology, life attitudes and cognition. It found no deterioration and some advantages, replicating the Hoasca findings in a different population with a stronger design.
A longitudinal design removes some of the problem and not the important part of it. Following users over time controls for measurement occasion, but it still cannot tell you what would have happened to those people had they not been drinking. Only randomisation does that, and randomising people into years of religious practice is not possible.
This is the ratio in miniature. Blossom holds 250 ayahuasca papers and 16 trials, and the great majority of the 250 are studies of this kind: rigorous, informative about safety and about who drinks, and structurally incapable of establishing that the drink caused the outcome.
2014-02
Federal University of Rio Grande do Norte
The randomised placebo-controlled trial
A trial in Natal randomises patients with treatment-resistant depression to a single dose of ayahuasca or an inert placebo, with response measured to seven days.
This is the study the rest of the canvas points at. Randomised, parallel-group, with a placebo built to match the taste and colour of the brew, in patients who had already failed at least two antidepressants. Depression scores were significantly lower in the ayahuasca group at one, two and seven days, and the effect was large enough to show up in a sample this size.
Three limitations belong next to that result and are stated in the paper. The sample is small, so the confidence intervals are wide. The follow-up is seven days, which says nothing about durability. And blinding almost certainly failed, because a substance that produces vivid visual imagery and vomiting is not hard to distinguish from mineral water; the placebo arm improved substantially too, which tells you the comparison was not psychologically inert.
It has been cited more than eight hundred times and it remains, a decade later, the only adequately powered randomised placebo-controlled ayahuasca trial for a psychiatric indication in this database. Compare the compounds that have a pharmaceutical sponsor: psilocybin went from a first randomised readout to multi-site Phase 3 in under a decade. Ayahuasca produced a positive randomised result in 2018 and nothing has been built on it, because there is nobody whose job it is to build on it.
2015-02
PLOS ONE
The default mode network changes
Functional imaging shows ayahuasca reducing activity and connectivity in the default mode network, the system associated with self-referential thought.
The default mode network is the set of brain regions most active when a person is not doing anything in particular: remembering, planning, ruminating, thinking about themselves. It is overactive in depression, and reducing its activity is the mechanistic story most commonly offered for why psychedelics might help.
This study found that reduction under ayahuasca, and later work from the same and other groups found related changes in signal complexity and in the persistence of connectivity patterns after the acute effects had passed. A secondary analysis of the randomised trial also reported a rise in brain-derived neurotrophic factor.
Mechanistic plausibility is not efficacy, and it is worth resisting the pull of a satisfying story. Plenty of interventions change the default mode network without treating anything. These findings explain how an effect could work; they cannot demonstrate that it does.
2015-09
University of São Paulo
Social anxiety pilot takes four years to recruit seventeen people
A randomised, double-blind pilot in social anxiety disorder screens 894 volunteers to enrol 17, and takes from September 2015 to July 2019 to finish.
The design is sound: randomised, double-blind, placebo-controlled, with an organoleptically matched placebo and a simulated public speaking test as the primary outcome. The brew was prepared by a Santo Daime church and its alkaloid concentrations verified by mass spectrometry.
The numbers are the story. Eight hundred and ninety-four people screened, seventeen randomised, nine to ayahuasca and eight to placebo, with equipment failure reducing some analyses to fourteen. A pilot of that size can establish feasibility and produce effect-size estimates for a later trial. It cannot establish whether ayahuasca treats social anxiety, and the authors did not claim it could.
Four years to enrol seventeen participants in the one country where this research is legal, with a preparation the researchers had to obtain from a church, and no commercial sponsor. That is the practical texture behind the sixteen-trial figure at the top of this page.
2017-11
The Global Ayahuasca Survey opens
A large international online survey begins collecting data from ayahuasca drinkers, eventually running to tens of thousands of responses across dozens of countries.
This is how the ayahuasca literature grew to 250 papers without 250 trials. A self-selecting online cohort cannot demonstrate that ayahuasca causes anything, but it can do things no trial can afford: describe who drinks and why, map settings from indigenous ceremony to European neo-shamanic circle, and count events too rare for a trial of thirty people to see.
Successive analyses from it have reported associations with lower problematic alcohol use, differences in outcome by context and setting, and long-term wellbeing measures in people who continue drinking. Read every one of those as a description of a population, not as an effect of a drug. People who feel a practice is helping them are the ones who keep doing it and the ones who answer surveys about it.
The survey has been most valuable where trials are weakest, which is on harms. That is the next marker.
2020-12
Psychiatric University Hospital, Zurich
Pharmahuasca enters the clinic
A Zurich Phase I study gives healthy volunteers a formulation of synthesised DMT and harmine, and measures brain network dynamics and prosocial processing.
This is a different object from the brew, and the distinction should be held firmly. Ayahuasca is a decoction of at least two plants whose alkaloid ratios vary between churches, batches and seasons. Pharmahuasca is isolated DMT plus an isolated monoamine oxidase inhibitor, in milligrams, in a capsule. Pharmacologically they are relatives. Culturally they have almost nothing to do with each other, and only one of them is anybody's sacrament.
The clinical logic for the substitute is strong. A regulator will not license a variable plant decoction prepared by a religious organisation, and a trialist cannot blind one. A fixed-dose formulation is reproducible, can be titrated, and can be given to a control group in an identical-looking capsule. Everything that makes it a good drug candidate is a consequence of stripping out what made it a ceremony.
It is also the point where money becomes possible. A formulation can be patented in a way a brew cannot, which is why the only ayahuasca-adjacent work on this canvas with commercial backing is in this lane. Whether the results transfer in either direction, from brew to formulation or back, is untested.
2021-10
Beckley Med Foundation
Ayahuasca-assisted therapy for grief
An open-label trial in Spain tests ayahuasca alongside a structured therapy for prolonged grief, one of the few ayahuasca studies with a defined psychotherapy protocol.
Most of the clinical work on this canvas gives a dose and measures a scale. This study did something else: it paired the session with a specific constructivist therapy aimed at reconstructing meaning after a bereavement, which is closer to how the compound is actually used outside research.
The trial is open-label, so expectancy is uncontrolled, and its registry status is unresolved. A non-randomised clinical report on ayahuasca-assisted meaning reconstruction for bereavement appeared in 2025 and reported improvement in grief severity.
Grief is an interesting choice of indication and an awkward one. It is not straightforwardly a disorder, the natural course is improvement, and an uncontrolled study over months will show recovery whatever the treatment does. That is not an argument against the work; it is an argument for a control group.
2022-01
The reciprocity question is put formally
A paper on epistemic injustice in ayahuasca research argues that biomedical study of the brew has systematically excluded the knowledge systems it draws on.
The argument is specific rather than rhetorical. Biomedical research treats indigenous and mestizo practice as a source of raw material, a plant and a hypothesis, while granting evidential authority only to the instruments that arrive later. Practitioners appear in the literature as subjects and almost never as investigators, co-authors or beneficiaries.
The practical stakes are visible on this canvas. Peru declared traditional ayahuasca use national cultural heritage in 2008 and Brazil protects religious use, but neither creates any obligation on a Swiss laboratory synthesising DMT and harmine, and neither returns anything to the communities involved. If the pharmahuasca work at the right of this canvas succeeds commercially, the mechanism by which value flows back to the Amazon does not currently exist.
There is a version of this argument that is sentimental and a version that is operational, and the operational one asks concrete questions. Who is on the protocol. Who holds the intellectual property. Where the vine came from and what was paid for it. Whether the results are reported back in a language the source communities read. Those are answerable, and almost none of the 250 papers in this database answers them.
2022-09
Harmine on its own is dose-ranged
A Phase I single-ascending-dose study establishes the maximum tolerated oral dose of pure harmine in 27 healthy volunteers.
Harmine is the vine without the visions. It inhibits monoamine oxidase, and separately it inhibits the kinase DYRK1A and stimulates proliferation of human neural progenitor cells in vitro, which is why a line of preclinical work treats it as a candidate in its own right for neurodegenerative and metabolic indications rather than as an enabling ingredient.
Taking the brew apart into its components is what pharmacology does, and it is genuinely informative: knowing what harmine does alone is a prerequisite for attributing anything to the combination. It also produces a compound with no ceremonial identity at all, which nobody would recognise as ayahuasca.
The study used a continual reassessment method to find the maximum tolerated dose, and the results were published in 2024. It is a dose-finding study in healthy people and says nothing about whether harmine treats anything.
2022-11
ended earlyAdverse effects get a denominator
The Global Ayahuasca Survey reports physical and mental health adverse effects across thousands of drinkers, with most experiencing at least one.
Vomiting and nausea are so routine in ceremony that they are often treated as part of the process rather than as side effects, which is exactly the sort of thing that goes unrecorded until someone asks systematically. This analysis did ask, and found that adverse physical effects were reported by the large majority of respondents, and adverse mental health effects by a substantial minority, with a small proportion needing professional help afterwards.
Risk was higher in people with a history of mental illness, and the specific serious events documented elsewhere in this literature include acute psychotic episodes, prolonged psychological distress after first-time ceremony use, and the well-established hazard of the MAO inhibition itself. Harmala alkaloids interacting with serotonergic antidepressants, tyramine-rich foods or stimulants is not a theoretical concern; it is the most predictable serious harm ayahuasca carries and it is a direct consequence of the mechanism described in 1984.
The framing matters. Reporting most drinkers experiencing an adverse effect sounds alarming and is not the same as reporting that most drinkers were harmed; a survey cannot tell you which of these events participants regarded as a cost worth paying. What it can tell you is that the practice is not benign, that screening matters, and that the medication interaction should be checked every time.
2023-02
Formulation tested inside a meditation retreat
A double-blind placebo-controlled study gives DMT and harmine to experienced meditators during a mindfulness retreat, with 40 participants.
An unusually thoughtful design. Rather than stripping the context away and then wondering what was lost, the researchers built one deliberately: a residential retreat with a trained meditator population, and the formulation randomised against placebo inside it. It is the closest thing in the interventional literature to a controlled test of set and setting.
Results were published in 2024 and 2025, covering subjective experience, brain connectivity at rest and mixed-methods accounts of what participants made of it. Healthy experienced meditators are not patients, so this is mechanism and phenomenology rather than treatment.
2023-05
Reconnect Labs
The first company-sponsored study
Reconnect Labs runs a randomised Phase I dose-finding study of a fixed DMT and harmine combination in 16 healthy volunteers across six dosing days.
This is the money lane's only real entry, and it arrives 172 years after the first marker on this canvas. A company sponsoring a dose-finding study of its own formulation is unremarkable in any other compound on Blossom. Here it is the exception that proves the rule, and it is a formulation rather than the brew.
The study compared six doses of a fixed combination with continuous psychological support, and the pharmacokinetic and pharmacodynamic results were published in 2025. Related work from the same lane has explored alternative routes of administration intended to shorten and control the onset, which is a direct attempt to solve the practical problems, the four to six hour duration and the nausea, that make the traditional preparation hard to run in a clinic.
Note what has to be true for capital to appear. A defined molecule ratio, a manufacturable product, an ownable claim. The brew fails all three, which is why the 250 papers on the left of this canvas were paid for by universities, foundations and churches rather than by anybody expecting a return.
2023-05
ended earlyThree registered trials never enrol anyone
Three Australian registrations, two microdose pilots and a DMT-harmala formulation study, sit at "not yet recruiting" with completion dates that have long passed.
Registered intent and delivered research are different quantities, and the gap is usually invisible because pipeline charts quietly drop the records that went nowhere. Three of the sixteen ayahuasca trials in this database are in that category, which is close to a fifth of the entire clinical record.
Two of them are registrations of the same small pilot, a single 30 mg microdose of ayahuasca alkaloids in four or five healthy adults with blood neurotransmitter and inflammatory marker outcomes. The third is a crossover study of two oral DMT and harmala formulations in eight volunteers. All three were listed as not yet recruiting with completion dates in 2023.
Leaving them on the canvas is the honest choice. A field with sixteen trials cannot afford to round three of them away, and the reasons small unfunded studies stall, which are money, supply and regulatory patience, are the same reasons this compound has so few trials in the first place.
2023-10
Advanced Integrative Medical Science Institute
A standardised preparation is dose-ranged in the US
An open-label Phase I study tests three single oral doses of SM-001 in twelve healthy adults, at a private integrative medicine institute in Seattle.
The registry entry describes an oral preparation dosed by volume per kilogram, with safety and plasma biomarkers as outcomes. That is about as much as the public record supports, and the study is listed with unknown status and no reported results.
It is included because it is the only US registration on this canvas for a whole ayahuasca-type preparation rather than an isolated constituent, and because it shows the shape the field would take if the regulatory route opened: a standardised, quantified product with a name, run by a clinic rather than a university.
2024-01
Insel Gruppe AG, University Hospital Bern
PET imaging of the formulation in Bern
A randomised single-blind crossover FDG-PET study measures what an oral harmine and DMT formulation does to brain glucose metabolism against placebo.
PET with fluorodeoxyglucose measures how much glucose brain tissue is consuming, which is a direct index of neural activity rather than the indirect blood-oxygen signal that fMRI reads. Running it against placebo in a crossover design is a clean way to ask what the formulation does to brain metabolism.
The result, published in 2026, was a global increase in cerebral glucose metabolism. That is a mechanistic finding in healthy male volunteers and carries no therapeutic claim, but it is the most physiologically direct measurement anyone has made of an ayahuasca-type preparation.
2025-12
University of São Paulo
Three new Brazilian trials open at once
Three randomised double-blind studies compare a single oral dose of ayahuasca with oral esketamine, in PTSD and in two basic-science questions in healthy women.
These are the only ayahuasca trials currently recruiting anywhere in this database, and they are the first new ones since 2023. All three are small, all three run at the same site, and all three use esketamine as an active comparator rather than an inert placebo, which is a sensible response to the blinding problem: two substances that both do something are harder to tell apart than one that does and one that does not.
The indications are worth noticing. One is PTSD, with ten participants. The other two are basic science in healthy volunteers, on body image perception and on premenstrual symptoms, with twenty participants each. Two of the three are not treatment studies at all.
Registry completion dates were set at July 2026 and all three remain listed as recruiting, so those dates will move. The Brazilian programme that produced the only randomised evidence on this canvas restarted after a decade, and it restarted at pilot scale.
2029-01
plannedNo Phase 3 trial is scheduled
Nothing beyond Phase II is registered for ayahuasca in this database, and no sponsor has announced one. The right-hand side of this canvas is empty on purpose.
This marker records an absence rather than an event, which is unusual and, for this compound, the most accurate thing the canvas can show. Sixteen trials, the largest with 84 planned participants, none above Phase II, and no registered plan to go further. On a timeline for any pharmaceutical compound this space would be full of readouts and filings.
It would be wrong to read that as a verdict on the evidence. The 2018 randomised result was positive and has never been contradicted; it has simply never been followed up, because following it up would cost a few hundred million dollars and nobody would own the result. The absence here is economic, not scientific.
The two drug candidates that do exist prove the point rather than contradicting it. Both are synthesised DMT plus harmala formulations, and both are owned. Capital arrived for ayahuasca only at the moment the brew was replaced by something a company could hold a claim on, which is a statement about intellectual property rather than about pharmacology.
It would also be wrong to read it as a problem that necessarily needs solving. Ayahuasca is currently drunk by hundreds of thousands of people in churches, ceremonies and retreats, most of whom are not seeking a licensed medicine and would not benefit from one existing. A regulatory approval would create a supervised route for patients who need one; it would not settle whether the practice on the left of this canvas should change at all. Those are separate questions, and this timeline is only equipped to answer the first.
Ayahuasca
Ayahuasca: the research journey