Research journey
Ibogaine: 13 trials, 84 papers, one canvas
Nine registered trials against seventy-nine papers. A century of Bwiti practice, sixty years of offshore treatment, a documented cardiac risk, and a clinical record that has only just started.
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.
Traditional use and observation
1864 to 1969Bwiti practice, and Lotsof's accidental discovery.1864
Iboga enters the European record
French naturalists send Tabernanthe iboga specimens to Europe. The Bwiti practice they were describing is far older than the record of it.
The date on this marker is the date Europe started writing things down, not the date the practice began. Iboga root bark has been central to Bwiti initiation in Gabon and neighbouring countries for generations, and the ceremony is a religious and social institution rather than a proto-clinical trial. Placing it 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.
What Western chemistry added, starting with the isolation of ibogaine from the root bark around 1901, was a single purified molecule and a dose that can be measured. What it left behind was everything Bwiti supplies around the substance: preparation, a community, people who have done this before, and a framework for interpreting what happens. Almost every safety argument further down this canvas is really an argument about how to rebuild some of that in a clinic.
Gabon has since treated iboga as national heritage rather than a freely tradeable botanical, and the question of who benefits from an industrialised ibogaine sits under every capital event on the money lane. It is a live question, not a historical footnote.
1962
Lotsof observes opioid withdrawal interrupted
Howard Lotsof, then a heroin user in his teens, takes ibogaine recreationally and finds his withdrawal symptoms gone.
The founding observation of the entire anti-addiction case for ibogaine came from someone taking it for fun and noticing that something else had happened. He was not withdrawing, and he did not want the drug. Lotsof spent the next four decades trying to turn that into a medicine.
It is an unusually honest origin story, and it sets the pattern for what follows. The signal here has always been strong enough to keep people interested and never been produced under conditions that would satisfy a regulator. Fifty-one years pass between this moment and the first ibogaine trial in our database.
What would it take to test a claim like this properly? A randomised comparison against the current standard of care in opioid use disorder, powered for relapse rather than for withdrawal scores. That study has still not been run.
The offshore decades
1970 to 2012Schedule I; clinics carry the caseload while the safety record accrues.1970
ended earlyUS Schedule I
Ibogaine is placed in the most restrictive US drug category, and the compound leaves the reach of ordinary research.
The US restricted ibogaine in 1967 and confirmed it as Schedule I under the Controlled Substances Act in 1970. Schedule I means no accepted medical use and no accepted safe use under supervision, which is a claim about evidence as much as about risk.
The practical effect was to make ibogaine research expensive and slow at exactly the moment Lotsof was trying to interest people in it. Meanwhile ibogaine-containing preparations had been sold openly in France as a tonic for decades and were withdrawn in the 1960s. Two jurisdictions, opposite directions, neither driven by a trial.
Read the rest of this canvas with the scheduling in view. Treatment did not stop. It moved to countries where the rules were looser and the oversight was thinner, which is the single most consequential fact about ibogaine safety.
1985-02
Howard Lotsof
Lotsof patents ibogaine for addiction
US patent 4,499,096 claims ibogaine for interrupting narcotic dependence, giving a Schedule I compound a commercial owner.
A patent is not a permission slip. Lotsof secured intellectual property covering the use he had observed twenty-three years earlier, then spent the following decades trying to attract the capital and the regulatory goodwill needed to test it.
The patent matters for a reason that only becomes visible further right on this canvas. Ibogaine itself is an old, unpatentable molecule with expired use claims, so nobody can build a conventional pharmaceutical return on it. Every well-funded programme on this timeline is instead working on a modified molecule. That is not a scientific accident. It is what the incentive structure pays for.
1993-07
Neuropharmacology
Cocaine self-administration falls in rats
Preclinical work shows ibogaine reducing cocaine self-administration, the result that put public money behind the compound.
Preclinical means animal and cell work, before any human trial. Rats given ibogaine pressed the lever for cocaine less often, and the effect outlasted the drug in the bloodstream. That combination is what made the field take the anti-addiction claim seriously, because it suggested something was being changed rather than merely blocked.
This is still one of the most cited papers in our ibogaine set. It also anchors a pattern worth naming early: the strongest evidence for ibogaine is preclinical and observational, and the strongest evidence against it is clinical.
1995
National Institute on Drug Abuse (NIDA)
ended earlyUS federal funding stops
NIDA declines to fund clinical development after animal work shows cerebellar damage at high doses. The programme loses its public sponsor.
The National Institute on Drug Abuse had funded preclinical work and a human study had cleared regulatory review at the University of Miami. Then rodent findings of Purkinje cell degeneration in the cerebellum (the part of the brain that coordinates movement) at high doses arrived, and the federal appetite went with them.
Whether those doses were relevant to human treatment is still argued. What is not arguable is the effect on the field. With no public funder and no patentable molecule, ibogaine development moved almost entirely to private clinics operating outside the countries that had restricted it.
This is the hinge of the whole timeline. Everything between 1995 and 2013 on this canvas happens in clinics rather than in trials, and it is the reason the safety record was written in coroners’ reports before it was written in protocols.
2007-08
The medical subculture is documented
A paper counts the people being treated with ibogaine outside conventional medicine, and finds the great majority of them there.
Alper, Lotsof and Kaplan did something the field badly needed: they counted. Their survey of the ibogaine treatment scene established that a substantial and growing population was receiving ibogaine, overwhelmingly in private clinics, guesthouses and lay settings rather than in hospitals or trials.
Blossom currently lists more than thirty ibogaine providers as organisations, in Mexico, Costa Rica, South Africa, Portugal, Bali and elsewhere. Nine registered trials sit against that. If you want one number that explains ibogaine, it is that ratio.
This is not a story of brave outsiders versus timid regulators, and it is not a story of reckless cowboys either. It is what happens when demand for a treatment is real, the evidence to authorise it does not exist, and nobody with the money to generate that evidence has a reason to.
2009
New Zealand allows prescription use
Ibogaine and noribogaine are classified as prescription medicines, creating one of the few legal supervised routes anywhere.
New Zealand took a different path from the US: rather than prohibiting ibogaine, it made it a prescription medicine, which means a doctor can lawfully prescribe it while carrying the responsibility for doing so safely.
This produced something the rest of the field lacked, which is a small population of legally treated patients with identifiable clinicians and traceable records. Several of the observational safety datasets in this database exist because of arrangements like this one.
Prescription status is not approval. No regulator anywhere has approved ibogaine for any indication, and prescribing it means an individual clinician has accepted the risk that a licensing authority has not.
Into the clinic
2013 to 2024First registered trials; the cardiac risk quantified.2013-07
Radboud University
First registered ibogaine trial
A Phase I study in healthy volunteers examines how the CYP2D6 enzyme changes ibogaine metabolism. Fifty-one years after Lotsof.
The earliest trial in Blossom’s ibogaine record is not a test of whether the compound works. It is a pharmacokinetic study of how the body processes a single 20 mg dose, and specifically of the enzyme CYP2D6, which converts ibogaine into noribogaine at very different rates in different people.
That question is not academic. People who metabolise the drug slowly hold higher blood levels for longer, which is directly relevant to the cardiac risk two markers down. A clinic that does not genotype or monitor is dosing blind.
It is also a small, short, healthy-volunteer study, and it is where the formal record starts. Everything before it on this canvas is either laboratory work or treatment that nobody registered.
2013-10
DemeRx
ended earlyNoribogaine tested in opioid-dependent patients
Two placebo-controlled ascending-dose studies test the metabolite rather than the parent compound, in people discontinuing methadone.
Noribogaine is what the liver turns ibogaine into, and it carries much of the anti-withdrawal activity without producing the same intense visionary state. Testing it directly was the field’s first serious attempt to keep the useful part and discard the rest.
These are the only randomised, double-blind, placebo-controlled studies on the early half of this canvas, and they were run in New Zealand under the prescription framework noted above. They were designed to answer safety and dosing questions, not efficacy ones.
The results, published in 2016, did not clear the compound. Dose-dependent QT prolongation (a delay in the heart’s electrical recovery that can trigger fatal arrhythmia) showed up here too, in a controlled setting, which is the strongest kind of evidence there is that the cardiac signal is real and not an artefact of unsupervised clinics.
2014-11
Radboud University
Radboud opioid pilot runs for six years
An open-label, single-fixed-dose pilot in opioid-dependent patients takes six and a half years to complete.
Open-label means everyone knows who is getting the drug, so expectation cannot be separated from effect. Single fixed dose means no titration. It is a modest design, and it still took six and a half years to finish.
That duration is the point. Recruiting opioid-dependent participants into an inpatient study of a cardiotoxic Schedule I compound, in an academic centre with no commercial sponsor, is close to the hardest thing to do in this field. Meanwhile the clinics on the lane above were treating people continuously throughout.
The pharmacokinetic and pharmacodynamic results reached print in 2024, a decade after the trial opened.
2015-01
Frontiers in Pharmacology
ended earlyThe heart problem is named
A review sets out ibogaine’s effect on cardiac conduction, and the field acquires a mechanism for the deaths it already knew about.
Ibogaine blocks a potassium channel in heart muscle called hERG. That lengthens the QT interval, the window during which the heart’s electrical system resets, and a long enough QT interval can tip into torsades de pointes, an arrhythmia that kills. This is not a rare idiosyncratic reaction. It is a dose-dependent pharmacological property of the compound.
Ibogaine is also usually given to people with several of the things that make QT prolongation more dangerous: opioid dependence, methadone (itself QT-prolonging), electrolyte disturbance from vomiting, dehydration, poor cardiac health, and sometimes slow CYP2D6 metabolism. The risks stack.
Two companion papers in 2016 asked the practical questions that follow: how toxic is ibogaine, and what dose is safe? Neither could give a clean answer, and the honest reading is that the therapeutic window is narrow enough that dosing without cardiac monitoring, electrolyte correction and resuscitation capability is indefensible. Some clinics do all of that. Not all of them do.
2018-06
ended earlyHarms beyond the heart
Case reports document cardiac arrest, mania and persisting perceptual disturbance after ibogaine treatment.
The cardiac signal dominates the conversation, and it should, but it is not the only harm in the record. Published cases describe mania following ibogaine, and hallucinogen persisting perception disorder (visual disturbances that continue long after the drug has cleared) in someone treated for opioid dependence.
Case reports cannot tell you how often something happens. They can tell you that it happens at all, which is enough to require screening and follow-up that most offshore treatment does not provide.
2020-10
ICEERS
ICEERS methadone detoxification study
A Phase II study of ibogaine for methadone detoxification, run by a non-profit research institute.
Methadone is the hardest opioid to come off, because its long half-life stretches withdrawal over weeks. It is also the population where the anti-withdrawal claim for ibogaine would matter most, and the population where the cardiac risk is highest, since methadone prolongs the QT interval on its own.
ICEERS is a non-profit, which matters for what happens to the data. Work funded by a foundation rather than a sponsor tends to be published whatever it shows.
2021-06
Ibogalog patent filed
A patent on simplified iboga-like molecules is filed, claiming the scaffold behind tabernanthalog.
The commercial answer to an unpatentable, cardiotoxic molecule is to build a new one that is neither. This filing claims a family of simplified iboga analogues, and it was granted in 2022.
Watch what this does to the shape of the field. Capital flows to the analogues lane while the ibogaine lane above it stays academic and underfunded, which means the compound people are actually taking is the one nobody is paying to study.
2021-08
Journal of Psychedelic Studies
ended earlyAdverse events systematically reviewed
An updated systematic review pulls the scattered ibogaine harm reports into one place.
Systematic reviews are how a field finds signals too rare for any single study to catch. This one covers 2015 to 2020 and updates an earlier review, which tells you something in itself: the harm literature has to be periodically re-gathered because it accumulates as isolated case reports rather than as trial safety data.
A companion observational study published in the same month reported on the safety of ibogaine administration during supervised detoxification. Both are worth reading together, because the gap between what happens under monitoring and what happens without it is the whole practical question.
2021-10
Federal University of São Paulo (UNIFESP)
The randomised placebo-controlled trial
A Brazilian Phase II in alcohol use disorder is the only randomised, double-blind, placebo-controlled ibogaine trial in the database.
One trial. Out of nine, across sixty-four years, this is the single study in Blossom’s ibogaine record that randomises participants, blinds them and includes a placebo arm, and it is in alcohol use disorder rather than in the opioid indication the compound is famous for.
It also uses an escalating-dose design, which is a serious attempt to find the point where benefit and cardiac risk trade off rather than assuming a single dose is right for everyone. Run by a public university in São Paulo, completing at the end of 2024.
Set this against the 5-MeO-DMT canvas, where a randomised readout arrived seven years after the first trial and a $2.8bn acquisition followed. Ibogaine has had six decades and one randomised trial. The question that raises is not whether ibogaine works. It is why nobody with money has been willing to find out.
2021-11
Stanford University
Magnesium-ibogaine in veterans
Stanford follows thirty veterans with blast exposure who travelled to a Mexican clinic, and publishes the results in Nature Medicine.
This study did something unusual and rather clever. Rather than trying to obtain approval to dose veterans with ibogaine in the US, the researchers assessed men who were already going to Mexico for treatment, before and after. Magnesium was co-administered specifically to blunt the QT effect, and no cardiac adverse events were reported.
The design is its own limitation, and the authors say so. There is no control group, no blinding and no randomisation, and the participants had selected themselves, paid to travel and expected to improve. Those are exactly the conditions under which self-reported symptom scores move a lot.
It is still the most consequential ibogaine paper of the decade, because it put the compound in Nature Medicine, in a veterans population, at a moment when veterans’ mental health had US political attention. The follow-up imaging and phenomenology papers landed in 2026, and the Texas appropriation two markers down is downstream of this study more than of any other.
2023-02
Nature
Analogues disrupt opioid use without the cardiac signal
A new class of iboga alkaloids reproduces the anti-addiction effect in animals while dropping the hERG activity that makes ibogaine dangerous.
This is the most important scientific development on the canvas, and it is easy to under-read. Chemists rebuilt the molecule to keep the kappa-opioid and neuroplasticity effects while removing the potassium-channel block responsible for the QT prolongation. In animals, the anti-addiction behaviour survives the surgery.
If it holds in humans, ibogaine’s central dilemma dissolves: you would have the effect without the arrhythmia risk, and a patentable molecule that a company can afford to develop properly. Follow-on work through 2026 extends the same approach to alcohol drinking.
The caution is the usual one, and it is not small. Animal models of addiction predict human results badly, and a compound that is safer on paper still has to prove it works. Note also what this does to the ibogaine question itself: if the analogues succeed, the parent compound may never get the trial it has been waiting sixty years for.
A funded future
2025 onwardsTexas money, and analogues built to keep the effect without the risk.2025-01
Gilgamesh Pharma
NIDA funds a cardio-safe analogue
A $14m NIDA grant backs GM-3009 through IND-enabling studies and early trials. Federal money returns to iboga, thirty years on.
The same agency that walked away in the 1990s is now the largest single funder in this space, and it is one of the biggest NIDA awards in psychedelic research. What changed is not the politics of psychedelics so much as the chemistry: GM-3009 is designed to retain the anti-addiction effect and eliminate the QT prolongation.
Non-dilutive public funding at this stage also means the company does not have to raise venture capital against an early safety question, which is precisely the constraint that killed every previous attempt to develop ibogaine.
It is worth stating plainly what this money is not for. It will not tell anyone whether ibogaine itself is safe or effective. It funds a different molecule.
2025-03
Johns Hopkins University
ended earlyJohns Hopkins study withdrawn
The Trifecta research study is withdrawn from the registry before enrolling anyone.
Withdrawn is a specific registry status: registered, then closed without a single participant enrolled. With only nine ibogaine trials in existence, losing one before it starts is a material fraction of the record.
Records like this are quietly dropped from most pipeline charts. Leaving it visible is the honest choice, because registered intent and delivered research are different quantities and the gap between them is larger for ibogaine than for almost any compound in this database.
2025-06
Texas puts state money into ibogaine trials
Texas commits up to $50m of matched state funding for ibogaine clinical trials aimed at FDA approval, the largest public commitment the compound has ever had.
A US state legislature, not a pharmaceutical company and not a federal agency, has become the largest prospective funder of ibogaine research. The appropriation requires matching private money and is framed around opioid use disorder and veterans, which reflects the two constituencies that carried it politically.
This is a genuinely unusual mechanism. Because ibogaine cannot generate a patent-protected return, no ordinary sponsor could justify the several hundred million dollars a full development programme costs. Public money is the only way the parent compound gets tested, and Texas is the first jurisdiction to act on that logic at scale.
The open questions are about execution rather than intent. Who runs the trials, whether the matching capital materialises, whether the cardiac monitoring requirements make the studies affordable, and whether a state-funded programme can survive a change of administration. None of that is answerable yet, and treating the announcement as a result would be a mistake.
2025-09
University of California, San Francisco
UCSF opens a neuroimaging study
A study mapping ibogaine’s neural dynamics in opioid use disorder begins recruiting, running to 2027.
A major US academic centre dosing ibogaine in an opioid-dependent population is new, and the design goes after mechanism rather than outcome. Watching what the compound does to brain activity is how you start to explain why a single session might change behaviour for months.
It is also the only ibogaine trial currently recruiting anywhere in this database. One open study, against thirty-odd clinics treating people continuously.
2025-12
Universal Ibogaine
ended earlyUniversal Ibogaine runs out of runway
A securities regulator issues a cease trade order after the company fails to file audited accounts. Its clinical trial application never gets filed.
The company had done the unglamorous groundwork: a contract research organisation engaged to compile a Health Canada clinical trial application, GMP-grade ibogaine supply secured, a regulatory consultancy appointed. In September 2025 it committed publicly to filing within ninety days.
It did not file. Over the same period it sold a subsidiary property to fund basic operations, lost its chief financial officer, settled executive debt by issuing shares, and saw a board member resign. In December the Alberta Securities Commission issued a management cease trade order over unfiled audited financial statements, and by January 2026 the application was postponed until after a restructuring that has no date.
This is what failure looks like in this field, and it is worth showing because the failure mode is financial rather than scientific. Nothing here says ibogaine does not work. It says a small company on a public market could not afford to find out, which is the same wall Lotsof hit in 1985.
2026-01
DemeRx
Noribogaine clears a Phase 1 on cardiac safety
A double-blind Phase 1 in fifty-five healthy volunteers reports DMX-1001 safe and well tolerated, with QT effects judged not clinically relevant.
Twelve years after the first noribogaine studies, a multiple-ascending-dose Phase 1 across a 20 to 80 mg range reports that the QT effect is not clinically meaningful at therapeutic doses. For a compound in this family that is the milestone that matters most, because the cardiac question is the one that has blocked everything else.
Healthy volunteers are not the population of interest, and a clean Phase 1 says nothing about whether the drug reduces drinking or drug use. A Phase 2 in alcohol use disorder is planned. Until it reports, this is a safety result in search of an efficacy one.
2026-06
Mortality risk quantified across 19,071 patients
A multisite analysis and updated fatality review finally give the safety question a denominator, stratified by indication.
For sixty years the ibogaine safety debate was conducted with a numerator and no denominator. Everyone could name the deaths. Nobody could say how many treatments they came out of, or which patients were most at risk.
This analysis covers 19,071 patients across multiple sites treated under contemporary protocols, and stratifies the mortality risk by indication, alongside an updated systematic review of fatalities. Stratification is the important word: the risk of treating an opioid-dependent patient on methadone with cardiac comorbidity is not the risk of treating a veteran with a brain injury, and averaging them helps nobody.
Two cautions. Clinic-derived datasets carry survivorship and reporting bias, because the sites that agree to be analysed are the ones with the best protocols. And a mortality figure, however careful, does not establish benefit. It tells you what the treatment costs, not what it buys.
A cost-effectiveness model published the same month put ibogaine for opioid use disorder against usual care in one US state, which is the first time anyone has tried to price the trade-off rather than argue it.
2026-09
Gilgamesh Pharma
plannedGM-3009 Phase 1 expected to start
Company guidance points to a Phase 1 opening in the third quarter of 2026, testing the cardiac profile against ibogaine’s.
The first human data on whether the cardiac risk can be engineered out. Everything in the analogues lane rests on that question, and until now it has only been answered in animals.
The date is company guidance rather than a registry-confirmed start, so treat it as the optimistic case. Early-phase trials slip routinely.
2027-05
University of California, San Francisco
plannedUCSF study completes
Registry-listed completion date for the only ibogaine trial currently recruiting.
Ibogaine
Iboga: the research journey