A Phase 1 single ascending dose study of pure oral harmine in healthy volunteers

This was a Phase I, open-label, single ascending dose trial of pharmaceutical-grade harmine hydrochloride (HCl) in healthy adult volunteers. The trial used a continuous reassessment method (CRM) for dose escalation and was designed to allow dosing up to 1,200 mg and enrolment of up to 40 participants, although the study was stopped earlier on medical grounds. Eligibility criteria reported in the extracted text required healthy adults with a normal body mass index (BMI 19–30 kg/m2) and no prior psychiatric illness; age limits are not clearly stated in the extracted Methods text, though other sections indicate adult participants. Participants were randomised to receive a single oral dose of harmine HCl at one of several planned dose levels. In practice, the doses administered were 100 mg, 200 mg, 300 mg and 500 mg. Safety, tolerability and any psychoactive effects were the primary focuses. A team of experienced psychiatric investigators assessed psychoactivity (including hallucinations). Clinical measures included vital signs and laboratory testing; the extracted Results text indicates that descriptive statistics were calculated with SPSS and/or SAS and that plotted measures display mean ± standard error of the mean. A retrospective analysis of adverse events by mg/kg body weight was performed to evaluate a weight-based threshold for dose-limiting toxicity (DLT).

Thirty-four adults were screened and 25 participants were enrolled and dosed under the CRM algorithm. Group sizes reported in the extracted text were: 100 mg (n = 10), 200 mg (n = 10), 300 mg (n = 4) and 500 mg (n = 1). The maximum tolerated dose (MTD) was determined to lie between 100 mg and 200 mg on an absolute-dose basis, but a weight-based analysis identified an apparent cutoff of 2.7 mg/kg. No participant below 2.7 mg/kg experienced a DLT; among participants receiving >2.7 mg/kg the frequency of DLTs was high (the abstract text reports 90% of those >2.7 mg/kg experienced a DLT). At the 100 mg dose all 10 participants tolerated harmine HCl without DLTs; two reported only mild adverse events (dizziness, drowsiness and/or impaired concentration). At 200 mg, 6 of 10 participants experienced DLTs, described as mild to moderate and typically consisting of nausea, vomiting, dizziness and drowsiness. At 300 mg all four participants experienced moderate adverse events and two had DLTs. A single participant received 500 mg and experienced nausea and vomiting. Gastrointestinal adverse events were the predominant DLTs; nausea and vomiting were the most common non‑serious DLTs at doses above the weight-based cutoff. Drowsiness, impaired concentration and dizziness were other frequently reported non‑serious adverse events across doses. Diarrhoea was not reported in any of the 25 participants. Cardiovascular measures (mean systolic and diastolic blood pressure and heart rate) remained in the normal range on average; two participants experienced brief hypotension associated with a vasovagal reaction during procedures (blood draw or vomiting), both transient and not attributed to direct harmine effects. No serious adverse events were identified. Regarding psychoactivity, the investigators report few psychoactive effects overall. No frank hallucinations were observed by the psychiatric assessment team. Some participants at doses >2.7 mg/kg had mild alterations in visual or tactile perception, and many participants across doses reported drowsiness or a sense of being intoxicated, but overt hallucinatory phenomena were not documented. The authors also note that retrospective consideration of body weight showed that participants who experienced DLTs at the 200 mg dose were women with smaller body mass who thus received greater than 2.7 mg/kg. The CRM algorithm had continued recommendations that might have included higher doses, but after observing the observed and predicted risk profile (including an estimated 25% risk of adverse events/DLTs at doses >200 mg), the investigators determined no further participants should receive doses above 200 mg and the study was terminated when 10 participants at the 100 mg and 200 mg dose levels had been enrolled. Statistical reporting in the extracted Results text indicates use of mean, standard deviation, ranges, counts and percentages computed with SPSS and/or SAS, with plotted measures showing mean ± standard error of the mean.

Ables and colleagues interpret these findings to indicate that single oral doses of harmine HCl up to 200 mg — and more precisely doses below a weight‑based threshold of approximately 2.7 mg/kg — are generally well tolerated in healthy adults, with no serious adverse events observed in this study. The primary dose‑limiting toxicities were nausea and vomiting, which the authors suggest may reflect central nervous system and/or intestinal effects possibly related to harmine’s monoamine oxidase (MAO)‑inhibiting or serotonergic properties. Drowsiness, also commonly reported, is considered consistent with MAOI‑class effects in the central nervous system. The authors position their results relative to findings from Ayahuasca studies by noting that harmine alone does not appear to produce the frank hallucinatory effects typically attributed to Ayahuasca at doses comparable to those used in ceremonial or recreational contexts (their data indicate limited perceptual changes but not overt hallucinations at studied doses). They suggest harmine contributes to vomiting associated with Ayahuasca but that diarrhoea observed in Ayahuasca users is likely attributable to other alkaloids in the traditional decoction rather than harmine alone. The authors also discuss the possibility that additive effects among β‑carbolines (harmine, harmaline, tetrahydroharmine) explain reports of hallucinations with larger or combined β‑carboline exposures. Key limitations acknowledged in the extracted text include the small sample size, early termination of dose escalation for medical reasons, and the variability in participants’ body mass within the protocol’s normal BMI range — limitations that motivated the retrospective mg/kg analysis. The authors note that the study assessed only single doses and therefore cannot address multi‑day dosing safety or longer‑term tolerability. They also highlight the absence in this report of detailed pharmacokinetic (absorption, distribution, metabolism, excretion) data and recommend that future studies investigate these parameters and consider alternate routes of administration (oral disintegrating tablet, buccal, intravenous) which may alter gastrointestinal side effects such as vomiting. Implications discussed by the authors include the utility of using weight‑based dosing (mg/kg) rather than an absolute mg/person cutoff in future trials, the need for finer dose increments (suggested 50 mg steps around the 100–300 mg range) to define the MTD more precisely, and the importance of multi‑day and pharmacokinetic studies to fully characterise safety and tolerability before broader clinical development.

The authors conclude that single oral doses of pharmaceutical‑grade harmine HCl can be administered to healthy adults with acceptable tolerability at doses below a weight‑based threshold of approximately 2.7 mg/kg (corresponding to an absolute MTD between 100 mg and 200 mg in this sample). The predominant adverse events at higher doses were nausea and vomiting; no serious cardiovascular or other serious adverse events were observed. Harmine alone produced limited psychoactivity in this study and did not produce frank hallucinations at the doses tested. The authors recommend future work employing weight‑based dosing, finer dose escalation steps, multi‑day safety studies, and formal pharmacokinetic investigations, and they note that route of administration could materially affect tolerability (for example, buccal versus oral).