Overcoming the clinical challenges of traditional ayahuasca: a first-in-human trial exploring novel routes of administration of N,N-Dimethyltryptamine and harmine

Affective disorders remain common and disabling, and current antidepressant treatments have limited efficacy, slow onset and frequent side effects. Researchers have renewed interest in psychedelics as a source of more rapid-acting therapies; among these, ayahuasca, a traditional Amazonian decoction containing N,N-dimethyltryptamine (DMT) together with β-carbolines such as harmine, has shown rapid and sustained antidepressant effects in prior clinical work. However, the botanical preparation of ayahuasca has pharmacokinetic and tolerability shortcomings: long and variable absorption, first-pass metabolism, batch-to-batch variability in alkaloid content, and common gastrointestinal adverse effects (nausea, vomiting, diarrhoea) that complicate standardised clinical use. Dornbierer and colleagues set out to improve the pharmacokinetic/pharmacodynamic (PKPD) profile and tolerability of ayahuasca-analogue formulations by testing purified DMT and harmine delivered by alternative routes that bypass the gastrointestinal tract. The trial compared peroral capsules containing DMT hemifumarate plus oral extended-/immediate-release harmine with a parenteral-oriented approach combining buccal (oromucosal) harmine tablets and an intranasal DMT spray. The main aims were to assess PKPD characteristics, subjective effects, vital signs and adverse events in a first-in-human, open-label, within-subject dose-finding pilot study in healthy volunteers, to determine whether these formulations reduce GI side effects and improve predictability of response relative to traditional ayahuasca.

The study used a non-randomised, open-label, within-subject design in which ten healthy male volunteers (mean age 30.7 ± 5.4 years, BMI 18.5–25) attended four study days with at least a 1-week washout between sessions. Inclusion criteria excluded current or past somatic, neurological or psychiatric disorders, family history of Axis-I disorders, regular medication, and significant lifetime drug use; all participants gave written informed consent and the protocol was approved by local ethics and regulatory bodies. Two administration strategies were compared. On study days 1 and 2 both DMT hemifumarate and harmine HCl were given perorally: harmine total 250 mg delivered as a mix of immediate-release (IR) and extended-release (ER) minitablets (150 mg IR + 100 mg ER) 30 minutes before DMT; DMT capsules contained 30 mg (day 1) or 50 mg (day 2) DMT hemifumarate. On study days 3 and 4 harmine was delivered buccally as orodispersible freeze-dried tablets (ODTs) — 150 mg on day 3 and 200 mg on day 4 — 30 minutes prior to an intranasal DMT hemifumarate spray. The intranasal DMT solution was prepared at 5 mg DMT hemifumarate per puff; day 3 allowed up to 50 mg total (increments of 5 mg), day 4 up to 90 mg total (increments of 5 or 10 mg). The intranasal regimen was repeated-intermittent with fixed small doses at early timepoints and participant self-determined dose increments at later timepoints to enhance safety and allow individual control of intensity. Pharmacokinetic sampling collected venous blood at -30, -15, 0 and 360 minutes after the first DMT administration; samples were centrifuged, plasma frozen and stored at -80 °C. Analyses quantified plasma DMT, harmine and metabolites indole-3-acetic acid (3-IAA) and N-methyltryptamine (NMT) using UHPLC coupled to tandem mass spectrometry. Calibration ranges and internal standards were reported. Continuous monitoring of vital signs occurred at prespecified intervals, and adverse effects were assessed by physician observation and questionnaire-based visual analogue scales (VAS 0-100) for intensity, liking and arousal at multiple timepoints up to 300–360 minutes. Liver enzymes were measured at baseline and after sessions.

All ten participants completed all four study days. Plasma samples were analysable from 9 participants on days 1 and 2, 8 participants on day 3 and all 10 on day 4; missing samples were due to sample quality issues. For DMT, mean peak plasma concentrations (Cmax) were reported as 27.9 ± 15.5 ng/mL (day 1, 30 mg oral), 23.6 ± 11.9 ng/mL (day 2, 50 mg oral), 23.9 ± 5.0 ng/mL (day 3, intranasal up to 50 mg with 150 mg buccal harmine) and 33.1 ± 9.3 ng/mL (day 4, intranasal up to 90 mg with 200 mg buccal harmine). Harmine peak concentrations were 125.5 ± 111.2 ng/mL (day 1), 104.8 ± 100.5 ng/mL (day 2), 93.9 ± 33.5 ng/mL (day 3) and 132.4 ± 111.6 ng/mL (day 4). The DMT metabolite 3-IAA showed higher concentrations on day 4, when the largest DMT/harmine doses were used; NMT tracked DMT kinetics but was much lower on day 3 than on days 2 and 4 despite similar DMT peaks on days 2 and 3. Marked interindividual variability was observed with oral dosing: the authors report that Cmax varied by a factor of ~6.8 for DMT and ~55.5 for harmine across subjects after oral administration, and harmine tmax ranged widely (60–270 minutes). Parenteral-oriented dosing (buccal harmine + intranasal DMT) produced more homogeneous PK profiles, with reported variability factors of ~2.3 for DMT Cmax and ~3.2 for harmine. Subjective effects measured by VAS increased robustly with all four dosing conditions. Peak intensity (mean ± SD as percent of scale) was 69.0% ± 26.3% on day 1, 57.2% ± 39.9% on day 2, 72.0% ± 21.5% on day 3 and 79.0% ± 20.8% on day 4. Peak liking was 86.0% ± 10.6% (day 1), 67.1% ± 34.5% (day 2), 92.0% ± 10.3% (day 3) and 86.0% ± 13.5% (day 4). Peak arousal values were 56.1% ± 34.9% (day 1), 55.7% ± 33.1% (day 2), 73.0% ± 27.1% (day 3) and 63.4% ± 27.0% (day 4). Subjective effects generally tracked plasma concentrations more closely on days 3 and 4. The coefficient of variation for maximum intensity ratings decreased from day 1 to day 4 (day 1: 37.6; day 2: 51.7; day 3: 28.6; day 4: 23.2), indicating reduced between-subject variability with the parenteral-oriented regimen. Adverse events were mostly low to moderate and transient. Oral dosing (days 1–2) produced more somatic discomfort and nausea during the first two hours after dosing; no participant vomited and no diarrhoea was reported. The buccal/intranasal regimen (days 3–4) showed fewer reported adverse effects and was described as better tolerated. Vital signs showed mild cardiovascular effects: heart rate generally remained 60–80 bpm, systolic blood pressure rose on average by 15 mmHg with peak increases at 30 minutes (days 1–3) and 120 minutes (day 4), diastolic pressure increased by about 10 mmHg, and body temperature changes were ≤1 °C with no readings above 37.8 °C. Liver enzyme measurements before and after the largest dosing session (day 4) showed no clinically relevant changes.

Dornbierer and colleagues interpret their findings as evidence that purified DMT/harmine formulations can reduce the gastrointestinal adverse effects and interindividual unpredictability associated with traditional botanical ayahuasca. The parenteral-oriented combination of buccal harmine and intranasal DMT produced more consistent plasma concentrations and lower between-subject variability in subjective intensity compared with oral dosing, while also permitting participant-controlled, repeated-intermittent intranasal dosing to titrate subjective strength. The study team suggests that bypassing the intestinal MAO-A–rich environment and avoiding emetogenic plant constituents likely underlies the attenuation of nausea and other GI side effects. The authors position these results relative to prior work by noting that oral ayahuasca shows variable alkaloid content and unpredictable pharmacokinetics, whereas short-acting parenteral DMT plus a localized MAO-A inhibitor may offer a dose-sparing, more controllable approach. They propose mechanistic explanations for observed variability after oral dosing, including interindividual differences in intestinal MAO-A activity and in hepatic CYP2D6-mediated harmine metabolism, which could lead to widely differing harmine and DMT bioavailability. Key limitations noted by the investigators include the small sample size, restriction to healthy male participants, the open-label non-randomised within-subject design and the possibility of expectation and habituation effects across sessions. They also acknowledge that intranasal DMT without harmine was not tested, so the independent effects of intranasal DMT hemifumarate remain unresolved. The dosing regimen involved repeated small increments rather than single bolus doses, and the analyses were descriptive rather than inferential; thus the authors call for placebo-controlled, double-blind, randomised crossover trials to validate these preliminary findings. Finally, the study team suggests clinical implications cautiously: the parenteral-oriented DMT/harmine kit may offer a safer, more predictable, and patient-oriented means of delivering rapid-acting psychedelic interventions, but further controlled studies are required before clinical adoption or regulatory consideration.