Influence of CYP2D6 activity on the pharmacokinetics and pharmacodynamics of a single 20 mg dose of ibogaine in healthy volunteers

Ibogaine is a naturally occurring psychoactive alkaloid from Tabernanthe iboga that has historical use at low doses for fatigue and hunger and at higher doses for ritual hallucinations. Interest in ibogaine resurfaced following anecdotal and nonmedical use suggesting single high doses might reduce opioid withdrawal and craving; mechanistic work has identified noribogaine, a demethylated long‑acting metabolite, as an active species and implicated CYP2D6 in ibogaine demethylation. Published human pharmacokinetic data are limited and heterogeneous, with few detailed plasma or safety data in healthy volunteers and most clinical reports focused on opioid‑dependent patients rather than controlled physiological assessments. Glue and colleagues set out to characterise how CYP2D6 activity influences the pharmacokinetics and pharmacodynamics of a single low (20 mg) oral dose of ibogaine in healthy volunteers. The study aimed to compare plasma ibogaine and noribogaine time‑courses, relate pharmacokinetics to CYP2D6 phenotype assessed by urinary dextromethorphan:dextrorphan metabolic ratios, and to record basic pharmacodynamic effects (pupillary response and self‑rated symptoms) and safety/tolerability at this low dose.

This was a double‑blind, placebo‑controlled phenotyping study in healthy male or sterilised female volunteers aged 20–40 years. Twenty‑one participants who were drug‑ and medication‑free were enrolled. No genotyping for CYP2D6 was performed; instead, phenotype was assessed using urinary dextromethorphan (DM) to dextrorphan (DX) metabolic ratios from 5‑hour urine collections after oral DM dosing. On day 1 participants received 30 mg dextromethorphan for baseline phenotyping. Subjects were randomised to receive paroxetine (10 mg days 2–3, 20 mg/day days 4–15) or placebo in a double‑blind fashion. A second dextromethorphan phenotyping was performed on day 7. On day 8 all participants received a single oral 20 mg dose of ibogaine after an overnight fast. Plasma samples were collected pre‑dose and at 0.5, 1.0, 1.5, 2, 3, 4, 6, 8, 12, 24, 48, 72, 96, 120, 144, and 168 hours post dose; samples were protected from light, centrifuged and stored at −70°C. Pupillary miosis was measured by dark‑adapted pupillometry pre‑dose and at several post‑dose time points up to 168 hours. Subjective effects (sleepy, energetic, nausea, anxious, calm) were recorded using 100 mm visual analogue scales at pre‑specified times up to 168 hours. Safety monitoring included clinical assessments, vital signs and adverse event recording. Plasma ibogaine and noribogaine were quantified by validated LC‑MS/MS assays; within‑ and between‑day precision and accuracy were reported to be acceptable and assay performance was not affected by paroxetine. DM and DX in urine were also measured by LC‑MS/MS after enzymatic hydrolysis. Pharmacokinetic parameters were calculated by noncompartmental methods: observed Cmax and Tmax, terminal elimination rate constant (Kel) from log‑linear plots, t1/2 = 0.693/Kel, and AUC0–t by trapezoidal rule. Active moiety was defined as the molar sum of ibogaine and noribogaine. Subjects with DM:DX MR >0.3 were classified as phenotypic poor metabolizers (PMs). Summary statistics were reported by pretreatment group and relationships were examined by regression methods.

All 21 enrolled healthy male volunteers completed the protocol; mean age was 23.5 years and mean BMI 24.5 kg/m2. One subject (subject 1) was phenotypically a poor metabolizer at baseline (day 1 MR = 0.51) and his ibogaine/noribogaine data were excluded from some group analyses. Baseline (day 1) DM:DX metabolic ratios were similar between groups. By day 7, paroxetine pretreatment produced an approximately 11‑fold increase in mean DM:DX MR (mean 0.160) versus no change in the placebo group (mean 0.018), demonstrating substantial CYP2D6 inhibition (P < .05). Plasma pharmacokinetics (excluding subject 1): in placebo‑pretreated participants ibogaine concentrations were generally barely detectable whereas substantial noribogaine exposures were observed. Paroxetine pretreatment markedly altered ibogaine exposure: mean ibogaine Cmax in the paroxetine group was 26‑fold higher than in the placebo group; mean ibogaine AUC0–t was 66‑fold greater and mean ibogaine t1/2 was longer (10.2 vs 2.5 hours); all comparisons P < .05. Noribogaine AUC0–t values were similar between groups, but noribogaine Cmax tended to be lower in the paroxetine group (12.7 vs 18.7 ng/mL, P = .05) and noribogaine t1/2 was longer (20.1 vs 13.0 hours, P = .07). Mean active moiety (molar sum) AUC0–t was nearly double in paroxetine‑pretreated subjects compared with placebo (P = .005). Correlations between log day 7 DM:DX MR and ibogaine AUC0–t and Cmax were strong (Pearson r = 0.82 and 0.77, respectively); correlations with noribogaine parameters were weaker. Pharmacodynamics: pupillometry showed mean maximal reductions in dark‑adapted pupil diameter of 0.4–0.6 mm at 4 hours post dose in both groups with return toward baseline by 12 hours. Repeated measures ANOVA showed a significant effect of time and a treatment × time interaction, but no main effect of treatment; there were no correlations between ibogaine, noribogaine, or active moiety exposure and change in pupil diameter. Visual analogue scale ratings showed modest trends toward decreased sleepiness and increased energy over 24 hours, with no difference between pretreatment groups; absence of a placebo‑ibogaine arm limits interpretation of these subjective trends. Safety: there were no clinically important changes in vital signs. Sixteen participants reported 25 adverse events in total; 18 events occurred prior to ibogaine dosing and 7 after dosing. Most events were mild and resolved without intervention. Adverse events were reported more frequently in the paroxetine group (17 events) than placebo (8 events), with the most common symptoms being nausea, gastrointestinal disturbance and dizziness. No hallucinations or perceptual changes were reported.

Glue and colleagues interpret their findings as confirmation that CYP2D6 plays an important role in the demethylation of ibogaine to noribogaine in humans and that reduced CYP2D6 activity substantially increases systemic exposure to ibogaine. In this controlled setting a single 20 mg dose of ibogaine was generally safe and well tolerated in healthy male volunteers. Paroxetine pretreatment produced a robust increase in the DM:DX metabolic ratio consistent with CYP2D6 inhibition; this was accompanied by large increases in ibogaine Cmax and AUC0–t, while noribogaine AUC0–t remained similar across groups, resulting in an almost twofold increase in active moiety exposure in the paroxetine group. The authors note that the lack of correlation between CYP2D6 phenotype and noribogaine PK is consistent with noribogaine clearance being dominated by phase 2 reactions (glucuronidation) rather than CYP2D6. They contrast their results with previous reports that used genotyped CYP2D6 poor metabolizers and different matrices, doses and assays, and caution that direct comparisons are complicated by those methodological differences. Individual data from a baseline phenotypic PM (subject 1) illustrated much higher ibogaine exposure relative to noribogaine, supporting the potential clinical relevance of CYP2D6 status. Several limitations are acknowledged. Only two of 11 paroxetine‑treated subjects reached phenotypic PM status by the study definition, so the observed differences may underestimate those between true genotypic PMs and extensive metabolizers. Subjects were not genotyped for CYP3A or CYP2C19, enzymes that may contribute to demethylation, and polymorphisms in these enzymes could have affected results. The single low 20 mg dose used for safety and detectability limits generalisability to higher therapeutic doses that have been used in opioid dependence settings. The absence of a placebo‑ibogaine arm prevents definitive attribution of modest VAS changes to drug effects rather than time‑of‑day. Pupillometry findings differed from those reported in a larger noribogaine study, and the authors suggest further work is needed to clarify ibogaine and noribogaine µ‑opioid pharmacodynamics. On the basis of the roughly doubled active‑moiety exposure observed with reduced CYP2D6 activity, the investigators conclude that it may be prudent to characterise CYP2D6 genotype or phenotype prior to ibogaine use and to consider halving intended doses in CYP2D6 poor metabolizers. They further recommend prospective study of combined reductions in CYP2D6, CYP3A and CYP2C19 activity given the uncertainty about their combined impact on ibogaine exposure.