Pharmacogenetics of ecstasy: CYP1A2, CYP2C19, and CYP2B6 polymorphisms moderate pharmacokinetics of MDMA in healthy subjects

MDMA (3,4-methylenedioxymethamphetamine; ecstasy) produces prosocial and empathic effects and is used recreationally and experimentally in psychotherapy, yet its use can cause severe toxicity including agitation, hypertension and hyperthermia. Earlier work established that CYP2D6 is primarily responsible for O-demethylenation of MDMA to HHMA and subsequent metabolism, while in vitro data also implicate CYP1A2, CYP2B6 and CYP2C19 in the N-demethylation of MDMA to the minor but active metabolite MDA. The extent to which genetic polymorphisms in CYP1A2, CYP2B6 and CYP2C19 influence MDMA pharmacokinetics and pharmacodynamics in humans remained unknown. Vizeli and colleagues designed the present study to test whether common genetic variants in CYP2C19, CYP2B6 and CYP1A2 alter conversion of MDMA to MDA and whether these genotypes affect MDMA’s physiological or subjective effects. Because CYP1A2 is inducible by tobacco smoking in carriers of the rs762551 A/A genotype, the investigators also examined interactions between smoking status and CYP1A2 genotype to evaluate CYP1A2’s contribution to MDMA N-demethylation in humans.

The study is a prospectively planned pooled analysis of eight double-blind, placebo-controlled, crossover studies in healthy subjects. Across those studies 142 European/Caucasian participants aged 18–45 years were recruited; after missing genotyping or incomplete concentration–time data the reported analyses included 139 participants (69 male, 70 female; mean age 24.9 ± 4.1 years). Seven studies used a 125 mg oral MDMA dose and one study used 75 mg; in total 110 subjects received 125 mg (mean 1.9 ± 0.3 mg/kg) and 29 received 75 mg (mean 1.1 ± 0.1 mg/kg). Washout periods between treatment periods were at least 7 days. Exclusion criteria included psychiatric or physical illness, substantial prior illicit drug use, recent illicit drug use, and concomitant use of drugs known to interact with CYP function. Heavy smokers (>10 cigarettes/day) were excluded, but very light (1–5/day) and light smokers (6–10/day) were included. MDMA (± MDMA hydrochloride) was given as a single oral dose of 75 or 125 mg. Blood samples for pharmacokinetic (PK) analysis were taken at 0, 0.33, 0.67, 1, 1.5, 2, 3, 4 and 6 h post-dose, with plasma stored at −20°C. Plasma concentrations of MDMA, MDA and HMMA were measured; HMMA was assessed after enzymatic deglucuronidation in 76 subjects. The lower limit of quantification for analytes was 1 ng/ml. Pharmacodynamic measures included repeated assessments of blood pressure, heart rate, tympanic core temperature and subjective effects on visual analogue scales; the rate pressure product (RPP; systolic blood pressure × heart rate) was used as an index of cardiostimulant effect. Genomic DNA was extracted from whole blood and genotyping performed via commercial TaqMan SNP assays to determine variants in CYP2C19, CYP2B6 (rs3745274) and CYP1A2 (rs762551). CYP2D6 activity was assessed phenotypically by the dextromethorphan/dextrorphan ratio and sensitivity analyses were performed restricted to CYP2D6 extensive metabolisers (EMs) and after excluding intermediate (IM) and poor metabolisers (PMs) to reduce confounding. Peak plasma concentrations (Cmax) were taken from observed data and area under the curve from 0–6 h (AUC6) was calculated by the linear trapezoidal method. Plasma values were dose-normalised to the mean dose per body weight (1.7 mg/kg) and mg/kg dose was included as a covariate in pharmacodynamic analyses. Statistical analyses used one-way ANOVA with genotype as the between-subjects factor, followed by Tukey post hoc tests. Smoking status was included as a factor in analyses involving CYP1A2. Sensitivity analyses confined to the 125 mg dose group were conducted to exclude dose-level confounding.

The analysed sample comprised 139 participants (69 male, 70 female), with MDMA plasma sampling up to 6 h. HMMA was assayed in 76 subjects. Dose groups differed as expected: Cmax of MDMA was greater after 125 mg versus 75 mg (mean ± SD: 230 ± 46 vs 125 ± 29 ng/ml; F1,137 = 140.20, P < 0.001). The 125 mg dose also produced larger peak subjective effects (80 ± 23 vs 57 ± 30%; F1,137 = 21.5, P < 0.001) and greater peak cardiostimulant responses measured by RPP (14728 ± 3278 vs 12067 ± 3159 mmHg × bpm; F1,137 = 15.3, P < 0.001). After dose normalisation, subjective and cardiovascular effects did not differ between dose groups, although dose-normalised MDMA Cmax showed a trend toward being higher at 125 mg (F1,137 = 4.08, P = 0.05), suggesting near-nonlinear kinetics across the studied dose range. CYP2C19: Genotype influenced MDMA→MDA conversion. The CYP2C19 genotype significantly affected MDA AUC6 (F3,135 = 3.54, P < 0.05) and the MDMA/MDA AUC6 ratio (F3,135 = 5.55, P < 0.01). Two CYP2C19 poor metabolisers showed more rapid increases in MDMA plasma levels, but that particular effect did not reach statistical significance. Pharmacodynamically, CYP2C19 genotype altered Emax of the RPP (F3,134 = 2.92, P < 0.05), with higher RPP values in CYP2C19 PMs compared with IMs and UMs (both P < 0.05). There were no CYP2C19 effects on HMMA concentrations or on body temperature and no effects on subjective measures of MDMA. CYP2B6: The rs3745274 SNP (G/G vs G/T vs T/T) significantly affected MDMA Cmax (F2,136 = 3.72, P < 0.05) with higher Cmax in T/T versus G/G carriers (P < 0.05). CYP2B6 genotype also influenced the MDMA/MDA AUC6 ratio (F3,136 = 3.67, P < 0.05), showing higher ratios in T/T compared with G/T or G/G groups (both P < 0.05). No significant effects of CYP2B6 genotype were observed on plasma MDA or HMMA levels. Autonomic and subjective responses to MDMA were not altered by CYP2B6 genotype. CYP1A2 and smoking interaction: There was a significant interaction between smoking status and CYP1A2 rs762551 genotype (inducible A/A vs A/C and C/C) on MDA Cmax and MDA AUC6 (F5,133 = 5.56, P < 0.001 and F = 4.04, P < 0.01, respectively). Among carriers of the inducible A/A genotype, light smokers (6–10 cigarettes/day) had higher MDA formation than nonsmokers and very light smokers (1–5/day), both P < 0.001. No smoking-related differences in MDA were found in A/C or C/C genotype carriers. CYP1A2 genotype, smoking, or their interaction did not affect MDMA or HMMA plasma concentrations or the pharmacodynamic autonomic and subjective outcomes. Sensitivity analyses accounting for CYP2D6 status were performed: analyses were replicated in 111 subjects phenotyped as CYP2D6 EMs and after exclusion of 19 IMs and 9 PMs, to address potential confounding by CYP2D6 activity. The extracted text does not provide detailed numeric results of these sensitivity analyses beyond stating they were conducted.

The investigators conclude that genetic polymorphisms in CYP2C19 and CYP2B6 contribute to the N-demethylation of MDMA to MDA in humans, corroborating prior in vitro findings. Evidence came from higher MDMA/MDA AUC6 ratios in subjects with lower CYP2C19 or CYP2B6 function, and from genotype-associated differences in MDMA Cmax for CYP2B6. Additionally, an interaction between smoking and the inducible CYP1A2 rs762551 A/A genotype led to increased MDA formation in smokers with that genotype, indicating a contribution of CYP1A2 to MDMA N-demethylation under induced conditions. Although MDA is pharmacologically active in vitro and in animal models, the authors note that altered conversion to MDA did not correspond to changes in subjective MDMA effects in this dataset. Paradoxically, subjects with slower MDMA→MDA conversion (CYP2C19 PMs) showed greater cardiostimulant responses, leading the authors to suggest that MDMA itself may contribute more to the cardiostimulant effects than MDA. The authors caution that some findings require confirmation: only two CYP2C19 PMs were present, and the greater RPP in these two subjects may be a chance finding. Likewise, the CYP1A2×smoking interaction was based on only four light smokers with the inducible genotype and therefore needs replication in larger samples, ideally including heavier smokers who would be expected to show greater CYP1A2 induction. Additional limitations acknowledged include the absence of subjects with rare combinations of impaired enzymes (for example concurrent CYP2D6 PM with CYP2C19 PM and CYP2B6 T/T), which could magnify pharmacokinetic consequences; plasma sampling was limited to 6 h, though the authors argue this covers the relevant pharmacodynamic window; and only doses up to 125 mg were tested. The authors propose that combinations of impaired CYP function could pose greater risk for MDMA toxicity and that larger studies are needed to confirm and extend these pharmacogenetic observations.