Oxytocin receptor gene variation predicts subjective responses to MDMA

MDMA (3,4-methylenedioxymethamphetamine) produces distinct prosocial effects such as increased empathy and sociability, which distinguish it from other stimulant drugs and contribute both to its recreational use and its potential as an adjunct to psychotherapy. Previous research indicates that MDMA increases oxytocin release via actions in hypothalamic nuclei, and that oxytocin administration in humans can produce prosocial effects; however, some studies have questioned whether peripheral oxytocin changes fully account for MDMA's empathogenic actions. Neuroimaging and behavioural work has also implicated altered processing of emotional faces and reward-related circuitry following MDMA, but the specific role of oxytocinergic signalling in these effects remains uncertain. Bershad and colleagues set out to test whether common variation in the oxytocin receptor gene (OXTR), specifically the rs53576 single nucleotide polymorphism (A/G), influences subjective responses to MDMA. Prior studies have associated the G/G genotype with higher empathy and sociability and greater sensitivity to intranasal oxytocin, whereas A-allele carriers show relative deficits in social processing. The investigators hypothesised that individuals homozygous for A (A/A) would show reduced prosocial responses to MDMA—particularly sociability—compared with G-allele carriers, when tested under controlled, double-blind conditions using two active oral MDMA doses and placebo.

The study combined data from two within-subject, double-blind, placebo-controlled experiments in occasional MDMA users. Sixty-eight healthy adults (39 male, 29 female), aged 18–35, who reported between 4 and 40 lifetime uses of MDMA, completed three laboratory sessions separated by at least 5 days. At each session participants received capsules containing placebo, 0.75 mg/kg MDMA, or 1.5 mg/kg MDMA (maximum single dose 125 mg); dosing order was counterbalanced under double-blind conditions. Sessions ran from 09:00 to 14:00, with baseline measures prior to capsule ingestion at 09:30 and repeated subjective and cardiovascular assessments at regular intervals for 5 hours. Saliva samples for genotyping were collected at the first session. Subjective effects were measured with visual analogue scales (1–100) for adjectives used in prior MDMA studies (for example “sociable,” “anxious,” “loving,” “playful”) and a Drug Effects Questionnaire (DEQ) assessing ‘‘feel high’’, ‘‘like drug’’, and related items. Cardiovascular measures (heart rate, systolic and diastolic blood pressure) were recorded every 30 minutes. Female participants not using hormonal contraceptives were tested in the follicular phase to limit hormonal variability. Compliance with substance restrictions was verified by breath and urine testing; pregnancy tests were performed each session. Genotyping for OXTR rs53576 (A/G) used the ABI TaqMan assay. For analysis, area under the curve (AUC) measures were calculated for each subjective VAS across the session using the trapezoidal method; change-from-placebo scores were computed for each active dose. Principal components analysis (PCA, promax rotation, eigenvalue = 1) was used to reduce the subjective measures into factors, iteratively excluding items with low or cross-loadings until four factors emerged (sociability, anxiety, euphoria, dizziness). Final factor scores were the mean AUC of items loading on each factor. Participants were grouped by rs53576 genotype (G/G, A/G, A/A) and factor scores were analysed with two-way repeated-measures ANOVAs (within-subject factor: dose change-from-placebo for 0.75 and 1.5 mg/kg; between-subject factor: genotype). Significant interactions were followed with Bonferroni-corrected post hoc tests. Analyses of cardiovascular effects used the same strategy. Statistical significance was set at p < 0.05.

Genotype distribution among the 68 participants was 28 G/G, 30 A/G, and 10 A/A; this conformed to Hardy–Weinberg equilibrium. Groups did not differ on age, sex distribution, education, BMI, or MDMA use history (mean prior MDMA uses = 14, SD = 11). Baseline (pre-capsule) averaged measures of euphoria, anxiety, sociability, and dizziness did not differ across genotypes. Both MDMA doses produced robust main effects on the derived subjective factors. Compared with placebo, MDMA increased sociability (dose F(2, 66) = 8.10, p < 0.001), anxiety (F(2, 66) = 28.84, p < 0.001), euphoria (F(2, 66) = 80.21, p < 0.001), and dizziness (F(2, 66) = 21.34, p < 0.001). Cardiovascular measures also increased: heart rate (F(2, 66) = 52.0, p < 0.001), systolic blood pressure (F(2, 66) = 51.4, p < 0.001), and diastolic blood pressure (F(2, 66) = 26.9, p < 0.001). A Dose × Genotype interaction was observed for sociability (F(2, 65) = 2.81, p = 0.03). Specifically, individuals with the A/A genotype did not report a significant increase in sociability at the higher 1.5 mg/kg dose, whereas carriers of the G allele showed increased sociability at that dose. There was a trend for a Dose × Genotype interaction on euphoria that did not reach conventional significance (F(2, 66) = 2.20, p = 0.07); A/A participants reported smaller increases in euphoria at 1.5 mg/kg compared with G-allele carriers. No genotype-dependent differences emerged for MDMA-induced anxiety, dizziness, or cardiovascular responses. Drug identification data showed increasing correct identification with dose: 55% correctly identified placebo, 59% the 0.75 mg/kg session, and 76% the 1.5 mg/kg session.

Bershad and colleagues interpret their findings as evidence that variation at OXTR rs53576 moderates some subjective effects of MDMA in healthy adults. They report that, while MDMA produced expected increases in sociability, euphoria, anxiety, dizziness, and cardiovascular measures across the sample, A/A homozygotes showed attenuated increases in sociability and, to a lesser extent, euphoria at the 1.5 mg/kg dose relative to G-allele carriers. Because prior work links the rs53576 G allele with greater social sensitivity and responsiveness to exogenous oxytocin, the authors view these genotype-dependent differences as consistent with an oxytocin-mediated component of MDMA's prosocial effects. The investigators situate their results within earlier literature showing dose-dependent increases in plasma oxytocin after MDMA and studies in animals where oxytocin receptor antagonists reduce MDMA’s prosocial actions. They note that baseline self-reported sociability did not differ across genotypes in their sample, suggesting that the genotype effect emerged specifically in response to an exogenous prosocial challenge at a higher dose rather than reflecting baseline trait differences detectable by the measures used. Several limitations acknowledged by the authors temper the conclusions. Sample size was modest, and the A/A genotype group was the smallest (N = 10), raising the possibility of limited power or instability of effects in the rare homozygote subgroup. The authors caution that use of summary measures (AUC, PCA-derived factor scores) may increase type I error risk and that they did not test genotype-by-timecourse interactions because they focused on session-wide summary scores. The sample was also restricted to healthy, recreational MDMA users with a narrow use-history range (4–40 lifetime uses), limiting generalisability to heavier users or clinical populations. In terms of implications, the authors propose that OXTR variation could contribute to inter-individual differences in both recreational responses to MDMA and its potential therapeutic efficacy when used adjunctively in psychotherapy. They suggest future work should replicate these findings in larger, prospectively genotyped samples, examine timecourse effects, include broader MDMA use histories, and assess whether genotype predicts clinical outcomes in MDMA-assisted therapies (for example for PTSD).