Effects of 3,4-Methylenedioxymethamphetamine on Conditioned Fear Extinction and Retention in a Crossover Study in Healthy Subjects

Abstract

Background: 3,4-Methylenedioxymethamphetamine (MDMA) has shown initial promise as an adjunct in psychotherapy to treat posttraumatic stress disorder (PTSD). Its efficacy and safety have been demonstrated across phase I-III studies. However, the mechanism underlying the potential utility of MDMA to treat PTSD in humans has not yet been thoroughly investigated. Preliminary evidence suggests that MDMA may facilitate fear extinction recall, which may be through the release of oxytocin. To test this hypothesis, we examined the efficacy of acute MDMA treatment to enhance fear extinction learning and recall. Methods: We used a two-period, double-blind, randomized, placebo-controlled crossover design in 30 healthy male subjects who received a placebo and a single dose of MDMA (125 mg). Fear extinction was tested using two separate Pavlovian fear conditioning paradigms, one using skin conductance response (SCR), and the other fear-potentiated startle (FPS) to conditioned cues. MDMA treatment occurred after fear conditioning and 2 h before extinction learning. Extinction recall was tested 23 h after MDMA intake. Additional outcome measures included subjective effects, emotion recognition tasks, plasma levels of oxytocin, and pharmacokinetics. Results: Fear conditioning and extinction learning were successful in both fear extinction paradigms (generalized eta-squared [ges] for SCR: 0.08; FPS: 0.07). Compared to placebo treatment, MDMA treatment significantly reduced SCRs to the reinforced conditioned stimulus (CS+) during extinction learning (ges = 0.03) and recall (ges = 0.06). Intensity of the subjective effects of MDMA (good effect, trust, and openness) during extinction learning negatively correlated with the discrimination between CS+ and the safety stimulus (CS-) during recall. MDMA did not influence FPS to conditioned cues. Oxytocin concentration was increased fourfold on average by MDMA during acute effects but was not associated with fear extinction outcomes. Conclusions: MDMA treatment facilitated rapid fear extinction and retention of extinction as measured by SCR to fear cues, in line with animal studies of MDMA facilitation of extinction. However, this effect may be limited to certain forms of learned fear responses, as it was not observed in the extinction model using startle reactivity as the outcome. This study provides further evidence for the facilitation of extinction with MDMA treatment and suggests this may be a component of its efficacy when paired with psychotherapy. Clinical Trial registration: clinicaltrials.gov identifier: NCT03527316.

Keywords: MDMA and fear extinction paradigms; fear extinction; fear-potentiated startle; healthy subjects; oxytocin; skin conductance response.

FIGURE 1

Schematic study day. Participants underwent every two sessions with either 125-mg 3,4-methylenedioxymethamphetamine (MDMA) or placebo. The order was randomized but balanced. The washout period between the study days was at least 30 days. Autonomic and subjective effects were assessed throughout the study day. Blood samples for MDMA and oxytocin blood plasma concentrations were collected before and periodically up to 24 h after drug ingestion.

FIGURE 2

MDMA treatment reduces CS+/CS− discrimination during fear extinction learning and enhances extinction recall. Standardized values (z-scores) of both trial types (conditioned fear stimulus [CS+] and conditioned safety stimulus [CS−]) of the skin conductance response task. (A) Subjects showed differentiation between CS+ and CS− in acquisition. (B) In the extinction learning phase, subjects after a placebo displayed discrimination of the conditioned stimuli in the early phase and extinction over time, but already early nondiscrimination of the conditioned stimuli after MDMA. (C) In the recall phase, subjects after the placebo displayed an early return of fear memory by discrimination of the conditioned stimuli. There was no discrimination of the conditioned stimuli after MDMA, indicating extinction retention. The acquisition was before drug intake. MDMA, 3,4-methylenedioxymethamphetamine. Values are mean ± standard error of the mean (SEM). Stats in (A) are main effects, *p < 0.05, ***p < 0.001. Stats in (B) and (C) are from Tukey posthoc tests, ^# p < 0.05, ^## p < 0.01, ^### p < 0.001.

FIGURE 3

MDMA promotes extinction recall. The difference in standardized values (Δz-scores) between the trial types (conditioned fear stimulus [CS+] minus conditioned safety stimulus [CS−]) of the skin conductance response task in the recall. Subjects after MDMA showed significantly less discrimination between CS+ and CS− compared to the placebo session. Stats are from Tukey posthoc tests following an interaction of drug × time in delta values [(F1,22) = 6.74, p = 0.017]. *p < 0.05. MDMA, 3,4-methylenedioxymethamphetamine.

FIGURE 4

No changes induced by MDMA in the fear-potentiated startle task. Difference between the standardized values of both conditioned trial types (conditioned fear stimulus [CS+] and conditioned safety stimulus [CS−]) and the pulse alone (NA) during the fear-potentiated startle task. (A) Subjects showed a clear differentiation between CS+ and CS− in acquisition. (B) In the extinction learning phase, subjects in both sessions displayed discrimination of the conditioned stimuli in the early trial blocks (^#CS+ vs. CS− of block 1 and 2, Tukey posthoc: p < 0.001 and p < 0.01, respectively) and extinction over time (trial blocks). (C) In the recall extinction phase, subjects in both sessions showed a return of fear of the conditioned stimulus in the early trial blocks and renewed extinction over time. The acquisition was before drug intake. MDMA, 3,4-methylenedioxymethamphetamine. Values are mean ± SEM. Stats in (A) are main effects. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 5

MDMA characteristic effects, such as trust and openness, correlate with extinction recall. Correlation matrix between MDMA and oxytocin concentrations and subjective effects of MDMA with extinction recall parameters. Characteristic MDMA effects during extinction learning correlated with the CS+ and subsequently ΔCS in the early phase of extinction recall. Pearson correlation coefficient was used. p-values in white numbers. *p < 0.05; **p < 0.01. Not corrected for multiple testing. CS+, conditioned fear stimulus; CS−, conditioned safety stimulus; ΔCS, CS+–CS−; AMRS, Adjective Mood Rating Scale; VAS, visual analog scale; FPS, fear-potentiated startle; SCR, skin conductance response; MDMA, 3,4-methylenedioxymethamphetamine. Concentrations at 2.5 h were generated as the mean of the respective 2 and 3 h time points. Data is from MDMA sessions only.

FIGURE 6

MDMA inhibits the recognition of negative emotion and promotes misclassification as happy. Results of the facial emotion recognition task. (A) Correctly identified emotions. (B) Emotion misclassification if not correctly recognized. Stats are from paired t-tests. *p < 0.05; **p < 0.01. Bonferroni corrected for multiple testing. MDMA, 3,4-methylenedioxymethamphetamine. Values are mean ± standard error of the mean. The task was performed 3 h after drug intake.