We employed a randomized, placebo-controlled, double-blind methodology for this exploratory phase 2 single-site study conducted from February 2014 through April 2017. The authors assert that all procedures contributing to this work comply with ethical standards of relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008. The study was approved by Los Angeles BioMedical Research Institute IRB and conducted in accordance with Good Clinical Practice. Twelve participants were enrolled and randomized to receive MDMA (n = 8) or inactive placebo (n = 4). MDMA was synthesized by David Nichols at Purdue University, compounded with lactose, and placed into gelatin capsules by a research pharmacist. The inactive placebo, lactose, was filled in equivalent weight in identical capsules. After three 60- to 90-min non-drug preparatory psychotherapy sessions, participants received two blinded experimental sessions with MDMA or placebo, spaced approximately 1 month apart. Following each experimental session, three 60- to 90-min non-drug integrative psychotherapy sessions occurred over 3 weeks. The blind was broken at 6-month follow-up; participants who received placebo in the first treatment phase returned for two optional open-label treatment sessions with MDMA (data not presented).
A dose-finding study design was selected in response to anecdotal data, suggesting that hyper-reactivity to sensory stimulation and emotion regulation challenges associated with autism might indicate the need for a lower, yet therapeutically active, MDMA dose range. Among participants receiving MDMA, the first subgroup (N = 4) received 75mg MDMA at the first session and 100-mg MDMA at the second session. The second subgroup (N = 4) received 100mg MDMA at the first session and 125 mg at the second session. All doses were tolerated well; no participants declined the option to escalate the dose for the second session.
An independent rater (IR) administered the Leibowitz Social Anxiety Scale (LSAS) (Liebowitz et al. 1985) at baseline, 1 day, 2 weeks, and 4 weeks after each experimental session and re-administered it before the blind was broken at 6 months. There was one LSAS IR for the entire study to minimize variance. The primary outcome was change from baseline to 1-month post second experimental session in LSAS total scores. At monthly intervals, between the 1-month post-treatment psychotherapy session and the 6-month follow-up visit, participants completed the Beck Depression Inventory (BDI-II) (Beck et al. 1996), Spielberger State-Trait Inventory (STAI Form Y-2) (Spielberger et al. 1983), and Perceived Stress Scale (PSS) (Cohen et al. 1983) through an electronic Patient Reported Outcome (ePRO) system (Medrio, CA, USA).
Screening, eligibility, and participants
Participants were recruited through Internet advertisements, word of mouth, and clinician referrals. No participants in this study were under conservatorship; all signed an informed consent after review with investigators. Eligibility was established through clinical interview and administration of diagnostic instruments, including the Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders- Fourth Edition Axis I Research Version (SCID-I-RV) (First et al. 2002), the Columbia Suicide Severity Rating Scale (C-SSRS) (Posner et al. 2011; Posner et al. 2007), LSAS (Liebowitz et al. 1985), and the Autism Diagnostic Observation Schedule (ADOS-2 Module 4) (Bastiaansen et al. 2011). To be eligible, a global LSAS score of 60 or higher, indicating marked to severe fear and avoidance of specific social situations, was required. Participants were 21 or older, MDMA naïve by self-report, physically healthy, and psychologically stable (see Supplemental for inclusion/exclusion criteria).
Preparatory and integrative psychotherapy
All participants received three preparatory psychotherapy sessions, during which past or current salient issues in the participant’s life were discussed. These sessions focused on establishing rapport between the participant and treatment team. In two instances, an additional preparatory session was required to accommodate clinical considerations.
Research findings support mindfulness-based therapies for autistic adults (Spek et al. 2013). Consequently, participants received standardized mindfulness-based therapy adapted from dialectical behavioral therapy (DBT) as part of their treatment (Linehan 1993). DBT was developed to support individuals struggling with interpersonal relationships, emotion regulation, and distress tolerance. In general, these psychosocial domains are challenging for autistic adults with SAD. A notable advantage of mindfulness-based preparatory psychotherapy was the introduction of vocabulary and skills that helped participants with transitioning into MDMA-influenced cognitive and affective states, as well as with communicating with others during novel, often ineffable, altered states of consciousness.
For experimental sessions, participants arrived around 09:30. They were required to refrain from eating after 24:00 (midnight) except for non-alcoholic fluids prior to the session. Study visits took place in a room with a den-like ambiance, which was designed to minimize sensory distress (e.g., soft lighting, noise abatement). Per consultation with members of the autistic community, features such as elements of nature (e.g., fresh flowers), “fidget” objects for self-regulating through repetitive movement (“stimming”), and suitable décor items were added to support common autistic preferences. Additionally, the room accommodated esthetic adjustments for comfort (e.g., seating arrangements, temperature) and had an adjacent private lavatory.
Study drug was administered around 10:30 after a guided progressive muscle relaxation exercise (McCallie et al. 2006). The experimental sessions were video-recorded, contingent upon participant consent for adherence rating and training purposes. Water intake was monitored to avoid dehydration or water intoxication; optional snacks and a light meal were made available 3 h after study drug administration. Sessions concluded in late afternoon, and participants were ready to leave around 17:30 after a brief closing. Participants were given a contact for urgent assistance from an investigator physician.
The morning following each experimental session, participants returned to the center for integrative psychotherapy. Safety data were collected, the content of the previous day’s experience was examined, and methods for adjusting back to daily life after treatment were reviewed. Two in-person integrative psychotherapy sessions were scheduled at 2-week intervals for 1 month and again at the 6-month follow-up point, with the option of adding an additional office visit, if needed. Telephone safety checks occurred each of the 7 days following experimental sessions. During these calls, spontaneously reported reactions were recorded; the call length was extended to provide additional time for processing cognitive and affective responses, as appropriate.
The primary outcome measure was the LSAS, a 24-item, semi-structured interview evaluating the severity of social anxiety symptoms. The LSAS has been used widely in studies, including research on SAD in autistic adults (Bejerot et al. 2014). In addition, change from baseline was assessed with secondary measures, including BDI-II, PSS, Interpersonal Reactivity Index (IRI) (Davis 1980; Davis 1983), Rosenberg Self-Esteem Scale (RSES) (Rosenberg 1965), STAI, Toronto Alexithymia Scale (TAS-20) (Bagby et al. 1994), The Awareness of Social Inference Test (TASIT) (McDonald et al. 2006), and Emotion Regulation Questionnaire (ERQ) (Gross and John 2003).
Blood pressure, heart rate (GE Medical Systems Information Technologies, Tampa, FL), and temperature (Braun, Kronberg im Taunus, Germany) monitoring was performed pre-drug, then hourly for 6 to 7 h following administration of active drug or placebo.
Investigators collected adverse events and concomitant medications at each visit and spontaneously reported reactions during experimental sessions and 7 days after. Suicidal ideation was assessed with the C-SSRS at the beginning and end of treatment days; subjective units of distress (SUDs) were assessed hourly to determine need for additional support. In addition, a consented study support partner (SSP) drove the participant to and from experimental sessions and from day-after integrative sessions. The participant could choose any trusted adult as their SSP. SSPs were instructed to serve as a nonintrusive supportive presence and to remain in the same location or close by to the participant after treatment through the visit the next morning.
Statistical Package for the Social Sciences (SPSS), version 20 (IBM Corp, Chicago, IL), was used for analyses. Data from the MDMA dose subgroups were combined into one MDMA group (75–125 mg) for analysis due to the small sample size and all doses being within the active therapeutic range of MDMA. Independent samples t tests were used to test for significant changes in LSAS total score from baseline to 1 month post-second experimental session, designated as the primary endpoint, and from baseline to 6-month follow-up. Analyses of secondary outcome measures were exploratory; descriptive statistics are presented. The alpha level indicating significance for primary analysis was 0.05 (two-tailed). Effect sizes were estimated using Cohen’s d independent-groups pretest-post-test design (Kadel and Kip 2012). Primary outcome results from one participant in the MDMA group are missing due to emerging medical history information, which indicated that the participant no longer satisfied inclusion criteria. For the intent-to-treat set, this participant’s baseline scores were included; missing data was not imputed.