Study design and oversight
This multi-site, randomized, double-blind, placebo-controlled study assessed the efficacy and safety of MDMA-AT versus placebo with therapy in participants diagnosed with moderate or severe PTSD (NCT04077437). Thirteen study sites (11 in the United States and two in Israel, both institutional and private) participated. The trial was conducted in accordance with the Good Clinical Practice guidelines of the International Council for Harmonization and with the ethical principles of the Declaration of Helsinki. An independent data monitoring committee ensured that the study was conducted safely and had sufficient sample size. The review boards and institutions that approved the study protocol are listed in the Supplementary Methods.
After written informed consent, participants were screened for eligibility. Adults (≥18 years of age) meeting the full DSM-5 criteria for current PTSD per CAPS-5 assessment40,41 and a CAPS-5 total severity score ≥28 (moderate or higher severity) with symptom duration of ≥6 months were eligible for enrollment confirmation. During the Preparation Period that preceded the Treatment Period, participants were tapered off all psychiatric medications before baseline to avoid potential drug interactions and confounding efficacy (Supplementary Fig. 1). Full inclusion and exclusion criteria are outlined in the Supplementary Methods.
Randomization and masking
Participants were randomized in a 1:1 allocation and in a blinded fashion to the MDMA-AT and placebo with therapy groups, stratified by clinical site. Randomization was managed via an interactive web randomization system (IWRS) (IT Clinical version 11.0.1) based on a centralized randomization schedule developed by an independent third-party vendor to maintain blinding.
A central pool of blinded independent assessors was used to mitigate the risk of functional unblinding42. Assessors were trained and supervised by independent consultants with expertise in PTSD diagnostics and the CAPS-5 to ensure inter-rater reliability and validity of assessments. Supervision involved reviewing each assessor’s first two assessments as well as 20% of all assessments (chosen at random) throughout the study, with each review resulting in detailed feedback for the assessor. The independent assessors were blinded to the general study design, study visit, treatment assignment, number of treatments received and any safety data for the participant. Participants were instructed to withhold their opinion on treatment group assignment and the number of completed visits from the independent assessors. Each assessor conducted no more than one CAPS-5 assessment with each participant to reduce potential bias and expectancy effect from having conducted repeat CAPS-5s with a participant. Assessors were also vetted before their onboarding to ensure that there were no conflicts of interest (such as other involvement within the Multidisciplinary Association for Psychedelic Studies (MAPS) organization or a bias toward MDMA-AT), and assessors were instructed to not expose themselves to scholarly presentations and papers related to MDMA-AT for PTSD to maintain their blinding to study design.
To ensure that all site and sponsor staff were shielded from study outcomes, the blinded independent assessor pool collected and stored outcome measures in a dedicated database that was separate from the blinded clinical database. A blinding survey was conducted at study termination (visit 20) to assess if participants thought that they received MDMA or placebo.
Trial procedures were consistent with MAPP1 (ref. 12). Enrolled participants underwent three 90-min preparation sessions with a two-person therapy team, including at least one licensed therapist, and were then randomized 1:1 to receive MDMA-AT or placebo with therapy for approximately 3 months. The treatment period consisted of three 8-h dosing sessions, in conjunction with therapy, spaced approximately 1 month apart. Therapy was conducted by trained personnel in accordance with the MAPS MDMA-AT treatment manual (https://maps.org/treatment-manual) and trial protocol. During experimental sessions, and in keeping with the dosing in MAPP1, participants received a split dose of 120–180 mg of MDMA or placebo. For the first experimental session, the initial dose of 80 mg was followed by a supplemental half-dose of 40 mg. In the second and third experimental sessions, the initial dose of 120 mg was followed by a supplemental half-dose of 60 mg. The supplemental half dose was administered 1.5–2 h after the initial dose. Participants in both treatment groups received identical therapy. The 120-mg (80 mg + 40 mg) split dose was selected for the first experimental session in phase 3 trials to allow patients to acclimate to the treatment regimen using a clinical titration approach based on clinician recommendations from a phase 2 trial in veterans and first responders13. During the second and third experimental sessions, doses were escalated to 180 mg (120 mg + 60 mg), as this was the most frequently studied efficacious dose in phase 2 trials. This dosing regimen also provides clinicians with the option of dose adjustments if needed.
Within the MDMA-AT group, three participants did not undergo dose escalation in experimental sessions 2 and 3, and two participants experienced dose administration timing errors (Supplementary Table 2). Each experimental session was followed by three 90-min integration sessions to support participants in processing and understanding their experience (Supplementary Fig. 1). Full procedures, including details on therapy teams and training, are outlined in the Supplementary Methods.
Independent assessors conducted CAPS-5 and SDS outcome assessments at baseline, after experimental sessions 1 and 2 and 6–8 weeks after experimental session 3 (18 weeks after baseline) via video interviews. Primary and secondary objectives were mean change in CAPS-5 total severity and SDS scores, respectively, for MDMA-AT versus placebo with therapy from baseline to 18 weeks after baseline.
Exploratory outcome measurements included characterization of the treatment response and differences between the treatment groups by demographics and characteristics. Responder analyses were based on categorical diagnostic assessment data and the CAPS-5 total severity score assessment. PTSD severity was defined using the CAPS-5 total severity score as follows: asymptomatic (0–10), mild (11–22), moderate (23–34), severe (35–46) and extreme (47+) (ref. 41). A ≥10-point reduction in CAPS-5 total severity score was considered to be clinically meaningful as agreed upon with the FDA through a Special Protocol Assessment. Four responder categories were derived and compared at each post-experimental session visit using CAPS-5 scores. These categories were: non-responder (<10-point reduction from baseline), responder (≥10-point reduction from baseline), loss of diagnosis (≥10-point reduction from baseline and no longer meeting PTSD diagnostic criteria) and remission (CAPS-5 total severity score of 11 or less and no longer meeting PTSD diagnostic criteria).
Safety objectives included assessment of differences between groups in severity, incidence and frequency of TEAEs, serious TEAEs, TEAESIs, suicidal ideation and behavior and vital signs. TEAEs were defined as any adverse event that occurred during the treatment period from the first experimental session to the last integration session. The severity of TEAEs was determined by the site physician as mild (no limitation in normal daily activity), moderate (some limitation in normal daily activity) or severe (unable to perform normal daily activity). A serious TEAE was defined as any unforeseen medical event at any dose of the drug that resulted in death; was life-threatening; required inpatient hospitalization; caused significant disability or incapacity; resulted in a congenital anomaly or birth defect; or required intervention to prevent permanent impairment or damage. Serious TEAEs also included any event, based on medical judgement, that jeopardized the participant or may have required intervention to prevent one of the events listed previously. With the exception of serious adverse event reporting, relatedness to study drug was not assessed by investigators, to preserve blinding. In an effort to identify common adverse events that may be most related to MDMA, TEAEs occurring with incidence >10% and at least twice the prevalence in the MDMA-AT group versus the placebo with therapy group are reported. Suicidality was tracked at each study visit using the C-SSRS (see the Supplementary Methods for more information).
SAS version 9.4 (SAS Institute) was used for analyses. Sample size was calculated to achieve a power of 90% at an alpha of 0.0499.
Efficacy was tested using an MMRM analysis comparing the change from baseline to 18 weeks after baseline in CAPS-5 and SDS scores between treatment groups in two-sided tests with alpha set at 0.0499. The alpha was adjusted to account for an administrative interim analysis for sample size re-estimation conducted after all participants were enrolled and 60% of primary endpoint data had been collected. Fixed effects were treatment, visit, treatment group by visit interaction and dissociative subtype; baseline CAPS-5 score was a covariate. Primary and secondary efficacy analyses used a de jure (related to initially randomized treatment) estimand and a supportive de facto (treatment policy) estimand of the modified intention-to-treat population, which required exposure to MDMA or placebo and at least one follow-up CAPS-5 assessment, as in MAPP1 (ref. 12). The de jure dataset included all available data, except for 12 (one MDMA-AT and 11 placebo with therapy) outcome measurements taken after treatment discontinuation in analysis of treatment efficacy (Supplementary Table 3). Missed observations were considered missing at random (MAR), and choice of this assumption was tested with a tipping point analysis (Supplementary Methods).
In additional exploratory analyses, 13 covariates were assessed in the model, with alpha set at 0.0499: age, sex (self-reported), prior use of selective SSRIs, work disability, disease severity, PTSD duration, dissociative subtype, overnight site stay, site ID, moderate depression (as measured by the BDI-II), severe adverse childhood experiences and moderate alcohol and substance use disorder risk (as measured by the Drug Use Disorders Identification Test and the Alcohol Use Disorders Identification Test). Analyses of primary or secondary outcomes by gender were not planned a priori; some exploratory analyses included sex as a covariate (Supplementary Methods).
Safety analysis evaluated TEAEs at the participant level, including all participants who received MDMA or placebo. Causal association with MDMA was determined based on relative incidence of TEAEs with at least a two-fold difference between groups.
Adverse events of special interest
In accordance with FDA guidance, special attention was paid to a subset of adverse events, TEAESIs, relating to cardiac function, suicide risk and MDMA abuse, misuse or diversion. TEAESIs involving cardiac function that could be indicative of QT prolongation or cardiac arrhythmias were collected, including torsade de pointes, sudden death, ventricular extrasystoles, ventricular tachycardia, ventricular fibrillation and flutter, non-postural syncope and seizures. TEAESIs involving suicide risk included suicide, suicide attempts, self-harm associated with suicidal ideation, suicide ideation assessed as a score of 4 or 5 on the C-SSRS and suicidal ideation judged by the investigator to be serious/severe. TEAESIs involving terms of MDMA abuse, misuse, drug diversion, dependence or overdose were also collected.
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.