Sleep Quality Improvements After MDMA-Assisted Psychotherapy for the Treatment of Posttraumatic Stress Disorder

Sleep disturbance is highly prevalent among people with posttraumatic stress disorder (PTSD) and is implicated in the onset, maintenance, and poorer treatment outcomes of the disorder. Earlier research shows that PTSD is associated with a range of sleep problems—insomnia, nightmares, fragmented sleep, atypical sleep behaviours and daytime dysfunction—and that these problems can impede fear extinction and recovery. Standard PTSD treatments often reduce core PTSD symptoms without fully resolving sleep disturbance, and many patients either drop out of or fail to respond to available sleep-specific interventions, indicating an unmet need for therapies that address both PTSD and sleep problems. Ponte and colleagues used pooled data from four Phase II, randomised, double-blind trials of MDMA-assisted psychotherapy to examine whether active MDMA (75–125 mg) given in conjunction with manualised psychotherapy would improve self-reported sleep quality, as measured by the Pittsburgh Sleep Quality Index (PSQI), and whether changes in sleep quality would be associated with reductions in PTSD symptom severity (measured by the CAPS-IV). The investigators hypothesised that PSQI decreases at the primary endpoint would correlate with reductions in CAPS-IV total scores and explored whether a clinically meaningful PSQI improvement (a decrease of 3 points or more) predicted greater PTSD symptom change at treatment exit and 12 months.

Data derive from four randomised, double-blind Phase II trials conducted at sites in the United States, Canada, and Israel. The pooled sample included civilians, veterans, and first responders aged 18 or older who had chronic PTSD (symptoms ≥ 6 months) and inadequate response to at least one prior pharmacotherapy or psychotherapy. Eligibility required a CAPS-IV total severity score threshold (≥ 50 for three studies, ≥ 60 for one). Key exclusions were current or past psychotic disorders, bipolar I, recent substance abuse, unstable medical conditions, pregnancy, and low body weight; psychiatric medications were tapered off before participation with limited permitted exceptions. Participants were randomised to receive two blinded, 8-hour psychotherapy sessions with either an active MDMA dose (initial 75, 100, or 125 mg with an optional supplemental half dose) or a control dose (0–40 mg with optional supplement), delivered alongside a manualised, largely non-directive co-therapy approach that included three 90-minute preparatory sessions and three 90-minute integrative sessions after each experimental session. After the primary endpoint (1–2 months after the blinded sessions) participants and investigators were unblinded; full-dose participants were offered an additional open-label session and control-group participants crossed over to receive active MDMA in the open-label segment. Participants stayed overnight and received night attendant support after experimental sessions. Sleep quality was assessed with the 19-item PSQI (global range 0–21; clinical cut-off ≥ 5), and PTSD symptoms with the CAPS-IV (total range 0–136). Cronbach's alpha for the PSQI in this sample was reported as .63 at baseline, rising to .80 at long-term follow-up. Because the PSQI was added as an amendment at one site, not all participants completed baseline PSQI. The pooled analytic sample (modified intent-to-treat) included participants who completed at least one blinded session and at least one follow-up; one participant missing baseline PSQI was excluded from PSQI analyses. Primary comparisons between active and control groups from baseline to the primary endpoint used independent-samples t tests (alpha = .05, two-tailed), with Hedges' g for between-group effect sizes. Given that most participants ultimately received active MDMA, longitudinal outcomes (baseline, treatment exit [TE], and 12-month follow-up) were analysed using a mixed-effects repeated-measures model (MMRM) with an unstructured covariance, restricted maximum likelihood estimation, and fixed effects for time, baseline CAPS-IV, and study; participant was modelled as a random effect. The investigators also examined whether achieving a clinically meaningful PSQI drop (≥ 3 points) at the primary endpoint related to CAPS-IV trajectories, and tested candidate covariates (age, sex, race, PTSD duration, prestudy ecstasy use) with prespecified selection thresholds. Pearson correlations were used to assess associations between change scores.

The pooled sample comprised 63 participants (active-dose n = 47; control-dose n = 16). Mean age was 40.8 years (SD = 11.49); 54.0% were male. The majority of participants were White/Caucasian (85.7%). Mean PTSD duration was lengthy (225.3 months, SD = 208.66). Trauma types included combat (71.4%), sexual assault (46.0%), child abuse (30.2%), accidents (15.9%), and other (69.8%). Lifetime suicidal ideation and behaviour were common (87.3% and 30.2% respectively). Prior ecstasy use was reported by 28.6% of participants. At the primary blinded endpoint, the active MDMA group showed significantly greater improvement than the control group on both primary measures. CAPS-IV total severity declined by a mean of -34.0 (SD = 26.46) in the active group versus -12.4 (SD = 16.38) in the control group, p = .003. PSQI global scores improved by a mean of -3.53 (SD = 5.03) in the active group compared with a mean worsening of 0.56 (SD = 3.05) in controls, p = .003. A clinically meaningful PSQI improvement (≥ 3-point drop) occurred in 53.2% of active participants versus 12.5% of controls (χ2 = 8.87, p = .003). Analyses of PSQI subscales at the primary endpoint revealed significant between-group differences for three domains. The Sleep Quality subscale improved more in the active group (mean change -0.62) than in controls (+0.19), p = .006. Sleep Latency showed a mean improvement of -0.42 in the active group versus a deterioration of +0.38 in controls, p = .005. Daytime Dysfunction improved in the active group (mean -0.82) while worsening in controls (+0.25), p < .001. The investigators note that sleep-specific items from the CAPS-IV were not removed from analyses. Because most participants received active MDMA during blinded or open-label phases, longitudinal analyses combined groups for TE and 12-month follow-up. In MMRM models, CAPS-IV scores decreased significantly from baseline to TE and from baseline to 12 months (both p < .001), with further reduction from TE to 12 months (p = .011). PSQI global scores also declined from baseline to TE and to 12 months (both p < .001), with additional improvement from TE to 12 months (p = .030). Age was the only covariate that met significance criteria, with participants under 30 showing greater sleep improvement (p = .017). Change in PSQI from baseline to TE correlated with change in CAPS-IV over the same interval (Pearson r = .371, p = .004). A subgroup analysis found that achieving a ≥ 3-point PSQI drop at the primary endpoint was associated with lower CAPS-IV scores at TE (least-squares mean group difference -14.41, SD = 7.06, p = .046) and at 12-month follow-up (group difference -23.06, SD = 6.00, p < .001). At 12 months, 21 of 63 participants had PSQI scores ≤ 5 (below the clinical cut-off), compared to only two participants at baseline. On the nights following experimental sessions, some active-group participants received sleep aids (e.g., zolpidem, melatonin, zopiclone); no control-group participants received sleep aids on the first night.

Ponte and colleagues interpret their pooled Phase II data as evidence that MDMA-assisted psychotherapy leads to clinically meaningful improvements in both sleep quality and PTSD symptoms among individuals with chronic, treatment-resistant PTSD. At the blinded endpoint, the active-dose group exhibited greater reductions in CAPS-IV scores and PSQI global scores than controls, and about half of active participants achieved a ≥ 3-point PSQI improvement. The investigators further report that gains in sleep and PTSD symptoms persisted and continued to improve through 12-month follow-up, with a modest but significant correlation between improvements in sleep and reductions in PTSD severity. The authors highlight improvements on PSQI subscales for sleep quality, sleep latency, and daytime dysfunction, and propose that decreases in hyperarousal—a core feature of PTSD that affects both sleep onset and daytime functioning—may underlie these changes. They also note that MDMA can acutely disrupt sleep in the days immediately following dosing, but that the sleep impairments reported acutely are transient and contrast with the longer-term sleep improvements observed at follow-up. Key limitations acknowledged include the relatively small sample size and limited racial/ethnic diversity (85.7% White/Caucasian), reliance on self-report sleep measures rather than objective polysomnography, and potential confounding by untreated comorbid sleep disorders such as obstructive sleep apnoea that may require separate treatment. The authors recommend future research to compare subjective and objective sleep measures after MDMA-assisted psychotherapy, to use PTSD-specific sleep assessments, and to examine whether MDMA differentially affects distinct sleep symptoms (for example, nightmares, acting-out dreams, temperature dysregulation) and whether observed benefits generalise to less treatment-resistant populations. Overall, the study team concludes that MDMA-assisted psychotherapy was associated with substantial and durable improvements in both self-reported sleep quality and clinician-rated PTSD symptoms in this pooled Phase II sample, and that sleep improvements were related to greater reductions in PTSD severity at follow-up.