Therapeutic effect of increased openness: investigating mechanism of action in MDMA-assisted psychotherapy

The paper frames PTSD not only as a cluster of re-experiencing, avoidance, and hyperarousal symptoms but also as a condition that, for some individuals, produces enduring shifts in personality—phenomena sometimes described as "complex PTSD" or, in ICD terminology, "enduring personality change following a catastrophic experience." Earlier research links PTSD and trauma exposure with elevated Neuroticism on the NEO PI(-R), and separate lines of inquiry have shown that classic psychedelics (notably psilocybin) can produce lasting increases in the personality domain of Openness, particularly following profound or mystical-type experiences. MDMA, which has prosocial and anxiolytic pharmacological effects, has been used as an adjunct to psychotherapy for PTSD and has shown sustained reductions in PTSD symptom severity in prior work by the same group. Wagner and colleagues set out to test whether changes in two NEO PI-R personality dimensions—Openness and Neuroticism—might operate as mechanisms linking MDMA-assisted psychotherapy to reductions in PTSD symptoms. Specifically, the study reanalysed data from a randomised, placebo-controlled trial to examine whether increases in Openness and decreases in Neuroticism were associated with CAPS (Clinician-Administered PTSD Scale) score reductions at 2-month follow-up, whether MDMA produced greater personality change than therapy with placebo, and whether any personality changes persisted at long-term follow-up (LTFU) after an open-label crossover for the original placebo group.

This report is a secondary analysis of data from a previously published randomised, double-blind trial of MDMA-assisted psychotherapy for chronic, treatment-resistant PTSD. Participants were screened with structured clinical interviews (SCID/SCID-II) and required to meet DSM-IV-R criteria for chronic crime- or war-related PTSD with a CAPS global severity score of at least 50, despite prior treatment (⩾ 3 months of an SSRI/SNRI plus ⩾ 6 months of psychotherapy). Exclusion criteria included major medical conditions, psychiatric disorders other than PTSD or major depression, and current substance abuse confirmed by medical and laboratory screening. Twenty subjects completed the protocol and the primary 2-month outcome; the sample was mostly female (17 female, 3 male), entirely Caucasian, with mean age 40.4 ± 7.2 years. Initially, participants were randomised to two experimental MDMA-assisted psychotherapy sessions (N = 12) or psychotherapy with an inactive lactose placebo capsule (N = 8). Participants, therapists, and the independent CAPS rater were blinded. The blind was broken after the 2-month follow-up and placebo participants were offered an open-label crossover; seven of eight placebo participants crossed over, yielding 16 participants with LTFU NEO PI-R data. LTFU occurred at a mean of 45.4 months (SD = 17.3). Therapy was delivered by a male–female co-therapist team, with two preparatory sessions, two (and for some a third) experimental sessions, and integration sessions after each experimental session. Experimental dosing was MDMA 125 mg, with an optional supplemental 62.5 mg in eight cases (four MDMA, four placebo). Psychotherapy was non-directive and focused on processing trauma-related material as it arose. Personality was measured with the 240-item NEO PI-R (focusing here on Openness and Neuroticism); PTSD symptoms were assessed with the CAPS administered by an independent rater. Statistical analyses used IBM SPSS Version 21.0. The investigators inspected distributions and tested baseline group equivalence with ANOVA and Chi-square where appropriate. Primary re-analyses used mixed-design repeated-measures ANOVAs: CAPS global scores at baseline and 2-month follow-up served as the dependent variable, with change scores in Openness and Neuroticism (baseline to 2-month follow-up) introduced as covariates to probe their role as mechanisms. Separate repeated-measures ANOVAs tested group (MDMA vs therapy-only) × time effects on NEO Openness and Neuroticism. Long-term persistence was evaluated by comparing baseline and LTFU NEO scores after the crossover. Correlational analyses examined the relationship between changes in Openness and Neuroticism. All p-values reported were two-tailed.

Baseline characteristics did not differ significantly between the MDMA and placebo groups for CAPS scores or trauma type (p > .10). In re-analyses of CAPS global scores from baseline to 2-month follow-up, inclusion of Openness change as a covariate showed a strong main effect of decreased CAPS scores irrespective of treatment: F(1, 17) = 40.60, p < .001. Crucially, there was a significant interaction between change in Openness and CAPS reduction, indicating that participants with the largest increases in Openness had the greatest decreases in PTSD symptoms: F(1, 17) = 5.68, p = .029. When Openness change was adjusted for in the model, the previously reported between-group treatment effect (MDMA vs therapy-only) on CAPS was no longer statistically significant: F(1, 17) = 1.44, p = .246. Analyses using change in Neuroticism as a covariate likewise showed a main effect of decreased CAPS scores regardless of condition: F(1, 17) = 20.48, p < .001. The group × treatment interaction on CAPS scores was significant in this model, with greater CAPS improvement in the MDMA group compared with therapy-only: F(1, 17) = 6.57, p = .02. Change in Neuroticism interacted strongly with CAPS reduction such that larger decreases in Neuroticism associated with larger CAPS decreases: F(1, 17) = 18.83, p < .001. Adjusting for Neuroticism attenuated between-group differences in CAPS: F(1, 17) = 2.85, p = .11. Examining personality scores from baseline to 2-month follow-up, there was no main effect of Openness across time: F(1, 18) = .001, p = .988, but a significant Openness × Group interaction was observed: F(1, 18) = 5.26, p = .034. Visual inspection showed Openness increased in the MDMA group and decreased in the therapy-only group. For Neuroticism there was a significant main effect across time: F(1, 18) = 6.94, p = .017, but no significant Group × Neuroticism interaction: F(1, 18) = 1.03, p = .324. Post-hoc group comparisons did not reach statistical significance for Openness or Neuroticism at baseline or 2-month follow-up (all p > .05), a result the authors note may reflect low observed power (< .20) for these comparisons. Correlational analysis across both groups found a trend for Openness increases to co-occur with Neuroticism decreases: r = -0.435, p = .055. Long-term follow-up analyses (after the open-label crossover) comparing baseline and LTFU NEO PI-R scores indicated significant and enduring changes in both Openness and Neuroticism following MDMA-assisted psychotherapy (p < .05). The extracted text does not report exact mean differences, confidence intervals, or F-statistics for the LTFU comparisons within this excerpt.

Wagner and colleagues interpret their findings as preliminary evidence that MDMA-assisted psychotherapy can engender persistent changes in personality—specifically, increases in Openness and decreases in Neuroticism—that are associated with reductions in PTSD symptoms. The strongest reported result is that increases in Openness moderated the relationship between treatment and CAPS reduction, such that participants with the largest Openness gains showed the greatest symptom improvement. Decreases in Neuroticism were also associated with symptom reduction, and adjusting for either personality change attenuated the between-group treatment effect on CAPS scores. The authors situate these findings within prior literature showing that classic psychedelics can produce sustained increases in Openness and that personality change has been observed after successful treatment in other psychiatric contexts. They discuss possible mechanisms: MDMA’s pharmacology (serotonin release and elevation of oxytocin, prolactin, cortisol) may reduce avoidance and fear responses, thereby enabling engagement with trauma memories and facilitating cathartic or epiphany-type experiences that could underlie lasting personality shifts. The investigators further speculate about potential epigenetic processes that might explain durable personality change, citing general evidence that profound environmental experiences can influence gene expression. Key limitations acknowledged by the authors include a small sample size, reliance on subjective measures (CAPS and NEO PI-R), and compromised blinding—many participants and therapists could correctly guess treatment assignment, raising the possibility of expectancy effects. The authors note that a placebo response seems unlikely to fully account for their results because an independent rater administered CAPS and because treatment effects persisted after the open-label crossover and at long-term follow-up, but they concede the possibility of expectancy influence and state that ongoing studies are using an active placebo/dose–response designs to address this. Finally, conflicts of interest are disclosed: several authors are employed by or received payment from the study sponsor, and one author serves as a medical monitor on other sponsor-conducted studies. The investigators recommend further research to probe causal pathways linking MDMA-facilitated subjective experiences, personality change, and clinical improvement, and to clarify whether the MDMA experience is therapeutic only within a psychotherapeutic context.