An Open-Label Study of Single-Dose Psilocybin for Borderline Personality Disorder With Co-Occurring Major Depressive Disorder

Borderline personality disorder (BPD) commonly co-occurs with major depressive disorder (MDD), and earlier research has suggested that BPD may make depression harder to treat. The authors note that psilocybin has shown promise for depression, but it was unclear whether people with BPD and MDD would benefit similarly, particularly because BPD symptoms themselves might change differently or interfere with antidepressant response. Grant and colleagues therefore set out to conduct a pilot open-label study of a single dose of psilocybin in adults with both BPD and MDD. Their main aim was to examine whether depressive symptoms and BPD symptoms changed after treatment, and whether co-occurring BPD seemed to limit any antidepressant effect. The study was exploratory and designed to generate early safety and efficacy signals for future trials.

Grant and colleagues conducted an open-label pilot study at the University of Chicago between September 2023 and August 2025. Adults aged 18 to 65 years with current DSM-5 diagnoses of both MDD and BPD were recruited through social media advertisements and referrals. Diagnosis was based on a psychiatrist-led clinical interview. Key inclusion criteria included a Montgomery-Asberg Depression Rating Scale (MADRS) score of at least 20 and a Borderline Personality Disorder Symptom Assessment Scale (BPD-SAS) score of at least 20. Major exclusions included unstable medical illness, bipolar I disorder, schizophrenia or schizoaffective disorder, recent substance use disorder, current suicide risk, pregnancy, recent psychotropic medication changes, current ketamine or esketamine treatment, and cognitive impairment affecting consent. Stable psychotropic medication use and ongoing psychotherapy were allowed. After telephone screening and an in-person baseline assessment, participants received a single oral 25 mg dose of psilocybin one week later. The dosing session lasted about 8 hours and included a supportive setting with trained medical staff present throughout, ambient music, and post-dose observation before discharge to a friend or family member. Follow-up visits occurred at 1, 2, and 4 weeks after dosing. Safety monitoring included the Columbia-Suicide Severity Rating Scale (C-SSRS) and adverse event documentation at each visit. The study had two preregistered co-primary outcomes: change from baseline to week 4 in depressive symptoms measured by the MADRS, and change in BPD symptoms measured by the BPD-SAS. A secondary outcome was the Clinical Global Impression-Improvement scale (CGI-I). The researchers used descriptive statistics and paired-samples t tests to compare baseline with week 4 scores, after checking normality with Q-Q plots and Shapiro-Wilk tests. Analyses were conducted in SPSS, and P<0.05 was considered statistically significant because this was a pilot study.

Of 50 screened individuals, 9 participants were enrolled; 5 were women, and the mean age was 31.33 years. The sample had moderate baseline symptom severity, with mean MADRS and BPD-SAS scores of 28.56 and 32.67, respectively. Seven participants completed all visits, although the two who missed some interim visits still attended the final visit. Four participants were receiving ongoing psychotherapy, and four were taking an antidepressant at a stable therapeutic dose. Depressive symptoms improved significantly over 4 weeks. The mean MADRS score fell from 28.56 at baseline to 17.22 at the final visit, corresponding to a paired t test of t(8) = -4.217, P = 0.003, with a large effect size (Cohen d = 1.41; Hedges g = 1.27). In contrast, BPD symptoms did not change significantly: mean BPD-SAS scores decreased from 32.67 to 28.11, but this was not statistically significant (t(8) = -1.70, P = 0.129; Cohen d = 0.565). On the CGI-I, 3 of 9 participants (33.3%) were rated as much improved and another 3 of 9 (33.3%) as minimally improved; 1 participant (11.1%) was minimally worse by the end of the study. Adverse effects were generally mild and limited to the dosing day. The most common physical events were nausea in 4 participants, fatigue or drowsiness in 1, and dizziness in 1. Emotional effects included anxiety during dosing in 2 participants and crying or emotionality in 3 participants a few hours after psilocybin, all of which resolved by the end of the session. Three participants had passive suicidal ideation at baseline, and one later had a recurrence of similar baseline-level suicidality about a week after dosing, which the participant associated with interpersonal difficulties rather than the psilocybin experience.

The authors interpret the study as providing early evidence that psilocybin may improve depressive symptoms in adults with co-occurring BPD and MDD, while not producing a meaningful change in BPD symptoms themselves. They emphasise that, in this very small open-label sample, BPD did not appear to prevent improvement in depression. They also note that the roughly 12-point reduction in MADRS score is similar to that reported in earlier psilocybin depression research. At the same time, the authors stress that the BPD findings are difficult to interpret. They suggest that the depression improvement could partly reflect placebo response, and they recommend measuring expectancy in future studies to help disentangle this. They also compare the lack of BPD change with the expectation that BPD can be associated with strong placebo responses, which makes the null finding noteworthy but uncertain. Several limitations are highlighted. The study was open-label, very small, and brief, so it cannot establish durability of benefit. The BPD diagnosis was made clinically rather than with a structured diagnostic instrument, which may reduce reliability. The use of two co-primary outcomes in an exploratory pilot study may also limit interpretation. The authors further note that the BPD-SAS is a new measure with unclear psychometric properties, which weakens confidence in the BPD result. They also raise the possibility that repeated psilocybin dosing, rather than a single dose, might have produced stronger or longer-lasting effects. In terms of implications, Grant and colleagues suggest that BPD may not need to be an exclusion criterion in future depression studies of psilocybin, at least based on this small sample. They also suggest it may be worthwhile to examine psilocybin combined with psychotherapy to see whether this could improve BPD symptoms in addition to depressive symptoms.