🍄🚨🍄🚨 Another day, another cool preprint from our group: An open-label, dose-escalation trial of psilocybin-assisted therapy for bipolar 2 depression We enrolled individuals with #bipolar 2 (BD-II) currently experiencing a major #depressive episode. Participants must have had at least one unsuccessful medication trial of >6 weeks duration for BD-II over their lifetime. Four participants were on medications with potential #psilocybin interactions, which were tapered under psychiatric supervision to ensure clearance (≥5 half lives) before each administration. Psychiatric exclusions included current hypomania, history of psychotic disorder, moderate substance use disorder in the past 12 months, psychedelic use in the past six months and imminent risk of suicide. In this open-label, single-arm dose-escalation pilot trial, 14 participants received 10 mg of psilocybin, followed by 25 mg if depressive symptoms persisted. As usual participants underwent psychotherapy before, during, and after psilocybin administration sessions and were proactively monitored for adverse events. Depression and quality of life were assessed using the Montgomery-Asberg Depression Rating Scale (MADRS) and Quality of Life in Bipolar Disorder Questionnaire (QoL-BD), along with exploratory measures. Following the 10 mg session, MADRS scores significantly improved at all timepoints (A21: -12.7 [2.7], n=14, p<0.001) with 4 participants (28.5%) meeting remission criteria (MADRS ≤ 6) at 21 days after the 10mg dose. Non-remitting patients were given the 25 mg dose per protocol. After the 25 mg session, MADRS scores improved from baseline (B21: -18.6 [3.1], n=9, p<0.001). Sustained improvements in MADRS scores were observed at the 90-day assessment compared to baseline (AB90: -14.3 [2.8], n=12, p<0.001); Hedges’ g=1.9. QoL-BD scores improved 21 days following both the 10 and 25 mg administration sessions (A21: 35.5 [9.4], p<.001; B21: 55.9 [10.9], p<.001) and at the 90-day assessment (AB90: 31.2 [10.2], p=0.004); Hedges’ g=1.6. As expected, common adverse events included mild-to-moderate anxiety, nausea, and headache on dosing day. There were three notable adverse events: active suicidal ideation 37 days after an administration session, passive suicidal ideation 11 days after an administration session, and a hypomanic episode two weeks after an administration session. The small sample size and open-label design warrant cautious interpretation. The placebo response likely explains some of the observed benefits. However, in line with previous findings, baseline expectancy was not associated with clinical outcomes - to the best of my knowledge no psychedelic trial has provided empirical evidence for a positive expectancy bias at this point. Our results offer initial evidence that psilocybin therapy may be clinically beneficial in BD-II when administered under controlled conditions with a safety profile similar to studies in other patient populations. Full preprint: tinyurl.com/52n6uv4t 🙏+♥️to all coauthors: @drdowneydoes @thebandlab @ellenbradshaw and many others from the @UCSFPsychiatry 's @TrPR_Program! #psychedelic #trial #MentalHealthMonday #medicalinnovation

Tired of the same arguments about unmasking in #MDMA / #psychedelic trails? Bored of people throwing around unmasking without quantitative analysis? Unsure of what is even unmasking bias? If you answered yes to any, this is your lucky day! In this short🚨preprint🚨, we investigate whether unmasking could explain the efficacy of MDMA-AT. As a recap, in both phase 3 trials the MDMA-AT vs. control effect size was ~0.8 standardized mean difference (SMD), which is about 2-3 times larger than currently approved PTSD treatments. Given this large effect, some (including me) have argued that this effect size is too large to be explained by unmasking. But is it? First, we provide a definition of unmasking bias: it is the difference between the 'treatment vs. control' effect size of two trials that are identical except one is completely unmasked, while the other is perfectly masked, i.e. how much the unmasked trial overestimates the 'between-treatment difference' relative to the truly masked trial. Such pairs of trials do not exists for MDMA (all trials are unmasked), but we can find examples for other drugs. If the magnitude of unmasking bias with other drugs is consistently lower than MDMA-AT's ~0.8 SMD, then unmasking is unlikely to explain the positive results. On the other hand, if unmasking bias is higher for other drugs, then concerns about unmasking are justified. We calculated the unmasking bias for ketamine and escitalopram: - In saline controlled #ketamine trials participants are unmasked (~80-90% correct guess rate), but masking is maintained when midazolam is used as control (correct guess rate is at chance level). The difference between the results of these trials (i.e. the difference between the 'ketamine vs. saline' and the 'ketamine vs. midalozam' effect size) is the unmasking bias! - Overt/covert studies can also be used to estimate unmasking bias (in these trials, all patients receive the same drug, but are randomized to be told what they receive (overt) or falsely told they are receiving placebo (covert)). Functionally, overt drug administration is same as an unmasked trial, while covert is same as an masked trial. Such studies exists for #escitalopram! Check figure below for results. In summary, the unmasking bias for ketamine in depression is 1.1 SMD and for escitalopram it ranges from ~0.5-1.4. Thus, the ~0.8 SMD MDMA-AT treatment vs. control condition effect sizes are within the range of unmasking biases found for other psychotropic treatments. These findings do not prove that effects of MDMA-AT are entirely or even partially due to functional unmasking. However, our findings indicate that MDMA-AT’s effect sizes are not too large to be due to unmasking effects - there is some mea culpa here as I was among those who have argued this before I run the current analysis. We will soon submit this paper for peer-review as a viewpoint/perspective type article, constructive criticism is welcomed. Work is in collaboration with @EllenBradleyMD and @thebandlab - thank you both! Preprint: osf.io/preprints/osf/ujtpf