LSD microdosing in major depressive disorder: results from an open-label trial

Major depressive disorder (MDD) is a leading global cause of disability, affecting roughly 5% of adults and often responding incompletely to conventional antidepressants, which have slow onset and variable tolerability. Interest has therefore grown in alternative treatments, including classic serotonergic psychedelics such as LSD, psilocybin and DMT. Microdosing—repeated administration of sub-perceptual doses of psychedelics—has become common in community settings and is frequently reported as an attempt to self-manage depressive symptoms. Controlled studies in healthy volunteers have shown that single microdoses of LSD (5–20 µg) can acutely alter neural connectivity, cognition and mood, and a small body of work suggests transient mood enhancement in mildly depressed individuals, but the therapeutic potential of repeated microdosing for clinical depression remains unclear. Daldegan-Bueno and colleagues set out to evaluate the tolerability, feasibility and preliminary clinical effects of an 8-week regimen of sublingual LSD microdosing in people with moderate MDD in a Phase 2A open-label trial (LSDDEP1). The investigators additionally monitored safety with routine blood tests, ECG and transthoracic echocardiography to assess the theoretical risk of 5-HT2B receptor-mediated valvulopathy. Secondary aims included measuring changes in depressive symptoms and other mental health and quality-of-life outcomes, and assessing whether the protocol would be practical to carry forward into a Phase 2B randomised controlled trial.

This open-label Phase 2A trial recruited adults aged 21–65 years with DSM-5 major depressive disorder and a Montgomery-Åsberg Depression Rating Scale (MADRS) score of 18–35 at screening. Key exclusions included current or past psychotic or bipolar disorder, recent problematic substance use, recent or lifetime psychedelic microdosing or psychedelic use to treat depression, significant medical or cardiac abnormalities, and pregnancy or lactation. Participants already taking stable antidepressant therapy were permitted to continue but were asked not to change treatments during the trial. All procedures were approved by the institutional ethics committee and relevant regulatory authorities. The investigational product was a GMP-quality LSD hemitartrate liquid formulation (MB-22001), dispensed in 20 µg units (0.1 ml = 1 µg). Participants received 16 doses over eight weeks, taken twice weekly. The first dose (8 µg) was administered under supervision in clinic; subsequent doses were self-administered sublingually at home with instructions to avoid risk activities for 6 hours post-dose and not to dose after 14:00. A custom smartphone application recorded self-administered dosing via video uploads and collected end-of-day subjective effect questionnaires used to guide individual titration. Dose titration began at 8 µg and could be adjusted within a 4–20 µg range according to two iterative titration schemes based on participant-reported subjective effects; the scheme was amended during the trial to allow larger incremental adjustments. The titrated dose for each participant was defined as the dose most frequently taken. Participants could opt into an 8-week extension that repeated the same protocol. Primary trial metrics were tolerability, operationalised as the percentage of participants who withdrew due to adverse events related to treatment, and feasibility, measured by the percentage of scheduled clinic visits attended. The MADRS (structured with the SIGMA interview guide) was the primary clinical outcome and was measured at baseline and at 2, 4, 6 and 8 weeks, with follow-up assessments at 1, 3 and 6 months after treatment. Participants achieving ≥50% MADRS reduction from baseline to 8 weeks were classified as responders; MADRS <10 at 8 weeks defined remitters. Secondary measures included HAM-A (anxiety), Ruminative Response Scale (RRS), DASS, Dimensional Anhedonia Rating Scale (DARS) and WHOQOL-BREF. Safety assessments comprised full blood count and metabolic panels, 12-lead ECG, transthoracic echocardiography before and after treatment to assess valvular function, and suicidality monitoring (C-SSRS and MADRS item 10). Discontinuation-like symptoms were assessed with the DESS at one month post-treatment. Analysis was descriptive only, as a pilot trial: means with standard deviations or 95% bootstrapped confidence intervals were reported for continuous variables, and counts and percentages for categorical variables. No inferential hypothesis tests or formal sample size calculation were performed; the recruitment target was approximately 20 participants.

Recruitment and sample: Of 173 people completing prescreening, 23 were allocated and 19 received the intervention. The analysed sample was predominantly male (79%) and New Zealand European ethnicity predominated; 26% identified as Māori or Pasifika. Baseline mean MADRS was 23.7 (SD 6.7), consistent with moderate depression. Most participants (n = 15, 79%) were taking antidepressant medication (primarily SSRIs). Tolerability and feasibility: Two participants withdrew for reasons unrelated to the trial and one withdrew due to an adverse event (anxiety on dosing days). By the study’s definitions, withdrawal due to adverse events attributed to treatment occurred in one participant. Clinic visit attendance among enrolled participants was 100%, and home dosing compliance (verified by video uploads) was reported as 100% for scheduled at-home doses. Overall, 342 doses were taken (23 on-site, 319 at home). The mean titrated dose was 14.61 µg (SD 3.63), range 6–20 µg, with 15 µg being the most frequent titrated dose (n = 6). Safety and adverse events: No serious or severe adverse events were reported. The most common dosing-day adverse event was headache (three events across two participants), representing under 1% of dosing days. One participant experienced anxiety on dosing days and withdrew after five microdoses; this participant had high baseline anxiety and the anxiety resolved after withdrawal. Echocardiography was completed pre- and post-intervention in 15 participants; four did not complete echocardiography. No clinically significant valvular abnormalities were identified after the main 8-week intervention, including in four participants who extended to 32 doses over 16 weeks. For those who had echocardiography after the main trial, the mean total LSD intake cited was 200.27 µg (SD 20.17, n = 11); for those completing the extension the mean total intake was 413 µg (SD 167.54, n = 4). No clinically significant changes were found in ECG parameters or routine laboratory tests. Mean QTc intervals were 408.84 ms (SD 20.23) at screening, 408.49 ms (SD 23.98) at Measure, and 413.58 ms (SD 27.65) at Extension Measure. One participant’s QTc rose from 434 ms at baseline to 460 ms at Measure. DESS screening for SSRI-like discontinuation symptoms one month after stopping microdosing was completed by 13 participants; one participant reported five symptoms and three participants reported one symptom. Depression outcomes: MADRS mean score decreased from 23.7 (SD 6.72) at baseline to 9.59 (SD 7.67) at the end of treatment, a 59.52% reduction. At 8 weeks, nine participants (52.94%) met the remitter criterion (MADRS <10) and nine participants (52.94%) met the responder criterion (≥50% reduction); these categories were not mutually exclusive. The reduction in depressive symptoms emerged by week 2, stabilised by week 4, and was reported to persist for up to six months in follow-up assessments. Other clinical measures: Anxiety measured by HAM-A fell by 51.9% at the end of treatment. DASS subscales showed reductions of 34.9% (stress), 59.1% (anxiety) and 40.6% (depression). Rumination (RRS) reduced by 14.7%. Anhedonia (DARS) scores improved by 14.8% (higher scores indicate less anhedonia). WHOQOL-BREF domains increased, with psychological quality of life showing the largest change (37.3%); physical, social and environmental domains increased by 14.1%, 18.6% and 9.15% respectively, and overall quality-of-life and general health items increased by 9.1% and 8.16% respectively. Supplemental figures and tables in the extracted text present score trajectories and percentage changes.

Daldegan-Bueno and colleagues interpret these findings as preliminary evidence that an 8-week, twice-weekly sublingual LSD microdosing regimen can be delivered feasibly and is generally well tolerated in people with moderate MDD in a naturalistic, home-based setting. High levels of medication compliance and clinic attendance were noted, and no serious or severe adverse events, clinically meaningful laboratory abnormalities, ECG changes, or echocardiographic signs of valvulopathy were observed in the sample and exposure range studied. The investigators highlight the single trial withdrawal due to anxiety (at 6 µg) and the low frequency of dosing-day adverse events (headaches in <1% of dosing days), noting that the titration scheme and low starting dose may have reduced the incidence of tolerability problems. The authors report a pronounced clinical signal: MADRS scores fell by approximately 60% from baseline to 8 weeks, with over half the sample classified as responders and remitters. Improvements were also observed across anxiety, stress, rumination, anhedonia and multiple aspects of quality of life. These effects appeared by two weeks, stabilised by four weeks and were reported to persist for up to six months. The study team cautions that, because this was an open-label pilot without a control condition, observed symptom reductions could reflect placebo effects, regression to the mean, expectancy, the mild psychosocial intervention embedded in the protocol (self-selected beneficial activities on dosing days), or other contextual influences. The investigators note that their sample’s relatively low baseline depression severity may increase susceptibility to placebo response. Safety considerations receive particular attention: this is reported as the first controlled human trial to include echocardiographic assessment for valvulopathy after repeated LSD exposure, and no clinically relevant valvular changes were detected with up to 32 microdoses over 16 weeks in the small subset assessed. Nonetheless, the authors emphasise that the sample size and limited cumulative exposure do not exclude a valvulopathy risk with larger populations or longer-term use, and that theoretical comparisons to drugs with continuous plasma exposure (e.g. pergolide, cabergoline) suggest substantially lower cumulative exposure in the microdosing regimen. Limitations acknowledged by the investigators include the open-label design, small sample size, potential underreporting of adverse events due to self-reporting in a home-dosing context, the possibility that prior psychedelic experience among over half the sample introduced selection bias, and the high proportion of participants concurrently taking antidepressants which may alter LSD effects (for example, SSRIs may attenuate or delay psychedelic effects). Given these uncertainties, the authors conclude that their results justify proceeding to a Phase 2B randomised, placebo-controlled trial to assess efficacy and safety more rigorously; such a trial is stated to be underway as a continuation of this work.