Single Treatment With MM120 (Lysergide) in Generalized Anxiety Disorder: A Randomized Clinical Trial

Generalized anxiety disorder (GAD) is a common, chronic condition marked by excessive, uncontrolled worry and associated symptoms such as fatigue, muscle tension, concentration problems and insomnia. Existing pharmacotherapies (for example, selective serotonin reuptake inhibitors and benzodiazepines) and psychotherapies have limitations: many patients do not achieve sustained relief, adverse effects and dependence risks reduce tolerability and adherence, and no new drugs for GAD have been approved in the US since 2007. Lysergide (LSD) has historical and emerging evidence suggesting acute psychological effects via 5-HT2A receptor agonism and possible durable benefits mediated by post-dose neuroplasticity; however, prior trials typically combined LSD with psychotherapy, leaving the independent contribution of a drug-only intervention unclear. Robison and colleagues therefore conducted a Phase 2b multicentre, randomised, double-blind, placebo-controlled trial to characterise the dose–response relationship, efficacy, safety and tolerability of a single oral dose of MM120 (lysergide D-tartrate) in adults with moderate to severe GAD. The study compared four fixed doses (25 μg, 50 μg, 100 μg and 200 μg, freebase equivalents) versus placebo, using central blinded raters and a prespecified multiple comparison procedure–modelling (MCP‑Mod) approach to identify the minimally efficacious and optimal dose to inform Phase III trials.

This Phase 2b trial enrolled adults aged 18–74 years with a primary DSM‑5 diagnosis of GAD confirmed by MINI and baseline Hamilton Anxiety Rating Scale (HAM‑A) score ≥20. The multicentre study ran at 22 outpatient psychiatric research sites in the United States. After screening and a medication taper/washout when required, 198 participants were randomised equally, without stratification, to a single oral dose of MM120 at 25 μg (n = 39), 50 μg (n = 40), 100 μg (n = 40), 200 μg (n = 40) or placebo (n = 39). Participants receiving psychotherapy at baseline could continue at the same frequency but could not initiate or change therapy during the trial. Recent or frequent psychedelic use and contraindicated medical/psychiatric conditions were exclusionary. Dosing sessions were individual, conducted in a private room with two trained dosing session monitors present for safety and comfort. Psychotherapeutic interactions were explicitly prohibited. Participants were continuously monitored for a minimum of 12 hours; hourly assessments began 8 hours post-dose to characterise resolution of acute effects. Efficacy and safety follow-up visits occurred on day 2 and at weeks 1, 2, 4, 8 and 12. Independent central raters, blind to treatment allocation and visit number, assessed HAM‑A and MADRS via telephone to reduce bias. The prespecified primary outcome was a dose–response relationship for change in HAM‑A from baseline to week 4, evaluated using the MCP‑Mod method at a two‑sided α = 0.05. The key secondary outcome applied the same MCP‑Mod approach to HAM‑A change at week 8. Secondary endpoints included HAM‑A at other time points, MADRS, Clinical Global Impression–Severity (CGI‑S), response (≥50% HAM‑A reduction), remission (HAM‑A ≤7), and measures of function/quality of life. The MCP‑Mod analysis tested six candidate dose–response models (linear, Emax and multiple sigmoid Emax variants) and used an AIC‑weighted average of the top models to estimate the dose–response curve and the minimally efficacious dose (with a minimal clinically important difference prespecified as 2.5 HAM‑A points). Efficacy analyses used the full analysis set (all randomised with baseline and ≥1 post‑baseline HAM‑A); missing primary end point data were handled with multiple imputation (20 iterations), with placebo‑based imputation for observations missing due to prohibited medication use. Secondary analyses used ANCOVA adjusted for baseline and treatment group and other standard statistical tests as prespecified.

Of 554 screened, 198 participants were randomised and dosed; 194 comprised the full analysis set (mean age 41.3 years, 56.7% female, 83.0% White). Baseline mean HAM‑A scores were similar across groups (≈29–31). Many participants required medication tapering before baseline (84%), 18% were receiving psychotherapy at enrolment, and 16% were treatment‑naive. The MCP‑Mod procedure identified statistically significant dose–response relationships for HAM‑A change at both week 4 (primary) and week 8 (key secondary). At week 4, the 100 μg and 200 μg groups showed significant placebo‑adjusted reductions in HAM‑A: least‑squares mean differences (LSMDs) versus placebo were −5.0 points (95% CI −9.6 to −0.4) for 100 μg and −6.0 points (95% CI −9.8 to −2.0) for 200 μg. The 25 μg and 50 μg groups did not differ significantly from placebo at week 4. At week 8, the 100 μg group had an LSMD of −2.9 points (95% CI −5.4 to −0.5) and the 200 μg group −5.4 points (95% CI −9.5 to −1.4) versus placebo; the lower doses again were not significant. Treatment effects persisted through week 12. At that time point, LSMDs versus placebo were −7.7 points (95% CI −12.6 to −2.8; Cohen d = 0.81) for 100 μg and −7.4 points (95% CI −12.4 to −2.4; Cohen d = 0.77) for 200 μg. Response rates (≥50% HAM‑A reduction) at week 12 were 65.0% for 100 μg, 62.5% for 200 μg and 30.8% for placebo; remission rates were 47.5%, 45.0% and 20.5%, respectively. The 100 μg and 200 μg doses also produced significant improvements on clinician‑rated CGI‑S from day 2 and on depressive symptoms (MADRS) from week 1 through week 12; the 25 μg and 50 μg doses did not reliably separate from placebo on these measures. Safety findings showed dose‑related adverse events. Rates of ≥1 treatment‑emergent adverse event (TEAE) were 76.9% (25 μg), 90.0% (50 μg), 97.5% (100 μg), 100% (200 μg) and 56.4% (placebo). Treatment‑related TEAEs followed a similar pattern. Most AEs occurred on the dosing day and resolved within the 12‑hour observation period. Perceptual changes were common and dose dependent: visual perceptual changes were reported in 92.5% (100 μg) and 100% (200 μg) on dosing day (lower-dose percentages and placebo were lower), and nausea and headache were also more frequent at higher doses. Two severe AEs of “feeling intoxicated” occurred on dosing day (one each in the 50 μg and 100 μg groups) and resolved during the session. One SAE (hospitalisation for a panic attack) occurred 97 days post‑dose in the 50 μg group and was judged unrelated to treatment. Five participants (2.5%) withdrew because of AEs. There were no reports of suicidal behaviour related to dosing; a small number reported suicidal ideation without intent, consistent with their baseline history. The MCP‑Mod analysis estimated the minimally efficacious dose to lie between 56.7 μg and 85.1 μg.

Robison and colleagues interpret the findings as evidence that a single oral dose of MM120 produces a dose‑dependent reduction in anxiety among adults with moderate to severe GAD, with the 100 μg dose identified as optimal for further development. A treatment effect appeared rapidly (by day 1) and was sustained through 12 weeks on clinician‑rated and central‑rater measures of anxiety and on depressive symptom scales. The investigators note that the 200 μg dose produced more frequent adverse events without clear additional efficacy, supporting selection of 100 μg for pivotal trials. The authors place these results in context by comparing effect sizes with prior meta‑analyses: pooled pharmacotherapy trials for GAD showed a Cohen d ≈ 0.39, while psychotherapies have shown larger pooled effects (Cohen d ≈ 0.76 in one meta‑analysis). They underscore that MM120 demonstrated an acute onset akin to benzodiazepines but without the need for repeated dosing in this single‑dose design, and that observed safety and tolerability were consistent with previous LSD research. Potential mechanisms discussed include 5‑HT2A receptor‑mediated acute effects and enhanced post‑dose neuroplasticity in brain regions such as the prefrontal cortex. The study team acknowledges several limitations. First, interpersonal interaction with dosing session monitors—despite explicit training to avoid psychotherapeutic interventions—could have contributed to outcomes. Second, perceptible drug effects likely led to participant unblinding; to mitigate this, independent central raters blinded to allocation and visit number conducted primary efficacy assessments and remained blinded for over 80% of HAM‑A evaluations, although approximately 85% of participants correctly identified active treatment. Third, only the primary and key secondary outcomes were multiplicity‑controlled; other hypothesis tests were not adjusted. Fourth, dropout was slightly higher than comparable trials, which may relate to the 12‑week follow‑up and required medication tapering. Fifth, the sample was physically healthy with limited psychiatric comorbidity and was less racially and ethnically diverse than the broader GAD population, so generalisability may be limited. Lastly, concomitant pharmacotherapy was excluded by tapering, and only a minority of participants (18%) were in psychotherapy during the trial, so the interaction of MM120 with ongoing treatments or systematic psychotherapeutic approaches was not addressed. Overall, the authors conclude that the trial supports dose‑dependent efficacy of single‑dose MM120 and informs the choice of 100 μg for Phase III evaluation, while recognising the need for further studies to confirm efficacy, durability and safety in broader and more diverse clinical populations.

In this Phase 2b trial of adults with moderate to severe GAD, a single oral dose of MM120 produced dose‑dependent reductions in anxiety symptoms, with the 100 μg dose showing the most favourable balance of efficacy and tolerability. These results support selecting 100 μg MM120 for pivotal Phase III trials to evaluate its therapeutic benefit and safety profile in the treatment of GAD.