Definium Therapeutics has reported positive results from its first Phase 3 LSD anxiety trial. The data push a psychedelic medicine closer to a regulatory filing in a condition that has not seen a new drug approval since 2007. The company said on 12 August that Voyage met its primary endpoint and every key secondary endpoint. Voyage tested the oral lysergide candidate DT120 in generalised anxiety disorder. It is the second late-stage win for the drug in under two months.
Definium trades on Nasdaq as DFTX and was known as MindMed until January 2026. It reported positive Phase 3 results in major depressive disorder in June. The company now runs late-stage programmes across three psychiatric indications. Voyage is the first of two pivotal LSD anxiety studies. Management has said it intends to file for approval in both anxiety and depression.
What the LSD anxiety trial measured
Voyage enrolled 214 participants aged 18 to 74 across roughly 35 sites in the United States. All had a DSM-5 confirmed diagnosis of generalised anxiety disorder. Diagnosis was established through clinical assessment and the Mini-International Neuropsychiatric Interview. All scored at least 20 on the Hamilton Anxiety Rating Scale at screening and baseline. Mean baseline scores were 28.4 on treatment and 27.4 on placebo, placing most participants in the severe range.
Participants were randomised one to one. They received either a single 100 microgram dose of DT120 or matching placebo. The double-blind period ran for 12 weeks. A 40-week open-label extension followed, in which participants could receive up to four further doses.
The primary endpoint was change in Hamilton Anxiety Rating Scale total score at week 12. Participants on DT120 improved by 11.6 points. Those on placebo improved by 6.2 points. The placebo-adjusted difference of 5.4 points was significant at p below 0.0001. The standardised effect size was 0.81. Effect sizes above 0.8 are conventionally described as large.
Rapid onset and a durable signal
The most striking feature of the LSD anxiety data is the speed of separation. By week one, the treatment group had improved by 11.9 points against 4.2 points on placebo. That is a gap of 7.7 points. On the Clinical Global Impression-Severity scale, clinicians rated treated patients as meaningfully less impaired by day two. The placebo-adjusted difference there was 0.8 points. All three key secondary endpoints were multiplicity controlled, and all reached p below 0.0001.
Response rates followed the same pattern. At week 12, 43 per cent of treated participants achieved at least a 50 per cent improvement, against 16 per cent on placebo. Just over half reached a mild or better symptom level, compared with 23 per cent on placebo. Full remission was reached by 14 per cent on DT120 and 4 per cent on placebo. That difference was significant at p equals 0.0222.
The LSD anxiety efficacy gap is narrower than in earlier testing. The company’s Phase 2b study, published in JAMA in October 2025, recorded a 7.7-point placebo-adjusted reduction at week 12 at the same dose. The standardised effect size, however, was identical in both at 0.81. The two studies are not directly comparable. The Phase 2b was a dose-ranging trial with roughly 40 participants per arm and a primary endpoint at week 4, against 107 per arm and a week 12 primary endpoint in Voyage, and absolute score changes in both the treatment and placebo arms were considerably larger in the earlier study. Analysts had broadly expected some attenuation in a larger population.
Safety and the monitoring question
Definium described the safety profile as consistent with prior clinical experience. Treatment-emergent adverse events were mild to moderate and transient. They clustered on the day of dosing. Definium reported no new safety signals. It also reported no signal of suicidality or suicidal behaviour, which matters for a Schedule I compound in a psychiatric population.
The practical constraint sits in the dosing session. DT120 produces transient perceptual, cognitive and affective effects. Participants were therefore monitored in clinic and assessed hourly from five hours after dosing. Assessment used a structured end-of-session checklist. The average time to meeting those criteria was 6.4 hours, with a median of 6.1 hours. By hour eight, 92 per cent of participants had cleared. Any approved product would require supervised administration, a delivery model closer to an infusion suite than a prescription pad.
Why generalised anxiety disorder has stalled
Commercial interest in the programme rests on a therapeutic gap of nearly two decades. The last new US approval for generalised anxiety disorder came in February 2007, when duloxetine was cleared for the indication. Serotonin and serotonin-noradrenaline reuptake inhibitors remain first-line treatment. Benzodiazepine use has declined on dependence concerns.
Definium cites an estimated 26 million affected adults in the United States. That figure draws on a national prevalence study combined with census data and internal company estimates, rather than a single independent source. Brian Barnett is Vice Chair of psychiatry at Cleveland Clinic and a Voyage principal investigator. In a company statement, he described generalised anxiety disorder as one of the most undertreated conditions in psychiatry. He added that many of his patients have found no relief despite trying multiple therapies. An approved LSD anxiety treatment, he said, would work through an entirely different mechanism from anything currently used in clinical practice.
How analysts read the LSD anxiety results
Sell-side reaction was positive, and in places better than expected. Andrew Tsai at Jefferies noted that the separation exceeded the roughly five-point figure the market had anticipated. Paul Matteis at Stifel called the outcome a “clean win”. He described the efficacy as rapid and consistent with what had been seen in earlier testing, and the durability as continuing to be robust. He also wrote that the run of catalysts ahead, with a second anxiety readout, a second depression study and a PTSD programme to come, sets the company up to pursue label expansion soon after any first approval.
Revenue models vary. Gavin Clark-Gartner at Evercore ISI models $2 billion in peak annual sales for anxiety, plus $2 billion from depression and $1 billion from post-traumatic stress disorder. Following June’s depression readout, Marc Goodman at Leerink Partners forecast $1.5 billion to $2 billion in depression alone. Shares opened around 15 per cent higher on 12 August at just above $47. They settled near $43 by mid-morning.
A supportive policy backdrop
The LSD anxiety data land in a favourable regulatory environment. On 18 April 2026, President Trump signed Executive Order 14401, Accelerating Medical Treatments for Serious Mental Illness. The order directs the Food and Drug Administration to issue Commissioner’s National Priority Vouchers to appropriate psychedelic candidates holding Breakthrough Therapy designation. It also instructs the Drug Enforcement Administration and other agencies to reduce research barriers. It creates a Right to Try pathway for eligible patients and commits $50 million to state-led research.
DT120 holds Breakthrough Therapy designation for generalised anxiety disorder. That places the candidate within the order’s stated criteria. It was not, however, among the first three voucher recipients on 24 April. Those went to Compass Pathways for psilocybin in treatment-resistant depression, the Usona Institute for psilocybin in major depression, and Transcend Therapeutics for methylone in post-traumatic stress disorder. The vouchers compress standard review timelines to roughly one to two months.
The order has not been universally welcomed. Bioethicists and regulatory scholars have questioned whether compressing review to one or two months is appropriate for a class of drugs the agency has limited experience assessing. They have also noted that the order approves no product and creates no enforceable rights for patients or providers.
Commercial capital has followed the policy shift. Eli Lilly’s acquisition of psychedelics developer AtaiBeckley, agreed on 16 July for up to $3.8 billion, comprises $2.8 billion upfront and up to $1 billion in contingent value rights. It is the clearest signal yet that large pharmaceutical companies view the class as investable.
Panorama and the road to filing
Attention now turns to Panorama, the second Phase 3 LSD anxiety study, with topline data expected in September. Its design differs in one significant respect. Participants are randomised two to one to two across 100 microgram DT120, 50 microgram DT120 and placebo. The low-dose arm is included specifically to make it harder for participants to identify their assignment. Functional unblinding is the central methodological objection to psychedelic trials. Panorama is Definium’s direct answer to it. The primary endpoint remains the 100 microgram comparison against placebo at week 12. A confirmatory result would give Definium two positive pivotal studies to support a filing.
Beyond Panorama, Definium is running Ascend, a second Phase 3 study in major depression. The company expects to enrol about 175 participants, with topline data in 2027. It expects to begin a late-stage post-traumatic stress disorder programme next year.
Two cautions are worth holding. First, no classic psychedelic has yet cleared the FDA. The agency rejected Lykos Therapeutics’ MDMA-assisted therapy for post-traumatic stress disorder in August 2024 and asked for a further Phase 3 study. The complete response letter, published in September 2025, cited a failure to collect abuse-related adverse events, alongside unreported events identified during inspections, a lack of durability data, and high rates of prior MDMA use among participants that the agency said suggested possible selection bias. Functional unblinding had been a prominent concern at the advisory committee that preceded the decision. Second, scheduling remains unresolved. Lysergide is a Schedule I controlled substance. Rescheduling by the Drug Enforcement Administration would be required before any approved product could be marketed. Neither the LSD anxiety result nor the executive order removes those hurdles.
