On 19 August a European regulator authorised a Lilly Phase 3 trial in depression. Not a site list and not an intention: a decision, recorded in the EU Clinical Trials Information System as trial 2025-522310-24-00, sponsor Eli Lilly & Co., status Authorised, recruitment pending.

That is RENEW-MDD-2, the second Phase 3 of brenipatide in major depressive disorder. It surfaced on ClinicalTrials.gov the next day as NCT07775300, and the registry entry is the weaker of the two facts. CTIS is a regulatory system rather than a noticeboard. An entry there means a member state read a protocol and said yes.

Search CTIS for brenipatide and five authorised Lilly trials come back:

Now the drugs Lilly bought for $3.8 billion. We searched CTIS for each of them.

BPL-003 returns one trial. 2024-513457-70-00, the intranasal dose-finding Phase 2b in treatment-resistant depression with an open-label extension, authorised 24 September 2024, sites in Poland, Spain and Germany. Status: Ended. It finished on 26 June 2025.

VLS-01 returns nothing. Zero results, on a search that returns results for everything else we asked it.

So the comparison is not presence against absence. It is direction. Lilly's own depression and addiction programmes have four authorisations dated this year, two of them Phase 3s in alcohol use disorder now running in Germany and Belgium, and two more decisions inside our seven-day window. The psychedelic assets it acquired in July have one European trial between them, at Phase 2b, and it stopped fourteen months ago.

Alcohol use disorder is where this bites hardest. Lilly has two Phase 3 trials in that indication authorised across Germany and Belgium, one recruiting since February. Beckley Psytech has one BPL-003 alcohol trial, NCT05674929, Phase 2, United Kingdom only, completed. Same company group since July. Same indication. One programme is in the EU regulatory system and moving; the other is neither.

We are not asserting anything about what brenipatide is. We are reporting what the register calls things. CTIS returns five authorised brenipatide trials sponsored by Lilly, and it returns the acquired assets separately, under their own names, on their own trial numbers, with Beckley Psytech rather than Lilly as sponsor of record. Lilly's own acquisition release uses those same names throughout and never mentions brenipatide: it describes "BPL-003 (mebufotenin benzoate)" and "VLS-01, the second most advanced program of the pipeline", and it keys every contingent value right to them, "$1.00 per share upon initiation of a Phase 3 clinical trial of VLS-01", "$0.50 per share upon U.S. regulatory approval and DEA rescheduling of BPL-003", "$1.00 per share upon U.S. regulatory approval and DEA rescheduling of VLS-01".

So the five brenipatide authorisations are not the acquired programmes under a new label, and none of this touches either prong of the Issue 7 call, which turns on whether BPL-003 or VLS-01 gets a registered Phase 3 with an EU-member site, or a disclosed EMA pathway, within twelve months of the deal closing. Nothing here falsifies that call and nothing here scores it. On the registry side, re-swept on 24 August across four sponsor and intervention queries, there is still no Phase 3 for either drug anywhere on earth.

What this week removes is an excuse. The weakest counter to Issue 7 was always that Lilly does not do European regulatory work in psychiatry. It does, in Greece, Germany, Belgium and Spain, across depression, alcohol use disorder and irritable bowel syndrome, with decisions dated this month and last January.

Three things we had to correct in our own reporting this week, and we would rather show you all three than quietly ship the version that survived.

We nearly published the claim that the acquired assets are absent from the EU system. They are not. Every ClinicalTrials.gov record returned for BPL-003, mebufotenin and VLS-01 lacks a CTIS number, and it would have been easy to report that as a European blank. It is not one. The CTIS identifier on ClinicalTrials.gov is filled in by sponsors, not regulators, and it is filled in unreliably. BPL-003's Phase 2b carries no CTIS number on the registry and has a full CTIS record. Searching the registry told us the opposite of what searching the regulator told us, and the regulator is right.

Compass is in that system too, at a scale worth stating plainly. COMP360 returns 2023-505268-12-00, COMP 006, a Phase III in treatment-resistant depression authorised 9 October 2023 and ongoing with recruitment ended, running in nine EU and EEA countries: France, Denmark, Sweden, Germany, Ireland, the Netherlands, Spain, the Czech Republic and Poland. That is a large European clinical footprint and we are not going to bury it.

It also does not contradict Issue 9, and the distinction is the one this whole beat runs on. A clinical trial authorisation lets you study a drug in a member state. A marketing authorisation application asks permission to sell it. Issue 9 said no European application is on file for COMP360, meaning the second kind, and that remains true: the EMA's list of applications under evaluation, extracted 4 August, contains no COMP360 and no psilocybin at all. Compass has spent years running trials across nine European countries and has not asked a single European regulator for permission to sell the result. Read one way that is stranger than an absence would have been.

The bipolar comparison is not like for like. RENEW-Bipolar-1 carries no European sites, and the record confirms it: Argentina, Brazil, China, India, Japan, Mexico, Puerto Rico, the United States, and no CTIS number. But it is a Phase 2, and Phase 2 programmes routinely run narrower geographies than Phase 3s. It is consistent with Lilly deciding per programme. It does not prove it.

One detail the CTIS records settle in passing. Both brenipatide depression pivotals are described in their official titles as adjunctive treatment, given on top of a stable standard-of-care regimen rather than after a washout. Readers who followed the Issue 5 lead will recognise the fault line. We flag the parallel and stop there. Two programmes can land on the same design for unrelated reasons, and we have read nothing from Lilly explaining why this one did.

Panorama has not reported. Definium Therapeutics has published nothing on the trial since guiding it to September. Its press release page runs 17 August (employee inducement grants), 12 August (the Voyage topline we covered last issue), 6 August (Q2 results), 3 August (inducement grants). No Panorama release and nothing dated September. The guidance was September, so an absence in August is what the guidance predicts. We flag it because Issue 4 scores on that number and the number is close.

AtaiBeckley's press release page has been silent for five weeks. The most recent item on it is the Lilly acquisition announcement of 16 July, with no August entry of any kind, so no first-patient-dosed announcement for the BPL-003 Phase 3 programme has appeared there. That page is the only channel we searched. The Issue 3 call resolves on a report rather than a registry record, so a dosing statement made in an SEC filing, at a conference or in an investor deck would count and would not show up in what we checked. The call resolves on 30 September, thirty-six days from today.

Compass is supplying a VA-sponsored psilocybin trial. Announced 6 August, so outside our window and not covered in Issue 10. The release describes PIVOT, "Psilocybin Intervention for Veterans Overcoming Treatment-Resistant Depression": a multi-site, double-blind randomised controlled trial in veterans with treatment-resistant depression, with or without concurrent PTSD, primary endpoint MADRS, lead site the Birmingham VA Health Care System in Alabama with Tuscaloosa, Portland, Seattle and Philadelphia. MADRS is the depression scale almost every trial in this field is scored on. It runs from 0 to 60, higher is worse, and the useful thing is the yardstick rather than the range: the drug-versus-placebo gaps we have reported on it run from 3.6 points in Compass's placebo-controlled pivotal to 8.1 in Definium's Emerge. That narrow band is where this field's arguments happen. Compass's stated role is narrow. It donates the study drug and supports investigator training with materials, guidance and consultation, and its own forward-looking statements call this "this investigational study sponsored by the U.S. Department of Veterans Affairs". The trial "will compare outcomes between two different doses of COMP360 psilocybin", so there is no inert comparator, which is the design question Issue 6 raised about Compass's own durability data. Europe appears in the release exactly where Issue 9 found it in the Q2 results: the London headquarters and the UK ILAP designation, both boilerplate, no filing named.

Sweden has opened a psilocybin trial in anorexia nervosa. NCT07169747, sponsored by Region Skåne, Phase 2, recruiting, registry updated 14 August. Two psilocybin doses with psychological support in young adults. It appears in CTIS as 2024-515163-63-00, authorised 18 June 2025. We have read those two records and not a protocol or any statement from the sponsor.

And while we were in CTIS, the thing worth noticing was who else is in it. A search for psilocybin returns thirty authorised or decided EU trials. We read the first twenty by decision date and every sponsor among them is a university, a hospital or a research institute, across Denmark, France, Italy, Spain, Ireland, Sweden, Austria, Belgium, the Netherlands, Czechia and Portugal, in cocaine addiction, gambling, fibromyalgia, palliative care, anorexia and alcohol. We did not read the remaining ten, so we are not telling you the commercial count is zero. Across twenty of thirty, Europe's authorised psilocybin research is being done by academics, in the same system where Lilly has five trials for a drug it developed itself.

Zero, for the fourth consecutive reading. No newly registered interventional trial with a site in the EU, EEA, UK or Switzerland was first posted between 18 and 24 August. Two records matched our fourteen drug terms in that window, both American: one observational study and one interventional psilocybin trial in methamphetamine use disorder with no European site.

The base rate we built in Issue 9 is the reason this stays a number rather than becoming a story. Ninety-one of 135 weeks are zero, runs of three or more have happened sixteen times, and August runs at roughly half the annual rate. Four in a row is unremarkable on a series that behaves like this one.

One note on method, because it decided the number. Our monitoring sweeps a broader term list to find stories than the fourteen-term list that produces this count, and this week the two disagree: the broad sweep surfaced the Lilly brenipatide registration as a new European trial and the count list does not, because brenipatide is not a psychedelic. The published figure is zero. The count list is the one the base rate was computed on, and a number that changes depending on which list you reach for is not a number.

A statement about our own coverage first, as in Issue 10. We ran the full register audit this week and it came back split. Four registers were queried directly and returned clean negatives with working controls. Two failed on their control queries, which means we have no answer for them rather than an empty one, and we are not going to present the second group as if it belonged to the first.

Verified clean. The EMA's medicines register returns nothing for psilocybin, MDMA, midomafetamine or lysergide, with esketamine returning results as a control. Its list of applications under evaluation, extracted 4 August, carries 77 applications with substance names and indications, includes psychiatry, and contains no psychedelic and no COMP360. The MHRA products database returns, verbatim, "There are no search results for psilocybin", against seventeen results for esketamine. Swissmedic's list of authorised narcotic-containing human medicines carries no psilocybin, MDMA or lysergide, and does return methylphenidate, morphine, esketamine and a cannabis extract, which is how we know the file is what it claims to be.

Germany, and this one is a genuine upgrade. Previous checks read the G-BA's current list of thirty live benefit assessments. This week we swept the full archive of 1,370 procedures across the substance letters those five names begin with. No psilocybin, psilocin, MDMA or lysergide procedure appears anywhere in that register. Esketamine returns three procedures, which is how we know the query works. The sweep filtered on four letters, so it is a claim about those substances rather than about every entry in the file.

Not verified, and we are saying so. NICE's guidance-in-development register returned HTTP 403 on every query including its controls, and Zorginstituut Nederland's query API returned HTTP 500 on every query including its control. Both were retried and failed again. These are outages on our side of the glass, not findings. Issue 9 reported all three of those registers closed with working controls, and that was true on 3 August. It is not something we can restate as current, and the Dutch prong in particular is the one the Issue 8 call turns on.

Nothing moved between 18 and 24 August in the four registers we could read. We cannot say that about NICE's development register or the Dutch one, and we are not going to imply it by rounding four up to six.

Last issue's call is not yet scoreable and no scoring action is owed. The Issue 10 call, on whether a pharmaceutical company is ever named as responsible party on a BfArM compassionate use entry for psilocybin or MDMA, resolves 31 December 2027. It is open and not falsified.

One piece of housekeeping on our own scoring rule. On 19 August we posted the Issue 4 scoring rule to the company page ahead of Panorama, and we owed you a check on whether that post had loosened a rule Issue 10 published as binary. We have read the posted copy. It did not. What went out was this:

"This is a company release, not a peer-reviewed paper, and 5.4 points is small enough that trial-level noise could swallow it. If we miss inside noise, we take the miss and say the call was badly specified. We do not widen the rule after the fact."

That is tighter than the printed rule, not looser. No amendment was made and no correction is owed. The rule stands as Issue 10 set it: on the placebo-adjusted difference between the 100 microgram arm and placebo on the primary endpoint at the primary timepoint, under 5.4 points we are right, at 5.4 or above we are wrong, and smaller means smaller in magnitude. Placebo-adjusted is doing real work in that sentence. It means the gap between the two arms, not how far either arm improved on its own. Patients on placebo in these trials get better too, sometimes a lot better, and the only number that tells you anything about the drug is the distance between the two lines.

This issue's call. Before 31 December 2026, the ClinicalTrials.gov record for NCT07775300 will show overall status RECRUITING and still carry at least one of Germany, Greece, Poland, Spain or the United Kingdom in its location list.

The falsifier, up front, as two conditions that both have to hold: if on 31 December the record's overall status is anything other than RECRUITING, or every one of those five countries has been dropped from its locations, we are wrong. We score on the record's overall status rather than per-site status because ClinicalTrials.gov does not report site-level recruitment consistently, and a rule you cannot read off the page is not a rule.

The reasoning. Our argument this week is that Lilly's European regulatory work is real intent rather than paperwork. That should cost us something if it is wrong, and this is the cheapest honest way to make it cost. Registered sites do not always activate, trials get cut, footprints get trimmed, and a country can sit on a record for a year without a patient being screened there. We are betting it activates. If it does not, our reading of that authorisation was too generous and we will say so.

We have deliberately not built this on the BPL-003 Phase 3 posting. Three of our open calls already resolve off that single event, Issue 9 named that concentration as a structural weakness in print, and a fourth would make the ledger worse.

Eight carried in; this issue makes the ninth.

If something here is wrong, please tell me, and I will correct it openly.

Thank you so much for reading.

Dhruv Shekhawat
Founder, Psychedelic Brief

Sources