Lilly’s depression trial went from one approval to four in seven days.

Correction, 8 September: we printed one new European trial in five weeks. CTIS carries three authorisations in the same period. Details in Issue 13.

Lead Story

Lilly's brenipatide programme now holds regulatory decisions in five EU member states, across four indications. Four of those decisions, all on the Phase 3 depression trial, landed inside seven days. The register moved.

Last week we read that site list as paperwork. The register has since answered: Greece on 19 August, Spain and Germany on 24 August, Poland on 25 August, four national approvals on one trial in seven days. That is not a company parking filings. It is a company running trials in Europe, for evidence it intends to spend somewhere else.

Update, on our own reporting. Issue 11 stated that CTIS listed Greece alone for the Phase 3 depression trial. The record now shows Germany, Greece, Poland and Spain. Two of those decisions are dated before we published; the record itself, by CTIS's own account, became public a day after. We can verify what the register says, not when it showed it.

Half of us still believes the old read. Decisions are not patients, and until someone is dosed at one of those 36 sites, a five-country footprint is still reversible at the stroke of a pen. The other half has watched four regulators say yes in one week and knows paperwork does not move like that.

---

Context

The alcohol-use trials collected their national approvals in January and February. The depression trial collected four inside seven days, 19 to 25 August. Applications sit with regulators for months, so approvals landing that close together mean the filings went in long before anyone was watching, ours included.

A gloss, because the register's language hides the point. A "member state concerned" is a country whose regulator has approved the application to run the trial there. Each row in the table below is a national yes, already given.

---

Data Point

One. The number of new European psychedelic trial registrations in the past five weeks: a 12-patient open-label psilocybin pilot in post-traumatic stress disorder at ARQ National Psychotrauma Centre, Oegstgeest, registered 28 August. Open-label means everyone knows who gets the drug; it answers feasibility, not efficacy. The run of silence is over, and it was broken by an academic, not a company.

---

Regulator Watch

We owed two checks from Issue 11.

NICE's guidance-in-development register returned HTTP 403 on every query, including its esketamine control, and the body it returns is a NICE page headed "Temporary Service Interruption". That is a broken query, not a zero, for the second issue running. NICE's main search does work, and it is clean: no results for psilocybin against six for esketamine. That is a narrower claim about a different register and it does not close the first one.

Zorginstituut Nederland's query API returned HTTP 500 with an empty body on every term including its control. Its web search route answers, but it returns a byte-identical page for esketamine, for psilocybine and for a nonsense word, so it is a shell that ignores the query and it is not evidence either.

A broken query and a genuine zero are different findings, and we print which is which.

---

The Call

Before 31 December 2026, at least one additional EU member state will appear across Lilly's five brenipatide CTIS trials, beyond the current five: Belgium, Germany, Greece, Poland and Spain.

Falsifier: on 31 December 2026 the union of member states concerned across those five trial records is five or fewer. Scored against the register via our trials table, in the first issue after resolution.

Our reasoning, labelled as opinion: trial engines expand. If Europe is where the evidence gets made, the footprint grows; if the old paperwork read was right after all, it freezes where it stands. Either way the register will say so, and we have put the date on it.

---

Second item: the number that appears in no filing

The atai and Beckley combination closed 5 November 2025. The 8 September vote is the Lilly merger, a different deal.

The figure attached to the atai and Beckley combination in nearly every report, roughly $390 million, does not appear as a stated term in any of the transaction's filings we examined. That includes our own H1 review, which carried the number the way everyone else did: unsourced.

The consideration was entirely in shares: 103,000,066 atai shares issued at closing, against 105,044,902 specified in the purchase agreement. No cash component, no fixed price. What Beckley holders received was worth whatever atai traded at on the day, which means the deal's most-quoted number was never a term of the deal. It was a snapshot of a share price on the day someone first multiplied it out.

Every outlet carried the figure; none of the coverage we reviewed pointed to a filing for it, because it is not a term in any we examined. We know, because we were one of them. Worth remembering every time this sector quotes a deal value: ask what document the number lives in. Usually the answer is nobody's.

---

The record

Nine calls carried in; this issue makes the tenth. One resolved, one void, none falsified. Issue 11's open call on NCT07775300 is unchanged: overall status remains NOT_YET_RECRUITING as of 31 August 2026, with Germany, Greece, Poland, Spain and the United Kingdom all still on the record.

If something here is wrong, tell me, and I will correct it in the open.

Thank you so much for reading.

Dhruv Shekhawat
Founder, Psychedelic Brief