Amylyx Avexitide Wins Phase 3: 55% Fewer Hypoglycemic Events, Filing at Stake

A company that pulled its only product off the market two years ago just posted a Phase 3 win with a drug it bought out of somebody else's bankruptcy — and the stock went up fifty-four percent.

BioMarin is buying a molecule and leaving the entire company behind. Leo Pharma is paying four hundred thirty-five million for a drug already sitting at the FDA. And EyePoint lost sixty-seven percent on a trial it says it actually won. Let's get into it.

Welcome to The Pharma Closeout for Tuesday, August 18th. I'm Alex Mercer.

And I'm Maya Patel. Unusually heavy data day for the middle of August.

The number is fifty-five percent. That's the reduction in serious-to-severe hypoglycemic events versus placebo for Amylyx's avexitide, primary endpoint of the Phase 3 Lucidity trial, reported this morning. Shares closed up fifty-four percent — a three-year high per Investor's Business Daily. We've been tracking this readout since June. It landed.

In a condition with nothing approved. Say the setup, because the setup is the story.

The setup is that this company had no business winning anything. Spring of 2024, Amylyx pulls Relyvrio off the market after the confirmatory Phase 3 came back no better than placebo on a fatal disease. That was their only product. So they go shopping in Eiger BioPharmaceuticals' bankruptcy proceedings and buy avexitide for thirty-five million dollars. The co-CEOs told Fierce their entire thesis was that Eiger had already done the mid-stage work, so one clean late-stage trial would be enough to file. That's not a pipeline strategy. That's a company betting its existence on a discount rack.

And they went the opposite direction on mechanism, which nobody wants to talk about because it's inconvenient. Everyone in this industry is chasing GLP-1 agonists. Avexitide blocks the receptor. Now watch what that does to the safety table — no severe adverse events, most common were diarrhea and injection-site reactions, and no changes in body weight.

Why is the weight line the one you're circling?

Because these are post-bariatric patients. If you hand them a drug that reverses what the surgery accomplished, you have solved the hypoglycemia and destroyed the product. Fifty-five percent wouldn't have saved it. That single line in the safety data is what makes this filable, and it's the thing a topline headline never mentions.

So the turnaround narrative writes itself. Distressed asset, no competition, filing by year end.

That's where I'd narrow it. Seventy-eight patients over sixteen weeks, in a condition patients carry for decades. I'm not questioning the win — every secondary endpoint hit too. But the unanswered question was never "does it work." It's how many of these patients are ever identified. Post-bariatric hypoglycemia affects roughly one-tenth to one-third of people who have the procedure, per research cited by the Cleveland Clinic Journal of Medicine — and almost none of them have it written down anywhere.

You can watch the street split on exactly that assumption. Mizuho models risk-adjusted revenue at one point three billion by 2040. TD Cowen's Joseph Thome models peak US sales at one and a half billion, and writes that the figure "may prove conservative as diagnosis and awareness improve." Same molecule, same data — the entire spread between those two models is a diagnosis-rate assumption.

Which means the risk migrated. It's not a molecule risk anymore and it isn't a competitive risk. It's a case-finding problem, and case-finding is a medical education budget, not a clinical one.

Here's the competitive read. This is a first-mover position in a rare endocrine indication where the payer conversation is about as clean as it gets — the company says every one of these events is a medical emergency, and endocrinologists on the steering committee reacted accordingly. But the transferable insight is the sourcing. Thirty-five million dollars in a bankruptcy auction to a two-point-three billion dollar market cap. If you sit in business development and you have been treating distressed asset auctions as scrap, someone is going to put this slide in front of you this quarter.

And the thread that's still open is whether the mechanism travels. Amylyx told Fierce it's looking at other procedures that change how the body handles nutrients — gastric and esophageal cancer surgery, Nissen fundoplication — and specifically flagged Japan, where gastric cancer surgery volume is high. If it works across those populations, the diagnosis ceiling I just described lifts substantially. None of that is in this filing, and it's the difference between a good rare disease drug and a franchise.

Shifting to the deal side — and this one has a structure I want to spend a minute on. BioMarin is acquiring Alesta Therapeutics: two hundred seventy-five million upfront, up to two hundred fifteen million in development and regulatory milestones, funded with cash on hand, closing this quarter. The asset is ALE1, an oral small molecule for hypophosphatasia, currently in a Phase 1/2a study. And immediately before close, Alesta spins out every non-ALE1 asset into a new entity — and every Alesta employee goes with it.

Nobody joins BioMarin. They bought a molecule and declined the company.

Which is a deliberate choice, and there's a precedent behind it. Let's bring in Marcus Webb on this one.

And the market they're aiming at is genuinely vacant, which is the part that justifies paying for a Phase 1 asset. Strensiq is approved only in infantile- and juvenile-onset disease, and AstraZeneca's next-generation enzyme replacement missed its primary endpoint in patients twelve and older. So adult hypophosphatasia is sitting open.

Open, but is oral enough to win it?

On dosing, it should be. Strensiq is subcutaneous three to six times a week. A pill against that is not a marginal convenience argument — it's a different disease experience. But BioMarin is buying a Phase 1/2a safety and pharmacokinetics study in healthy volunteers and adults. There is no efficacy story here yet. They're paying for the shape of the market, not for evidence, and they know it.

Also on the deal side, Leo Pharma is licensing worldwide rights to dersimelagon from Tanabe Pharma for up to four hundred thirty-five million in combined upfront and near-term milestones. Oral MC1R agonist for erythropoietic protoporphyria and X-linked protoporphyria. Tanabe filed the NDA in June — so Leo is buying an asset already under FDA review. Clinical risk close to zero, regulatory timing as the value trigger.

And the differentiation is delivery, not biology. The only approved therapy in EPP is Scenesse — same MC1R mechanism, but it's an implant every eight weeks. In the Phase 3 INSPIRE study, Tanabe reported dersimelagon extended daily sunlight exposure before prodromal symptoms by a placebo-adjusted twenty-three point one nine minutes across weeks twelve through sixteen. Leo says that if approved it would be the first oral therapy in either condition.

So the pattern in Leo's dealmaking is now unmistakable. 'A hundred and five million range for Spevigo, fifty million for Replay's platform, four hundred thirty-five million here for a filed NDA.' They aren't buying science risk — they're buying regulatory proximity. For a mid-size dermatology specialist without a big pharma balance sheet, that's the only version of this strategy that survives a miss.

On the regulatory side, the one worth arguing about is EyePoint — and we flagged this readout on July 30th. Monday, Duravyu missed its primary endpoint against Eylea on best-corrected visual acuity, and the stock crashed sixty-seven percent. But read the actual failure. Nine patients out of two hundred eleven on Duravyu had vision loss unrelated to wet AMD. Zero in the Eylea arm. CEO Jay Duker called that highly unusual against historical rates of three to five percent in prior Phase 3 trials — including Eylea's own.

Zero in the comparator arm is the part that doesn't parse.

It doesn't, and that asymmetry is the entire company's argument. Strip those nine out in an ad hoc analysis and Duravyu is non-inferior. Citi's Yigal Nochomovitz wrote that there's precedent in ophthalmology for one failed trial and one successful trial to be enough for approval.

I'll take the other side of that for a second. An ad hoc analysis that rescues a missed primary endpoint is the oldest move in the book, and investors priced it accordingly.

Investors priced the headline. The regulator will price the mechanism of failure, and those aren't the same question. If the nine patients lost vision for reasons unrelated to the disease and unrelated to the implant, that's noise. If it's the implant, Lucia reproduces it this fall and there is no path. That's why the secondary data matters — injection burden dropped forty-two percent, two fewer injections on average. The durability thesis worked. The vision endpoint is what broke.

And the read-through for the incumbent is that Eylea didn't need rescuing. Quarterly sales dropped below a billion dollars for the first time in eight years — but that's biosimilars eating it, not durability competition. Ocular Therapeutix moved up as much as eleven percent Monday on somebody else's bad day.

Also worth noting on the regulatory side — Moderna's mFlusiva. BioPharma Dive reported Monday that this month's approval makes it the first approved mRNA flu vaccine in the US, using accelerated approval for adults sixty-five and older. Relative vaccine efficacy was twenty-six point six percent against a standard-dose shot.

So the approval isn't the hard part.

The approval was the easy part. The platform's advantage is speed — you could theoretically pick strains two months out instead of six to nine, which improves the match. But that only exists if the WHO and FDA change the strain-selection framework, and nobody has committed to doing that. Until they do, Moderna owns a faster factory pointed at a recommendation set on someone else's calendar. The technology is ahead of the regulation, and the regulation is what gets paid.

Looking ahead. Amylyx files with the FDA by year end, and with breakthrough therapy designation in hand the company says it anticipates priority review — that's the next real catalyst. BioMarin closes Alesta by the end of September and updates full-year guidance after. And for Leo, dersimelagon's FDA decision is now the whole value trigger.

Two clinical items I'd add. EyePoint's Lucia readout this fall carries the entire Duravyu thesis, and there's no hedging it. And on AstraZeneca — the volrustomig lung trial we covered yesterday died at an interim survival check, but three more Phase 3s continue in mesothelioma, cervical, and head and neck cancer. Watch tolerability, not efficacy. A third of enrollees in the first-in-human study stopped because of side effects, and Leerink has been asking publicly whether a therapeutic window exists at all. If that CTLA-4 toxicity shows up again outside lung, this stops being one failed trial and becomes a class problem inside a pipeline AstraZeneca is counting on for one of its sales targets.

Lucia is the one I'll be sitting on. If those nine patients were bad luck, EyePoint gets a second act. If they weren't, we learn something uncomfortable about putting an implant in an eye. Have a good evening.

And that is your Pharma Closeout for Tuesday, August 18th — Amylyx turning a thirty-five-million-dollar bankruptcy asset into a Phase 3 win, BioMarin buying the molecule and walking away from the company, Leo paying up for a filed NDA, and EyePoint arguing with its own primary endpoint. This is the version of this industry I find hardest to look away from. Follow the show on Spotify, Apple Podcasts, or wherever you listen — tomorrow's briefing lands on its own.