UPDATE: Amylyx Pharmaceuticals (AMLX) - The Coin Landed Heads

I told you yesterday to set your alarm for 8:00 a.m. ET. Here is what came out of it: a 55% reduction in the composite rate of Level 2 and Level 3 hypoglycemic events versus placebo, rate ratio 0.45 (95% CI 0.32–0.63), p=0.000003, in a randomized, double-blind, placebo-controlled trial of 78 participants. All three secondary endpoints hit. No treatment-related serious adverse events.

The Event. The lock-and-key pharmacology I have been describing since the deep dive survived contact with a placebo arm. Avexitide is now the first drug to show a statistically significant, clinically meaningful reduction in real-world hypoglycemic events in post-bariatric hypoglycemia (PBH) under blinded conditions.

The Reaction. AMLX closed at $35.11, up +63.8% from yesterday’s $21.43. That is roughly what a clean pivotal win looks like on a name where the market had priced something close to a coin flip.

The Catch. Within hours of the data, Amylyx launched a $350.0M underwritten common stock offering (plus a $52.5M underwriters’ option) at a stock price that had just doubled off its August lows. The clinical risk is retired. The dilution I said wasn’t coming before the readout arrived the instant the readout was in hand.

The primary endpoint, precisely. n=78, randomized 3:2 to 90 mg avexitide subcutaneously once daily or placebo, 21 U.S. sites, 16-week double-blind period. Primary: reduction in the composite rate of Level 2 (glucose <54 mg/dL) and Level 3 (events requiring assistance from another person) hypoglycemic events through Week 16. Result: 55% reduction, rate ratio 0.45, 95% CI 0.32–0.63, p=0.000003. Greater than 90% of participants completed the double-blind period, and all eligible participants rolled into the 32-week open-label extension.

The implication: This lands squarely inside the 40–60% placebo-adjusted range I said yesterday would constitute a clean win, not the murky 25–35% “statistically approvable, commercially ambiguous” zone. The confidence interval is the number I care about most: the upper bound of 0.63 means even the pessimistic end of the estimate is a 37% event reduction. In a trial this small, a wide CI crossing toward 1.0 would have been the tell that the headline percentage was carried by a handful of outlier patients. It isn’t.

All secondary endpoints hit. Significant reductions in Level 2 events by self-monitored blood glucose (SMBG), Level 2 events by continuous glucose monitoring (CGM), and independently adjudicated Level 3 events.

The implication: This is the internal-consistency check, and it is the part that separates a real drug from a lucky primary. SMBG is patient-recorded fingerstick data — subject to reporting behavior. CGM is a sensor strapped to the arm recording glucose whether the patient is paying attention or not. When a patient-dependent measure and a machine-independent measure move in the same direction by similar magnitudes, the effect is almost certainly real biology rather than measurement artifact. And the Level 3 events — the severe, neuroglycopenic ones involving confusion, loss of consciousness, seizures — were independently adjudicated, meaning a blinded committee, not the investigator, classified them. That is the endpoint the FDA and payers will care about most, and it is the hardest one to massage.

The run-in trap didn’t spring. I flagged the three-week run-in period yesterday as the single most plausible mechanism by which a genuinely effective drug posts a disappointing trial — stabilize diet and behavior before randomization, suppress the placebo arm’s event rate, compress the delta. Entry required at least three hypoglycemic events during the 21-day run-in despite adherence to consistent dietary management.

The implication: The enrichment worked in the drug’s favor rather than against it. By requiring documented events on optimized diet, Amylyx selected patients whose hypoglycemia is refractory to the current standard of care — which is nutrition therapy, full stop. Those are exactly the patients who would be prescribed the drug commercially. A 55% reduction on top of stabilized diet is a more useful number than a 55% reduction in an unselected population.

Safety was clean. Adverse events generally mild or moderate. No treatment-related serious adverse events. Most common AEs: diarrhea, injection site erythema, injection site bruising. Notably, no changes in body weight in either arm over 16 weeks.

The implication: The weight-neutrality line matters more than it looks. Blocking the GLP-1 receptor in a post-bariatric population raises the obvious mechanistic worry — does antagonizing the satiety pathway un-do the surgery and put the weight back on? Sixteen weeks says no. That is short, and the open-label extension will need to confirm it, but it removes the most intuitive objection a prescribing bariatric surgeon would raise.

The $350M raise. Same day. Leerink, Morgan Stanley, Guggenheim, LifeSci as bookrunners. Stated use of proceeds: pre-commercial activities for avexitide including securing additional manufacturing capacity, R&D, and working capital. The 424B5 also discloses the company is “currently in discussions with contract manufacturing organizations to secure commercial supply of avexitide, if approved,” with no definitive agreements signed.

The implication: Read those two sentences together. The CMC risk I graded High in the original deep dive — outsourced peptide manufacturing, cold chain, a global GLP-1 capacity crunch — is what this money is for. Amylyx has $250.8M and roughly $29M a quarter of burn; it did not need $350M to survive. It needed $350M to buy peptide manufacturing slots in a market where everyone making GLP-1 drugs is fighting for the same fill-finish lines. The company is converting the readout into hard capacity while its cost of capital is at a multi-year low. That is opportunistic and, frankly, correct. It is still dilution: roughly 10M new shares at the $35.11 close against 111.4M outstanding, so call it ~9% before the greenshoe.

The regulatory path. NDA submission to the FDA by the end of 2026. Breakthrough Therapy Designation already in hand for PBH, Orphan Drug Designation for hyperinsulinemic hypoglycemia, positioned for NCE exclusivity. Company reiterates anticipated commercial launch in 2027, if approved. Data to be presented at an upcoming medical meeting.

The implication: Breakthrough designation buys rolling review and intensive FDA guidance, and orphan status carries 7 years of U.S. market exclusivity from approval — which is the answer to the patent problem I spent a section on in the deep dive, where the earliest Stanford in-licensed patents began expiring in May 2026 and the University of Pennsylvania family goes as early as January 2028. Regulatory exclusivity, not the patent estate, is the near-term moat. The liquid formulation patents to November 21, 2037 remain the long-dated backstop.