This week, Amylyx Pharmaceuticals reported that its experimental drug avexitide cut the rate of serious low blood sugar events by 55 percent in a placebo-controlled Phase 3 trial in patients with post-bariatric hypoglycemia. The study was designed to detect a 35 percent improvement; it found 55, with a p-value of 0.000003. No treatment for the condition is approved anywhere, so the result, covered by STAT News, opens the door to the first therapy for the condition, with an FDA filing planned by year's end.

The condition deserves the attention those numbers just earned it. Post-bariatric hypoglycemia is not a low-blood-sugar dip you snack your way through. It is a crash of blood glucose one to four hours after eating, in people who had gastric bypass or sleeve surgery, and in its severe form it can mean confusion, loss of consciousness, or a seizure. The trial counted exactly those events as its endpoint, and the timing of them is the story. The bill for bariatric surgery was always there. It was deferred, not created.

The success was real, and so was the delay

Post-bariatric hypoglycemia does not announce itself in the recovery room. It typically appears one to three years after surgery, when the dramatic phase of the operation's effects, the weight loss, the resolved diabetes, is complete and the patient has long since been discharged from the surgery's success story. A Swedish registry study of more than 5,000 patients found that hypoglycemia severe enough to require hospitalization arrived on average 2.7 years after the operation, and sometimes more than a decade later.

Estimates of how common it is vary with how it is measured. Prospective cohorts have diagnosed it in roughly a third of gastric bypass patients within the first few years, as reported in Surgery for Obesity and Related Diseases, while Amylyx's own estimate puts the U.S. total near 160,000, about 8 percent of people who had sleeve or bypass surgery. The gap between the two numbers is revealing: the condition is counted only when someone thinks to look, and for years nobody looked, because the successes were being counted and the clock had not run.

The operation did work. That is exactly the point. The surgery's cost was never absent, only parked, and it was parked in physiology from the first meal after the operation.

The same machinery that does the work is the machinery that crashes

The mechanism makes the deferral legible. After Roux-en-Y bypass, food arrives in the small intestine almost directly, and the gut, confronted with that sudden load, releases a surge of incretin hormones, above all GLP-1, as reviews of the condition's pathogenesis describe. GLP-1 normally amplifies insulin secretion in response to glucose. In these patients it amplifies too far: insulin keeps clearing glucose out of the bloodstream even after the meal's sugar is gone, and blood glucose drops into the 50s, or lower, one to four hours later. The exaggerated GLP-1 response is part of the same hormonal signature that helps explain why bypass improves glucose control in the first place, often before much weight is lost. The surgery's metabolic benefit and its most frightening late complication run on the same circuit.

Avexitide is a GLP-1 receptor antagonist. It does not undo the rerouting, the rapid emptying, or the weight loss. It blocks the signal the surgery turned up, damping the insulin overshoot and holding the glucose floor in place. In the LUCIDITY trial, 78 adults with the condition took 90 milligrams of the drug by daily injection or placebo for 16 weeks. The drug cut the rate of Level 2 events, blood glucose below 54 milligrams per deciliter, and Level 3 events, severe episodes in which a person needs help, by 55 percent versus placebo, and all secondary endpoints met. Side effects were mostly mild, with diarrhea and injection-site reactions the most common, there were no treatment-related serious adverse events, and neither group's weight changed. The drug treats the complication without cancelling the surgery.

The people the surgery helped most carry the higher risk

The risk of the complication is not spread evenly, and the pattern is the hard part of the story. Cohort studies find it more common after gastric bypass than sleeve, more common in people who were younger at surgery, and less common in people who had type 2 diabetes before the operation, as in this multicenter retrospective study. The metabolic failure the surgery is famous for fixing predicts a lower chance of this late complication. The patients with the cleanest post-operative glucose profile, the ones whose outcomes get written up as the surgery's victories, face the higher late risk. The bill lands on the success stories.

That is why the trial mattered to the people who ran it. The study's principal investigator put it plainly: "Preventing even one Level 2 or Level 3 event is medically meaningful." For a patient whose worst days are measured in episodes like that, one prevented event is not a rounding error in a data table. It is a dinner eaten without planning the crash around it.

The bill matured as the procedure counts fell

The surgery that carries this deferred cost is, at this moment, in decline. According to research presented at the American Society for Metabolic and Bariatric Surgery annual meeting, U.S. procedure counts fell from a 2022 peak of 230,207 to 177,297 in 2024, the first year below 200,000 since 2020, a drop of more than 20 percent. The researchers, led by Tyler Cohn of Loyola University Chicago, noted the decline coincided with the surge in GLP-1 weight-loss drugs, while making clear the study did not establish causation. Sleeve gastrectomy fell to 58 percent of cases, and gastric bypass, the procedure most associated with post-bariatric hypoglycemia, rose to its highest share in five years.

The timing is the thing. The patients now entering the complication window are the ones who had surgery in the peak years. The people avexitide would serve, if approved, are not the people scheduling surgery this year. They are the people who already had it, and that population matures on a schedule set years ago, nearly independent of current volumes. Surgery counts fall; the cohort carrying the deferred cost only ages forward. The decline in the procedure and the arrival of the bill are the same story, told from two ends of a clock. As Cohn also said, "Millions more are living with obesity and are not receiving treatment at all."

A company that has been billed before

Amylyx arrives at this moment with its own history of deferred costs. In 2022 the FDA approved the company's ALS drug Relyvrio on the strength of a 127-patient Phase 2 trial, after the founders had publicly promised to pull the drug voluntarily if the confirmatory study failed. The ALS Association backed that trade, arguing that patients facing a fatal disease were willing to accept the risk of a therapy whose proof was incomplete. Critics, including Public Citizen, argued the drug should never have reached patients without demonstrated benefit. This analysis takes no position on that fight. The structural point is the same one: the cost of the bet was deferred by an explicit promise, and in March 2024, when the 664-patient PHOENIX trial failed, the promise was kept. Amylyx withdrew Relyvrio the next month and cut about 70 percent of its staff.

Three months later, in July 2024, the company bought avexitide for $35.1 million at the bankruptcy auction of Eiger BioPharmaceuticals, as Pharmaceutical Executive reported. The drug that may become the first approved treatment for a complication of bariatric surgery was itself rescued from someone else's deferred bill. Costs and assets alike get parked, and both come due in the manner of their parking: Relyvrio's cost was parked in a promise and came due in a failed trial, while the surgery's cost was parked in the body and came due years later, in an emergency room.

A trial measures sixteen weeks; the condition runs for years

The LUCIDITY result is a 16-week measurement of a condition with a years-long clock, and the honest reading of the trial includes its limits. Seventy-eight patients is a small foundation on which to build a disease-modifying claim; the open-label extension is still running; the long-term safety of daily injections in this population is not yet established. The FDA's Breakthrough Therapy and Orphan Drug designations and the planned end-of-2026 filing suggest the company sees a clear path, but the condition itself, which outruns the trial's window, argues for humility about what one result settles.

What the result does settle is the point of accounting. For the first time, the deferred cost of bariatric surgery has a measurable price: a 55 percent reduction in the events that dominate patients' lives, at a tolerable side-effect profile, without reversing the surgery's benefits. The cost was never absent. It was parked, and the parking was the success itself, the years of better glucose control, the resolved diabetes, the weight lost and kept off, all of it real, and all of it coexisting with a bill that was accruing the whole time. Success that is measured only in the first year will keep deferring costs into the fifth, the tenth, the fifteenth. The patients the surgery served best are the ones now owed the treatment. What changed this week is not that the bill became payable. It was always payable. What changed is that there is finally something to pay it with.

Primary sources

  1. STAT News, reporting by Elaine Chen, for the lede facts on the LUCIDITY result, including the trial's 35 percent powering assumption and the description of severe events.
  2. Amylyx Pharmaceuticals' topline press release for the trial design, efficacy and safety data, regulatory designations, and the company's prevalence estimate.
  3. The American Society for Metabolic and Bariatric Surgery for 2020 to 2024 U.S. procedure counts from research led by Tyler Cohn of Loyola University Chicago, including the noted association with GLP-1 drug use.
  4. PubMed abstracts of Lüscher et al. in Surgery for Obesity and Related Diseases and of the Honka and Salehi pathogenesis review for prevalence timing and the mechanism of the condition, and published multicenter cohort data for the risk-factor pattern.
  5. The Associated Press for the history of Relyvrio's approval, trial failure, and withdrawal, and Pharmaceutical Executive for the details of Amylyx's acquisition of avexitide from Eiger BioPharmaceuticals.