Amylin vs GLP-1: Why Amylin Is the 2026 Story

Ask what the next obesity medicine looks like and the honest answer in mid-2026 is that nobody knows, but three companies have placed the same bet. Amylin is not a new hormone. A version of it has carried a United States prescription label since 2005. What is new is that Novo Nordisk, Zealand Pharma with Roche, and Eli Lilly all have a long-acting amylin analogue in mid-stage human trials at the same time, and each of them argues that this pathway does something the GLP-1 drugs do not. Some of that argument rests on published trials. Some of it rests on rat data and press releases.

What amylin actually is

Amylin is a pancreatic hormone, released from beta cells alongside insulin after a meal. A 2026 narrative review in Diabetes, Obesity and Metabolism sets out its physiology in three parts: it slows gastric emptying, it suppresses glucagon secretion, and it promotes meal termination through central mechanisms. The third part is the one the obesity field cares about. A 2026 review in Peptides adds the receptor biology. The amylin receptor is not a standalone protein; it is a complex assembled around the calcitonin receptor, and working that structure out is what let chemists design a second generation of non-aggregating, long-acting analogues. Native amylin clumps. That is why early attempts at turning it into a medicine were so awkward, and why every molecule now in trials is an engineered version rather than the hormone itself.^{[1, 2]}

Where that diverges from a GLP-1 drug

GLP-1 receptor agonists work through a different receptor and a different hormone family. Semaglutide acts at the GLP-1 receptor; tirzepatide adds the GIP receptor; retatrutide adds glucagon on top of both. Amylin is not an incretin and does not act at any of those receptors. From a reader's point of view the two routes end in the same place, which is eating less, but they get there through separate circuits. That separation is the whole commercial argument. The plain-language summary of the 2026 Diabetes, Obesity and Metabolism review states it directly: amylin-based therapies target appetite regulation through pathways different from those the current GLP-1 based medicines work through. The 2026 Peptides review adds the observed consequence, which is that in clinical trials the combination of cagrilintide and semaglutide has achieved greater effects than either component alone.^{[1, 2]}

The amylin medicine that already exists

Pramlintide reached the United States market in 2005 and is sold as SYMLIN, later as the SymlinPen device, under application NDA 021332, now labelled by AstraZeneca Pharmaceuticals. Its prescribing information describes a narrow role: an adjunctive therapy in patients with type 1 or type 2 diabetes who use mealtime insulin and who have not reached their glucose targets on optimal insulin alone. It is not sold as a weight-loss product. Both 2026 reviews make the same point about it. Pramlintide established that the pathway could produce weight loss in people, and then ran into two walls: modest efficacy, and a schedule that required repeated daily doses. That is the history the long-acting analogues are trying to get past, and it is why a hormone described decades ago is only now a headline.^{[1, 2, 6]}

Three programmes, one receptor family

The three long-acting amylin analogues furthest along are cagrilintide from Novo Nordisk, petrelintide from Zealand Pharma with Roche, and eloralintide from Eli Lilly. They sit at different stages and they are being developed for different jobs. Cagrilintide has been carried furthest as half of a two-drug combination with semaglutide. Petrelintide and eloralintide are both being run as standalone weekly compounds. The trial-by-trial numbers live on the entity pages rather than here: the petrelintide data sheet and what petrelintide is, the eloralintide data sheet and eloralintide explained, and the cagrilintide data sheet. For the two standalone programmes lined up against each other, see petrelintide vs eloralintide. A fourth approach folds both pathways into a single molecule rather than combining two: amycretin is a unimolecular GLP-1 and amylin receptor agonist, and it has its own data sheet.^[7]

Selectivity is the design variable

The amylin receptor family has more than one member, and that is where the three molecules differ on paper. A 2025 paper in Molecular Metabolism, from the Eli Lilly group that discovered eloralintide, reports that in cell assays the compound preferentially activated the human AMY1 receptor, by roughly twelve-fold over the calcitonin receptor and eleven-fold over the AMY3 receptor. The same paper describes cagrilintide as a non-selective amylin receptor agonist, and reports that in lean rats eloralintide induced significantly less conditioned taste avoidance than cagrilintide. Conditioned taste avoidance is a rodent behavioural measure researchers read as a nausea proxy. It is a rat finding published by the company that owns the compound, and it circulates online as though it were a human tolerability result.^[3]

What the combination trial changed

REDEFINE 1 was published in the New England Journal of Medicine in 2025. It was a phase 3a trial of 3417 adults with overweight or obesity and without diabetes, running 68 weeks, with four arms: cagrilintide and semaglutide together, semaglutide alone, cagrilintide alone, and placebo. Under the treatment-policy estimand, which counts everyone as randomised whether or not they stayed on the medicine, the combination arm showed a mean body-weight change of -20.4 percent against -3.0 percent for placebo. Gastrointestinal adverse events were reported in 79.6 percent of the combination group against 39.9 percent on placebo, and the authors describe them as mainly transient and mild to moderate. Read the estimand line whenever you see a figure from this trial quoted, because company materials and journal tables do not always report the same one.^[4]

Where the head-to-head evidence actually is

One published trial has compared an amylin analogue directly against a GLP-1 receptor agonist, and it is older and narrower than the discussion around it suggests. The 2021 cagrilintide dose-finding phase 2 trial in The Lancet randomised 706 participants across five cagrilintide dose groups, with a liraglutide comparator arm and placebo arms, over a 26-week treatment period. Under the trial-product estimand, mean weight reduction from baseline ran from 6.0 to 10.8 percent across the cagrilintide groups against 3.0 percent for placebo. The top cagrilintide group came in at 10.8 percent against 9.0 percent for liraglutide, an estimated difference of 1.8 percentage points. That is a real comparison, and it is also one dose pair inside a dose-finding trial, against a first-generation daily GLP-1 medicine rather than against semaglutide or tirzepatide. The trial that would let anyone rank the two classes has not been run.^[5]

What the evidence does not show

Three gaps. First, no long-acting amylin analogue has reported a cardiovascular or other hard-outcome trial. The 2026 review that surveyed the field found pramlintide to be the only amylin-based medicine with a marketing authorisation as of its 30 April 2026 cut-off, with everything else still in clinical development. Second, we could not find a published body-composition substudy for cagrilintide, petrelintide or eloralintide, of the kind the GLP-1 trials now run as standard. That gap matters, because lean-mass loss is the live argument about the incretin drugs and nobody has measured it in this class yet; our guide to what the DEXA data shows covers where those numbers come from. Third, the tolerability claim these programmes rest on has not been tested against a modern GLP-1 comparator in a trial designed to answer that question.^[1]

Tolerability is the pitch, and it is not settled

Each of these programmes is being positioned on how it feels rather than only on the number. The 2026 Peptides review is blunt about the current state: gastrointestinal side effects, especially nausea, similar to those reported for GLP-1 receptor agonists, are commonplace when people start on amylin analogues and when the dose steps up, and mostly settle with continued exposure. That is a review author's summary rather than a pooled estimate, and the individual trial figures vary enormously between programmes and between dose arms inside the same programme. The REDEFINE 1 number above, where four in five people on the combination reported a gastrointestinal event, is the most solid figure in the class, and it describes a combination rather than an amylin analogue on its own.^{[2, 4]}

Regulatory status and what is being sold

Cagrilintide, petrelintide and eloralintide are not approved by the FDA for any use, and they are not approved anywhere else. They exist inside company trial programmes, and the only people with legitimate access to them are participants in those trials. Material advertised online under any of these names sits outside that supply chain, and no published analysis tells you what is in it. Our page on gray-market peptides covers what is and is not knowable about that trade. Anything you would actually put in your body belongs in a conversation with a clinician, not a forum thread.

References & sources

1. Alhazmi A, le Roux CW. Amylin Analogs: The Next Major Class of Weight Loss Therapy: A Review of Experimental Data and Early-Phase Clinical Trials. Diabetes Obes Metab. 2026 Jul 14. doi:10.1111/dom.71074. PMID 42452898. Narrative review with a 30 April 2026 evidence cut-off; source for amylin physiology, pramlintide's limits, and the roster of compounds in development.
2. Bailey CJ, Flatt PR, Conlon JM. Long-acting amylin-related peptides as therapies for obesity and type 2 diabetes. Peptides. 2026 Mar;196:171480. PMID 41747885. Source for the amylin-calcitonin receptor complex, the second-generation non-aggregating analogue design, the CagriSema combination effect, and the gastrointestinal tolerability summary.
3. Briere DA, et al. Eloralintide (LY3841136), a novel amylin receptor agonist for the treatment of obesity: From discovery to clinical proof of concept. Mol Metab. 2025 Dec;102:102271. PMCID PMC12640043. Eli Lilly. Source for the AMY1 receptor selectivity ratios and the rat conditioned-taste-avoidance comparison against cagrilintide.
4. Garvey WT, et al. Coadministered Cagrilintide and Semaglutide in Adults with Overweight or Obesity. N Engl J Med. 2025 Aug 14;393(7):635-647. PMID 40544433. REDEFINE 1, ClinicalTrials.gov NCT05567796. Phase 3a, 3417 participants, 68 weeks, treatment-policy estimand.
5. Lau DCW, et al. Once-weekly cagrilintide for weight management in people with overweight and obesity: a multicentre, randomised, double-blind, placebo-controlled and active-controlled, dose-finding phase 2 trial. Lancet. 2021 Dec 11;398(10317):2160-2172. PMID 34798060. ClinicalTrials.gov NCT03856047. The one published comparison against a GLP-1 receptor agonist.
6. SYMLIN (pramlintide acetate) prescribing information, application NDA 021332, AstraZeneca Pharmaceuticals LP. DailyMed label set 4aea30ff-eb0d-45c1-b114-3127966328ff, retrieved 26 July 2026. Source for the indication wording and the marketing history.
7. Dahl K, et al. Amycretin, a novel, unimolecular GLP-1 and amylin receptor agonist: results from a phase 1b/2a randomised controlled study. Lancet. 2025 Jul 12;406(10499):149-162. PMID 40550231. ClinicalTrials.gov NCT06064006. Novo Nordisk. Source for the unimolecular co-agonist approach.