TL;DR: GLP-1 is not the only hormone that tells your brain a meal is over. Amylin, co-released with insulin from the pancreas, runs a separate satiety pathway through the area postrema, and it’s why every major obesity-drug maker now has an amylin analog in the clinic. Pramlintide proved the biology decades ago but needed constant injections. Cagrilintide made it once-weekly and paired it with semaglutide as CagriSema. Petrelintide is chasing amylin alone, with early signs of unusually clean tolerability. Zenagamtide (formerly amycretin) skips co-formulation entirely and engineers both jobs — GLP-1 and amylin — into one molecule. None of the newer three is approved yet.
What is amylin, actually?
Amylin, also called islet amyloid polypeptide (IAPP), is a 37-amino-acid hormone secreted by the same pancreatic beta cells that make insulin — and released in the same moment, in roughly a 1:100 ratio alongside it. Every time a meal triggers an insulin pulse, amylin rides along.
Once in circulation, amylin does three things that matter for weight and glucose control:
- Slows gastric emptying, stretching out how long a meal sits in the stomach.
- Suppresses postprandial glucagon, tempering the liver’s tendency to dump extra glucose into the blood right after eating.
- Drives satiety centrally, largely through the area postrema, a circumventricular organ in the hindbrain that sits outside the blood-brain barrier. That location matters mechanistically: it gives a circulating peptide direct access to brainstem satiety and nausea circuitry without needing to cross the barrier, and it feeds forward into hypothalamic appetite circuits as well.
Amylin doesn’t act through its own dedicated gene-encoded receptor. Instead, it signals through the calcitonin receptor (CTR) paired with a receptor activity-modifying protein (RAMP) — combinations that produce the AMY1, AMY2, and AMY3 receptor subtypes. That shared-receptor architecture is why cagrilintide and its relatives are often described in the pharmacology literature as dual amylin/calcitonin receptor agonists rather than “pure” amylin mimetics.
A satiety pathway that doesn’t overlap with GLP-1
GLP-1 receptor agonists like semaglutide work through the incretin system — a gut-and-brain pathway tied to nutrient sensing after a meal. Amylin runs on a structurally distinct receptor system that happens to converge on some of the same downstream hypothalamic and brainstem circuits, without being the same signal.
That non-redundancy is the entire mechanistic case for stacking the two pathways, which is exactly what CagriSema and zenagamtide are built to do in different ways — see our deep dive on what CagriSema is for how a two-molecule combination expresses that logic.
It’s also the source of a claim worth interrogating rather than repeating: that amylin agonism is better tolerated than incretin agonism. The evidence so far leans that direction — Zealand Pharma’s ZUPREME-1 phase 2 data for petrelintide reported vomiting in about 3% of participants versus 6.2% on placebo, a strikingly clean profile for this drug class — but that’s a placebo-controlled read within one trial, not a head-to-head against a GLP-1 comparator, and the participant pools and durations across amylin trials still differ. “Placebo-like tolerability” is the sponsor’s framing; treat it as an encouraging early signal in a still-thin dataset, not an established property of the mechanism.
Four compounds, three design philosophies
The most useful way to sort the amylin field isn’t by company — it’s by how many molecules are doing the work.
| Compound | Molecule(s) | Mechanism | Status (mid-2026) |
|---|---|---|---|
| Pramlintide | Single peptide, amylin analog | Amylin/calcitonin receptor agonist | FDA-approved (2005) for diabetes; short-acting, dosed with meals |
| Cagrilintide | Single peptide, amylin analog (studied alone and co-formulated with semaglutide) | Amylin/calcitonin receptor agonist (DACRA) | Investigational; phase 3 CagriSema program (REDEFINE) |
| Petrelintide | Single peptide, amylin only | Amylin/calcitonin receptor agonist | Phase 2 complete (ZUPREME-1); phase 3 announced Apr 2026, initiation planned 2H 2026 |
| Zenagamtide (amycretin) | Single 68-amino-acid unimolecular peptide | GLP-1 receptor + amylin receptor co-agonist, one molecule | Phase 2 complete in type 2 diabetes; phase 3 planned H2 2026 |
Pramlintide: the cautionary tale
Pramlintide (marketed as Symlin) is the compound that proved amylin biology works in humans, and it did so more than two decades ago. It’s approved for use alongside insulin in diabetes, and it genuinely slows gastric emptying and blunts postprandial glucose the way the mechanism predicts. The catch is pharmacokinetics: native amylin’s short half-life carries over into pramlintide, which needs to be injected before meals — up to three times a day. That dosing burden is precisely why pramlintide never became a weight-management mainstay despite validating the pathway, and it’s the reason every compound that followed was engineered first and foremost for duration.
Cagrilintide: making amylin last a week
Cagrilintide took the pramlintide-validated biology and rebuilt it for once-weekly dosing using the same fatty-acid acylation approach that extended semaglutide’s half-life. It’s studied both as a standalone amylin analog and — more prominently — co-formulated with semaglutide as CagriSema, Novo Nordisk’s two-molecule combination now in phase 3. For the full mechanism and trial record, see what cagrilintide is and the CagriSema overview.
Petrelintide: amylin, and only amylin
Petrelintide, developed by Zealand Pharma with Roche as co-development partner, is the cleanest test of amylin monotherapy at long-acting, once-weekly dosing — no GLP-1 component at all. The phase 2 ZUPREME-1 trial reported roughly 10.7% mean weight loss at week 42, against about 1.7% on placebo, alongside that notably low vomiting rate. Zealand and Roche announced on April 29, 2026 that they would advance petrelintide into phase 3, with initiation planned for the second half of 2026 — phase 2 data in hand, phase 3 not yet underway. More detail is in our dedicated petrelintide guide.
Zenagamtide: one peptide, two receptors
Zenagamtide — Novo Nordisk’s renamed amycretin — takes the opposite engineering path from CagriSema. Instead of co-formulating two peptides, it’s a single 68-amino-acid chain with a GLP-1-receptor-agonist segment and an amylin-receptor-agonist segment joined by a short glycine/glutamic-acid linker, fatty-acid acylated for albumin binding and stabilized against DPP-4 degradation. One molecule, two targets, by design rather than by mixing two drugs in a vial.
At ADA 2026, Novo Nordisk reported phase 2 data in adults with type 2 diabetes: up to 14.6% mean weight loss at 36 weeks at the top studied dose, against about 2.1% on placebo, with A1C reductions of up to 1.71 percentage points. Phase 3 development is slated to begin in the second half of 2026. See the zenagamtide guide for the full trial breakdown.
Different trials, durations, and populations — not a head-to-head. Petrelintide and cagrilintide-alone figures are from adults with obesity/overweight; zenagamtide's figure is from adults with type 2 diabetes, a population that typically shows an attenuated weight-loss response.
Why every GLP-1 maker is chasing amylin
Look across the current obesity pipeline and the pattern is hard to miss: every company with a leading GLP-1 asset is now also running an amylin program, either as a companion combination or folded into a single molecule. The logic is straightforward once you separate the two hormone systems the way this article has — GLP-1 alone has a ceiling, and a genuinely non-overlapping satiety pathway is one of the few credible ways to push past it without simply raising the incretin dose and its GI burden. Whether that shows up as a two-molecule combination like CagriSema, a single-molecule co-agonist like zenagamtide, or amylin running solo like petrelintide is now the live design question in the field. For where these sit alongside multi-agonist incretins and other emerging candidates, see the next-gen weight loss peptides overview; for the incretin side of the comparison, our semaglutide explainer covers the GLP-1 mechanism these amylin compounds are designed to complement.
Frequently asked questions
What is amylin and how is it different from GLP-1?
Amylin is a 37-amino-acid hormone co-released with insulin from pancreatic beta cells; it slows gastric emptying, suppresses glucagon after meals, and drives satiety mainly through the area postrema. GLP-1 is a separate incretin hormone acting through its own receptor system. The two converge on overlapping brain circuits but signal through structurally distinct receptors, which is why researchers treat them as complementary rather than redundant pathways.
What’s the actual difference between CagriSema and zenagamtide?
CagriSema is two molecules — cagrilintide and semaglutide — combined in one formulation. Zenagamtide is a single 68-amino-acid peptide engineered to activate both the GLP-1 receptor and the amylin receptor on its own. Both aim at the same two pathways; they just get there through different molecular architectures.
Is petrelintide the same thing as cagrilintide?
No. Both are long-acting amylin analogs, but petrelintide is being developed by Zealand Pharma with Roche and studied strictly as amylin monotherapy, while cagrilintide is a Novo Nordisk molecule studied both alone and — more prominently — co-formulated with semaglutide as CagriSema.
Has petrelintide or zenagamtide been approved?
No, neither has been approved anywhere as of this writing. Petrelintide has completed phase 2 (ZUPREME-1) and phase 3 was announced in April 2026 with initiation planned for the second half of 2026. Zenagamtide has completed phase 2 in type 2 diabetes, with phase 3 also planned for the second half of 2026.
Why did pramlintide never become a major weight-loss drug?
Pramlintide proved that amylin agonism works in humans, but its short half-life meant it needed injecting before meals, up to three times a day. That dosing burden — not a lack of efficacy — is why the field waited for long-acting engineering, first with cagrilintide and now petrelintide and zenagamtide, before amylin became a serious weight-management contender.
Is amylin agonism really better tolerated than GLP-1 agonism?
The early data leans that way — petrelintide’s ZUPREME-1 trial reported a notably low vomiting rate close to placebo — but that’s a within-trial, placebo-controlled comparison, not a head-to-head against a GLP-1 drug. It’s a promising signal worth watching through phase 3 rather than a settled fact about the mechanism.