After GLP-1, Amylin: The Obesity Drugs Betting on Tolerability Instead of Magnitude
Petrelintide and amycretin put amylin receptor agonism to the test: less weight loss than incretins, far better GI tolerability. RUO reference overview.
Amylin analogues such as petrelintide and zenagamtide (amycretin) target amylin receptors, a mechanism distinct from GLP-1. Phase 2 trials report less weight loss than leading incretins but markedly better gastrointestinal tolerability — petrelintide 10.7% at 42 weeks with 1.5% GI discontinuation. All are investigational, unapproved, and discussed here as research-use-only reference material.
For three years the obesity pipeline has been an arms race over a single number: percentage of body weight lost. Amylin receptor agonists are the first serious attempt to compete on a different axis. Petrelintide and zenagamtide (amycretin) do not win that arms race on magnitude, and their sponsors are arguing — with data now on the record — that magnitude was never the binding constraint. Everything discussed below is investigational, unapproved by any regulator, and covered here strictly as research-use-only (RUO) reference material.
What is amylin, and why is it not simply another incretin?
Amylin is a 37-amino-acid peptide hormone co-secreted with insulin by pancreatic beta cells after a meal. Its physiological job is short-term meal termination: it slows gastric emptying, suppresses postprandial glucagon and signals satiety through amylin receptors in the hindbrain. That last part matters, because it is a different neural route from the one incretin drugs use. GLP-1 and GIP receptor agonism — the pharmacology that produced the current generation of metabolic agents, and whose origin story runs through Gila monster venom — acts largely through separate circuits. Amylin analogues also appear to restore sensitivity to leptin, a mechanism incretins are not credited with.5
The practical consequence of different circuitry is a different side-effect profile. Nausea and vomiting are the dose-limiting problems of incretin agonism, and the reason many people stop treatment before reaching a maintenance dose. If satiety can be produced without recruiting the same emetic pathways, a drug does not need to be as strong on paper to be as useful in practice. That is the entire bet.
What did the petrelintide Phase 2 actually report?
Petrelintide is a long-acting amylin analogue given once weekly by subcutaneous injection, developed by Zealand Pharma with Roche. The Phase 2 ZUPREME-1 trial (NCT06662539) randomised 485 adults with overweight or obesity8 and reported topline results in March 2026, with the full dataset presented at the American Diabetes Association Scientific Sessions on 5 June 2026.12
The trial reported a mean body weight reduction of up to 10.7% at week 42, against 1.7% for placebo, meeting its primary endpoint.23 Taken alone that figure is unremarkable next to what the incretin field has published. The ADA presentation was built around the second set of numbers. The trial reported nausea in 19.6% of participants on petrelintide versus 6.2% on placebo — and vomiting in 3.0% versus 6.2% on placebo, meaning the reported vomiting rate fell below the placebo arm. Discontinuation attributed to gastrointestinal adverse events was 1.5%. Overall discontinuation for adverse events was 4.8% on drug against 4.9% on placebo. Between 88% and 98% of participants, depending on arm, successfully escalated to their target maintenance dose.156
1.5% of participants discontinued petrelintide for gastrointestinal reasons in the Phase 2 trial — the number the tolerability thesis rests on.
A 10.7% mean reduction is not the largest number in obesity research. The argument is that it is the number most people would still be seeing two years later — and that argument is a hypothesis, not a result.
Where does zenagamtide (amycretin) fit?
Novo Nordisk’s zenagamtide, previously known as amycretin and coded NN 9487, takes a different structural approach. It is a unimolecular peptide agonist of both GLP-1 and amylin receptors: one molecule engaging two receptor families. This is a genuine distinction from CagriSema, which pairs two separate molecules — cagrilintide and an incretin agonist — in a fixed-dose combination. Zenagamtide is in development in both once-weekly subcutaneous and once-daily oral formulations.9
The Phase 2 dose-finding study in type 2 diabetes was presented at ADA on 5 June 2026. It enrolled 262 adults with type 2 diabetes inadequately controlled on metformin with or without an SGLT2 inhibitor, testing six subcutaneous dose levels from 0.4 mg to 40 mg against placebo over 36 weeks. The trial met its primary endpoint of A1C change: from a baseline of 7.8%, the trial reported a reduction of up to 1.71 percentage points at the 40 mg dose, an estimated treatment difference versus placebo of −1.56 (95% CI −2.05 to −1.07, p<0.0001), with up to 89.1% of participants reaching A1C below 7%. The key supportive secondary endpoint, body weight, reported up to 14.6% mean loss at week 36 versus 2.1% on placebo, with no apparent plateau by the end of the treatment period. Most adverse events were gastrointestinal and mostly mild to moderate.9 Novo Nordisk stated it planned to begin Phase 3 in type 2 diabetes in the second half of 2026.10
The earlier phase 1b/2a study in overweight and obesity, published in The Lancet in June 2025, reported up to 24.3% mean weight loss at 36 weeks with subcutaneous amycretin at the highest dose tested (60 mg) against 1.1% on placebo, in 125 participants.1112 That is a much larger number than petrelintide’s — but it is also a much smaller, earlier and shorter study, and zenagamtide is a dual agonist that retains GLP-1 pharmacology. It is not a test of amylin agonism in isolation.
| Programme | Mechanism | Study reported | Reported weight change | Reported GI signal |
|---|---|---|---|---|
| Petrelintide | Amylin receptor agonist (monotherapy) | Phase 2, 42 weeks, 485 adults | up to 10.7% vs 1.7% placebo | Nausea 19.6% vs 6.2%; vomiting 3.0% vs 6.2%; 1.5% GI discontinuation |
| Zenagamtide (amycretin) | Unimolecular GLP-1 + amylin agonist | Phase 2 in T2D, 36 weeks, 262 adults | up to 14.6% vs 2.1% placebo | Mostly mild–moderate GI events; rates not directly comparable |
| Amycretin, subcutaneous | Unimolecular GLP-1 + amylin agonist | Phase 1b/2a, 36 weeks, 125 adults | up to 24.3% vs 1.1% placebo | GI most common; mostly mild–moderate |
| Retatrutide | Triple GIP/GLP-1/glucagon agonist | Phase 2, 48 weeks | 24.2% at highest dose vs 2.1% placebo13 | Predominantly GI, dose-related |
| Tirzepatide | Dual GIP/GLP-1 agonist | Phase 3 SURMOUNT-1, 72 weeks | 20.9% at 15 mg vs 3.1% placebo14 | Predominantly GI, dose-related |
These are separate trials with different durations, populations, endpoints and eras of trial design. None of these comparisons is head-to-head, and cross-trial percentages should not be read as a ranking. All figures are reported trial outcomes for investigational compounds; none is approved by the FDA, EMA or any other regulator for any indication, and none of this constitutes dosing guidance. Research-use-only reference materials are not medicines.
Why the combination strategy matters more than the monotherapy number
The commercial signal is worth reading alongside the clinical one. In March 2025, Roche paid Zealand Pharma $1.65 billion upfront for petrelintide, with up to $5.3 billion more in milestones and sales-based payments — an unusually large sum for a molecule whose monotherapy weight loss would later land in the low double digits.7 The deal explicitly included a fixed-dose combination of petrelintide with Roche’s dual GLP-1/GIP agonist enicepatide (CT-388).4
On 29 April 2026, Roche and Zealand announced that petrelintide would advance to Phase 3 for chronic weight management, beginning in the second half of 2026, with a Phase 2 study of the combination planned earlier in the year.415 Read that way, amylin agonism is not being positioned as a replacement for incretin pharmacology. It is being positioned as a tolerability-preserving partner for it — a way to add efficacy without adding proportionally to the nausea burden. Zenagamtide encodes the same idea in a single molecule instead of a combination product.
An honest read of the evidence
Several caveats deserve equal billing with the headline numbers.
Phase 2 results routinely shrink in Phase 3. Larger, longer, more heterogeneous populations with stricter estimand handling tend to produce smaller treatment effects and worse tolerability than dose-finding studies, and there is no reason to assume amylin analogues are exempt. Petrelintide’s 10.7% is a Phase 2 figure; so is zenagamtide’s 14.6%, and amycretin’s 24.3% comes from a phase 1b/2a study of 125 people.
Cross-trial comparison is not evidence. Placing retatrutide‘s 24.2% next to petrelintide’s 10.7% is a category error: different populations, different durations, different placebo responses, different trial eras. The only valid comparisons will come from head-to-head studies that have not been run. Our own comparison of the incretin agents carries the same caveat for the same reason.
Most importantly, the central claim is still unproven. What has been measured is tolerability — nausea rates, vomiting rates, discontinuation rates, dose-escalation success. What has not been demonstrated is the inference built on top of it: that better tolerability produces better adherence, that better adherence produces better real-world weight outcomes, and that this eventually outweighs a larger effect achieved in a controlled trial. Every step of that chain is plausible. None of it has outcomes data behind it yet. Real-world persistence with metabolic agents is influenced by cost, access, supply and motivation at least as much as by nausea, and a trial cannot tell you how those interact.
There is a broader shift here too. The interesting question is moving from “how much weight can a molecule remove” to “what happens to the person while it does” — body composition, muscle preservation, durability after stopping. Amylin analogues are one answer; work on exercise mimetics is another. Both are early.
Nothing here is medical advice, and nothing here is dosing guidance. Petrelintide, zenagamtide and amycretin are investigational compounds that no regulator has approved for any indication, and every dose figure above is reported as a fact about what a clinical trial administered, not as a recommendation. Reference materials supplied for laboratory research are not medicines and are not intended for human or veterinary use.
Condor Research · Scientific desk
Atrio Sciences s.r.o., IČO 57 669 651, Nitra (SK) · info@condorresearch.com
- Amylin is a 37-amino-acid peptide hormone co-secreted with insulin by pancreatic beta cells; amylin receptor agonism acts through largely different neural circuits from GLP-1/GIP incretin agonism.
- Petrelintide (Zealand Pharma/Roche), a once-weekly subcutaneous amylin analogue, reported up to 10.7% mean weight loss at 42 weeks versus 1.7% placebo in Phase 2 ZUPREME-1 (NCT06662539, 485 adults randomised).
- Full ZUPREME-1 data at ADA on 5 June 2026 quantified the tolerability claim: nausea 19.6% vs 6.2% placebo, vomiting 3.0% vs 6.2% placebo (below the placebo rate), 1.5% GI-related discontinuation, and 88–98% of participants reaching target maintenance dose.
- Zenagamtide (amycretin, NN 9487) is a unimolecular peptide agonist of both GLP-1 and amylin receptors — one molecule, two receptors — distinct from CagriSema, which combines two separate molecules.
- Zenagamtide Phase 2 in type 2 diabetes (262 adults, 36 weeks, ADA 5 June 2026) met its A1C primary endpoint at up to −1.71 percentage points from a 7.8% baseline, with up to 14.6% body weight loss versus 2.1% placebo.
- Roche paid Zealand $1.65bn upfront (up to $5.3bn total) in March 2025, including a fixed-dose combination of petrelintide with the dual GLP-1/GIP agonist enicepatide (CT-388); Phase 3 in chronic weight management was announced on 29 April 2026 to begin in H2 2026.
- None of petrelintide, zenagamtide or amycretin is approved by the FDA, EMA or any other regulator for any indication — all are investigational.
Is petrelintide approved anywhere?
No. Petrelintide is investigational. It has not been approved by the FDA, the EMA or any other regulator for any indication. Roche and Zealand announced in April 2026 that it would advance to Phase 3 for chronic weight management beginning in the second half of 2026, which means pivotal data does not yet exist.
How is an amylin analogue different from a GLP-1 receptor agonist?
Amylin is a 37-amino-acid hormone co-secreted with insulin by pancreatic beta cells, and amylin analogues act on amylin receptors through largely different neural circuits from incretin agonists. They are reported to slow gastric emptying and signal satiety, and are also associated with restoring leptin sensitivity. The practical difference observed in trials so far is a gastrointestinal side-effect profile much closer to placebo.
Does petrelintide produce less weight loss than incretin agents?
In the trials published to date, yes. The Phase 2 trial reported up to 10.7% mean loss at 42 weeks, while incretin trials have reported figures around 20% and above over longer periods. Those are separate studies and cannot be compared directly, but nobody in the field claims amylin monotherapy currently matches incretins on magnitude. The argument is about tolerability and staying on treatment, not about peak effect.
What is the difference between zenagamtide, amycretin and CagriSema?
Zenagamtide and amycretin are two names for the same Novo Nordisk compound (code NN 9487), a single molecule that activates both GLP-1 and amylin receptors. CagriSema is a fixed-dose combination of two separate molecules, one of which is the amylin analogue cagrilintide. Unimolecular versus two-molecule is a real pharmacological and manufacturing distinction, not a marketing one.
Why did Roche pay so much for a drug with 10.7% weight loss?
Because the strategy is not monotherapy. The March 2025 agreement — $1.65 billion upfront and up to $5.3 billion in further payments — explicitly covered a fixed-dose combination of petrelintide with Roche's dual GLP-1/GIP agonist enicepatide. The apparent thesis is that amylin agonism adds efficacy to an incretin backbone without adding proportional nausea.
Can these compounds be used to treat obesity?
No. They are investigational compounds under clinical study and are not approved for the treatment of obesity, type 2 diabetes or anything else. Any reference material of this kind supplied for laboratory work is for research use only, is not a pharmaceutical product, and is not for human or veterinary consumption. Questions about weight management should go to a qualified clinician.