Petrelintide's phase 2 trial: 493 participants, up to 10.7% mean weight loss at 42 weeks, no vomiting at the top dose. What the published paper supports.
Petrelintide is an investigational long-acting amylin analogue developed by Zealand Pharma as a once-weekly injection. The ZUPREME-1 phase 2 trial randomized 493 adults with overweight or obesity and reported up to 10.7% mean weight loss at week 42 versus 1.7% with placebo, with no vomiting and no gastrointestinal discontinuations at the maximally effective dose. The full paper is now published in The Lancet Diabetes & Endocrinology.
Evidence checked October 22, 2026. This page explains human research, not purchasing, dosing or self-administration.
What is petrelintide?
Petrelintide is a human amylin analogue with a half-life of about 10 days. Amylin is a hormone released with insulin after meals that slows gastric emptying and increases satiety. Petrelintide sits in the same mechanistic family as eloralintide and the amylin half of CagriSema, but it is a distinct molecule.
Zealand is developing it two ways at once: as a monotherapy, and as a combination partner for other weight-management drugs. Those are separate evidence questions, and the combination data do not exist yet.
What did ZUPREME-1 measure?
ZUPREME-1 was a multicentre, randomized, double-blind, placebo-controlled dose-finding trial across 32 sites in Poland, Romania and the USA. It enrolled 493 adults without type 2 diabetes, mean baseline BMI of 37 kg/m², 53% female. Participants received once-weekly petrelintide, escalated every four weeks, across five dose arms, or placebo. The primary endpoint was weight change at week 28; week 42 was a secondary endpoint.
| Measure | ZUPREME-1 result | How to read it |
|---|---|---|
| Participants | 493 (404 petrelintide, 81 placebo pooled) | Larger than most phase 2 amylin trials |
| Primary endpoint | Significant weight loss versus placebo in all five arms at week 28 | Dose-finding, not a single winning dose |
| Greatest mean loss at week 42 | 10.7%, efficacy estimand | Best-performing arm, not the trial average |
| Placebo mean loss at week 42 | 1.7% | Within-trial comparator |
| Vomiting at the maximally effective dose | Zero cases | A tolerability signal in one arm of one trial |
| GI discontinuations at that dose | Zero | 98% of that cohort reached the maintenance dose |
Sources: the published Lancet Diabetes & Endocrinology paper00213-5/fulltext) (DOI 10.1016/S2213-8587(26)00213-5, published September 29, 2026) and Zealand's March 5, 2026 topline announcement.
The tolerability claim, in context
"Placebo-like tolerability" is the phrase both Zealand and Roche used. The published adverse-event table supports a real signal: pooled nausea was 19.6% on petrelintide versus 6.2% on placebo, and most gastrointestinal events were mild. That is genuinely lower than the nausea rates seen in GLP-1 trials.
Two cautions. First, the zero-vomiting result belongs to one dose arm, not every arm. Second, tolerability comparisons across trials are informal. A proper claim that petrelintide is easier to stay on than semaglutide or tirzepatide requires a head-to-head trial, which does not exist.
One more detail worth keeping: female participants lost considerably more weight than male participants in this trial, per Roche's announcement. Sex differences in amylin response are not yet explained.
How does 10.7% compare?
Petrelintide's best arm is below the headline numbers for tirzepatide, retatrutide or eloralintide. That gap is real, and the honest framing is the one the companies themselves use: petrelintide's pitch is tolerability and combination potential, not maximum weight loss.
Cross-trial percentage rankings still mislead. ZUPREME-1 ran 42 weeks with a 42-week endpoint; the trials behind bigger numbers ran longer, used different populations and different analyses. Our eloralintide evidence review shows how much the estimand alone can move a headline.
What remains unknown
Phase 3 results do not exist. Neither do cardiovascular outcome data, long-term maintenance data after stopping, or any combination-therapy results. The amylin class is promising precisely because several companies are now testing it, but promise is not evidence.
Sources and related research
- Published ZUPREME-1 paper, The Lancet Diabetes & Endocrinology00213-5/fulltext)
- Zealand Pharma topline announcement, March 5, 2026
- Roche announcement, March 5, 2026
- ZUPREME-1 ADA 2026 poster (adverse-event tables)
- Eloralintide: the other amylin agonist with published phase 2 data
- Amycretin: GLP-1 plus amylin in one molecule
Frequently asked
Related Articles
New guides, by email
New compound guides and research notes when we publish them. No set schedule. Unsubscribe anytime.
No spam. Unsubscribe anytime.
Evidence notes from published research and trial registries. Educational information, not medical advice.