Petrelintide phase 3 begins as Zealand and Roche bank two phase 2 readouts

Zealand Pharma and Roche have moved their amylin analog into petrelintide phase 3 testing, with three trials that together plan about 7,000 participants. The step follows two phase 2 readouts: ZUPREME-1 in people without diabetes, now published in The Lancet Diabetes and Endocrinology, and ZUPREME-2 in people with type 2 diabetes, reported as topline results on 7 October. In both, weight loss came with gastrointestinal side effects that the companies describe as largely comparable to placebo, the feature they are betting the programme on.

The three petrelintide phase 3 trials

ClinicalTrials.gov lists three phase 3 studies with Roche as lead sponsor, all recruiting and all with a start date of 30 September 2026. ZUPREME-3 (NCT07843498) plans 3,900 adults with overweight or obesity, ZUPREME-4 (NCT07843485) plans 600 adults with overweight or obesity and type 2 diabetes, and ZUPREME-5 (NCT07843472) plans 2,500 adults with overweight or obesity and established cardiovascular disease. Each tests once weekly petrelintide against placebo. The ZUPREME-3 registry record gives an estimated primary completion in December 2028, while ZUPREME-4 is due in October 2028 and ZUPREME-5 in July 2030. Zealand said on 7 October that the first participants had started treatment.

Petrelintide mimics amylin, a hormone released by the pancreas after meals that signals fullness to the brain and slows the emptying of the stomach. It is built as a long acting analog so that it can be given once a week. Zealand Pharma discovered it and now develops it with Roche, which is why the phase 3 trials carry the Roche name on the registry. A further phase 2 study, ZYNERGY (NCT07589686), will test petrelintide together with enicepatide, a Roche incretin also known as CT-388, in 486 adults; the record lists it as not yet recruiting. The molecule sits on our amylin class page next to its closest rivals.

What ZUPREME 1 showed in people without diabetes

ZUPREME-1 (NCT06662539) randomised 493 adults without type 2 diabetes, of whom 485 received the drug or placebo, at 32 sites in Poland, Romania and the United States. Participants had a body mass index of at least 30, or at least 27 with high blood pressure or abnormal blood lipids. Five petrelintide maintenance doses were compared with placebo in a five to one ratio, with dose escalation in fixed four week steps and lifestyle counselling for everyone. At week 28, the primary endpoint, weight fell by 7.9 to 9.8 percent on petrelintide against 1.7 percent on placebo. At week 42 the drug arms had lost 8.7 to 10.7 percent of body weight, and placebo had not moved further.

At week 42 the drug arms had lost 8.7 to 10.7 percent of body weight, and placebo had not moved further.

The tolerability numbers explain the interest. Nausea affected 20 percent on the drug against 6 percent on placebo, mostly mild and mostly during escalation. Vomiting was 3 percent against 6 percent, diarrhoea 7 percent in both groups and constipation 7 against 4 percent, according to the paper as summarised by Pharmacy Times. Only 1 percent stopped permanently because of gastrointestinal events. Other findings need watching: antibodies to the drug appeared in 38 percent of treated people without an apparent effect on weight loss, hair loss was reported in 5 percent against 1 percent, and there were three treatment related serious adverse events, two gallstone cases and one obstructive pancreatitis. Higher doses gave similar results, suggesting a plateau.

ZUPREME 2 adds people with type 2 diabetes

ZUPREME-2 (NCT06926842) enrolled 220 adults in the United States with overweight or obesity and type 2 diabetes, on metformin with or without an SGLT2 inhibitor. Mean body mass index was 36.3 and mean HbA1c 8.0 percent. Three petrelintide doses were compared with placebo, with escalation every fourth week for up to 16 weeks and maintenance to week 28. Zealand reported weight loss of 7.4 to 9.2 percent across the drug arms against 2.0 percent on placebo, under the efficacy estimand. HbA1c fell by up to 0.65 percentage points on petrelintide and rose by 0.23 points on placebo. Stopping treatment because of gastrointestinal events was 1.9 percent on the drug against 1.7 percent on placebo.

Weight loss in obesity trials of people with type 2 diabetes often runs lower than in trials without diabetes, and the ZUPREME-2 figures sit a little below ZUPREME-1, though the trials differ in length and design. These are topline numbers from a company announcement; Zealand said full results will be presented at a scientific meeting, without naming it. The glucose effect is modest next to incretin drugs, which is expected for a molecule that does not act on the GLP-1 receptor. What the two readouts share is a pattern of steady weight loss with side effect rates that rarely separate from placebo. The table sets the phase 2 and phase 3 trials side by side.

What phase 3 has to prove for amylin alone

On our board the molecule moves from phase 2 to phase 3, with the signal up. The weight loss seen so far, around 10 percent at 42 weeks, is below what the leading incretin drugs report, so the petrelintide phase 3 case rests on tolerability and on how long people stay on treatment rather than on headline efficacy. The ZUPREME-1 authors warned against comparing across trials, noting a figure of 11.5 percent at 68 weeks reported for cagrilintide, another amylin analog. They also flagged limits of their own study: 86 percent of participants were White, dose groups were not masked from one another, and several cardiometabolic analyses were done after the fact.

Three checkpoints come next. Full ZUPREME-2 results at a scientific meeting will show the dose by dose detail that the topline omitted. ZYNERGY, once it starts recruiting, will test whether petrelintide adds to an incretin in the way that rival combinations aim to. And the petrelintide phase 3 readouts, from late 2028, will show whether the gentle side effect profile holds in thousands of people treated for longer. The cardiovascular study, ZUPREME-5, runs to 2030. We log each confirmed step in our stage moves feed. Petrelintide is not approved anywhere, and nothing in this story is a guide to treatment.

Conclusion

With petrelintide phase 3 under way in about 7,000 people, Zealand Pharma and Roche are testing whether an amylin analog can win on comfort rather than size of effect. Phase 2 gave weight loss near 10 percent with gastrointestinal side effects close to placebo, plus antibody and gallbladder signals to watch. The next checkpoints are full ZUPREME-2 data, the start of ZYNERGY and phase 3 readouts from late 2028.