Petrelintide in type 2 diabetes cannot yet be ranked against tirzepatide or semaglutide, because the trial designs differ. Zealand Pharma, which develops the drug with Roche, reported Phase 2b ZUPREME-2 topline data on 7 October 2026. Mean weight loss reached up to 9.2 per cent at week 28, against 2.0 per cent on placebo.
Fierce Biotech reported the readout on 8 October. The question readers will keep asking is how the drug compares with approved options. This article sets out the figures, the limits on comparison, and what would settle it.
What petrelintide is and who is developing it
Petrelintide is a once-weekly subcutaneous amylin analogue. Zealand Pharma, a Copenhagen biotechnology company, develops it with Roche. The two signed their collaboration in March 2025.
David Kendall, Chief Medical Officer at Zealand Pharma, said the data “demonstrate its promise as a standalone treatment”. He was referring to people with overweight or obesity and type 2 diabetes. That is a sponsor statement, not a regulatory finding.
How does petrelintide in type 2 diabetes compare with tirzepatide and semaglutide?
The published figures differ in length, population and statistical method. That makes a ranking unsafe today. ZUPREME-2 ran for 28 weeks and reported weight loss under the efficacy estimand.
The tirzepatide trial SURMOUNT-2 ran for 72 weeks and used a treatment-regimen estimand. The semaglutide 2.4 mg trial STEP 2 ran for 68 weeks. Both enrolled adults with obesity or overweight and type 2 diabetes.
In SURMOUNT-2, mean weight fell 12.8 per cent on tirzepatide 10 mg and 14.7 per cent on 15 mg. The placebo group lost 3.2 per cent. Placebo-adjusted differences were 9.6 and 11.6 percentage points.
In STEP 2, semaglutide 2.4 mg cut weight by 9.6 per cent, against 3.4 per cent on placebo. The treatment difference was 6.2 percentage points.
By LSDN arithmetic, ZUPREME-2 gave a gap of up to 7.2 percentage points. That assumes one placebo figure of 2.0 per cent across arms. That figure cannot be set against the 72 and 68 week results without the caveats below.
It is not a ranking. The comparator abstracts report week 72 and week 68 results, not week 28 results. Baselines look similar, though. ZUPREME-2 participants had a mean BMI of 36.3 kg/m² and a mean HbA1c of 8.0 per cent. SURMOUNT-2 reported 36.1 kg/m² and 8.02 per cent.
Why the estimand and the timepoint matter
An efficacy estimand describes the effect in people who take treatment as intended. A treatment-regimen estimand counts everyone randomised, whatever they went on to do. The two can give different numbers.
Zealand gave its headline figure under the efficacy estimand. It said results under the treatment-regimen estimand were “largely consistent”, but gave no figures. SURMOUNT-2 used the treatment-regimen estimand. Figures from the two estimands should not be swapped.
What ZUPREME-2 reported for petrelintide in type 2 diabetes
The trial randomised 220 adults in the United States. All had overweight or obesity and type 2 diabetes. Participants took metformin, with or without an SGLT2 inhibitor. Three once-weekly petrelintide doses were compared with placebo, with escalation every fourth week. The release gives no dose amounts.
Weight loss
All three arms beat placebo on the primary endpoint at week 28. Mean loss ranged from 7.4 to 9.2 per cent across arms, under the efficacy estimand. The placebo group lost 2.0 per cent. The release gives no per-arm participant numbers.
HbA1c
HbA1c fell by up to 0.65 percentage points at week 28. It rose by 0.23 points on placebo. Placebo-adjusted reductions ranged from 0.61 to 0.88 percentage points across arms.
William Blair analysts, in a note reported by Fierce Biotech, called a placebo-arm rise in HbA1c “unusual” in such studies. The same note said HbA1c disappointed in ZUPREME-2.
Safety and tolerability
Zealand reported no unexpected safety signals. Discontinuation for gastrointestinal adverse events was 1.9 per cent on petrelintide and 1.7 per cent on placebo. Most gastrointestinal events were mild, the release said, and occurred during dose escalation.
The company described tolerability as “largely comparable to placebo”. That is the sponsor’s characterisation, not an independent assessment.
What the ZUPREME-2 release does not report
The release as issued on 7 October omits several items. It gives no dose amounts and no per-arm participant numbers. It reports no incidence rates for nausea, vomiting or other gastrointestinal events.
LSDN found no hypoglycaemia, serious adverse event or overall discontinuation figures in it. These are gaps in the release, not negative findings. Zealand said full results would come at an upcoming scientific conference, but named none.
How ZUPREME-2 follows ZUPREME-1
Genentech reported ZUPREME-1 topline data on 5 March 2026. That Phase 2 trial randomised 493 adults with overweight or obesity. Full results were published in The Lancet Diabetes & Endocrinology on 29 September 2026. On the week 28 primary endpoint, weight loss reached up to 9.8 per cent under the efficacy estimand. The placebo group lost 1.7 per cent. Weight loss reached up to 10.7 per cent at week 42, a secondary measure.
The paper allows a same-timepoint comparison within petrelintide. At week 28, it reached up to 9.8 per cent in ZUPREME-1 and 9.2 per cent in ZUPREME-2. By LSDN arithmetic, the placebo-adjusted gaps were up to 8.1 and 7.2 percentage points. The populations differ, so this is context rather than a finding.
Adverse events led to discontinuation in 4.8 per cent of the most effective arm. The placebo figure was 4.9 per cent. That measure covers all adverse events. It differs from the gastrointestinal-only figure in ZUPREME-2. The published paper also reports nausea in 20 per cent of petrelintide participants, against 6 per cent on placebo.
The market reaction to ZUPREME-1 was sharp. Zealand shares fell more than 30 per cent on 6 March, Reuters reported.
The share reaction to ZUPREME-2 was smaller. Investing.com reported that Zealand shares fell about 3 per cent on 8 October.
After ZUPREME-2, William Blair analysts said weight loss in both trials disappointed. They still expect an “improved treatment experience” because of tolerability.
The analysts wrote that petrelintide “will likely be used primarily in the maintenance setting”, Fierce Biotech reported. That is one firm’s view. It is not a finding from either trial.
LSDN’s explainer on amylin analogues covers how the class works.
What this means for UK patients
For petrelintide in type 2 diabetes, ZUPREME-2 enrolled participants in the United States only, according to Zealand’s release. UK patients are not represented in this dataset. The Phase 3a release names no countries.
Tirzepatide is already licensed in Great Britain for both type 2 diabetes and weight management. The Mounjaro KwikPen SmPC on the electronic medicines compendium shows both indications and carries a black triangle. LSDN read it on 8 October 2026. The page showed an emc update date of 8 September 2026.
Petrelintide is in Phase 3a development. The release describes no regulatory filing.
Where petrelintide in type 2 diabetes goes next
Zealand and Roche have started a Phase 3a programme of three trials. ZUPREME-3 covers obesity or overweight without type 2 diabetes. ZUPREME-4 covers people with type 2 diabetes. ZUPREME-5 covers established cardiovascular disease.
About 7,000 participants are planned in total, and the first have started treatment. Fierce Biotech reported that ZUPREME-4 is recruiting 600 people.
Roche licensed petrelintide in a deal announced on 12 March 2025. Zealand was due $1.65 billion upfront, with $1.4 billion payable at closing. Profits and losses on US and European sales are shared equally.
What would settle the comparison
Three things would settle the question of petrelintide in type 2 diabetes. The first is the full ZUPREME-2 dataset, with doses, arm sizes and treatment-regimen figures. The second is ZUPREME-4, which studies petrelintide in type 2 diabetes in the registrational programme. The third is a time-matched or head-to-head comparison.
Independent comment on ZUPREME-1 points the same way. Dr Marie Spreckley of the University of Cambridge called the results “promising”. She said “larger and longer trials, including a direct comparison” were needed before tolerability claims could be made. She was speaking to the Science Media Centre on 29 September 2026.
LSDN found no announced head-to-head trial in the sources it read on 8 October 2026. Zealand’s release gives no date for ZUPREME-4 results. Readers can follow new readouts on LSDN’s obesity drug news tracker.
