Mazdutide

Origin and structure

Mazdutide is a modified version of oxyntomodulin, a hormone that cells in the gut release after a meal and that naturally activates two receptors: the GLP-1 receptor and the glucagon receptor. Eli Lilly designed it under the code LY3305677 and first gave it to healthy volunteers in the UK in 2016. In 2019 Lilly licensed the rights for China to Innovent Biologics, where the compound, then called OXM3, became IBI362. Innovent ran the large trials and won approval in China under the brand name Xinermei.

The molecule is a chain of 34 amino acids: H-Aib-QGTFTSDYSKYLDEKKAKEFVEWLLEGGPSSG-NH2. Position 2 carries 2-aminoisobutyric acid (Aib), an unnatural amino acid that GLP-1 drugs commonly use to resist breakdown by the enzyme DPP-4. The lysine at position 20 carries a 20-carbon fatty acid on a short linker, a side chain much like those on semaglutide and tirzepatide. It binds albumin in the blood and keeps the peptide circulating long enough for one injection under the skin per week.

How it works

At the GLP-1 receptor mazdutide behaves much like semaglutide: the pancreas releases more insulin when blood sugar is high, the stomach empties more slowly and appetite falls. The glucagon receptor adds a second signal, aimed mainly at the liver, where glucagon increases energy use and the breakdown of fat. The bet behind the design is that this second signal adds weight loss and liver-fat reduction on top of the GLP-1 effect.

Glucagon on its own pushes blood sugar up, so the GLP-1 arm has to outweigh it. In the diabetes trials HbA1c fell clearly, so the balance works out in that direction. Whether the glucagon arm adds anything in people is less clear. The active comparators so far were GLP-1-only drugs at diabetes doses (dulaglutide 1.5 mg in DREAMS-2, semaglutide 1 mg in DREAMS-3), and no trial has yet set mazdutide against a GLP-1 drug at its full obesity dose. Heart rate rose by a few beats per minute in the trials, a known effect of GLP-1 drugs to which glucagon may add.

Is mazdutide approved?

Only in China. The Chinese medicines regulator (NMPA) approved it in June 2025 for weight management and in September 2025 for type 2 diabetes. No other country has approved it.

In the EU and the US it has no marketing authorisation, and no mazdutide medicine is registered in the Netherlands. Lilly licensed only the Chinese rights to Innovent; it has run phase 1 and phase 2 trials in the US, including a 48-week obesity trial and a trial in alcohol use disorder, but had registered no phase 3 trial by October 2026.

Research-grade mazdutide, the lyophilised peptide sold for laboratory use, is not Xinermei. It has no marketing authorisation anywhere, is not made under a medicines licence, and may not be used in people.

• China: approved June 2025 as Xinermei for long-term weight management, alongside diet and exercise, in adults with a BMI of 28 or more, or 24 or more with at least one weight-related condition (China uses lower BMI cut-offs than Europe); September 2025 for blood-sugar control in adults with type 2 diabetes.
• EU: not authorised. The only trials in Europe were Lilly's early phase 1 studies in the UK (2016 to 2018) and Germany (2019 to 2021).
• US: not approved. Lilly completed a high-dose phase 1 trial, a phase 2 obesity trial (2023 to 2025) and a phase 2 trial in alcohol use disorder (2025 to 2026); a phase 2 master protocol for weight management that includes mazdutide is still recruiting in the US and Argentina. No phase 3 trial is registered.

What the research shows

Four phase 3 trials have been published, all in China and all funded by Innovent: GLORY-1 and GLORY-2 in obesity, DREAMS-1 and DREAMS-2 in type 2 diabetes, together about 2,100 participants. Lilly's US phase 2 trial is the only sizeable published study outside China. The headline results:

• Phase 2, obesity, China (Nature Communications 2023): 248 adults; after 24 weeks −6.7% body weight on 3 mg, −10.4% on 4.5 mg and −11.3% on 6 mg, against +1.0% on placebo.
• GLORY-1 (NEJM 2025): 610 adults with a mean BMI of 31.1; after 48 weeks −11.0% on 4 mg and −14.0% on 6 mg, against +0.3% on placebo. At least 15% was lost by 35.7% and 49.5%, against 2.0% on placebo.
• GLORY-2 (JAMA 2026): 462 adults with a BMI of 30 or more (mean 34.3), 16% of them with type 2 diabetes; after 60 weeks on 9 mg, −16.7% against −1.5% on placebo, and 84.3% against 33.1% lost at least 5%.
• DREAMS-1 (Nature 2026): 320 adults with type 2 diabetes treated with diet and exercise alone; after 24 weeks HbA1c fell by 1.57 points on 4 mg and 2.15 on 6 mg, against 0.14 on placebo, with weight loss of 5.6% and 7.8% against 1.3%.
• DREAMS-2 (Nature 2026): 731 adults with type 2 diabetes on tablets, against dulaglutide 1.5 mg weekly; after 28 weeks HbA1c fell 0.24 and 0.30 points more than on dulaglutide with 4 and 6 mg, and weight loss was 3.8 and 5.8 percentage points greater.
• US phase 2 (Lancet Diabetes & Endocrinology 2026): 179 US adults without diabetes; after 32 weeks −7.3% on 3 mg, −15.6% on 10 mg and −18.1% on 16 mg, against −0.9% on placebo, with further loss by week 48.
• DREAMS-3 (company press release, October 2025; not yet in a journal): 349 adults with type 2 diabetes and obesity; after 32 weeks 48.0% on mazdutide 6 mg against 21.0% on semaglutide 1 mg reached an HbA1c below 7% together with at least 10% weight loss, and weight fell 10.3% against 6.0%.
• Still running: GLORY-3 against semaglutide, and trials in sleep apnoea, adolescents, high blood pressure, MASH and heart failure.

Side effects and warnings

As with other GLP-1 drugs, the side effects are mostly gastrointestinal and rise with the dose. At 4 and 6 mg in GLORY-1 they were mostly mild to moderate, and 1.5% and 0.5% stopped treatment because of side effects, against 1.0% on placebo. At 9 mg in GLORY-2, 53.1% vomited (placebo 1.3%), 46.9% had nausea (3.2%) and 39.4% diarrhoea (6.5%), and 2.9% stopped, against none on placebo. In the US phase 2 trial, 20% of those on 16 mg stopped because of side effects, mainly gastrointestinal.

In the Chinese phase 2 trial, heart rate rose by 5.4 to 8.8 beats per minute on mazdutide and by 4.8 on placebo, and returned to normal after treatment. That trial found no suspected pancreatitis, no calcitonin values of 20 ng/L or higher and no thyroid tumours, but two participants had rises of the pancreatic enzyme lipase above three times the upper limit of normal. The evidence has clear limits: almost all participants were Chinese, the longest published treatment lasted 60 weeks, and there is no cardiovascular outcome trial. Rare problems known from other GLP-1 drugs, such as pancreatitis and gallbladder disease, need larger and longer studies to assess.

Mazdutide versus semaglutide and tirzepatide

All three act on the GLP-1 receptor; they differ in the second target. Semaglutide works on GLP-1 alone, tirzepatide adds the GIP receptor, and mazdutide adds the glucagon receptor. Retatrutide combines all three, and survodutide from Boehringer Ingelheim is another GLP-1/glucagon dual agonist in late-stage development.

No head-to-head trial against semaglutide or tirzepatide has been published in a journal. DREAMS-3 compared mazdutide 6 mg with semaglutide 1 mg, a diabetes dose rather than the 2.4 mg obesity dose, in 349 Chinese adults with type 2 diabetes and obesity. Innovent announced in October 2025 that mazdutide came out ahead on the combined goal of an HbA1c below 7% plus at least 10% weight loss at 32 weeks; these are company figures that have not appeared in a peer-reviewed journal. GLORY-3 compares the two in 479 adults with overweight or obesity and fatty liver, with weight and liver fat at 48 weeks, and is still running. The only published comparison is with dulaglutide (DREAMS-2), an older GLP-1 drug. Setting the GLORY numbers beside Western semaglutide or tirzepatide trials is misleading: the populations, BMI thresholds and trial lengths differ.

Frequently asked questions

Is mazdutide approved by the EMA or the FDA?

No. China is the only country that has approved it: in June 2025 for weight management and in September 2025 for type 2 diabetes, under the brand name Xinermei. Lilly, which licensed only the Chinese rights to Innovent, has run phase 1 and 2 trials in the US but has not registered a phase 3 trial, which an EU or US approval would normally require.

Is mazdutide legal in the Netherlands?

No mazdutide medicine is authorised in the Netherlands or anywhere else in the EU, so there is no registered product for a doctor to prescribe or a pharmacy to dispense. Research-grade mazdutide is laboratory material, not a medicine, and may not be used in people.

How is mazdutide different from tirzepatide (Mounjaro)?

Both activate the GLP-1 receptor, but tirzepatide pairs it with the GIP receptor and mazdutide with the glucagon receptor, which is meant to raise energy use and act on liver fat. Tirzepatide is approved in the EU and the US with large international trials behind it; mazdutide is approved only in China, and the two have never been compared directly.

How much weight did people lose on mazdutide in trials?

In GLORY-1, Chinese adults lost 11.0% on 4 mg and 14.0% on 6 mg after 48 weeks, against a 0.3% gain on placebo. In GLORY-2, adults with a BMI of 30 or more lost 16.7% on 9 mg after 60 weeks, against 1.5% on placebo. In a smaller US phase 2 trial, 16 mg reached 18.1% at 32 weeks, but one in five stopped because of side effects.

Is mazdutide banned in sport?

It is not named on the WADA Prohibited List, and GLP-1 drugs are not prohibited. The catch-all class S0 bans substances that no government health authority has approved for human use, but mazdutide has been approved in China since June 2025, so S0 does not cover it. In the Netherlands the Dopingautoriteit applies the same list.

Is research-grade mazdutide the same as Xinermei?

No. Xinermei is a licensed medicine in China, made and tested under pharmaceutical rules with a known dose per injection. Research-grade mazdutide is a lyophilised peptide sold for laboratory use; its identity and purity are what a certificate of analysis says, nothing more, and it may not be used in people. Our lab results show what we test for.

Storage and handling

Store according to product-label instructions and batch handling notes. Lyophilized research peptides are commonly kept dry, protected from light, and handled with clean laboratory technique.

References

Checked against these sources on 28 October 2026.