Mazdutide: real Phase 3 data, but almost all of it from China

Mazdutide is an injectable drug for weight loss and type 2 diabetes that switches on two of the body’s metabolic hormone receptors at the same time. Most of the buzzy GLP-1 compounds are still stuck at the press-release stage, but mazdutide is further along: it has a full Phase 3 trial published in a major journal, and it is actually approved in one country. The catch is that the country is China, the trials were paid for (and partly written) by the company developing the drug, and there is still no long-term outcome data and very little evidence from outside China.

What it is

Mazdutide is a once-weekly peptide given by subcutaneous (under-the-skin) injection. Chemically, it is a long-acting version of oxyntomodulin — a gut hormone the body makes naturally that triggers both the GLP-1 and glucagon receptors. Scientists tweaked that natural molecule by lipidating it (attaching a fatty chain so the drug lasts long enough to dose just once a week). Its molecular formula is C₂₁₀H₃₂₂N₄₆O₆₇, with a molar mass of roughly 4,563 g/mol and CAS number 2259884-03-0. No one has published the drug’s full amino-acid sequence (the exact order of its building blocks) in an original scientific source, so anyone who states a precise sequence is guessing.

The “dual agonism” — hitting two receptors at once — is the whole point of the drug. The GLP-1 side does the familiar job you see in drugs like semaglutide: it nudges the body to release insulin when blood sugar is high, slows how fast the stomach empties, and dials down appetite. The glucagon side is what sets mazdutide apart. Switching on the glucagon receptor is thought to raise the rate at which you burn energy and to burn fat in the liver, possibly adding weight loss and liver-fat reduction on top of what GLP-1 does alone. But the glucagon receptor cuts both ways: glucagon naturally pushes blood sugar and heart rate up. In the trials the net effect on blood sugar was still downward, but the rise in heart rate is real, gets bigger at higher doses, and is the clearest safety theme running through all the studies.

The claims

Marketing and online forums pitch mazdutide as a next-generation GLP-1 that beats semaglutide and adds liver benefits the older drugs lack. The uses that are actually approved and well-studied are chronic weight management and blood-sugar control in type 2 diabetes. The weaker, more experimental claims are about liver-fat reduction (in fatty-liver disease, now called MASLD/MASH) and broad heart-and-metabolism improvements — blood pressure, cholesterol and other lipids, uric acid, waist size. Those show up as secondary endpoints (extra things the trial measured on the side) rather than as the main thing the trial set out to prove.

Outside China, mazdutide is also sold by research-chemical and “peptide” vendors cashing in on the GLP-1 wave. Those products are not approved, not quality-checked for human use, and their claims are not regulated by anyone. Treat vendor pages as marketing, not evidence.

The evidence

The honest framing comes first: almost all of the major data come from Chinese patients, in trials funded by Innovent (the developer), with Innovent’s own employees among the study authors. That does not make the results wrong, but it is a built-in limitation that runs through the entire body of evidence.

The early work was small. A 2022 Phase 1b trial (eClinicalMedicine, n=24) gradually raised the dose across groups and reported about 11.7% weight loss in the 9 mg group versus 1.8% on placebo at week 12 — but it also recorded heart-rate jumps of up to about +17.4 bpm (beats per minute) at 9 mg, the first clear sign of the heart-rate effect. Two Phase 2 trials in 2023 added more confidence: an obesity trial (Nature Communications, n=248) found weight loss of roughly 6.7% to 11.3% across the doses over 24 weeks, and a diabetes trial (Diabetes Care, n=250) showed drops in HbA1c (a measure of average blood sugar over the past few months) that were as good as or better than dulaglutide, an existing diabetes drug used for comparison, with more weight loss on top.

The big weight-loss trial is GLORY-1, published in the New England Journal of Medicine in 2025 (n=610 Chinese adults, over 48 weeks). At 6 mg it produced about 14.8% mean weight loss versus roughly 0.5% on placebo, and around 83% of participants lost at least 5% of their body weight. It also reported a large drop in liver fat. (The exact figure varies by subgroup — Innovent’s own summary cites up to about −80% in people who started with very high liver fat — so the liver benefit is best described in broad terms rather than pinned to one precise number.) Blood pressure, lipids, uric acid, liver enzymes (transaminases), and waist size all improved, and the trial reported no sign of increased heart-and-blood-vessel risk, though it was not built to measure that.

For diabetes, the Phase 3 DREAMS program matters because two of its trials — DREAMS-1 (mazdutide on its own) and DREAMS-2 (mazdutide versus dulaglutide) — were published one right after the other in Nature on 17 December 2025. DREAMS-1 was the basis China used to approve the drug for diabetes. This is genuine, peer-reviewed Phase 3 evidence, not a press release.

Some other eye-catching numbers are still only at the press-release stage and should be taken with caution. GLORY-2 (n=462, over 60 weeks) reported mean weight loss of about 18.55% overall at the higher 9 mg dose, and about 20.08% in participants without diabetes — but it has not been peer-reviewed. DREAMS-3, a head-to-head against semaglutide, reported that mazdutide came out ahead on a combined blood-sugar-and-weight measure, but it too is still only at the conference- and press-release level. There is also a 2025 animal study (eBioMedicine) hinting at brain/thinking benefits in diabetic mice; that is preclinical (done in animals, not people) and is not a basis for any human claim.

What is missing matters as much as what exists. There is no published trial tracking real heart outcomes, and no data on hard outcomes or deaths — only stand-in measures and safety signals. The main proof of how well it works is essentially all from Chinese patients, so we do not yet know how well it carries over to other groups. We also lack long-term published data on whether the weight loss lasts and how much comes back after stopping, and there is no published head-to-head against tirzepatide. The sponsor funded the trials and had several of its own people among the authors throughout.

Safety and side effects

The most common side effects are digestive — nausea, diarrhea, vomiting, and a smaller appetite — and they are mostly mild to moderate and cluster in the early weeks while the dose is being raised. In the Phase 2 diabetes trial, diarrhea happened in about 36% of participants, nausea in 23%, and vomiting in 14%.

The signature concern, tied to how the drug works, is heart rate. The rise gets bigger at higher doses, reaching up to about +17.4 bpm at 9 mg in the small Phase 1b study, and was more modest at the 4 and 6 mg doses in GLORY-1. Anyone with heart risk should weigh this carefully, and there is no heart-outcome trial to fall back on. Hypoglycemia (blood sugar dropping too low) ran around 10% in the Phase 2 diabetes trial (versus 8% on placebo), with no severe cases reported. As for the liver, the published trials point toward improvement rather than harm. The real unknowns are long-term, multi-year safety; the pancreatitis, gallbladder, and thyroid concerns that apply to the whole incretin drug class (the family of gut-hormone-based drugs mazdutide belongs to); safety in pregnancy; and how the drug behaves in people who are not Chinese.

Legal and regulatory status

In China, the NMPA (China’s drug regulator, its version of the FDA) approved mazdutide for chronic weight management on 27 June 2025 — billed as the world’s first GCG/GLP-1 dual agonist approved for weight loss — and for blood-sugar control in type 2 diabetes on 18 September 2025. An application to add the 9 mg dose was accepted for NMPA review in November 2025 and is still pending.

In the United States, mazdutide is not approved. It is investigational (still in testing, not cleared for sale), with Lilly holding the rights outside China at an earlier stage of development. Any “FDA timeline” you see comes from commercial or secondhand sites, not the FDA, and is guesswork. Outside China there is no legal, approved product you can buy; anything sold online as a “research chemical” or unapproved peptide is unregulated, of unverified identity and purity, and not legal for human use in the US or EU.

For athletes: mazdutide is not named on the WADA (World Anti-Doping Agency) Prohibited List, and the GLP-1 class is not banned — semaglutide is on WADA’s Monitoring Program, meaning it is being watched but not prohibited. By comparison with its drug class, mazdutide is currently allowed in sport, but the category is under watch and WADA has funded work on tests to detect GLP-1 drugs, so that status could change.

Bottom line

Mazdutide has more real, peer-reviewed human evidence behind it than most of the hyped GLP-1 compounds — a Phase 3 NEJM weight-loss trial, two Phase 3 diabetes trials in Nature, and actual regulatory approval in China for two uses. That is well past “early.” But it falls short of “Strong” (grade 8–10) because the main proof of how well it works comes only from Chinese, sponsor-funded patients, the most eye-catching weight-loss numbers (the 9 mg ~20% result and the win over semaglutide) are still only at the press-release level, there is no published heart-outcome or long-term safety trial, and there is no FDA or EMA approval. Outside China, nothing sold as mazdutide is approved or quality-checked. Nothing here is medical advice.

Evidence grade: 7/10 · Moderate.

Sources

• Mazdutide. Wikipedia.
• Mazdutide (IBI362) Phase 1b in Chinese adults with overweight/obesity. eClinicalMedicine, 2022.
• Mazdutide Phase 2 in obesity/overweight. Nature Communications, 2023.
• Mazdutide Phase 2 in obesity/overweight (PubMed record). 2023.
• Efficacy and Safety of Mazdutide in Chinese Adults with Type 2 Diabetes. Diabetes Care, 2023/2024.
• Efficacy and Safety of Mazdutide in Type 2 Diabetes (full text). Diabetes Care.
• Phase 3 GLORY-1 study of mazdutide published in NEJM. Innovent/PRNewswire, 2025.
• Mazdutide 9 mg achieves up to 20.1% weight loss (GLORY-2). PRNewswire, 2025.
• Two Phase 3 results of mazdutide in type 2 diabetes published back-to-back in Nature (DREAMS-1/-2). PRNewswire, 2025.
• Mazdutide shows superiority over semaglutide in head-to-head Phase 3 (DREAMS-3). PRNewswire, 2025.
• Mazdutide receives NMPA approval for chronic weight management. Innovent/PRNewswire, 2025.
• Mazdutide receives NMPA approval for glycemic control in type 2 diabetes. Innovent/PRNewswire, 2025.
• China approval for Lilly and Innovent’s mazdutide breaks new class of GLP-1 obesity drugs. Fierce Pharma.
• Innovent enters licensing agreement with Lilly (2019). PRNewswire.
• Mazdutide improves diabetes-associated cognitive dysfunction in db/db mice. eBioMedicine, 2025.
• WADA Prohibited List. World Anti-Doping Agency.
• Ozempic on WADA’s Monitoring List for 2025. SwimSwam.