Mazdutide

Summary

Mazdutide is a once-weekly injectable that activates two receptors: GLP-1 and glucagon. Chemically it is a long-acting analog of oxyntomodulin, a gut hormone that naturally hits both. That pairing is the same bet survodutide makes — GLP-1 suppresses appetite, glucagon is thought to raise energy expenditure — and a different one from tirzepatide, which pairs GLP-1 with GIP.

What separates mazdutide from most of the next-generation obesity drugs is that it is no longer investigational. China's NMPA approved it in June 2025 for weight management and again in September 2025 for type 2 diabetes, making it the first approved GLP-1/glucagon dual agonist anywhere. The evidence behind that is real: two Phase 3 obesity trials published in the New England Journal of Medicine and JAMA, and two Phase 3 diabetes trials published in Nature.

The two things to hold onto: every pivotal trial was run entirely in Chinese participants, and the approval is good in China and nowhere else. There is no FDA or EMA approval, no legal prescription route in the US or Europe, and nothing sold online under this name is the approved product.

What people use it for

Weight loss, mostly, and this is the peptide-forum use that has grown fastest since the Chinese approval — an approved obesity drug that most of the world cannot get is exactly the situation gray-market suppliers exploit. Type 2 diabetes is the second approved indication and the subject of its own Phase 3 programme.

The pitch that circulates online usually goes further than either approval: that the glucagon component makes mazdutide a better fat-loss drug than semaglutide because it burns calories rather than only suppressing appetite. The mechanism behind that framing is real. Whether it delivers a better outcome than the drugs already on the market is a separate claim, and it is graded separately below.

Human evidence

GLORY-1 — the trial the approval rests on. Published in the New England Journal of Medicine in 2025, this randomized, double-blind, placebo-controlled Phase 3 trial enrolled 610 Chinese adults with obesity or overweight (mean weight 87.2 kg, mean BMI 31.1) and gave once-weekly subcutaneous mazdutide or placebo for 48 weeks.

The trial also reported benefits across all prespecified cardiometabolic measures. Discontinuation for adverse events was low — between 0.5% and 1.5%.

GLORY-2 — the higher dose. Published in JAMA, this trial randomized 461 Chinese adults with obesity (mean BMI 34.3, 16.1% with type 2 diabetes) 2:1 to 9 mg once weekly or placebo for 60 weeks. Mean weight change was −16.65% (95% CI −18.19 to −15.12) against −1.50% on placebo, with 84.3% of the mazdutide group losing at least 5% versus 33.1% on placebo.

Note the number if you have seen this trial reported elsewhere: the sponsor's press release describes the same trial as producing 18.55% weight loss overall and 20.08% in participants without diabetes, and that "20%" figure is what most coverage repeated. Both sets of numbers come from the same data analysed under different estimands — broadly, whether people who stopped the drug or added rescue therapy are counted as failures or excluded. We quote the peer-reviewed figure, because that is the one that survived review.

Type 2 diabetes. Two Phase 3 trials were published back-to-back in Nature. DREAMS-1 randomized 320 adults whose diabetes was inadequately controlled by diet and exercise alone (mean HbA1c 8.24%) to 4 mg, 6 mg or placebo for 24 weeks: HbA1c fell 1.57% on 4 mg and 2.15% on 6 mg against 0.14% on placebo, with weight down 5.61% and 7.81% respectively. DREAMS-2 compared mazdutide against an active drug — 731 adults on background oral medication, randomized to mazdutide 4 mg, 6 mg or dulaglutide 1.5 mg for 28 weeks. Both mazdutide doses beat dulaglutide on HbA1c (by 0.24% and 0.30%) and by considerably more on weight (3.78% and 5.76% greater reduction).

Two published Phase 3 trials per indication, in more than a thousand randomized participants, plus a regulatory approval in a major market, is why both the weight-loss and blood-sugar claims grade S+. That grade describes the strength of the evidence, not its breadth: every one of these trials was conducted in China, in Chinese participants, and no trial has yet reported cardiovascular outcomes — the endpoint that ultimately separated semaglutide from its predecessors.

What has not been shown. There is no published trial establishing that mazdutide beats semaglutide or tirzepatide for weight loss. A head-to-head trial against semaglutide exists — DREAMS-3 — and the sponsor reported in October 2025 that mazdutide 6 mg outperformed semaglutide on a composite endpoint in 349 adults with early type 2 diabetes and obesity. But that result is a topline press release, not a peer-reviewed publication; only the trial's design and baseline data have appeared in a journal. Two further limits matter: the comparator was semaglutide 1 mg, the type 2 diabetes dose, not the 2.4 mg dose used for weight loss, and the population had diabetes, not obesity alone. An independent 2026 BMJ network meta-analysis that pooled 43 obesity-drug trials placed mazdutide in a 13.1–14.6% weight-loss band alongside retatrutide and ecnoglutide, explicitly rated the certainty of that estimate as very low to low, and ranked tirzepatide first at −14.9%. So the "stronger than the GLP-1s" claim grades C — not disproved, but currently resting on indirect comparison and an unpublished trial.

Animal / preclinical evidence

The dual-agonist rationale predates the trials. Mazdutide is built from oxyntomodulin, a naturally occurring gut hormone that activates the GLP-1 and glucagon receptors together; in preclinical mouse work the molecule improved glucose control, reduced body weight and increased energy expenditure. Glucagon-receptor agonism is what supplies the energy-expenditure and hepatic-fat arm of the effect, while GLP-1 agonism suppresses food intake and offsets glucagon's tendency to raise blood glucose — each partner covering the other's weakness.

That is the mechanism, and it is sound. It is also no longer the main evidence: with four published Phase 3 trials, the grades above rest on the human data, and the preclinical work is here as background rather than support.

Anecdotal / community reports

Low-confidence. These are community reports, not evidence. Not medical guidance.

Mazdutide is prescribed in China and unavailable everywhere else, which produces an unusual split. Some online discussion comes from people with genuine prescriptions; far more comes from people who bought vials labelled "mazdutide" from research-chemical suppliers outside any medical system. Those two groups are not describing the same product, and posts rarely distinguish them.

The commonly reported effects — appetite suppression, nausea, rapid early weight loss — match the pharmacology and the trials, which makes them easy to believe and no more evidentiary for it. Nothing about a report confirms that a vial contained mazdutide, contained the stated amount, or was sterile.

Doses used in published studies

Context only — not a recommendation. PeptideIQ Base does not provide dosing advice.

The Phase 3 obesity trial GLORY-1 used 4 mg or 6 mg once weekly by subcutaneous injection over 48 weeks, and its registration record gives the exact escalation: the 4 mg arm started at 2 mg weekly for four weeks before stepping up, and the 6 mg arm spent four weeks at 2 mg and four more at 4 mg before reaching target. These two doses, 4 mg and 6 mg, are what China approved. GLORY-2 tested 9 mg once weekly for 60 weeks; that dose is not approved and a supplementary application for it is under review. The diabetes trials used the same 4 mg and 6 mg doses — for 24 weeks alone (DREAMS-1) and 28 weeks on top of oral medication (DREAMS-2).

Two things these figures do not tell you. First, the published abstracts for GLORY-2, DREAMS-1 and DREAMS-2 do not state their titration schedules, so the step-up detail above applies to GLORY-1 only and should not be assumed to carry across. Second, and more importantly, every one of these regimens was administered under clinical supervision, with a known product, monitoring, and a defined stopping rule. Nothing about them transfers to unverified material bought outside a medical system.

Safety & side effects

The side-effect profile is dominated by gastrointestinal events, as with every drug in this class. GLORY-1 reported gastrointestinal adverse events as the most common, mostly mild to moderate, with discontinuation for adverse events between 0.5% and 1.5%. GLORY-2, at the higher 9 mg dose, reported markedly higher rates: vomiting 53.1%, nausea 46.9%, diarrhea 39.4%, again mostly mild to moderate, with 2.9% discontinuing versus 0% on placebo. In the diabetes trials the commonest events were diarrhea, decreased appetite and nausea, and gastrointestinal events were more frequent on mazdutide than on dulaglutide.

One signal is specific to the glucagon component and worth naming. The Phase 1b dose-escalation study, which pushed to 9 mg and 10 mg in 24 participants, recorded heart-rate increases of up to 17.4 beats per minute. Glucagon-receptor agonism raises heart rate and can affect glucose and hepatic parameters in ways a GLP-1 agonist alone does not. No serious adverse events or pancreatitis were seen in that small study, but a 24-person Phase 1b is not where cardiovascular safety gets settled — and no cardiovascular-outcomes trial for mazdutide has reported.

Long-term safety is therefore only partly characterised: the longest published exposure is 60 weeks, entirely within one population. And none of this addresses unregulated product, where unknown identity, purity and sterility sit on top of the drug's own profile.

Regulatory / legal status

Mazdutide is approved in China and nowhere else. The NMPA approved it on 27 June 2025 under the brand name Xinermei, for chronic weight management in adults with a BMI of 28 or above — or 24 or above with at least one weight-related comorbidity such as high blood sugar, high blood pressure, dyslipidemia, fatty liver or obstructive sleep apnea — used alongside a reduced-calorie diet and increased physical activity. A second approval for glycemic control in type 2 diabetes followed in September 2025. Approved doses are 4 mg and 6 mg once weekly; the 9 mg dose from GLORY-2 is the subject of a supplementary application under review.

Outside China there is no approval and no legal prescription route. The drug originated at Eli Lilly as LY3305677 and was licensed to Innovent Biologics for Greater China; Lilly has completed a Phase 2 obesity study of the same molecule outside China, testing doses as high as 16 mg, so the molecule is in development elsewhere — but development is not approval. The FDA has not approved it, and neither has the EMA.

That matters for a specific reason. An approved drug that most of the world cannot obtain creates precisely the conditions gray-market suppliers sell into, and material listed online as "mazdutide for research use" is not the approved Chinese product: it has had no regulatory review, no verified identity or purity, and no pharmacovigilance behind it. Treating the trial and label evidence on this page as though it applies to that material is the error the C grade above is about. Status can change and varies by country; this is not legal or medical advice.

Podcast / media mentions

Mazdutide entered Western coverage in mid-2025 as "China approves the first GLP-1/glucagon drug," and again in late 2025 when GLORY-2 was reported as 20% weight loss. Both framings are recognisable from the sources, and both compress something.

The approval is genuine and the evidence behind it is a proper Phase 3 programme — this is not a case of a weak drug clearing a low bar. But an approval is jurisdictional: it says a Chinese regulator reviewed Chinese trial data, and it creates no legal path to the drug in the US or Europe. The 20% figure, meanwhile, is the sponsor's analysis of the non-diabetic subgroup; the peer-reviewed JAMA result for the whole trial is 16.65%. Neither is dishonest, but only one of them went through review, and the gap between the two is a useful illustration of how much a headline number can move depending on who is counted.