Mazdutide 10mg Peptide

Mazdutide is a synthetic analogue of oxyntomodulin, the proglucagon fragment the gut releases after a meal that activates the GLP-1 receptor and the glucagon receptor at the same time. It carries two development codes, IBI362 from Innovent Biologics and LY3305677 from Eli Lilly, and it is the molecule that moved the glucagon arm of that pairing from hypothesis to marketed product: in June 2025 China's National Medical Products Administration cleared it for long-term weight management in adults, the first clearance granted anywhere to a GLP-1 and glucagon receptor dual agonist. A second Chinese indication, glycaemic control in type 2 diabetes, followed in September 2025.

The molecule is a 34 residue peptide, CAS 2259884-03-0, built on the glucagon and oxyntomodulin backbone rather than on a GLP-1 or GIP scaffold. Position 2 carries alpha-aminoisobutyric acid instead of serine, position 20 carries a lysine acylated with a C20 fatty diacid through a gamma-glutamate and two AEEA spacers, and the chain ends in a glycinamide. Those three changes are what turn a hormone that clears from plasma in minutes into a peptide with a terminal half-life measured in days.

Almost every published efficacy dataset comes from China, at once-weekly doses between 3 mg and 9 mg. Eli Lilly, which holds rights outside China and ran the first human studies of the molecule in 2016, has taken it considerably higher in western cohorts, to 16 mg in a completed phase 2. Mazdutide supplied here is a research reagent for laboratory work and is not an approved medicine in North America or Europe.

Oxyntomodulin is glucagon plus an eight residue C-terminal extension, and it is a weak agonist at both the GLP-1 receptor and the glucagon receptor. That native dual activity is the whole design premise: rather than bolting a second pharmacology onto an incretin, mazdutide keeps the hormone that already has both and fixes its pharmacokinetics. The USAN substem the compound was assigned, -dutide, is defined for oxyntomodulin analogues, which is the cleanest statement of where the molecule sits in chemical space.

The GLP-1 receptor arm does what the class does: glucose-dependent insulin secretion, slowed gastric emptying, and reduced food intake through hindbrain and hypothalamic circuits. The glucagon receptor arm is the differentiating half. Glucagon signalling at hepatocytes raises fatty acid oxidation and energy expenditure, and in the preclinical work that supported this molecule the two arms were separable: IBI362 lowered body weight in Gcgr knockout mice and in Glp1r knockout mice, so neither receptor alone accounted for the effect, and unlike semaglutide it raised energy expenditure against vehicle.

Glucagon agonism on its own raises hepatic glucose output, which is why glucagon receptor agonists were considered unusable in diabetes for decades. The dual design resolves that: the GLP-1 component offsets the glycaemic cost of the glucagon component, and the balance between the two potencies is a deliberate engineering choice rather than an accident of the scaffold. In the phase 3 monotherapy trial in Chinese adults with type 2 diabetes, HbA1c fell 2.15 percentage points on the 6 mg arm against 0.14 on placebo at week 24, which is the practical demonstration that the offset works.

Duration comes from the C20 fatty diacid at Lys20. The diacid binds circulating albumin reversibly, which keeps the peptide out of renal filtration and releases it slowly; the alpha-aminoisobutyric acid at position 2 blocks the dipeptidyl peptidase-4 cleavage that destroys native glucagon-family peptides within minutes. In the phase 1b type 2 diabetes cohorts, median time to peak concentration was around 72 hours after a subcutaneous dose and terminal half-life ranged from 147 to 673 hours, which is what supports once-weekly exposure.

Most multi-receptor metabolic peptides start from an incretin and acquire a second pharmacology by mutation. Tirzepatide is built on a GIP scaffold and gained GLP-1 activity; retatrutide extends that logic to a third receptor. Mazdutide goes the other way. Oxyntomodulin is the 37 residue product the intestinal L cell cleaves out of proglucagon, and it already binds both the GLP-1 receptor and the glucagon receptor. The engineering problem was never how to add a receptor. It was how to keep a hormone alive in plasma for a week.

Three modifications do that work. Serine at position 2 is replaced by alpha-aminoisobutyric acid, a non-coded residue whose quaternary alpha carbon prevents dipeptidyl peptidase-4 from making the N-terminal cut that inactivates glucagon-family peptides. Lysine at position 20 is acylated with eicosanedioic acid, a twenty-carbon dicarboxylic fatty acid, attached through a gamma-glutamate and two 8-amino-3,6-dioxaoctanoic acid spacers; the free distal carboxylate is what gives this class of linker its albumin affinity. The chain terminates in a glycinamide rather than a free acid, removing the terminal charge.

That combination is the reason the receptor balance of mazdutide is not the same as the receptor balance of any GLP-1 mono-agonist scaled up, and it is why comparisons drawn from tirzepatide or semaglutide data have limited predictive value here. The USAN nomenclature committee assigned mazdutide to the -dutide substem, defined specifically for oxyntomodulin analogues, and that classification is the shortest accurate summary of what the compound is.

GLORY-1 (NCT05607680) is the trial that produced the Chinese registration. It ran at multiple Chinese centres from November 2022, randomising 610 adults aged 18 to 75 in a 1:1:1 ratio to once-weekly 4 mg, 6 mg or placebo for 48 weeks. Baseline mean body weight was 87.2 kg and mean BMI 31.1. The two primary endpoints were the percentage change in body weight and the proportion reaching at least 5% loss at week 32, both under a treatment-policy estimand that counts participants who stopped the study drug or started another anti-obesity agent.

At week 32 the 4 mg arm had lost 10.09% of baseline weight and the 6 mg arm 12.55%, against a 0.45% gain on placebo, with 73.9% and 82.0% respectively reaching the 5% threshold against 10.5%. By week 48 the figures were 11.00% and 14.01%. Prespecified cardiometabolic measures moved in the same direction. Adverse events were predominantly gastrointestinal and mild to moderate, and discontinuation for adverse events was low across all three arms at 1.5%, 0.5% and 1.0%.

GLORY-2 (NCT06164873) tested the higher 9 mg dose in a heavier population: 461 Chinese adults with a BMI of at least 30, 16.1% of whom had type 2 diabetes, randomised 2:1 against placebo for 60 weeks at 27 hospitals. Mean body weight change was -16.65% against -1.50%. The tolerability figures are the ones worth reading alongside that number: vomiting was reported by 53.1% of the treated arm against 1.3% on placebo, nausea by 46.9% against 3.2%, and diarrhoea by 39.4% against 6.5%, with 2.9% discontinuing for adverse events.

Two further phase 3 studies in the weight programme are running rather than reported: an obstructive sleep apnoea trial in Chinese adults with a BMI of at least 28 (NCT06931028) and an adolescent study (NCT07255209). A phase 2 in metabolic dysfunction-associated steatohepatitis (NCT06937749) and a phase 2 in alcohol use disorder sponsored by Eli Lilly (NCT06817356) extend the programme past weight and glycaemia.

The diabetes programme reported two phase 3 trials in Nature in 2026. The monotherapy trial randomised 320 Chinese adults whose diabetes was inadequately controlled on diet and exercise alone, mean HbA1c 8.24% and mean diabetes duration 1.9 years, to once-weekly 4 mg, 6 mg or placebo for 24 weeks. HbA1c fell 1.57 and 2.15 percentage points against 0.14 on placebo, and body weight fell 5.61% and 7.81% against 1.26%.

The active-comparator trial randomised 731 participants on background oral agents to 4 mg mazdutide, 6 mg mazdutide or 1.5 mg dulaglutide for 28 weeks. Both mazdutide arms were non-inferior and then superior to dulaglutide on HbA1c, by 0.24 and 0.30 percentage points, and the weight separation was an order of magnitude larger, 3.78% and 5.76%. That gap is the clearest read available on what the glucagon receptor arm contributes on top of pure GLP-1 agonism, because dulaglutide is a GLP-1 mono-agonist tested in the same population under the same protocol.

The earlier phase 2 in this population, published in Diabetes Care in 2024, had already shown the pattern in 250 participants over 20 weeks: HbA1c reductions of 1.41 to 1.67 percentage points across the 3, 4.5 and 6 mg arms against 1.35 with dulaglutide and 0.03 with placebo, with body weight reduction up to 7.1% against 2.7% for dulaglutide. Hypoglycaemia was reported in 10% of mazdutide participants against 8% on placebo.

DREAMS-3 (NCT06184568) is the head-to-head against semaglutide 1 mg: 349 Chinese adults with type 2 diabetes of up to ten years duration and a BMI of at least 28, randomised to once-weekly 6 mg mazdutide or semaglutide over a 32 week active-controlled period with a 24 week extension. Its primary endpoint is composite, the proportion reaching HbA1c under 7.0% together with at least 10% weight reduction at week 32.

China's National Medical Products Administration cleared mazdutide in June 2025 for long-term body weight management in adults with a BMI of at least 28, or at least 24 with one or more weight-related comorbidity, alongside diet and increased physical activity. A second Chinese clearance for glycaemic control in adults with type 2 diabetes followed in September 2025. It is marketed in China under the trade name Xinermei. There is no United States, European Union, United Kingdom, Canadian or Australian marketing authorisation, and none of the trials above was conducted with that intent.

The development history is routinely reported backwards. The molecule originated at Eli Lilly as LY3305677 and entered human study under Lilly sponsorship: NCT02972645, a multiple-dose study in healthy participants, started in December 2016, followed by NCT03325387 in 2017 and NCT03928379 in type 2 diabetes in 2019. Innovent Biologics in-licensed Chinese rights in 2019 and ran the phase 1b, phase 2 and phase 3 programme in China under the code IBI362. Lilly retained rights elsewhere and is the sponsor of the US phase 2 (NCT06124807) and the alcohol use disorder phase 2 (NCT06817356).

That split matters for reading the literature. The Chinese trials use once-weekly doses of 3 to 9 mg in participants whose mean baseline body weight was 87 to 94 kg. The Lilly studies in western cohorts run to 16 mg: a 32 participant phase 1 reached 20.0% and 21.0% mean body weight reduction at 20 weeks across two escalation regimens, and the 179 participant phase 2 reached 18.1% at 32 weeks on 16 mg. Comparing an unqualified mazdutide percentage against a retatrutide or semaglutide percentage without checking which programme and which population produced it is the most common error in secondary write-ups of this compound.

Survodutide, stocked here as survodutide-10mg, is the other GLP-1 and glucagon receptor dual agonist in this catalogue and the closest pharmacological analogue. Boehringer Ingelheim selected it from 19 candidate dual agonists on the basis of balanced receptor potency, and its phase 3 SYNCHRONIZE-1 trial randomised 725 adults with obesity and without diabetes to once-weekly 3.6 mg, 6.0 mg or placebo for 76 weeks, reporting mean body weight change of -12.2% and -13.0% against -5.4% on placebo. The two molecules differ in scaffold, in the geography of their trial populations and in placebo response: SYNCHRONIZE-1 recorded a 5.4% placebo weight loss where GLORY-1 recorded a 0.3% placebo weight gain, which alone forbids reading the headline percentages side by side.

Retatrutide, stocked here across several vial sizes, adds a third receptor. Its 338 participant phase 2 reported -24.2% mean body weight change at 48 weeks on the 12 mg arm against -2.1% on placebo, the largest figure any incretin-family peptide had published at that time point. The comparison worth drawing is mechanistic rather than numerical: retatrutide adds GIP receptor agonism to the same GLP-1 and glucagon pairing mazdutide has, so the two compounds isolate what the GIP arm contributes, while survodutide and mazdutide isolate what a different balance between the same two receptors contributes.

No head-to-head trial exists between any pair drawn from mazdutide, survodutide and retatrutide. Every cross-molecule comparison in the literature is indirect, drawn between trials that differ in duration, population, escalation schedule and estimand. Mazdutide is the only one of the three holding a marketing authorisation anywhere.

The strongest preclinical evidence separating mazdutide from a GLP-1 mono-agonist is hepatic. In a 2025 Radiology study, 42 male mice were fed a high-fat diet for 13 weeks and then given mazdutide, semaglutide or phosphate-buffered saline every three days for four weeks, with quantitative imaging on a 9.4 tesla scanner at baseline, one week and four weeks. Median proton density fat fraction fell 5.59% from baseline with the dual agonist against 3.30% with semaglutide at four weeks, a difference not yet present at one week. Fat fraction correlated with hepatic triglyceride content and with histological steatosis score.

A separate mouse study of early metabolic dysfunction-associated steatotic liver disease, using a 12 week high-fat diet followed by four weeks of subcutaneous mazdutide at 100, 200 or 400 micrograms per kilogram, plus a free fatty acid loaded hepatocyte model at 10 to 50 nM, reported reduced hepatic steatosis alongside suppression of the PERK arm of the endoplasmic reticulum stress response, lower NF-kappaB signalling, and downregulation of SREBP-1, C/EBP beta and PPAR gamma.

The energy expenditure claim rests on knockout work rather than on indirect calorimetry in humans. The compound reduced body weight in Gcgr knockout mice and in Glp1r knockout mice, so neither receptor alone accounts for the effect, and it raised energy expenditure against vehicle in a comparison where semaglutide did not. In a db/db mouse model of type 2 diabetes, mazdutide improved performance on cognitive behavioural testing relative to dulaglutide, with transcriptomic, proteomic and metabolomic analysis pointing to neuroprotection and energy metabolism pathways; those findings are in male mice only and have no human counterpart yet.

Mazdutide ships as a lyophilised solid. Lyophilised acylated peptides of this class are stable at ambient temperature for the duration of transit and are held long term at -20 °C or below, protected from light and moisture; vials should be equilibrated to room temperature before opening so that condensation does not enter the powder. Once in solution the peptide is held at 2 to 8 °C and aliquoted so that a working stock is not repeatedly frozen and thawed. Every one of those points is a laboratory storage condition, not a preparation instruction.

Identity confirmation is where catalogue listings for this compound most often fail. The molecular formula is C210H322N46O67 and the average mass 4,563.06 g/mol, the value consistent with the FDA substance registry entry (UNII MB76Z4IBZ5) and with the USAN and INN chemical name: a 34 residue chain, alpha-aminoisobutyric acid at position 2, a C20 diacid conjugate at Lys20, glycinamide at the C-terminus. Deposited structure records in the public chemistry databases carry a chain one serine residue shorter, which is why masses near 4,476 g/mol also circulate, and figures near 4,234 g/mol count the peptide without its fatty diacid and linker, which together account for roughly 740 g/mol of the total. A mass-spectrometric identity check is only meaningful against a stated reference mass.

Purity by reversed-phase HPLC is the usual specification, and Volta releases at or above 99%, with mass spectrometry as the orthogonal identity method. Acylated peptides of this length carry a characteristic impurity profile: deletion and des-amino sequences from synthesis, and the free acid at the C-terminus where amidation is incomplete. A certificate of analysis reporting a single purity number without a chromatogram and a mass spectrum for the same lot is not a characterisation.

Immunogenicity is worth noting for anyone designing repeat-dose animal work. In the 12 week phase 1b in Chinese patients with type 2 diabetes, 4 of 24 participants receiving the compound, 16.7%, developed anti-drug antibodies against 1 of 12 on placebo, and no neutralising antibodies were detected in post-baseline samples.