Mazdutide: what it is, how it works and what studies have found

Mazdutide is, according to its developer, the world’s first dual glucagon/GLP-1 receptor agonist approved for weight management [1]: China approved it in 2025 for weight management and then for type 2 diabetes. Outside China it is still investigational. This overview sets out its chemistry, mechanism, pharmacokinetics, the published GLORY and DREAMS trial data and its regulatory position, with each quantitative statement linked to a primary source. It reports published research and is not medical guidance.

What is mazdutide?

Mazdutide (development codes IBI362 and LY3305677) is a once-weekly synthetic peptide analogue of mammalian oxyntomodulin, the proglucagon-derived gut hormone that activates both the GLP-1 receptor (GLP-1R) and the glucagon receptor (GCGR) [2][3]. A fatty-acyl side chain extends its half-life [2]. Its INN and USAN is mazdutide, and KEGG classifies it as an oxyntomodulin analogue targeting GLP1R and GCGR [4].

It has been studied in obesity and type 2 diabetes in China and in obesity in the United States [5][6][7].

Mazdutide structure, sequence and molecular weight

Mazdutide has a 34-residue main chain. Position 2 is 2-aminoisobutyric acid (Aib), the C-terminus is a glycinamide, and lysine 20 carries the side chain: two 8-amino-3,6-dioxaoctanoic acid (AEEA) units, a gamma-glutamate and a 20-carbon diacid (icosanedioic acid, written in registry names as 19-carboxynonadecanoyl) [8][4].

Property	Mazdutide
INN / USAN	Mazdutide [4]
Development codes	IBI362, LY3305677 [2]
CAS number	2259884-03-0 [4][9]
Molecular formula	C210H322N46O67 [4]
Average molecular weight	4563.07 g/mol (KEGG) [4]
Monoisotopic mass	4560.3203 Da [4]
Main-chain length	34 amino acids [4][8]
Sequence (one-letter)	H-X-QGTFTSDYSKYLDEKKAKEFVEWLLEGGPSSG-NH2, X = Aib [4][8]
Side chain	Lys20 N6-[AEEA-AEEA-gamma-Glu-C20 diacid] [8]
Receptors	GLP-1 receptor and glucagon receptor (agonist) [4]
Salt form	Not defined by the INN; synthetic research material is usually an acetate or trifluoroacetate salt

A note on conflicting database values. The PubChem compound record for mazdutide lists C207H317N45O65 and a molecular weight of about 4,476 g/mol [9]. The difference from the KEGG formula (C3H5NO2, about 87 Da) is exactly one serine residue. Recalculating the formula from the registered 34-residue sequence with the Aib, amide and side-chain modifications gives C210H322N46O67 (about 4,563 g/mol), matching KEGG [4][8]. Laboratories setting LC-MS identity criteria should use the KEGG value.

Mazdutide mechanism of action

Oxyntomodulin activates both GLP-1R and GCGR and, given pharmacologically, can lower both glucose and body weight [3]. Mazdutide reproduces that dual activity in a long-acting molecule. GLP-1R agonism underlies the glucose-lowering and appetite effects shared with selective GLP-1 receptor agonists; the developer describes glucagon receptor activation as potentially adding energy expenditure and hepatic fat metabolism effects [10].

Human evidence of glucagon receptor engagement comes from biomarkers. In a US phase 1 trial, participants treated with mazdutide showed reductions in gluconeogenic amino acids and large increases in fasting beta-hydroxybutyrate from baseline (not significantly different from placebo); the authors interpreted the amino-acid change as consistent with GCGR target engagement and the ketone rise as likely reflecting GCGR-mediated hepatic fatty-acid oxidation, while noting the latter was not directly measured [11]. Evidence level: phase 1 biomarker.

Mazdutide half-life and pharmacokinetics

Two phase 1b trials in China report long and variable terminal half-lives. In patients with type 2 diabetes, concentrations declined slowly after Tmax with half-lives of 147.3 to 673.3 hours (6.1 to 28.1 days) [2]. In adults with overweight or obesity in two higher-dose cohorts, half-life ranged from 174.8 to 1075.7 hours (7.3 to 44.8 days) [3]. Both studies were small (42 and 24 participants), so these ranges are imprecise [2][3]. They are consistent with the once-weekly regimen used in all later trials [5].

Development history and timeline

Mazdutide began as LY3305677 at Eli Lilly, which ran the first phase 1 study in healthy participants from December 2016 [12]. Innovent Biologics entered an exclusive licence agreement with Lilly for development and potential commercialisation in China [10], and ran the Chinese phase 1b to phase 3 programme [2][5]. Lilly has continued US development, publishing high-dose phase 1 and phase 2 obesity trials [11][7].

Year	Milestone	Ref
2016	First Lilly phase 1 study (LY3305677) in healthy participants starts	[12]
2022	Phase 1b results in Chinese patients with type 2 diabetes and in adults with obesity	[2][3]
2023	Phase 2 obesity results (China)	[13]
2024	Phase 2 type 2 diabetes results (China)	[14]
2025	GLORY-1 phase 3 published in NEJM; US high-dose phase 1 published	[5][11]
27 Jun 2025	NMPA approval for chronic weight management announced	[1][15]
19 Sep 2025	NMPA approval for glycaemic control in type 2 diabetes announced	[10]
Oct 2025	DREAMS-3 head-to-head topline versus semaglutide announced	[16]
2026	DREAMS-1 and DREAMS-2 (Nature), GLORY-2 (JAMA) and US phase 2 (Lancet Diabetes & Endocrinology) published	[6][17][18][7]

Mazdutide clinical trials: key studies and findings

Study	Phase / design	Population	n	Duration	Primary endpoint	Main finding	Ref
NCT04466904	1b, three cohorts vs placebo and open-label dulaglutide	Chinese type 2 diabetes	42 treated	12 weeks	Safety	HbA1c and glucose fell in all cohorts	[2]
NCT04440345	1b, two higher-dose cohorts vs placebo	Chinese overweight or obesity	24	12 or 16 weeks	Safety	Weight -11.7% vs -1.8% (12-week cohort)	[3]
NCT04904913	2, double-blind	Chinese overweight or obesity	248	24 weeks	% weight change	-6.7% to -11.3% vs +1.0% placebo	[13]
Phase 2 T2D (China)	2, double-blind; open-label dulaglutide	Chinese type 2 diabetes	250	20 weeks	HbA1c change	-1.41 to -1.67 points vs -1.35 dulaglutide, +0.03 placebo	[14]
GLORY-1 (NCT05607680)	3, double-blind	Chinese overweight or obesity	610	48 weeks	% weight change at week 32	Week 48: -11.00% and -14.01% vs +0.30%	[5]
GLORY-2 (NCT06164873)	3, double-blind, 2:1	Chinese adults, BMI 30 or more	461	60 weeks	% weight change at week 60	-16.65% vs -1.50%	[18]
DREAMS-1	3, monotherapy vs placebo	Chinese type 2 diabetes, diet and exercise only	320	24 + 24 weeks	HbA1c change	-1.57 and -2.15 points vs -0.14	[6]
DREAMS-2	3, vs dulaglutide	Chinese type 2 diabetes on oral agents	731	28 weeks	HbA1c change	Superior to dulaglutide on HbA1c and weight	[17]
DREAMS-3 (NCT06184568)	3, open-label vs semaglutide	Chinese type 2 diabetes with obesity	349	32 weeks	HbA1c below 7% plus 10% or more weight loss	Topline only: weight -10.29% vs -6%	[19][16]
NCT06124807	2, double-blind	US adults with obesity or overweight, no diabetes	179	48 weeks	% weight change at week 32	-7.3% to -18.1% vs -0.9%	[7]
NCT05623839	1, two escalation schemes vs placebo	US adults with overweight or obesity	32	20 weeks	Safety	Weight -20.0% and -21.0% vs -0.1%	[11]

GLORY-1: weight management in China

GLORY-1 randomised 610 adults aged 18 to 75 with a BMI of at least 28 kg/m², or 24 to under 28 with a weight-related condition, 1:1:1 to two mazdutide maintenance regimens or placebo for 48 weeks [5]. Mean baseline weight was 87.2 kg and BMI 31.1 [5]. At the week 32 primary timepoint (treatment-policy estimand) weight changed by -10.09% and -12.55% versus +0.45% with placebo, and 73.9%, 82.0% and 10.5% lost at least 5% [5]. At week 48 the changes were -11.00% and -14.01% versus +0.30%, with 35.7%, 49.5% and 2.0% losing at least 15% [5]. This trial was the main basis for the June 2025 NMPA approval [1]. Evidence level: phase 3.

GLORY-2: higher-dose regimen in moderate to severe obesity

GLORY-2 enrolled Chinese adults with a BMI of 30 kg/m² or more, with or without type 2 diabetes, at 27 hospitals between December 2023 and November 2025, and randomised them 2:1 to a higher once-weekly mazdutide maintenance dose or placebo for 60 weeks [18]. Of 461 treated participants, 64.0% were female, 16.1% had type 2 diabetes and mean BMI was 34.3 [18]. At week 60, weight changed by -16.65% versus -1.50% (difference -15.15 percentage points), and 84.3% versus 33.1% lost at least 5% [18]. Evidence level: phase 3.

Earlier phase 2 obesity trials

In the first phase 2 obesity trial, 248 Chinese adults were randomised to three mazdutide dose levels or placebo for 24 weeks; weight changed by -6.7%, -10.4% and -11.3% versus +1.0% [13]. A separate higher-dose phase 2 trial in 80 adults with a BMI of 30 kg/m² or more reported -12.78% versus +1.80% at week 24, with 81.7% of the mazdutide group and no placebo participants reaching 5% weight loss [20]. Evidence level: phase 2.

DREAMS-1 and DREAMS-2: type 2 diabetes

DREAMS-1 tested mazdutide monotherapy in 320 Chinese adults with type 2 diabetes managed by diet and exercise alone (mean HbA1c 8.24%, diabetes duration 1.9 years) [6][10]. At week 24, HbA1c fell by 1.57 and 2.15 percentage points in the two mazdutide arms versus 0.14 with placebo, and weight changed by -5.61% and -7.81% versus -1.26% [6].

DREAMS-2 randomised 731 participants on background oral agents to two mazdutide regimens or dulaglutide for 28 weeks [17]. Both mazdutide arms were non-inferior and superior to dulaglutide for HbA1c (least-squares mean differences -0.24 and -0.30 percentage points) and gave greater weight loss (differences -3.78 and -5.76 percentage points) [17]. These two trials supported the September 2025 type 2 diabetes approval [10]. Evidence level: phase 3.

DREAMS-3: head-to-head with semaglutide (topline only)

DREAMS-3 is an open-label phase 3 trial that randomised 349 Chinese adults with type 2 diabetes and obesity 1:1 to mazdutide or semaglutide for 32 weeks, with a composite primary endpoint of HbA1c below 7.0% plus at least 10% weight loss [19]. The developer’s topline announcement reported HbA1c changes of -2.03 versus -1.84 percentage points and mean weight changes of -10.29% versus -6% [16]. These are company-reported topline data, not yet peer reviewed, from an open-label trial. Evidence level: phase 3 topline.

US development

A US phase 2 trial randomised 179 adults without diabetes at 24 US centres to three mazdutide regimens, including higher doses than those used in China, or placebo [7]. At week 32 (efficacy estimand), weight changed by -7.3%, -15.6% and -18.1% versus -0.9%, with further reductions by week 48 [7]. In an earlier 20-week phase 1 study in 32 US adults, two escalation schemes to a high target dose gave -20.0% and -21.0% versus -0.1% [11]. Evidence level: phase 2 and phase 1.

Safety findings reported in mazdutide studies

Gastrointestinal events were the most frequent adverse events in every published trial.

• Phase 2 type 2 diabetes: diarrhoea 36%, decreased appetite 29%, nausea 23%, vomiting 14% and hypoglycaemia 10% (8% with placebo) [14].
• GLORY-1: adverse events were mostly gastrointestinal and mild to moderate; discontinuation for adverse events was 1.5% and 0.5% with mazdutide versus 1.0% with placebo [5].
• GLORY-2: vomiting 53.1% versus 1.3%, nausea 46.9% versus 3.2% and diarrhoea 39.4% versus 6.5%; discontinuation for adverse events 2.9% versus 0% [18].
• Higher-dose phase 2 (China): nausea 50.0% versus 0%, diarrhoea 38.3% versus 10.0%, vomiting 36.7% versus 10.0% [20].
• US phase 2: discontinuation for adverse events was highest in the top dose arm at 20%, mainly gastrointestinal [7].
• Phase 1b type 2 diabetes: diarrhoea 29.2% with mazdutide versus 33.3% with dulaglutide and 0% with placebo [2].

Regulatory status of mazdutide by region

Region	Status (October 2026)	Ref
China	Approved by NMPA: chronic weight management in adults with BMI of 28 kg/m² or more, or 24 or more with a weight-related condition (announced 27 June 2025); glycaemic control in adults with type 2 diabetes (announced 19 September 2025)	[1][15][10]
United States	Investigational; phase 2 completed	[7]
European Union	Investigational; no EMA authorisation	[7]
United Kingdom	Investigational; no marketing authorisation identified	[7]
India	Investigational; no marketing authorisation identified	[7]

The Chinese approvals cover mazdutide as a finished medicine sold by the licence holder. They do not change the status of research-grade material, which remains for laboratory use only.

Mazdutide in laboratory research

• Analytical identity. A 2026 validated LC-HRMS method identifies and quantifies mazdutide together with eight other GLP-1-based peptides (including survodutide, retatrutide and tirzepatide) using accurate-mass full scan with data-dependent MS/MS on a C18 column, reaching mass accuracy of 0.0 to 0.5 ppm and fragment coverage of 76% to 100% [21]. Because database formulas disagree, intact-mass confirmation should be checked against 4563.07 g/mol average mass [4].
• Biomarkers of dual agonism. Clinical studies used fasting amino acids and beta-hydroxybutyrate as markers of glucagon receptor engagement, alongside glucose, insulin and C-peptide for GLP-1R effects [11]. These markers translate to rodent studies.
• Pharmacokinetic sampling. Half-lives of one to several weeks mean washout and sampling windows must be long; phase 1b studies characterised Cmax, Tmax, terminal half-life and AUC over the first 168 hours [2][3].

Purity is normally assessed by reversed-phase HPLC or UPLC, and identity by LC-MS intact mass plus peptide mapping across the side-chain-bearing Lys20 fragment.

Synthesis, purity and impurities

Mazdutide is produced by chemical synthesis, typically Fmoc solid-phase peptide synthesis (SPPS) with an orthogonally protected Lys20 for on-resin assembly of the AEEA-AEEA-gamma-Glu-C20 diacid side chain. See SPPS versus LPPS for route considerations.

Structure-linked impurity classes include deletion sequences along the 34-residue chain (a des-Ser variant has exactly the formula of the alternative database entry, so intact-mass methods must resolve it), D-amino acid variants, Gln3 deamidation, Trp25 oxidation, aspartimide-related variants at Asp9 and Asp15, and incomplete or truncated side chains. Purity thresholds and impurity reporting are covered in peptide API impurities and specifications.

Storage and stability of lyophilized mazdutide

Research mazdutide is supplied as a lyophilized (freeze-dried) powder. Typical laboratory practice is storage at -20 °C or below, protected from light and moisture, with vials equilibrated to room temperature before opening. Like other C20-acylated peptides, it can self-associate and adsorb to surfaces in dilute solution, so assay buffers often include a carrier protein. See lyophilised peptide stability and storage.

Mazdutide vs tirzepatide

Mazdutide and tirzepatide share a GLP-1 receptor arm and a Gly-Gly-Pro-Ser-Ser-Gly motif near the C-terminus, but differ in their second receptor: glucagon for mazdutide and GIP for tirzepatide [4][22][23].

Feature	Mazdutide	Tirzepatide
Receptors	GLP-1R + GCGR [4]	GIPR + GLP-1R [23]
Main-chain length	34 amino acids [4]	39 amino acids [22]
Molecular formula	C210H322N46O67 [4]	C225H348N48O68 [22]
Approval	China only (2025) [1][10]	FDA, chronic weight management (2023) [23]
Large obesity trial	GLORY-1, China, 48 weeks: -14.01% (higher arm) vs +0.30% [5]	SURMOUNT-1, global, 72 weeks: -20.9% (highest arm) vs -3.1% [24]
US phase 2	32 weeks, efficacy estimand: up to -18.1% vs -0.9% [7]	Not applicable (phase 3 completed)

No head-to-head trial of mazdutide versus tirzepatide has been published. GLORY-1 and SURMOUNT-1 differ in population (Chinese adults with lower baseline BMI versus a multinational population), duration (48 versus 72 weeks) and dose ranges, and the US phase 2 used an efficacy estimand at 32 weeks, so the percentages in the table should not be ranked [5][24][7]. The open question is whether glucagon receptor agonism adds hepatic and energy-expenditure effects beyond GIP co-agonism; survodutide, the other glucagon/GLP-1 dual agonist in late development, is reviewed in survodutide: structure, mechanism and research findings. Tirzepatide supply is covered on our tirzepatide page.

How Winhycare Pharma supplies mazdutide

Winhycare Pharma supplies mazdutide as research-use-only material to research organisations and laboratories, at a purity of 99% or higher as stated in our product brochure, with a certificate of analysis covering HPLC purity and mass confirmation against the 34-residue structure. Custom synthesis of mazdutide analogues, labelled variants and fragments is available. Product details are on the mazdutide research peptide page, and enquiries go through our research peptide contact form.

For laboratory research use only. Not for human or veterinary use.