Mazdutide (IBI362): weight-loss & diabetes drug
A lab-made gut-hormone mimic that curbs appetite and helps burn liver fat; being studied for obesity, type 2 diabetes, and fatty liver disease.
- Class
- Dual GLP-1/glucagon receptor agonist (incretin mimetic); oxyntomodulin analog
- Status
- Approved prescription drug (China / NMPA, 2025) — not approved by FDA, EMA, Health Canada, MHRA, or any other major regulator outside China
- Best-supported effect
- Significant body-weight reduction in Chinese adults with obesity (Phase 3 RCTs); improved glycemic control in Chinese adults with type 2 diabetes (Phase 3 RCTs)
- Main caveat
- All pivotal trials enrolled exclusively Chinese participants; no completed Western regulatory review, no cardiovascular outcomes trial, and no independent replication outside the Chinese development program
A researcher, an agent, or an algorithm wrote down the sequence and picked a target to hit.
An AI model like OpenFold3 or AlphaFold built a 3D structure and scored how well it fits the binding site.
A second contributor repeated the computation on their own hardware and the scores matched.
A chemistry service or a researcher ordered the sequence, it was manufactured, and mass spectrometry confirmed the right molecule was produced.
A binding or activity measurement confirmed that it actually does what the computer predicted — or didn't.
What this is
Mazdutide (also known as IBI362 and LY3305677, brand name Xinermei) is a synthetic analog of oxyntomodulin — a naturally occurring gut hormone that acts on two receptors simultaneously: the GLP-1 receptor and the glucagon receptor. This dual mechanism sets it apart from GLP-1/GIP dual agonists like tirzepatide: where tirzepatide pairs GLP-1 with GIP, mazdutide pairs GLP-1 with glucagon, a combination intended to reduce appetite while also increasing energy expenditure and promoting fat burning in the liver. To enable once-weekly dosing, the molecule carries a C18 fatty acid side chain attached to the oxyntomodulin backbone — that acylation is what extends its half-life to support weekly injection and is not represented in the raw 30-residue sequence stored here. Mazdutide was originally discovered by Eli Lilly under the code LY3305677; in 2019 Lilly licensed Greater China rights to Innovent Biologics, which advanced the compound under the code IBI362. In June 2025, China's NMPA approved it as Xinermei for chronic weight management — the first regulatory approval anywhere of a dual GLP-1/glucagon receptor agonist — followed by a type 2 diabetes indication in September 2025. Eli Lilly retains rights outside Greater China but has not announced a US or EU filing, and no Western approval pathway is currently active.
History
Mazdutide traces its origins to Eli Lilly's exploration of oxyntomodulin-derived dual agonists, where the compound was designated LY3305677. In 2019, Lilly out-licensed Greater China rights (mainland China, Hong Kong, Macau, and Taiwan) to Innovent Biologics, which took it forward under the clinical code IBI362. Innovent ran two parallel Phase 3 program families in Chinese populations: GLORY for obesity and DREAMS for type 2 diabetes. Both programs read out across 2024 and 2025. China's NMPA granted approval for chronic weight management in June 2025, making mazdutide the first dual GLP-1/glucagon receptor agonist to receive regulatory approval anywhere in the world. A type 2 diabetes indication followed from the NMPA in September 2025 under the brand name Xinermei. Lilly retains ex-Greater China rights but has not signaled a US or EU development path, in part because tirzepatide already occupies that portfolio space.
What it does
Mazdutide reduces body weight and improves blood sugar control by working through two complementary pathways at once. The GLP-1 receptor component tells the brain that the body has eaten enough, slows digestion, and increases insulin release in response to meals. The glucagon receptor component signals the liver to burn more fat and use more energy at rest — a lever that GLP-1-only drugs do not pull. In Phase 3 trials conducted in Chinese adults with obesity, this dual action produced weight loss of 14.0% at the 6 mg dose over 48 weeks (GLORY-1, n=610) and 20.1% at the 9 mg dose over 60 weeks (GLORY-2, n=462). In Chinese adults with type 2 diabetes, DREAMS-1 (n=320, 24 weeks) showed a mean HbA1c reduction of 2.02 percentage points at 6 mg, and DREAMS-3 (n=349, 32 weeks) showed superiority over semaglutide on a combined endpoint of HbA1c below 7% plus at least 10% weight loss, with 48% of mazdutide participants versus 21% on semaglutide reaching that benchmark. A concern sometimes raised about glucagon co-agonism — that it would raise blood sugar in diabetic patients — was not borne out: the GLP-1 component dominates net glycemic effect at therapeutic doses.
Evidence
- Human: Multiple Phase 3 RCTs in Chinese adults. GLORY-1 (n=610, 48 weeks, obesity): 14.0% mean weight loss at 6 mg; 49.5% achieved at least 15% body-weight reduction. GLORY-2 (n=462, 60 weeks, obesity): 20.1% mean weight loss at 9 mg. DREAMS-1 (n=320, 24 weeks, type 2 diabetes): HbA1c −2.02 percentage points at 6 mg. DREAMS-2 (n=731, type 2 diabetes): superior to dulaglutide on glycemic control. DREAMS-3 (n=349, 32 weeks, type 2 diabetes vs. semaglutide): 48% versus 21% reached the combined endpoint of HbA1c <7% plus ≥10% weight loss. All pivotal trials enrolled exclusively Chinese participants; no independent Western replication has been published (Madsbad and colleagues, Expert Opinion on Investigational Drugs, 2025).
- Animal: Preclinical pharmacology supporting dual GCG/GLP-1 agonism and oxyntomodulin receptor biology is established; not the primary evidence base for the approved indications.
- In vitro: Receptor binding and pharmacology characterization from Phase 1b work; not individually detailed in available sources.
Myths and misconceptions
- "Glucagon agonism raises blood sugar, making mazdutide unsafe in diabetes." The DREAMS Phase 3 program showed substantial HbA1c reductions across multiple trials in Chinese adults with type 2 diabetes, including superiority over semaglutide on a combined glycemic-and-weight endpoint in DREAMS-3. The GLP-1 component dominates net glycemic effect at therapeutic doses; the glucagon component contributes hepatic fat oxidation and energy expenditure without producing net hyperglycemia.
- "Mazdutide and survodutide are effectively the same drug." Both are dual GLP-1/glucagon receptor agonists derived from oxyntomodulin biology, but they are distinct molecules with different sequences, receptor potency profiles, and developers (Innovent/Lilly versus Boehringer Ingelheim). They have not been compared head-to-head in any published trial.
- "Mazdutide is available outside China through grey-market suppliers." Xinermei is the licensed Innovent product manufactured under regulated pharmaceutical conditions. Material sold under the mazdutide name outside China cannot be assumed to share sequence, salt form, purity, or potency with the approved drug.
- "Lilly will file mazdutide in the US." Lilly holds ex-Greater China rights but has not announced a US or EU regulatory submission. The lack of a public Western filing timeline likely reflects portfolio overlap with tirzepatide.
Known effects
- Body-weight reduction (obesity) — Phase 3 RCTs (Chinese population); NMPA-approved indication
- Glycemic control improvement (type 2 diabetes) — Phase 3 RCTs (Chinese population); NMPA-approved indication
- Hepatic fat oxidation and increased resting energy expenditure — Mechanistic (glucagon receptor pathway); not isolated as a standalone clinical endpoint in published trials
- GI motility slowing — Class effect of GLP-1 receptor agonism; consistent with trial adverse event profiles
Safety signals
Nausea is the most commonly reported adverse event across the GLORY and DREAMS programs, described as dose-dependent and generally mild to moderate, consistent with the broader GLP-1 receptor agonist class. Vomiting, diarrhea, and decreased appetite were also reported. Transient elevation of hepatic glucose output is a mechanistic consequence of glucagon receptor agonism; in Phase 3 the net glycemic effect was HbA1c reduction, not hyperglycemia. Mazdutide carries the same class-level boxed warning concern regarding medullary thyroid carcinoma (MTC) that applies to GLP-1 receptor agonists as a group, based on rodent C-cell tumor findings; no mazdutide-specific human thyroid cancer data have been extracted. A rare pancreatitis signal applies by class analogy. No completed cardiovascular outcomes trial has been published for mazdutide; long-term cardiovascular, gallbladder, and diabetic retinopathy signals require multi-year post-marketing surveillance data beyond the 24–60-week Phase 3 durations.
Regulatory status
- China (NMPA): Approved prescription drug — chronic weight management (June 2025) and type 2 diabetes (September 2025) under the brand name Xinermei. This is the first regulatory approval anywhere of a dual GLP-1/glucagon receptor agonist.
- US (FDA): Not approved; no regulatory submission announced. Eli Lilly holds ex-Greater China rights but has not filed in the US; no FDA-approved, compounded, or imported form is legally available domestically.
- EU (EMA): Not approved; no filing announced.
- UK (MHRA), Canada, Australia (TGA): Not approved; no public timeline for ex-China filings.
- WADA: Likely prohibited under S2 (peptide hormones, growth factors, related substances and mimetics) in jurisdictions where approved, and under S0 (non-approved substances) in jurisdictions where not approved; current list status should be independently verified for athletes subject to the WADA code.
Mechanism
Mazdutide is a synthetic analog of oxyntomodulin, a naturally occurring gut-derived peptide hormone that activates both the GLP-1 receptor (GLP-1R) and the glucagon receptor (GCGR). A C18 fatty acid acylation extends plasma half-life to support once-weekly subcutaneous dosing; this lipid chain is not part of the raw backbone sequence.
GLP-1 receptor pathway: GLP-1R activation in the hypothalamus suppresses appetite; in the pancreas it enhances glucose-dependent insulin secretion; in the gastrointestinal tract it slows gastric emptying. The net effect at therapeutic doses is reduced caloric intake and improved postprandial glycemic control.
Glucagon receptor pathway: GCGR activation in hepatocytes promotes fatty acid oxidation and increases resting energy expenditure while also stimulating glycogenolysis. The GLP-1 component dominates net glycemic effect at therapeutic doses, preventing the hyperglycemia that would otherwise result from GCGR-mediated hepatic glucose output.
The dual mechanism is proposed to confer additive benefit over GLP-1-only agonists: reduced food intake from GLP-1R agonism combined with increased energy expenditure and hepatic fat oxidation from GCGR agonism. This is mechanistically distinct from tirzepatide's GLP-1/GIP dual agonism. The Phase 3 GLORY and DREAMS results — 20.1% weight loss at 9 mg in GLORY-2 and substantial HbA1c reductions across DREAMS — support the mechanistic hypothesis at the endpoint level, though direct head-to-head comparisons with GLP-1/GIP dual agonists or GLP-1-only agents on a mechanism-isolation basis have not been published.
Open questions
- Generalizability beyond Chinese populations: All pivotal Phase 3 trials enrolled exclusively Chinese participants. Whether the 20.1% weight-loss result and DREAMS glycemic outcomes translate to other ethnic groups, dietary patterns, and healthcare systems remains unknown.
- Cardiovascular outcomes: No completed cardiovascular outcomes trial (CVOT) has been published for mazdutide. Whether it produces MACE reduction comparable to semaglutide or tirzepatide is not established.
- Long-term safety: Phase 3 trial durations were 24–60 weeks. Pancreatitis incidence, gallbladder disease, diabetic retinopathy progression, and rare oncologic signals require multi-year post-marketing surveillance.
- Hepatic outcomes (MASH/MASLD): Given GCGR-mediated hepatic fat oxidation, dedicated trials in metabolic-associated steatohepatitis or steatotic liver disease would be informative; none have been published.
- Head-to-head with tirzepatide and retatrutide: No direct comparison between mazdutide and tirzepatide (GLP-1/GIP) or retatrutide (GLP-1/GIP/GCGR triple agonist) has been published. The comparative benefit of glucagon co-agonism over GIP co-agonism cannot be estimated from available data.
- Body composition at 9 mg: Lean-mass versus fat-mass loss data at the high-dose 9 mg regimen are not published in detail in available sources.
- Western regulatory filing: Whether and when Lilly will seek US or EU approval remains unknown.
Related peptides
- Survodutide — also a dual GLP-1/glucagon receptor agonist derived from oxyntomodulin biology; developed by Zealand Pharma and Boehringer Ingelheim; in global Phase 3 (SYNCHRONIZE program)
- Semaglutide — GLP-1-only receptor agonist; the most-studied GLP-1 agonist and the active comparator in DREAMS-3
- Tirzepatide — dual GLP-1/GIP receptor agonist; mechanistically distinct from mazdutide (GIP rather than glucagon co-agonism); FDA- and EMA-approved
- Retatrutide — triple GLP-1/GIP/glucagon receptor agonist; adds glucagon agonism on top of GLP-1/GIP, making it a superset of mazdutide's receptor profile
Research directions for this peptide, selected from the current sources — hypotheses you can explore and model. None of it is proven yet; tap any one to see the full thinking.
Do mazdutide's two receptor targets each work in a different organ, making the combination more powerful than either alone?
If this is true, patients with both obesity and fatty liver disease could get two independent benefits from a single weekly injection, potentially making it more effective than existing GLP-1 drugs for liver complications.
Could changing one building block near the start of the mazdutide chain shift its action more toward burning liver fat and less toward suppressing appetite?
This could lead to a version of the drug that is better tolerated (less nausea) while still treating the fatty liver complications of obesity, expanding the patient population who can benefit.
After losing weight on mazdutide, could patients stay on a lower dose and still keep the weight off because the drug also raises their metabolism?
If true, patients could reduce their dose over time, lowering side effects and cost while still maintaining the weight loss, which would be a major advantage over current weekly GLP-1 injections.
Could mazdutide help control blood sugar in the type of diabetes that starts in adulthood as an autoimmune disease, where some insulin-making cells still work?
If so, patients with this often-missed form of diabetes could have a once-weekly injection option that stabilizes blood sugar without the hypoglycemia risk of insulin, potentially slowing the loss of their remaining insulin-producing cells.
▸full evidence table2 metrics
| metric | value | tool |
|---|---|---|
| ipTM | 0.8440924286842346 | openfold3-mlx |
| ranking score | 0.8946712613105774 | openfold3-mlx |
▸structural qualityopenfold3
| metric | value | note |
|---|---|---|
| gpde | 0.708 | global PDE — lower = better |
| disorder | 0.142 | fraction disordered |
| chain pair ipTM (A, B) | 0.844 | interface quality |
▸3-letter notation
▸recipeopenfold3-mlx 0.3.1
| parameter | value |
|---|---|
| model | openfold3-mlx 0.3.1 |
| weights | aedd8f3eb814e392… |
| hardware | apple_m4_base_16gb |
| mlx version | 0.31.1 |
| python | 3.14.3 |
| random seed | 42 |
| msa strategy | colabfold |
| diffusion samples | 1 |
| runtime | 453s |
| predicted by | mlx@peptide |
| predicted at | 2026-04-30 |
python3 openfold3/run_openfold.py predict --query_json {query.json} --runner_yaml examples/example_runner_yamls/mlx_runner.yml --output_dir {output_dir} --num_diffusion_samples 1 ▸citationbibtex
@peptide{pep10904,
sequence = {HSQGTFTSDYSKYLDSRRAQDFVQWLMNTX},
target = {gcgr},
author = {peptidemodel},
year = {2026},
status = {computed}
}