Cotadutide: experimental weight-loss & liver-disease drug (MEDI0382)
An experimental once-daily injectable developed by AstraZeneca that targets two gut hormones to lower blood sugar, reduce weight, and treat fatty liver disease; not yet an approved drug.
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
Cotadutide (development code MEDI0382) is an experimental once-daily injectable peptide that activates two related gut-hormone receptors at the same time: the GLP-1 receptor — the same target as Ozempic and Victoza — and the glucagon receptor. It was developed by MedImmune/AstraZeneca for type 2 diabetes, obesity, fatty liver disease (NASH/MASH), and diabetic kidney disease. The stored 31-residue chimeric sequence is the bare peptide backbone built from key residues of both glucagon and GLP-1; the active drug additionally carries a C16 palmitic-acid chain tethered to an internal lysine via a γ-glutamate spacer, which lets it bind albumin and circulate long enough for once-daily dosing (Henderson and colleagues, Diabetes, Obesity and Metabolism, 2016). AstraZeneca terminated the cotadutide program in April 2023 in favour of a once-weekly successor, so cotadutide reached Phase 2b but never Phase 3 and is not approved.
History
Cotadutide came out of a structure-based engineering program at MedImmune aimed at combining glucagon's hepatic lipid-burning activity with GLP-1's glucose-lowering safety in a single molecule. The molecule was designed with a roughly five-fold bias toward the GLP-1 receptor over the glucagon receptor, intended to keep the glucagon arm's hepatic benefit without provoking clinically significant hyperglycaemia (Henderson and colleagues, 2016). First-in-human Phase 1/2a data in patients with type 2 diabetes were published by Ambery and colleagues in The Lancet in 2018, showing reductions in post-meal glucose and body weight versus placebo over 41 days. Cotadutide then became the first GLP-1/glucagon dual agonist to generate meaningful Phase 2b data in metabolic-disease populations (Nahra and colleagues, Diabetes Care, 2021). In April 2023, AstraZeneca announced it was stopping cotadutide development to focus instead on a once-weekly GLP-1/glucagon dual-agonist successor, citing the shift in the diabetes and obesity market toward weekly dosing rather than any safety concern.
What it does
Cotadutide simultaneously turns on two receptors that sit at the centre of energy and liver metabolism. Through the GLP-1 receptor it lowers blood glucose by stimulating glucose-dependent insulin secretion, slows gastric emptying, and suppresses appetite. Through the glucagon receptor it acts on the liver to increase fat oxidation and reduce hepatic lipid accumulation. In clinical trials this translated into lower HbA1c, weight loss, and reductions in liver fat and liver enzymes (ALT/AST) in adults with type 2 diabetes and overweight or obesity (Nahra and colleagues, Diabetes Care, 2021). The 5:1 GLP-1:glucagon-receptor bias was selected so the GLP-1 arm's insulin-secretion effect counterbalances any glucagon-driven rise in hepatic glucose output (Henderson and colleagues, 2016; Boland and colleagues, Nature Metabolism, 2020).
Evidence
Human. The Phase 2b trial by Nahra and colleagues (Diabetes Care, 2021) randomised 834 adults with type 2 diabetes inadequately controlled on metformin to cotadutide 100 µg, 200 µg, or 300 µg once daily, placebo, or open-label liraglutide 1.8 mg for 54 weeks. Cotadutide significantly reduced HbA1c and body weight versus placebo at week 14 and week 54 (all P < 0.001); ad-hoc analyses also showed improvements in hepatic parameters. A mechanistic Phase 2a study (Parker and colleagues, Nature Metabolism, 2023) directly compared cotadutide with liraglutide in people with type 2 diabetes and obesity and reported that cotadutide reduced fasting hepatic glycogen and produced about 12.5% greater postprandial glucose reduction than liraglutide — evidence that the glucagon arm contributed real, GLP-1-distinct hepatic effects. In diabetic kidney disease, Selvarajah and colleagues (Kidney International, 2024) reported that 300 µg and 600 µg once-daily cotadutide reduced urinary albumin-to-creatinine ratio by roughly 44% and 50% versus placebo on top of standard care.
Animal. Boland and colleagues (Nature Metabolism, 2020) showed that cotadutide resolved NASH and hepatic fibrosis in two mouse models, with the liver-specific benefits — reduced hepatic lipid content, improved mitochondrial function, normalised glycogen flux — attributable to the glucagon-receptor arm rather than to GLP-1 signalling alone. Henderson and colleagues (2016) had earlier shown anti-obesity and metabolic effects in diet-induced obese mice and cynomolgus monkeys, with weight loss exceeding that of liraglutide at comparable doses.
Background. Cotadutide sits within the broader pharmacology of proglucagon-derived peptides, reviewed by Lafferty and colleagues (Frontiers in Endocrinology, 2021).
Known effects
- Glycaemic control (HbA1c reduction) — Phase 2b in type 2 diabetes (Nahra 2021)
- Body weight reduction — Phase 2b (Nahra 2021)
- Hepatic fat and liver-enzyme reduction — Phase 2b ad-hoc analyses (Nahra 2021)
- Hepatic glycogen modulation and superior postprandial glucose reduction vs liraglutide — mechanistic Phase 2a (Parker 2023)
- Urinary albumin reduction in diabetic kidney disease — Phase 2b (Selvarajah 2024)
- NASH histological improvement and fibrosis reversal — preclinical (Boland 2020)
- Anti-obesity and metabolic effects in non-human primates — preclinical (Henderson 2016)
Safety signals
Gastrointestinal adverse events — nausea and vomiting — were the dominant tolerability signal in cotadutide trials, consistent with the GLP-1 receptor agonist class. In the Phase 2b diabetic kidney disease trial, treatment-discontinuation rates from adverse events reached approximately 21% at the highest 600 µg dose, with most events gastrointestinal (Selvarajah and colleagues, Kidney International, 2024). Glucagon co-activation did not produce clinically significant hyperglycaemia at the 5:1 GLP-1:glucagon-receptor bias used (Henderson and colleagues, 2016; Nahra and colleagues, 2021). The palmitoyl-γ-glutamate albumin-binding modification — chemically analogous to the design used in liraglutide — was generally well tolerated.
Regulatory status
- US/EU: Not approved. Development was active through Phase 2b across type 2 diabetes, NASH/MASH, and diabetic kidney disease.
- Program status: AstraZeneca announced termination of the cotadutide program in April 2023, redirecting effort to a once-weekly GLP-1/glucagon dual-agonist successor. No Phase 3 trial was initiated.
Mechanism
Cotadutide is a 31-residue chimeric peptide whose sequence draws structural elements from both glucagon and GLP-1, with a γ-glutamate-linked C16 palmitic acid attached to an internal lysine for reversible albumin binding and once-daily pharmacokinetics (Henderson and colleagues, Diabetes, Obesity and Metabolism, 2016). The peptide engages the GLP-1 receptor and the glucagon receptor — both Gαs-coupled — with a roughly 5:1 potency bias toward GLP-1R. At pancreatic β-cells the GLP-1R arm drives glucose-dependent insulin secretion; at hypothalamic appetite circuits it suppresses food intake; at the stomach it delays gastric emptying. The glucagon-receptor arm acts predominantly on hepatocytes, where it increases fatty-acid oxidation, reduces de novo lipogenesis, and modulates glycogen turnover. Boland and colleagues (Nature Metabolism, 2020) used receptor-specific dissection in mouse models to show that cotadutide's hepatic lipid-lowering and mitochondrial-function effects depend on the glucagon-receptor arm and are not reproduced by GLP-1R activation alone, while body-weight and food-intake effects are predominantly GLP-1R-mediated. Parker and colleagues (Nature Metabolism, 2023) demonstrated in humans that cotadutide reduces fasting hepatic glycogen and produces greater postprandial glucose reduction than liraglutide at comparable GLP-1R-mediated weight loss — direct human-pharmacology evidence that the glucagon arm contributes biology beyond GLP-1 monoagonism.
Open questions
- Whether cotadutide's 5:1 GLP-1:glucagon-receptor bias is the right balance for hepatic-fibrosis resolution, or whether a more glucagon-weighted ratio provides deeper liver benefit.
- Whether AstraZeneca's 2023 termination reflected the competitive shift to weekly dosing and the GLP-1 monoagonist explosion, or whether internal Phase 2b readouts on registration-grade liver-fibrosis endpoints fell short.
- Whether the kidney-protective albumin-reduction signal seen in the diabetic kidney disease Phase 2b (Selvarajah 2024) translates to the once-weekly successor dual-agonist program.
Related peptides
- Liraglutide — single-target GLP-1 receptor agonist using the same palmitoyl-γ-Glu-Lys albumin-binding chemistry; cotadutide's direct GLP-1-only comparator in the Phase 2b trial (Nahra 2021).
- Semaglutide — once-weekly GLP-1 receptor agonist; the dominant GLP-1 monoagonist whose commercial success contributed to AstraZeneca's strategic deprioritisation of cotadutide.
- Tirzepatide — once-weekly GLP-1/GIP dual agonist; a different dual-agonist pairing that achieved Phase 3 success in diabetes and obesity while cotadutide did not.
- Survodutide, pemvidutide, mazdutide — competing GLP-1/glucagon dual agonists with different receptor-bias ratios, still in active clinical development as of 2024.
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.
Does cotadutide's glucagon-activating component fix liver damage through a separate biological route, on top of what weight loss achieves?
If so, patients with fatty liver disease could benefit from drugs like cotadutide even if they lose only modest weight, offering a new treatment path for a condition that has very limited options today.
Do a handful of specific residues in cotadutide determine whether the drug leans more toward the GLP-1 side or the glucagon side, and could disease change that balance?
If true, drug designers could fine-tune future weight-loss and liver medicines to be safer in people whose livers have unusually high glucagon sensitivity, reducing the risk of blood-sugar spikes.
Could a drug abandoned for diabetes still be useful for preventing the liver cancers that fatty liver disease can cause?
Fatty liver disease is now one of the leading causes of liver cancer worldwide, and there are almost no preventive drugs. If cotadutide-type medicines could interrupt that progression, millions of patients with chronic liver disease might have a new protective option.
Could engineers modify cotadutide to also repair the gut lining, which often drives the liver damage that cotadutide targets?
Many patients with fatty liver disease also have a leaky gut that feeds inflammation into the liver. A drug addressing both problems at once could work better than current medicines and might be needed in smaller doses, potentially reducing side effects.
▸full evidence table2 metrics
| metric | value | tool |
|---|---|---|
| ipTM | 0.8578919172286987 | openfold3-mlx |
| ranking score | 0.9070388674736023 | openfold3-mlx |
▸structural qualityopenfold3
| metric | value | note |
|---|---|---|
| gpde | 0.687 | global PDE — lower = better |
| disorder | 0.142 | fraction disordered |
| chain pair ipTM (A, B) | 0.858 | interface quality |
▸3-letter notation
▸recipeopenfold3-mlx 0.3.1
| parameter | value |
|---|---|
| model | openfold3-mlx 0.3.1 |
| weights | — |
| hardware | — |
| mlx version | — |
| python | — |
| random seed | — |
| msa strategy | — |
| diffusion samples | 1 |
| runtime | 623s |
| predicted by | mlx@peptide |
| predicted at | 2026-04-30 |
▸citationbibtex
@peptide{pep10970,
sequence = {HSQGTFTSDYSKYLDSERAAKEFIAWLVKGR},
target = {gcgr},
author = {peptidemodel},
year = {2026},
status = {computed}
}