Survodutide in brief
Survodutide, code BI 456906, is a synthetic 29-residue peptide on the human glucagon backbone, first reported by Boehringer Ingelheim in 2022. It carries a ring amino acid at position 2, a C-terminal amide, and a C18 fatty diacid on Lys24 attached through a glycine and serine linker and a gamma-glutamate. It is a dual agonist at the human glucagon and GLP-1 receptors, characterised by cAMP potency in cells rather than by binding affinity, which has not been reported. Nearly all of its molecular pharmacology comes from the originating company, and no structure of the peptide exists, alone or bound to either receptor.[1,2,3,4,5]
Survodutide regulatory status
Survodutide is investigational: no marketing authorisation was found in Drugs@FDA (queried through the openFDA Drugs@FDA dataset), the EMA medicines database or the Health Canada Drug Product Database (searched November 2026). The highest development phase on ClinicalTrials.gov is phase 3, with a completed phase 3 record in adults with overweight or obesity without type 2 diabetes and a recruiting phase 3 record in metabolic dysfunction-associated steatohepatitis with liver fibrosis. It is not named on the 2026 WADA Prohibited List, whose S0 class covers any pharmacological substance under clinical development with no current approval by a governmental health authority for human therapeutic use. It does not appear on the FDA's list of bulk drug substances that may present significant safety risks, in either the category 2 table or the table of nominations withdrawn by their nominators (page content current as of 22 April 2026), and no nomination for it was found there.[6,7,8,9,10,11,12]
Survodutide at a glance
- Investigational, and not approved anywhere. No marketing authorisation was found in Drugs@FDA, the European Medicines Agency medicines database or the Health Canada Drug Product Database, the registered programme reaches phase 3, and a 2026 review records the molecule as licensed by Zealand Pharma to Boehringer Ingelheim, which is solely responsible for its global development.[13,6,7,8,9,10]
- Registry identity is unambiguous: four independent registries hold the same 29-residue chain and the same six-residue linker, and two of them pin the acylation to the lysine at position 24. CAS 2805997-46-8, PubChem CID 171378821, FDA unique ingredient identifier 2ALA66NS64, KEGG D12898, InChIKey MEDXQFAHWBMVIM-PCLHIQTFSA-N. The whole conjugate is C192H289N47O61, about 4232 g/mol. The development code BI 456906 names the same molecule throughout the earlier literature.[2,3,14,15]
- Functional potency was measured as cAMP accumulation in CHO-K1 cells expressing one human receptor at a time: 0.33 nM at the GLP-1 receptor and 0.52 nM at the glucagon receptor, against 60 pM for GLP-1 and 20 pM for glucagon in the same assay (discovery report). Both figures are assay conditions, and both move with the assay matrix.[1]
- The glucagon arm of this chemotype was localised to hepatic glucagon receptor signalling in liver-specific knockout mice, and that experiment used BI 456908, a sibling analogue, rather than survodutide itself.[16]
- No molecular picture exists. There is no deposited experimental structure of this peptide, alone or in a receptor complex, and no computed conformer in PubChem, so every statement about how it sits in either receptor pocket is inference from the native hormones' structures in the same two receptors.[5,2]
What is Survodutide?
Survodutide is a laboratory-designed peptide first reported in 2022 by a group at Boehringer Ingelheim under the code BI 456906. It belongs to a class sometimes called unimolecular dual agonists: single synthetic peptides engineered so that one chain activates two related receptors that normally each have their own hormone. The two receptors here are the human glucagon receptor (GCGR, UniProt P47871) and the glucagon-like peptide 1 receptor (GLP-1R, P43220), both class B1 G protein-coupled receptors that couple to the stimulatory G protein Gs.[1,17,18,19,20,21]
Its development status is stated plainly in the literature and is worth having in one place. The peptide is investigational and is not approved for clinical use anywhere; the registered programme reaches phase 3, with a completed record in adults with overweight or obesity without type 2 diabetes and a recruiting record in metabolic dysfunction-associated steatohepatitis with liver fibrosis; and the chemistry did not originate with the company running those trials. A 2026 review records that Zealand Pharma licensed the molecule to Boehringer Ingelheim, which is solely responsible for its global development, which is why the preclinical authorship and the trial sponsorship carry the same name while the design lineage does not.[13,9,10]
The published record runs under two names, which is the first practical obstacle to reading it. The earliest human studies appeared in 2023 under the development code, the name survodutide replaces that code from the 2024 reports onward, and reports of larger trials followed in 2026. Three of the trial reports cited here carry published corrections rather than retractions, and a fourth carries published correspondence; none of them is retracted. Those trials are listed in the references bibliographically, as the record of where the molecule has been studied and under which name. This page does not reproduce their designs, their regimens or their results.[22,23,24,25,26,27,28]
The design route matters for reading the sequence. The chain is the human glucagon sequence with nine positions changed, which puts survodutide on the glucagon side of this class: mazdutide starts from oxyntomodulin and GLP-3 RT from GIP. Starting from glucagon and adding GLP-1 receptor activity, rather than the other way round, is the choice the whole molecule follows from, and the natural precedent for it is oxyntomodulin, a proglucagon product that activates both receptors on its own.[1,29,19]
It is routinely confused with the peptides nearest it in the same class, and they are chemically separate molecules rather than variants of one another. Mazdutide is the other glucagon and GLP-1 dual agonist of this generation, GLP-3 RT adds a third receptor arm, GLP-1 SM activates one receptor, GLP-2 TZ activates two different ones. A validated multiplexed liquid chromatography and high-resolution mass spectrometry method identifies nine of these peptides, survodutide among them, by accurate mass with data-dependent fragment confirmation, reporting 0.0 to 0.5 ppm mass accuracy and 76 to 100 percent fragment coverage across the set. Confirming identity on real material is therefore an analytical question rather than an academic one, and this molecule is awkward to measure for a reason built into its design: the C18 diacid gives the peptide a long plasma residence, so residual peptide persists in a sample and interferes with antibody assays for it, and one assay-development report had to deplete that residual peptide with activated charcoal to raise its antibody assay's drug tolerance about a hundredfold. A label is a claim, and only a measurement settles what a preparation contains.[30,31,1]
Laid out side by side, the five peptides readers most often bring together separate on four axes: how long the chain is and which hormone it is built on, which receptors it activates, what occupies position 2, and where the lipid arm attaches. Survodutide is 29 residues on glucagon, glucagon and GLP-1 receptors, a cyclobutane ring residue at position 2, a C18 diacid on Lys24. Mazdutide is 34 residues on oxyntomodulin, the same two receptors, 2-aminoisobutyric acid at position 2, a C20 diacid on Lys20. GLP-1 SM is 31 residues on GLP-1, the GLP-1 receptor alone, 2-aminoisobutyric acid at position 2, an acylated lysine at chain position 20. GLP-2 TZ is 39 residues on the GIP backbone, the GIP and GLP-1 receptors, 2-aminoisobutyric acid at positions 2 and 13, a C20 diacid on Lys20. GLP-3 RT is 39 residues on the same backbone and adds a glucagon receptor arm to those two. Survodutide is the only one of the five built on glucagon itself and the only one carrying a cyclobutane ring at position 2, which is why its position-2 residue is named so often on this page.[32,33,15,3]
One caution about that comparison, because it is the kind of thing a reader would otherwise inherit. The FDA registry records the position-2 residue, the acylated lysine and the terminal amide for mazdutide, GLP-1 SM and GLP-2 TZ as explicit structural modifications, and it records all three for survodutide. For GLP-3 RT the same registry holds the 39-residue chain and no structural-modification entries at all, so that molecule's lipid arm and its non-coded residues are not pinned by that record and have to be read from its own literature. The receptor assignments above come from KEGG's target fields, which are curated statements about each molecule rather than measurements made in one laboratory across all five.[33,32]
Not to be confused with
- Glucagon
- The endogenous 29-residue hormone, residues 53 to 81 of human pro-glucagon. Survodutide is not glucagon. Aligned position by position against it, the peptide replaces Ser2 with 1-aminocyclobutane-1-carboxylic acid, substitutes eight further residues (Ser16 to Glu, Arg18 to Ala, Gln20 to Lys, Val23 to Ile, Gln24 to Lys, Met27 to Glu, Asn28 to Ser, Thr29 to Ala), ends in an amide rather than a free acid, and carries a linker and a C18 diacid on the lysine at position 24 that the hormone does not have.[29,2,3,1]
- Oxyntomodulin
- The endogenous 37-residue proglucagon product, residues 53 to 89, which activates both the glucagon and the GLP-1 receptor and is the natural precedent this design class points back to. It is a longer chain with glucagon's own sequence at its N-terminal end and a basic tail after position 29, and it carries no non-coded residue, no amide terminus and no lipid arm.[29,1]
- Mazdutide
- The other published glucagon and GLP-1 receptor dual agonist of this generation, and a separate molecule. It is 34 residues built on the oxyntomodulin sequence with 2-aminoisobutyric acid at position 2 and its diacid on Lys20; survodutide is 29 residues on the glucagon sequence with 1-aminocyclobutane-1-carboxylic acid at position 2 and its diacid on Lys24. A validated chromatographic method identifies the two as separate species by accurate mass with fragment confirmation.[30,2,3]
- GLP-3 RT
- A 39-residue peptide developed from the GIP backbone that adds a deliberate GIP receptor arm to the glucagon and GLP-1 pair. Survodutide has no GIP receptor arm in its published characterisation. The two appear beside each other in the same nine-agonist analytical panel as distinct molecules.[30,1]
- GLP-1 SM
- An agonist at the GLP-1 receptor only, built on the GLP-1 backbone. It is the comparator in most of the preclinical work on survodutide precisely because it lacks the glucagon receptor arm, which is how that arm's contribution is isolated. Chemically it is a different peptide, not a variant.[1,16]
- BI 456908 and BI 456897
- Two sibling compounds from the same 19-member series. BI 456908 is described by its authors as a structural and pharmacological back-up analogue of survodutide that was not taken into clinical development, and it is the molecule actually used in the liver-specific knockout experiment that localised the glucagon arm. Results obtained with it are evidence about the chemotype rather than about survodutide.[4,16]
- GLP-2 TZ
- A dual GIP and GLP-1 receptor agonist. It shares the lipidated multi-receptor design idea and has neither a glucagon receptor arm nor any sequence relationship to the glucagon backbone survodutide is built on.[30]
Survodutide structure and sequence
Read along the chain, the sequence is His-Ac4c-Gln-Gly-Thr-Phe-Thr-Ser-Asp-Tyr-Ser-Lys-Tyr-Leu-Asp-Glu-Arg-Ala-Ala-Lys-Asp-Phe-Ile-Lys-Trp-Leu-Glu-Ser-Ala, ending in an amide rather than a free acid. Ac4c is 1-aminocyclobutane-1-carboxylic acid: an alanine whose alpha carbon has been closed into a four-membered ring, which removes the chiral centre and constrains the local backbone. The second of the molecule's late lysines, Lys24, carries the lipid arm: its side-chain nitrogen is acylated by a hexapeptide linker, Gly-Gly-Ser-Gly-Ser-Gly reading outward, then a gamma-linked L-glutamate, then octadecanedioic acid, an eighteen-carbon chain with a carboxylic acid at each end. The distal acid stays free, which is what gives the arm its albumin affinity.[2,3,1]
Set against glucagon the design reads clearly. Human glucagon is HSQGTFTSDYSKYLDSRRAQDFVQWLMNT; survodutide changes nine positions in it. Position 2 becomes the ring residue, which sits exactly on the dipeptidyl peptidase-4 cleavage site this family is cut at. Positions 16, 18, 20, 23, 24, 27, 28 and 29 are substituted, and two of those changes are the ones the rest of the molecule hangs off: Gln20 becomes a lysine and Gln24 becomes the lysine that is acylated. The amide terminus replaces the free acid. Nothing else in the chain moves.[29,2,3,1]
The arithmetic is worth doing once, because it confirms that the registry figures describe the same object the sequence does. Summing the 29 residues with the position-2 ring residue and the C-terminal amide, then adding the six linker residues, the glutamate and the diacid less the water lost at each new amide bond, gives C192H289N47O61, which is exactly what PubChem carries for CID 171378821 at about 4232 g/mol average and 4229.0957 Da monoisotopic. There is no 3D view on this page because nothing has been deposited and PubChem holds no computed conformer, so the molecule is described here by its graph alone.[2,5]
Modifications
- 2 1-aminocyclobutane-1-carboxylic acid (Ac4c) in place of the serine glucagon carries here. The alpha carbon is part of a cyclobutane ring, which makes the residue achiral and alpha,alpha-dialkyl. Position 2 is the dipeptidyl peptidase-4 cleavage site on this family of peptides, and the discovery report names proteolytic stability as the reason for the substitution.[1,2,3]
- 24 The lysine side-chain nitrogen carries a hexapeptide linker read Gly-Gly-Ser-Gly-Ser-Gly outward from the lysine, then a gamma-linked L-glutamate, then octadecanedioic acid, a C18 chain with a carboxylic acid at each end. The distal carboxylate stays free. The FDA registry records the acyl fragment as N-(17-carboxy-1-oxoheptadecyl)-L-gamma-glutamylglycine and the junction as an amide bond between chain residue 24 and the sixth residue of the linker. The discovery report describes the diacid as an albumin-binding, half-life-extending group.[2,3,1,14]
- 29 C-terminal alanine amide rather than a free carboxylate. Named in the discovery report as a second proteolytic-stability measure and confirmed independently by the PubChem IUPAC name, which ends the chain in a 1-amino-1-oxopropan-2-yl group.[1,2]
Every proteinogenic residue in the chain is the L enantiomer apart from the single glycine at position 4 and the linker's four glycines, which are achiral, and the position-2 Ac4c residue, which is achiral because its alpha carbon bears two ring bonds. The linker glutamate is the L enantiomer and is attached through its gamma carboxyl. There is no cysteine, therefore no disulfide; the peptide is linear and uncyclised; the FDA registry records zero N-, O- and C-glycosylation sites, and there is no metal coordination. The only branch point in the molecule is the Lys24 side-chain nitrogen bearing the linker.[2,3]
What studies on Survodutide show
Each card names the experimental system and what was measured. A result in one system is not evidence about another.
Discovery report: cell-based cAMP accumulation assays, islet perifusion, and receptor knockout mice · CHO-K1 cells stably expressing one human receptor at a time (GCGR or GLP-1R); MIN6 insulinoma cells for the endogenous mouse GLP-1 receptor; primary mouse, rat, cynomolgus monkey and human hepatocytes for the endogenous glucagon receptor; mouse, rat and perifused human pancreatic islets; wild-type and GLP-1R knockout mice
Potency at the two human receptors was 0.33 nM (GLP-1R) and 0.52 nM (GCGR), roughly tenfold weaker than GLP-1 at 60 pM and glucagon at 20 pM measured alongside, and 0.36 nM at the endogenous mouse GLP-1 receptor in MIN6 cells. Potency is strongly matrix dependent, measured in a separate cAMP-response-element luciferase assay in HEK293 cells rather than in the CHO-K1 accumulation assay: in undiluted human plasma the figures were 1.0 nM at the GLP-1 receptor and 8.3 nM at the glucagon receptor, in undiluted mouse plasma 1.6 nM and 16 nM, against 0.29 nM and 0.17 nM at the glucagon receptor in 0.5 percent human and mouse plasma, which is the albumin binding of the diacid arm showing up as an apparent loss of potency. The food-intake and glucose-tolerance readouts were absent in GLP-1R knockout mice, which assigns those readouts to the GLP-1 receptor arm.[1]
- Material
- BI 456906, the molecule later named survodutide, as described above
- Measured
- Functional potency (EC50) for cAMP accumulation at each receptor and in several plasma matrices; glucose-stimulated insulin secretion in islets; food intake, glucose tolerance and gastric emptying readouts in wild-type and GLP-1R knockout mice; hepatic Nnmt mRNA and plasma FGF-21 and amino acids as glucagon receptor engagement readouts
- Compared with
- Native GLP-1 and native glucagon in the same assays; GLP-1 SM and GLP-1 LR as GLP-1 receptor selective comparators
- Limitation
- Every functional reading is cAMP accumulation, so the study establishes nothing about arrestin recruitment, receptor trafficking or pathway bias, and nothing about binding affinity as distinct from pathway activation. The potency figures are assay conditions and are not comparable across matrices or across assay formats. All of the work comes from the originating company.
Zimmermann T et al., 2022, Molecular Metabolism
Candidate selection study correlating in vitro potency with in vivo receptor engagement biomarkers · CHO-K1 cells expressing human GCGR and GLP-1R; MIN6 cells; primary rat hepatocytes; lean, diet-induced obese and db/db mice
In vitro potency at each receptor correlated with its matching in vivo engagement biomarker across the 19-compound series (Spearman, p below 0.05), which is what made the biomarker pair usable as a selection tool. Survodutide was chosen on the balance between its two receptor arms rather than on maximum potency at either, and the paper states that selection reasoning explicitly.[4]
- Material
- Survodutide alongside 18 other dual glucagon and GLP-1 receptor agonists from the same series, including BI 456908 and BI 456897
- Measured
- cAMP potency at both human receptors for all 19 compounds, against two acute in vivo engagement biomarkers, hepatic nicotinamide N-methyltransferase mRNA and plasma FGF21
- Compared with
- GLP-1 SM as a GLP-1 receptor selective reference; the other 18 series members against each other
- Limitation
- Read as an abstract only, so the per-compound potency table was not seen. The correlation is within one chemical series from one laboratory, so it is not established that the same biomarkers track receptor engagement for other chemotypes. Biomarker correlation is not receptor occupancy.
Thomas L et al., 2024, Diabetes, Obesity and Metabolism
Tissue-specific receptor knockout study · Diet-induced obese wild-type mice and liver-specific glucagon receptor knockout mice, around ten animals per arm, a 28 to 29 day study period, with an intraperitoneal glucose tolerance test at day 21
The increment this chemotype produced over the GLP-1-selective comparator fell back to that comparator's level in liver-specific glucagon receptor knockouts, and the rise in energy expenditure, the fall in circulating amino acids and the triacylglycerol reduction were all absent there too. The glucose-tolerance change was retained in knockouts. Together that assigns the glucagon arm of this chemotype to hepatic glucagon receptor signalling and the glycaemic readouts to the GLP-1 arm.[16]
- Material
- BI-456908, described by its authors as a structural and pharmacological back-up analogue of survodutide that was not advanced to clinical development, explicitly not survodutide itself
- Measured
- Body weight and composition, energy expenditure, intraperitoneal glucose tolerance, fasting glucose and insulin, plasma amino acids, cholesterol, triacylglycerol, glycerol and non-esterified fatty acids, liver triacylglycerol, plasma ALT
- Compared with
- GLP-1 SM, as a GLP-1 receptor selective agonist, in both genotypes; vehicle in both genotypes
- Limitation
- The molecule tested is an analogue, so this is evidence about the chemotype and not about survodutide. The knockout is liver specific, so extrahepatic glucagon receptors are untested rather than excluded. Group sizes are around ten animals, and these are measured endpoints in one mouse model.
Long F et al., 2026, Diabetes, Obesity and Metabolism
Receptor expression mapping, labelled-peptide distribution imaging and c-Fos mapping · Human and mouse circumventricular organs, single-cell expression data; mouse brain imaging; mouse c-Fos labelling
GCGR transcript was barely detectable in mouse and human area postrema and arcuate nucleus at single-cell level, while GLP1R was present in both. The labelled peptide was detected in circumventricular organs and adjacent hypothalamic and hindbrain nuclei without evidence of uniformly crossing the blood-brain barrier, and the glucagon-receptor-selective comparator produced no neuronal activation in the regions examined.[34]
- Material
- A fluorophore-labelled survodutide conjugate for the imaging work
- Measured
- GCGR and GLP1R transcript presence in area postrema and arcuate nucleus at single-cell resolution; where the labelled peptide was detected in the mouse brain; which nuclei showed c-Fos labelling
- Compared with
- A separate long-acting glucagon receptor selective agonist
- Limitation
- Read as an abstract only. A fluorophore-labelled conjugate is a different molecule from the unlabelled peptide and its distribution is not receptor occupancy. c-Fos labelling marks neuronal activity, not a receptor interaction, and transcript presence is not protein.
Zimmermann T et al., 2026, Molecular Metabolism
Engineered rodent disease model with histology and quantitative 3D whole-organ imaging · Female db/db C57BKS mice, uninephrectomised, with adeno-associated-virus-mediated renin overexpression; repeated exposure over the study period
The measured renal endpoints moved in that model, and plasma amino acids including alanine fell, which the authors read as hepatic glucagon receptor activation. The amino acid readout is the molecular link back to the discovery work, which used the same measurement as its glucagon receptor engagement marker.[35]
- Material
- Survodutide
- Measured
- Albuminuria, histological scores for glomerulosclerosis, tubular injury and inflammation, kidney and glomerular volume, glomerular density and sphericity by 3D imaging, and plasma amino acids including alanine
- Compared with
- Vehicle
- Limitation
- Read as an abstract only. One engineered model, female animals only, vehicle as the sole comparator, so nothing is established about this molecule relative to a GLP-1-selective agonist here. The study characterises endpoints in that model and not a receptor-level mechanism.
Hamilton A et al., 2026, Diabetes, Obesity and Metabolism
Analytical method development and validation · Reversed-phase liquid chromatography on a C18 column with high-resolution full-scan mass spectrometry and data-dependent HRMS/MS fragmentation
Top-down identification achieved 0.0 to 0.5 ppm mass accuracy with 76 to 100 percent fragment coverage and baseline chromatographic separation for most of the set inside a 20 minute run. This establishes that survodutide is distinguishable from the eight related agonists by accurate mass with fragment confirmation rather than by retention time alone.[30]
- Material
- Survodutide as one of nine peptide reference standards, beside bofanglutide, ecnoglutide, exenatide, GLP-1 LR, mazdutide, GLP-3 RT, GLP-1 SM and GLP-2 TZ
- Measured
- Chromatographic separation, accurate mass, fragment ion coverage, linearity, limits of detection and quantitation across the nine peptides
- Compared with
- The eight other agonists in the same run
- Limitation
- Read as an abstract only, so per-peptide retention times and fragment coverage were not seen. The method was applied to marketed GLP-1 LR, GLP-1 SM and GLP-2 TZ formulations, not to survodutide material, so its performance on real survodutide preparations is untested here.
Tong LH et al., 2026, Journal of Chromatography A
Immunoassay method development · Neutralising anti-drug-antibody assays in clinical matrix, with activated-charcoal sample pretreatment
Charcoal depletion of residual peptide raised drug tolerance about a hundredfold. The relevance to chemistry is direct: the albumin-binding C18 arm that gives the molecule its long residence is the same feature that keeps it in the sample and masks the signal, so a design choice at Lys24 shows up as an analytical problem downstream.[31]
- Material
- Survodutide, used as the model half-life-extended dual agonist
- Measured
- Assay drug tolerance, that is the residual peptide concentration an antibody assay can tolerate before it stops detecting antibodies
- Compared with
- The same assays without charcoal pretreatment
- Limitation
- Read as an abstract only. An assay-development paper with no receptor pharmacology, and the hundredfold figure describes these assays rather than a general property.
Sanville M et al., 2026, The AAPS Journal
Survodutide compared with Mazdutide and GLP-3 RT
Four things about the 3 compounds most often read beside it, on axes that can be checked: what kind of substance it is, its formula and mass, what public records settle about which substance it is, and the strongest three-dimensional evidence that exists for it. Every figure is the one stated on that compound's own page, which is where it is cited, so the two can never disagree. The table compares no effect, no potency and nothing a compound has been reported to do, and these compounds are not alternatives to one another.
| Compound | Kind | Formula and mass | Identity on the public record | Strongest structure evidence |
|---|---|---|---|---|
| Survodutide | Peptide | 29 residues, C192H289N47O61, 4232 g/mol | One species, pinned by public records | No three-dimensional model |
| Mazdutide | Peptide | 34 residues, C210H322N46O67, 4563 g/mol | One species, pinned by public records | No three-dimensional model |
| GLP-3 RT | Peptide | 39 residues, C221H342N46O68, 4731 g/mol | One species, pinned by public records | Experimental structure, resolved inside a complex: 8YW3, Single-particle cryo-electron microscopy at 2.68 Å |
| GLP-1 SM | Peptide | 31 residues, C187H291N45O59, 4114 g/mol | One species, pinned by public records | Experimental structure, resolved inside a complex: 7KI0, Single-particle cryo-electron microscopy at 2.5 Å |
How survodutide is characterised at the glucagon and GLP-1 receptors
The published characterisation rests on functional activation rather than on binding. In cells expressing one human receptor at a time, the peptide raised cAMP with an EC50, the concentration producing half of that assay's maximal response, of 0.33 nM at the GLP-1 receptor and 0.52 nM at the glucagon receptor, about tenfold weaker than the native hormones measured in the same assay. Those two figures being close to each other is the point of the molecule: a 19-compound series was profiled and this one was selected on the balance between the two arms, not on potency at either, with hepatic nicotinamide N-methyltransferase mRNA and plasma FGF21 used as the in vivo readouts that the in vitro potencies were checked against. A post hoc biomarker analysis of two trial populations reports circulating glucagon falling in both, which is the counter-regulatory consequence of agonism at the glucagon receptor rather than an outcome.[1,4,36]
Which arm does what has been separated experimentally in two directions. Deleting the GLP-1 receptor abolished the food-intake and glucose-tolerance readouts, so those belong to that arm. Deleting the glucagon receptor in liver alone removed the energy-expenditure rise, the fall in circulating amino acids and the lipid changes while leaving glucose tolerance intact, so the glucagon arm of this chemotype acts through hepatic receptors. That second experiment used BI 456908, a sibling analogue, which is why it is evidence about the chemotype rather than about this exact peptide. A brain-mapping study adds a consistent asymmetry at the level of receptor distribution: GCGR transcript was barely detectable in the mouse and human circumventricular regions examined, while GLP1R was present there.[1,16,34]
The albumin-binding arm is part of the mechanism rather than an accessory to it. Albumin binding here means the reversible association of the free carboxylate at the end of the C18 arm with serum albumin, the most abundant plasma protein, which keeps the peptide in circulation and, while it lasts, keeps it away from a receptor. In a cAMP-response-element luciferase assay, a different format from the accumulation assay above, glucagon receptor potency moved from 0.29 nM in 0.5 percent human plasma to 8.3 nM in undiluted human plasma, and from 0.17 nM to 16 nM in the mouse equivalent; the undiluted GLP-1 receptor figures were 1.0 and 1.6 nM, and the discovery report reads the surviving ratio between the two arms in human plasma as about 1 to 8. A peptide bound to albumin is not a peptide at a receptor. A quoted potency figure is therefore meaningless without both its matrix and its assay format, which is why this molecule's behaviour in plasma, in an immunoassay and at an isolated receptor are three different numbers.[1,31]

What is still unknown about Survodutide
There is no structural picture at all. No deposited experimental structure contains this peptide, alone or bound to either receptor, and PubChem holds no computed conformer for its CID. For the nearest molecules in the class this is not the case: both GLP-3 RT and GLP-2 TZ have cryo-EM complexes with their receptors, so their contacts can be named residue by residue. For survodutide every statement about how the chain sits in either pocket is inference from the native hormones' structures in the same receptors. Alongside that, no binding affinity at either receptor was found in any source read here, so affinity and the geometry that would explain it are both missing.[5,2,1]
The identity records agree with each other, with one caution that is worth stating because it is the kind of thing a reader would otherwise inherit. PubChem, the FDA registry, ChEMBL and KEGG all hold the same 29-residue chain and the same six-residue linker, some of them written in opposite directions, and the FDA registry and KEGG independently pin the acylation junction to chain residue 24, with the FDA registry also naming the acyl fragment. But the ChEMBL record carries a third biocomponent labelled as a light chain whose sequence is an immunoglobulin kappa light chain, which has no relationship to this peptide and appears to be a record error. The FDA registry also lists no C-terminal amidation among its structural modifications even though PubChem's own IUPAC name and the discovery report both state it. Neither gap changes the molecule; both are reasons to read a registry record field by field rather than as a whole.[14,3,2,15,1]
Nearly all of the molecular pharmacology comes from one source. The discovery report, the candidate-selection study, the brain-mapping study and a study combining the peptide with a neuropeptide Y2 receptor agonist were all produced by the originating company or with its support, and the knockout study that localised the glucagon arm was funded by it and used its back-up compound. None of that makes the measurements wrong, and the figures are stated clearly enough to be checked. It does mean that the central potency numbers are unreplicated by an independent laboratory, and the one fully independent measurement located here is an analytical identification method from a government laboratory, which establishes that the molecule can be told apart from its neighbours and nothing about its pharmacology.[1,4,34,37,16,30]
Whether the glucagon arm contributes anything that does not simply follow from the GLP-1 arm is still being argued, and at present it is argued statistically rather than experimentally. A mediation analysis of the phase 2 liver dataset set out to separate the part of the liver endpoint change that tracks body-weight change from the part that does not, which is a modelling argument over one trial rather than an experiment, and the most recent compound-specific review frames the remaining uncertainties as uncertainties. The clean version of that question is the knockout experiment, and that was run on the analogue.[38,13,16]
Key papers on Survodutide
- BI 456906: Discovery and preclinical pharmacology of a novel GCGR/GLP-1R dual agonist with robust anti-obesity efficacy.
Zimmermann T, Thomas L, Baader-Pagler T, Haebel P, Simon E, Reindl W, Bajrami B, Rist W, Uphues I, Drucker DJ, Klein H, Santhanam R, Hamprecht D, Neubauer H, Augustin R (2022), Molecular Metabolism · Discovery report with in vitro receptor pharmacology and receptor knockout mice
The discovery report, and the only place every design feature of the molecule is named with the reason it is there, alongside the potency figures and the assay matrices they were measured in.
- PubChem Compound Summary for CID 171378821, Survodutide.
PubChem, National Center for Biotechnology Information (2026), PubChem · Database record
The PubChem record: formula, both masses, CAS number, and a deposited synonym that spells the entire construct out including the linker and the diacid.
- SURVODUTIDE: substance record, unique ingredient identifier 2ALA66NS64.
US Food and Drug Administration, Global Substance Registration System (2026), FDA Global Substance Registration System · Database record
The FDA registry record, which is the only source that pins the acylation to chain residue 24 by an explicit linkage and names the acyl fragment as its own registered substance.
- Hepatic GCGR is required for the superior weight loss and metabolic effects of a structurally related analogue of the dual GCGR/GLP-1R agonist survodutide in mice.
Long F, Challa TD, Efthymiou V, Klug M, Klein T, Neubauer H, Wolfrum C, Horvath C (2026), Diabetes, Obesity and Metabolism · Tissue-specific receptor knockout study
The liver-specific knockout experiment that separates the glucagon arm from the GLP-1 arm, and the clearest worked example of why an analogue result is not a result about the molecule.
- The dual GCGR/GLP-1R agonist survodutide: Biomarkers and pharmacological profiling for clinical candidate selection.
Thomas L, Martel E, Rist W, Uphues I, Hamprecht D, Neubauer H, Augustin R (2024), Diabetes, Obesity and Metabolism · Candidate selection study
Shows how the molecule was picked out of a 19-compound series on the balance between its two receptor arms, which is the design decision the whole chemotype turns on.
- Development and validation of a multiplexed LC-HRMS method for nine GLP-1 receptor agonists and its pharmaceutical application.
Tong LH, Leung KK, Hung CT (2026), Journal of Chromatography A · Analytical method development and validation
The independent analytical method, and the answer to the practical question of how survodutide is told apart from the eight peptides it is most often confused with.
- IUPHAR review: From foe to friend: Repurposing glucagon to treat obesity and type 2 diabetes.
Elmendorf AJ, Yousefian M, Kim IM, Hardaway JA, Habegger K, Flak JN (2026), Pharmacological Research · Review
Read last, for class context: why a glucagon receptor arm is deliberately added to a GLP-1 receptor agonist, written for a pharmacology readership.
References
- 1.Zimmermann T, Thomas L, Baader-Pagler T, Haebel P, Simon E, Reindl W, Bajrami B, Rist W, Uphues I, Drucker DJ, Klein H, Santhanam R, Hamprecht D, Neubauer H, Augustin R. (2022). BI 456906: Discovery and preclinical pharmacology of a novel GCGR/GLP-1R dual agonist with robust anti-obesity efficacy. Molecular Metabolism. 66:101633. doi:10.1016/j.molmet.2022.101633 · full text read
- 2.PubChem, National Center for Biotechnology Information. (2026). PubChem Compound Summary for CID 171378821, Survodutide. PubChem. source · database record
- 3.US Food and Drug Administration, Global Substance Registration System. (2026). SURVODUTIDE: substance record, unique ingredient identifier 2ALA66NS64. FDA Global Substance Registration System. source · database record
- 4.Thomas L, Martel E, Rist W, Uphues I, Hamprecht D, Neubauer H, Augustin R. (2024). The dual GCGR/GLP-1R agonist survodutide: Biomarkers and pharmacological profiling for clinical candidate selection. Diabetes, Obesity and Metabolism. 26:2368-2378. doi:10.1111/dom.15551 · abstract only
- 5.RCSB Protein Data Bank. (2026). RCSB PDB full-text and sequence searches for survodutide and BI 456906, run 2 November 2026. RCSB PDB. source · database record
- 6.U.S. Food and Drug Administration. (2026). openFDA Drugs@FDA dataset query for active ingredient survodutide: no matching application. openFDA. source · database record
- 7.European Medicines Agency. (2026). Medicines output: human and veterinary medicines (JSON report), searched for survodutide, 2 November 2026. European Medicines Agency. source · database record
- 8.Health Canada. (2026). Drug Product Database active-ingredient query for survodutide: no matching product. Health Canada Drug Product Database. source · database record
- 9.U.S. National Library of Medicine. (2026). ClinicalTrials.gov record NCT06066515: phase 3, randomised, double-blind, placebo-controlled study of BI 456906 in participants with overweight or obesity without type 2 diabetes (completed). ClinicalTrials.gov. source · database record
- 10.U.S. National Library of Medicine. (2026). ClinicalTrials.gov record NCT06632444: phase 3, randomised, double-blind, placebo-controlled trial of survodutide in adults with non-cirrhotic MASH and F2 to F3 liver fibrosis (recruiting). ClinicalTrials.gov. source · database record
- 11.World Anti-Doping Agency. (2025). World Anti-Doping Code International Standard: Prohibited List 2026. World Anti-Doping Agency. source · full text read
- 12.U.S. Food and Drug Administration. (2026). Certain Bulk Drug Substances for Use in Compounding that May Present Significant Safety Risks (category 2 of the substances nominated under sections 503A or 503B, and the table of nominations withdrawn by their nominators), content current as of 04/22/2026. U.S. Food and Drug Administration. source · full text read
- 13.Sergi CM. (2026). Survodutide for the treatment of obesity: Mechanistic rationale, clinical evidence, and remaining uncertainties. Contemporary Clinical Trials. 170:108493. doi:10.1016/j.cct.2026.108493 · abstract only
- 14.ChEMBL, European Bioinformatics Institute. (2026). ChEMBL molecule record CHEMBL5314776: SURVODUTIDE. ChEMBL. source · database record
- 15.Kanehisa Laboratories. (2026). KEGG DRUG entry D12898: Survodutide (USAN/INN). KEGG. source · database record
- 16.Long F, Challa TD, Efthymiou V, Klug M, Klein T, Neubauer H, Wolfrum C, Horvath C. (2026). Hepatic GCGR is required for the superior weight loss and metabolic effects of a structurally related analogue of the dual GCGR/GLP-1R agonist survodutide in mice. Diabetes, Obesity and Metabolism. 28:2437-2442. doi:10.1111/dom.70359 · full text read
- 17.UniProt Consortium. (2026). UniProtKB P47871 (GLR_HUMAN): Glucagon receptor, Homo sapiens. UniProtKB. source · database record
- 18.UniProt Consortium. (2026). UniProtKB P43220 (GLP1R_HUMAN): Glucagon-like peptide 1 receptor, Homo sapiens. UniProtKB. source · database record
- 19.Elmendorf AJ, Yousefian M, Kim IM, Hardaway JA, Habegger K, Flak JN. (2026). IUPHAR review: From foe to friend: Repurposing glucagon to treat obesity and type 2 diabetes. Pharmacological Research. 223:108077. doi:10.1016/j.phrs.2025.108077 · abstract only
- 20.Neff GW. (2025). Shared mechanistic pathways of glucagon signalling: Unlocking its potential for treating obesity, metabolic dysfunction-associated steatotic liver disease, and other cardio-kidney-metabolic conditions. Diabetes, Obesity and Metabolism. 27:6869-6883. doi:10.1111/dom.70148 · abstract only
- 21.Vinton TY. (2025). Survodutide: a novel peptide for treatment of obesity and metabolic diseases. Proceedings (Baylor University Medical Center). 38:523. doi:10.1080/08998280.2025.2498840 · abstract only
- 22.Jungnik A, Arrubla Martinez J, Plum-Morschel L, Kapitza C, Lamers D, Thamer C, Scholch C, Desch M, Hennige AM. (2023). Phase I studies of the safety, tolerability, pharmacokinetics and pharmacodynamics of the dual glucagon receptor/glucagon-like peptide-1 receptor agonist BI 456906. Diabetes, Obesity and Metabolism. 25:1011-1023. doi:10.1111/dom.14948 · abstract only
- 23.Yazawa R, Ishida M, Balavarca Y, Hennige AM. (2023). A randomized Phase I study of the safety, tolerability, pharmacokinetics and pharmacodynamics of BI 456906, a dual glucagon receptor/glucagon-like peptide-1 receptor agonist, in healthy Japanese men with overweight/obesity. Diabetes, Obesity and Metabolism. 25:1973-1984. doi:10.1111/dom.15064 · abstract only
- 24.Bluher M, Rosenstock J, Hoefler J, Manuel R, Hennige AM. (2024). Dose-response effects on HbA1c and bodyweight reduction of survodutide, a dual glucagon/GLP-1 receptor agonist, compared with placebo and open-label [GLP-1 SM] in people with type 2 diabetes: a randomised clinical trial. Diabetologia. 67:470-482. doi:10.1007/s00125-023-06053-9 · abstract only · corrected: Correction published as Diabetologia 2024;67(4):758, doi 10.1007/s00125-024-06095-7, PMID 38349400. Corrected, not retracted.
- 25.le Roux CW, Steen O, Lucas KJ, Startseva E, Unseld A, Hennige AM. (2024). Glucagon and GLP-1 receptor dual agonist survodutide for obesity: a randomised, double-blind, placebo-controlled, dose-finding phase 2 trial. The Lancet Diabetes & Endocrinology. 12:162-173. doi:10.1016/S2213-8587(23)00356-X · abstract only · corrected: The PubMed record lists an erratum in The Lancet Diabetes & Endocrinology 2026;14(2):e2. Corrected, not retracted.
- 26.Sanyal AJ, Bedossa P, Fraessdorf M, Neff GW, Lawitz E, Bugianesi E, Anstee QM, Hussain SA, Newsome PN, Ratziu V, Hosseini-Tabatabaei A, Schattenberg JM, Noureddin M, Alkhouri N, Younes R; 1404-0043 Trial Investigators. (2024). A Phase 2 Randomized Trial of Survodutide in MASH and Fibrosis. The New England Journal of Medicine. 391:311-319. doi:10.1056/NEJMoa2401755 · abstract only
- 27.le Roux CW, Wharton S, Startseva E, Kloer IM, Hussain SA, Unseld A, Bozkurt B, Ard JD, Bays HE, Bogdanski P, Ekinci EI, Jastreboff AM, Ji L, Ogawa W, Pedersen SD, Pietilainen KH, Sattar N, Seufert J, Stenlof K, van Beek AP, Vangoitsenhoven R, Brueckmann M, Younes R, Kaplan LM; SYNCHRONIZE-1 Investigators. (2026). Survodutide Once Weekly for the Treatment of Adults with Obesity. The New England Journal of Medicine. 395:776-787. doi:10.1056/NEJMoa2600751 · abstract only
- 28.Kaplan LM, Startseva E, le Roux CW, Wharton S, Bozkurt B, Mazo DF, von Schlippenbach J, Gonzalez Maldonado S, Ajaz Hussain S, Neff GW, Gonzalez Rojas Y, Smith C, Younes R, Sanyal AJ; SYNCHRONIZE-MASLD Investigators. (2026). Survodutide in adults with obesity and metabolic dysfunction-associated steatotic liver disease: SYNCHRONIZE-MASLD, a randomized, double-blind, placebo-controlled phase 3 trial. Nature Medicine. 32:2948-2958. doi:10.1038/s41591-026-04479-3 · abstract only · corrected: Author Correction published as Nature Medicine 2026;32(10):3913, doi 10.1038/s41591-026-04650-w, PMID 42642663. Corrected, not retracted.
- 29.UniProt Consortium. (2026). UniProtKB P01275 (GLUC_HUMAN): Pro-glucagon, Homo sapiens. UniProtKB. source · database record
- 30.Tong LH, Leung KK, Hung CT. (2026). Development and validation of a multiplexed LC-HRMS method for nine GLP-1 receptor agonists and its pharmaceutical application. Journal of Chromatography A. 1786:467288. doi:10.1016/j.chroma.2026.467288 · abstract only
- 31.Sanville M, Dambra R, Werth E, Oquendo E, Herrmann M, Saini G. (2026). Charcoal-Based Depletion Method to Improve Drug Tolerance in Neutralizing Antibody Assays for Therapeutic Peptides. The AAPS Journal. 28:173. doi:10.1208/s12248-026-01311-1 · abstract only
- 32.Kanehisa Laboratories. (2026). KEGG DRUG records for the four comparator peptides read for the comparison in this entry: D12225 mazdutide, D10025 [GLP-1 SM], D11360 [GLP-2 TZ], D12430 [GLP-3 RT]. Read 3 November 2026. KEGG DRUG. source · database record
- 33.U.S. Food and Drug Administration. (2026). Global Substance Registration System substance records for the four comparator peptides: unique ingredient identifiers MB76Z4IBZ5 (mazdutide), 53AXN4NNHX ([GLP-1 SM]), OYN3CCI6QE ([GLP-2 TZ]), NOP2Y096GV ([GLP-3 RT]). Read 3 November 2026. FDA Global Substance Registration System. source · database record
- 34.Zimmermann T, Bleymehl K, Haebel P, Perens J, Roostalu U, Hecksher-Sorensen J, Doerr J, Jarosch S, Lam D, Klein H, Pekcec A, Chehimi SN, Crist RC, Reiner BC, Hayes MR, Augustin R. (2026). Survodutide acts through circumventricular organs in the brain and activates neuronal regions associated with appetite regulation. Molecular Metabolism. 105:102326. doi:10.1016/j.molmet.2026.102326 · abstract only
- 35.Hamilton A, Thisted L, Ougaard M, Hahn M, Andersen K, Perens J, Augustin R, Bajrami B, Rist W, Christensen M, Hansen HH, Klein T. (2026). Nephroprotective Effects of the Dual GCGR/GLP-1R Agonist Survodutide in Hypertension-Accelerated Diabetic Kidney Disease. Diabetes, Obesity and Metabolism. doi:10.1111/dom.71387 · abstract only
- 36.Ekinci EI, Maldonado SG, Unseld A, Martinez JA, Hennige AM, Schoelch C. (2026). Dual Glucagon and GLP-1 Receptor Agonist Survodutide Improves Biomarkers of Beta-Cell Function and Insulin Sensitivity in People With Type 2 Diabetes or Living With Overweight/Obesity. Diabetes, Obesity and Metabolism. 28:8242-8253. doi:10.1111/dom.70861 · abstract only
- 37.Augustin R, Oldenburger A, Baader-Pagler T, Zimmermann T, Borghardt J, Hecksher-Sorensen J, Baljuls A, Reindl W, Krawczyk B, Martel E, Brennauer A, Peters S, Grube A, Rudkjaer LB, Haebel P. (2025). Novel NPY2R agonist BI 1820237 provides synergistic anti-obesity efficacy when combined with the GCGR/GLP-1R dual agonist survodutide. Molecular Metabolism. 99:102205. doi:10.1016/j.molmet.2025.102205 · abstract only
- 38.Noureddin M, Sanyal AJ, Bedossa P, Fraessdorf M, Schoelch C, Startseva E, Marshall SR, Alhussein A, Neff GW, Lawitz EJ, Bugianesi E, Anstee QM, Newsome PN, Ratziu V, Hosseini-Tabatabaei A, Schattenberg JM, Alkhouri N, Younes R. (2026). Weight reduction-dependent/-independent effects of survodutide on liver endpoints: Mediation analysis of a phase 2 trial in MASH. Hepatology. doi:10.1097/HEP.0000000000001825 · abstract only
- 39.World Health Organization, INN Programme. (2023). International Nonproprietary Names for Pharmaceutical Substances (INN). Recommended International Nonproprietary Names: List 90. Entry survodutidum / survodutide, page 855. WHO Drug Information. Vol. 37, No. 3:855. source ·
Related entries
- MazdutideMazdutide is a synthetic 34-residue peptide built on the human oxyntomodulin sequence, carrying a C20 fatty diacid on the Lys20 side chain. It also carries 2-aminoisobutyric acid at position 2 and a C-terminal glycinamide, and is characterised in the literature as an agonist at both the human GLP-1 receptor and the human glucagon receptor.
- GLP-3 RTGLP-3 RT is an investigational synthetic 39-residue peptide from Eli Lilly, carrying a fatty-diacid side chain on Lys17. Within one molecule it activates three human class B1 G protein-coupled receptors, the GIP, GLP-1 and glucagon receptors.
- GLP-1 SMGLP-1 SM is a long-acting synthetic GLP-1 receptor agonist, a 31-residue analogue of human glucagon-like peptide 1 with a fatty-acid side chain. That side chain binds serum albumin, and two backbone substitutions stiffen the cleavage site and leave the side chain only one place to attach.
- GLP-2 TZGLP-2 TZ is a synthetic 39-residue peptide carrying a twenty-carbon fatty-diacid side chain on Lys20. It activates two human class B1 G protein-coupled receptors, the GIP receptor and the GLP-1 receptor, with measurably different signalling behaviour at each.
