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

Semaglutide is a long-acting GLP-1 receptor agonist: a lipidated peptide built on the human GLP-1 sequence, engineered so that a hormone with a half-life of minutes becomes one that circulates for about a week. This overview covers its structure, how it acts at the receptor, how its development unfolded, what the main clinical programmes reported, and how the molecule is characterised in the laboratory. It summarises published findings only; it is not medical or usage guidance.

What is semaglutide?

Semaglutide is a GLP-1 (glucagon-like peptide-1) analogue with 94% sequence homology to human GLP-1. It selectively binds and activates the GLP-1 receptor, the same receptor targeted by the native hormone [1]. It belongs to the incretin-based class of peptides and is the active ingredient of approved medicines for type 2 diabetes, chronic weight management and several related indications in the US and EU [1] [2] [11].

The molecule was designed for once-weekly exposure. Its developers set out to increase albumin affinity and secure full stability against metabolic degradation, and selected semaglutide from a series of acylated GLP-1 analogues as the optimal once-weekly candidate [3].

Semaglutide structure and chemistry

Semaglutide is a 31-residue peptide corresponding to positions 7–37 of human GLP-1. Relative to GLP-1 it has two amino acid substitutions, α-aminoisobutyric acid (Aib) at position 8 and arginine at position 34, and it is derivatised at lysine 26 [3]. The US label explains the purpose of each change: position 8 is modified to stabilise against dipeptidyl peptidase-4 (DPP-4), position 34 is changed so that only one fatty di-acid attaches, and lysine 26 carries a hydrophilic spacer and a C18 fatty di-acid that drive albumin binding [1].

Property	Semaglutide
INN	Semaglutide
CAS number	910463-68-2 [5] [7]
UNII	53AXN4NNHX [5]
Molecular formula	C187H291N45O59 [1] [6]
Molecular weight	4113.58 g/mol [1] [6]
Length	31 amino acids (GLP-1 positions 7–37) [3] [6]
Backbone sequence (one-letter)	H-Aib-EGTFTSDVSSYLEGQAAK-EFIAWLVRGRG, free C-terminal acid [5] [6]
Substitutions vs GLP-1	Aib8 (DPP-4 resistance), Arg34 (single acylation site) [1] [3]
Side chain	Lys26 acylated with a C18 fatty di-acid (17-carboxyheptadecanoyl) through γGlu-2×OEG [4] [5]
Plasma protein binding	>99% bound to albumin [1]

The linker choice was empirical. In the developers’ screening of C12 to C20 fatty acids, “a C18 di-acid together with a γGlu-2xOEG linker resulted in the highest albumin affinity”, and the OEG (8-amino-3,6-dioxaoctanoic acid) units gave both high receptor potency and albumin affinity [4]. The FDA substance registration system’s systematic name spells out the same chain: N6-[N-(17-carboxy-1-oxoheptadecyl)-L-γ-glutamyl-2×(2-[2-(2-aminoethoxy)ethoxy]acetyl)]-L-lysine at position 26 [5].

Semaglutide mechanism of action

The GLP-1 receptor is a class B G protein-coupled receptor expressed in the pancreas, gastrointestinal tract, heart, lungs, kidneys and brain [4]. Semaglutide stimulates insulin secretion and lowers glucagon secretion, both in a glucose-dependent manner: when blood glucose is high, insulin secretion is stimulated and glucagon secretion is inhibited [1]. Receptors in the pancreas and the brain have been shown to account, respectively, for the glycaemic and body-weight effects seen with liraglutide and semaglutide [4].

Why the half-life is about one week

The principal protraction mechanism is reversible albumin binding, which reduces renal clearance and protects the peptide from metabolic degradation; the Aib8 substitution adds resistance to DPP-4 [1]. The pharmacology follows from those design choices:

Parameter	Value	Evidence level	Ref
GLP-1 receptor affinity	0.38 ± 0.06 nM (three-fold lower than liraglutide)	In vitro	[3]
Plasma half-life, intravenous	46.1 h	Mini-pig	[3]
Mean residence time, subcutaneous	63.6 h	Mini-pig	[3]
Half-life, subcutaneous	145–165 h	Phase 1b, adults with overweight or obesity	[25]
Elimination half-life	Approximately 1 week; present in circulation about 5 weeks after the last dose	Approved US label	[1]
Metabolism	Proteolytic cleavage of the backbone and sequential beta-oxidation of the fatty acid side chain	Approved US label	[1]
Intact drug in urine	Approximately 3% of the dose	Approved US label	[1]

The receptor affinity trade-off is worth noting: semaglutide binds the GLP-1 receptor three times more weakly than liraglutide in vitro, but its higher albumin affinity and full metabolic stability produce much longer exposure [3].

Development history and timeline

Semaglutide was discovered at Novo Nordisk as a successor to liraglutide, the company’s once-daily acylated GLP-1 analogue, which received EU marketing authorisation in 2009 [4]. The 2015 discovery paper described semaglutide as “currently in phase 3 clinical testing” [3].

Date	Milestone	Ref
2015	Discovery and selection published; phase 3 under way	[3]
2016	SUSTAIN-6 cardiovascular outcomes trial published	[13]
5 Dec 2017	First US approval (subcutaneous, type 2 diabetes), NDA 209637	[8]
8 Feb 2018	First EU marketing authorisation (type 2 diabetes)	[11]
20 Sep 2019	US approval of the oral tablet, NDA 213051	[9]
3 Apr 2020	EU authorisation of the oral tablet (type 2 diabetes)	[11]
4 Jun 2021	US approval for chronic weight management, NDA 215256 (priority review)	[10]
6 Jan 2022	EU authorisation for obesity and overweight	[11]
8 Mar 2024, 15 Aug 2025, 21 Nov 2025	New-indication supplements approved on NDA 215256; the label now covers cardiovascular risk reduction and noncirrhotic MASH (accelerated approval)	[2] [10]
24 Nov 2025	EU authorisation of a fixed combination with insulin icodec	[11]
26 Mar 2026	EU authorisation of a semaglutide medicine for MASH	[11]
2026	Phase 3 evoke trials in early Alzheimer’s disease reported as negative	[23]

For how compound patent expiry from 2026 is opening the market to generic semaglutide, see semaglutide patent expiry by country.

Semaglutide clinical trials: key studies and findings

The programme names are SUSTAIN (subcutaneous, type 2 diabetes), PIONEER and SOUL (oral, type 2 diabetes), STEP (weight management), OASIS (oral, weight management), SELECT (cardiovascular outcomes without diabetes), FLOW (kidney outcomes) and ESSENCE (MASH). Results below are as published; “semaglutide” means the trial’s active arm, and dose levels are deliberately omitted.

Study	Phase / design	Population	n	Duration	Primary endpoint	Main finding	Ref
SUSTAIN 1	3a, placebo-controlled	Type 2 diabetes, treatment-naive	388	30 weeks	HbA1c change	HbA1c fell 1.45–1.55 points vs 0.02 with placebo	[12]
SUSTAIN-6	3, CV outcomes, noninferiority	Type 2 diabetes, high CV risk	3,297	104 weeks	CV death, non-fatal MI or stroke	6.6% vs 8.9%; HR 0.74 (0.58–0.95)	[13]
PIONEER 6	3, oral, CV safety	Type 2 diabetes, high CV risk	3,183	Median 15.9 months	MACE	3.8% vs 4.8%; HR 0.79 (0.57–1.11), noninferior	[14]
STEP 1	3, placebo-controlled	Obesity or overweight, no diabetes	1,961	68 weeks	% body-weight change	−14.9% vs −2.4%	[15]
OASIS 1	3, oral, placebo-controlled	Obesity or overweight, no diabetes	667	68 weeks	% body-weight change	−15.1% vs −2.4%	[17]
STEP-HFpEF	3, placebo-controlled	HFpEF with BMI ≥30	529	52 weeks	KCCQ-CSS and weight	KCCQ-CSS +16.6 vs +8.7; weight −13.3% vs −2.6%	[18]
SELECT	3, CV outcomes, superiority	CV disease, BMI ≥27, no diabetes	17,604	Mean 39.8 months follow-up	CV death, non-fatal MI or stroke	6.5% vs 8.0%; HR 0.80 (0.72–0.90)	[19]
FLOW	3, kidney outcomes	Type 2 diabetes with CKD	3,533	Median 3.4 years	Major kidney disease events	HR 0.76 (0.66–0.88)	[20]
SOUL	3, oral, CV outcomes	Type 2 diabetes with ASCVD and/or CKD	9,650	Median 49.5 months	MACE	12.0% vs 13.8%; HR 0.86 (0.77–0.96)	[21]
ESSENCE (part 1)	3, placebo-controlled, interim	Biopsy-defined MASH, fibrosis F2–F3	800 analysed	72 weeks	Histology	Steatohepatitis resolution 62.9% vs 34.3%	[22]
evoke / evoke+	3, oral, placebo-controlled	Early Alzheimer’s disease	3,808	104 weeks (primary)	CDR-SB change	No difference vs placebo	[23]

SUSTAIN: type 2 diabetes and cardiovascular safety

SUSTAIN 1 randomised 388 treatment-naive adults with type 2 diabetes at 72 sites in eight countries. From a mean baseline HbA1c of 8.05%, HbA1c fell by 1.45 and 1.55 percentage points in the two semaglutide arms at week 30, against 0.02 with placebo; the main reason for stopping treatment was gastrointestinal adverse events [12].

SUSTAIN-6 tested cardiovascular safety in 3,297 people with type 2 diabetes, 83.0% of whom had established cardiovascular disease, chronic kidney disease or both. The primary composite of cardiovascular death, non-fatal myocardial infarction or non-fatal stroke occurred in 108 of 1,648 (6.6%) on semaglutide and 146 of 1,649 (8.9%) on placebo, hazard ratio 0.74 (95% CI 0.58–0.95), confirming noninferiority. Non-fatal stroke was lower (HR 0.61), while retinopathy complications were higher (HR 1.76, 95% CI 1.11–2.78) [13].

STEP: weight management

STEP 1 enrolled 1,961 adults with a BMI of 30 or more (or 27 or more with a weight-related condition) and no diabetes, randomised 2:1 to once-weekly subcutaneous semaglutide or placebo plus lifestyle intervention for 68 weeks. Mean body-weight change was −14.9% versus −2.4%, a treatment difference of −12.4 percentage points (95% CI −13.4 to −11.5). A 15% or greater reduction was reached by 50.5% of the semaglutide group and 4.9% of the placebo group [15].

STEP-HFpEF extended the programme to 529 people with heart failure with preserved ejection fraction and obesity. At 52 weeks, the KCCQ clinical summary score improved by 16.6 points versus 8.7 with placebo, body weight changed by −13.3% versus −2.6%, and C-reactive protein fell by 43.5% versus 7.3% [18].

SELECT: cardiovascular outcomes without diabetes

SELECT asked whether the cardiovascular benefit seen in diabetes extends to people with overweight or obesity who do not have diabetes. It enrolled 17,604 patients aged 45 or older with pre-existing cardiovascular disease and a BMI of 27 or more. Over a mean follow-up of 39.8 months, the primary outcome occurred in 569 of 8,803 (6.5%) on semaglutide and 701 of 8,801 (8.0%) on placebo, hazard ratio 0.80 (95% CI 0.72–0.90) [19].

FLOW and ESSENCE: kidney and liver

FLOW randomised 3,533 people with type 2 diabetes and chronic kidney disease and was stopped early for efficacy at a prespecified interim analysis. Major kidney disease events were 24% lower with semaglutide (331 vs 410 first events; HR 0.76, 95% CI 0.66–0.88), and death from cardiovascular causes was also lower (HR 0.71) [20].

ESSENCE is a 240-week phase 3 trial in biopsy-defined MASH with moderate or advanced fibrosis. In the week-72 interim analysis of the first 800 patients, steatohepatitis resolved without worsening of fibrosis in 62.9% versus 34.3%, and fibrosis improved without worsening of steatohepatitis in 36.8% versus 22.4% [22]. The US label now includes noncirrhotic MASH with F2 to F3 fibrosis under accelerated approval, based on improvement of MASH and fibrosis, with continued approval contingent on a confirmatory trial [2].

Oral semaglutide research

Peptides are poorly absorbed from the gut and are degraded by proteolytic enzymes. Oral semaglutide is co-formulated in a tablet with the absorption enhancer SNAC (sodium N-[8-(2-hydroxybenzoyl) aminocaprylate]). Clinical and dog studies showed that absorption happens in the stomach, is confined to an area close to the tablet surface, is transcellular with no evidence of an effect on tight junctions, and requires SNAC, which protects the peptide through local buffering [16].

The oral form has its own outcome trials. PIONEER 6 (3,183 patients) showed cardiovascular noninferiority, with MACE in 3.8% versus 4.8% (HR 0.79) [14]. The larger SOUL trial (9,650 patients, median follow-up 49.5 months) showed a lower MACE risk, 12.0% versus 13.8% (HR 0.86, 95% CI 0.77–0.96), although the confirmatory secondary outcomes did not differ [21]. In weight management, OASIS 1 reported −15.1% versus −2.4% at 68 weeks [17].

Not every hypothesis held. The evoke and evoke+ trials randomised 3,808 people with amyloid-confirmed early Alzheimer’s disease to oral semaglutide or placebo. At week 104 the change in CDR-SB score did not differ (estimated difference −0.08, p=0.57 in evoke; 0.10, p=0.46 in evoke+), and both trials were discontinued for a negative clinical outcome [23].

Safety findings reported in studies

Across the programmes the dominant adverse events were gastrointestinal. The US label lists nausea, vomiting, diarrhoea, abdominal pain and constipation as the most common adverse reactions [1]. In STEP 1, 4.5% of the semaglutide group and 0.8% of the placebo group stopped treatment because of gastrointestinal events [15]. In SELECT, adverse events leading to permanent discontinuation occurred in 16.6% versus 8.2% [19]. In OASIS 1, gastrointestinal adverse events were reported by 80% with oral semaglutide and 46% with placebo [17].

Other reported findings:

• Thyroid C-cell tumours (rodents): semaglutide causes dose- and duration-dependent thyroid C-cell tumours in rodents at clinically relevant exposures; human relevance has not been determined, and the label carries a boxed warning [1].
• Pancreatitis: acute pancreatitis has been observed with GLP-1 receptor agonists, including semaglutide [1].
• Diabetic retinopathy complications: higher with semaglutide in SUSTAIN-6 (HR 1.76) [13].
• Serious adverse events: reported less often with semaglutide than placebo in STEP-HFpEF (13.3% vs 26.7%) and FLOW (49.6% vs 53.8%) [18] [20].

Regulatory status by region

Region	Status	Ref
United States	Approved: subcutaneous (2017) and oral (2019) for type 2 diabetes; subcutaneous (2021) and tablets for weight management; labels also cover cardiovascular risk reduction, kidney outcomes in type 2 diabetes with CKD, and noncirrhotic MASH (accelerated approval)	[1] [2] [8] [9] [10]
European Union	Authorised: type 2 diabetes (2018), oral (2020), obesity and overweight (2022), combination with insulin icodec (2025), MASH (2026)	[11]
Japan	Marketed products listed under the JAN name	[6]
India, China, Canada, UK and others	Varies by country and by patent position; see our country-by-country guide and generic regulatory pathways

Semaglutide in laboratory research

Semaglutide is widely used as a reference GLP-1 receptor agonist in pharmacology and analytical work. Typical laboratory applications, all described in its discovery literature or in regulatory characterisation, include:

• Receptor assays: GLP-1 receptor binding and cAMP potency in recombinant cell lines; the discovery work reported receptor affinity and potency alongside albumin binding [3].
• Albumin-shift assays: comparing potency with and without albumin to rank lipidated analogues; the C18 di-acid showed the largest affinity shift when albumin was added [4].
• Pharmacokinetic models: rat studies were used to compare half-life across fatty acid lengths, from about 1.2 hours for a C12 di-acid to around 7 hours for semaglutide in rats [4].
• Absorption research: SNAC co-formulation and gastric absorption studies [16].
• Analytical characterisation: reversed-phase HPLC or UPLC for purity and related substances, LC-MS for intact mass (monoisotopic mass about 4111.12 Da [6] [7]), peptide mapping to confirm the Aib8, Arg34 and Lys26 modifications, and tests for water, counter-ion and peptide content. Our guide to reading a semaglutide API COA explains each test.

Synthesis, purity and impurities

The US reference product’s peptide backbone is produced by yeast fermentation, and the side chain is then attached at lysine 26 [1]. Liraglutide, the predecessor, was made the same way: recombinant backbone, then a simple chemical acylation [4]. Semaglutide can also be made fully synthetically, by solid-phase peptide synthesis (SPPS) or hybrid routes. The EU’s synthetic peptide guideline, effective 1 June 2026, covers development of synthetic peptides that use biological peptides as the reference medicinal product, and names deletion, truncated and insertion sequences among the impurity types to control [24].

Semaglutide is a demanding target: the 31-residue chain, the sterically hindered Aib at position 8 and the branched lipid side chain each create opportunities for deletion sequences, diastereomers and incomplete or misplaced acylation. Purity therefore needs orthogonal confirmation, not a single HPLC number. For route comparisons see SPPS vs LPPS; for impurity classes and limits see peptide API impurities and specifications.

Storage and stability of lyophilized semaglutide

Research and API material is supplied as a white to off-white lyophilized (freeze-dried, or lyophilisation-processed) powder. Winhycare’s specification for semaglutide is storage at −20 °C, protected from light and moisture. The general science of why dry peptide powders are more stable than solutions, how moisture uptake and temperature excursions drive degradation, and how to handle vials in the lab is covered in lyophilised peptide stability and storage.

Semaglutide versus related peptides

Property	Semaglutide	Tirzepatide	Retatrutide	Cagrilintide
Receptor targets	GLP-1R [1]	GIPR and GLP-1R [26]	GIPR, GLP-1R, GCGR [27]	Amylin receptors (see the cagrilintide overview)
Backbone	Human GLP-1(7–37) [3]	GIP-based, 39 residues [26]	39 residues [28]	Amylin analogue, 37 residues [29]
Lipid	C18 di-acid on Lys26 [1]	C20 di-acid on Lys20 [26]	C20 di-acid on Lys17 [28]	C20 di-acid on the N-terminal amine [29]
Half-life (human)	About 1 week [1]	About 5 days [26]	About 6 days [27]	159–195 h [25]
Status (Oct 2026)	Approved (US, EU) [11]	Approved (US, EU) [11] [26]	Investigational, phase 3 [27]	Investigational, phase 3

Details and citations for each comparator are in the tirzepatide, retatrutide and cagrilintide overviews, and the side-by-side analytical comparison is in semaglutide vs tirzepatide vs liraglutide.

How Winhycare Pharma supplies semaglutide

Winhycare Pharma supplies semaglutide in two separate streams: as research-use-only material for laboratories, and as API to licensed pharmaceutical manufacturers. Material is supplied at ≥99% purity by HPLC with a certificate of analysis for every batch, and custom synthesis (specified salt forms, purity grades, reference-standard quantities) is available. For semaglutide API we support US-DMF Type II and CTD Module 3 documentation. Specifications and documentation are on the semaglutide supplier page; to request a quotation, use the peptide API enquiry form.

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