What it is
Semaglutide is a lab-made copy of a gut hormone called GLP-1. Your body releases GLP-1 after you eat. It tells the pancreas to make insulin, tells the stomach to empty more slowly, and tells the brain you are full. The natural hormone lasts only a couple of minutes. Semaglutide was engineered to last about a week, which is why it is a once-weekly shot.
It is sold under three brand names by Novo Nordisk. Ozempic is the diabetes version. Wegovy is the higher-dose weight-management version. Rybelsus is a daily tablet.
Semaglutide is a 31-residue GLP-1(7-37) analogue with three modifications to native GLP-1: an Aib substitution at position 8 conferring DPP-4 resistance, an Arg substitution at position 34 to prevent acylation at that lysine, and a C18 fatty diacid attached at Lys26 through a γGlu-2xOEG linker. The diacid drives high-affinity, reversible albumin binding, reducing renal clearance and yielding a plasma half-life of roughly 165 hours. It is a full GLP-1 receptor agonist acting on pancreatic β-cells (glucose-dependent insulin secretion), α-cells (glucagon suppression), gastric motility and hypothalamic and hindbrain appetite circuits.
The oral formulation co-formulates semaglutide with SNAC (sodium N-(8-[2-hydroxybenzoyl]amino)caprylate), which locally raises gastric pH and promotes transcellular absorption across the gastric epithelium. Oral bioavailability is about 1%, hence the 3–14 mg daily doses.
Who made it and when
Novo Nordisk chemists, led by Lotte Bjerre Knudsen, designed semaglutide in the early 2010s as a longer-lasting follow-up to their daily drug liraglutide. The company owns the core patents, which run into the early 2030s in the United States.
Semaglutide emerged from Novo Nordisk’s programme to extend the albumin-binding strategy used in liraglutide. Lau and colleagues described the structure–activity work in the Journal of Medicinal Chemistry in 2015. The compound patent, US 8,129,343, is held by Novo Nordisk; regulatory exclusivity and secondary formulation and dosing patents extend protection beyond the compound patent in most markets.
What the data say
The big trials are consistent. In people with obesity and no diabetes, the 2.4 mg weekly dose cut body weight by about 15% over 16 months, against roughly 2% on placebo. A separate trial in people with existing heart disease found about one fifth fewer heart attacks, strokes and cardiovascular deaths. It also protected the kidneys in people with diabetes, and improved liver inflammation in MASH. A higher 7.2 mg dose produced about 21% weight loss. The Alzheimer’s trials failed.
STEP 1 (n=1,961) reported a treatment difference of −12.4 percentage points in body weight at 68 weeks. SELECT (n=17,604) reported a hazard ratio of 0.80 for 3-point MACE over a mean 39.8 months. FLOW (n=3,533) reported a hazard ratio of 0.76 for the composite kidney outcome and was stopped early for efficacy. ESSENCE part 1 met both histological primary endpoints at 72 weeks. STEP UP extended the dose–response curve to 7.2 mg. EVOKE and EVOKE+ in early Alzheimer’s disease did not meet their primary endpoint on CDR-SB.
Regulatory picture
Semaglutide is fully FDA approved under its brand names. What is not approved is anyone else’s version. During the 2022–2025 shortage, compounding pharmacies were allowed to make copies. The FDA declared the shortage over in February 2025 and set deadlines in April and May 2025 for compounders to stop. Since then the FDA has sent warning letters to sellers of “compounded” or “research” semaglutide.
Semaglutide is not on the 503A bulk drug substances list and does not need to be, because it is a component of FDA-approved drugs. Compounding of essentially-copies of an approved product is permitted only while the product appears on the FDA drug shortage list under section 503A(b)(1)(D) and 503B(a)(2)(A). With the shortage removed on 2025-02-21, that pathway closed on 2025-04-22 for 503A and 2025-05-22 for 503B. Personalised compounded formulations that differ materially from the approved product remain a contested grey area.