Semaglutide: the most heavily evidenced compound here, and the most legally eventful

The short version

Semaglutide is a laboratory-modified copy of GLP-1, a hormone the gut releases after a meal to tell the pancreas to release insulin and the brain that eating can stop. Natural GLP-1 is destroyed almost immediately by an enzyme in the blood. Semaglutide is built to survive that enzyme and to cling to a blood protein, which is why a weekly schedule is possible at all.

It is marketed under several trade names, including Ozempic, Wegovy and Rybelsus — one molecule with more than one commercial identity, which is itself a jurisdictional fact rather than a pharmacological one.

The evidence behind it is unusually large for anything discussed on this desk. In a 68-week trial, mean body weight fell 14.9% with once-weekly semaglutide against 2.4% with placebo ^[14]. In 17,604 adults with existing cardiovascular disease and no diabetes, it reduced major adverse cardiovascular events ^[13]. In 3,533 people with type 2 diabetes and chronic kidney disease, it reduced major kidney events ^[12]. It is also, by a clear margin, the compound here with the most documented side effects.

What it is

Semaglutide is a 31-amino-acid acylated analogue of human glucagon-like peptide-1, sharing roughly 94% sequence homology with the natural hormone. Three deliberate structural changes account for everything that makes it usable as a medicine.

At position 8, alanine is replaced by alpha-aminoisobutyric acid, which blocks cleavage by dipeptidyl peptidase-4 — the enzyme that dismantles native GLP-1 within minutes. At position 34, lysine is replaced by arginine. And the single remaining lysine at position 26 carries a C18 fatty di-acid side chain attached through a glutamic-acid and ADO spacer. That lipid tail binds strongly but reversibly to serum albumin, shielding the peptide from kidney clearance and metabolism.

That albumin trick is the structural basis for once-weekly administration, and it is the reason a molecule this fragile in principle behaves like a long-acting drug in practice. In development it carried the designations NNC0113-0217, NN9535 and OG217SC; it reaches patients as a once-weekly subcutaneous injection and as a once-daily oral tablet.

How it works

Semaglutide is a long-acting agonist at the GLP-1 receptor. Where native GLP-1 is dismantled within moments of release, semaglutide persists long enough to support a once-weekly schedule, and every downstream effect follows from that combination of the same signal and a far longer exposure.

At the pancreas it potentiates glucose-dependent insulin secretion from beta cells and suppresses inappropriate glucagon release — glucose-dependent meaning the insulin push only happens when blood sugar is high, which is why the drug alone carries a low hypoglycaemia risk. In the stomach it slows gastric emptying.

The weight effect, though, is mostly central rather than digestive. Rodent work traced it directly: semaglutide accessed the brainstem, area postrema, hypothalamic arcuate nucleus and parabrachial nucleus, reduced food intake and modified food preference — and did so without decreasing energy expenditure ^[16]. In the arcuate nucleus it activates the appetite-suppressing POMC and CART neurons and inhibits the appetite-driving NPY and AgRP neurons. Weight comes off because less food goes in, not because the metabolic furnace burns hotter.

What the research shows

The semaglutide record is built from large randomised outcome trials, which is what separates it from most of what gets discussed as a research peptide.

Weight. In the STEP 1 randomised trial, once-weekly subcutaneous semaglutide 2.4 mg produced a mean body-weight change of -14.9% from baseline to week 68, against -2.4% with placebo, in adults with overweight or obesity and without diabetes ^[14].

Cardiovascular outcomes. In SELECT, 17,604 adults with pre-existing cardiovascular disease and overweight or obesity but no diabetes received once-weekly semaglutide 2.4 mg or placebo; major adverse cardiovascular events fell, with a hazard ratio of 0.80 (95% CI 0.72-0.90; P<0.001) ^[13]. Earlier, in SUSTAIN-6, once-weekly semaglutide at 0.5 or 1.0 mg reduced the composite of cardiovascular death, non-fatal myocardial infarction and non-fatal stroke in type 2 diabetes at high cardiovascular risk, with a hazard ratio of 0.74 (95% CI 0.58-0.95) ^[17].

Kidney outcomes. In the FLOW trial, 3,533 people with type 2 diabetes and chronic kidney disease received once-weekly semaglutide 1.0 mg or placebo; major kidney-disease events — kidney failure, a fall in eGFR of 50% or more, or death from kidney or cardiovascular causes — fell, with a hazard ratio of 0.76 (95% CI 0.66-0.88) ^[12].

Head to head. In SURMOUNT-5, 751 adults with obesity were randomised to tirzepatide or semaglutide for 72 weeks. Mean weight change was -20.2% with tirzepatide against -13.7% with semaglutide, a statistically significant difference (P<0.001) ^[11]. On this endpoint, in this population, semaglutide came second.

Safety in aggregate. A dedicated safety review concluded that semaglutide has an overall favourable risk-benefit profile in type 2 diabetes, with mostly mild-to-moderate transient gastrointestinal effects — nausea in roughly one-third of patients — an increased risk of biliary disease, and pancreatic and thyroid-cancer signals about which definitive conclusions cannot yet be drawn because the incidence is low ^[15].

All amounts named above are what the cited trials administered under supervision. They are reported here as study design and are not a recommendation to anyone.

Reported effects, cautions and safety

What follows first is anecdotal, not clinical evidence — patterns people describe in patient communities and product reviews, unverified and never attached to an amount here.

The benefit reported most often is not weight loss itself but the quieting of what people call food noise: the constant background chatter about the next meal simply going silent, usually within the first week or two. Alongside it, sharply reduced cravings for sweet, fried and high-fat food, with some describing those foods becoming actively off-putting. Weight loss is reported by the large majority, typically as steady over months and slowing after an early stretch, and people usually attribute it to eating far less rather than to any change in activity. Among those using it for type 2 diabetes, markedly better blood-sugar and A1C readings are a common theme. A recurring secondary observation, discussed widely in patient communities, is that the urge to drink alcohol fades along with food cravings.

On the adverse side, nausea is the single most reported effect, sometimes escalating to vomiting, peaking early and after each increase and usually easing as the body adjusts. A distinctive complaint is foul, sulfurous burping that arrives after an increase and can persist. Bowel habits are disrupted in both directions, sometimes alternating. Reflux and heartburn are common, as is tiredness in the first day or two after an injection. Some describe active food aversions, a metallic taste and a heightened, unpleasant sensitivity to smell. A smaller group reports hair shedding and a gaunter face months in, both widely attributed to losing weight quickly rather than to the drug. Headaches, lightheadedness and mild injection-site reactions round out the list. None of this is trial data.

The documented cautions belong to a different class of evidence:

• Gastrointestinal intolerance, especially during escalation. Nausea, vomiting, diarrhoea and constipation dominate the adverse-effect profile in trials and are the leading cause of discontinuation; they are predominantly mild to moderate and transient, and concentrated around the titration period, with nausea in roughly one-third of patients ^[15].
• Thyroid C-cell tumours (boxed warning). The class carries a boxed warning derived from rodent studies in which C-cell tumours occurred at supratherapeutic exposures; a personal or family history of medullary thyroid carcinoma or MEN-2 is a contraindication. Available human data do not establish a clear increase in thyroid cancer attributable to semaglutide, and the signal is best framed as unresolved rather than demonstrated ^[15].
• Acute pancreatitis (class warning). A recognised class warning; treatment is conventionally stopped where pancreatitis is suspected. Pancreatic-cancer signals remain ones on which definitive conclusions cannot yet be drawn owing to low incidence ^[15].
• Gallbladder and biliary disease. An increased risk of cholelithiasis is a real trial and pharmacovigilance finding, attributed largely to the rate and magnitude of weight loss rather than to direct drug toxicity ^[15].
• Pre-existing diabetic retinopathy. In SUSTAIN-6, retinopathy complications were significantly more frequent, with a hazard ratio of 1.76 (95% CI 1.11-2.78), concentrated among participants with pre-existing retinopathy undergoing rapid lowering of HbA1c; the leading interpretation is early worsening driven by the speed of glycaemic correction rather than retinal toxicity ^{[17] [15]}.
• Lean-mass loss. A body-composition substudy found that the weight lost comprised both fat and a meaningful proportion of lean mass, which raises a sarcopenia question particularly in older adults and has motivated research into protein intake and resistance training.
• Weight regain after stopping. Discontinuation is followed by substantial regain, with cardiometabolic improvements reverting toward baseline. This frames obesity pharmacotherapy as a chronic rather than a curative intervention.
• Hair shedding. A pharmacovigilance reporting signal for alopecia exists; it is most consistent with rapid-weight-loss-associated telogen effluvium, a reversible diffuse shedding, rather than direct toxicity.
• Pregnancy. Semaglutide is contraindicated in pregnancy per approved labelling, and because it clears slowly the label advises discontinuation well in advance of a planned pregnancy.
• The oral formulation is unforgiving. Oral semaglutide is co-formulated with the absorption enhancer SNAC and has very low oral bioavailability, so it must be taken fasted, with only a small amount of water, and separated from food, drink and other oral medication. Administration errors substantially reduce the absorbed amount.
• Narrow-therapeutic-index oral drugs. A systematic review of interactions with oral medicines found that delayed gastric emptying does not generally cause clinically significant interactions, but advised monitoring for narrow-therapeutic-index drugs, especially during escalation.

Where it sits on the jurisdictional map

Semaglutide sits at the top rung of the ladder this site is organised around: an approved prescription medicine, with the FDA approving it across type 2 diabetes, chronic weight management, reduction of major adverse cardiovascular events in adults with established cardiovascular disease and overweight or obesity, and, in 2025, metabolic dysfunction-associated steatohepatitis. It is not specifically prohibited by the World Anti-Doping Agency, though athletes are expected to verify current status themselves.

And yet it is the compound here with the most eventful recent legal history — which is exactly why it leads this desk.

During a federally declared shortage in the United States, compounding pharmacies produced semaglutide. The FDA documented dosing errors, adverse events requiring hospitalisation, and products with unverified or non-pharmaceutical active ingredients. When the shortage was declared resolved in 2025, those compounding pathways were curtailed, generating continuing regulatory and telehealth uncertainty. Note what did not happen in that sequence: no trial reported, no finding reversed, no mechanism revised. Availability expanded and then contracted on the strength of a supply declaration, and the evidence base sat unchanged throughout.

The multiple trade names make a smaller version of the same point. One active ingredient carries several commercial identities, and which of them a reader recognises depends on where they live and which indication they encountered it through. A person convinced that two of those names are two different drugs has been misled by a marketing artefact, not by pharmacology.

The reading lesson generalises. When an approval status, a brand name or an availability window differs between two countries, the first question is which of the three has changed — because it is almost never the science.