Semaglutide: what the trials show

What it is

Semaglutide is a lab-made version of GLP-1, a gut hormone, built to last about a week in the body.

Weight loss in trials

The STEP 1 trial randomized 1,961 adults with overweight or obesity and no diabetes to weekly injections of 2.4 mg or placebo for 68 weeks. The semaglutide group lost 14.9 percent of their body weight versus 2.4 percent on placebo. STEP 5 confirmed the result over two years: 15.2 percent versus 2.6 percent at 104 weeks in a randomized trial of 304 adults.

A higher dose does more. STEP UP randomized 1,407 adults to 7.2 mg, 2.4 mg or placebo for 72 weeks. Weight loss was 18.7 percent, 15.6 percent and 3.9 percent. An oral tablet at 25 mg daily produced 13.6 percent versus 2.2 percent at 64 weeks in a randomized trial of 307 adults.

Not all the loss is fat. A 2024 review of six studies covering 1,541 adults found that lean mass made up anywhere from nearly 0 to 40 percent of the weight lost, with clearer lean loss in the larger trials. The data does not show that muscle is spared.

Heart attacks, strokes, kidneys and liver

SELECT enrolled 17,604 people with overweight or obesity (BMI 27 or more) and existing heart disease, but no diabetes. Over a mean of 39.8 months, cardiovascular death, heart attack or stroke occurred in 6.5 percent on semaglutide versus 8.0 percent on placebo, a hazard ratio of 0.80. In type 2 diabetes, SUSTAIN-6 (3,297 people, randomized, 104 weeks) found a hazard ratio of 0.74 for the same endpoint. SOUL tested the oral form in 9,650 people with type 2 diabetes who also had heart disease, kidney disease or both for a mean of 47.5 months: 12.0 percent versus 13.8 percent, hazard ratio 0.86.

For kidneys, FLOW randomized 3,533 people with type 2 diabetes and chronic kidney disease to 1.0 mg weekly or placebo for a median of 3.4 years. Major kidney events (kidney failure, a large drop in kidney function, or death from kidney or heart causes) fell with a hazard ratio of 0.76, and death from any cause with a hazard ratio of 0.80.

For liver disease, ESSENCE randomized 1,197 adults with MASH (a form of fatty liver disease with inflammation) and stage F2 to F3 fibrosis to 2.4 mg weekly or placebo. A planned analysis of the first 800 at 72 weeks found MASH resolved in 62.9 percent on semaglutide versus 34.3 percent on placebo. Fibrosis improved in 36.8 percent versus 22.4 percent.

Alzheimer's: tested in two large trials, both failed

The evoke and evoke+ trials randomized 3,808 people with early Alzheimer's disease to oral semaglutide 14 mg daily or placebo. Semaglutide did not slow cognitive decline.

What happens when you stop

The STEP 1 extension followed 327 participants who stopped semaglutide at week 68. By week 120, they had regained 11.6 of the 17.3 percentage points they originally lost. The net weight loss at week 120 was 5.6 percent. Cardiometabolic improvements also moved back toward baseline.

That is roughly two thirds of the lost weight returning within one year.

Is it approved

Semaglutide has been FDA approved since December 2017, first as Ozempic for type 2 diabetes. The current approved uses in the US include weight management (Wegovy, June 2021), cardiovascular risk reduction (March 2024, based on SELECT), chronic kidney disease in type 2 diabetes (January 2025, based on FLOW), and MASH with F2 to F3 fibrosis (August 2025, accelerated approval). A 25 mg daily tablet was approved in December 2025. A higher injection dose of 7.2 mg weekly was approved in March 2026.

A gray-market vial is not the tested product

Every trial above used pharmaceutical-grade semaglutide. A 2024 analysis tested three vials bought from illegal online pharmacies. Measured purity was only 7.7 to 14.4 percent against the 99 percent on the label, the semaglutide content was 29 to 39 percent above the labelled amount, and all three contained endotoxin (2.2 to 9.0 EU/mg).

In May 2026 the FDA proposed not adding semaglutide to the list of bulk substances that outsourcing facilities may compound from. The comment period was extended to 30 July 2026. No final decision has been published.

Doses used in the studies

Weight loss trials used 2.4 mg injected once weekly after a 16 week step-up. STEP UP tested 7.2 mg weekly. The oral weight loss dose was 25 mg once daily. Diabetes trials mostly used 0.5 to 1.0 mg weekly by injection or up to 14 mg daily by mouth.

These are the doses the studies tested. They are not recommendations.

Side effects and safety

GI side effects are the main problem. In OASIS 4, 74 percent of the semaglutide group reported gastrointestinal events versus 42 percent on placebo. In SELECT, 16.6 percent stopped the drug because of side effects versus 8.2 percent on placebo.

Skin tingling or burning (dysesthesia) affected 22.9 percent on 7.2 mg, 6.0 percent on 2.4 mg and 0.5 percent on placebo.

NAION (sudden vision loss from the optic nerve): In June 2025 the EMA safety committee concluded NAION is a "very rare" side effect of semaglutide, affecting up to 1 in 10,000 users. The signal is weak and low-certainty in both trial data and observational studies. A meta-analysis of 78 randomized trials (73,640 people) found an odds ratio of 3.92, but from very few events and with a wide confidence interval of 1.02 to 15.02. An observational meta-analysis found a raised rate in people with type 2 diabetes (adjusted hazard ratio 1.70) but not in people with obesity alone (adjusted hazard ratio 0.47, confidence interval 0.19 to 1.13).

In SOUL, followed for a mean of 47.5 months, serious adverse events were 47.9 percent on semaglutide versus 50.3 percent on placebo. In SUSTAIN-6, diabetic eye complications were more common on semaglutide (hazard ratio 1.76). Data beyond about four to five years of continuous use is limited.

The takeaway

Semaglutide has the kind of trial data most compounds never get: randomized studies in tens of thousands of people, across weight loss, heart disease, kidney disease and liver disease. The weight loss and the cardiovascular benefit both hold up in large randomized trials.

The limits are equally clear. Most of the weight comes back if you stop. Part of what you lose is lean tissue. GI side effects are common, especially at higher doses, and in SELECT 16.6 percent stopped the drug because of side effects, against 8.2 percent on placebo. And a gray-market vial is not the product any of these trials tested.

I don't gatekeep.

Tell me what you're after and where you live. I'll email you my source.

Get George's source