GLP-1 medications: what they do, and what happens when you stop
Few drug classes have arrived this quickly into general use, and few are surrounded by this much noise. The short version: the weight loss is real, the cardiovascular benefit is real, the side effects are real, and the part that gets least attention — what happens when you stop — matters more than most of the marketing.
What they are
GLP-1 is a hormone the gut releases after eating. It slows gastric emptying, prompts insulin release when glucose is high, and acts on appetite regulation in the brain. These drugs are engineered versions that persist far longer than the natural hormone.
The practical effect people describe is a reduction in what has come to be called food noise — the background preoccupation with eating. That is a pharmacological effect on appetite regulation, which is worth stating plainly, because obesity has long been treated as a failure of character rather than a condition with physiology behind it. A drug that works on that physiology is evidence about the condition, not just a treatment for it.
What the trials show
Weight loss varies by agent and dose. In trials, semaglutide at its obesity dose produced roughly 15% mean body weight reduction, and tirzepatide roughly 20% at the highest dose. The oral options approved this year land somewhat lower — around 13.6% for oral semaglutide in OASIS 4, and about 12.4% for orforglipron at its top dose.
More important than the weight numbers is the outcome data. The SELECT trial tested semaglutide in people with established cardiovascular disease and obesity but without diabetes, and found roughly a 20% reduction in major adverse cardiovascular events. That is the finding that moved these from weight-loss drugs to cardiometabolic drugs, and it is the reason a clinician may bring them up with someone who has not raised weight at all.
The options as of now
The class expanded significantly in 2026, and injections are no longer the only route:
- Semaglutide injection — the longest track record in obesity, and the agent behind the cardiovascular outcome data
- Tirzepatide injection — acts on two receptors rather than one, and produces the largest average weight loss of the agents currently approved
- Oral semaglutide — approved for weight management and cardiovascular risk reduction, launched at the start of 2026
- Orforglipron — approved in April 2026, a once-daily pill with no food or water timing restrictions, which is a meaningful practical difference
Which one fits depends on tolerance, cost, coverage, and whether a daily pill or a weekly injection is more realistic for a given person.
What is coming: retatrutide
The next agent likely to matter is retatrutide, which adds a third target. Where semaglutide acts on the GLP-1 receptor and tirzepatide on GLP-1 and GIP, retatrutide also acts on the glucagon receptor — a counterintuitive addition, since glucagon raises blood sugar, but one that appears to increase energy expenditure alongside the appetite effects.
It is well past early trials. Results from the Phase 3 TRIUMPH programme have been reporting through 2025 and 2026:
- TRIUMPH-1 — 25.0% average weight loss at the 12 mg dose at 80 weeks, against 3.9% on placebo
- TRIUMPH-3 — 22.6% in adults with severe obesity and established cardiovascular disease
- TRIUMPH-4 — 28.7% at 68 weeks in people with obesity and knee osteoarthritis, alongside substantial reductions in joint pain scores
Those are the largest averages the class has produced. Several caveats belong with them. Seven further Phase 3 trials are due to complete during 2026. Lilly has said it plans to file for approval in the first quarter of 2027, which puts any realistic availability well after that. The tolerability profile has drawn attention from analysts reading the same results. And there is as yet no cardiovascular outcome trial of the kind that made semaglutide’s case — larger weight loss is a reasonable proxy for benefit, not a demonstration of it.
The side effects are mostly gastrointestinal
Nausea, vomiting, constipation, and diarrhea are common, usually worst during dose escalation, and usually improve. They are also the main reason people stop. Slower titration helps.
Less common but more serious concerns include pancreatitis and gallbladder disease. There is a boxed warning regarding thyroid C-cell tumors based on rodent data, which makes a personal or family history of medullary thyroid carcinoma or MEN2 a contraindication.
The part that gets undersold: muscle
Weight lost on these drugs is not all fat. A meaningful fraction is lean mass, which matters more with age than the scale does — muscle mass and strength predict falls, fractures, independence, and mortality.
Resistance training and adequate protein are not optional add-ons during weight loss. They are what determines whether the weight you lose is the weight you wanted to lose.
What happens when you stop
This is the honest centre of the conversation and it is rarely the headline.
In the extension of the STEP 1 trial, participants who stopped semaglutide regained about two-thirds of the weight they had lost within a year, and cardiometabolic markers moved back toward baseline with it.
That is not a failure of the drug. It is what treating a chronic condition looks like — nobody expects blood pressure to stay down after stopping an antihypertensive. But it reframes the decision. Starting one of these is closer to beginning long-term treatment than to completing a course, and the cost and access questions should be answered on that basis rather than on a few months of results.
Compounded versions are now largely closed off
For roughly two years, the drug shortages permitted compounding pharmacies to produce semaglutide and tirzepatide, and a substantial telehealth industry grew on top of that. Those shortages have since been resolved, and with them the legal basis.
The FDA removed injectable semaglutide from the shortage list in February 2025; 503A pharmacies had until April 2025 to stop compounding it and 503B facilities until May 2025, with parallel earlier deadlines for tirzepatide. In April 2026 the agency proposed excluding semaglutide, tirzepatide, and liraglutide from the 503B bulks list entirely, on the basis that there is no clinical need for outsourcing facilities to make them.
Safety data sits behind that. The FDA has received hundreds of adverse event reports involving compounded versions of these drugs, a substantial share of them dosing errors — unsurprising when the same molecule is supplied at varying concentrations, in vials rather than pre-set pens, with patients drawing up their own doses.
If you are currently getting one of these from a telehealth service at a price well below the branded product, it is worth asking exactly what you are receiving and under what authority.