GLP-1 drugs and the trial results

What GLP-1 drugs are approved for, the main trials in people, the known risks, and how FDA-approved products differ from pharmacy-made or other unapproved versions.

Bottom line

Several medicines based on GLP-1, a gut hormone, have uses approved by the FDA (the U.S. Food and Drug Administration). They also have strong evidence from randomized trials, which assign people to a drug or a comparison by chance. Their benefits and risks hold for a specific product, group of people, and approved use. They can't be assumed for every GLP-1 product or every use. They also can't be assumed for a compounded version (a custom version of a drug mixed by a pharmacy).451

Evidence in people
Established for specific approved uses
Approval status
Semaglutide and tirzepatide are each found in more than one FDA-approved product, and the products have different approved uses on their labels. Compounded versions are not FDA-approved. The FDA does not review them for safety, effectiveness, or quality before they are sold.46571
Biggest unknown
Trial averages cannot predict how one person will respond or how well they will handle the drug. They also can't predict whether that person will keep taking it, or how they will do over the longer term. Results for an approved product can't settle what is really in an unapproved version. Nor can they settle its quality, sterility, or how well it performs.9111

Where research exists

Filled markers show which stages of research this guide covers. They don't rate how well something works, how safe it is, or how good the studies are.

  • Health results in peopleCovered in this guide
  • Blood tests and body measurements in peopleCovered in this guide
  • Animal studiesCovered in this guide
  • Lab studies in cellsCovered in this guide
  • Untested idea or marketing claimNo useful studies found

What GLP-1 medicines are

GLP-1 is a hormone that the gut releases after eating. It helps control insulin release, which depends on the level of blood sugar. It also helps control glucagon (a hormone that raises blood sugar), how fast the stomach empties, appetite, and how much a person eats. GLP-1 receptor agonists are medicines designed to act on the GLP-1 receptor (the spot on cells that GLP-1 works through). They act for longer than the body's own signal.46

Tirzepatide is often grouped with GLP-1 receptor agonists. But its FDA labels describe it as acting on two receptors, for GIP and GLP-1 (GIP is another gut hormone). That difference matters when comparing how the drugs work and the evidence for them.57

Each brand name refers to a specific product. Wegovy and Ozempic both contain semaglutide, but they have different approved uses on their labels. The same is true of Zepbound and Mounjaro, which both contain tirzepatide. Each product's approval rests on its own evidence and label.4657

As of this guide's last evidence check, the labels for these products include these uses:

  • blood sugar control in certain people with type 2 diabetes
  • lowering certain heart and blood vessel risks, or kidney risks, in certain high-risk groups
  • long-term weight management in certain groups
  • moderate-to-severe obstructive sleep apnea in adults with obesity (the airway keeps getting blocked during sleep, so breathing stops and starts)
  • for Wegovy, noncirrhotic MASH with moderate-to-advanced fibrosis

The exact label sets what is approved. MASH is a liver disease in which fat in the liver causes inflammation and damage. The Wegovy use covers MASH with moderate to advanced scarring (fibrosis) that has not reached cirrhosis. This use was approved under accelerated approval (a faster FDA route based on early measures that are likely to predict a benefit). That approval may depend on further evidence confirming the benefit, as its label explains.4657

Why GLP-1 drugs draw interest

The interest goes beyond weight. Trials of specific products have looked at blood sugar, body weight, and heart and blood vessel events. They have also looked at kidney health, signs of liver disease, and how severe sleep apnea is. Each result holds for its own outcome and group of people.4651012

Large average changes can come with wide differences between people. They can also come with side effects, people stopping treatment, and weight coming back after treatment changes. An average from a trial cannot predict one person's result.911

These medicines are studied as part of ongoing care, usually along with programs for diet and physical activity. Trials leave some questions open. These include access (whether people can get the drug), sticking with treatment, and lean mass (muscle and other tissue that isn't fat). They also include keeping results over the long term and what happens after a person stops treatment.9114

The main trials

STEP 1 (a trial of semaglutide for weight) included 1,961 adults with overweight or obesity who did not have diabetes. They were assigned by chance to semaglutide or a placebo (a dummy treatment) for 68 weeks. Both groups also got a lifestyle program (help with diet and exercise). On average, people on semaglutide lost 14.9% of their body weight, and people on placebo lost 2.4%. Stomach and gut problems led 4.5% of the semaglutide group to stop treatment, compared with 0.8% of the placebo group.9 These results describe one trial group and one product. They can't promise how much weight any one person will lose. They also can't serve as evidence for a compounded version.

SELECT (a trial testing whether semaglutide lowers heart and blood vessel events) included 17,604 adults with overweight or obesity. All had existing heart or blood vessel disease, and none had diabetes. They were assigned by chance to semaglutide or placebo. A major heart or blood vessel event, such as a heart attack or stroke, happened in 6.5% of the semaglutide group and 8.0% of the placebo group. The hazard ratio was 0.80. (A hazard ratio compares how often events happen in two groups over time. A number below 1 means a lower rate in the semaglutide group.) Adverse events are side effects or other health problems during a trial. They led 16.6% of the semaglutide group to stop treatment for good, compared with 8.2% of the placebo group.10 The heart and blood vessel result applies to this high-risk group and this product.

SURMOUNT-1 (a trial of tirzepatide for weight) included 2,539 adults with obesity or overweight who did not have diabetes. They were assigned by chance to one of three tirzepatide groups or to placebo for 72 weeks. On average, people in the three tirzepatide groups lost 15.0%, 19.5%, and 20.9% of their body weight, and people on placebo lost 3.1%. Adverse events led 4.3%, 7.1%, and 6.2% of those tirzepatide groups to stop treatment, compared with 2.6% of the placebo group.11 These are trial averages. They can't predict one person's result. They also can't show that a compounded product is the same as the one tested.

SURMOUNT-OSA was made up of two randomized trials with 469 adults in total. All had obesity and moderate-to-severe obstructive sleep apnea. After 52 weeks, tirzepatide lowered the apnea–hypopnea index more than placebo did. (This index counts how often breathing stops or becomes shallow during each hour of sleep.) That was true in the group not using positive airway pressure therapy. This therapy uses a machine that blows air through a mask to keep the airway open. It was also true in the group that kept using it.125 The trials studied a defined group of people with sleep apnea. They tell us nothing about improving sleep in general.

Don't rank the percent changes from separate trials as if the studies had compared the drugs directly. The trials differ in who took part, which products were used, and how each study was run. They also differ in exactly what each one set out to measure, how long people were followed, and how missing data was handled.911

Animal and cell research

For approved medicines based on GLP-1, treatment decisions can draw on large trials in people and on product labels. Findings from cells and animals about how the drugs work can add background. But the benefits and harms seen in people, and evidence about the specific product, should count for more.1045

The current Wegovy and Zepbound labels both carry boxed warnings. A boxed warning is the FDA's most serious kind of warning, printed in a box on the label. These warnings are there because semaglutide and tirzepatide caused tumors of thyroid C-cells (one type of cell in the thyroid gland) in rodents. The labels state that it is unknown whether this matters for humans. They also say the drugs should not be used by anyone with a personal or family history of two conditions. One is medullary thyroid carcinoma (a type of thyroid cancer). The other is MEN 2 (an inherited condition that causes tumors in hormone-making glands).45 An accurate summary keeps both facts. The warning is serious, and whether it matters for humans is unknown.

Known risks

Stomach and gut problems are among the most common side effects listed on the labels. They include nausea, diarrhea, vomiting, constipation, belly symptoms, and indigestion. Severe stomach and gut reactions can happen. Both labels advise against use in people with severe gastroparesis (a condition in which the stomach empties very slowly).45

The current labels also cover pancreatitis (an inflamed pancreas), gallbladder disease, and sudden kidney injury linked to losing too much body fluid. They address serious allergic reactions, and the risk of low blood sugar when these drugs are used with some diabetes medicines. Further topics are diabetic eye disease (diabetic retinopathy) and food or liquid getting into the lungs during anesthesia or deep sedation.4567

These medicines slow how fast the stomach empties. That can affect how the body absorbs medicines taken by mouth. Tirzepatide's labels also include a specific precaution about birth control after starting treatment or raising the dose. So checking a person's other medicines is part of judging the risk of a given product.457

Under the current labels, using these drugs for weight loss is not compatible with pregnancy. Each label gives its own instructions about pregnancy and about people who could become pregnant. This guide leaves out dosing schedules on purpose.45

In January 2026, the FDA reported on its full review of GLP-1 receptor agonists. The review did not find a higher risk of suicidal thoughts or behavior with these drugs. The FDA asked for that warning to be removed from the labels that carried it. The updated Wegovy and Zepbound labels show that it has been removed.845

Compounded and unapproved versions

Before an approved drug goes on sale, the FDA reviews that specific product. Compounded and other unapproved GLP-1 versions don't get that FDA review for safety, effectiveness, or quality. That holds even when an ingredient name looks like the name of an approved medicine.1

The FDA says semaglutide sodium and semaglutide acetate are different active ingredients from the semaglutide in approved drugs. These are salt forms (semaglutide paired with another chemical). The agency says it has no information showing that these salt forms have the same chemical properties or act the same way in the body. It also says it knows of no lawful basis for using them in compounding.1

The FDA has described problems with compounded semaglutide and tirzepatide. These include dosing errors, fake labels, concerns about shipping and storage, and reports of adverse events (health problems after use). Reports like these come in only when someone chooses to send one. They can't show how often a problem happens, and they can't prove the drug caused it. It is also likely that some problems go unreported, because many state-licensed pharmacies don't have to send reports to the FDA.1

As of May 31, 2026, the FDA said it had received 990 adverse-event reports linked to compounded semaglutide. It had received more than 730 linked to compounded tirzepatide. The FDA also warned that it may not be possible to tell whether the drugs caused these events. It added that the counts are likely incomplete.1 These are counts from safety monitoring. Without knowing how many people used these products in all, the counts give no estimate of risk. They also can't show that the product caused every reported event.

In April 2026, the FDA updated its policy for compounders. It said semaglutide and tirzepatide were not on the 503B bulks list or the FDA drug shortage list. (The 503B bulks list names raw ingredients that outsourcing facilities, which compound drugs on a larger scale, may use.) The update also restated the limits on making products that are essentially copies of drugs already on the market. Sections 503A and 503B of federal law set different conditions for compounding. Neither one is a way to get the compounded drug approved.2

On April 30, 2026, the FDA proposed leaving semaglutide, tirzepatide, and liraglutide off the 503B bulks list. It had found no clinical need for outsourcing facilities to compound them from bulk ingredients. The FDA called this a proposal. It said it would consider public comments before making a final decision.3

Questions to ask your doctor

Which exact product, and which use on its label, are we talking about? Do my diagnosis and my risks match the people studied in the evidence behind it?4657

Which result matters most to me: blood sugar control, weight, heart and blood vessel risk, kidney risk, sleep apnea, or another goal? What is the evidence for that result in absolute terms, such as how many people out of 100 were helped?10126

Which of these change the balance for me: reasons I shouldn't take it, past stomach or gut problems, or past gallbladder or pancreas problems? What about kidney risks, eye disease, pregnancy, upcoming medical procedures, and other medicines that could interact with it?45

If the product is compounded or unapproved in another way, what evidence shows its exact active ingredient and what it really is? What shows its quality, its sterility, how it was stored, and that it works the same in patients as the approved drug? A trial of an approved product can't answer those questions for it.1

What is the long-term plan for checkups, diet, physical activity, side effects, getting the drug, sticking with it, and changes to treatment? What is still uncertain about the things that matter most to me?4911

References

  1. 1
    FDA's Concerns with Unapproved GLP-1 Drugs Used for Weight LossU.S. Food and Drug Administration · 2026-05-31 · FDA document

    The page separates approved products from unapproved versions. It notes that side-effect reports sent in voluntarily are incomplete and can't by themselves show that a drug caused a problem.

  2. 2
    FDA Clarifies Policies for Compounders as National GLP-1 Supply Begins to StabilizeU.S. Food and Drug Administration · 2026-04-01 · FDA document

    This FDA policy page says semaglutide and tirzepatide were not on the 503B bulks list or the FDA drug-shortage list on April 1, 2026. It explains the separate rules for the two kinds of compounding pharmacy, 503A and 503B.

  3. 3
    FDA Proposes to Exclude Semaglutide, Tirzepatide, and Liraglutide on 503B Bulks ListU.S. Food and Drug Administration · 2026-04-30 · FDA document

    The FDA called this a proposal and said it would read public comments before making a final decision.

  4. 4
    Wegovy (semaglutide) Prescribing InformationU.S. Food and Drug Administration · 2026-02-01 · FDA-approved drug label

    What the label says about approved uses and safety applies to Wegovy itself and the people the label names. It does not carry over to other semaglutide products or pharmacy-made versions.

  5. 5
    Zepbound (tirzepatide) Prescribing InformationU.S. Food and Drug Administration · 2026-02-01 · FDA-approved drug label

    What the label says about approved uses and safety applies to Zepbound itself and the people the label names. It does not carry over to other tirzepatide products or pharmacy-made versions.

  6. 6
    Ozempic (semaglutide) Prescribing InformationU.S. Food and Drug Administration · 2026-06-01 · FDA-approved drug label

    Ozempic's approved uses, for type 2 diabetes and for lowering heart and kidney risks, apply to Ozempic itself and the people its label names. They do not carry over to other semaglutide products.

  7. 7
    Mounjaro (tirzepatide) Prescribing InformationU.S. Food and Drug Administration · 2026-01-01 · FDA-approved drug label

    Mounjaro's approved use, controlling blood sugar, applies to Mounjaro itself and the people its label names. It does not carry over to Zepbound or pharmacy-made tirzepatide.

  8. 8
    FDA Requests Removal of Suicidal Behavior and Ideation Warning from GLP-1 Receptor Agonist MedicationsU.S. Food and Drug Administration · 2026-01-13 · FDA document

    The FDA's full review did not find a higher risk, and the FDA asked for the warning to be removed. This replaces its early 2024 notice.

  9. 9
    Once-Weekly Semaglutide in Adults with Overweight or ObesityThe New England Journal of Medicine (PubMed record) · 2021-03-18 · Trial in people

    STEP 1 tested one semaglutide plan, along with diet and exercise advice, in adults without diabetes. Its averages can't predict one person's result, and they say nothing about pharmacy-made versions.

  10. 10
    Semaglutide and Cardiovascular Outcomes in Obesity without DiabetesThe New England Journal of Medicine (PubMed record) · 2023-12-14 · Trial in people

    SELECT included adults who already had heart and blood vessel disease and overweight or obesity, but not diabetes. Its result applies to those people and that product.

  11. 11
    Tirzepatide Once Weekly for the Treatment of ObesityThe New England Journal of Medicine (PubMed record) · 2022-07-21 · Trial in people

    SURMOUNT-1 tested one tirzepatide product in adults without diabetes. Its group averages promise no result for any one person, and they say nothing about pharmacy-made versions.

  12. 12
    Tirzepatide for the Treatment of Obstructive Sleep Apnea and ObesityThe New England Journal of Medicine (PubMed record) · 2024-10-03 · Trial in people

    The two trials included adults with obesity and moderate to severe sleep apnea (breathing that stops and starts during sleep). The results apply to those people and that product.

Evidence last checked August 22, 2026.

Educational use only. This guide does not diagnose, prescribe, recommend a product or replace care from a licensed clinician.