Semaglutide

Wegovy has product-specific indications for weight management and cardiovascular risk reduction. Its injection also has accelerated approval for adults with noncirrhotic MASH and moderate-to-advanced fibrosis (F2–F3); that indication should not be transferred to oral semaglutide or research products.^[1]

Jurisdiction

United States

Product

WEGOVY (semaglutide)

Indication

Product-specific weight management and cardiovascular indications; injection for noncirrhotic MASH with F2-F3 fibrosis in adults

Record date

2026-09-08

• Eligibility, ages and indications differ by formulation; this is not a comprehensive list of semaglutide product approvals.

Wegovy carries a boxed warning for thyroid C-cell tumors and is contraindicated with a personal or family history of medullary thyroid carcinoma or MEN2. The warning derives from rodent findings; human relevance remains uncertain. The current label records removal of the suicidal-behavior/ideation warning in February 2026.^[1]

• A rodent tumor finding should not be presented as established human thyroid-cancer causation; this summary is not the complete label.

Wegovy labeling warns of hypoglycemia, especially with insulin or insulin secretagogues. The label distinguishes supporting semaglutide glycemic-control studies from the unevaluated Wegovy-specific insulin combination.^[1]

• This label warning does not establish that every insulin formulation and Wegovy combination was studied.

Wegovy can cause severe gastrointestinal reactions and is not recommended with severe gastroparesis. Its label describes postmarketing acute kidney injury reports, mostly following gastrointestinal symptoms that led to dehydration.^[1]

• Postmarketing reports do not provide a reliable incidence or prove causality in every case.

Wegovy labeling warns that pulmonary aspiration has been reported rarely in patients receiving GLP-1 receptor agonists during general anesthesia or deep sedation, in some cases despite reported adherence to preoperative fasting. The label states available data are insufficient to recommend specific fasting or drug-discontinuation changes to reduce this risk.^[1]

• These are rare postmarketing reports, not an established incidence rate or a tested mitigation protocol.

FDA has reported that, as of May 31, 2026, it had received about 990 adverse-event reports associated with compounded semaglutide products. FDA states it cannot always determine whether the drug directly caused a reported event, and that compounded products are not FDA-approved or reviewed for safety, effectiveness or quality before marketing.^[3]

• A raw report count is not an incidence rate, is likely affected by underreporting (state-licensed non-outsourcing pharmacies need not report to FDA), and describes compounded products rather than the FDA-approved Wegovy/Ozempic products.

Wegovy labeling reports a semaglutide elimination half-life of approximately one week. Oral and injected formulations have different absorption and dosing characteristics, so equal amounts do not imply equal exposure.^[1]

• The estimate describes semaglutide elimination in the labeled product evidence, not the quality of unapproved preparations.

In STEP 1, adults with overweight or obesity without diabetes lost a mean 14.9% of body weight with weekly injected semaglutide versus 2.4% with placebo over 68 weeks, alongside lifestyle intervention. Gastrointestinal events caused discontinuation more often with semaglutide.^[2]

• A trial average is not an individual prediction or a comparison with other drugs.
• Only the published abstract was inspected; no PMC full text exists for this article, so full methods, tables and supplementary appendices were not reviewed.

A Bayesian network meta-analysis published in 2025 pooled 19 randomized controlled trials covering 29,506 adults with a body-mass index of 25 kg/m2 or more, comparing liraglutide, semaglutide, survodutide, tirzepatide, retatrutide and placebo at 36 weeks or later. Its abstract reports that retatrutide and the dual agonists produced equivalent mean weight loss of 11.0 kg, more than the GLP-1 receptor agonists at 9.0 kg, and that for reaching at least 15% weight loss the odds ratios were 54.6 for retatrutide, 16.4 for the dual agonists and 9.0 for the GLP-1 receptor agonists. It also reports that retatrutide carried the highest risk of adverse events, and that in meta-regression the presence of type 2 diabetes was associated with less weight loss. These are indirect comparisons across trials, not head-to-head results.^[7]

• Abstract-only inspection. The included-trial list, risk-of-bias assessment, network structure and heterogeneity statistics were not read, because no lawful full-text copy was reachable.
• The abstract does not name the 19 pooled trials or state their development phase, so this record cannot show which trial of any compound contributed to which estimate. The survodutide and retatrutide results recorded elsewhere in this reference come from phase 2 dose-finding trials, the same records also list completed phase 3 trials of both compounds, and semaglutide and tirzepatide are approved for weight management, so the pooled classes may combine evidence of unequal maturity.
• A network meta-analysis estimates comparisons between treatments that were mostly never tested against each other. Its rankings are model outputs, not measured differences.
• The GLP-1 receptor agonists and the dual agonists are pooled by class in the reported results, so the abstract gives no separate estimate for semaglutide, survodutide or tirzepatide; retatrutide is reported on its own because the abstract treats it as a group of one.
• The meta-regression associations are observational comparisons across trials, not randomized ones.
• This finding is recorded as cross-compound context. It is not evidence about any single compound's own trial results, which are recorded separately on each record.

Cagrilintide and semaglutide together are not an extrapolation from separate studies: the pair itself has been tested in randomized trials, which is unusual among the combinations described in this reference. REDEFINE 1 randomized 3,417 adults to the combination, to each agent alone, or to placebo, so both single agents were studied as active comparator groups alongside the combination; REDEFINE 2 and REIMAGINE 1 tested the combination against placebo in adults with type 2 diabetes. In the blood-pressure analysis of REDEFINE 1, the combination showed larger reductions in systolic and diastolic blood pressure, and a larger proportion of participants losing at least 20% of body weight, than either agent alone, although each estimate is reported against placebo rather than as a test between the groups. Gastrointestinal adverse events were reported by more participants on the combination than on placebo in both REDEFINE trials. This is evidence about one specific named pair at the doses studied, not about combining amylin and GLP-1 receptor agonists generally.^[4][5][6]

• The arm-by-arm figures come from the blood-pressure analysis of REDEFINE 1, in which the single-agent comparisons against placebo are described as post hoc and blood pressure was not the outcome the trial was designed to test. No statistical comparison of the combination against either single agent was reported in the inspected passages, and the trial's own primary comparison published in its abstract is against placebo only.
• Two of the three sources were inspected as abstracts, so no adverse-event table was read, and none of the passages inspected for this finding contains a comparison of discontinuations between the combination and either single agent. The direction of that comparison is published separately, in the curated interaction finding for this pair.
• The trials studied fixed doses of both components. Nothing here describes other doses, other amylin analogues, other GLP-1 receptor agonists, or products combined outside a trial.
• Larger weight loss and lower blood pressure in a trial are measured outcomes of a studied product. They are not a recommendation to combine these substances, and this reference gives no combination or regimen guidance.

OZEMPIC labeling warns that concomitant insulin can increase hypoglycemia risk and may require clinician-directed insulin adjustment. In SUSTAIN 5, semaglutide improved glycemic control on basal insulin; the trial pre-emptively reduced participants' insulin dose 20% at entry, and the numerical excess in hypoglycemia was not statistically conclusive.^[8][9]

• A supervised diabetes combination can have clinical benefit; this is not a blanket prohibition.
• The trial did not establish a doubling of risk, and its own protocol-mandated insulin-dose reduction at entry limits comparison to unmodified add-on use.

ZEPBOUND labeling does not recommend coadministration with another tirzepatide product or a GLP-1 receptor agonist such as semaglutide.^[10]

• Not classified as a formal contraindication; no amount of additive toxicity is established.

REDEFINE 1 assigned 3,417 adults with overweight or obesity and without diabetes to cagrilintide and semaglutide together, to semaglutide alone, to cagrilintide alone or to placebo for 68 weeks, with 2,108 on the combination and 302 on each single agent, so adverse events on the combination could be compared directly against each agent by itself. That comparison sits in the trial report, which this reference could not reach; an editorial in the same journal issue describes what it showed, in direction only. Gastrointestinal effects leading to permanent discontinuation of the trial product were more frequent on the combination than on semaglutide alone. Injection-site reactions were more common on the combination and on cagrilintide alone than on semaglutide alone or placebo. Serious adverse events were more common on the combination than on semaglutide. The editorial describes the pattern of adverse events in REDEFINE 2, a separate trial of the combination in adults with type 2 diabetes, as similar, and records six deaths across the two trials combined, all in groups receiving the combination and including one suicide in each trial; it adds that earlier studies of suicidality during treatment with GLP-1 receptor agonists have not supported a causal link, while saying that the question continues to be of concern.^[4][11]

• Every one of these comparisons is a direction without a size. The editorial states no percentage, count, confidence interval or statistical test for any of them, so how much more frequent each was is unknown to this reference. The trial's own safety table was never read: the publisher returned HTTP 403, there is no PubMed Central copy of the trial report, and the trial registration has no posted results.
• The account of the cross-arm comparison is a commentator's, not the trial's. An editorial reports no new data and performs no independent analysis. Where this editorial's figures can be checked against the trial's own blood-pressure analysis they differ, apparently because it quotes a different estimand, which is recorded on its source record.
• The deaths are pooled across two trials, one of them in adults with type 2 diabetes, whom the rest of this finding does not describe. They are reported without denominators and without a placebo comparison, and the groups receiving the combination were by far the largest. A death or a suicide during a trial is not thereby shown to have been caused by the treatment.
• The trial gave the two agents in a single combined injection rather than as two separate products, at the fixed amounts it selected, in adults without diabetes, for 68 weeks. How that product is supplied is not otherwise established here, and nothing in this finding describes other amounts, other amylin analogues, other GLP-1 receptor agonists, other populations, or these substances combined outside a trial.
• The two records cited here hold the same two publications already cited on the cagrilintide compound record, under separate identifiers because the interaction dataset is authored in its own file. Meeting them in both places is one body of evidence, not two.
• Six correspondence items about the trial report are indexed in PubMed and none was inspected. None is an erratum, retraction or expression of concern.
• A trial result is a measured outcome of a product under study. It is not a reason to combine these substances, and this reference gives no combination guidance.