How Strong Is the Evidence That Semaglutide Protects the Heart?

Semaglutide has moved beyond metabolic management into cardiovascular medicine, but the strength of that evidence depends heavily on who was actually studied. The SELECT trial is the foundation of the cardiovascular claim, and reading it carefully reveals both a real signal and strict limits on how far that signal can be generalized.

What SELECT actually tested

SELECT enrolled adults with overweight or obesity and established cardiovascular disease without diabetes. This population is specific: participants already had documented cardiovascular disease, and the trial explicitly excluded people with diabetes. The primary endpoint was a composite of major adverse cardiovascular events (MACE), defined as cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke, measured over a follow-up period intended to capture sustained cardiovascular outcomes. The intervention was subcutaneous semaglutide at a maintenance dose of 2.4 mg, the same formulation marketed as Wegovy, compared against placebo NEJM.

What changed and what did not

The trial reported a statistically significant reduction in MACE for semaglutide relative to placebo. That is a meaningful outcome for a drug originally developed for glycemic control and weight management. However, the result was driven by specific components of the composite endpoint, and not every individual cardiovascular outcome reached the same threshold of benefit. Understanding which endpoints moved, and by how much, is essential before translating the headline result into broader clinical or regulatory expectations PMC.

Relative risk versus absolute risk

The most commonly cited figure from SELECT is the relative risk reduction in MACE. Relative risk describes the percentage decrease in risk between the treatment group and the placebo group, and it tends to produce larger, more attention-grabbing numbers. Absolute risk reduction is different: it describes the actual difference in events experienced by the two groups, expressed as a raw percentage. A meaningful relative reduction can correspond to a modest absolute reduction, depending on the underlying event rate in the studied population.

This distinction matters because SELECT studied a higher-risk population. Participants had established cardiovascular disease, meaning their baseline probability of experiencing a MACE event was elevated compared with a general population. In such a group, even a modest absolute reduction can produce a notable relative reduction. The same relative percentage applied to a lower-risk population would yield a smaller absolute benefit. This is why extrapolating the SELECT outcome reduction to every Wegovy user is not supported by the trial design NEJM.

Who is excluded by the evidence

The SELECT population does not represent everyone now prescribed semaglutide. The trial excluded patients with diabetes, a group that constitutes a substantial share of real-world GLP-1 receptor agonist use. It also required established cardiovascular disease, meaning participants had already experienced or been diagnosed with conditions such as prior myocardial infarction, stroke, or peripheral arterial disease. Individuals using semaglutide for weight management who do not have established cardiovascular disease were not the subjects of this trial, and the cardiovascular outcome data cannot be directly transferred to them PMC.

This is a material limitation. Semaglutide is now prescribed across a broad demographic, including younger adults seeking weight loss and patients with metabolic risk factors but no prior cardiovascular events. For those patients, the cardiovascular benefit demonstrated in SELECT remains unproven rather than confirmed.

Endpoints, composites, and interpretation

Composite endpoints are standard in cardiovascular outcomes trials because they allow researchers to capture multiple clinically relevant events within a single statistical analysis. They also require careful interpretation. A reduction in MACE indicates that the combined rate of cardiovascular death, nonfatal myocardial infarction, and nonfatal stroke was lower in the semaglutide group. It does not necessarily mean that every individual component was reduced to the same degree, nor that the drug prevents all cardiovascular events.

The SELECT results are best read as evidence that semaglutide reduced the incidence of a defined composite cardiovascular endpoint in a specific high-risk population over the trial follow-up period. They are not evidence that the drug eliminates cardiovascular risk, nor that the same magnitude of effect applies across all patient groups. The distinction between statistical significance, clinical significance, and personal medical decision-making is important here, and treatment decisions should remain individualized NEJM.

Why the development matters

Before SELECT, GLP-1 receptor agonists were understood primarily as metabolic therapies. The trial shifted that frame by demonstrating a cardiovascular benefit in a non-diabetic population with established cardiovascular disease. That is a genuine expansion of the evidence base. But the strength of the claim is bounded by the trial design: the population, the endpoints, the follow-up duration, and the distinction between relative and absolute risk.

For clinicians, regulators, and patients, the practical takeaway is not that semaglutide is a universal cardiovascular therapy. It is that, within a defined group of adults with overweight or obesity and established cardiovascular disease without diabetes, semaglutide reduced the relative risk of a composite cardiovascular endpoint. Translating that finding into broader use requires evidence that does not yet exist, and assuming equivalence across populations would misrepresent what SELECT actually demonstrated PMC.