SELECT: semaglutide cut major heart events by 20% in people who already had heart disease

Weekly semaglutide 2.4 mg cut major cardiovascular events by 20% in 17,604 adults who had overweight or obesity and existing heart or artery disease but not diabetes: a heart attack, stroke or cardiovascular death happened to 6.5% of the semaglutide group and 8.0% of the placebo group over an average of 39.8 months [1]. That is the result the FDA acted on in March 2024, and it belongs to the people SELECT enrolled — adults with established cardiovascular disease, taking semaglutide for their hearts on top of good standard care.

Why this study matters

Obesity treatments have long been judged by the number on the scale. The SELECT authors point out that earlier lifestyle and drug trials for obesity had failed to show fewer heart attacks or strokes, and that nobody knew whether a GLP-1 drug could lower cardiovascular risk in people without diabetes [1]. SELECT was built to answer that.

It changed the label. In March 2024 the FDA approved semaglutide (as Wegovy) to reduce the risk of cardiovascular death, heart attack and stroke in adults with cardiovascular disease and either obesity or overweight [2]. Note the wording: with cardiovascular disease. That's exactly who SELECT enrolled.

What they did

To get in, you needed to be 45 or older, have a BMI of 27 or more, and already have had a heart attack, a stroke or symptomatic peripheral artery disease. Anyone with diabetes, or an HbA1c of 6.5% or above, was excluded [1].

This was a sick, well-treated group. Average age was 61.6, 72.3% were men, and mean BMI was 33.3. More than three-quarters had had a heart attack and nearly a quarter had chronic heart failure. Two-thirds (66.4%) had prediabetes. Most were already on the standard drugs: 90.1% on lipid-lowering medication and 86.2% on antiplatelet drugs [1].

Unlike the weight-loss trials, there was no structured diet programme. Semaglutide was simply added to whatever care people were already getting.

What they found

The primary result was clearly positive (p below 0.001) [1]. In absolute terms the gap was 1.5 percentage points. By our own rough arithmetic, that means around 67 people would need to be treated for a little over three years to prevent one of these events; the paper doesn't report this figure and it's a crude estimate that ignores timing.

The secondary outcomes need careful reading. The trial tested them in a fixed order, and the first, cardiovascular death, fell short of statistical significance (p = 0.07) [1]. Under the trial's rules that stopped formal testing, so the heart failure and all-cause death results are reported as estimates rather than confirmed effects [1]. The primary outcome, the one the trial was designed to answer, is where the certainty sits.

People on semaglutide lost 9.39% of their body weight by week 104, against 0.88% on placebo [1]. That's less than the 14.9% seen in STEP 1 [3], which isn't surprising in an older, mostly male population without a lifestyle programme. The authors noted that the benefit appeared early after starting treatment [1], which has fed a debate about how much of the effect comes from weight loss itself and how much from the drug's other actions.

Side effects

Serious adverse events were actually less common with semaglutide: 33.4% versus 36.4% [1]. But twice as many people stopped the drug because of side effects, 16.6% against 8.2%, mostly for gut symptoms (10.0% vs 2.0%). Gallbladder problems were slightly more common (2.8% vs 2.3%) [1]. Investigator-reported pancreatitis was the same in both arms, at 0.3% [1].

One design point matters here. SELECT used targeted safety collection: it systematically recorded only serious adverse events, events that led to stopping treatment, and pre-specified events of interest [1]. Milder, common side effects weren't all captured, so this trial isn't the place to judge how often people feel sick on the drug.

Caveats

• Only people with existing cardiovascular disease. The authors call this an important limitation: the trial didn't study people without prior heart or artery disease [1].
• Few women. Only 27.7% of participants were women, and 3.8% were Black [1].
• Many stopped treatment. Permanent discontinuation reached 26.7% on semaglutide and 23.6% on placebo [1]. The analysis still counted everyone as randomised, which, if anything, dilutes the effect.
• Follow-up was near-complete. 96.9% completed the trial and vital status was known for 99.4% [1].
• Sponsor. Novo Nordisk funded the trial and designed the protocol with an academic steering committee [1].

What it means in practice

For people like those in the trial, adults with overweight or obesity and established cardiovascular disease but no diabetes, SELECT is solid evidence that semaglutide 2.4 mg reduces heart attacks, strokes and cardiovascular deaths as a combined outcome. The effect is moderate in absolute terms and came on top of good standard care. That population — adults with existing cardiovascular disease — is the one the evidence and the label cover.

What is still open

• Whether the benefit applies to people with obesity but no existing cardiovascular disease.
• How much of the effect depends on weight loss itself versus other effects of the drug.
• Whether semaglutide reduces cardiovascular or all-cause death on its own; the trial's hierarchy stopped before those could be formally confirmed.
• Whether women benefit to the same degree, given how few took part.

What this means

Our position is that SELECT makes semaglutide 2.4 mg a heart drug for the people it studied, not a weight-loss drug that happens to have a cardiovascular footnote. For adults with overweight or obesity who already have heart or artery disease and no diabetes, we would treat it as part of standard cardiovascular care, alongside the lipid-lowering and antiplatelet drugs that 90.1% and 86.2% of participants were already taking [1]. That is a claim we are willing to own, and readers are free to disagree.

We also hold the boundaries as firmly as the result. SELECT enrolled people who had already had a heart attack, a stroke or symptomatic peripheral artery disease, and the FDA label it produced says the same [2]. That is the population the evidence covers, and we would not stretch it over everyone taking semaglutide for weight loss.

Where we read the trial more warmly than the fine print is on deaths. Cardiovascular death, the heart failure composite and all-cause death all pointed the same way, and it is the fixed testing order, not a contradictory result, that leaves them reported as estimates rather than confirmed effects [1]. We think that signal is strong enough to inform how these drugs are prescribed and discussed with patients, and we say so.

And we think the framing around weight loss misses the point. Participants lost less weight than in the dedicated weight-loss trials, 9.39% against 14.9% in STEP 1 [3], and still had fewer heart attacks and strokes. A patient's reason for taking semaglutide may be the scale, but what SELECT measured — and what we would tell them it buys — is a lower chance of a heart attack, a stroke or cardiovascular death.

Educational content only — not medical advice. Many peptides discussed on HPR are not approved for human use. Talk to a qualified clinician before making any decision about your health.