SELECT Trial: Semaglutide Cut Major Cardiovascular Events by 20% in Adults Without Diabetes

What SELECT actually tested

SELECT was a multicenter, double-blind, randomized, placebo-controlled, event-driven superiority trial [1]. It enrolled 17,604 people and assigned 8,803 to semaglutide 2.4 mg once weekly and 8,801 to placebo in a 1:1 ratio [1]. Enrollment ran from October 31, 2018 to March 31, 2021 across 41 countries [3], and the primary results were published in the New England Journal of Medicine in 2023 [1].

Eligibility defined the question. Participants had to be 45 or older, have preexisting cardiovascular disease, and have a BMI of 27 or higher, with no history of diabetes [1]. Mean age was 61.6 years [3].

The primary endpoint was the time to the first occurrence of cardiovascular death, nonfatal myocardial infarction or nonfatal stroke [1]. Dosing was escalated over 16 weeks to a target of 2.4 mg weekly [3], matching the Wegovy label schedule: 0.25 mg for weeks 1 to 4, 0.5 mg for weeks 5 to 8, 1 mg for weeks 9 to 12, 1.7 mg for weeks 13 to 16, then maintenance from week 17 [2].

The 20% result, number by number

A primary endpoint event occurred in 569 of 8,803 people on semaglutide (6.5%) and 701 of 8,801 people on placebo (8.0%): hazard ratio 0.80, 95% CI 0.72 to 0.90, P<0.001 [1]. Mean duration of exposure was 34.2 months and mean duration of follow-up was 39.8 months [1].

The absolute difference was 1.5 percentage points, or 1.5 fewer events per 100 people treated over that follow-up. The 20% relative figure and the 1.5-point absolute figure describe the same result.

Which parts of the composite moved

A composite endpoint can hide as much as it shows. The Wegovy label reports cardiovascular death in 2.5% versus 3.0% (HR 0.85; 95% CI 0.71 to 1.01), heart attack in 2.8% versus 3.8% (HR 0.72), and all-cause death in 4.3% versus 5.2% (HR 0.81) [2]. The same label gives a stroke hazard ratio of 0.89 (95% CI 0.72 to 1.11) and notes that superiority for cardiovascular death was not confirmed in the prespecified testing hierarchy [2]. The significant primary result was for cardiovascular death, nonfatal heart attack or nonfatal stroke combined [1]. A separate analysis of the same trial reported all-cause death HR 0.81 (95% CI 0.71 to 0.93), cardiovascular death HR 0.85 (0.71 to 1.01) and non-cardiovascular death HR 0.77 (0.62 to 0.95) [4].

Why does a GLP-1 drug help people who do not have diabetes?

Semaglutide is a GLP-1 receptor agonist: it binds and activates the receptor for glucagon-like peptide-1, a natural hormone that helps regulate appetite and calorie intake and, in a glucose-dependent way, insulin release [2]. Participants lost weight, with mean body weight down 9.4% on semaglutide versus 0.9% on placebo at week 104 [2].

Beyond that, the mechanism is open. The Wegovy label states that the exact mechanism of cardiovascular risk reduction is not established [2]. SELECT was not designed to separate weight loss from any other effect of semaglutide.

Who the label covers

Both Wegovy injection and tablets have a cardiovascular risk-reduction indication: to reduce the risk of major adverse cardiovascular events, defined as cardiovascular death, nonfatal myocardial infarction or nonfatal stroke, in adults with established cardiovascular disease and either obesity or overweight [2]. Drugs@FDA records a Wegovy supplement for a new indication dated March 8, 2024 [5]. For this indication, the injection maintenance dose is 2.4 mg weekly (recommended) or 1.7 mg weekly; the tablet maintenance dose is 25 mg daily [2].

The label does not recommend Wegovy with another semaglutide-containing product or with any other GLP-1 receptor agonist [2].

Safety in SELECT and on the label

In the trial, 16.6% of people on semaglutide and 8.2% on placebo stopped treatment because of an adverse event [1]. The Wegovy label notes that in its cardiovascular outcomes trial, safety data collection was limited to serious adverse events including death, events leading to discontinuation, and events of special interest [2]. SELECT therefore did not generate a full side-effect table the way the weight-management trials did.

The label's warnings for semaglutide 2.4 mg include a boxed warning for thyroid C-cell tumors seen in rodents, with human relevance unknown, and contraindication in people with a personal or family history of medullary thyroid carcinoma or MEN 2 [2]; acute pancreatitis, with the label directing that the drug be discontinued if pancreatitis is suspected [2]; acute gallbladder disease [2]; severe gastrointestinal reactions, with the drug not recommended in severe gastroparesis [2]; acute kidney injury from volume depletion [2]; hypersensitivity reactions including anaphylaxis and angioedema [2]; heart rate increase [2]; and pulmonary aspiration during general anesthesia or deep sedation [2].

The rates people usually ask about come from the weight-management trials, not SELECT. At 2.4 mg weekly, nausea occurred in 44% versus 16% on placebo, diarrhea in 30% versus 16%, vomiting in 24% versus 6% and constipation in 24% versus 11% [2].

SELECT beside the earlier GLP-1 heart trials

Before SELECT, the cardiovascular outcome trials in this class enrolled people with type 2 diabetes.

• SUSTAIN-6: 3,297 adults with type 2 diabetes, semaglutide 0.5 or 1 mg, 104 weeks; the composite occurred in 6.6% versus 8.9% (HR 0.74; 95% CI 0.58 to 0.95). The trial was designed to test noninferiority [6].
• LEADER: 9,340 adults with type 2 diabetes and high cardiovascular risk, liraglutide (Victoza, 1.8 mg target dose) versus placebo; the primary outcome occurred in 13.0% versus 14.9% [7].
• SOUL: 9,650 adults with type 2 diabetes and atherosclerotic cardiovascular disease, chronic kidney disease or both, oral semaglutide up to 14 mg daily; 12.0% versus 13.8% (HR 0.86; 95% CI 0.77 to 0.96) [8].
• Meta-analysis: 11 randomized trials with 85,373 participants in all, 67,769 of them with type 2 diabetes, found a 13% reduction in major adverse cardiovascular events with GLP-1 receptor agonists in the participants with type 2 diabetes (HR 0.87; 95% CI 0.81 to 0.93); SELECT was added to the analysis post hoc [9].

Observational data point the same direction. A target trial emulation using US claims data compared 7,422 matched adults with atherosclerotic cardiovascular disease, a BMI of 27 or higher and no diabetes who started any GLP-1 receptor agonist with 7,422 who did not: all-cause mortality was lower (HR 0.68; 95% CI 0.53 to 0.88) and there was no significant association with stroke [10]. The authors note that observational data cannot establish cause and effect [10].

What SELECT does not answer

• Everyone enrolled already had cardiovascular disease, so the trial does not describe people without it [1].
• It compared one dose with placebo, so it cannot rank 2.4 mg against 1 mg, against the oral tablets, or against other drugs [1].