- SELECT enrolled 17,604 adults aged 45 or older who already had cardiovascular disease, a BMI of 27 or more, and no diabetes.1
- Over about 3.3 years of follow-up, a heart attack, stroke or cardiovascular death happened to 6.5% of people on semaglutide 2.4 mg and 8.0% on placebo, a 20% lower relative risk.1
- More people stopped semaglutide because of side effects: 16.6% vs 8.2% on placebo.1
Why this trial mattered
Earlier trials showed that semaglutide lowered the risk of heart attack and stroke in people with type 2 diabetes. For people with overweight or obesity but no diabetes, nobody knew. Weight loss improves blood pressure and other risk markers, but better numbers on a lab report do not always mean fewer heart attacks. Only a large outcomes trial can answer that.1
SELECT was that trial. It was not designed to measure weight loss. It was designed to count serious heart and blood vessel events.
Who was in it
Everyone in SELECT already had established cardiovascular disease, defined as at least one of: a previous heart attack, a previous stroke, or symptomatic narrowing of the arteries in the legs (peripheral artery disease). To join, people also had to be 45 or older, have a BMI of 27 or higher, and have no history of diabetes.1,2
That makes SELECT quite different from the big weight-loss trials, which enrolled younger people with a higher average BMI and mostly women.2 Keep that in mind when you compare results across trials.
Half the participants (8,803) were randomly assigned to weekly semaglutide 2.4 mg injections and half (8,801) to placebo. Neither they nor their doctors knew which. On average, people took the study medicine or placebo for about 34 months and were followed for about 40 months.1
The main result
The trial counted the first of three events: death from cardiovascular causes, a heart attack that was not fatal, or a stroke that was not fatal. Cardiologists call this group of events MACE, short for major adverse cardiovascular events.
One of these events happened to 569 people on semaglutide and 701 people on placebo. The hazard ratio was 0.80, which means the risk at any point during the trial was about 20% lower on semaglutide. The result was statistically clear (P<0.001).1
Relative risk vs absolute risk
"20% lower risk" is the headline number, and it is a relative one. It compares the two groups with each other. The absolute difference tells you how many people that actually affected.
In SELECT, the gap was 1.5 percentage points: 8.0% minus 6.5%. Put another way, over a little more than three years, there were about 15 fewer of these events for every 1,000 people on semaglutide than on placebo. Both ways of saying it are true. The absolute number depends heavily on how high the starting risk is, and everyone in SELECT already had heart disease.
What happened to weight
Weight was a secondary result. A later analysis of the same trial reported that weight loss on semaglutide continued for about 65 weeks and then held steady for up to four years.2
| At week 208 (4 years) | Semaglutide | Placebo |
|---|---|---|
| Change in body weight | −10.2% | −1.5% |
| Change in waist size | −7.7 cm | −1.3 cm |
At two years, 67.8% of people on semaglutide had lost at least 5% of their weight and 44.2% had lost at least 10%, compared with 21.3% and 6.9% on placebo.2
That is less than the roughly 15% average seen in the STEP 1 weight-loss trial. The authors give several reasons. SELECT participants did not join to lose weight and got no structured diet and exercise program. Doctors were allowed to slow, lower or pause the dose. And the group was older, mostly men, and had a lower starting BMI.2 If you want the weight-loss trials themselves, see what STEP 1 and SURMOUNT-1 found.
Side effects and stopping
Side effects that led people to stop the study medicine for good were about twice as common on semaglutide: 1,461 people (16.6%) vs 718 people (8.2%) on placebo.1 In the weight analysis, stopping because of side effects was more common in people who started with a lower BMI.2
Serious adverse events were reported less often on semaglutide than on placebo in every BMI group.2 For how often stomach side effects such as nausea show up in the weight-loss trials, and when, see our article on GI side effects.
What SELECT does not tell us
- Why it worked. The trial shows fewer events, not the reason. Weight loss is one possible explanation, but the trial was not built to separate it from other effects of the medicine.
- Other medicines. SELECT tested semaglutide 2.4 mg only. Its results do not automatically carry over to tirzepatide, oral semaglutide or other doses.
- What happens after stopping. Other trials show that weight tends to come back when these medicines stop. See what those trials found.
- Your own risk. Averages from a trial do not tell any one person how much they would benefit. That is a conversation for your cardiologist or prescriber, who knows your history.
SELECT was funded by Novo Nordisk, which makes semaglutide.1 That is normal for trials of this size, and the trial was double-blind and placebo-controlled, but it is worth knowing.
Sources
- Lincoff AM, Brown-Frandsen K, Colhoun HM, et al. Semaglutide and Cardiovascular Outcomes in Obesity without Diabetes. N Engl J Med. 2023;389(24):2221-2232. PubMed · doi:10.1056/NEJMoa2307563
- Ryan DH, Lingvay I, Deanfield J, et al. Long-term weight loss effects of semaglutide in obesity without diabetes in the SELECT trial. Nat Med. 2024;30(7):2049-2057. doi:10.1038/s41591-024-02996-7 · Full text (PMC)
