Cardiovascular outcomes · Landmark trial
The SELECT trial: does Wegovy actually reduce heart-attack risk?
SELECT (Lincoff et al., NEJM 2023) is why the FDA expanded the Wegovy label in March 2024 to include cardiovascular risk reduction. It is the first cardiovascular outcomes trial to show benefit in adults with overweight or obesity without diabetes. This is what the trial actually measured, what it did not, and what it means for the population it enrolled.
The endpoints, side-by-side
| Endpoint | Semaglutide | Placebo | Effect |
|---|---|---|---|
| Primary: 3-point MACE (CV death + non-fatal MI + non-fatal stroke) | 6.5% | 8.0% | HR 0.80 (95% CI 0.72-0.90, p<0.001) — 20% relative reduction |
| All-cause mortality | 4.3% | 5.2% | HR 0.81 (0.71-0.93) — statistically significant |
| Cardiovascular death | 2.5% | 3.0% | HR 0.85 (0.71-1.01) — trend, not significant on its own |
| Non-fatal myocardial infarction | 2.7% | 3.7% | HR 0.72 (0.61-0.85) — the strongest individual component |
| Non-fatal stroke | 1.7% | 1.8% | HR 0.93 (0.74-1.16) — directionally consistent, not significant |
| Heart failure hospitalization (secondary) | 3.4% | 4.1% | HR 0.82 — significant reduction |
| Weight change at 104 weeks | −9.4% | −0.9% | Difference: −8.5 percentage points |
Data from Lincoff AM et al., NEJM 389:2221-2232, 2023. Full trial details in the primary source linked below.
Who SELECT actually enrolled
Understanding the SELECT population is the difference between over-extrapolating the result and applying it appropriately.
- Age: ≥45 years. Mean age 61.6.
- BMI: ≥27. Mean BMI 33.4.
- CV disease: prior myocardial infarction, prior stroke, or symptomatic peripheral artery disease. Everyone had documented atherosclerotic disease.
- Diabetes: excluded. HbA1c <6.5% was required. This is the population the trial addresses; extrapolation to patients with T2D is done via separate trials (SUSTAIN-6, LEADER for liraglutide).
- Sex: ~72% male. Reflects the higher CV disease prevalence in men in this age band. Subgroup analysis found the benefit similar in women, though the female subgroup was smaller.
- Race/ethnicity: ~84% White. Under-representation of Black, Hispanic, and Asian populations is a known limitation.
Weight loss vs weight-independent mechanism
The clinical question everyone asked after SELECT: is this benefit just because patients lost weight?
The best evidence says no — or at least not fully. Three observations from SELECT and follow-up analyses point to weight-independent mechanism:
- Timing. CV event curves separated within the first 12-24 months, before most of the weight loss had accumulated. Weight loss on semaglutide 2.4 mg follows the STEP-1 curve — most loss between weeks 20 and 44. But MACE curves diverged earlier.
- hs-CRP drop. High-sensitivity C-reactive protein, a marker of systemic inflammation, dropped ~30% more on semaglutide vs placebo. Inflammation reduction is a plausible independent CV-benefit driver, well established in the CANTOS (canakinumab) trial.
- Adjusted analysis. Secondary analyses that controlled for weight change, HbA1c change, BP change, and lipid change still found a residual benefit — meaning the drug is doing something beyond the cardiometabolic improvements it drives.
Not fully resolved. The exact split between weight-driven and weight-independent mechanism is under active study. But SELECT + subsequent analyses reject the null hypothesis that the benefit is purely from weight loss.
FDA label expansion — March 2024
In March 2024, the FDA expanded the Wegovy label to include cardiovascular risk reduction in adults with established cardiovascular disease + BMI ≥27, without diabetes — the SELECT population. This was the first time a chronic weight-management medication received a cardiovascular outcomes indication.
Practical impact: prior authorization criteria at many payers now include an obesity + CV disease pathway that is easier to clear than the primary weight-loss criteria. If you have documented CV disease and BMI ≥27, coverage is materially easier than the pure obesity indication.
Safety signals in SELECT
SELECT was also the largest single semaglutide safety database. Key findings:
- GI adverse events: ~28% of semaglutide arm reported GI events vs ~14% placebo — consistent with STEP-1. Trial dropout for GI was ~7%.
- Pancreatitis: incidence was low and did not differ significantly between arms (0.2% vs 0.2%).
- Malignancy: total cancer incidence did not differ. Notably, medullary thyroid carcinoma was not increased — see our MTC evidence review for the full read.
- NAION: not pre-specified in SELECT. The Hathaway 2024 signal came from a separate single-center cohort at Mass Eye and Ear. See our NAION FAERS surveillance report.
Frequently asked
What did the SELECT trial actually show?
The SELECT trial randomized 17,604 adults ≥45 with established cardiovascular disease + BMI ≥27 (no diabetes) to semaglutide 2.4 mg weekly or placebo, both with standard care. Over a mean 3.3 years, the primary endpoint (3-point MACE = CV death, non-fatal MI, non-fatal stroke) occurred in 6.5% of the semaglutide arm vs 8.0% of placebo — a 20% relative risk reduction (HR 0.80, 95% CI 0.72-0.90, p<0.001). This is the landmark result that led the FDA to expand the Wegovy label to include cardiovascular risk reduction in March 2024.
Does the SELECT benefit come from weight loss, or from something else?
Not fully explained by weight loss. SELECT patients on semaglutide lost roughly 9.4% of body weight vs 0.9% on placebo. But the cardiovascular benefit was measurable within the first year — before most of the weight loss had accumulated. The Lincoff group and subsequent secondary analyses argue for weight-independent mechanisms: reduced inflammation (hs-CRP dropped ~30% more on semaglutide), improved endothelial function, lipid profile improvement, and potentially direct effects on plaque biology. The exact mechanism split is not resolved, but the direction is: SELECT tells us semaglutide reduces cardiovascular events beyond what weight loss alone would predict.
Does this mean everyone with obesity should be on semaglutide?
No. SELECT enrolled a specific population: adults ≥45 with prior cardiovascular disease (prior MI, prior stroke, or symptomatic peripheral artery disease) plus BMI ≥27, and no type 2 diabetes. The FDA-approved CV indication tracks this population. Extrapolation to lower-risk primary-prevention groups is not supported by the SELECT dataset. That said, other trials (STEP-HFpEF for heart failure, FLOW for kidney disease) and observational data increasingly suggest broader benefit.
Was the SELECT benefit driven by lower blood pressure?
Partly. SELECT patients on semaglutide had systolic BP drops of roughly 3.8 mmHg vs placebo, LDL cholesterol down ~5.4 mg/dL, and HbA1c down ~0.3%. These cardiometabolic improvements together explain some of the benefit but not all. Adjusted analyses suggest weight-independent mechanisms account for a meaningful share.
What about stroke specifically?
Non-fatal stroke was one of the three components of the primary MACE endpoint. The point estimate for stroke was HR 0.93 (95% CI 0.74-1.16) — not statistically significant on its own. The 20% MACE reduction was driven primarily by cardiovascular death (HR 0.85) and non-fatal MI (HR 0.72). Stroke benefit is directionally consistent but under-powered in SELECT.
How do I know if I qualify for the SELECT-population Wegovy indication?
The FDA-approved CV indication for Wegovy applies to adults with established cardiovascular disease + overweight (BMI ≥27) or obesity, without type 2 diabetes. "Established CV disease" means a prior heart attack, prior stroke, or symptomatic peripheral artery disease. Talk to your prescriber — they will make the individual determination based on your cardiac history and BMI.
Sources
Editorial notes
- Freshness: reviewed July 17, 2026. Refresh triggered by any secondary SELECT analysis publication, FDA label change, or new head-to-head CV outcomes trial.
- Not medical advice. Individual decisions about cardiovascular risk reduction are for your prescriber, cardiologist, or preventive medicine specialist — this page describes the published evidence.
- Citation: journalists and researchers are welcome to cite — attribution to Glpverdict is the only requirement.