The SELECT trial (Lincoff 2023, NEJM, N=17,604, median follow-up 33.5 months) was a watershed moment in cardiovascular medicine: semaglutide 2.4mg reduced major adverse cardiovascular events (MACE — cardiovascular death, non-fatal myocardial infarction, non-fatal stroke) by 20% in adults with overweight or obesity and established cardiovascular disease, without diabetes. This was the first cardiovascular outcomes trial of a weight-loss drug in a non-diabetic population to demonstrate a significant MACE reduction, and it fundamentally repositioned GLP-1 receptor agonists from metabolic drugs to cardiovascular medicines.
Critically, the SELECT trial's cardiovascular benefit appeared to emerge before substantial weight loss occurred — suggesting that weight reduction alone does not fully explain the cardioprotection. GLP-1 receptors are expressed in cardiomyocytes, vascular endothelial cells, and macrophages (the foam cells of atherosclerotic plaques), providing multiple potential direct cardioprotective mechanisms independent of the weight or glucose effects that first brought these drugs to prominence.
| Trial | Drug | N | Population | MACE Result | Key Finding |
|---|---|---|---|---|---|
| SELECT 2023 | Semaglutide 2.4mg | 17,604 | Obese, no T2DM, established CVD | -20% (HR 0.80, p<0.001) | First MACE reduction in non-diabetic population; repositions semaglutide as CV drug |
| SUSTAIN-6 2016 | Semaglutide 0.5+1mg | 3,297 | T2DM, high CV risk | -26% (HR 0.74) | Strongest MACE effect of any GLP-1 trial; mainly driven by stroke reduction (-39%) |
| LEADER 2016 | Liraglutide 1.8mg | 9,340 | T2DM, high CV risk | -13% (HR 0.87) | First positive GLP-1 CV trial; -22% CV death; FDA approved for CV risk reduction |
| REWIND 2019 | Dulaglutide 1.5mg | 9,901 | T2DM, mixed CV risk | -12% (HR 0.88) | Only GLP-1 CV trial to show benefit in primary prevention population (no prior CVD) |
| EXSCEL 2017 | Exenatide weekly | 14,752 | T2DM, mixed CV risk | Neutral (HR 0.91, NS) | Trend toward benefit not significant; exenatide may be less cardioprotective than semaglutide/liraglutide |
Highest evidence-based indication (per SELECT): Adults with BMI ≥27, established cardiovascular disease (prior MI, stroke, or peripheral artery disease), without type 2 diabetes; the -20% MACE reduction represents a larger absolute benefit in this high-risk population than most other available cardiovascular medications; the SELECT NNT of ~67 over 33.5 months is clinically competitive with statins in secondary prevention (statin NNT for MACE: ~50–100 over 5 years); this is the population where cost-benefit calculation is clearest for semaglutide as a cardiovascular medicine.
Type 2 diabetes with CVD: Multiple trials (LEADER, SUSTAIN-6, REWIND) confirm MACE reduction; ADA/ACC guidelines now recommend GLP-1 RA as preferred add-on to metformin in T2DM patients with established CVD or high CV risk, independent of HbA1c control; the cardiovascular benefit is considered class-effect for the longer-acting GLP-1 RAs (semaglutide, liraglutide, dulaglutide) but NOT clearly established for shorter-acting agents (exenatide twice daily, lixisenatide — both showed neutral CV outcomes).
Heart failure with preserved ejection fraction (HFpEF): STEP-HFpEF represents the first drug class to show meaningful symptom improvement in HFpEF; for obese HFpEF patients, semaglutide is now a reasonable therapeutic consideration in addition to standard HF care; ongoing FLOW trial (semaglutide for chronic kidney disease) — results expected 2024–2025; GLP-1 applications continue to expand beyond the original metabolic indication.
The unresolved question — weight loss vs direct effect: Post-hoc analyses of SELECT show that the MACE reduction was similar in individuals who lost more vs less weight — suggesting a weight-independent component to cardioprotection; but randomized trials comparing matched weight loss (by diet vs semaglutide) don't yet exist in CV populations; the mechanistic data (direct GLP-1 receptor expression in heart and vasculature, early MACE benefit before weight loss, CRP reduction independent of weight) strongly support a direct cardioprotective effect beyond weight, but the definitive magnitude attribution is still an active research question.