The 10 Cardiovascular Outcome Trials Behind the GLP-1 Class (2026)

The one that changed the category. 17,604 adults with established heart disease and excess weight but no diabetes, followed a mean 39.8 months. Major cardiovascular events hit 6.5% on semaglutide against 8.0% on placebo — a hazard ratio of 0.80, with cardiovascular death down 15%. Nobody had previously shown that treating obesity prevents hard cardiac events in people without diabetes. It produced the March 2024 cardiovascular indication, and it is why coverage arguments for this drug no longer have to be about weight at all.

PMID 37952131 ↗NCT03574597 ↗DOI 10.1056/NEJMoa2307563 ↗

The trial that started it. 9,340 high-risk patients with type 2 diabetes on liraglutide or placebo over a median 3.8 years, with major events at 13.0% against 14.9% — a hazard ratio of 0.87, and superiority rather than the mere safety the trial was obliged to demonstrate. Cardiovascular death fell 22% and death from any cause 15%. Before LEADER these were glucose drugs; after it they were cardiovascular ones.

PMID 27295427 ↗NCT01179048 ↗DOI 10.1056/NEJMoa1603827 ↗

Formally a safety study — 3,297 high-risk patients over 104 weeks, powered only to show semaglutide did no harm. It found a 26% reduction in major events instead (6.6% against 8.9%, hazard ratio 0.74), with nonfatal stroke down 39%. ⚠ Worth reading with the caveat that a trial designed for noninferiority which lands on superiority is a weaker basis for the claim than one built to test it. The finding held up in everything that followed.

PMID 27633186 ↗NCT01720446 ↗DOI 10.1056/NEJMoa1607141 ↗

The longest of them, at a median 5.4 years, and the most useful for anyone who has not already had a cardiac event: 69% of its 9,901 participants had no prior cardiovascular event. Major events came in at 12.0% against 13.4%, a hazard ratio of 0.88, and the benefit held in the primary-prevention group as much as the secondary one. That distinction is what earned Trulicity a 2020 indication covering patients without established disease.

PMID 31189511 ↗NCT01394952 ↗DOI 10.1016/S0140-6736(19)31149-3 ↗

The first trial built specifically around kidneys. 3,533 people whose diabetes had already damaged their kidneys, halted ahead of schedule once the benefit was clear: the kidney endpoint fell 24%, major cardiac events 18%, and death from any cause 20%. It produced the January 2025 kidney indication for Ozempic — the first label in this class granted for something other than glucose or weight.

PMID 38785209 ↗NCT03819153 ↗DOI 10.1056/NEJMoa2403347 ↗

9,463 patients with diabetes and established heart disease, over a median 1.6 years, with major events at 7% against 9% — a 22% reduction, and a robust one. The postscript is the interesting part: GSK had already pulled albiglutide from the market for commercial reasons before the result was published. The trial survives as evidence about the class rather than about a drug anyone can be prescribed.

PMID 30291013 ↗NCT02465515 ↗DOI 10.1016/S0140-6736(18)32261-X ↗

529 patients carrying obesity alongside heart failure of the preserved-ejection-fraction type, over 52 weeks — a condition with almost nothing to offer it therapeutically. Semaglutide improved symptom scores by 7.8 points more than placebo and cut weight by 10.7 percentage points more, with six-minute walk distance and inflammatory markers improving alongside. It reframed obesity-related HFpEF as something treatable rather than merely comorbid.

PMID 37622681 ↗NCT04788511 ↗DOI 10.1056/NEJMoa2306963 ↗

The safety trial for the oral formulation: 3,183 high-risk patients over a median 15.9 months, with major events at 3.8% against 4.8%. That cleared noninferiority but did not reach superiority, and the confidence interval crossed 1. Cardiovascular death fell 51%, which reads dramatically but rests on small numbers over a short follow-up. It did its job — clearing the regulatory hurdle for the first oral GLP-1 in 2019 — and should not be quoted as proof of benefit.

PMID 31185157 ↗NCT02692716 ↗DOI 10.1056/NEJMoa1901118 ↗

4,076 patients with diabetes plus cardiovascular or kidney disease, over a median 1.8 years, with major events at 7.0% against 9.2% — a 27% reduction — and a kidney composite down 32%. Its most useful feature is often overlooked: 15% of participants were on an SGLT2 inhibitor already, and the benefit persisted on top of it, which is direct evidence the two classes add rather than overlap. Sanofi discontinued the drug before it reached market.

PMID 34215025 ↗NCT03496298 ↗DOI 10.1056/NEJMoa2108269 ↗

The first cardiovascular outcomes trial in the class, and it found nothing: 6,068 patients after an acute coronary syndrome, with events at 13.4% against 13.2% — a hazard ratio of 1.02. ★ It belongs on this list precisely because it is negative. It establishes that cardiovascular benefit is not automatic for anything labeled a GLP-1, and that the short-acting agents behave differently from the long-acting ones that followed. Any claim about the class should have to explain ELIXA.

PMID 26630143 ↗NCT01147250 ↗DOI 10.1056/NEJMoa1509225 ↗