LEADER, SUSTAIN-6, REWIND, SELECT: A Cross-Trial Comparison of GLP-1 Receptor Agonist Cardiovascular Outcomes

What did each of the four trials find?

LEADER. 9,340 adults with type 2 diabetes and high cardiovascular risk were randomized to liraglutide or placebo and followed for a median of 3.8 years. The primary composite occurred in 13.0% of the liraglutide group and 14.9% of the placebo group, hazard ratio 0.87 (95% CI 0.78 to 0.97; P<0.001 for noninferiority, P=0.01 for superiority) [1].

SUSTAIN-6. 3,297 adults with type 2 diabetes on standard care received semaglutide 0.5 mg or 1.0 mg weekly or placebo for 104 weeks; 83.0% had established cardiovascular disease, chronic kidney disease or both. The primary composite occurred in 6.6% versus 8.9%, hazard ratio 0.74 (95% CI 0.58 to 0.95; P<0.001 for noninferiority) [2]. The stated hypothesis was noninferiority, with a margin of 1.8 on the upper confidence bound [2].

REWIND. 9,901 adults aged 50 or older with type 2 diabetes who had either a previous cardiovascular event or cardiovascular risk factors were randomized to dulaglutide 1.5 mg weekly or placebo. Over a median follow-up of 5.4 years the primary composite occurred in 12.0% versus 13.4%, hazard ratio 0.88 (95% CI 0.79 to 0.99; p=0.026) [3]. Median HbA1c at entry was 7.2% [3].

SELECT. 17,604 adults aged 45 or older with preexisting cardiovascular disease and a BMI of 27 or higher, and no history of diabetes, received semaglutide 2.4 mg weekly or placebo in an event-driven superiority trial. Over a mean follow-up of 39.8 months the primary composite occurred in 6.5% versus 8.0%, hazard ratio 0.80 (95% CI 0.72 to 0.90; P<0.001) [4].

Who was enrolled, and why does it matter?

LEADER and SUSTAIN-6 evaluated cardiovascular outcomes in adults with type 2 diabetes, and REWIND tested whether dulaglutide reduced major cardiovascular events in adults with type 2 diabetes who had either a previous cardiovascular event or cardiovascular risk factors [1][2][3]. SELECT tested semaglutide in adults with established cardiovascular disease and excess weight but no history of diabetes [4].

The recruitment rules differed in detail. LEADER enrolled adults with type 2 diabetes and high cardiovascular risk [1]. SUSTAIN-6 enrolled adults with type 2 diabetes on a standard-care regimen, 83.0% of whom had established cardiovascular disease, chronic kidney disease or both [2]. REWIND deliberately included people with risk factors but no previous cardiovascular event alongside those with one. Median HbA1c at entry was 7.2% [3]. SELECT required preexisting cardiovascular disease plus a BMI of 27 or higher and excluded anyone with a diabetes history [4].

How do the four results compare?

All four hazard ratios fall between 0.74 and 0.88, and every 95% confidence interval excludes 1.0 [1][2][3][4]. The intervals overlap, so none of these results shows that one of the three drugs is stronger than another. No trial in this set randomized patients to two of these drugs head to head.

The absolute differences are smaller than the hazard ratios suggest. LEADER saw 1.9 fewer events per 100 patients over a median 3.8 years [1]; SUSTAIN-6, 2.3 fewer per 100 over 104 weeks [2]; REWIND, 1.4 fewer per 100 over a median 5.4 years [3]; SELECT, 1.5 fewer per 100 over a mean 39.8 months [4]. Placebo event rates ranged from 8.0% in SELECT to 14.9% in LEADER, but the trials enrolled different populations and followed them for different lengths of time. Their absolute event differences are not directly comparable [4][1].

Did any trial show a mortality benefit?

In LEADER, cardiovascular death was 4.7% with liraglutide versus 6.0% with placebo, and death from any cause was 8.2% versus 9.6%. These were exploratory outcomes without adjustment for multiple comparisons [1].

REWIND reported no difference in all-cause mortality: 10.8% versus 12.0%, hazard ratio 0.90 [3].

Death from any cause was not the primary endpoint of any of the four trials; in each, the primary endpoint was the cardiovascular composite [1][2][3][4]. In SELECT the prespecified primary endpoint was the composite of cardiovascular death, nonfatal myocardial infarction or nonfatal stroke [4].

Stroke and heart attack, looked at separately

SUSTAIN-6 gives the clearest component data. Nonfatal myocardial infarction occurred in 2.9% of the semaglutide group versus 3.9% of the placebo group (hazard ratio 0.74, 95% CI 0.51 to 1.08; P=0.12), and nonfatal stroke in 1.6% versus 2.7% (hazard ratio 0.61) [2]. In LEADER, nonfatal myocardial infarction, nonfatal stroke and hospitalization for heart failure were each nonsignificantly lower, while the composite itself was significantly reduced [1].

Component results in a single trial carry less weight than the composite they belong to, and the four trials were not designed to compare a stroke effect against a heart attack effect.

What the four trials establish together

Three different molecules — liraglutide, semaglutide and dulaglutide — lowered the same three-part cardiovascular composite in four separately recruited high-risk populations [1][2][3][4]. The Wegovy, Victoza, Trulicity and Ozempic labels each describe cardiovascular risk reduction for their specified populations: the Wegovy label lists reduction of major adverse cardiovascular events in adults with established cardiovascular disease and obesity or overweight [5]; the Victoza label lists cardiovascular risk reduction in adults with type 2 diabetes and established cardiovascular disease [6]; the Trulicity label lists cardiovascular risk reduction in adults with type 2 diabetes who have established cardiovascular disease or multiple cardiovascular risk factors [7]; and the Ozempic label lists cardiovascular risk reduction in type 2 diabetes with established cardiovascular disease [8].

SELECT adds the part that diabetes-only trials could not answer: the composite fell in people who never had diabetes, from 8.0% to 6.5% [4].

What they do not settle

Dose is unresolved. SELECT used semaglutide 2.4 mg in adults without diabetes and SUSTAIN-6 used 0.5 mg or 1.0 mg in adults with diabetes, so the two results cannot separate the effect of a higher dose from the effect of a different population [4][2].

Prevention in people without existing cardiovascular disease is also open. REWIND included adults with risk factors but no previous event, and its published primary result is for the whole randomized cohort, not for that subgroup alone [3].

Finally, these four trials tested three GLP-1 receptor agonists and do not stand in for every drug in or near the class.