SUSTAIN-6: Semaglutide Cuts Major Cardiovascular Events by 26 Percent in High-Risk Type 2 Diabetes
In the SUSTAIN-6 cardiovascular outcomes trial, once-weekly semaglutide reduced the composite of cardiovascular death, nonfatal heart attack, and nonfatal stroke by 26 percent in people with type 2 diabetes at high cardiovascular risk — with an important caveat on diabetic retinopathy.
For most of the modern era of diabetes treatment, proving that a glucose-lowering drug was safe for the heart was optional. That changed after the rosiglitazone controversy and the FDA’s 2008 guidance, which required new diabetes therapies to demonstrate cardiovascular safety in dedicated outcomes trials before or soon after approval. SUSTAIN-6, published in the New England Journal of Medicine in 2016 and led by Steven Marso, was semaglutide’s answer to that requirement — and it ended up showing considerably more than safety.
What the researchers did
The trial enrolled 3,297 adults with type 2 diabetes at high cardiovascular risk: those aged 50 or older with established cardiovascular or chronic kidney disease, or aged 60 or older with at least one cardiovascular risk factor. A clear majority — about 83 percent — already had established cardiovascular disease at entry. Participants were randomized to semaglutide 0.5 mg once weekly, semaglutide 1.0 mg once weekly, or matching placebo, all on top of standard care, with a planned treatment duration of 104 weeks and a median follow-up of just over two years.
The primary endpoint was the classic three-point MACE composite: time to first occurrence of cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke. The trial was formally powered for noninferiority — ruling out cardiovascular harm — with superiority testing as the secondary question.
What they found
Semaglutide cleared the noninferiority bar and went well past it. The primary composite occurred in 6.6 percent of participants on semaglutide versus 8.9 percent on placebo, a hazard ratio of 0.74 (95 percent confidence interval 0.58 to 0.95) — significant both for noninferiority (P less than 0.001) and for superiority (P = 0.02). There was no clear dose gradient between the 0.5 mg and 1.0 mg arms.
| Endpoint | Semaglutide | Placebo | Hazard ratio (95% CI) |
|---|---|---|---|
| MACE composite | 6.6% | 8.9% | 0.74 (0.58–0.95) |
| Nonfatal myocardial infarction | 2.9% | 3.9% | 0.74 (0.51–1.08) |
| Nonfatal stroke | 1.6% | 2.7% | 0.61 (0.38–0.99) |
| Cardiovascular death | 2.7% | 2.8% | 0.98 (0.65–1.48) |
The composite was driven primarily by a 39 percent relative reduction in nonfatal stroke. Nonfatal myocardial infarction also trended lower but did not reach significance on its own, and cardiovascular death was essentially identical between groups. The mechanism behind this pattern — stroke protection emerging within about two years, without a mortality signal — remains debated; it does not map neatly onto weight loss or glucose lowering alone.
The trial’s most important safety finding concerned the eye. Diabetic retinopathy complications — defined as the need for photocoagulation, intravitreal agents, vitreous hemorrhage, or diabetes-related blindness — occurred in 3.0 percent of participants on semaglutide versus 1.8 percent on placebo (hazard ratio 1.76; 95 percent CI 1.11 to 2.78). Crucially, the excess clustered in participants who already had retinopathy at baseline, often alongside insulin use, and is thought to reflect the well-known early worsening that can accompany rapid improvements in glycemic control — semaglutide lowered HbA1c by roughly 0.7 to 1.0 percentage points more than placebo — rather than direct retinal toxicity. Even so, the signal prompted a dedicated retinopathy outcomes trial (FOCUS), and guidelines now recommend retinal screening before rapid intensification of glycemic control in patients with existing retinopathy.
Weight effects, while not the trial’s aim, were real: participants lost an average of 2.9 kilograms on 0.5 mg and 4.3 kilograms on 1.0 mg versus placebo. Gastrointestinal adverse events — nausea, vomiting, diarrhea — were more frequent with semaglutide and were the most common reason for stopping treatment.
The investigators concluded that in patients with type 2 diabetes at high cardiovascular risk, the rate of cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke was significantly lower with semaglutide than with placebo, while rates of retinopathy complications were significantly higher.
Marso et al., N Engl J Med 2016 (SUSTAIN-6)
Limitations
SUSTAIN-6 was a safety trial that grew into an efficacy finding, not a dedicated prevention study: with a median follow-up of about two years, it is short by outcomes-trial standards, and it could not assess durability of benefit. It was not designed or powered to analyze weight loss, and the doses used (0.5 and 1.0 mg) are the diabetes doses — nothing here speaks to the 2.4 mg obesity dose, whose cardiovascular effects in people without diabetes were later tested in SELECT. The stroke-dominated benefit, the retinopathy signal, and the absence of a mortality effect all called for confirmation in larger trials, some of which have since arrived. Treatment decisions for people with type 2 diabetes — including the choice of glucose-lowering therapy — belong with a treating physician.
SUSTAIN-6 nonetheless changed the conversation. Together with the LEADER trial of liraglutide, it helped establish that GLP-1 receptor agonists are not merely safe for the cardiovascular system but, in high-risk patients, actively protective — the foundation on which the obesity-era trials STEP and SELECT would later build.