What was the exact hazard ratio for MACE in SUSTAIN-6?

The primary endpoint hazard ratio was 0.74 (95% CI 0.58-0.95, p = 0.02), representing a 26% relative reduction in the composite of cardiovascular death, nonfatal MI, and nonfatal stroke with semaglutide versus placebo.

Was the MACE reduction driven by one specific component?

Yes, nonfatal stroke drove most of the composite benefit (HR 0.61 to 95% CI 0.38-0.99). Nonfatal MI trended lower (HR 0.74, CI crossing 1.0), and cardiovascular death was neutral (HR 0.98).

Did the 0.5 mg and 1.0 mg doses produce different cardiovascular results?

Both doses produced similar MACE event rates (6.7% and 6.4%, respectively) despite dose-dependent differences in HbA1c reduction, suggesting the cardiovascular benefit was not purely glycemia-mediated.

When did the Kaplan-Meier curves begin to separate?

Visible separation in the primary composite appeared between weeks 40 and 60, consistent with an anti-atherosclerotic mechanism rather than a hemodynamic one.

What happened with retinopathy complications in SUSTAIN-6?

Retinopathy complications were significantly higher with semaglutide (HR 1.76 to 95% CI 1.11-2.78), likely related to rapid glycemic improvement in patients with pre-existing retinopathy. The effect mirrors the well-known "early worsening" phenomenon seen with insulin intensification.

Did SUSTAIN-6 show a kidney benefit?

Yes. New or worsening nephropathy was reduced by 36% (HR 0.64 to 95% CI 0.46-0.88, p = 0.005). This was driven by less progression to macroalbuminuria rather than hard renal endpoints like dialysis.

How does SUSTAIN-6 compare with the SELECT trial for semaglutide?

SELECT studied a higher dose (2.4 mg) in obese patients without diabetes over a longer period and confirmed a 20% MACE reduction (HR 0.80). SUSTAIN-6 used lower doses (0.5-1.0 mg) in a diabetic population and found a 26% reduction. Both trials support cardiovascular benefit, though the populations and dose regimens differ.

Was SUSTAIN-6 designed to prove cardiovascular benefit?

No. It was powered for noninferiority (upper CI bound <1.8) to satisfy the FDA's post-2008 CVOT requirement. The finding of superiority was a stronger result than the trial was designed to detect.

Who was eligible for SUSTAIN-6?

Adults with type 2 diabetes (HbA1c ≥7.0%) who were ≥50 with established cardiovascular disease or ≥60 with cardiovascular risk factors. Approximately 83% had established CVD at enrollment.

Does the Ozempic FDA label include a cardiovascular indication based on SUSTAIN-6?

Yes. The FDA granted Ozempic a labeled indication for reducing MACE risk in adults with type 2 diabetes and established cardiovascular disease, based primarily on SUSTAIN-6 data.

SUSTAIN-6 Results in Detail: Numbers, Subgroups, and Time Course

Q: Who was eligible for SUSTAIN-6?

Adults with type 2 diabetes (HbA1c ≥7.0%) who were ≥50 with established cardiovascular disease or ≥60 with cardiovascular risk factors. Approximately 83% had established CVD at enrollment.

Q: Does the Ozempic FDA label include a cardiovascular indication based on SUSTAIN-6?

Yes. The FDA granted Ozempic a labeled indication for reducing MACE risk in adults with type 2 diabetes and established cardiovascular disease, based primarily on SUSTAIN-6 data.

By HealthRX Medical Team

Published May 25, 2026Updated May 25, 2026Last reviewed May 25, 2026

At a glance

| Parameter | Detail | |---|---| | N | 3,297 (1,648 semaglutide; 1,649 placebo) | | Intervention | Semaglutide 0.5 mg or 1.0 mg subcutaneous once weekly | | Comparator | Placebo (volume-matched, once weekly) | | Duration | Median 2.1 years on treatment; minimum 104-week exposure | | Primary endpoint | First occurrence of 3-point MACE (CV death, nonfatal MI, nonfatal stroke) | | Key result | HR 0.74 to 95% CI 0.58-0.95, p = 0.02 for noninferiority and superiority | | Registration | NCT01720446 |

Trial design and what it actually tested

SUSTAIN-6 was a preapproval cardiovascular outcomes trial (CVOT) mandated by the FDA's 2008 guidance for new antidiabetic agents. Novo Nordisk designed it primarily to satisfy the noninferiority requirement (upper 95% CI bound for MACE <1.8), meaning the trial was powered to rule out unacceptable cardiovascular harm, not to prove benefit. That the study achieved superiority with a relatively modest sample size was unexpected.

Patients were randomized 1:1:1:1 across four arms: semaglutide 0.5 mg, semaglutide 1.0 mg, and their respective volume-matched placebos. For reporting, the two semaglutide doses were pooled against pooled placebo. All participants remained on standard-of-care diabetes therapy, and investigators could adjust background medications (including insulin) to meet glycemic targets. This open adjustment of background therapy is a design feature that dilutes between-group HbA1c differences and, theoretically, isolates the non-glycemic cardiovascular effects of the study drug.

Eligibility required type 2 diabetes with HbA1c ≥7.0% and either established cardiovascular disease (age ≥50) or cardiovascular risk factors (age ≥60). Roughly 83% of participants had established CVD, prior MI, prior stroke, or peripheral arterial disease at baseline, making this a secondary-prevention-dominant cohort (Marso et al., NEJM 2016).

Primary endpoint: the numbers behind the 26% reduction

The primary composite outcome (first occurrence of cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke) occurred in 108 of 1,648 patients (6.6%) in the semaglutide group versus 146 of 1,649 patients (8.9%) in the placebo group.

HealthRX MACE Component Breakdown

| Component | Semaglutide n (%) | Placebo n (%) | HR (95% CI) | |---|---|---|---| | 3-point MACE (composite) | 108 (6.6%) | 146 (8.9%) | 0.74 (0.58-0.95) | | Cardiovascular death | 44 (2.7%) | 46 (2.8%) | 0.98 (0.65-1.48) | | Nonfatal MI | 47 (2.9%) | 64 (3.9%) | 0.74 (0.51-1.08) | | Nonfatal stroke | 27 (1.6%) | 44 (2.7%) | 0.61 (0.38-0.99) |

Several points stand out from this breakdown. The composite result was driven disproportionately by nonfatal stroke reduction (HR 0.61). Nonfatal MI showed a consistent 26% point estimate reduction but did not reach statistical significance on its own (HR 0.74, upper CI crossing 1.0). Cardiovascular death was essentially neutral (HR 0.98), contributing almost nothing to the composite benefit.

This pattern matters clinically. The stroke signal was large and statistically significant. Whether this reflects a specific vascular-protective mechanism of semaglutide (anti-inflammatory effects, plaque stabilization, blood pressure reduction of 1.3-2.4 mmHg) or represents random variation in a trial not powered for individual components remains debated. The SUSTAIN-6 supplementary appendix reports that most nonfatal strokes were ischemic, not hemorrhagic, consistent with an atherothrombotic mechanism.

Dose-level results: 0.5 mg versus 1.0 mg

Although the primary analysis pooled both doses, the publication provides dose-stratified data that is often overlooked.

| Dose | MACE events / N | Event rate | |---|---|---| | Semaglutide 0.5 mg | 55 / 826 | 6.7% | | Semaglutide 1.0 mg | 53 / 822 | 6.4% | | Placebo (pooled) | 146 / 1,649 | 8.9% |

Both doses showed similar absolute event rates, suggesting the cardiovascular benefit was not strongly dose-dependent across the 0.5-1.0 mg range. HbA1c reduction, by contrast, was dose-dependent (1.1% with 0.5 mg vs. 1.4% with 1.0 mg at week 104), reinforcing the theory that the MACE reduction was not simply a glucose-lowering effect (Marso et al., 2016).

Time course of separation

The Kaplan-Meier curves for the primary composite began separating visibly between weeks 40 and 60 (roughly months 10-15). This is later than the separation seen in EMPA-REG OUTCOME for empagliflozin (which diverged within the first 3 months, driven by heart failure and CV death), but earlier than what was observed in some DPP-4 inhibitor trials that never separated at all.

The delayed separation is consistent with an anti-atherosclerotic mechanism rather than a hemodynamic one. Plaque stabilization, reduction in systemic inflammation (hsCRP decreased 36% with semaglutide in the trial), and modest blood pressure lowering all require months to translate into fewer hard thrombotic events. By contrast, the rapid EMPA-REG separation likely reflected volume and hemodynamic effects on heart failure, a different mechanism entirely.

At 104 weeks, the curves continued to diverge with no sign of plateauing. This observation became important when the FDA approved the cardiovascular indication for semaglutide: the sustained, widening separation suggested that longer treatment could yield larger absolute benefits.

Secondary and exploratory endpoints

Glycemic and metabolic parameters (week 104)

| Parameter | Semaglutide (pooled) | Placebo | Difference | |---|---|---|---| | HbA1c change from baseline | -1.1% to -1.4% | -0.4% | -0.7 to -1.0% | | Body weight change | -3.6 to -4.9 kg | -0.7 kg | -2.9 to -4.2 kg | | Systolic BP change | -1.3 to -2.4 mmHg | -0.6 mmHg | -0.7 to -1.8 mmHg |

Nephropathy

A prespecified secondary composite of new or worsening nephropathy (new macroalbuminuria, doubling of serum creatinine, need for dialysis, or renal death) favored semaglutide: HR 0.64 to 95% CI 0.46-0.88, p = 0.005. This was driven almost entirely by reduced progression to macroalbuminuria (new persistent macroalbuminuria: HR 0.54). Hard renal endpoints (creatinine doubling, dialysis) were too infrequent to analyze independently.

Retinopathy

Retinopathy complications (vitreous hemorrhage, blindness, need for photocoagulation or intravitreal agents) were more frequent with semaglutide: HR 1.76 to 95% CI 1.11-2.78, p = 0.02. This was unexpected and concerning. The investigators attributed it to rapid HbA1c reduction in patients with pre-existing retinopathy, a phenomenon well-documented with insulin intensification. Among patients without baseline retinopathy, rates were similar between groups. The FDA label for Ozempic carries a warning about this finding and recommends monitoring in patients with diabetic retinopathy.

All-cause mortality

All-cause mortality was 62 (3.8%) with semaglutide versus 60 (3.6%) with placebo (HR 1.05 to 95% CI 0.74-1.50). The trial was underpowered for mortality, and this result should be interpreted as uninformative rather than as evidence of harm.

Subgroup analyses

The supplementary appendix reports prespecified subgroup analyses for the primary composite.

| Subgroup | HR (95% CI) | Interaction p-value | |---|---|---| | Age <65 years | 0.71 (0.51-0.98) | 0.67 | | Age ≥65 years | 0.78 (0.55-1.11) | | | Male | 0.82 (0.61-1.10) | 0.28 | | Female | 0.59 (0.38-0.91) | | | Established CVD | 0.72 (0.55-0.93) | 0.50 | | Risk factors only | 0.87 (0.48-1.56) | | | HbA1c <8.5% | 0.77 (0.55-1.07) | 0.74 | | HbA1c ≥8.5% | 0.71 (0.49-1.03) | |

No significant interactions emerged, suggesting the MACE benefit was consistent across prespecified subgroups. The numerically larger effect in women (HR 0.59) is notable but the interaction test was nonsignificant (p = 0.28), and the female subgroup had fewer events, making the confidence interval wide.

Limitations the authors acknowledged

The SUSTAIN-6 investigators were transparent about several constraints. The trial was designed and powered for noninferiority, not superiority. The superiority finding, while statistically significant, emerged from a study with a relatively small event count (254 total primary events). This limits the precision of component-level analyses and subgroup findings.

The open adjustment of background diabetes medications means the between-group HbA1c difference narrowed over time (investigators titrated insulin more aggressively in placebo patients to maintain glycemic targets). While this design choice strengthens the case for non-glycemic cardiovascular effects, it complicates interpretation of glucose-mediated benefits.

The retinopathy signal was an important safety finding, but the trial was not designed to systematically assess retinal outcomes with standardized fundoscopy. Event ascertainment relied on investigator reporting, which may have missed mild cases or overcounted in patients already under ophthalmologic surveillance.

Trial duration of 2.1 years is relatively short for atherosclerotic outcomes. The SELECT trial later evaluated semaglutide 2.4 mg in an obese, non-diabetic population over a longer period and confirmed a 20% MACE reduction (HR 0.80 to 95% CI 0.72-0.90), providing further evidence that the cardiovascular benefit extends beyond glucose lowering and persists with longer follow-up.

What SUSTAIN-6 changed in practice

The 2019 ADA/EASD consensus report cited SUSTAIN-6 (alongside LEADER for liraglutide) as the basis for recommending GLP-1 receptor agonists with proven cardiovascular benefit as preferred second-line therapy in type 2 diabetes patients with established atherosclerotic CVD. This represented a shift from choosing diabetes drugs based solely on glucose lowering to a cardiovascular-outcome-driven selection algorithm.

The FDA approved Ozempic's cardiovascular risk reduction indication partly on the strength of SUSTAIN-6 data, making semaglutide one of a small group of diabetes medications with a labeled CV benefit alongside empagliflozin and liraglutide.

Frequently asked questions

›

References

Marso SP, Bain SC, Consoli A, et al. Semaglutide and cardiovascular outcomes in patients with type 2 diabetes. N Engl J Med. 2016;375(19):1834-1844. PubMed
Ozempic (semaglutide) prescribing information. Novo Nordisk. FDA label. AccessData
Lincoff AM, Brown-Frandsen K, Colhoun HM, et al. Semaglutide and cardiovascular outcomes in obesity without diabetes. N Engl J Med. 2023;389(24):2221-2232. PubMed
Davies MJ, D'Alessio DA, Fradkin J, et al. Management of hyperglycemia in type 2 diabetes, 2018. A consensus report by the ADA and EASD. Diabetes Care. 2018;41(12):2669-2701. PubMed
Zinman B, Wanner C, Lachin JM, et al. Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes (EMPA-REG OUTCOME). N Engl J Med. 2015;373(22):2117-2128. PubMed