What SUSTAIN-6 Actually Changes in Clinical Practice

By
Published Updated Last reviewed
GLP-1 medication and metabolic health image for What SUSTAIN-6 Actually Changes in Clinical Practice

At a glance

| Field | Detail | |---|---| | Trial name | SUSTAIN-6 (Trial Evaluating Cardiovascular Outcomes with Semaglutide in Subjects with Type 2 Diabetes) | | N | 3,297 | | Intervention | Semaglutide 0.5 mg or 1.0 mg subcutaneous, once weekly | | Comparator | Placebo (volume-matched), once weekly | | Duration | Median 2.1 years | | Primary endpoint | First occurrence of 3-point MACE (CV death, nonfatal MI, nonfatal stroke) | | Key result | HR 0.74 (95% CI 0.58, 0.95; p <0.001 for noninferiority, p = 0.02 for superiority) |

Why This Trial Exists

The FDA's 2008 guidance on diabetes drugs demanded that every new glucose-lowering agent prove it does not increase cardiovascular risk. Most sponsors designed their CV outcomes trials (CVOTs) to meet the minimum bar: noninferiority. SUSTAIN-6 was no exception. Novo Nordisk powered the study to exclude an upper 95% CI bound of 1.8 for the hazard ratio, a noninferiority margin. The primary publication in the New England Journal of Medicine makes this explicit: the trial was not originally designed or powered to prove superiority.

That the superiority p-value (0.02) cleared anyway, in a trial of only 3,297 patients followed for roughly two years, caught the cardiology world off guard. By comparison, the LEADER trial of liraglutide enrolled 9,340 patients over 3.8 years to reach a similar conclusion. SUSTAIN-6 did it faster, in a smaller cohort, with a larger point estimate of benefit.

The Population That Was Actually Studied

Eligibility required type 2 diabetes with HbA1c ≥7.0% and one of three risk profiles:

  • Established cardiovascular disease (prior MI, stroke, TIA, revascularization, or >50% stenosis in coronary, carotid, or lower-extremity arteries)
  • Chronic heart failure (NYHA class II, III)
  • Chronic kidney disease stage 3 or higher, or age ≥60 with at least one additional CV risk factor

About 83% of enrolled patients had established CV disease, prior MI, or prior stroke. The remaining 17% qualified on risk factors alone. This skew matters: the trial tells us the most about semaglutide's CV effects in patients who already carry a heavy atherosclerotic burden. It tells us comparatively little about primary prevention in younger, lower-risk patients with newly diagnosed diabetes.

Baseline characteristics were typical for a CVOT: mean age 65, mean diabetes duration 13.9 years, mean HbA1c 8.7%, and roughly 61% male. About 93% were on metformin, 43% on insulin, and 48% on sulfonylureas. The trial protocol permitted background therapy adjustments, meaning investigators could titrate insulin and other agents as needed. This pragmatic design makes the results more generalizable to real-world endocrinology clinics, but it also introduces noise: some of the CV benefit could theoretically trace to differences in overall glycemic management rather than a direct vascular effect of semaglutide.

Methodology Notes That Matter

Randomization and blinding. Patients were randomized 1:1:1:1 to semaglutide 0.5 mg, semaglutide 1.0 mg, or matching placebo for each dose. All patients and investigators were blinded. The two dose groups were pooled for the primary analysis, which was pre-specified.

Event adjudication. An independent, blinded committee adjudicated all potential MACE events using standardized definitions. This is standard for CVOTs but worth noting because the event rate was relatively low (254 first MACE events across the entire trial), meaning even small adjudication disagreements could shift the hazard ratio.

Glycemic separation. By week 104, the semaglutide group had an HbA1c reduction approximately 0.7, 1.0 percentage points greater than placebo. This separation is clinically meaningful and raises the question every CVOT faces: is the CV benefit driven by glucose lowering, by the drug's direct vascular effects, or both? SUSTAIN-6 cannot answer this definitively. The primary analysis did not include mediation modeling for HbA1c.

Weight and blood pressure. Semaglutide-treated patients lost 2.9 to 4.3 kg more than placebo and had systolic blood pressure reductions of 1.3 to 2.6 mmHg more. These metabolic benefits likely contribute to the CV signal but cannot be isolated from the drug's other effects in this trial design.

The Actual Numbers

| Outcome | Semaglutide (n = 1,648) | Placebo (n = 1,649) | HR (95% CI) | |---|---|---|---| | 3-point MACE | 108 (6.6%) | 146 (8.9%) | 0.74 (0.58, 0.95) | | 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) | | CV death | 44 (2.7%) | 46 (2.8%) | 0.98 (0.65, 1.48) | | All-cause mortality | 62 (3.8%) | 60 (3.6%) | 1.05 (0.74, 1.50) |

The composite was driven primarily by reductions in nonfatal stroke and, to a lesser extent, nonfatal MI. Cardiovascular death showed no signal at all (HR 0.98). All-cause mortality was numerically higher in the semaglutide group, though not statistically significant.

This pattern is worth pausing on. A drug that reduces atherosclerotic events but does not reduce CV death after two years could mean: (a) the benefit on mortality simply needs more time to emerge, (b) the trial was underpowered for the mortality endpoint, or (c) the drug's benefit is genuinely limited to nonthrombotic pathways. The SELECT trial, published in 2023, later showed semaglutide 2.4 mg reduced MACE by 20% in patients with obesity but without diabetes, with a similar pattern of stronger effects on nonfatal events. The SELECT results suggest the SUSTAIN-6 pattern was likely a power issue rather than a ceiling on benefit.

Retinopathy: The Complication Nobody Expected

SUSTAIN-6 reported a statistically significant increase in diabetic retinopathy complications (HR 1.76 to 95% CI 1.11, 2.78). This finding alarmed ophthalmologists and generated its own literature. The most likely explanation is the "early worsening" phenomenon seen with rapid glycemic improvement: when HbA1c drops quickly, retinal microvasculature that adapted to chronic hyperglycemia can destabilize temporarily.

This effect is well-documented with insulin initiation and was seen in the DCCT. It is not unique to semaglutide. The Ozempic prescribing information carries a warning about this risk and recommends monitoring patients with existing retinopathy when initiating therapy. For clinicians, the practical takeaway: get a dilated eye exam before starting semaglutide in anyone with known retinopathy, and repeat it within 6 to 12 months.

What the Guidelines Actually Changed

Before SUSTAIN-6 and the preceding LEADER trial, the ADA/EASD consensus algorithm positioned GLP-1 receptor agonists as one option among many after metformin failure. The drug class competed with sulfonylureas, SGLT2 inhibitors, DPP-4 inhibitors, and insulin on roughly equal footing.

After SUSTAIN-6, the 2018 ADA/EASD consensus report restructured the algorithm entirely. For patients with type 2 diabetes and established atherosclerotic CV disease, a GLP-1 receptor agonist with proven CV benefit became the recommended first injectable, independent of HbA1c. This was a categorical shift: the indication was no longer glycemic control but cardiovascular protection.

The 2019 ESC/EASD guidelines on diabetes and cardiovascular disease went further, recommending GLP-1 receptor agonists or SGLT2 inhibitors as first-line therapy in treatment-naive patients with T2D and CV disease, ahead of metformin in some scenarios. By 2023, the ADA Standards of Care no longer required metformin as the universal first step; cardiometabolic risk now drove initial drug selection.

Which Prescribing Patterns Actually Shifted

The guideline changes translated unevenly into clinical practice. Several real-world patterns emerged:

Cardiology adoption was slow. Despite the CV data, most semaglutide prescriptions through 2020 originated from endocrinology and primary care, not cardiology. Cardiologists who routinely prescribed statins and PCSK9 inhibitors were slower to add GLP-1 agonists to their toolkit. This has changed gradually, but cardiologist-initiated GLP-1 prescribing remains a minority of total prescriptions.

Insurance barriers persisted. Many payers required prior authorization demonstrating metformin failure before covering semaglutide, even for patients with clear CV indications. Step therapy requirements often forced patients through cheaper agents first, delaying access to a drug with proven event reduction.

Dose confusion arose. SUSTAIN-6 tested 0.5 mg and 1.0 mg weekly doses. The commercial Ozempic launch added a 2.0 mg dose for glycemic control. The SELECT trial used 2.4 mg (the Wegovy dose) for CV outcomes in obesity. Clinicians sometimes conflated these doses and indications, prescribing Wegovy for diabetes or Ozempic for obesity-related CV protection without matching the evidence base to the specific product.

What the Trial Does Not Tell You

Primary prevention in low-risk diabetes. Only 17% of SUSTAIN-6 participants lacked established CV disease. The trial cannot confirm that semaglutide reduces MACE in a 45-year-old with newly diagnosed T2D, no prior events, and a 10-year ASCVD risk below 7.5%.

Heart failure outcomes. SUSTAIN-6 excluded patients with NYHA class IV heart failure and did not adjudicate heart failure hospitalizations as a primary or key secondary endpoint. The relationship between GLP-1 agonists and heart failure remains complex, and SUSTAIN-6 should not be cited as evidence for or against use in HFrEF.

Head-to-head comparisons. There is no CVOT directly comparing semaglutide to an SGLT2 inhibitor. The choice between empagliflozin (EMPA-REG OUTCOME) and semaglutide for a given patient remains an exercise in indirect comparison, patient phenotyping, and clinical judgment.

Duration beyond two years. The median follow-up was 2.1 years. Whether the MACE benefit grows, plateaus, or attenuates over five or ten years is unknown from this dataset.

The Clinical Translation, Plainly Stated

For a patient sitting in your office with type 2 diabetes, a prior MI, and an HbA1c of 8.2% on metformin, SUSTAIN-6 provides direct evidence that adding once-weekly semaglutide 0.5 to 1.0 mg will reduce their composite risk of CV death, nonfatal MI, and nonfatal stroke by roughly one quarter over two years. The number needed to treat (NNT) for MACE prevention over the trial period is approximately 45.

For patients who do not match this profile closely (no prior CV events, normal kidney function, younger age), the evidence is weaker but not absent. The totality of GLP-1 RA cardiovascular data, spanning LEADER, SUSTAIN-6, SELECT, and HARMONY Outcomes, supports a class effect. Current ADA guidance reflects this by recommending GLP-1 RAs for patients with T2D and established or high-risk ASCVD without requiring exact trial-population matching.

Frequently asked questions

›
›
›
›
›
›
›
›
›
›

References

  1. 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

  2. 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

  3. Cosentino F, Grant PJ, Aboyans V, et al. 2019 ESC Guidelines on Diabetes, Pre-Diabetes, and Cardiovascular Diseases. Eur Heart J. 2020;41(2):255-323. PubMed

  4. Lincoff AM, Brown-Frandsen K, Colhoun HM, et al. Semaglutide and Cardiovascular Outcomes in Obesity without Diabetes (SELECT). N Engl J Med. 2023;389(24):2221-2232. PubMed

  5. Ozempic (semaglutide) Prescribing Information. Novo Nordisk. Revised 2020. FDA Label