SUSTAIN-6

Published

Marso SP, et al. "Semaglutide and Cardiovascular Outcomes in Patients with Type 2 Diabetes". The New England Journal of Medicine. 2016. 375(18):1834-1844.
PubMed • Full text • PDF • ClinicalTrials.gov

Clinical Question

In patients with T2DM at high CV risk, is semaglutide non-inferior to placebo in terms of CVD events?

Bottom Line

In patients with T2DM at high CV risk, semaglutide is non-inferior to placebo in terms of CVD events. Semaglutide was in fact found to have lower risk of CVD events in a superiority analysis.

Major Points

Patients with T2DM are at elevated risk for CVD events. In 2008, the FDA recommended that DM medications be evaluated for CVD safety. Semaglutide is a glucagon-like peptide 1 (GLP-1) agonist intended for use in persons with T2DM, but had not been assessed for CVD events. Published in 2016, the Trial to Evaluate Cardiovascular and Other Long-term Outcomes with Semaglutide in Subjects with Type 2 Diabetes (SUSTAIN-6) sought to determine whether semaglutide was non-inferior to placebo in terms of CVD events among individuals with T2DM at high risk for CVD events.

The randomized, double-blind, placebo-controlled, noninferiority trial randomized 3297 patients to receive semaglutide (once-weekly, 0.5 mg or 1.0 mg) or placebo for 104 weeks. The primary outcome was the composite of CV mortality, nonfatal MI, or nonfatal stroke. The outcome was observed in 6.6% and 8.9% of the semaglutide and placebo groups, respectively (hazard ratio 0.74; 95%CI 0.58-0.95; P<0.001 for noninferiority). Importantly, the authors also found semaglutide to be superior to placebo for this composite CVD outcome (P=0.02). As compared to placebo, semaglutide was associated with more frequent gastrointestinal events. The trial also noted that the semaglutide group had a higher incidence of diabetic retinopathy as compared to placebo (3 vs. 1.8%, hazard ratio 1.76; 95% CI 1.11-2.78; P=0.02). It was concluded that the noninferiority of semaglutide was confirmed.

The use of semaglutide for CVD risk reduction among non-diabetics with BMI ≥27 kg/m² and prior CVD events was confirmed in the 2023 SELECT trial. Of note, 2016 also saw the publication of LEADER, which found similar CVD benefit with the GLP-1 agonist, liraglutide.

Guidelines

ADA Diabetes Care (2023, adapted)

• Among persons with T2DM and established ASCVD or at high risk for ASCVD, GLP-1 agonists with known CVD benefit lowers risk of CVD events (A)

Design

• Multicenter, double-blind, parallel-group, placebo-controlled trial
• N=3,297
  • Semaglutide 0.5 mg (n=826)
  • Semaglutide 1 mg (n=822)
  • Placebo 0.5 mg (n=824)
  • Placebo 1 mg (n=825)
• Setting: 230 sites in 20 countries
• Enrollment: February-December 2013
• Median follow-up: 2.1 years
• Analysis: Non-inferiority
• Primary outcome: CV mortality, nonfatal MI, or nonfatal stroke

Population

Inclusion Criteria

• Type 2 DM
  • Either if patients have not been treated with anti-hyperglycemic drug, or treated with ≤2 anti-hyperglycemic drug with or without insulin (basal or premixed)
• Hemoglobin A1c ≥7%
• Age ≥50 years with ≥1 criteria below:
  • Established CV disease (previous CV, cerebrovascular, or peripheral vascular disease)
  • heart failure (NYHA class II or III)
  • CKD ≥stage 3
• Age ≥60 years with ≥1 criteria below:
  • persistent microalbuminuria or proteinuria
  • LVH (on ECG or imaging)
  • LV systolic or diastolic dysfunction (on imaging)
  • ankle/brachial pressure index<0.9

Exclusion Criteria

• Received a dipeptidyl-peptidase (DPP) 4 inhibitor ≤30 days or a GLP-1–receptor agonist or insulin (except basal or premixed) ≤90 days before screening
• ACS or cerebrovascular event ≤90 days before randomization
• Planned coronary, carotid, or peripheral artery revascularization
• HD
• T1DM
• Chronic pancreatitis or idiopathic acute pancreatitis
• ESLD
• Solid organ transplant (previous or planned)
• Cancer in prior 5 years
• Personal or family history of MEN2, familial, or non-familial medullary thyroid carcinoma
• Calcitonin ≥50 ng/L
• NYHA class IV heart failure
• Worsening of glycemic control ≤90 days prior to screening
• Hypersensitivity (known or suspected) to semaglutide
• Pregnant, breast-feeding, or not on adequate contraception

Baseline Characteristics

From the semaglutide 1 mg group (n=822); some details are in the supplementary appendix Tables S6S8

• Demographics: Age 65 years; male sex 63%; White race 84.1%, Black race 7&, Asian 7%
• Weight: 93 kg; BMI 33 kg/m²
• Duration of diabetes: 14 years; HbA1c: 8.7%
• Treatment for diabetes: Biguanides 72%; sulfonylurea 42%; thiazolidinedione 3%; alpha-glucosidase inhibitors 1%; meglitinides 3%; SGLT-2 inhibitors 0.1%; insulin 58%
• Other medications: Statins 73%, aspirin 66%
• PMH: IHD 60%; MI 32%; heart failure 22%; ischemic stroke 11%; hemorrhagic stroke 3%; hypertension 94%; non-smoker: 44%
• BP: 136/77 mm Hg
• Cholesterol: LDL-C 83 mg/dL; HDL-C 43 mg/dL
• Renal function: normal 30%; eGFR<15 ml/min/1.73m² 0.5%

Interventions

• Randomization in a 1:1:1:1 ratio to a group:
  • Semaglutide 0.5 mg weekly
  • Semaglutide 1 mg weekly
  • Placebo 0.5 mg weekly
  • Placebo 1 mg weekly
• Doses were titrated up from 0.25mg for 4 weeks that escalated to 0.5mg for 4 weeks until the maintenance dose was reached, either 0.5mg or 1mg, and no changing of the maintenance dose was permitted.

Outcomes

Comparisons are semaglutide vs. placebo

Primary Outcomes

CV mortality, nonfatal MI, or nonfatal stroke

6.6% vs. 8.9% (HR 0.74; 95% CI 0.58-0.95; P<0.001 for noninferiority and P=0.02 for superiority)

Secondary Outcomes

CV mortality, nonfatal MI, nonfatal stroke, coronary or peripheral revascularization, hospitalization for heart failure or unstable angina

12.1% vs. 16.0% (HR 0.74; 95% CI 0.62-0.89; P=0.002)

All-cause mortality, nonfatal MI, or nonfatal stroke

7.4% vs. 9.6% (HR 0.77; 95% CI 0.61-0.97; P=0.03)

All-cause mortality

3.8% vs. 3.6% (HR 1.05; 95% CI 0.74-1.50; P=0.79)

CV mortality

2.7% vs. 2.8% (HR 0.98; 95% CI 0.65-1.48; P=0.92)

Nonfatal MI

2.9% vs. 3.9% (HR 0.74; 95% CI 0.51-1.08; P=0.12)

Nonfatal stroke

1.6% vs. 2.7% (HR 0.61; 95% CI 0.38-0.99; P=0.04)

Hospitalization for heart failure or unstable angina

1.3% vs. 1.6% (HR 0.82; 95% CI 0.47-1.44; P=0.49)

Revascularization

5% vs. 7.6% (HR 0.65; 95% CI 0.50-0.86; P=0.003)

Hospitalization for heart failure

3.6% vs. 3.3% (HR 1.11; 95% CI 0.77-1.61; P=0.57)

Retinopathy complications

3% vs. 1.8% (HR 1.76; 95% CI 1.11-2.78; P=0.02)

New or worsening nephropathy

3.8% vs. 6.1% (HR 0.64; 95% CI 0.46-0.88; P=0.005)

Other analyses

From Table S10 in the supplement.

HbA_1C difference post-treatment

Semaglutide 0.5 mg vs. placebo: −1.1% vs. -0.4% (0.7% point lower than placebo; P<0.001)

Semaglutide 1 mg vs. placebo: −1.4% vs. -0.4% (1% point lower than placebo; P<0.001)

Body weight difference post-treatment

semaglutide 0.5 mg vs. placebo: −3.6 kg vs. −0.7 kg (2.9 kg lower than placebo; P<0.001)

semaglutide 1 mg vs. placebo: −4.9 kg vs. −0.5 kg (4.3 kg lower than placebo; P<0.001)

Systolic BP difference post-treatment

semaglutide 0.5 mg vs. placebo: -3.4 mm Hg vs. -2.2 mm Hg (1.3 mm Hg lower than placebo; P=0.1)

semaglutide 1 mg vs. placebo: -5.4 mm Hg vs. -2.8 mm Hg (2.6 mm Hg lower than placebo; P<0.001)

Pulse rate difference post-treatment

semaglutide 0.5 mg vs. placebo: +2.1 bpm vs. +0.1 bpm (2 bpm higher than placebo; P<0.001)

semaglutide 1 mg vs. placebo: +2.4 bpm vs. -0.1 bpm (2.5 bpm higher than placebo; P<0.001)

Subgroup Analysis

No significant interactions for age, gender, race, ethnicity BMI, HbA_1C, duration of diabetes, heart failure severity, history of vascular disease, insulin treatment or renal function

Adverse Events

P-values were not provided for these comparisons.

Any

Semaglutide 0.5 mg vs. placebo: 89.6% vs. 90.8%

Semaglutide 1 mg vs. placebo: 89.1% vs. 89.2%

Serious

Semaglutide 0.5 mg vs. placebo: 35% vs. 39.9%

Semaglutide 1 mg vs. placebo: 33.6% vs. 36.1%

Severe

Semaglutide 0.5 mg vs. placebo: 24.2% vs. 26.2%

Semaglutide 1 mg vs. placebo: 25.2% vs. 23.5%

Leading to discontinuation

Diarrhea

Semaglutide 0.5 mg vs. placebo: 1.8% vs. 0.6%

Semaglutide 1 mg vs. placebo: 2.3% vs. 0.2%

Nausea

Semaglutide 0.5 mg vs. placebo: 2.2% vs. 0.2%

Semaglutide 1 mg vs. placebo: 4.6% vs. 0.2%

Vomiting

Semaglutide 0.5 mg vs. placebo: 1.7% vs. 0.4%

Semaglutide 1 mg vs. placebo: 2.8% vs. 0.2%

GI

Semaglutide 0.5 mg vs. placebo: 50.7% vs. 35.7%

Semaglutide 1 mg vs. placebo: 52.3% vs. 35.2%

Acute pancreatitis

Semaglutide 0.5 mg vs. placebo: 0.7 vs. 0.4%

Semaglutide 1 mg vs. placebo: 0.4% vs. 1.1%

Gallbladder disorder

Semaglutide 0.5 mg vs. placebo: 3.9% vs. 4.6%

Semaglutide 1 mg vs. placebo: 3.2% vs. 2.8%

Neoplasm (benign and malignant)

Semaglutide 0.5 mg vs. placebo: 8% vs. 8.5%

Semaglutide 1 mg vs. placebo: 10.8% vs. 8.4%

Pancreatic malignancy

Semaglutide 0.5 mg vs. placebo: 0% vs. 0.2%

Semaglutide 1 mg vs. placebo: 0.1% vs. 0.2%

Hypoglycemia

Semaglutide 0.5 mg vs. placebo: 23.1% vs. 21.5%

Semaglutide 1 mg vs. placebo: 21.7% vs. 21%

Criticisms

• The primary outcome was only significant as a composite, with only nonfatal stroke being statistically significant. However, the outcomes of cardiovascular death and nonfatal MI may not have shown significance due to an inadequately sized study population.
• Participants probably have higher baseline CVD risk than the general DM population
• Adjustments to other antihyperglycemic medications may have been differed between patients
• Individuals in the 0.5mg group may have needed a 1mg dose of semaglutide to achieve the appropriate response
• Designed for non-inferiority.

Funding

Novo Nordisk