STEP-HFpEF

Published

Kosiborod MN, et al. "Semaglutide in Patients with Heart Failure with Preserved Ejection Fraction and Obesity". The New England Journal of Medicine. 2023. 389(12):1069-1084.
PubMed • Full text • PDF • ClinicalTrials.gov

Clinical Question

In patients with heart failure with preserved ejection fraction (HFpEF) and obesity (BMI ≥30 kg/m²) but no diabetes, does treatment with semaglutide improve symptoms, physical limitations, exercise function, and induce weight loss compared to placebo?

Bottom Line

In patients with HFpEF and obesity but no DM, semaglutide significantly improved heart failure–related symptoms and physical limitations, increased exercise capacity, and led to substantial weight loss over 52 weeks compared to placebo.

Major Points

GLP-1 antagonists were found to improve CVD outcomes in people with diabetes (e.g., SUSTAIN-6 in 2016) and obesity without DM (e.g., SELECT in 2023). These medications are also found to promote weight loss among persons with obesity but no diabetes (e.g., STEP 1 in 2021). HFpEF is a common heart failure subtype that relates to obesity. In the early 2020s, there were limited options to improve outcomes in HFpEF. Whether GLP-1 agonists benefit persons with HFpEF was unknown.

Published in 2023, the Semaglutide Treatment Effect in People with Heart Failure with Preserved Ejection Fraction (STEP-HFpEF) trial randomized 529 adults with obesity and symptomatic HFpEF but no diabetes to semaglutide (with dosing as done for the weight loss-promoting Wegovy brand rather than the diabetes Ozempic brand). Over a 52-week period, semaglutide led to significant improvements in heart failure symptoms and physical limitations (measured by the KCCQ-CSS), increased the 6-minute walk distance, and resulted in substantial weight loss compared to placebo. Additionally, semaglutide reduced C-reactive protein levels, suggesting decreased inflammation. (CRP is commonly lowered with weight loss.) The treatment was associated with fewer serious adverse events, primarily a reduction in cardiac events. There were more adverse events that led to discontinuation with semaglutide than placebo. These findings suggest that semaglutide may be an effective therapeutic option for managing HFpEF in patients with obesity, addressing a significant unmet need in this population.

Similar benefits were seen with semaglutide among patients with HFpEF, obesity, and diabetes in the STEP-HFpEF DM trial (2024).

Guidelines

As of June 2025, no guidelines have been published that reflect the results of this trial.

Design

• Multicenter, double-blind, parallel-group, randomized, placebo-controlled trial
• N=529
  • Semaglutide group (n=263)
  • Placebo group (n=266)
• Setting: 96 sites across 13 countries in Asia, Europe, North America, and South America
• Enrollment: March 2021 to March 2022
• Follow-up: 52 weeks of treatment plus a 5-week follow-up period
• Analysis: Intention-to-treat, per-protocol
• Dual primary end points:
  • Change in Kansas City Cardiomyopathy Questionnaire Clinical Summary Score
  • Change in body weight

Population

Inclusion Criteria

• Age ≥18 years
• Diagnosed with HFpEF:
  • Left ventricular ejection fraction ≥45%
  • NYHA functional class II–IV symptoms
• Evidence of elevated left ventricular filling pressures, elevated natriuretic peptide levels with echocardiographic abnormalities, or recent hospitalization for heart failure with ongoing diuretic treatment or echocardiographic abnormalities
• KCCQ-CSS score <90
• 6-minute walk distance ≥100 meters
• Body mass index (BMI) ≥30 kg/m²

Exclusion Criteria

• Body weight change >5 kg within 90 days before screening
• History of diabetes (glycated hemoglobin ≥6.5% or known diagnosis)
• Recent or planned bariatric surgery
• Use of other weight-management pharmacotherapy
• Conditions interfering with trial participation or data interpretation

Baseline Characteristics

• Demographics: Age 69 years, 56% female, 96% White, 7% Hispanic
• BMI: 37 kg/m²
• Weight: 105.1 kg
• KCCQ-CSS score: 58.9 points
• 6-minute walk distance: 320 meters
• Labs: NT-proBNP 450.8 pg/mL, CRP 3.9 mg/L
• LVEF:
  • 45% to <50%: 16%
  • 50% to <60%: 41%
  • ≥60%: 43%
• NYHA class: Two 66%, three or four 34%
• Medical problems: AF 52%, hypertension 82%, CAD 18%
• Medications at baseline:
  • ACEI, ARB, ARNI: 81%
  • MRA: 35%
  • Diuretic: 81%
    • Loop: 62%
    • Thiazide: 17%
  • Beta-blocker: 79%
  • SGLT2 inhibitor: 4%

Interventions

• Randomized to a group:
• Semaglutide- Semaglutide 2.4 mg administered once weekly via subcutaneous injection
  • Dose escalation from 0.25 mg to 2.4 mg over 16 weeks
• Placebo Group - Matching placebo administered once weekly via subcutaneous injection with identical dose escalation schedule.

Outcomes

Comparisons are semaglutide therapy vs. placebo.

Primary Outcomes

Mean change in KCCQ-CSS from baseline to week 52

The Kansas City Cardiomyopathy Questionnaire; higher scores denote fewer symptoms and better physical function

+16.6 vs. +8.7; difference 7.8 (95% CI 4.8-10.9; P<0.001)

≥5-point increase: 75% vs. 64%; OR 1.9 (95% CI 1.3 to 2.8)

≥10-point increase: 63% vs. 48%; OR 2.1 (95% CI 1.4 to 3.1)

≥15-point increase: 51% vs. 34%; OR 2.2 (95% CI 1.5 to 3.2)

Percentage change in body weight from baseline to week 52

-13.3% vs. -2.6%; difference -10.7 (95% CI -11.9 to -9.4; P<0.001)

≥10% reduction: 66% vs. 10%; OR 15.5 (95% CI 9.4 to 25.4)

≥15% reduction: 44% vs. 2%; OR 30.6 (95% CI 12.2 to 76.6)

≥20% reduction: 24% vs. 0.4%; OR 56.0% (7.8 to 400.8)

Secondary Outcomes

Change in 6-minute walk distance from baseline to week 52

21.5 vs. 1.2 m; difference 20.3 m (95% CI 8.6-32.1; P<0.001)

Percentage change in CRP levels from baseline to week 52

-43.5% vs. -7.3%; geometric mean ratio 0.61 (95% CI 0.51-0.72; P<0.001)

Additional Outcomes

Hierarchical composite endpoint (death, heart failure events, changes in KCCQ-CSS and 6-minute walk distance), crude win percentage

60.1 vs. 34.9; OR 1.72 (95% CI 1.37-2.15); P<0.001

Reduction in NT-proBNP from baseline to 52 weeks

-20.9% vs. -5.3% geometric mean treatment ratio 0.84 (95% CI 0.71 to 0.98)

HF hospitalization or urgent visit

1 vs. 12 events; HR 0.08 (95% CI 0.00 to 0.42)

Adverse Events

Comparisons are semaglutide vs. placebo

Serious adverse events

13% vs. 27%; P<0.001

Cardiac: 3% vs. 11%; P<0.001

Infection: 1.5% vs. 6%; P=0.006

GI: 3% vs. 3%; P=1.00

Nervous system: 3% vs. 3%; P=0.80

Renal/GU: 2.3% vs. 1.5%; P=0.54

Adverse events leading to discontinuation

13.3% vs. 5.3%

Fatal events

1.1% vs. 1.5%

Criticisms

• Limited diversity, 95.8% were white, which may limit generalizability
• Relatively short follow-up duration. Semaglutide requires continued use to maintain weight loss
• Low Use of SGLT2 Inhibitors: Only 3.6% of participants were on SGLT2 inhibitors, which are guideline directed in HFpEF management
• Exclusion of patients with diabetes
• No subgroup analysis

Funding

• Sponsor: Novo Nordisk, the manufacturer of semaglutide.
• Conflicts of Interest: Several authors disclosed relationships with pharmaceutical companies, including Novo Nordisk.