Semaglutide in obesity-related heart failure with preserved ejection fraction and type 2 diabetes across baseline HbA1c levels (STEP-HFpEF DM): a prespecified analysis of heart failure and metabolic outcomes from a randomised, placebo-controlled trial

Abstract

Background

About half of patients with heart failure with mildly reduced or preserved ejection fraction (HFpEF) have type 2 diabetes. In the STEP-HFpEF DM trial of adults with obesity-related HFpEF and type 2 diabetes, subcutaneous once weekly semaglutide 2·4 mg conferred improvements in heart failure-related symptoms and physical limitations, bodyweight, and other heart failure outcomes. We aimed to determine whether these effects of semaglutide differ according to baseline HbA_1c.

Methods

STEP-HFpEF DM, a double-blind, randomised, placebo-controlled trial conducted at 108 clinical research sites across 16 countries in Asia, Europe, and North and South America, included individuals aged 18 years or older with documented HFpEF (left ventricular ejection fraction ≥45%), type 2 diabetes, and obesity (BMI ≥30 kg/m²). Participants were randomly assigned (1:1), with a block size of four within each stratum using an interactive web response system, stratified by baseline BMI (<35 kg/m² vs ≥35 kg/m²), to receive either semaglutide 2·4 mg or placebo subcutaneously. The effects of semaglutide versus placebo on the efficacy endpoints were evaluated by HbA_1c categories at baseline: low (<6·5%; <48 mmol/mol), medium (6·5% to <7·5%; 48 mmol/mol to <58 mmol/mol), and high (≥7·5%; ≥58 mmol/mol). The dual primary endpoints were change in the Kansas City Cardiomyopathy Questionnaire Clinical Summary Score (KCCQ-CSS) and bodyweight percentage from baseline to 52 weeks and were assessed in all randomly assigned participants by intention to treat. Hypoglycaemia events were also analysed to assess safety in all randomly assigned participants who received at least one dose of study drug. This trial is registered with ClinicalTrials.gov, NCT04916470.

Findings

Between June 27, 2021 and Sept 2, 2022, 616 participants were enrolled and randomly assigned (mean age 68·4 years [SD 8·9]; 273 [44%] were female, 343 [56%] were male, and 519 [84%] were White). The low baseline HbA_1c group included 227 participants (112 assigned to semaglutide and 115 to placebo), the medium baseline HbA_1c group included 226 participants (124 assigned to semaglutide and 102 to placebo), and the high baseline HbA_1c group included 163 participants (74 assigned to semaglutide and 89 to placebo). The median duration of follow-up in the overall trial was 401 days (IQR 400–405). The change in KCCQ-CSS from baseline to 52 weeks was 12·4 points (95% CI 8·8 to 16·0) with semaglutide versus 5·7 points (2·1 to 9·2) with placebo (mean difference 6·7 points [1·6 to 11·8]) in the low baseline HbA_1c group, 14·5 points (11·0 to 17·9) versus 8·5 points (4·8 to 12·2; 6·0 points [0·9 to 11·1]) in the medium baseline HbA_1c group, and 14·5 points (10·0 to 19·0) versus 4·8 points (0·7 to 8·9; 9·6 points [3·5 to 15·7]) in the high baseline HbA_1c group (p_interaction=0·64; p_trend=0·46). The change in bodyweight percentage from baseline to 52 weeks was –10·8 (95% CI –12·1 to –9·5) with semaglutide versus –3·3% (–4·6 to –2·0) with placebo (mean difference –7·5% [–9·4 to –5·6]) in the low baseline HbA_1c group, –9·6% (–10·8 to –8·3) versus –3·3% (–4·7 to –1·9; –6·3 [–8·2 to –4·4]) in the medium baseline HbA_1c group, and –8·6% (–10·2 to –7·0) versus –3·6% (–5·2 to –2·1; –5·0 [–7·2 to –2·7]) in the high baseline HbA_1c group (p_interaction=0·22; p_trend=0·083). Hypoglycaemia events occurred in 30 (10%) of 310 participants (70 events; 22·9 events per 100 person-years) in the semaglutide group compared with 21 (7%) of 306 participants in the placebo group (90 events; 29·5 events per 100 person-years).

Interpretation

Semaglutide 2·4 mg improved heart failure-related symptoms and physical limitations, and reduced bodyweight in patients with obesity-related HFpEF and type 2 diabetes, all independently of baseline HbA_1c, and resulted in lower rates of hypoglycaemia than placebo, despite a well controlled baseline HbA_1c and broad use of concomitant glucose-lowering medications.

Funding

Novo Nordisk.