Semaglutide 2.4 mg scored 7.3 points above placebo on KCCQ-CSS in HFpEF with type 2 diabetes

Participants on semaglutide 2.4 mg once weekly had a mean change of 14.4 points on the KCCQ Clinical Summary Score at week 52, against 7.6 points on placebo, a treatment difference of 7.3 points (95 percent confidence interval 4.1 to 10.4), the ClinicalTrials.gov record shows. The Novo Nordisk trial (STEP HFpEF DM) enrolled 617 people with obesity-related heart failure with preserved ejection fraction and type 2 diabetes, so this population now has posted results from a randomized, placebo-controlled phase 3 trial of this dose.

Why it matters

The registry says the study was designed to compare semaglutide with a "dummy" medicine on body weight and heart failure symptoms (the study record). The DailyMed label for RYBELSUS and OZEMPIC tablets, version published Aug 19, 2026, lists two uses: as an adjunct to diet and exercise to improve glycemic control in adults with type 2 diabetes mellitus, and to reduce the risk of major adverse cardiovascular events in adults with type 2 diabetes mellitus who are at high risk for these events. The same label states that safety and effectiveness have not been established in pediatric patients. That label covers tablets, while this trial tested a once-weekly injection.

Primary outcomes at week 52

The record gives November 2, 2026 as the date results were first posted. It lists two primary outcomes, both analyzed in the full analysis set with missing week 52 observations imputed.

Change in KCCQ Clinical Summary Score: 14.4 (standard deviation 18.3) with semaglutide, 281 participants analyzed, and 7.6 (standard deviation 18.8) with placebo, 272 analyzed. The treatment difference was 7.3 points (95 percent confidence interval 4.1 to 10.4), p-value <0.0001, by ANCOVA.
Percentage change in body weight: -10.2 (standard deviation 6.8) with semaglutide, 286 analyzed, and -3.2 (standard deviation 5.9) with placebo, 278 analyzed. The treatment difference was -6.4 (95 percent confidence interval -7.6 to -5.2), p-value <0.0001.

The KCCQ is a standardised 23-item, self-administered instrument on heart failure symptoms, physical limitation, quality of life and social limitation. Scores run from 0 to 100, and higher scores reflect better health status. The Clinical Summary Score covers the symptom and physical limitation domains.

Secondary measures

The record lists these week 52 results for secondary outcomes.

Six-minute walking distance, change from baseline to week 52: an increase of 16.9 meters (standard deviation 54.3) with semaglutide, 281 analyzed, against an increase of 3.8 meters (standard deviation 49.8) with placebo, 265 analyzed.
Participants with 10 percent weight loss or more: 147 of 286 with semaglutide and 29 of 278 with placebo.
Hypoglycemia, counted from baseline through week 57 during treatment rather than at week 52: the record lists 20 treatment-emergent severe or clinically significant episodes in the semaglutide group and 30 in the placebo group, in the safety analysis set of 310 and 306 participants. Clinically significant means blood glucose below 54 milligrams per deciliter.

Design and participants

Participants were randomized in a parallel design with quadruple masking to semaglutide 2.4 mg once weekly or to matching placebo, both given by subcutaneous injection for 52 weeks. The dose rose from 0.25 mg in weeks 1 to 4, to 0.5 mg in weeks 5 to 8, to 1.0 mg in weeks 9 to 12, to 1.7 mg in weeks 13 to 16, and to 2.4 mg from week 17 to week 52. A 5-week follow-up ran to week 57.

The trial began on 2021-06-15 and reached primary completion on 2023-10-11. It ran at 108 sites in 16 countries. Inclusion criteria required age of 18 years or older, a body mass index of 30.0 kg/m^2 or more, and a left ventricular ejection fraction of 45% or more at screening. Participants also needed a type 2 diabetes diagnosis at least 90 days before screening and HbA1c of 10.0% or less. In the 617 participants who started, 310 were assigned to semaglutide and 307 to placebo. The full analysis set held 310 and 306 participants, because one participant randomized to placebo by administrative error was excluded.

Safety measures

The record counts participants with at least one event. Serious adverse events affected 55 of 310 participants on semaglutide and 88 of 306 on placebo. Deaths affected 6 participants on semaglutide and 10 on placebo.

Among adverse events listed above the 5 percent frequency threshold, nausea affected 55 participants on semaglutide and 18 on placebo. Diarrhoea affected 42 and 20 participants, and vomiting affected 26 and 7. Cardiac failure was reported as an adverse event in 13 participants on semaglutide and 30 on placebo.

The record states that Novo Nordisk reserves the right to postpone publication and communication of results for up to 60 days.