Question 1

How is SUMMIT different from STEP-HFpEF, the semaglutide HFpEF trial?

Accepted Answer

STEP-HFpEF (Kosiborod 2023 NEJM, semaglutide 2.4 mg) and STEP-HFpEF DM (Kosiborod 2024 NEJM, the diabetes-cohort companion) showed improvements in patient-reported symptoms (KCCQ-CSS), 6-minute walk distance, hsCRP, and weight in HFpEF and obesity but were not powered to reduce hard cardiovascular outcomes. SUMMIT enrolled 731 patients (larger and longer than STEP-HFpEF's two trials of about 529 and 616 patients), tested tirzepatide instead of semaglutide, and was the first to demonstrate a statistically significant reduction in the composite of cardiovascular death or worsening heart-failure event (HR 0.62, P=0.026) in addition to the KCCQ-CSS improvement. The two programs are complementary rather than head-to-head.

Question 2

Does the SUMMIT result apply to heart failure with reduced ejection fraction (HFrEF)?

Accepted Answer

No. SUMMIT enrolled only patients with an ejection fraction of at least 50 percent, the HFpEF phenotype. HFrEF (ejection fraction below 40 percent) has a separate evidence base built on ACE inhibitors and ARBs, beta blockers, mineralocorticoid antagonists, sacubitril-valsartan, and SGLT2 inhibitors. There is no completed phase-3 trial of tirzepatide in HFrEF, and clinicians should not extrapolate the SUMMIT result to that population. Ongoing programs such as Lilly's broader cardiovascular outcomes trial portfolio may eventually address HFrEF separately.

Question 3

Does the Zepbound HFpEF FDA label require the patient to also have obesity?

Accepted Answer

Yes. The FDA-expanded indication, like the SUMMIT eligibility criteria, applies to adults with HFpEF and obesity, defined in the trial as a body-mass index of at least 30 kg/m². A patient with HFpEF but a BMI in the normal range is outside the studied population and outside the label. Patients should consult their cardiologist or endocrinologist about whether the label criteria apply to their specific situation; the trial did not study patients with BMI below 30.

Question 4

Was the cardiovascular benefit larger in patients without diabetes, given they lost more weight?

Accepted Answer

No, and that is one of the most clinically important findings from the prespecified diabetes-stratified analysis. Patients with type 2 diabetes lost roughly 10.4 percent of body weight on tirzepatide versus 12.9 percent in patients without diabetes, but the hazard ratio for the composite cardiovascular endpoint was effectively identical in the two strata (HR 0.64 with diabetes versus 0.61 without; Pinteraction 0.95). Improvements in KCCQ-CSS, 6-minute walk distance, quality of life, and NYHA functional class were also similar between strata. This suggests the heart-failure benefit is not simply a function of weight-loss magnitude and may reflect mechanisms beyond weight loss, including reductions in blood volume, inflammation, and myocardial injury.

Question 5

Should a patient with HFpEF currently on Wegovy switch to Zepbound based on SUMMIT?

Accepted Answer

There is no head-to-head comparison of Wegovy and Zepbound in HFpEF, and the two drugs are now both supported by phase-3 evidence in this population through different programs (semaglutide via STEP-HFpEF and STEP-HFpEF DM; tirzepatide via SUMMIT). SUMMIT showed a hard cardiovascular outcomes benefit that the STEP-HFpEF program was not designed to detect, while STEP-HFpEF demonstrated KCCQ-CSS and 6-minute walk benefits in line with what SUMMIT later confirmed for tirzepatide. The decision to switch is a clinical one that should weigh symptom control on the current regimen, tolerability, insurance coverage, and the patient's cardiovascular risk profile. Patients should not switch GLP-1 medications without consulting their prescriber.

Question 6

What was the absolute risk reduction in SUMMIT, and what is the number needed to treat?

Accepted Answer

The composite of cardiovascular death or worsening heart-failure event occurred in 9.9 percent of the tirzepatide group versus 15.3 percent of the placebo group over a median follow-up of 104 weeks, an absolute risk reduction of about 5.4 percentage points. That corresponds to a number needed to treat (NNT) of approximately 19 over roughly two years to prevent one composite cardiovascular event. The NNT for the KCCQ-CSS improvement is more difficult to express because it is a continuous quality-of-life measure rather than a binary outcome, but the 6.9-point between-group difference exceeds the 5-point threshold widely used as a clinically meaningful within-patient change.

Endpoint	Treatment	Comparator	Difference
Change from baseline in KCCQ Clinical Summary Score at 52 weeks (coprimary endpoint, range 0-100, higher = better quality of life) A between-group difference of approximately 5 points on the KCCQ-CSS is widely considered clinically meaningful in HFpEF trials; SUMMIT exceeded that threshold.	+19.5 points (SD 1.2)	+12.7 points (SD 1.3)	Between-group difference 6.9 points (95% CI 3.3 to 10.6; P<0.001)
Change from baseline in 6-minute walk distance at 52 weeks Mechanistic improvement in functional exercise capacity, consistent with the symptom-burden reduction captured by KCCQ-CSS.	Increased on tirzepatide	Smaller change on placebo	Between-group difference approximately 18 meters favoring tirzepatide (P<0.001)
Percent change from baseline in body weight at 52 weeks Weight loss in HFpEF participants mirrors the magnitudes seen in obesity-only SURMOUNT trials; the diabetes subgroup lost 10.4% versus 12.9% in the non-diabetes subgroup.	Approximately -13.9%	Approximately -2.2%	Between-group difference roughly -11.6 percentage points (P<0.001)
Percent change from baseline in high-sensitivity C-reactive protein at 52 weeks Anti-inflammatory effect was tightly correlated with reductions in troponin T and improved 6-minute walk distance in mechanistic analyses.	Reduced	Smaller change	Estimated treatment difference -37.2% (95% CI -45.7 to -27.3; P<0.001)
Percent change from baseline in NT-proBNP at 52 weeks Numerical reduction in the canonical heart-failure biomarker that fell just short of statistical significance, consistent with concurrent reductions in blood volume and filling pressures.	Reduced	Smaller change	Estimated treatment difference -10.5% (95% CI -20.7 to 1.0; P=0.07)
Change from baseline in systolic blood pressure at 52 weeks Modest blood-pressure reduction contributing to the volume-pressure offloading hypothesis advanced in the Borlaug mechanistic companion.	Reduced	Smaller change	Estimated treatment difference -5 mmHg (95% CI -7 to -3; P<0.001)
Percent change from baseline in troponin T at 52 weeks Reduction in a marker of myocardial injury, supporting a structural rather than purely hemodynamic benefit.	Reduced	Smaller change	Estimated treatment difference -10.4% (95% CI -16.7 to -3.6; P=0.003)
NYHA functional-class improvement at 52 weeks	Higher proportion improved	Lower proportion improved	Favors tirzepatide; magnitude similar in patients with and without diabetes in prespecified subgroup analysis

Event	Treatment rate	Comparator rate
Adverse event leading to discontinuation of trial drug Predominantly gastrointestinal.	23/364 (6.3%)	5/367 (1.4%)
Nausea (gastrointestinal events were the most common AE class) Consistent with the GI profile across the SURMOUNT and SURPASS programs.	Higher on tirzepatide than placebo	Lower
Serious adverse events No new safety signal identified relative to prior tirzepatide cardiovascular and obesity programs.	Similar between arms	Similar

SUMMIT trial deep-dive: tirzepatide for HFpEF and obesity (Packer 2025 NEJM)

Primary endpoint

Secondary endpoints

Adverse events

Subgroup analyses

Clinical significance

Frequently Asked Questions

References

Related trial deep-dives