Zepbound for Heart Failure: Off-Label Use, Evidence, and Dosing Protocol

Is Zepbound approved for heart failure?

No. The Zepbound label lists two indications: to reduce excess body weight and maintain weight reduction long term in adults with obesity, or with overweight plus at least one weight-related condition; and to treat moderate-to-severe obstructive sleep apnea in adults with obesity [1]. The indications section names no form of heart failure [1]. A prescription written for heart failure therefore falls outside the labeled indication.

Heart failure with preserved ejection fraction (HFpEF) is common: in the community, roughly half of people with heart failure have HFpEF, and obesity is one of its risk factors [3].

What SUMMIT found in HFpEF with obesity

SUMMIT was double-blind and placebo-controlled, assigning 731 adults in a 1:1 ratio: 364 to tirzepatide and 367 to placebo [2]. Entry required heart failure with an ejection fraction of at least 50% and a body-mass index of at least 30 [2]. Tirzepatide was given subcutaneously once weekly at up to 15 mg for at least 52 weeks, and median follow-up was 104 weeks [2]. The trial had two primary endpoints [2].

Cardiovascular death or worsening heart failure. This composite occurred in 36 patients (9.9%) on tirzepatide and 56 (15.3%) on placebo, hazard ratio 0.62 (95% CI 0.41 to 0.95; P=0.026) [2]. By component, worsening heart failure events occurred in 29 (8.0%) versus 52 (14.2%), hazard ratio 0.54 (95% CI 0.34 to 0.85), and adjudicated cardiovascular death in 8 (2.2%) versus 5 (1.4%), hazard ratio 1.58 (95% CI 0.52 to 4.83) [2]. The difference in the composite came from worsening heart failure events; the number of cardiovascular deaths was small and the confidence interval for that hazard ratio was wide [2].

Quality of life. The Kansas City Cardiomyopathy Questionnaire clinical summary score rose 19.5 points with tirzepatide and 12.7 with placebo at 52 weeks, a between-group difference of 6.9 points (95% CI 3.3 to 10.6; P<0.001) [2].

Adverse events, mainly gastrointestinal, led to discontinuation of the trial drug in 23 patients (6.3%) in the tirzepatide group [2].

What SUMMIT did not answer

• Ejection fraction below 50%. Entry required at least 50%, so heart failure with reduced or mildly reduced ejection fraction was not part of the trial [2].
• People without obesity. A body-mass index of at least 30 was required [2].
• Comparison with other heart failure drugs. The trial was placebo-controlled [2].

What dosing protocol applies off-label?

The label's dosage sections cover weight reduction and obstructive sleep apnea, not heart failure [1]. SUMMIT studied tirzepatide at doses up to 15 mg weekly in people with HFpEF and obesity [2]. For Zepbound's approved indications, the label starts dosing at 2.5 mg once weekly for 4 weeks, followed by 5 mg and 2.5-mg increases after at least 4 weeks on the current dose; maintenance doses are indication-specific and the maximum is 15 mg weekly [1]. The label says to consider a lower maintenance dose if a maintenance dose is not tolerated, and it calls for no dose adjustment for renal or hepatic impairment [1].

Other label dosing details: a missed dose can be given within 4 days (96 hours), otherwise it is skipped and the regular day resumes [1]. The weekly day can be changed if two doses are at least 3 days (72 hours) apart [1]. The dose can be given at any time of day, with or without food, into the abdomen, thigh, or back of the upper arm (the upper arm by another person), rotating sites [1]. The label names three presentations (a vial, a prefilled single-dose pen, and a single-patient-use KwikPen) and says patients should be trained on the specific one they receive [1].

Label warnings and adverse reactions

The Zepbound label's warnings cover volume depletion and acute kidney injury, severe gastrointestinal reactions, gallbladder disease, pancreatitis, hypersensitivity, pulmonary aspiration, and pregnancy; hypotension and heart rate changes are reported in its adverse reactions section [1].

• Volume depletion and kidney injury. Severe gastrointestinal reactions were reported in 1.7% (5 mg), 2.5% (10 mg) and 3.1% (15 mg) of patients versus 1% on placebo, and the drug is not recommended in severe gastroparesis [1]. Postmarketing reports describe acute kidney injury, sometimes requiring hemodialysis, mostly after nausea, vomiting or diarrhea caused dehydration [1]. The label says to monitor renal function in patients reporting reactions that could lead to volume depletion, especially during initiation and escalation [1].
• Blood pressure. Hypotension was reported in 1.6% versus 0.1% on placebo, and was more frequent in patients taking antihypertensive therapy (2.2%) than in those not taking it (1.2%) [1].
• Heart rate. Heart rate rose a mean 1 to 3 beats per minute [1].
• Gallbladder and pancreas. Cholecystitis occurred in 0.7% versus 0.2% on placebo. The label warns that acute pancreatitis has been observed and says to discontinue Zepbound if pancreatitis is suspected [1].
• Procedures. The label reports pulmonary aspiration during general anesthesia or deep sedation and tells patients to inform healthcare providers of planned surgeries or procedures [1].
• Contraindications. A personal or family history of medullary thyroid carcinoma or MEN 2, and known serious hypersensitivity to tirzepatide or an excipient [1].
• Pregnancy. Weight loss offers no benefit in pregnancy and may cause fetal harm; the label says to discontinue when pregnancy is recognized [1].

Does weight loss by itself help HFpEF?

A randomized trial tested calorie restriction and aerobic exercise in 100 older adults with obesity and stable HFpEF. Participants' mean age was 67 and mean body-mass index 39.3, and they were assigned to 20 weeks of diet, exercise, both, or attention control [4]. Peak oxygen consumption rose 1.3 mL/kg/min with diet (95% CI 0.8 to 1.8; P<0.001) and 1.2 mL/kg/min with exercise (95% CI 0.7 to 1.7; P<0.001), with an additive effect of 2.5 mL/kg/min for both together [4]. The change in Minnesota Living with Heart Failure score with diet was minus 6 units (95% CI minus 12 to 1; P=0.08), which was not statistically significant [4].

How it compares with semaglutide and SGLT2 inhibitors

STEP-HFpEF assigned 529 adults with HFpEF and a body-mass index of at least 30 to semaglutide 2.4 mg weekly or placebo for 52 weeks, with KCCQ-CSS change and weight change as the two primary endpoints [5]. KCCQ-CSS improved by 7.8 points more than placebo (95% CI 4.8 to 10.9), body weight changed minus 13.3% versus minus 2.6%, six-minute walk distance improved by 20.3 m, and a hierarchical composite favored semaglutide (win ratio 1.72, 95% CI 1.37 to 2.15) [5]. SUMMIT's primary endpoints were a time-to-first-event composite and KCCQ-CSS, a different pair from STEP-HFpEF's [2][5].

The SGLT2 inhibitor trials used broader entry criteria. EMPEROR-Preserved assigned 5,988 patients with an ejection fraction above 40%; cardiovascular death or heart failure hospitalization occurred in 13.8% with empagliflozin and 17.1% with placebo, hazard ratio 0.79 (95% CI 0.69 to 0.90) [6]. DELIVER assigned 6,263 patients with an ejection fraction above 40%; the primary outcome occurred in 16.4% with dapagliflozin and 19.5% with placebo, hazard ratio 0.82 (95% CI 0.73 to 0.92) [7]. SUMMIT, by contrast, required an ejection fraction of at least 50% and a body-mass index of at least 30 [2].

What about heart failure with reduced ejection fraction?

Tirzepatide was not tested in reduced ejection fraction in SUMMIT; trials of liraglutide, a different GLP-1 drug, exist in this setting. The LIVE trial randomized 241 patients with an ejection fraction of 45% or less to liraglutide 1.8 mg daily or placebo for 24 weeks: ejection fraction did not differ between groups (mean difference minus 0.8%, 95% CI minus 2.1 to 0.5; P=0.24), heart rate rose by 7 beats per minute, and serious cardiac events occurred in 12 patients (10%) on liraglutide versus 3 (3%) on placebo (P=0.04) [8]. A separate 300-patient trial of liraglutide after hospitalization for acute heart failure found no significant effect on its primary endpoint, with deaths in 19 (12%) versus 16 (11%) [9].