Comparative Effectiveness of Tirzepatide and Semaglutide in Individuals at Cardiovascular Risk

This is a non-randomized, non-interventional study that is part of the Randomized Controlled Trials Duplicated Using Prospective Longitudinal Insurance Claims: Applying Techniques of Epidemiology (RCT-DUPLICATE) initiative (www.rctduplicate.org) of the Brigham and Women's Hospital, Harvard Medical School. It is intended to assess the comparative effectiveness of

1. Tirzepatide vs dulaglutide,
2. Semaglutide vs sitagliptin,
3. Tirzepatide vs semaglutide

on cardiovascular outcomes in individuals typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes (T2DM) and overweight but might not meet the eligibility criteria of pivotal RCTs for each drug (SUSTAIN-6 and SURPASS-CVOT trials), used to support regulatory approval in patients at cardiovascular risk.Although many features of the target trials cannot be directly replicated in healthcare claims, key design features, including outcomes, exposures, and inclusion/exclusion criteria, were selected to proxy those features from the target trial. Randomization cannot be achieved in healthcare claims data but was proxied through a statistical balancing of measured covariates according to standard practice.

The three database studies will be new-user active-comparative studies, conducted using 3 national United States claims databases, where investigators compare the effect of semaglutide vs sitagliptin (used as an active comparator placebo proxy), tirzepatide vs dulaglutide, and tirzepatide vs semaglutide on the composite end point of all-cause mortality, myocardial infarction, or stroke. Clinical guidelines during the study period recommended both tirzepatide and semaglutide for the same indications of glucose lowering and weight reduction.

• Tirzepatide (Drug): New use of tirzepatide dispensing claim is used as the exposure.
• Dulaglutide (Drug): New use of dulaglutide dispensing claim is used as the comparator.
• Semaglutide (Drug): New use of semaglutide dispensing claim is used as the exposure/comparator.
• Sitagliptin (Drug): New use of sitagliptin dispensing claim is used as the comparator.

Primary outcomes

• Composite of all-cause mortality, myocardial infarction or stroke (Tirzepatide vs. dulaglutide) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the comparative effect of tirzepatide vs dulaglutide on the composite of all-cause mortality, myocardial infarction, or death in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Composite of all-cause mortality, myocardial infarction or stroke (Injectable semaglutide vs sitagliptin) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the comparative effect of injectable semaglutide vs sitagliptin on the composite of all-cause mortality, myocardial infarction, or death in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Composite of all-cause mortality, myocardial infarction or stroke (Tirzepatide vs injectable semaglutide) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the comparative effect of tirzepatide vs semaglutide on the composite of all-cause mortality, myocardial infarction, or death in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.

Secondary outcomes

• Individual components of the primary endpoint, i.e., all-cause mortality, myocardial infarction, or stroke (Tirzepatide vs dulaglutide) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the comparative effect of tirzepatide vs dulaglutide on the individual components of the primary endpoint, i.e., all-cause mortality, myocardial infarction, or stroke in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Composite of all-cause mortality, hospitalization for heart failure, or urgent heart failure visits requiring intravenous diuretics (Tirzepatide vs dulaglutide) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the comparative effect of tirzepatide vs dulaglutide on all-cause mortality, hospitalization for heart failure, or urgent heart failure visits in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Urinary tract infections (Tirzepatide vs dulaglutide) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the comparative effect of tirzepatide vs dulaglutide on the safety outcome of urinary tract infections in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Serious bacterial infections (Tirzepatide vs dulaglutide) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the comparative effect of tirzepatide vs dulaglutide on the safety outcome of serious bacterial infections in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Gastrointestinal adverse events (Tirzepatide vs dulaglutide) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the comparative effect of tirzepatide vs dulaglutide on the safety outcome of gastrointestinal adverse events in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Individual components of the primary endpoint, i.e., all-cause mortality, myocardial infarction, or stroke (Injectable semaglutide vs sitagliptin) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the comparative effect of injectable semaglutide vs sitagliptin on the individual components of the primary endpoint, i.e., all-cause mortality, myocardial infarction, or stroke in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Composite of all-cause mortality, hospitalization for heart failure, or urgent heart failure visits requiring intravenous diuretics (Injectable semaglutide vs sitagliptin) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the comparative effect of injectable semaglutide vs sitagliptin on all-cause mortality, hospitalization for heart failure, or urgent heart failure visits in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Urinary tract infections (Injectable semaglutide vs sitagliptin) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the comparative effect of injectable semaglutide vs sitagliptin on the safety outcome of urinary tract infections in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Serious bacterial infections (Injectable semaglutide vs sitagliptin) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the comparative effect of injectable semaglutide vs sitagliptin on the safety outcome of serious bacterial infections in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Gastrointestinal adverse events (Injectable semaglutide vs sitagliptin) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the comparative effect of injectable semaglutide vs sitagliptin on the safety outcome of gastrointestinal adverse events in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Individual components of the primary endpoint, i.e., all-cause mortality, myocardial infarction, or stroke (Tirzepatide vs injectable semaglutide) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the comparative effect of tirzepatide vs injectable semaglutide on the individual components of the primary endpoint, i.e., all-cause mortality, myocardial infarction, or stroke in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Composite of all-cause mortality, hospitalization for heart failure, or urgent heart failure visits requiring intravenous diuretics (Tirzepatide vs injectable semaglutide) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the comparative effect of tirzepatide vs injectable semaglutide on all-cause mortality, hospitalization for heart failure, or urgent heart failure visits in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Urinary tract infections (Tirzepatide vs injectable semaglutide) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the comparative effect of tirzepatide vs injectable semaglutide on the safety outcome of urinary tract infections in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Serious bacterial infections (Tirzepatide vs injectable semaglutide) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the comparative effect of tirzepatide vs injectable semaglutide on the safety outcome of serious bacterial infections in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Gastrointestinal adverse events (Tirzepatide vs injectable semaglutide) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the comparative effect of tirzepatide vs injectable semaglutide on the safety outcome of gastrointestinal adverse events in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.

Other outcomes

• Hernia (Tirzepatide vs dulaglutide) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the effect of tirzepatide vs dulaglutide on negative control outcomes, including hernia in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Lumbar radiculopathy (Tirzepatide vs dulaglutide) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the effect of tirzepatide vs dulaglutide on negative control outcomes, including lumbar radiculopathy in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Hernia (Injectable semaglutide vs sitagliptin) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the effect of semaglutide vs sitagliptin on negative control outcomes, including hernia in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Lumbar radiculopathy (Injectable semaglutide vs sitagliptin) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the effect of semaglutide vs sitagliptin on negative control outcomes, including lumbar radiculopathy in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Hernia (Tirzepatide vs injectable semaglutide) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the effect of tirzepatide vs injectable semaglutide on negative control outcomes, including hernia, in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.
• Lumbar radiculopathy (Tirzepatide vs injectable semaglutide) [1 day after cohort entry date until the first of outcome or censoring, up to 365 days]: To evaluate the effect of tirzepatide vs injectable semaglutide on negative control outcomes, including lumbar radiculopathy, in patients typically treated in clinical practice who are at cardiovascular risk with type 2 diabetes and overweight.

Sex: All, ages: 18 Years to any

ELIGIBILITY FOR TIRZEPATIDE VS DULAGLUTIDE

Inclusion criteria

* History of MI, stroke, any surgical or percutaneous revascularization procedure, use of any antihypertensive/lipid-lowering drugs, coronary/carotid/peripheral artery disease, hypertension
* Type 2 diabetes
* BMI ≥25.0kg/m2
* Age ≥18 years
* Male or female sex

Exclusion Criteria:

* Medullary thyroid carcinoma, MEN syndrome type 2
* Malignancy
* Type 1 diabetes or secondary diabetes
* Chronic kidney disease or dialysis
* Uncontrolled diabetic retinopathy or maculopathy
* Pregnancy
* Prior use of pramlintide or any GLP-1-RA

ELIGIBILITY FOR SEMAGLUTIDE VS SITAGLIPTIN

Inclusion Criteria:

* History of MI, stroke, any surgical or percutaneous revascularization procedure, use of any antihypertensive/lipid-lowering drugs, coronary/carotid/peripheral artery disease, hypertension
* Type 2 diabetes
* BMI ≥25.0kg/m2
* Age ≥18 years
* Male or female sex

Exclusion Criteria:

* Medullary thyroid carcinoma, MEN syndrome type 2
* Malignancy
* Type 1 diabetes or secondary diabetes
* Chronic kidney disease or dialysis
* Uncontrolled diabetic retinopathy or maculopathy
* Pregnancy
* Prior use of pramlintide or any GLP-1-RA or DPP4i

ELIGIBILITY FOR TIRZEPATIDE VS SEMAGLUTIDE

Inclusion Criteria:

* History of MI, stroke, any surgical or percutaneous revascularization procedure, use of any antihypertensive/lipid-lowering drugs, coronary/carotid/peripheral artery disease, hypertension
* Type 2 diabetes
* BMI ≥25.0kg/m2
* Age ≥18 years
* Male or female sex

Exclusion Criteria:

* Medullary thyroid carcinoma, MEN syndrome type 2
* Malignancy
* Type 1 diabetes or secondary diabetes
* Chronic kidney disease or dialysis
* Uncontrolled diabetic retinopathy or maculopathy
* Pregnancy
* Prior use of pramlintide or any GLP-1-RA

• Results publication: Kruger N, Schneeweiss S, Desai RJ, Sreedhara SK, Kehoe AR, Fuse K, Hahn G, Schunkert H, Wang SV. Cardiovascular outcomes of semaglutide and tirzepatide for patients with type 2 diabetes in clinical practice. Nat Med. 2026 Jan;32(1):342-352. doi: 10.1038/s41591-025-04102-x. Epub 2025 Nov 9. PMID 41207920

• Study Protocol and Statistical Analysis Plan: Study Protocol and Statistical Analysis Plan (PDF, 2025-08-10)
• Study Protocol and Statistical Analysis Plan: Amendment to Study Protocol and Statistical Analysis Plan (PDF, 2025-10-05)

• Shirley Wang, PhD, ScM, Principal investigator, Brigham and Women's Hospital
• Nils Krüger, MD, Principal investigator, Brigham and Women's Hospital