Cardiovascular Outcomes of Tirzepatide vs Semaglutide in Obesity Without Type 2 Diabetes: A Matched Cohort Study

Technical Summary: Cardiovascular Outcomes of Tirzepatide vs. Semaglutide in Obesity Without Type 2 Diabetes

Problem Statement
Obesity is a primary driver of cardiovascular disease (CVD) and premature mortality. While incretin-based therapies like semaglutide (a selective GLP-1 receptor agonist) and tirzepatide (a dual GIP/GLP-1 receptor agonist) have transformed obesity management, a critical knowledge gap remains regarding their comparative cardiovascular safety and efficacy in patients with obesity but without type 2 diabetes (T2D).

Existing evidence establishes that semaglutide reduces major adverse cardiovascular events (MACE) in this population (SELECT trial). Furthermore, head-to-head trials (SURMOUNT-5) demonstrate that tirzepatide achieves superior weight loss compared to semaglutide. However, no randomized controlled trial (RCT) has directly compared the cardiovascular outcomes of these two agents in adults with obesity without T2D. The SURMOUNT-MMO trial, which aims to address this, is not expected to report until 2027. Consequently, it is unknown whether the dual receptor agonism of tirzepatide confers incremental cardiovascular protection beyond that of selective GLP-1 agonism.

Methodology
This study employed a retrospective, active-comparator, propensity score–matched cohort design using data from the TriNetX Global Collaborative Network, which aggregates de-identified electronic health records (EHR) from 178 healthcare organizations.

• Study Population: The cohort included adults aged 18–80 years with obesity (BMI ≥30 kg/m²) initiating either tirzepatide or subcutaneous semaglutide between January 1 and December 31, 2024.
• Exclusion Criteria: Patients with prior T2D (ICD-10 E11.x or HbA1c ≥6.5%), type 1 diabetes, end-stage renal disease, active malignancy, pregnancy, or prior use of GLP-1/GIP agonists were excluded.
• Matching: A 1:1 nearest-neighbor propensity score matching (caliper 0.1 SD) was performed using 12-month lookback covariates, including demographics, BMI, HbA1c, eGFR, comorbidities (hypertension, dyslipidemia, prior CVD), and concomitant medications. This resulted in 54,978 matched pairs (109,956 total patients).
• Study Design & Landmark: A 6-month landmark analysis was applied to minimize bias from early non-adherence or baseline severity. Outcomes were assessed at two horizons: 6–12 months and 6–18 months post-index date.
• Outcomes:
  • Primary: A modified 3-point MACE composite (acute myocardial infarction, ischemic stroke, or all-cause mortality). All-cause mortality was used in lieu of cardiovascular death due to the lack of national death registry linkage in TriNetX.
  • Secondary: Incident atrial fibrillation (AF), acute MI, ischemic stroke, and a chronic kidney disease (CKD) progression composite (new-onset CKD stage 3–5 or dialysis dependence).
  • Negative Controls: Hernia and age-related cataract were prespecified to detect residual confounding.
• Statistical Analysis: Cox proportional hazards regression estimated hazard ratios (HRs). Sensitivity analyses included E-values to quantify the strength of unmeasured confounding required to nullify the results and BMI-stratified subgroup analyses.

Key Results
After matching, the baseline characteristics of the two groups were well-balanced (mean age ~46.6 years, 74% female, mean BMI 39.5 kg/m²).

• Primary Outcome (MACE): Tirzepatide was associated with a significantly lower risk of MACE compared to semaglutide.
  • At 6–12 months: HR 0.80 (95% CI 0.69–0.92), representing a 20% relative risk reduction.
  • At 6–18 months: HR 0.84 (95% CI 0.75–0.94), representing a 16% relative risk reduction.
• Secondary Outcomes:
  • Incident Atrial Fibrillation: Significant reductions were observed at both time points (HR 0.83 at 12 months; HR 0.84 at 18 months).
  • CKD Progression: Significant reductions were observed (HR 0.84 at 12 months; HR 0.82 at 18 months).
  • Ischemic Stroke: A significant reduction was observed at 12 months (HR 0.81), though this attenuated and became non-significant at 18 months (HR 0.90).
  • Acute MI and All-Cause Mortality: Both showed directionally concordant reductions favoring tirzepatide (HR 0.83 for MI; HR 0.75 for mortality at 18 months) but did not reach statistical significance, likely due to low event rates.
• Robustness Checks:
  • Negative Controls: Both prespecified negative controls (hernia and cataract) showed null associations (HRs ~1.00), suggesting no systematic bias favoring tirzepatide.
  • E-Values: The E-value for the primary 12-month MACE point estimate was 1.81. This indicates that an unmeasured confounder would need to be associated with both the treatment and the outcome with a risk ratio of at least 1.81 to fully explain away the observed effect.
  • Subgroup Analysis: The benefit of tirzepatide was consistent across obesity classes I, II, and III.

Significance and Claims
The authors conclude that in adults with obesity without T2D, tirzepatide is associated with lower risks of MACE, incident AF, and CKD progression compared to semaglutide.

The paper posits that these findings generate the hypothesis that dual GIP/GLP-1 receptor agonism may confer cardiovascular benefits beyond those achieved by selective GLP-1 receptor agonism alone. The authors suggest that the superior weight loss efficacy of tirzepatide (as seen in SURMOUNT-5) may mediate these incremental benefits, though the study design cannot distinguish between weight-dependent and weight-independent mechanisms.

The authors explicitly state that these findings are observational and require confirmation in randomized trials before they can guide clinical decision-making regarding drug-class distinctions. They acknowledge limitations including the inability to reliably ascertain cardiovascular death, the use of ICD-10 coding for CKD progression (which may include mild cases), and the potential for residual confounding despite rigorous matching and negative control validation. The study does not claim definitive superiority but rather provides real-world evidence supporting the need for a dedicated cardiovascular outcomes trial comparing these agents directly in this specific population.