Introduction

Cardiovascular trials of the incretin-based medicines tirzepatide and semaglutide have demonstrated benefits across patients with diverse risk profile. Yet, in the absence of direct head-to-head comparisons, clinicians’ decision relies on indirect data from varied populations, making it challenging to determine the most effective treatment option for individual patients.

Aim

To ascertain the effectiveness of tirzepatide and semaglutide in patients having high cardiovascular risk, including obesity and type-2 diabetes mellitus (T2DM).

Patient Population

• The study population comprised of subjects enrolled in insurance programs in the United States between 2018 and 2025.
• Patients had low, moderate, or high cardiovascular risk and were diagnosed with obesity and T2DM. 
• Subgroups of patients with specific cardiovascular conditions such as coronary heart disease, cardiomyopathies, valve disorders, heart failure, as well as patients under high-intensive lipid-lowering therapy were also included.

Methods 

Study Design

• Two cardiovascular outcome trials, SUSTAIN-6 (semaglutide vs. sitagliptin as placebo proxy) and SURPASS-CVOT (tirzepatide vs. dulaglutide) were emulated to scale and critically evaluate the design, data, and analytic framework. 
• The efficacy of each drug was assessed in expanded populations representing patients routinely seen in clinical practice. Tirzepatide was directly compared with semaglutide, baseline confounders were balanced using propensity score matching.

Outcomes

Primary Composite Outcome

Results

• Overall, 158,310 patients met the eligibility criteria for the emulation of SUSTAIN-6 trial and 44,671 patients for the emulation of SURPASS-CVOT trial.
• A total of 453,201 individuals initiating semaglutide or sitagliptin (expanding SUSTAIN-6 eligibility) and 136,089 initiating tirzepatide or dulaglutide (expanding SURPASS-CVOT eligibility) were identified to represent a broader patient population typically encountered in clinical practice. 
• A total of 297,842 initiators were considered for the head-to-head comparison of tirzepatide vs. semaglutide. 
• The mean age for the matched expanded population ranged from 59.2 to 69.2 years, 50.4% to 55.8% were women. The mean body mass index (BMI) of the study population ranged from 34.5 to 38.7 kg/m².
• Approximately 2.9% to 9.3% patients had a history of prior myocardial infarction (MI) or stroke, and 18.9% to 36.7% of patients had chronic kidney disease (CKD).
• In the emulation of SUSTAIN-6 comparing semaglutide vs. sitagliptin (a proxy for placebo), patients at moderate and high cardiovascular risk had a 32% reduction in the risk of the primary outcome [hazard ratio (HR): 0.68, 95% confidence interval (CI): 0.60 to 0.77] as compared to the trial estimate of 26% risk reduction (HR: 0.74, 95% CI, 0.58 to 0.95). The four-agreement metrics were met (Table 1). 
• Analysis of the individual components of the primary outcome revealed closely concordant results for MI and stroke. Nevertheless, the results for all-cause mortality were divergent, indicating a residual confounding. 
• Secondary outcomes in both the trial emulations and their reference trials lacked sufficient statistical power to confirm agreement.
• The emulation of SURPASS-CVOT compared tirzepatide vs. dulaglutide in patients at high cardiovascular risk. Treatment with tirzepatide reduced the risk of primary outcome by 17% (HR: 0.83, 95% CI, 0.69 to 1.01) as against the trial estimates of an 8% risk reduction (HR: 0.92, 95% CI, 0.83 to 1.01). The benchmarking confirmed all agreement metrics, for both the primary composite outcome and all-cause mortality, supporting the validity of this approach.

Table 1: Benchmarking of results from trial emulations vs. reference trials

	Outcome	Trial emulation estimate	Trial estimate	SA	DA	EA	SD
SUSTAIN-6	Major adverse cardiovascular events (MACE)	0.68 (0.60 to 0.77)	0.74 (0.58 to 0.95)	Yes	Yes	Yes	Yes
	MI	0.70 (0.57 to 0.86)	0.74 (0.51–1.08)	N/A*	Yes	Yes	Yes
	Stroke	0.82 (0.65 to 1.03)	0.61 (0.38–0.99)	N/A*	Yes	Yes	Yes
	All-cause mortality	0.58 (0.48 to 0.71)	1.05 (0.74 to 1.50)	N/A*	No	No	No
SURPASS-CVOT	MACE	0.83 (0.69 to 1.01)	0.92 (0.83 to 1.01)	Yes	Yes	Yes	Yes
	MI	0.81 (0.61 to 1.06)	0.86 (0.74 to 1.00)	N/A*	Yes	Yes	Yes
	Stroke	0.92 (0.65 to 1.29)	0.91 (0.76 to 1.09)	N/A*	Yes	Yes	Yes
	All-cause mortality	0.76 (0.52 to 1.11	0.84 (0.75 to 0.94)	N/A*	Yes	Yes	Yes

DA: directional agreement; EA: estimate agreement; N/A: not applicable; SA: statistical agreement; SD: standardized difference agreement.

• A protocol of the expansion study comparing semaglutide vs. sitagliptin was amended based on the data benchmarks to target the end points that did not include death from any cause. A composite end point of MI or stroke without death of any cause was additionally included. A composite end point of hospitalization for heart failure or urgent care visit requiring intravenous diuretics (excluding death) was added similarly.
• The endpoints for the comparison of tirzepatide vs. dulaglutide remained unchanged in the expanded population. 
• Considering that the confounding structure for the head-to-head comparison of tirzepatide vs. semaglutide would be more similar to the tirzepatide vs. dulaglutide benchmarking study, a pre-specified analysis that included death in the primary composite major adverse cardiovascular event (MACE) outcome was also conducted. 

Primary End Point in Expanded Populations and High-Risk Subgroups

Semaglutide vs. Sitagliptin

• In patients with obesity and T2DM across cardiovascular risk levels, semaglutide reduced the 1-year risk of composite endpoint of 1-year MI or stroke by 18%, as compared to sitagliptin [1.5% (95% CI: 1.4–1.6%) vs. 1.7% (95% CI: 1.6–1.9%); risk difference: -0.3%; HR: 0.82; 95% CI: 0.74–0.91). 
• Mean follow-up was ~193 days; 46% of the patients discontinued the treatment. Similar benefit was seen in high-risk patients (HR 0.80; 95% CI: 0.71–0.91).

Tirzepatide vs. Dulaglutide

• In an expanded cohort, tirzepatide users had a 1-year primary endpoint risk (including all-cause mortality) of 1.4% (95% CI: 1.3–1.6%) vs. 1.8% (95% CI: 1.5–2.0%) for dulaglutide, with a risk difference of -0.3% (95% CI: -0.6 to 0.04%). This translated into a 13% risk reduction (HR: 0.87; 95% CI: 0.75–1.01). 
• Mean on-treatment follow-up was 189 days for tirzepatide and 173 days for dulaglutide. Treatment discontinuation occurred in 37% of cases. In high cardiovascular risk patients, the HR was 0.88 (95% CI: 0.73–1.07), suggesting a consistent trend in favor of tirzepatide, though not statistically significant.

Tirzepatide vs. Semaglutide

• In a direct comparison, tirzepatide and semaglutide both showed a 1-year primary endpoint risk of 1.3% (95% CI: 1.2–1.5%), with a risk difference of 0.0% (95% CI: -0.2 to 0.2%) and HR 1.06 (95% CI: 0.95–1.18). Mean on-treatment follow-up was 181 days for tirzepatide group and 174 days for semaglutide group. 
• Overall, 34% of the patients discontinued the treatment. In high cardiovascular risk individuals, the HR was 1.11 (95% CI: 0.96–1.27), indicating no significant difference in outcomes.

Secondary, Safety, and Negative Control End Points

• Semaglutide significantly reduced the risk of MI (HR: 0.81; 95% CI: 0.70–0.92) and stroke (HR: 0.84; 95% CI: 0.71–0.99) vs. sitagliptin across cardiovascular risk levels. 
• Tirzepatide showed non-inferior reductions in the risk of all-cause mortality (HR: 0.88), MI (HR: 0.91), and stroke (HR: 0.78) vs. dulaglutide, though CIs included no difference. 
• Head-to-head, tirzepatide and semaglutide had comparable risks for mortality (HR: 1.03; 95% CI 0.84 to 1.27), MI (HR: 1.03; 95% CI 0.88 to 1.21), and stroke (HR: 1.15; 95% CI 0.92 to 1.45).
• In the expanded populations, treatment with semaglutide was associated with a 39% lower risk of the secondary composite end point of heart failure hospitalization or urgent heart failure visit, as compared to sitagliptin (HR: 0.61; 95% CI, 0.57 to 0.66). 
• Tirzepatide lowered the risk of the secondary composite end point in heart failure hospitalization, urgent heart failure visit, or all-cause mortality, as compared to dulaglutide by 25% (HR: 0.75; 95% CI, 0.65 to 0.86). 
• With regards to tirzepatide vs. semaglutide, contrary to the primary end point in the head-to-head comparison, tirzepatide had a modest benefit, although the 95% CI overlapped the null (HR: 0.91; 95% CI, 0.81 to 1.01).
• In the expanded cohort, semaglutide significantly reduced the risk of heart failure hospitalization or urgent visits vs. sitagliptin (HR: 0.61; 95% CI: 0.57–0.66). 
• Tirzepatide lowered the composite risk of heart failure events or all-cause mortality compared to dulaglutide (HR: 0.75; 95% CI: 0.65–0.86). 
• In the head-to-head analysis, tirzepatide showed a modest, non-significant benefit over semaglutide (HR 0.91; 95% CI: 0.81–1.01).
• The risk of urinary tract infections or gastrointestinal adverse events did not differ between the treatment groups. 
• Subgroup analyses showed no significant treatment effect heterogeneity by age. across sex subgroups, semaglutide vs. sitagliptin and tirzepatide vs. semaglutide had similar effects, while tirzepatide vs. dulaglutide suggested greater benefit in male patients. 
• SGLT2 inhibitor use at baseline did not alter treatment outcomes meaningfully.
• Sensitivity analyses using an ‘as-started’ approach showed modestly attenuated estimates. Results remained robust when adjusting for glycosylated hemoglobin (HbA1c). 
• A 2-year on-treatment analysis confirmed consistency with 1-year results, reinforcing the comparative effectiveness and durability of observed treatment effects over time.

Conclusions

• Treatment with semaglutide was associated with a lower risk of MACE, as compared to sitagliptin.
• The clinical benefits with tirzepatide were at least comparable to that seen with dulaglutide, an older GLP-1 receptor agonist with established cardiovascular efficacy. 
• The direct head-to-head comparison demonstrated that tirzepatide was beneficial as semaglutide in reducing MACE. 
• These findings offer valuable, timely evidence on the cardiovascular effectiveness of contemporary GLP-1 receptor agonist therapies, supporting clinical decision-making when head-to-head randomized trials are unavailable.

Nat Med. 2025 Nov 9 (Online ahead of print). doi: 10.1038/s41591-025-04102-x.