Can Dual Incretin Receptor Agonists Exert Better Cardiovascular Protection than Selective GLP-1 Receptor Agonists? Highlights from SURPASS-CVOT

Abstract

Despite advances in cardiovascular risk reduction in type 2 diabetes (T2D), a persistent gap remains compared to individuals without diabetes. Glucagon-like peptide-1 receptor agonists (GLP-1RA) have provided consistent cardiovascular benefits. With more cardiovascular protective agents available for diabetes management, their incremental effect may be nearing a ceiling. The SURPASS-CVOT trial innovatively compared the dual GIP/GLP-1RA tirzepatide with the selective GLP-1RA dulaglutide, demonstrating noninferiority for major adverse cardiovascular events (MACE; HR 0.92; 95.3% CI 0.83–1.01; p = 0.086) and suggesting a potential 28% MACE risk reduction versus an imputed placebo. However, superiority over dulaglutide was narrowly missed. Despite greater improvements in glycemia (0.8% greater HbA1c reduction) and weight (7% greater weight loss), tirzepatide appeared to confer limited incremental cardiovascular benefit, raising questions about mechanism saturation or trial design constraints. Exploratory analyses showed promising benefits on mortality and renal function but require cautious interpretation. The trial's active comparator/imputed placebo design reflects an evolving ethical and therapeutic landscape in diabetes care. Whether dual incretin receptor agonism can meaningfully exceed current cardioprotective thresholds remains uncertain. By now, we may need new paradigms to overcome what may turn out to be a therapeutic ceiling for cardiovascular protection in the T2D population.

Opportunities for Reduction of Residual Risk in Type 2 Diabetes

Type 2 diabetes (T2D) remains one of the most potent accelerators of cardiovascular disease (CVD), even as therapeutic strategies have evolved beyond glycemic control alone. CVD mortality has declined in people with and without diabetes in the last decades, but the relative reductions were smaller in adults with diabetes, largely preserving the excess CVD risk gap [1].

The advent of glucagon-like peptide-1 receptor agonists (GLP-1RA) has transformed diabetes care, not only through improvements in HbA1c and body weight, but by demonstrating a consistent reduction in major adverse cardiovascular events (MACE).

Yet these benefits, though significant, may be approaching a ceiling: in REWIND, dulaglutide achieved a hazard ratio (HR) of 0.88 for MACE compared to placebo, but with only a 2% absolute risk reduction over a median of 5.4 years [2]. Compounding this plateau, placebo-controlled cardiovascular outcome trials (CVOTs) are increasingly deemed unethical in patients with established CVD, as current standards of care now include agents with proven cardiovascular efficacy, namely GLP-1RA and SGLT-2 inhibitors. Though the diabetes-attributable risk of MACE do not appear to be decreasing in the last decades, it is expected that the absolute rates of events will decline owing to an expanding proportion of patients receiving cardiorenal protective drugs and to the adherence to guideline-recommended targets for blood pressure and serum lipids. Notably, the most recent CVOT on a GLP-1RA shows that oral semaglutide can further reduce the risk of MACE even in a population with a high uptake of SGLT2 inhibitors [3]. Several meta-analysis, however, seem to imply that the mean effect on MACE with various strategies and drug classes (GLP-1RA; SGLT2 inhibitors, or PCKS9 inhibitors) hardly exceeds 15% relative risk reduction [4].

The question thus emerges: can we enhance cardioprotection beyond what is offered by GLP-1RA alone, and how do we evaluate this in a changing therapeutic and ethical landscape?

Ethical Approval

No ethical approval was sought as this article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by the author.

An Innovative Trial Design in SURPASS-CVOT

SURPASS-CVOT broke new ground as the first CVOT to compare two incretin-based therapies directly: the dual GIP/GLP-1RA tirzepatide versus the selective GLP-1RA dulaglutide. The trial was designed not only to test tirzepatide versus dulaglutide but also to assess the superiority of tirzepatide relative to an imputed placebo, the effect of which was derived from meta-analytic extrapolation of previous GLP-1RA trials [5].

This methodological pivot reflects a pragmatic shift. Rather than randomizing individuals with T2D and established CVD to placebo, participants received either tirzepatide or dulaglutide, with the assumption that dulaglutide preserved a cardiovascular benefit of at least 12% over placebo as observed in REWIND. The primary analysis tested noninferiority versus dulaglutide, to an upper limit of the 95% CI of the HR of below 1.05, supporting the superiority of tirzepatide versus putative placebo. An upper limit of the 95% CI of the HR of ≥ 1.05 but < 1.23 for tirzepatide versus dulaglutide would indicate noninferiority and proven safety of tirzepatide to putative placebo. The noninferiority margin of 1.05 was pre-specified and statistically conservative, preserving at least 50% of dulaglutide’s efficacy.

This trial design is bold enough to lead a new generation of CVOTs, but it introduces interpretive complexity. Imputed placebo comparisons rely on assumptions of stability in background therapies, population risk, and class effects. With more and more cardioprotective agents on board and with the pursuit of ambitious targets, further risk reduction may be harder to achieve.

Pathophysiological Basis

The hypothesis that tirzepatide might exceed GLP-1RA efficacy stems from dual incretin agonism. Tirzepatide activates both GIP and GLP-1 receptors, theoretically leveraging synergistic effects on weight loss, glycemic control, and metabolic modulation. In a 26-week phase 2 trial, tirzepatide at doses of 10 mg and 15 mg reduced HbA1c by 1.89% and 1.94%, respectively, compared to 1.21% with dulaglutide 1.5 mg [6]. Weight loss was similarly amplified: tirzepatide achieved reductions of up to 11.3 kg versus only 2.7 kg with dulaglutide. These benefits extended to secondary markers, including fasting glucose, waist circumference, triglyceride levels, and inflammatory molecules. These comprehensive effects may translate into cardiovascular benefits through multiple mechanisms. Both GLP-1 and GIP participate in nutrient-dependent cardiovascular homeostasis [7]. They promote endothelial nitric oxide production, enhance vasodilation, and improve myocardial substrate use, especially during ischemia. Additionally, GLP-1 exhibits mild inotropic effects and anti-inflammatory properties, potentially mitigating atherosclerosis. Dual agonism may amplify these effects, especially under postprandial conditions when nutrient and hormonal fluxes are highest. However, these hypotheses remain speculative. It is uncertain whether GIP contributes an additive or synergistic cardiovascular benefit beyond what GLP-1RA alone can achieve.

While awaiting the results of SURPASS-CVOT, real-world studies have provided some evidence supporting a greater cardiovascular protection achievable with tirzepatide. In two retrospective cohort studies of electronic medical records of individuals with T2D, initiation of tirzepatide was followed by significantly lower risk of a composite outcome including all-cause mortality, cerebral infarction, acute coronary syndrome, and heart failure: the HR was 0.54 versus semaglutide [8] and 0.60 versus the GLP-1RA class [9].

Topline Results of SURPASS-CVOT

SURPASS-CVOT enrolled 13,299 participants with T2D (28.9% female), with a mean age of 64.1 years and established CVD (65% coronary artery disease, 19% prior stroke, 25% peripheral arterial disease). Thirty percent of the participants were already on an SGLT2 inhibitor at baseline [10].

The topline findings from SURPASS-CVOT, disclosed before publication of the main manuscript [11], reported a hazard ratio of 0.92 (95.3% CI 0.83–1.01) for 3P-MACE in favor of tirzepatide, confirming noninferiority to dulaglutide. Some secondary analyses had nominal significance in favor of tirzepatide, with a 16% reduction in all-cause mortality versus dulaglutide (HR 0.84; 95% CI 0.75–0.94), and renal benefit in patients at high CKD risk, showcased by eGFR decline slowed by 3.54 ml/min/1.73 m² over 36 months. Yet, these analyses were not adjusted for multiplicity of testing and must be considered hypothesis-generating and interpreted with caution. No formal claim of superiority can be derived from these secondary endpoints under the original statistical plan, especially from regulatory standpoints. Notably, however, the pre-specified indirect comparison with a putative placebo suggested a significant 28% reduction in MACE (HR 0.72; 95% CI 0.55–0.94) and 39% reduction in all-cause mortality (HR 0.61; 95% CI 0.45–0.82), which projects tirzepatide among the most effective diabetes drugs for outcome prevention in people with established atherosclerosis (Fig. 1).

Fig. 1.

Summary of the SURPASS-CVOT results disclosed in topline announcement. NI noninferiority, Plb placebo, MACE major adverse cardiovascular events (non-fatal myocardial infarction or stroke, or cardiovascular mortality)

Clinical Interpretation

SURPASS-CVOT failed by a small amount to demonstrate superiority of tirzepatide over dulaglutide on the primary outcome but indicated a signal that warrants further exploration. At present, it is challenging to explain why achieving greater HbA1c (− 0.8%) and weight (− 7%) reductions adds little to cardiovascular protection [11]. When combined with the robust metabolic data from phase 2 and 3 trials, and the pathophysiological rationale rooted in nutrient-cardiovascular signaling, the case for dual agonism remains scientifically plausible. The new trial design itself may mark a turning point. Active comparator trials with imputed placebo could become the new standard in diabetes CVOTs, provided their statistical assumptions are rigorously justified. In fact, such a design incorporates substantial uncertainty if background therapy, population risk, or adherence differs significantly from historical trials, especially in view of the rapidly changing scenario of T2D management with a treat-to-benefit approach. Such uncertainty may have complicated the trial design, making its assumptions less reliable and ultimately limiting power. There are other possible reasons why statistical significance was narrowly missed. Given the potent glycemic effect of tirzepatide, more SGLT2 inhibitors may have been introduced in the dulaglutide group for equipoise, reducing the margin to see a significant effect. Dulaglutide may have proven more effective (estimated HR ~ 0.78 versus a putative placebo) than expected in a population of participants different than those enrolled in the REWIND trial (HR 0.88). The cardiovascular benefits of GLP-1RA observed across trials seem unrelated to doses and duration of action, suggesting a saturation effect. Thus, adding GIP receptor agonism may converge on mechanisms of vascular protection that are already maximally stimulated via the GLP-1 receptor, thereby yielding a marginal further potentiation. Alternatively, it may be necessary to wait longer for those added benefits in terms of HbA1c and body weight to translate into a further improvement in cardiovascular outcome.

If we are approaching a ceiling in the cardiovascular protection in diabetes, residual risk may be driven by entirely different mechanisms, including genetic drivers that are poorly modifiable. Also, with the changing pattern of mortality causes in diabetes, from cardiovascular to non-vascular (12), all-cause mortality may emerge as a new reasonable target for comprehensive diabetes management. Although exploratory, the superiority of tirzepatide versus dulaglutide on total mortality is a particularly appealing result.

In conclusion, in the face of a clear metabolic superiority, tirzepatide has demonstrated noninferiority for cardiovascular outcomes compared to GLP-1RA. New signals from SURPASS-CVOT are being explored as initial evidence that dual incretin receptor agonists may exert additional protection from adverse cardio-renal outcomes of diabetes. Unfortunately, this will remain speculative until the next active-comparator trial within the broad incretin family.

Author Contribution

Gian Paolo Fadini conceived the study, collected and analyzed the data, and wrote the manuscript.

Funding

No funding or sponsorship was received for this study or publication of this article.

Declarations

Conflict of Interest

Gian Paolo Fadini received honoraria, consultancy or lecture fees from AstraZeneca, Boehringer, Guidotti, Lilly, Novartis, and Sanofi. GPF is an Editorial Board member of Diabetes Therapy. GPF was not involved in the selection of peer reviewers for the manuscript nor any of the subsequent editorial decisions.

Ethical Approval

No ethical approval was sought as this article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by the author.