As the onslaught of anti-obesity drugs hit the market in recent years, more research is being released about their potential side effects as well as their benefits. Endocrine News investigates two of these studies to gain more insight to how these “miracle drugs” could affect cardiac health.
Two studies published in late 2025 examined cardiovascular outcomes among patients with type 2 diabetes treated with various glucagon-like peptide-1 receptor agonists (GLP-1RAs), providing real data suggesting that, although they clearly have multiple therapeutic benefits, not all agents within this drug class are created the same.
With recent federal policy changes aiming to make the GLP-1RAs and related drugs more affordable to potentially reach millions more patients who could see significant health improvements with treatment, this is a good time to scrutinize their safety and efficacy profiles. Collectively, these studies advance our understanding of within-class differences among GLP-1RAs, which may allow more individualized therapy, depending on patient-specific factors.
Tirzepatide Versus Dulaglutide
In “Cardiovascular Outcomes with Tirzepatide versus Dulaglutide in Type 2 Diabetes,” from NEJM in December,” Stephen J. Nicholls, MB, BS, PhD, of Monash University in Clayton, Australia, and team compared tirzepatide and dulaglutide head-to-head. They conducted the Study of Tirzepatide Compared with Dulaglutide on Major Cardiovascular Events in Participants with Type 2 Diabetes (SURPASS-CVOT) to settle a couple of questions, explains Nicholls: “Given that there is a lot of interest in developing new incretin therapies, can we achieve better outcomes (metabolic or clinical) with combination therapy, and can we develop agents that are better tolerated? For all of these agents, there will ultimately be interest in knowing their effect on cardiovascular events.”
For their active comparator–controlled, double-blind, noninferiority trial that began in 2020, 6,586 patients with type 2 diabetes were randomized to a tirzepatide (dual glucose-dependent insulinotropic polypeptide [GIP]/GLP-1RA) treatment group and 6,579 to a dulaglutide (selective GLP-1RA) group and followed for four years. The primary end point was time to first major cardiovascular adverse event (MACE), a composite of death from cardiovascular causes, myocardial infarction (MI), or stroke. Secondary outcomes included cardiovascular and all-cause death; additional composite cardiovascular outcomes (including coronary revascularization and heart-failure events), kidney function change over 36 months; and changes in A1c, weight, blood pressure (36 months), and triglycerides/low-density lipoprotein cholesterol (24 months). Tirzepatide proved as safe and effective as dulaglutide for major cardiovascular events (death, MI, or stroke) in high-risk patients with type 2 diabetes (noninferior, not clearly superior), with more gastrointestinal (GI) side effects, but better A1c and weight loss. However, the broader composite including revascularization was lower with tirzepatide, as was all-cause mortality.
Nicholls says they were not surprised that tirzepatide demonstrated noninferiority, but he underscores the significance of the latter findings. “While we expected that tirzepatide would have better metabolic benefits (weight loss, glycemic control),” he says, “we didn’t know if it would have better effects on clinical outcomes. The benefits on extended MACE including coronary revascularization and all-cause mortality are important. They highlight that people with CVD and type 2 diabetes are at risk of a range of complications. Tirzepatide fared well from that perspective.” Digging a little deeper into why and how tirzepatide prevented more deaths, Nicholls says, “the death benefit is really interesting and appears to involve a reduction in non-cardiovascular death. The degree to which that is less death in the setting of infection has been reported previously. We need to do more investigation.”
“Given that there is a lot of interest in developing new incretin therapies, can we achieve better outcomes (metabolic or clinical) with combination therapy, and can we develop agents that are better tolerated? For all of these agents, there will ultimately be interest in knowing their effect on cardiovascular events.” – Stephen J. Nicholls, MB, BS, PhD, Monash University, Clayton, Australia
A big question remains: what is it about the incretin agonists’ mechanism of action that makes them effective against CVD/events? “I think we’re still trying to work that out from all of the studies,” says Nicholls. “It is likely to be a combination of factors. It’s not all about the weight loss or the improved glycemic control. The combined metabolic benefits are likely to be important, direct effects on blood vessels and the heart may be important, and there may be additional factors that remain unknown. It highlights that there is more research needed — we need to understand this data to a greater degree.” For example, he cites an ongoing placebo-controlled trial of tirzepatide in high-risk patients with CVD but without type 2 diabetes. “That will be an important study to follow,” he says. He adds that researchers should continue to strive for diversity in the makeup of study participants, “so our findings more reflect the diversity of patients we see in the clinic.”
The bottom line here is that tirzepatide has proven its worth. Regarding the increased incidence of GI side effects, Nicholls points out that overall tirzepatide was similar to dulaglutide in terms of adverse events and drug discontinuation and thus tolerability. “Tirzepatide has good metabolic benefits and is at least comparable to dulaglutide regarding cardiovascular events,” he says. “The benefits on mortality and extended MACE further highlight its potential utility. This is good for patients as it suggests we have more choice of therapies.”
Dulaglutide, Exenatide, Liraglutide, and Semaglutide
From the September issue of Diabetes Research and Clinical Practice, “Comparative effectiveness of GLP-1 receptor agonists on cardiovascular outcomes among adults with type 2 diabetes and moderate cardiovascular risk: emulation of a target trial” compared the cardiovascular outcomes of four GLP-1RAs among adults with type 2 diabetes at moderate cardiovascular risk. Stacey M. Sklepinski, MD, Family Medicine Resident Physician (PGY1) at Advocate Lutheran General Hospital in Park Ridge, Ill. and Rozalina G. McCoy, MD, MS, of the University of Maryland School of Public Health THRIVE (Transforming Health through Real-world Insights, Values, and Evidence) Lab in Baltimore, Md., and team sought to address a research gap: comparing cardiovascular outcomes following initiation of GLP-1RAs among people with type 2 diabetes at moderate cardiovascular risk (recalling Nicholls’ stated research need), which they defined as individuals whose predicted risk of experiencing a MACE like MI or stroke, or dying from any cause, was 1% to 5% in the coming year. “More fundamentally,” say Sklepinski and McCoy, “there has been no head-to-head evidence comparing different GLP-1RAs for cardiovascular outcomes. Clinicians and patients often select agents based on insurance formulary preferences rather than comparative efficacy or effectiveness data. Payors make formulary decisions based, in large part, on manufacturer rebates and contract negotiations, particularly when comparative efficacy/effectiveness information is limited. With dramatic increases in GLP-1RA prescribing, potential shortages of semaglutide, in particular, and the recent availability of generic options for liraglutide, evidence-based guidance on preferential agents within this drug class became even more important.”
They chose to emulate a target trial as their study design because such a framework provides the closest approximation to randomized-controlled trial (RCT)-level evidence when head-to-head trials are unlikely to be conducted due to time, cost, and feasibility constraints.
Their study included adults with type 2 diabetes and moderate CVD risk who initiated dulaglutide (35,572), exenatide (4,376), liraglutide (8,843), or semaglutide (33,063) between 2019 and 2021. Semaglutide and liraglutide fared best: semaglutide was associated with lower risk of MACE, expanded MACE, all-cause mortality, acute stroke, and arterial revascularization compared to dulaglutide. Liraglutide was also associated with a lower risk of MACE and all-cause mortality compared to dulaglutide.
“It will also be important to examine use of GLP-1RAs and SGLT2is together. Both classes are highly effective in reducing cardiovascular events in patients with type 2 diabetes across the spectrum of CVD risk (notably, our prior work comparing across mediation classes suggests that SGLT2is may be even more favorable than GLP-1RAs for some outcomes, and are comparable on others), and we assume that taking both together is likely to be even more beneficial. But we do not have direct evidence to back this up.” – Rozalina G. McCoy, MD, MS, University of Maryland School of Public Health THRIVE (Transforming Health through Real-world Insights, Values, and Evidence) Lab, Baltimore, Md. (pictured) and Stacey M. Sklepinski, MD, Family Medicine Resident Physician (PGY1), Advocate Lutheran General Hospital, Park Ridge, Illinois
Possible mechanisms include differences in receptor-binding and pharmacokinetic properties, say Sklepinski and McCoy. “While all GLP-1RAs activate the same receptor, they differ in their structures (e.g., semaglutide has 94% homology to native GLP-1 with specific amino acid substitutions and fatty acid side chain allowing albumin binding) and half-life (~ 1 week for semaglutide, five days for dulaglutide, and 13 hours for liraglutide). Longer engagement with the receptor may translate to more sustained signaling effects. Beyond glycemic control and weight loss, GLP-1RAs exert direct cardiovascular effects including endothelial function improvement, reduction in inflammatory markers and inflammatory cytokines, anti-atherosclerotic effects, blood pressure reduction, and potential direct myocardial effects. The magnitude of these effects may vary by agent based on receptor pharmacology and tissue penetration.”
Of note, exenatide showed noninferiority to each of the other three agents, a finding that surprised the team, as it had not demonstrated cardiovascular benefits in the EXSCEL trial. Sklepinski and McCoy explain that treatment contamination in the EXSCEL trial could be diluting exenatide’s effects, specifically, higher rates of open-label use of other GLP-1RAs and sodium-glucose cotransporter 2 inhibitors (SGLT2is) in the placebo arm. However, post-hoc analyses revealed statistically significant benefits similar to liraglutide and approaching those of dulaglutide. “This highlights the value of real-world evidence when trial execution issues may obscure true drug effects and suggests that analysis of RCT effects may benefit from incorporating some of the same causal methods as observational studies. Indeed, RCTs are only randomized on treatment assignment (i.e., what medication the participant is assigned to take) not treatment exposure (i.e., what medication the patient actually takes and is exposed to), and when RCTs exhibit significant post-randomization variation, we need to account for those or at least consider them when interpreting study results,” they say.
Another surprising result — that there were no significant differences between semaglutide and liraglutide for any outcome despite prior studies demonstrating semaglutide’s superiority — led the researchers to make other important observations about study design. “The finding suggests that liraglutide’s cardiovascular benefits may be underappreciated, study populations and contexts matter (in higher-risk patients, differences in efficacy may be more pronounced), and/or that real-world effectiveness may differ from efficacy in selected trial populations,” they say.
They likewise found no difference between the GLP-1RA drugs in heart failure (HF) hospitalization risk, despite semaglutide’s benefits in the STEP-HFpEF trial for HF symptoms in patients with HFpEF, obesity, and type 2 diabetes (see the Targeting Frailty section of “Beyond the Scale: What Else Glucagon-like Peptide-1 Receptor Agonists Can Do” in this issue). “The current study found no differences in these risks between any GLP-1RAs, suggesting that HF benefits may be restricted to patients with significant obesity (as in STEP-HFpEF), HF benefits may emerge primarily in those with existing HF rather than for primary prevention, and/or that the mechanism of semaglutide’s benefit may involve weight loss rather than direct cardiac effects,” say Sklepinski and McCoy. They conclude that a future study should focus specifically on potential heterogeneous treatment effects as a function of obesity and HF risk.
Speaking of future studies, the team has identified several avenues to explore, including further research on how to best care for people with type 2 diabetes, with obesity, and cardiovascular-kidney-metabolic (CKM) syndrome risk factors more broadly. “First and foremost, it will be important to examine the potential mediating effects of weight loss and glucose-lowering on the effectiveness of different GLP-1RAs to understand both why we are seeing the observed benefits and why differences between drugs exist,” they say. “This will also help us better tailor treatment, particularly in patients who may not respond to GLP-1RA therapy with respect to weight loss or glucose-lowering as expected. It is important to understand whether the cardiovascular benefits of GLP-1RAs are limited to those who also experience glycemic and/or weight benefits, or we see benefits in hard outcomes more broadly. Current indirect evidence from the cardiovascular outcomes trials suggests that cardiovascular benefits are at least partly independent of glucose-lowering, but the same analysis was not done for weight loss.”
They also want to examine oral semaglutide separately from the subcutaneous formulation, particularly now that an oral version has been approved for weight loss, and to consider dose effects. Tirzepatide was not clinically available for the current study, so they would include it in future comparisons. Future work should also focus on other outcomes important to patients like MASLD/MASH, arthritis, sleep apnea, and others.
“It will also be important to examine use of GLP-1RAs and SGLT2is together,” state Sklepinski and McCoy. “Both classes are highly effective in reducing cardiovascular events in patients with type 2 diabetes across the spectrum of CVD risk (notably, our prior work comparing across mediation classes suggests that SGLT2is may be even more favorable than GLP-1RAs for some outcomes, and are comparable on others), and we assume that taking both together is likely to be even more beneficial. But we do not have direct evidence to back this up.”
They also want to highlight the importance of implementation research, asking, “how do we improve prescribing, access, affordability, and tolerability? We continue to see many patients who would benefit from these drugs not receive them, and there are many reasons for this: clinicians do not consistently prescribe them due to lack of awareness or training, high administrative burden, bias, or other reasons; patients cannot fill them due to cost, insurance coverage, or shortages; or patients stop them, again, due to cost, insurance coverage, shortages, or side effects (see the Discontinuation section of “Beyond the Scale” in this issue). Thus, administrative hurdles, worsening insurance coverage, and high costs are major barriers needing to be addressed. “Indeed,” say Sklepinski and McCoy, “most patients with type 2 diabetes would benefit from a GLP-1RA (or SGLT2i) for cardiovascular risk reduction, and we should continue to work on multilevel strategies to reduce barriers to prescribing, access, and use.”
Horvath is a freelance writer based in Baltimore, Md. In the January issue, she wrote about a number of new directions for treating patients with thyroid cancer.
Loading ...