Researchers compare tirzepatide and semaglutide in largest real-world study

Cardiovascular disease still claims more lives than any other illness, and many adults living with obesity or type 2 diabetes carry a heavy risk of heart attack and stroke. Two medications, semaglutide and tirzepatide, have become widely used because they lower blood sugar and support weight loss.

For years, researchers have wondered whether these medicines protect the heart in the same way, especially when used outside the tight controls of a clinical trial. A new nationwide analysis now offers the clearest real-world comparison to date, and it suggests that both drugs provide strong and early cardiovascular protection.

The research team, led by investigators at Mass General Brigham, used U.S. insurance claims data representing nearly a million adults treated with tirzepatide or semaglutide between 2018 and 2025. Their approach was straightforward and practical: Do tirzepatide and semaglutide protect against heart attack, stroke, and death when used in routine care and not in a clinical trial? Since no randomized trial has been completed that has compared semaglutide and tirzepatide, the authors filled this gap by observing a population of those using semaglutide and tirzepatide in the everyday care setting.

Before making a direct comparison about the two medications, the research team tested its methods using two well-known trials. To check the integrity of their methods, they replicated the design of two trials: SUSTAIN 6 which investigated semaglutide, and SURPASS-CVOT which studied tirzepatide. By maintaining eligibility rules and follow-up patterns, they confirmed the integrity of their real-world data using known trial data.

In other words, the emulations did what they wanted them to do. The terms of semaglutide hazard ratios were quite similar to those observed in SUSTAIN 6 for heart attack and stroke, and in fact, tirzepatide hazard ratios had also been similar to prior trials in SURPASS. The first notable difference was seen in all-cause mortality in the emulation of the semaglutide trials, which differed from the trial report. That finding led researchers to think more about non-mortality events in later analyses.

After developing confidence that their design had worked, the group expanded to adults who were representative of patients seen in clinics week after week. The sizeable cohorts included more than 450,000 adults starting either semaglutide or sitagliptin and more than 136,000 starting tirzepatide or dulaglutide. Nearly 300,000 adults were included in the more direct comparison of tirzepatide to semaglutide.

People starting semaglutide or tirzepatide were generally younger adults, and more frequently reported White race compared to those taking sitagliptin or dulaglutide. After matching for age and sex, kidney health, past cardiac concerns, and dozens of other measures, the groups balanced across a number of factors. Average BMI ranged from the mid-30s to the high 30s. Between about 50 and 56 percent of adults were female. A number had a history of stroke or heart attack, plus up to a third had chronic kidney disease.

Follow-up was representative of real-world practice. Most patients did not complete the study, and there are reasons for this that are not captured in claims data. The on-treatment follow-up duration of six months was representative across the classes of medications, yet between one-third and nearly one-half of patients discontinued their medications before the end of the data collection or observation period.

Semaglutide reduced the risk of a heart attack or stroke at one year compared to sitagliptin. The risk of risk for recurrence was reduced from 1.7% to 1.5%, or an 18% reduction in risk. This protection was consistent across the spectrum of patients who were at greater risk for cardiovascular events.

Tirzepatide produced advantages as well. When compared to dulaglutide, which is a GLP-1 agent with an established cardiovascular value, patients treated with tirzepatide had a 13% lower risk of cardiovascular events such as heart attacks, strokes, or death. Differences were modest, but meaningful, especially given the early onset. As the first author, Nils Krüger, MD stated, the benefits appear to extend beyond weight reduction.

The head-to-head comparisons are arguably most important to patients and clinicians. There was almost an identical one year rate of heart attacks, strokes, or deaths in the tirzepatide and semaglutide groups. The risk of the outcomes in both groups was 1.3%, and statistical analyses provided no meaningful differences between the groups. For heart failure outcomes, there was slightly less risk in the tirzepatide group, but confidence intervals of the data overlapped, and the signal for reduced risk is tenuous.

Safety outcomes corroborated the main findings. Individual safety outcomes, whether serious bacterial infections or urinary tract and digestive side effects, were generally lower risk compared to the comparison groups. The negative control outcomes also exhibited no spurious associations or signals of harm, further corroborating the integrity of the findings.

This study advances what can be gained from observational studies when other trial data are either lacking or incomplete. "Randomized controlled trials remain the gold standard," Krüger says, "but not all questions can be answered using a time- and resource-consuming design."

Importantly, he stated that the use of data from daily clinical practice allows for the expression of overlapping clinical circumstances in most patients who rarely enroll in tight research studies. The senior author, Shirley Wang, PhD, stated the aim of a transparent project was to allow for scientific merit regardless of outcome, and omitted barriers to engagement by preregistering their project and sharing data analytic code.

The timing of the results comes as both medications become more available. Many adult patients are turning to medications for either diabetes management or weight loss, and it is critical that clinicians have reliable information to impart on whether one medication might offer more heart-protective strategies. Based on these findings, both appear to be safe and provide measurable cardiovascular benefit that performs similarly when compared directly.

Research findings are available online in the journal Nature Medicine.

Like these kind of feel good stories? Get The Brighter Side of News' newsletter.