Reading the Heart Trials: Semaglutide Cardiovascular Benefits in SUSTAIN-6, SELECT and SOUL

Semaglutide cardiovascular benefits make for big headlines, but the numbers behind them are easy to misread. Before you trust a claim, it helps to know what each trial counted, who it enrolled, and where its answers stop. This guide walks through the key studies and explains how to read cardiovascular outcome trials so you can judge the next headline for yourself.

Why heart trials exist for these drugs

Heart trials for diabetes drugs began with a regulatory question. In 2008, the US Food and Drug Administration issued guidance asking companies developing new glucose-lowering drugs for type 2 diabetes to show that those drugs did not meaningfully raise cardiovascular risk. That request shaped how many of these studies were designed.

Later, the question shifted from safety toward benefit. A weight-loss claim and a heart claim rest on different evidence. Weight is measured on a scale over months. A heart claim means counting serious events, such as heart attacks and strokes, across thousands of people for years. That's why these trials are big and slow, and why they can't be read like a weight chart.

What MACE means and why it matters

Most of these trials use a composite endpoint called MACE, short for major adverse cardiovascular events. In the standard version, MACE counts cardiovascular death, nonfatal heart attack, and nonfatal stroke. Each trial's protocol spells out its exact definition, and small wording differences can change what gets counted. At OzemNews, our coverage of the trial data walks through these definitions step by step, since they're easy to skip in a headline.

A composite endpoint is useful because serious events are relatively rare. It can also hide which piece moved. If heart attacks fall sharply while strokes barely change, the combined number can still look good. Ask which component drove the result. Published papers typically break out each part in their tables.

Trials also report a hazard ratio, which compares the rate of events in the drug group with the placebo group over the study period. A hazard ratio of 0.80 suggests roughly a 20 percent lower rate of events. That's a relative figure, and it says nothing by itself about how many people avoided an event. Imagine that risk falls from 4 in 100 to 3 in 100 over a few years. That's a 25 percent relative reduction, yet only one fewer event for every 100 people. Your baseline risk decides how much a relative drop means for you.

SUSTAIN-6: the diabetes population

SUSTAIN-6 was published in the New England Journal of Medicine in 2016. It enrolled 3,297 adults with type 2 diabetes who were aged 50 or older and had established cardiovascular disease, chronic kidney disease at stage 3 or higher, or both. Participants received weekly semaglutide or placebo alongside their usual care. The main result was a MACE hazard ratio of 0.74, with a 95 percent confidence interval of 0.58 to 0.95. Median follow-up was about 2.1 years.

Follow-up length shapes how much weight a result can carry. With a median of about 2.1 years, longer-term questions remain open. The confidence interval shows the uncertainty around the estimate. Here its upper end sits just under 1, so the result points in a favorable direction but leaves room for a smaller effect. The paper also reported a higher rate of diabetic retinopathy complications with semaglutide, a reminder that one trial rarely answers every question.

Because everyone in SUSTAIN-6 had type 2 diabetes, its results don't automatically transfer to people without it.

SELECT: the trial that changed the conversation

SELECT was published in the New England Journal of Medicine in 2023 and looked at a different group. It enrolled 17,604 adults with overweight or obesity and established cardiovascular disease, but without diabetes. Participants received a weekly 2.4 mg dose of semaglutide or placebo. Median follow-up was a little over three years, and the MACE hazard ratio was 0.80, with a 95 percent confidence interval of 0.72 to 0.90. For semaglutide heart risk without diabetes, SELECT is the key reference point.

Because SELECT tested people without diabetes, it answered a question the earlier diabetes trials couldn't. Regulators took notice. In March 2024, the FDA approved a cardiovascular indication for Wegovy, to reduce the risk of major adverse cardiovascular events in adults with established cardiovascular disease and either obesity or overweight. Check the current prescribing information on Drugs@FDA before drawing conclusions for a specific person, since labels can change.

All-cause mortality needs a careful read. Death from any cause was a secondary outcome in SELECT, not the main test the trial was built around. The paper reports fewer deaths in the semaglutide group, which is encouraging. Still, a secondary result from a trial designed around MACE should be treated as supportive evidence, not a settled answer.

Where the evidence is thinner

Not every heart question has a hard-endpoint answer. STEP-HFpEF studied adults with obesity and heart failure with preserved ejection fraction. Its primary endpoints were symptoms, measured with a patient questionnaire called the KCCQ, and change in body weight. Those outcomes matter for how people feel and move through the day, but they aren't the same as fewer hospitalizations or deaths.

The oral version, tested in SOUL, deserves its own reading. Rosenstock and colleagues published that trial in the New England Journal of Medicine in 2025. It enrolled 9,650 adults with type 2 diabetes aged 50 or older who had established cardiovascular disease, chronic kidney disease, or both. Participants took a daily 14 mg oral semaglutide tablet, and the MACE hazard ratio was 0.86, with a 95 percent confidence interval of 0.77 to 0.97. A different formulation, dose, and population means its results shouldn't be stacked onto the injectable data as if they came from one study.

Inclusion criteria also decide who a result describes. The trials set age, diagnosis, and risk thresholds, so people outside those limits have less direct evidence. Trial populations rarely mirror everyone, so check each paper's baseline table for age, sex, race, and region before assuming a result applies to you. Readers who want to trace how these trials were reported over time can browse the OzemNews reporting archive. Trials also run under conditions that may differ from routine care, with scheduled visits and structured follow-up, so results from that setting don't map perfectly onto every clinic.

A reader's checklist for talking to your prescriber

Trial data only becomes useful when someone applies it to your own history, and that conversation belongs with your clinician. Headlines from OzemNews and other outlets can help you prepare, but they can't replace your own numbers. Consider bringing these three questions to your next appointment:

• Which trial supports my situation, and does my risk profile match its population?
• What is my estimated absolute risk, not only the relative reduction percentage?
• How would any cardiovascular benefit fit with my current medicines and my blood pressure, cholesterol, and glucose goals?

You can also ask which changes would prompt a review and which side effects you should report. Semaglutide may help reduce heart events in the populations these trials studied. No single trial promises prevention for every person, so let the evidence and your own numbers guide the decision together.

FAQ

Does semaglutide reduce heart attacks and strokes?

In SUSTAIN-6 and SELECT, semaglutide was linked with fewer MACE events than placebo, and MACE includes heart attacks and strokes. Those results come from specific populations, so semaglutide may help reduce risk rather than guarantee it.

Did SELECT only include people with diabetes?

No. SELECT enrolled adults with overweight or obesity and established cardiovascular disease, and participants did not have diabetes.

Does the oral version have the same heart evidence as the injection?

Not in the same way. SOUL tested a daily 14 mg oral tablet in adults with type 2 diabetes, so it should be read as its own study.

What does a hazard ratio of 0.80 mean?

It means the event rate in the drug group was about 20 percent lower than in the placebo group during the study period. It's a relative measure, so your own baseline risk still matters.

Sources

• FDA Guidance: Diabetes Mellitus: Evaluating Cardiovascular Risk in New Antidiabetic Therapies to Treat Type 2 Diabetes (2008)
• Marso et al., Semaglutide and Cardiovascular Outcomes in Patients with Type 2 Diabetes (SUSTAIN-6), New England Journal of Medicine, 2016
• Lincoff et al., Semaglutide and Cardiovascular Outcomes in Obesity without Diabetes (SELECT), New England Journal of Medicine, 2023
• Kosiborod et al., Semaglutide in Patients with Heart Failure with Preserved Ejection Fraction and Obesity (STEP-HFpEF), New England Journal of Medicine, 2023
• Rosenstock et al., Oral Semaglutide and Cardiovascular Outcomes in Patients with Type 2 Diabetes (SOUL), New England Journal of Medicine, 2025
• Drugs@FDA: FDA-Approved Drugs (Wegovy prescribing information)
• ClinicalTrials.gov: SELECT study record NCT03574597
• ClinicalTrials.gov: SOUL study record NCT03914326