What explains semaglutide's heart benefit?

On 23 September 2026, the European Heart Journal published a mediation analysis of the SELECT trial asking a specific question: what actually explains the 20% drop in cardiovascular events on semaglutide^[1]? The paper's answer is not the one a marketing summary would give. Weight loss, smaller waistlines, lower inflammation and better blood sugar control each moved in the right direction, but combined they accounted for only about a third of the benefit. The other two-thirds is still unaccounted for.

What the mediation analysis actually tested

SELECT enrolled 17,604 adults aged 45 and older with established cardiovascular disease (a prior heart attack, stroke or symptomatic peripheral artery disease) and a body mass index of 27 or higher. Anyone with diabetes, an HbA1c of 6.5% or above at screening, or recent use of a glucose-lowering medicine was excluded, which is what makes the trial a clean test of semaglutide's cardiovascular effect independent of its diabetes effect^[2]. Enrolment ran across more than 35 countries starting in October 2018. Participants took weekly subcutaneous semaglutide at a target dose of 2.4 mg or a matching placebo, and the trial tracked cardiovascular death, non-fatal heart attack and non-fatal stroke as a combined endpoint through 240 weeks of follow-up. A primary event occurred in 6.5% of the semaglutide group versus 8.0% on placebo, a 20% relative reduction (hazard ratio 0.80, P<0.001)^[2].

The new analysis, led by Helen Colhoun at the University of Edinburgh with SELECT's original investigators, used a statistical method called counterfactual mediation to ask how much of that 20% reduction could be attributed to measurable changes the drug produces: body weight, waist circumference, high-sensitivity C-reactive protein (a blood marker of inflammation), HbA1c, cholesterol, blood pressure, kidney filtration rate and urine albumin[1]. Each factor is checked individually, then together, to see how much of the treatment effect on MACE (major adverse cardiovascular events) they can account for.

The numbers that came back

Semaglutide improved every one of those markers. But when the authors calculated how much of the cardiovascular benefit each change explained, the individual point estimates were modest: waist circumference reduction accounted for an estimated 64.0% of the effect, hsCRP reduction 42.1%, and HbA1c improvement 29.0%[1]. The paper describes the confidence intervals around all three estimates as wide, meaning the true contribution of any single factor could be considerably smaller than the headline number suggests. When the authors combined all the measured mediators into one multivariable model, the joint estimate came out at 31.4%[1]. The factors doctors already know how to measure and treat explain less than a third of why fewer people in the semaglutide group had heart attacks, strokes or cardiovascular death.

What this is not

The SELECT result itself is not in doubt. A 20% MACE reduction is the trial's confirmed primary result and is the basis on which the FDA approved Wegovy, the branded semaglutide product, for cardiovascular risk reduction in adults with obesity or overweight and existing heart disease, in March 2024^[3]. What the mediation analysis questions is the explanation, not the effect. Its authors say mediation analyses could not fully ascribe the effects of semaglutide on MACE to known risk factors in SELECT with any certainty^[1].

It also does not identify what the missing mechanism is. The paper is explicit that these are exploratory statistical estimates, not a settled biological explanation, and it stops short of naming what else semaglutide might be doing to lower cardiovascular risk beyond the factors it measured^[1]. In the EU, the EMA's Wegovy authorisation already lists cardiovascular disease as one of the qualifying comorbidities for the weight-management indication^[4], a narrower route into the same population SELECT studied. Readers who want the full regulatory and prescribing picture, including how the cardiovascular indication interacts with semaglutide's existing diabetes and weight-management approvals, can see the current status on our semaglutide regulation page.

SELECT also tracked secondary outcomes beyond the primary composite: cardiovascular death alone, heart-failure hospitalisation or urgent visits, coronary revascularisation, and changes in blood pressure and lipids over the same follow-up period^[2]. The FDA's approval of the cardiovascular indication carries the same safety information as semaglutide's other approvals: a boxed warning for thyroid C-cell tumours seen in rodent studies, a contraindication for anyone with a personal or family history of medullary thyroid carcinoma or multiple endocrine neoplasia type 2, and the same gastrointestinal side-effect profile (nausea, diarrhoea, vomiting, constipation) reported across semaglutide's other trial programmes^[3]. None of that changes with this mediation analysis. It is about mechanism, not safety.

Why this matters

For a patient already on semaglutide for its cardiovascular indication, nothing about this paper changes the treatment recommendation. The trial that established the benefit stands. What changes is the confidence with which anyone, including the drug's own investigators, can currently explain why it works. The authors say further analyses of SELECT, and comparable data from other trials in the same drug class, are still to come. Semaglutide reduces cardiovascular events in this population. Why it does that beyond weight, waist size, inflammation and blood sugar is still an open question.