SELECT trial parses GLP-1 non-weight MACE benefit

On April 25, 2026, The Lancet published a prespecified secondary analysis from Novo Nordisk’s SELECT trial that tried to answer a question hanging over GLP-1 drugs since the original cardiovascular readout: how much of semaglutide’s benefit comes from weight loss itself, and how much appears to come through other pathways. The analysis, led by John Deanfield and colleagues, used data from 17,604 adults with overweight or obesity, established cardiovascular disease and no diabetes. The original SELECT trial, published in 2023, showed semaglutide cut major adverse cardiovascular events, or MACE, by 20% versus placebo. The new paper said changes in adiposity explained only part of that effect. ### What did the new SELECT analysis actually measure? The Lancet paper examined whether baseline body-mass index, baseline waist circumference, and treatment-related changes in those measures tracked with later cardiovascular outcomes in SELECT. The researchers reported that baseline body weight itself was not associated with MACE risk in the placebo group, while smaller baseline waist circumference and reductions in waist circumference over time were associated with lower risk. (nejm.org) The authors then used mediation analyses to estimate how much of semaglutide’s cardiovascular effect could be attributed to adiposity change. The most-cited estimate in the paper was that 32.9% of the treatment effect on MACE was mediated by change in waist circumference, meaning the remainder was not explained by that marker in the model. That is the basis for social-media shorthand that roughly two-thirds of the benefit was “non-weight-mediated.” (thelancet.com) ### Does that mean semaglutide’s heart benefit is independent of weight loss? The Lancet authors did not say weight loss is irrelevant. John Deanfield and co-authors wrote that semaglutide’s cardiovascular benefit was “partly” mediated by adiposity reduction and said the findings point to additional mechanisms beyond body-fat change alone. A separate Lancet correspondence published alongside the analysis challenged stronger readings of that conclusion. (thelancet.com) The letter said the inference that benefit is unrelated to weight loss could be a methodological artifact, and cautioned against discarding weight reduction as a primary mechanism based on this analysis. ### Why did waist circumference matter more than body weight in this paper? The SELECT analysis focused attention on central adiposity rather than scale weight alone. (thelancet.com) The paper reported that waist circumference, both at baseline and in change over time, showed a clearer relationship with MACE risk than body weight did. A Lancet commentary describing the analysis said this approach suggests fat distribution may be more informative than total body weight for cardiovascular risk in this population. (thelancet.com) Nature Medicine’s clinician commentary on SELECT said semaglutide’s effects likely extend beyond weight reduction, citing improvements seen across cardiovascular, renal and metabolic measures in the broader trial program. That article did not quantify the mediation estimate itself, but it said the new analyses add to evidence that obesity treatment with semaglutide has systemic effects. ### Where do the liver-fat and very-low-calorie-diet comparisons fit in? (thelancet.com) Recent online discussion has paired the SELECT mediation paper with older work on rapid liver-fat reduction during very-low-calorie diets, often to argue that cardiometabolic benefit can appear before large changes on the scale. The SELECT paper itself did not test a Counterpoint-style very-low-calorie protocol, and it did not directly measure short-term liver-fat clearance as the mediator of MACE reduction. (nature.com) What the paper did provide was a trial-based estimate showing that adiposity change, as captured by waist circumference, did not account for all of semaglutide’s cardiovascular benefit. Any claim that the residual effect is specifically due to liver fat, inflammation, endothelial effects or another pathway goes beyond what this analysis directly proved, unless backed by separate named studies. (thelancet.com) ### What remains established from SELECT regardless of the mechanism debate? The November 2023 New England Journal of Medicine report remains the primary trial result: semaglutide 2.4 mg reduced the composite of cardiovascular death, nonfatal myocardial infarction and nonfatal stroke by 20% in adults age 45 or older with preexisting cardiovascular disease, BMI of 27 or higher, and no history of diabetes. The trial enrolled 17,604 participants. (thelancet.com) The April 2026 Lancet analysis did not change that outcome. It narrowed the question to mechanism, and it did so with a prespecified but model-based decomposition that has already drawn both support and criticism in the medical literature. The next step for readers is the source material itself: the main analysis, the accompanying commentary and the published correspondence in The Lancet. (thelancet.com) (nejm.org)