The obvious explanation for a weight-loss drug preventing heart attacks is the weight loss. A prespecified substudy of SELECT tested that by following high-sensitivity C-reactive protein, an inflammation marker, and found a pattern the weight explanation does not fit[1]. The same trial, cut a different way, is covered in the frailty analysis.
Baseline hsCRP was almost identical in the two arms, at a geometric mean of 1.96 mg/L on semaglutide and 1.91 mg/L on placebo, and it predicted what happened next. Event risk rose across the three baseline strata of below 2, 2 to below 10, and 10 mg/L or above, including significant associations with cardiovascular and all-cause death. Semaglutide then reduced hsCRP by 37.8% at 104 weeks, and reduced event risk in every one of those strata.
The timing is the part that matters. Larger falls in hsCRP went with larger weight loss, which on its own would suggest the marker was simply tracking the scale. But the fall was already evident at 4 and 8 weeks, before major weight loss had happened, and it occurred among participants who did not lose weight at all. It was also independent of LDL cholesterol, of statin use, and of which cardiovascular entry criterion had let someone into the trial.
What it does establish is that the benefit is not purely a function of pounds lost, which matters to anyone weighing whether a smaller response on the scale means a smaller cardiovascular return. The comparison with the other intervention that produces large sustained weight loss is drawn in surgery against a GLP-1 on cardiovascular outcomes, and the older drug whose cardiovascular case rested on different reasoning is set out in where metformin ranks.
One limit deserves stating plainly. SELECT enrolled people with established atherosclerotic disease, overweight or obesity, and no diabetes, most of them already well treated with statins, so these numbers describe a specific and fairly narrow population. Whether they extend to people starting the drug for weight alone, with no cardiovascular diagnosis, is not something this substudy can answer, and the gap between who gets studied and who gets prescribed is a running problem examined in uptake after a heart attack or stroke.