Preprint: Semaglutide Reduced Biological Age of the Heart by 2–4 Years and of Four Other Organs by 1–4 Years on Proteomic Aging Clocks Across Five Trials
Preprint: Semaglutide Reduced Biological Age of the Heart by 2–4 Years and of Four Other Organs by 1–4 Years on Proteomic Aging Clocks Across Five Trials
In a new unreviewed preprint, a team that includes Novo Nordisk employees (the company manufactures semaglutide) and Harvard aging biologist Vadim Gladyshev applied proteomic aging clocks, models that estimate organ age from blood proteins, to samples from 10,052 participants in five completed semaglutide trials involving people with heart failure, obesity, diabetes, and chronic kidney disease. In those who received the drug, the clocks indicated younger ages for several organs relative to placebo, and the difference persisted through nearly three years of follow-up.
All five semaglutide trials (SELECT, STEP-HFpEF, STEP-HFpEF DM, FLOW, and SOUL) had already demonstrated that the drug reduces mortality, major cardiovascular events, and loss of kidney function. Rather than a sixth trial, the new study runs the blood samples already collected through a different instrument: the same type of clock, now built separately for the heart, kidneys, brain, pancreas, and lungs. The clocks themselves are not new. A year earlier, part of the same group, including co-author Ludger Goeminne, constructed them from UK volunteer blood data by matching protein levels to subsequent illness and death. At that point the models had only been validated on cross-sectional differences between individuals; whether they would respond to a pharmacological intervention was unknown.
The answer is yes, and directionally. Biological age of the heart fell by 2–4 years relative to placebo; for the remaining four organs, by 1–4 years. The effect persisted or recurred through 156 weeks, nearly three years. The clocks also predict clinical outcomes: among placebo participants, a proteomically "older" heart or kidneys at the same chronological age corresponded to higher risk of death and major cardiovascular events. For the heart and kidneys, the biological-age reduction was not statistically accounted for by weight loss alone, blood glucose, or the inflammatory marker CRP. A similar signal had appeared earlier: an unreviewed trial in 108 people with accelerated cellular aging also found a 3–5 year reduction in biological age after adjusting for weight, though by a different measurement method and in a far smaller sample. The study was conducted and funded by Novo Nordisk, sponsor of all five trials; at least seven of the twelve authors are company employees who hold its stock.
The adipose-tissue clock partly depends on leptin, a hormone whose level drops with weight loss. The model was trained on individuals in whom low leptin was typically a marker of wasting and disease rather than healthy weight reduction, so the leptin decline caused by semaglutide was initially read as accelerated fat aging; the signal reversed at later timepoints. In SELECT, only the adipose clock statistically attenuated the link between semaglutide and lower mortality: by 48% for all-cause mortality and by 54% for cardiovascular mortality. Heart biological age, by contrast, statistically explained part of the benefit: 49% of the reduction in major cardiovascular events in SOUL and 36% of the reduction in all-cause mortality in SELECT.
The same compound was the subject of a debate over a 92-day lifespan extension in aged mice, where it remained unclear whether the effect could be separated from simple caloric restriction. The new study addresses the same question by a different route: molecular clocks in humans across five trials.
For future geroprotector trials, this offers a candidate surrogate marker: a measurable molecular signal in place of years-long waits for mortality and complications. If organ-specific clocks reproducibly respond to a drug already proven to reduce hard endpoints, they can be applied to trials of other compounds, tracking which organs shift.