The short answer: Early, company-sponsored analyses suggest that semaglutide and tirzepatide nudge some blood-based “aging clock” readings in a younger direction in people with obesity, heart disease or diabetes. These results are not peer reviewed, none of them tested healthy people, and a better clock reading has not been shown to mean a longer or healthier life. They are a reason for more research, not a reason to take a GLP-1 medicine for anti-aging.
In early October 2026, Novo Nordisk and Eli Lilly each presented data at the Aging Research and Drug Discovery (ARDD) meeting suggesting that their weight-loss medicines may slow “biological aging.” Headlines followed quickly, along with a predictable question from patients: should I be on one of these to stay younger? This article goes back to the underlying documents to answer it. We read the full text of Novo’s analysis, which is posted as a preprint, and we describe Lilly’s data from the best available reports, since no paper or abstract was available for us to read. We explain what an aging clock is, what each study measured, and what can and cannot be concluded.
What is a “biological age” clock?
Your chronological age is the number of birthdays you have had. A biological age clock is a blood-test-based estimate that tries to capture how your body is doing compared with others of the same age. Researchers build these clocks by measuring thousands of molecules in large groups of people, then using statistics to find the patterns that best predict outcomes such as death or disease. Two kinds appear in the new studies:
Proteomic clocks measure the levels of many proteins in blood. Some are trained to predict time to death and can be built for individual organs, such as the heart or kidney. The clocks Novo used were trained on the UK Biobank, a large population database, and are described in a 2025 paper by Goeminne and colleagues.
Epigenetic clocks look at chemical tags on DNA (methylation) that change with age and with exposures such as smoking. Lilly and the HIV trial described below used this type.
Two cautions apply to all of them. First, a clock reading is a statistical prediction, not a measurement of how young a person feels or functions. Second, when a result says someone is “two years younger,” it means the clock’s score moved by an amount equal to two years of typical aging, not that the person gained two years of life.
Study 1: Novo Nordisk’s semaglutide analysis across five trials
The most detailed data come from a preprint posted on Research Square on September 22, 2026 by Maria Dermit and colleagues at Novo Nordisk and academic collaborators. A preprint has not been through journal peer review. An earlier version of the work was shown as a poster at the American Diabetes Association meeting in June 2026.
Rather than run a new trial, the team re-analyzed blood samples that had been stored from five earlier randomized, placebo-controlled semaglutide trials, 10,052 participants in total. They measured about the same set of proteins in each, applied the organ clocks, and compared the change from baseline in the semaglutide and placebo groups.
| Trial |
Who was studied |
Semaglutide dose |
Participants (placebo / semaglutide) |
Blood samples analyzed |
| SELECT |
Overweight or obesity with heart disease, no diabetes |
2.4 mg weekly injection |
867 / 774 |
Week 20 |
| STEP-HFpEF |
Obesity with heart failure (preserved ejection fraction) |
2.4 mg weekly injection |
208 / 208 |
Weeks 20 and 52 |
| STEP-HFpEF DM |
Same, with type 2 diabetes |
2.4 mg weekly injection |
235 / 234 |
Weeks 20 and 52 |
| FLOW |
Type 2 diabetes with chronic kidney disease |
1.0 mg weekly injection |
1,143 / 1,128 |
Weeks 12, 52 and 156 |
| SOUL |
Type 2 diabetes with high cardiovascular risk |
14 mg daily tablet |
2,604 / 2,651 |
Weeks 13 and 104 |
Source: Dermit et al., Research Square preprint (Table 1 and text). Everyone studied had obesity, heart disease, kidney disease or diabetes.
What they found
The heart clock was lower in people on semaglutide than on placebo in all five trials, by roughly 2 to 4 years of “clock age.” The change appeared at the first check after starting treatment (weeks 12 to 20) and persisted or recurred through follow-up of up to 156 weeks. Kidney and artery clocks were lower in some trials, especially FLOW and SOUL, and the abstract reports reductions of about 1 to 4 years across kidney, brain, pancreas, lung and multi-organ clocks. A separate protein-based frailty score improved by 52 weeks in both heart failure trials, by 52 and 156 weeks in FLOW, and by 104 weeks in SOUL, but not at 20 weeks in SELECT.
The picture was not uniformly favorable. The liver and immune clocks rose early, and the fat-tissue clock rose at first because the drug lowers leptin, a hormone made by fat cells. The authors interpret these as transient, and the adipose-clock rise was largely explained by weight change. That detail is a useful reminder that clocks respond to what a drug does to the body’s chemistry, not only to “aging.” The ADA poster of the same work also reported that a whole-body composite clock did not change, while the multi-organ clock did.
Two further findings stand out. First, most of the heart and kidney clock changes were not statistically explained by weight loss, blood sugar (HbA1c) or the inflammation marker CRP, which suggests effects beyond weight. Second, in exploratory analyses the heart clock accounted for about 36% of semaglutide’s effect on all-cause death and 40% of its effect on cardiovascular death in SELECT, and 34% to 49% of selected outcomes in SOUL. The authors label these exploratory and state that they do not show causation.
How the authors describe their own results
The preprint is careful. It concludes that semaglutide produces “favorable remodeling” of protein signatures linked to mortality, and it says the findings do not establish biological rejuvenation, causal mediation or that these markers can stand in for real outcomes. Their stated limitations include that the clocks were trained in a different population (UK Biobank), that protein levels are relative and hard to compare across trials, that the heart and fat clocks rest on only a few proteins, that the fat clock was trained in people whose weight was stable, and that a protein frailty score does not prove better strength or function. No healthy group was analyzed. Novo funded the analysis and all five trials, and many authors are Novo employees or shareholders.
Study 2: Lilly’s tirzepatide data
Lilly’s data were presented at the conference and have not been published as a paper or abstract that we could read. The details below come from the trade reports by Longevity.Technology and MedPath, which describe the presentation; they should be treated as secondary sources.
The work is a side analysis of SURMOUNT-5, the head-to-head trial of tirzepatide against semaglutide, and it used only people on tirzepatide, with no comparison group. Blood from 82 people passed quality control, and 71 had both a baseline and a 72-week sample; by week 72, 91.5% were taking the 15 mg dose. The analysts applied 15 epigenetic clocks that report an age. Over the 1.38 years that elapsed, all 15 advanced by less than 1.38 years, but only two reached statistical significance, with median differences of about 1.1 and 0.7 years. A pace-of-aging clock (DunedinPACE) fell from a median of 1.13 to 1.08. Organ-level clocks pointed toward lower predicted age in the immune system, kidney, liver and cardiovascular system.
Lilly itself calls the results preliminary and hypothesis-generating. The reported caveats are significant: no control group, a short follow-up compared with the measurement error of the clocks, some clocks pointing different ways, several predicted protein changes that did not match the measured levels, and clocks that were not built from trial or long-term follow-up data. In a before-and-after design with no placebo group, any change could reflect weight loss itself, or ordinary variation, rather than a specific effect of the drug.
A third signal: a small randomized trial in people with HIV
A separate line of evidence comes from a preprint on a randomized, placebo-controlled trial of semaglutide in adults with HIV-associated abdominal fat (lipohypertrophy). In a post hoc analysis of 84 participants (45 on semaglutide, 39 on placebo), 32 weeks of treatment was associated with a pace of aging about 9% slower on DunedinPACE (−0.09 units per year; 95% CI −0.17 to −0.02), and several other clocks moved favorably. Others did not, including a clock of intrinsic capacity. The group was small, short-term and a very specific population, and we could not confirm that the analysis has been published in a journal. It adds a placebo-controlled signal, but not proof.
What these studies do and do not show
| |
Novo (semaglutide) |
Lilly (tirzepatide) |
HIV trial (semaglutide) |
| Participants |
10,052 across five trials |
71 with paired samples |
84 with paired samples |
| Comparison group |
Placebo in each trial |
None |
Placebo |
| Clock type |
Protein (proteomic) |
DNA methylation (epigenetic) |
DNA methylation (epigenetic) |
| Peer reviewed |
No (preprint) |
No (conference presentation) |
Preprint; publication not confirmed |
| Healthy people included |
No |
No |
No |
| Who wrote it |
Novo employees and academics; Novo-funded |
Lilly |
UC San Diego-led academic group |
Sources: Research Square preprint; Longevity.Technology and MedPath reports of the Lilly presentation; medRxiv preprint (HIV trial).
Taken together, the studies show a consistent direction: in people who are overweight, diabetic or have heart or kidney disease, clock readings tend to move favorably on these drugs. That is worth taking seriously, and it fits what we already know about these medicines. For example, in the SELECT trial, semaglutide lowered the risk of major cardiovascular events (see our guide to tirzepatide and cardiovascular risk), and we reviewed a study of dementia risk in our dementia-risk article. A heart clock that improves in the same direction as heart outcomes is plausible.
They do not show that clock changes cause longer or healthier life, that the effect holds in people who are not sick, or that tirzepatide’s effect is real, since the only tirzepatide data have no control group. Clocks are still research tools; the Novo authors themselves state that these markers are not yet accepted substitutes for real outcomes. Some of the changes could also reflect the ordinary health gains that follow weight loss and better blood sugar control, although the Novo analysis found that weight, HbA1c and CRP did not fully explain the heart and kidney changes.
What is being tested next
A genuine test in healthier people is planned but has not begun. A federally funded study called VITAL-H, led by UT Health San Antonio with up to $38 million from ARPA-H, will compare rapamycin, dapagliflozin and semaglutide, each against placebo, in 726 adults in their 60s. Enrollment is expected to begin in 2027 on a five-year contract, so answers are years away. Lilly says it plans to extend its epigenetic work to semaglutide and other studies.
BeLite Clinical Perspective
The honest reading is “interesting and early.” It would be premature to prescribe a GLP-1 medicine to anyone for anti-aging, and neither company is making an approved claim of that kind. These drugs are approved to treat obesity, diabetes and, for semaglutide, to reduce major cardiovascular events. If you have one of those conditions, the established benefits are the reason to treat, and any effect on aging clocks would be a possible bonus, not the goal. You can see how we use them on our Wegovy and Zepbound pages, and review all of our medication options.
For people who are older, the more practical longevity question is how to lose weight without losing strength. Rapid weight loss can include lean mass, which is why protein and resistance training are part of any plan; see our guide to preserving muscle on GLP-1 therapy, and our discussion of how we tailor care at different life stages and GLP-1 medicines in perimenopause and menopause. If you are already taking one of these medicines, remember that stopping has consequences, which we describe in what happens after stopping Wegovy or Zepbound.
Finally, expect “anti-aging” to appear in marketing. Be wary of any clinic or website offering compounded or unapproved GLP-1 products for longevity.
FAQs
Do GLP-1 drugs slow aging?
Not established. In people with obesity, heart disease, kidney disease or diabetes, blood-based aging clocks tended to move in a younger direction on semaglutide and, in a small sample with no comparison group, on tirzepatide. Nobody has shown that this translates into living longer or aging better, and healthy people have not been studied.
What does “two to four years younger” mean?
It means a statistical score built to predict mortality risk was lower by an amount equivalent to two to four years of typical aging in the heart clock. It does not mean a person’s body is two to four years younger or that they will live two to four years longer.
Has this been published in a medical journal?
Not yet. Novo’s analysis is a preprint that has not been peer reviewed, Lilly’s was a conference presentation, and the HIV trial analysis is a preprint we could not confirm was published.
Should I take Wegovy or Zepbound to stay younger?
No. These medicines are for treating obesity, diabetes and related conditions, and the aging data are early and come from sick populations. If you qualify for treatment for those reasons, a possible aging benefit is a bonus, not a reason by itself.
Is the effect just from weight loss?
Partly unknown. In the Novo analysis, most heart and kidney clock changes were not statistically explained by weight loss, blood sugar or CRP, but that is a statistical finding in a re-analysis, not proof of a separate mechanism. Lilly’s data cannot answer this because there is no comparison group.
When will we know more?
Not soon. Journal publication of the existing analyses may come first. The first dedicated test in adults in their 60s, VITAL-H, expects to begin enrolling in 2027.
Last medically reviewed: November 1, 2026
References
- Dermit M, Colhoun HM, Goeminne LJE, Ryan D, et al. Semaglutide reduces proteomic organ-specific biological age across randomized clinical trials. Research Square preprint, posted September 22, 2026 (not peer reviewed).
- Dermit M, Aguayo-Orozco A, Voisin SAC, et al. Semaglutide reduces multi-organ proteomic-based biological age across cohorts. Poster, American Diabetes Association 86th Scientific Sessions, June 5–8, 2026.
- Goeminne LJE, et al. Organ-specific proteomic aging clocks predict disease and longevity across diverse populations. PubMed record.
- Longevity.Technology. Lilly and Novo share GLP-1 longevity data. October 4, 2026.
- MedPath. Novo Nordisk and Eli Lilly report GLP-1 signals of slowed biological aging at ARDD.
- Semaglutide slows epigenetic aging in people with HIV-associated lipohypertrophy: evidence from a randomized controlled trial. medRxiv preprint, 2025.
- Lincoff AM, et al. Semaglutide and cardiovascular outcomes in obesity without diabetes (SELECT). N Engl J Med. 2023.
- Aronne LJ, et al. Tirzepatide as compared with semaglutide for the treatment of obesity (SURMOUNT-5). N Engl J Med. 2025.
- UT Health San Antonio. UT San Antonio to lead $38 million national trial testing drugs to extend healthspan. March 2026.
BeLite can help patients translate emerging obesity-medicine evidence into an individualized plan that considers efficacy, tolerability, nutrition, muscle preservation, long-term maintenance, and access.BeLite Medical Center has provided physician-supervised medical weight management in Fairfax, Virginia since 1995. Our clinic offers in-person medical supervision — with the ability to adjust medications and monitor your progress in real time — that fully virtual services cannot match.Booking appointments is simple with our online scheduling portal, available 24/7. You can also call or text us at (703) 359-9200, or email belitemed@gmail.com. We offer free consultations for new patients — if you choose not to enroll, there is no charge. We serve patients across Virginia, Maryland, Washington D.C., the DMV area, Pennsylvania, and West Virginia. Patients do not sign treatment contracts and can stop coming at any time
Medical Disclaimer
This blog post is provided for informational purposes only and does not constitute medical advice, diagnosis, or treatment. It is not intended to substitute for consultation with a licensed healthcare provider. Readers should not use this information to diagnose or treat a medical or health condition without consulting an appropriately qualified healthcare professional. Any reliance on the information in this blog is at the readers own risk. Individual weight loss results vary based on personal health status and adherence to medical advice. No doctor-patient relationship is created by accessing or reading this content.
Reviewed by Richard Rothman, MD, PhD — Medical Director, BeLite Medical Center.