Semaglutide Slows Biological Aging by Up to 5 Years in Randomized Trial

Aging biology has identified epigenetic clocks — DNA methylation patterns that predict biological age and mortality risk — as some of the most sensitive biomarkers for evaluating potential anti-aging interventions. Until now, no randomized controlled trial had tested whether semaglutide, the widely prescribed GLP-1 receptor agonist, could actually slow biological aging as measured by these clocks. This study fills that gap, providing the first RCT-level evidence that semaglutide meaningfully reduces epigenetic age estimates across multiple validated tools.

The analysis is a post-hoc examination of a completed 32-week, double-blind, placebo-controlled phase 2b trial (NCT04019197) that originally investigated semaglutide's effects on adipose tissue distribution in people with HIV-associated lipohypertrophy. Adults with HIV on stable antiretroviral therapy, BMI ≥25 kg/m², and lipohypertrophy — but without diabetes or cardiovascular disease — were randomized 1:1 to once-weekly subcutaneous semaglutide (titrated to 1.0 mg) or placebo. Peripheral blood mononuclear cells were collected at baseline and 32 weeks and biobanked; 84 participants (45 semaglutide, 39 placebo) had paired methylome data available for this analysis. The cohort averaged 49 years of age, was 42% women, 58% Black, and had median BMI of 32.9 kg/m².

Across 17 epigenetic clocks spanning first, second, and third generations, semaglutide showed consistently favorable effects. The most pronounced results came from mortality-predictive and systems-level clocks. PhenoAge showed a 4.9-year reduction (p=0.004); PCGrimAge showed a 3.1-year reduction (p=0.007); GrimAge V1 and V2 (calculated via the independent Biolearn framework) showed reductions of 1.4 years (p=0.042) and 2.3 years (p=0.008) respectively; SystemsAge decreased by 4.2 years (p=0.009); and DunedinPACE — which measures the rate of aging rather than age itself — slowed by 0.09 units, approximately 9% (p=0.01). The multi-omic OMICmAge clock, which integrates proteomic, metabolomic, and clinical data proxies, also decreased by 2.2 years (p=0.009). Even the RetroAge clock, which specifically captures retrotransposon-associated epigenetic aging, declined by 2.2 years (p=0.030).

Organ-system clock analysis across 11 biological domains revealed semaglutide's effects were broad rather than tissue-specific. The largest reductions were in the Inflammation clock (−5.0 years, p=0.006), Brain clock (−5.0 years, p=0.005), and Blood clock (−4.4 years, p=0.011), with significant effects also in Heart (−4.3 years, p=0.009), Kidney (−4.2 years, p=0.014), Liver (−4.2 years, p=0.042), and Metabolic (−4.7 years, p=0.009) clocks. Lung, Hormone, Immune, and Musculoskeletal clocks showed directionally favorable but non-significant trends. Notably, a first-generation Intrinsic Capacity epigenetic clock — which measures physical and cognitive resilience — did not significantly change (p=0.31), suggesting semaglutide's epigenetic effects may be more prominent in mortality-risk and disease-related aging pathways.

The study has important caveats. It was conducted in people with HIV, who exhibit accelerated biological aging due to chronic low-grade inflammation and metabolic dysfunction, meaning the magnitude of effects observed here may not translate directly to the general population. The epigenetic analysis was post-hoc and not pre-specified as a primary outcome. Sample size was relatively modest at 84 participants, and the 32-week duration limits conclusions about long-term durability. Nonetheless, the consistency of findings across 17 independent clocks, two independent computational frameworks (TruDiagnostic and Biolearn), and 11 organ-system clocks substantially strengthens confidence in the signal. These results justify prospective, pre-registered trials specifically powered to evaluate GLP-1 agonists as gerotherapeutics in broader populations.