Proteomic aging clock predicts mortality and risk of common age-related diseases in diverse populations
Comprehensive Overview
Published in Nature Medicine in 2024, this study asked whether measuring hundreds of proteins circulating in a person's blood could estimate their "biological age," a measure of how quickly the body is actually aging that may better predict disease and death risk than simply counting birthdays. Researchers used a machine-learning method to analyze 2,897 blood proteins from 45,441 participants in the UK Biobank, a large long-term health study, training a model to predict each person's chronological age from their protein levels and identifying 204 proteins whose combined levels closely tracked age. The difference between this protein-predicted age and a person's actual age, termed the proteomic age gap, served as the study's central biological aging measure, and researchers tested whether a larger gap predicted the future onset of 18 major chronic diseases, overall risk of death, and measures of physical and cognitive decline such as walking speed, grip strength, and reaction time, tracking participants for up to 16 years. The model was then validated in two independent groups from China and Finland to confirm it worked outside the original British population. The topline finding was that this proteomic age gap reliably predicted disease risk, multimorbidity, and mortality years in advance.
Outcomes and Implications
The single most important finding was that people whose proteomic age gap fell in the top 5 percent, meaning their blood proteins suggested a biological age roughly 12 years older than their actual age, had on average a 2.6-fold higher risk of developing Alzheimer's disease and a nearly fourfold higher risk of death compared with those whose biological and chronological ages closely matched. This accelerated proteomic aging was significantly linked to nearly all 18 chronic diseases studied, including heart disease, stroke, chronic kidney and liver disease, type 2 diabetes, and several cancers, as well as to weaker grip strength, slower walking pace, poorer lung function, shorter telomeres, and worse cognitive performance. The proteins that contributed most to the aging signal were involved in maintaining the scaffolding between cells, regulating immune and inflammatory responses, hormone signaling, and the structure of neurons, pointing to specific biological processes that may drive the aging process rather than being simply markers of it. Because the model performed comparably well in independent populations from China and Finland, the authors argue it captures a form of biological aging that generalizes across genetically and geographically diverse groups rather than being specific to one population. They suggest that measuring this proteomic age gap could eventually help clinicians identify people at elevated risk for age-related disease before symptoms appear and could serve as a tool for testing whether new drugs or lifestyle interventions actually slow the biological aging process.