Blood Proteomics Maps Cellular Aging and Links It to Disease Risk Years Before Diagnosis

A large study in Nature Medicine shows that a simple blood draw may capture how different cell types in the body are aging—and that these signals can predict future disease risk years in advance. The work was led by Tony Wyss-Coray (Stanford University), with first author Daisy Yi Ding, and analyzed plasma from more than 60,000 individuals using proteomic profiling of over 7,000 proteins.

The researchers built machine-learning models that estimate the “biological age” of more than 40 cell types—including neurons, immune cells, muscle, and lung epithelial cells—by linking blood protein patterns to their likely cellular origins. They found that aging is highly uneven across the body: a substantial fraction of people showed accelerated aging in specific cell types, while only a small proportion exhibited widespread multi-cell-type aging.

These cell-type-specific aging signals were not just descriptive. Over 15 years of follow-up in independent cohorts, they were associated with major diseases and mortality. For example, accelerated aging in skeletal muscle cells was strongly linked to amyotrophic lateral sclerosis (ALS), while astrocyte aging in the brain was associated with higher Alzheimer’s disease risk, particularly in individuals carrying the APOE4 genetic variant. In some cases, risk increases were large—for instance, extreme skeletal muscle aging was associated with more than a tenfold higher risk of ALS, and extreme astrocyte aging substantially elevated Alzheimer’s risk.

The study also highlighted how lifestyle and biology intersect: smoking and obesity were associated with broadly “older” cellular profiles, while healthier lifestyle patterns aligned with younger cellular states across multiple tissues. Importantly, a composite “polycellular aging” score built from multiple cell types could stratify mortality risk across cohorts and proteomics platforms.

Together, the findings suggest that aging is not a single uniform process but a patchwork of aging trajectories across different cell types, each potentially tied to specific diseases. While the work is observational in humans and does not prove causality, it points toward a future in which blood-based proteomics could help identify which organ systems are aging fastest long before clinical disease appears.

If validated further, this framework could reshape how risk for neurodegeneration, cancer, and chronic disease is assessed—moving from single biomarkers to a multi-cellular view of aging in the bloodstream.

Reference: Ding, D.Y. et al. "Plasma proteomic signatures of cellular aging predict human disease." Nature Medicine (2026). https://doi.org/10.1038/s41591-026-04446-y