Background

• Disruption of proteostasis driving Aβ and tau pathologies may show early signs in blood proteomes [1].
• Proteomic ageing clocks summarize proteomes into biological ageing or morbidity risk estimates [2].
• These can detect systemic and organ-specific protein changes in blood related to disease [3].

Objectives

• Evaluate the extent to which proteomic and epigenetic ageing clocks capture variation in Alzheimer's-related phenotypes before clinical manifestation.
• Investigate how common dementia risk factors impact ageing clock-outcome associations.

Approach

• We used Generalized Estimating Equation models to test associations of ageing clocks with global cognitive scores, tau, neurodegeneration and neuroinflammation biomarkers.
• In interaction analyses we tested the potential effect of common dementia risk factors and genetic predisposition to them (e.g. diabetes, blood pressure, cholesterol and glucose levels) on the clock-outcome associations.
• In multivariable analyses we tested how much information on each outcome proteomic clocks collectively provide over epigenetic clocks.

Findings

• Accelerated systemic and brain-specific proteomic ageing were associated with poorer cognition and higher levels of plasma NfL, a neurodegeneration biomarker.
• Interaction analyses suggested weaker brain ageing-cognition associations in individuals with above-average genetic liability for type II diabetes.
• Proteomic clocks collectively explained up to 23% of variance in AD-related phenotypes not captured by epigenetic clocks.

References

[1] López-Otín, C. et al. Cell, 2023, https://doi.org/10.1016/j.cell.2022.11.001.
[2] Rutledge, J. et al. Nat Rev Genet, 2022, https://doi.org/10.1038/s41576-022-00511-7.
[3] Goeminne, L. J. E. et al. Cell Metabolism, 2025, https://doi.org/10.1016/j.cmet.2024.10.005.