Abstract Body: Introduction
The circulating proteome has emerged as a crucial resource for predicting biological aging. Recent studies identified life-stage specific waves of proteomic expression; however, little is known about the relations between life-stage proteomic waves and age-related disorders.

Hypothesis
Life-stage proteomic waves during middle and late adulthood are differentially associated with diseases and mortality.

Methods
The study included a total of 53,105 participants (aged 39-70 years) from the UK Biobank study, with the measurements of 2,922 Olink plasma proteins. We used differential expression-sliding window analysis (DE-SWAN) to analyze differential expression of protein levels within a window of 10 years, while sliding the window in increments of 2 year from middle to old. We constructed a coefficient-weighted protein scores for aging waves, and then used the Cox models to assess the associations between protein scores and the risk of 22 leading diseases and mortality.

Results
We identified three distinct life-stage proteomic waves peaking at ages 50, 56, and 67, comprising 943, 708, and 1,103 differentially expressed proteins, respectively. Protein scores from all waves showed significant and varying associations with 10 diseases across neurodegenerative, cardiovascular, autoimmune, respiratory, metabolic, and liver categories. The protein scores of age 56 wave demonstrated the strongest associations with most diseases including ischaemic stroke, ischaemic heart disease, rheumatoid arthritis, systemic lupus erythematosus, chronic obstructive pulmonary disease, lung cancer, and mortality (Hazard ratio [HR] ranges: 1.30-2.10). There were distinct association patterns for specific diseases. For instance, Alzheimer's dementia showed a peak risk at age 56 (HR 1.34), rising from age 50 (HR 1.25) and remaining essentially unchanged at age 67 (HR 1.31). In contrast, vascular dementia displayed consistently higher risk at ages 50 and 56 (HR 1.59), followed by a decline at age 67 (HR 1.48). Liver diseases showed a unique pattern, with the highest risk at age 50 (HR 2.74), followed by a gradual but monotonically decrease through ages 56 (HR 2.38) and 67 (HR 2.00).

Conclusions
Our findings indicate life-stage specific association patterns between circulating proteomic profiles and risks of a wide spectrum of diseases, highlighting the essentiality of age-based strategies in prevention and treatment of diseases.