Deep-learning SASP score predicts mortality; exercise changed its trajectory
A composite SASP Score built with deep learning from blood proteins independently predicted mortality and serious chronic conditions in UK Biobank data; in a separate trial cohort, multimodal exercise changed the score's trajectory over 18 months. [1]
Automatic summary · not editorially reviewed
The key findings
Senescent cells release a secretory profile (SASP) that can spread senescence to neighboring and distant tissues. [1]
The deep-learning SASP Score independently predicted mortality risk and incident serious chronic conditions in UK Biobank data. [1]
In an independent trial cohort, multimodal exercise significantly changed the SASP Score trajectory over 18 months. [1]
The source at a glance
- Study type / format
- Biomarker development and validation study using population proteomics and a deep learning model; abstract only.
- Population / context
- UK Biobank participants and an independent randomized clinical trial cohort (sizes not reported in the abstract).
- Summary based on
- Study abstract
- Original publication
- October 1, 2026
SASP proteins as a blood proxy for senescence
The abstract describes senescent cells as a recognized hallmark of biological aging, linked to multiple chronic conditions. These cells release a distinct secretory profile, the senescence-associated secretory phenotype (SASP), which can spread senescence to neighboring and distant tissues. Measuring SASP factors in blood is presented as a practical proxy for cellular senescence burden that may help track disease states and intervention outcomes.
Supporting source [1]
UK Biobank prediction of mortality and chronic disease
The researchers built a composite SASP Score by combining large-scale population protein data with a semi-supervised deep learning model trained on UK Biobank Pharma Proteomics Project data, then evaluated it within UK Biobank. The score was a strong, independent predictor of mortality risk and of incident serious chronic conditions such as dementia, COPD, myocardial infarction and stroke. The abstract does not report effect sizes, follow-up length or how much the score added beyond other predictors.
Supporting source [1]
Exercise trial cohort and biomarker scope
In an independent randomized clinical trial cohort, the score was externally validated and assessed over time. There, multimodal exercise significantly changed the SASP Score trajectory over 18 months. The abstract does not state whether this change translated into health or lifespan benefits, and it does not give the trial's size or comparator. The authors present the score as a potential biomarker for systemic cellular senescence burden and as a tool for aging research and geroscience-guided interventions, not as a proven clinical test.
Supporting source [1]
How to interpret this
Abstract only; the full text was not reviewed. Effect sizes, follow-up, trial size and comparator, and whether score changes reflect health benefits are not reported. Associations are not causes, and the score is described as a potential biomarker, not a validated clinical test.
Based on the study abstract; the full paper, tables and supplementary data were not reviewed.
Source & supporting passages
Zhao S, Kuo CL, Lenze EJ, Wetherell JL, Haynes L, El-Ahmad P, Fortinsky R, Kuchel G, Harris T, Diniz BS. · europepmc.org · October 1, 2026