The Biomarkers That Predict Mortality Risk

Researchers have discovered key biomarkers that are strongly linked to all-cause mortality and identified four measurements with independent predictive power.

Biological versus chronological

This paper opens with a discussion of a familiar fact for people interested in living longer: one’s biological age is not the same as one’s chronological age [1]. Biological age gives a more accurate indication of a person’s current and future health [2]. The authors touch upon the benefits and drawbacks of several well-known biomarkers; grip strength and gait speed are reliable [3] but, as these authors note, only begin to decline much later in life; neurological imaging is similarly reliable [4] but is expensive to perform and, like physical function tests, does not directly capture molecular changes. Telomere attrition, once a key focus of aging research, may not be as useful as previously hoped [5].

Epigenetic clocks and proteomic biomarkers have been developed to more accurately gauge aging and mortality risk. This includes the well-known GrimAge epigenetic mortality clock and its second version [6], which is referenced in this study. Organ-specific proteomics have been found to be strongly related to aging and disease as well [7].

This paper directly pits these various biomarkers against each other, gauging how valuable they are for predicting mortality and how much they overlap. The data source is LBC1936, a cohort of 861 people who were born in 1936, participated in a mental examination at age 11, and had their proteomic, physical, and neurological biomarkers examined every three years between the ages of 70 and 89.

GrimAge2 comes out on top

The researchers first examined this cohort’s organ age gaps: the difference between the chronological and biological age of each major organ. Consistent with previous research [7], these differences were only loosely correlated in the same individual. Even people with an extreme age gap in one organ were unlikely to similarly age in other organs.

Unsurprisingly, of all the various biomarkers that the researchers used, GrimAge2 was found to be the most closely connected with all-cause mortality. Organ age gaps were also highly associated with all-cause mortality, most notably in the liver, immune system, heart, pancreas, and brain.

However, some metrics of physical function were more closely related to mortality than any organ-specific clock. These included total brain volume and grey matter volume, overall cognitive function (g), and forced expiratory ratio and forced vital capacity, two metrics of lung function. Telomere attrition in leukocytes, on the other hand, was found to have no significant relationship.

The researchers then used statistical methods to examine how closely these biomarkers overlap with one another. As a composite biomarker, GrimAge2 was strongly related to all of the organ-specific proteomic markers along with many of the other markers. Walking time, a biomarker with similar predictive power as immune and heart proteomics, was strongly correlated with other biomarkers. While organ age gaps were loosely correlated with each other, they were uncorrelated with physical biomarkers as a whole.

The researchers then sought the biomarkers with the most independent predictive power when excluding other biomarkers, aiming to minimize redundancy. Four stood out in this regard: white matter volume, total brain volume, walking time, and the cognitive metric g. Combining these four markers accounted for 19% of the variance in mortality risk, and including the other 17 only raised this by 4%.

One protein stands out among many

Finally, this study took a look at proteins themselves. GDF15, which has been associated with cellular senescence, was found to be the protein most closely related to mortality risk after adjustment for lifestyle factors, and, of all the biomarkers in this study, it was only surpassed by GrimAge2 in overall predictive power. The neuropeptide NPS was found to have the strongest negative association: people who express more NPS are less likely to die of any cause.

As expected, the proteins most associated with mortality were associated with inflammation, including chemokine and interleukin signaling. The proteins that were associated with longer lifespans, on the other hand, were related to the maintenance of genomic stability, such as retroelement regulation and chromatin management.

The researchers acknowledge this study’s limitations, including the fact that this relatively small cohort only includes healthy Scottish people. This study also used time-point measurements and did not consider trajectories. This was also solely an observational study and did not delve into mechanisms. However, its core finding, that brain imaging and lung measurements are more effective than organ-specific proteomic clocks in predicting overall mortality, offers sobering insight.

Literature

[1] Salih, A., Nichols, T., Szabo, L., Petersen, S. E., & Raisi-Estabragh, Z. (2023). Conceptual overview of biological age estimation. Aging and disease, 14(3), 583.

[2] Jylhävä, J., Pedersen, N. L., & Hägg, S. (2017). Biological age predictors. EBioMedicine, 21, 29-36.

[3] de Souza, A. F., de Oliveira, D. C., Ramírez, P. C., de Oliveira Máximo, R., Luiz, M. M., Delinocente, M. L. B., … & da Silva Alexandre, T. (2025). Low gait speed is better than frailty and sarcopenia at identifying the risk of disability in older adults. Age and Ageing, 54(4), afaf104.

[4] Fletcher, E., Gavett, B., Harvey, D., Farias, S. T., Olichney, J., Beckett, L., … & Mungas, D. (2018). Brain volume change and cognitive trajectories in aging. Neuropsychology, 32(4), 436.

[5] Wang, Q., Zhan, Y., Pedersen, N. L., Fang, F., & Hägg, S. (2018). Telomere length and all-cause mortality: a meta-analysis. Ageing research reviews, 48, 11-20.

[6] Lu, A. T., Binder, A. M., Zhang, J., Yan, Q., Reiner, A. P., Cox, S. R., … & Horvath, S. (2022). DNA methylation GrimAge version 2. Aging (Albany NY), 14(23), 9484.

[7] Oh, H. S. H., Rutledge, J., Nachun, D., Pálovics, R., Abiose, O., Moran-Losada, P., … & Wyss-Coray, T. (2023). Organ aging signatures in the plasma proteome track health and disease. Nature, 624(7990), 164-172.