New Study Reveals Key Biomarkers and Brain Metrics Outperform Organ Clocks in Predicting All-Cause Mortality

The pursuit of understanding human aging has long been defined by a fundamental distinction: the gap between chronological age, measured simply by the passage of calendar years, and biological age, which reflects the physiological state of the body and its cells. While chronological age is immutable, biological age offers a remarkably accurate window into a person’s current health status and future mortality risks. Recently, a comprehensive study published in the academic literature has shed new light on this dynamic, identifying critical biomarkers strongly linked to all-cause mortality and pinpointing four distinct physical and cognitive measurements with unmatched independent predictive power.

By analyzing a well-characterized cohort, researchers sought to pit various modern biological metrics—ranging from epigenetic clocks and proteomic markers to neurological scans and physical function tests—against one another. The goal was not only to determine which metrics hold the highest predictive value for lifespan and health span, but also to understand how these diverse markers overlap, complement, or contradict one another in clinical applications.

A Historical Timeline of Aging Biomarkers

To appreciate the significance of this recent study, it is necessary to examine the evolution of aging research over the past several decades. For a long time, biogerontologists relied on simple physiological observations and rudimentary cellular metrics to estimate biological age. In the late 20th and early 21st centuries, telomere attrition—the progressive shortening of the protective caps at the ends of chromosomes—became a central focus of longevity research. Scientists hypothesized that telomere length in leukocytes would serve as a dependable molecular clock for human aging.

However, subsequent meta-analyses and longitudinal studies revealed significant limitations. While telomere shortening is broadly associated with cellular senescence and age-related pathology, its predictive power for individual all-cause mortality proved remarkably weak and inconsistent compared to newer, more sophisticated tools.

The next major leap came with the advent of epigenetic clocks in the 2010s. By analyzing DNA methylation patterns across the genome, researchers developed tools capable of estimating biological age with unprecedented precision. Among these, the GrimAge epigenetic mortality clock—and its recently developed successor, GrimAge2—emerged as powerful predictors of morbidity and mortality, capturing molecular changes driven by smoking, metabolic dysfunction, and other systemic stressors.

Concurrently, the field of proteomics advanced rapidly. Scientists gained the ability to measure thousands of circulating plasma proteins, leading to the development of organ-specific proteomic clocks. These tools allowed researchers to track the biological age of major organ systems, such as the heart, brain, liver, and lungs, revealing that organs within the same individual often age at vastly different rates. Despite these remarkable technological leaps, clinicians and researchers remained divided on which specific metrics—molecular, neurological, or physical—provide the most accurate and actionable assessment of overall mortality risk.

Methodology and Cohort Analysis

To settle these long-standing questions, the authors of the new study turned to the Lothian Birth Cohort 1936 (LBC1936), a uniquely rich longitudinal dataset. The LBC1936 comprises 861 individuals, all born in 1936, who underwent a comprehensive mental examination at the age of 11. Decades later, as they entered their golden years, these participants were subjected to rigorous and repeated evaluations every three years between the ages of 70 and 89.

This extensive battery of tests included proteomic profiling of major organ systems, physical function assessments (such as gait speed and grip strength), and advanced neurological imaging. By tracking this cohort over nearly two decades, the researchers were able to evaluate how effectively different classes of biomarkers predicted all-cause mortality within a homogenous population of healthy Scottish adults.

GrimAge2 and Organ Age Gaps: A Complex Landscape

The study’s initial findings reaffirmed the dominance of certain advanced molecular markers. When examining organ age gaps—the mathematical difference between an organ’s predicted biological age and the participant’s chronological age—the researchers observed that these gaps were only loosely correlated within the same individual. In practical terms, an individual exhibiting accelerated biological aging in their liver was statistically unlikely to show a similarly rapid rate of aging in their brain or heart. This supports the growing consensus that aging is not a uniform systemic collapse, but rather a mosaic process where different organ systems degrade at independent rates.

Nevertheless, when evaluated across the entire cohort, the GrimAge2 epigenetic clock emerged once again as the single metric most closely connected with all-cause mortality. Organ-specific age gaps also demonstrated robust associations with mortality risk, particularly those corresponding to the liver, immune system, heart, pancreas, and brain.

Paradoxically, however, traditional physical and neurological metrics frequently outperformed organ-specific proteomic clocks in predicting overall survival. Total brain volume, grey matter volume, overall cognitive function (quantified by the general cognitive ability factor g), and pulmonary metrics—specifically forced expiratory ratio and forced vital capacity—exhibited a stronger relationship with mortality than many sophisticated molecular clocks. Meanwhile, leukocyte telomere attrition was once again confirmed to have no statistically significant relationship with mortality outcomes in this cohort.

Unraveling Independent Predictive Power

To move beyond simple correlations and eliminate redundancy, the researchers applied advanced statistical modeling to determine which biomarkers retained independent predictive power when all other variables were accounted for.

Out of dozens of tested metrics, four stood out with extraordinary clarity:

1. White matter volume (derived from neuroimaging)
2. Total brain volume (derived from neuroimaging)
3. Walking time (a measure of physical mobility)
4. General cognitive function g (derived from cognitive testing)

Strikingly, the combination of these four non-invasive or functional markers accounted for 19% of the total variance in mortality risk within the cohort. Adding the remaining 17 biomarkers into the statistical model only marginally increased the explained variance by an additional 4%. This surprising finding suggests that macro-level indicators of neurological health and physical mobility encapsulate a massive amount of underlying physiological and biological degradation, serving as powerful proxy indicators of whole-body resilience.

Protein Signatures of Life and Death

In addition to evaluating composite clocks and physical tests, the study took a granular look at individual plasma proteins to identify specific biochemical drivers of mortality.

Following adjustments for lifestyle factors, the protein GDF15 (growth differentiation factor 15)—long associated with cellular senescence, tissue stress, and inflammation—emerged as the protein most strongly correlated with mortality risk. Across all biomarkers evaluated in the study, GDF15 was surpassed in overall predictive power only by GrimAge2. Conversely, the neuropeptide NPS (neuropeptide S) demonstrated the strongest negative association with mortality, indicating that individuals with higher circulating levels of NPS experienced a lower likelihood of all-cause death.

Broadly speaking, the proteomic analysis confirmed expected immunological pathways: proteins positively associated with mortality were heavily enriched in markers of chronic inflammation, including chemokine and interleukin signaling cascades. On the other end of the spectrum, proteins associated with extended lifespans were functionally linked to the maintenance of genomic stability, such as retroelement regulation and chromatin management mechanisms.

Implications for Clinical Practice and Future Research

While the study provides profound insights into the architecture of human aging, the authors acknowledge several important limitations that must be considered when interpreting the results. The cohort, while deeply characterized, was relatively small and restricted to healthy Scottish individuals of European ancestry, which may limit the generalizability of the findings to more ethnically diverse or globally representative populations. Furthermore, the study relied on time-point measurements rather than continuous trajectories, capturing static snapshots of aging rather than dynamic rates of physiological decline. Finally, as an observational study, it establishes strong correlations but cannot definitively prove causal mechanisms.

Despite these methodological caveats, the core findings carry sobering and transformative implications for the future of geriatric medicine and longevity science. The revelation that macro-level neurological imaging and pulmonary measurements can outperform complex, expensive organ-specific proteomic clocks in predicting overall mortality highlights the enduring value of holistic physiological assessments.

As healthcare systems worldwide grapple with aging populations, the identification of these four independent predictors—white matter volume, total brain volume, walking speed, and cognitive function—provides a practical roadmap for clinicians. Rather than relying solely on expensive multi-omic panels, future longevity interventions may increasingly integrate accessible neuroimaging and functional testing to accurately gauge patient resilience, stratify mortality risk, and evaluate the true efficacy of emerging anti-aging therapeutics.