DunedinPACE Epigenetic Clock Outperforms Clinical Measures in Mortality Prediction

It comes as no surprise that epigenetic clocks offer superior predictions of mortality risk compared to straightforward clinical assessments, such as isolated biomarkers reflecting age-related chronic inflammation. These epigenetic clocks draw from a vast array of data points, and certain ones, like

It comes as no surprise that epigenetic clocks offer superior predictions of mortality risk compared to straightforward clinical assessments, such as isolated biomarkers reflecting age-related chronic inflammation. These epigenetic clocks draw from a vast array of data points, and certain ones, like the Dunedin Pace of Aging clock, were specifically engineered with the goal of forecasting mortality outcomes. That said, empirical validation through rigorous analysis remains essential. In this investigation, scientists evaluated the predictive capabilities of diverse measures and their combinations against mortality in an extensive human epidemiological dataset. The findings reveal that the epigenetic clock surpasses alternative approaches, though its performance is further enhanced when integrated with select clinical indicators.

Comprehensive Analysis from the Berlin Aging Study II

Researchers drew upon data from the Berlin Aging Study II (BASE-II), involving participants aged 60 to 80 years at baseline, with an average follow-up duration of 7.4 years (standard deviation ±1.5 years, ranging from 3.9 to 10.4 years, and a total sample size of 1,083 individuals). This study compared 14 biomarkers of aging, which had been recently endorsed by an expert panel for application as outcome measures in intervention trials. These biomarkers encompassed a wide range of categories: physiological markers such as insulin-like growth factor 1 (IGF-1) and growth-differentiation factor-15 (derived from DNA methylation, known as DNAmGDF15); inflammatory indicators including high-sensitivity C-reactive protein (CRP) and interleukin-6 (IL-6); functional assessments like muscle mass, muscle strength, hand grip strength (HGS), Timed-Up-and-Go (TUG) test, gait speed, standing balance test, frailty phenotype (FP), cognitive health, and blood pressure; as well as epigenetic measures represented by the epigenetic clock and DunedinPACE.

To examine their predictive power, the team employed Cox proportional hazard regression analyses, focusing on all-cause mortality as well as mortality due to specific causes. All models were meticulously adjusted for key confounders, including age, sex, lifestyle factors, and genetic ancestry, ensuring robust and reliable insights into the independent contributions of each biomarker.

Key Findings on Predictive Strength

Within the adjusted models for all-cause mortality, several biomarkers demonstrated statistically significant associations with mortality risk: hand grip strength (HGS), interleukin-6 (IL-6), standing balance, cognitive health, and notably the epigenetic clock (DunedinPACE). Among these, the DunedinPACE epigenetic clock stood out as the most potent predictor, exhibiting the strongest correlation with mortality outcomes. In contrast, other markers such as CRP, gait speed, IGF-1, blood pressure, muscle mass, DNAmGDF15, frailty phenotype (FP), and TUG did not show significant associations with mortality in this cohort.

These primary results held firm across subgroup analyses that stratified participants by specific causes of death, underscoring the consistency and reliability of the observations. To optimize predictive utility, the researchers conducted feature selection procedures, identifying a streamlined set of biomarkers comprising muscle mass, standing balance, and the DunedinPACE epigenetic clock. This minimal combination achieved a discriminative accuracy (C-index of 0.63) that was nearly identical to the comprehensive model incorporating all 14 biomarkers (C-index of 0.65), demonstrating that a focused panel can deliver comparable performance without unnecessary complexity.

Implications for Aging Research and Clinical Practice

In summary, out of the 14 biomarkers of aging scrutinized in this study, the DunedinPACE epigenetic clock displayed the most robust and consistent link to mortality risk. This superior performance highlights the value of epigenetic measures in capturing the multifaceted biological processes driving aging and its consequences. Unlike traditional clinical metrics that often focus on singular aspects of health decline, epigenetic clocks integrate thousands of molecular signals from DNA methylation patterns, providing a holistic gauge of biological age progression.

The DunedinPACE, in particular, quantifies the pace of aging rather than just chronological age, offering dynamic insights into how rapidly an individual's physiology is deteriorating. Its design, rooted in longitudinal data from the Dunedin Multidisciplinary Health and Development Study, was explicitly optimized for mortality prediction, which explains its edge over static biomarkers. This study's use of a well-characterized cohort like BASE-II, with detailed follow-up data, strengthens confidence in these conclusions, as it minimizes biases common in shorter-term or less comprehensive epidemiological efforts.

While individual clinical measures like grip strength and inflammatory markers provide valuable information, their predictive power is limited by their narrow scope. Grip strength, for instance, reflects sarcopenia and overall physical function but misses systemic molecular changes. Similarly, IL-6 indicates inflammaging but does not encompass the full spectrum of age-related dysregulation. The epigenetic clock's ability to synthesize these and other signals into a single, powerful metric positions it as a game-changer for risk stratification in aging populations.

Looking ahead, these findings advocate for incorporating DunedinPACE into routine assessments for older adults, particularly in clinical trials targeting age-related diseases. Combining it with a few high-yield clinical tests, as shown here, could refine patient selection, monitor intervention efficacy, and guide personalized medicine strategies. As research progresses, validating this clock across diverse populations and integrating it with emerging multi-omics data will further solidify its role in advancing longevity science and improving healthspan outcomes.