People age at different rates determined by both their environment/ lifestyle and genetics. Biological aging is the gradual and progressive decline in body integrity that occurs over time eventually leading to loss of vitality or illness/ disability in some cases. Measures of biological aging can be standardized. Given that life expectancy in regions such as Canada, Chile, Western Europe, Scandinavia, Japan and Australia is over 80 years, such standardized measures of biological aging are needed for clinical trials of interventions designed to improve healthspan, the length of time a person lives without chronic non-communicable illness.
Biological clocks based on DNA methylation have been previously developed, however they may be biased or confounded by the early-life environment of the participants. For example, differing levels of pollution exposure or nutrition. A new preprint study with senior author Terrie E. Moffitt of King’s College London, UK, released on BioRxiv, aimed to develop a DNA methylation signature of biological age in blood that was independent of biomarkers that could be altered by early-life environment. Participants who were born in the same year were selected from a long-running cohort study of people born between 1 April 1972 and 31 March 1973 in Dunedin, New Zealand, with extensive measures first taken at age 3.
The current study authors had previously shown that over the course of twelve years the pace of aging determined by 18 biomarkers was normally distributed, and showed marked variation among study participants who were all the same chronological age, 38, by the end of the analysis.
In their new study they used the previously mentioned data and combined it with whole-genome methylation data from blood at age 38 years. An algorithm called “mPoA” was derived from a machine-learning elastic net of DNA methylation patterns linked with variation in pace of aging.
More rapid “mPoA” at age 38 correlated with poorer performance at age 45 on all physical tests, with the exception of grip strength, and these participants reported more functional limitations. More rapid “mPoA” also correlated with poorer performance on all cognitive tests.
Using “mPoA” on the CALERIE Trial participants, the first randomized trial of long-term caloric restriction in nonobese adults the authors were able to show that faster baseline mPoA in control participants predicted faster biological aging over the 24-month follow-up, although this was non-statistically significant. The correlation was weaker in the calorie restricted cohort. The authors noted that the control cohort was small.
Limitations of the study include that the training cohort, Dunedin, was small and that machine learning algorithms generally improve with greater data training. It is not clear how the decreases in physical and cognitive performance that “mPoA” can predict will impact disease, disability, or mortality as no follow-up was available.
The authors concluded that “Within the bounds of these limitations, our analysis establishes proof-of-concept for mPoA as a single-time-point measure that quantifies Pace of Aging from a blood test. It can be implemented in Illumina 450k and EPIC array data, making it immediately available for testing in a wide range of existing datasets as a complement to existing methylation measures of aging. Critically, mPoA offers a unique measurement for intervention trials and natural experiment studies investigating how the rate of aging may be changed by behavioral or drug therapy, or by environmental modification.”