Slowing, stopping and even reversing aging – how can we research progress if we can’t measure success?
Biomarkers of aging are pretty important in the Longevity world – after all, how can we check if anti-aging therapies work and to what degree if the only thing we can do is wait for people to die so we can see if their lives have been extended?
Biomarkers allow us to understand the biological ages of tissues and cells and compare them with chronological age. Being biologically younger than you are chronologically is big tick on the Longevity list of ‘must-haves’ and biomarkers provided a tool that allows the measurement and evaluation of effectiveness of different interventions without waiting around for a post-mortem. But what biomarkers should we use and how should we actually measure them?
This is the challenge that Longevity.Technology has set itself – setting out the journey towards a recognised, international consensus on the biomarkers of aging. It’s certainly needed; as Dr Alex Zhavoronkov, CEO of Insilico Medicine, points out: “No aging biomarker consensus or hierarchical structuring has yet emerged.” It’s not about whether we should define biological aging, it’s about how.
It’s a big ask – evaluating the current contenders (DNA methylation or other epigenetic clocks, mitochondrial health, genomic instability, frailty index, protein expression, telomere length, stem cell exhaustion, grip strength, circulating factors, autophagy…) is an undertaking in itself, but the job won’t stop there.
How often should parameters be measured to establish a baseline? What are the criteria to evaluate biomarkers – chronological age correlation? Age-related phenotypes prediction? Different interventions or lifestyle changes responsiveness?
Previously, DNA methylation was considered to be the only biomarker that could properly be used as a marker for biological age. Steve Horvath devised one of the most-used methylation clocks the eponymous “Horvath Clock”. The Horvath Clock gives a measurement of biological age that can be used in a wide range of cell types and and has been used to calculate “age acceleration” in a variety of tissues and environments.
However, a recent paper [1] claims that: “The model systematically underestimates age in tissues from older people … is consistently observed in multiple datasets. Age acceleration is thus age-dependent, and this can lead to spurious associations.”
If it is correct that the Horvath Clock loses validity for samples from older patients, it would seem appropriate to review the landscape – Dr Morgan Levine’s PhenoAge clock, or Horvath and Ake Lu’s GrimAge clock, for example. Do either clocks rival Horvath’s Pearson correlation coefficient of r=0.96 [2]?
The paper by El Khoury et al concludes: “The concept of an epigenetic clock is compelling, but caution should be taken in interpreting associations with age acceleration. Association tests of age acceleration should include age as a covariate.”
It’s a big challenge, but an exciting one. We’ll be weighing up disease and intervention relevance, the use of longitudinal data to track damage accumulation over time and trying to ascertain what existing data is most relevant.
AI will play a vital role, from genomic prediction through to studying photographs to investigate how appearance can be used as a biomarker.
We at Longevity.Technology feel a consensus is desperately needed to allow research to adopt accurate and meaningful biomarkers. As data sets increase, studies translate from animal to human and machine learning reveals new information, only with an international biomarker standard can we make progress towards Longevity. Watch this space!
[1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6915902/
[2] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4015143
