Read More at Fight Aging!
Today’s research materials illustrate the primary challenge faced by those who want to develop and use biomarkers of aging, ways to measure biological age rather than chronological age. All reasonable biomarkers of aging are actually useful and informative when it comes to unmodified aging. This is true of everything from combinations of simple tests, such as walking speed and grip strength, through to more modern contrivances based on machine learning techniques applied to epigenetic, transcriptomic, proteomic, or other voluminous data on the state of our biology that can be easily produced these days. One can see clear correlations between these biomarkers and mortality, and between these biomarkers and state of health in later life.
The problem arises once we start considering the effective treatment of aging as a medical condition. Not aging as a whole, of course, because aging is a set of diverse processes and their consequences that are very different from one another, and require very different strategies in order to build meaningful therapies. When testing a rejuvenation therapy that repairs one form of damage, or reverses one process of the many processes of aging, how will that treatment affect measures of aging? The answer probably differs on a case by case basis, and at present, despite the existence of at least one approach to rejuvenation that has actual, working therapies, meaning the clearance of senescent cells via senolytic treatments, there is no map to connect treatment to effect on biomarkers.
In practice, this means that biomarkers can’t be trusted as tools to evaluate whether or not potential rejuvenation therapies are actually any good, at least until after researchers have run many life span studies using both those therapies and those biomarkers in order to produce a calibration. That will take a good deal of time and effort, and makes biomarkers for aging a very much less useful than hoped at the present time.
Biomarkers used to track benefits of anti-ageing therapies can be misleading
The nematode C. elegans begin adulthood vigorously exploring their environment. Over time, they slow and stop crawling, a behavioural stage known as vigorous movement cessation (VMC). VMC is a biomarker of ageing and a proxy for nematode health. Studies of genetically identical nematodes have shown it is a powerful predictor of a worm’s lifespan, but at the same time, interventions designed to alter ageing can disproportionately affect VMC in comparison to lifespan and vice versa.
Researchers developed the ‘Lifespan Machine‘, a device that can follow the life and death of tens of thousands of nematodes at once. The worms live in a petri dish under the watchful eye of a scanner that monitors their entire lives. By imaging the nematodes once per hour for months, the device gathers data at unprecedented statistical resolution and scale. The research team found that nematodes have at least two partially independent ageing processes taking place at the same time – one that determines VMC and the other determines time of death. While both processes follow different trajectories, their rates are correlated to each other, in other words, in individuals for whom VMC occurred at an accelerated rate, so did time of death, and vice versa. In other words, the study revealed that each individual nematode has at least two distinct biological ages.
The study calls into question a crucial assumption of ageing biomarkers, that when interventions such as exercise or diet “rejuvenate” a biomarker, it’s a good sign that the underlying biology of ageing has similarly changed. “Our model shows that biomarkers can be trivially decoupled from outcomes because they measure an ageing process that is not directly involved in the outcome but simply correlates with it in a system of hierarchical processes. In simple terms, just because two parts of an individual tend to correlate in their biological age across individuals, it doesn’t mean that one causes the other, or that they are likely to involve shared ageing mechanisms.”
A Hierarchical Process Model Links Behavioral Aging and Lifespan in C. elegans
Individuals who remain vigorous longer tend to live longer, supporting the design of predictive behavioral biomarkers of aging. In C. elegans, the timing of age-associated vigorous movement cessation (VMC) and lifespan correlate strongly between individuals. However, many genetic and pharmaceutical interventions that alter aging produce disproportional effects on VMC and lifespan, appearing to “uncouple” the rate of behavioral aging and lifespan. To study the causal structure underlying such uncoupling, we developed a high-throughput, automated imaging platform to quantify behavioral aging and lifespan at an unprecedented scale.
Our method reveals an inverse correlation between each individuals’ vigorous movement span and their remaining lifespan. Robust across many lifespan-altering interventions, our data shows that individual C. elegans experience at least two distinct but coupled physical declines – one governing VMC and the other governing lifespan. Through simulations and modeling, we clarify the causal relationship between these two “biological ages” and highlight a crucial but often untested assumption in conventional aging biomarker research: predictive biomarkers may not always report on the same biological age as that which determines long-term health outcomes.