I’ve been committed to understanding the biology of aging since I was a teenager, and my education and career took aim at this problem from many angles. One aspect that still perplexes me is that there isn’t a good, easily communicable answer to this simple question: What is biological aging?
When it comes to biological aging research or, to use a fancier term, translational geroscience, scientists finally have a pretty good understanding of the major components of aging. But there’s no consensus definition of it that consolidates the existing framework.
Why do we need such a definition of biological aging? A good definition can grab the essential characteristics of an entity and put them to good use. Two examples illustrate this.
Here is an example from medicine, published this month in Nature: “Cancer is a catch-all term used to denote a set of diseases characterized by autonomous expansion and spread of a somatic clone.” That is a more exact way of saying, “Cancer is a disease caused by uncontrolled division of abnormal cells.” This definition captures the universal mechanism behind all cancers. As such, it also offers therapeutic options. No matter how diverse cancers get, keeping them under one umbrella is easier compared to the broad-spectrum of biological aging.
A definition from mathematics is also instructive: “The derivative of a function is the measure of the rate of change of the value of the function dependent on changes in the input.” It is a solid definition as it offers a procedure to compute the extreme values of a function.
Here are three consecutive steps — empirical, philosophical, and computational — that can be taken to create a good definition of biological aging:
The empirical step involves collecting what is already out there. Over the years, researchers have invented their own idiosyncratic definitions of biological aging, though these generally miss parts of the story.
Scientists often start papers with a summary referring to the consensus knowledge in the field and then ask the particular question they want to address and highlight the results. These summaries, which often contain definitions, are important educational windows into science, used by mainstream media to publicize results and form relevant narratives.
To illustrate the empirical step, I extracted four definitions from scientific papers exploring different aspects of aging that reveal the conceptual mess around defining biological aging.
“Aging is characterized by a progressive loss of physiological integrity, leading to impaired function and increased vulnerability to death” came from a 2013 paper in the journal Cell by Carlos López-Otín and colleagues.
“Aging underlies progressive changes in organ functions and is the primary risk factor for a large number of human diseases” was the definition in a 2019 report in Nature Medicine by Benoit Lehallier and colleagues.
“Aging is a progressive decline in functional integrity and homeostasis, culminating in death” was used in a 2019 review of the genetics of aging in Cell by Param Priya Singh and colleagues.
Finally, a 2020 paper in Nature Medicine on personal markers of aging by Sara Ahadi and colleagues offered this: “Aging is a universal process of physiological and molecular changes that are strongly associated with susceptibility to disease and ultimately death.”
I analyzed several components of these definitions of biological aging, as indicated by the column headers in the table below, and identified some recurring themes. The final column indicates logical connections between these components.
|Source||Types of changes||Nature of changes||Association with disease||Association with death||Logical connection between components|
|I. López-Otín, Cell 2013||Loss of physiological integrity||Progressive loss||Not mentioned||Increased vulnerability to death||Causal: Loss of physiological integrity leading to impaired function and increased vulnerability to death|
|II. Lehallier, Nature Medicine 2019
|Changes in organ functions||Progressive changes||Primary risk factor for a large number of human diseases||Not mentioned||Simple logical conjunction|
|III. Singh, Cell 2019||Decline in functional integrity and homeostasis||Progressive decline||Not mentioned||Culminating in death||Causal: Progressive decline in functional integrity and homeostasis culminating in death|
|IV. Ahadi, Nature Medicine 2020||Process of physiological and molecular changes||Not mentioned||Susceptibility to disease||Ultimately death||Associative: Physiological and molecular changes strongly associated with diseases and death|
This analysis offers two lessons, one negative and one positive. The negative lesson is that some definitions have hardly any overlap, as seen in I and II — it’s apples and oranges. The positive lesson is that the recurring themes suggest the possibility of creating a core definition for biological aging using a bottom-up, empirical approach by analyzing many attempted definitions.
However, I don’t believe that such a process would be sufficient.
The myriad definitions of biological aging help identify some necessary components of it. But an aggregated mash-up won’t guarantee a formally correct and useful definition. Identifying the content itself is not enough, especially when dealing with such a complex and lifelong process. Just because we have found most of the puzzle pieces does not mean we can put the puzzle together without a clue to its shape.
This is where the philosophical step comes into the picture. Here, biologists will benefit from recruiting people trained to come up with a formal definition: philosophers, mathematicians, computer scientists, and the like.
The philosophical step involves identifying a list of criteria that a consensus definition of biological aging should meet. I believe that such a definition should meet at least these five criteria:
- exact, using a well-connected system of scientific concepts to consolidate
- fruitful or productive in terms of application to the real world
- formally correct and uncompromised by faulty logic
- simple, compared to other alternatives
- flexible and open, ready to be modified and include new knowledge.
Completing the empirical and philosophical steps would yield a good starting point for a well-formed definition that captures the essentials of biological aging.
A consensus definition that meets both content and formal criteria, achieved through the empirical and philosophical steps, might help stabilize not just scientific consensus but consensus on public policy. Here the main issues are the relationship between biological aging and disease; and regulatory, clinical, and social aspects of healthy longevity. But a completed computational step will give us actual tools, helping the biomedical technology that advances healthy lifespans.
Applicability is perhaps the most important feature of a good definition, and this where the computational step comes in. The definition should suggest future experiments and, even more important, lend itself to computability so a formal model of biological aging can be built from it. Such a model can be used to simulate and compute biological aging scores based on input data and assess the effects of planned or real interventions to slow or stop negative aging processes.
Biomedical researchers now have a solid core of knowledge on biological aging, but do not have a working consensus definition to consolidate and represent this core knowledge and capture this so far elusive life process. The lack of an unambiguous and computable formal consensus definition of biological aging severely limits the applicability of this core knowledge to design comprehensive interventions to slow or stop negative aging processes.
A confident answer to the question “What is biological aging?” in humans will help us ensure that complexity does not hide any magical mysteries. Controlling that complexity to maximize a healthy lifespan wouldn’t need a magic wand, either.
Attila Csordas is a longevity biologist and philosopher and the founding director of AgeCurve Limited, based in Cambridge, U.K.