What follows are some thoughts on the question "what is aging". Initially there is a meandering back and forth around the concept using a non-biological example (pyramids), later we get to a more concrete definition of aging, and using that to cast some light on various questions like whether aging is a disease or not.

You can see a video companion to this essay here.

A paper from last year (Cohen et al., 2020a) pointed to the lack of consensus within gerontology ("aging research") as to what aging is. Multiple researchers consulted offered different definitions, some of which are conceptually different, with some researchers stressing the role of reproduction (aging starting after the reproductive age, an extreme case being Pacific salmons, which die shortly after laying eggs), others mentioning programmed factors, damage accumulation, or homeostasis decay.

Shortly after we got another paper from one of the same authors, Cohen et al. 2020b that argues that maybe aging doesn't refer to anything, that there is no such thing as aging.

A coherent definition is even essential for the field: there are intensive efforts to measure aging, to slow aging, and to treat aging, and it will be impossible to know if they are succeeding without a clear definition of the subject of our research. Is it accumulation of molecular damage? Is it loss of function with increasing age? Is it increases in mortality (or decreases in reproductive rate) with age? Underlying the discussion to date is an assumption so basic it goes unnoticed: that there is an underlying biological phenomenon of aging. We have a word for aging, and therefore we assume that science will accommodate us, providing a phenomenon to match our word. And in a colloquial sense this is certainly the case: no one can doubt that we see ourselves, our relatives, and our friends age. But is this colloquial usage scientifically justified? Is there really a “thing” or a phenomenon we can call aging? We argue here that our understanding of the biology is now sufficient to say definitively that this is not the case, that from a scientific perspective there is no such thing as aging, but rather a collection of disparate phenomena and mechanisms – sometimes interacting with each other – that relate in one way or another to our colloquial sense of the word. Accordingly, our desire to find a single reality of aging has created a great deal of confusion in the field. [...]

we are not arguing that the term “aging” should never be used. It will certainly continue to be used colloquially, and it may be useful to retain it to describe our field, or in certain situations when the very broad set of phenomena is really of some interest. But we believe scientists should move away from the term, particularly when they are only referring to a subset of the phenomena in question. For example, when we mean “damage accumulation,” we should say this rather than “aging.” Likewise, when we mean age-associated increase in mortality risk, we should find a term such as “demographic aging” or “Gompertzian mortality patterns.”

The paper itself seems to be framed as a natural extension of recent directions in the field; whereas years ago there were a bunch of seemingly monocausal theories of aging (oxidative stress, or telomere shortening being some of the most popular ones) now the field has moved towards an understanding of aging as made up of different processes that are interconnected. The authors are not arguing that these are not connected, so it's a bit unclear in what sense their analysis is an extension of said trends. But they do say that they would admit that aging is a unitary phenomenon (I take that to mean a phenomenon worthy of its own label) if at least there were a single upstream mechanistic cause that is always present in aging. The only argument they present against there being such phenomenon is that "it seems highly unlikely (though not impossible) that so many biologists working so hard for so long have missed a single upstream cause that would neatly unify our field". I'll show later that it is not so hard to find unitary causes that can be integrated in a coherent explanation of aging. But even if we couldn't find such single cause, if there is a family of similar phenomena, it may still make sense to call something "aging". See Freund (2019) for some of this thinking.

Ultimately, they do not argue in favor of abandoning the core idea in Geroscience (That a given intervention can impact multiple diseases as one target can sit upstream of multiple diseases), but they argue that on their view, rather than thinking of, say, metformin as something that can be broadly administered to prolong lifespan, it has to be targeted to a subset of the population

This example is telling: suggesting that there is no such thing as biological aging is not a brake on research, but helps fine-tune the question. If metformin is “anti-aging” and we all age, everyone could take it and expect to slow aging. However, if metformin makes specific modulations to the signaling pathways in Fig 1. with a diverse set of impacts on the outputs, a careful understanding of who it affects how is necessary before proceeding. In this sense, abandoning the notion of a unitary aging process may be necessary for the safe deployment of multi-disease therapies, and may save the Geroscience movement the profound embarrassment and setbacks that could come from an overly broad application. [...]

Here we have argued that the concept of aging does not reflect any underlying biological reality, and has in fact hindered progress in our field. Aging is not like development, nor like a clear pathological process such as Type-II diabetes, and our attempts to treat it as such explain why we have had so much difficulty arriving at a common language and definitions, and at integrating across different perspectives, including ecological, evolutionary, molecular, genetic, mechanistic, epidemiological, demographic, and clinical. [...]

Nonetheless, we believe that even facultative definitions are dangerous: we have often seen that a very circumscribed definition of aging in one paragraph is followed by a much less circumscribed application of the concept, showing how profoundly our linguistic categories influence our thinking. This will not be easy, given our identity as “aging researchers” and the depth with which the concept is anchored in our field and in the culture more broadly. Nonetheless, the biological arguments against aging as a unitary phenomenon are clear-cut, and the advantages to the field both numerous and important. Indeed, as the field moves toward potential anti-aging therapies, large numbers of human lives may depend on fully knowing what we are measuring and what the risks are of getting it wrong.

Their argument, that aging is not a single phenomenon, and that even having the concept of aging may bias us is reminiscent of my recent post on the great stagnation, where I argued something similar, that the narrative nudges one towards one direction and not others (That I regard as more fruitful).

But here I want to make the case for the opposite, that there's more to aging than they make it seem and that on the margin it is useful to have a unified concept of aging. They are, to be sure, absolutely right that aging is a collection of processes that can be singled out, some of which are species-specific. For example, if we take atherosclerosis (The buildup of plaque inside the arteries over time), that does not happen in C. elegans simply because they do not have blood. Most animals don't live long enough to get Alzheimer's like we do. If a handful of cells die in C. elegans the whole organism dies whereas in our case our stem cells will replenish them so that's less of a problem.

We can ask: Does aging cause anything at all? Once we catalog age-related diseases is there anything left for 'aging' to describe? What would it mean to treat aging? Does aging have a cause? To answer we need something for that 'aging' word there to refer to.

First, I think it is helpful to discuss aging in a generalized way and applied to a simpler system than the usual system in biology; so do pyramids age?

We could think of something that we might still can aging in pyramids. Pyramids were initially built in a given shape and over time they have lost some of that shape due to erosion and theft of the stones that comprise the pyramid themselves.

An aged pyramid?

In what sense has the pyramid "aged" or decayed? Could we swap the pictures and argue that the shining, more smoothly platonic pyramid is the one that is old. Certainly when the pyramids were being built they were not as nice and smooth as they would end up being when finished.

Aging, as is the case in biological system, is not the same as chronological aging, which is just counting how long the system has been around. I argued in a prior post that some fields have built in values or teleology. Aging and health are such value-laden concepts. For example, one could take the fact that one keeps growing in height after childhood and argue that that increase in height is a bad thing (e.g. because it imposes the need for larger, more expensive, beds, or roomier airplane seats) and consider that height increase as part of a decay process away from the 'healthy' shorter stature. Similarly one could take an age-related change (gaining memories) as a being part of aging if one considers that having that mental baggage is bad (Maybe smooth brains are good).

Most people wouldn't think of height that way because height may also be good for other things (e.g. reaching high shelves), but there is a set of capabilities that is commonly regarded as good and mostly lacking in tradeoffs. Possessing these is what is commonly referred to as being healthy.

For the pyramid we could then define a 'healthy' pyramid by a set of parameters (Like being relatively smooth rather than ragged and being bright in color rather than dull, if that's what their architects intended).

Why would a pyramid age? I mentioned earlier two reasons why the pyramid transitioned from one state to another: First, due to constant sand abrasion over the centuries. Second, due to generations of stone theft.

But consider what would happen if someone drops a nuclear bomb on the pyramid, or if it gets dismantled in a month by means of heavy machinery in a sort of it's time to build crazy, but backwards. We probably wouldn't consider that aging for the same reason we wouldn't consider dying of ebola aging. Sure there has been damage accumulation, as a popular account of aging goes, but there seems to be some element missing in that story.

Before continuing, it's worth noting that the kind of answer we are after is not the same kind of answer we would be after in physics. In physics we could reasonable expect to find some crisply defined formula or formalism that describes the problem. For example the Standard Model says there is a given finite number of particles which interact in very precise ways. A proton is a proton wherever it is. Aging in contrast is more of a nebulous concept. It's the kind of thing we can more or less describe, and the kind of thing is useful to talk about for certain ends, but not the kind of thing that we can crisply define as we can for the basic building blocks of physics. Asking of nature to give us a crisply defined phenomenon of aging is like asking an elm tree for pears. Insisting on that is the kind of thinking that leads one to the 'there is no such thing as aging' thinking. Conversely, trying to impose some very opinionated structure onto aging (aging is damage accumulation) runs into inevitable counterexamples (ebola, childhood cancers), so all we can hope to do is examine the biological phenomenon of aging as it manifests in various forms of life, and then reflect on the socially accepted meaning of the word aging, and making sure we all mean the same thing when we use the word.

With ebola, we can either consider that dying of ebola is indeed accelerated aging or we can exclude that from the phenomenon of aging. If this latter, then we need to explain why. The why may involve a reference to its suddenness (It's a large change, from health to death), or not being widespread (Only a small number of people will experience ebola), or being a cause that is external to the agent (When we think of aging we think of a process of internally driven deterioration). This latter is not fully right because you could imagine as well a scenario where certain epigenetic marks causally drive aging and intervening to add them could cause accelerated bona fide aging. In that case, if someone hypothetically infects you with a modified virus that sets the methylation marks to that of an aged person, you would end up aging and dying faster (If those methylation marks are causing aging rather than being correlated with aging).

The reason, then, why ebola wouldn't be accelerated aging is that there are many ways of breaking a system, and aging is just one of them. Inasmuch as a system breaks in a very narrowly defined way without concurrent systemic degradation, we will be inclined to say that such a thing is not aging. Someone dying of ebola won't necessarily have, say, systemic formation of crosslinks or a brain ridden with amyloids.

Another thought experiment here would be imagining someone who is in all regards young, and yet their LDL cholesterol rises over time, leading to atherosclerosis and then cardiovascular disease. Their cognition, and muscle strength and hair color all would look young, but their mortality risk would increase exponentially regardless. That wouldn't be because of aging, but because of a very narrowly defined deterioration process. With the pyramid we can still talk about aging despite its aging being simple to explain: in that case systemic is almost equivalent with a single process.

I said earlier that one reason we may not consider ebola as part of aging is that it is not widespread; that it makes sense to think of aging as if ebola didn't exist. This is the same reasoning that is typically used to exclude aging from the category of diseases.

Of course whether or not aging is a disease is itself its own protracted philosophical discussion fraught with better or worse arguments. For example Fulop et al. (2019) consider that aging is not a disease because

First, the extent of aging is systemic and complex while that of a disease is mostly limited. Aging is an inevitable, universal process (concerning all humans living long enough) while most diseases are associated with individuals’ susceptibilities/vulnerabilities, and most of them, even chronic, are preventable. The most important cause of aging is time, while diseases usually have specific known causes. In other words, aging is irreversible and progressive while diseases are reversible and discontinuous. Finally, and most importantly, aging may be modulable but not treatable, while diseases are ultimately treatable even if we do not know presently how, which is only a question of progress of science. So many essential differences clearly speak against the notion that aging is “just another” disease.

This is however not true in that many things, including the leading causes of mortality that we regard as diseases will probably occur given enough time, as Alzheimer's or cancer, or cardiovascular disease. Second, aging need not be irreversible. It may be hard to reverse, but that is a technical, not intrinsic impossibility per most theorizing on what aging is. Some–admittedly simple–animals seem to be able to avoid, or even reverse aging (Fahy, 2010. Chapter 6). Similarly, diseases are not discontinuous (having or not having a disease). There is a reason why there exists a category of pre-diabetic in-between healthy and full onset of the disease. Perhaps for medical purposes one has or doesn't have a disease if a doctor has determined so, but biology hardly operates in such binary terms.

The authors are right though that aging is not just another disease. And indeed the kind of thing it is is unlike other diseases, as it subsumes most of them (It wouldn't include, for example, sickle cell anemia, that is too rare a disease and aging has some pretension of being, agreeing with the authors here, universal to a class of systems).

Returning to the pyramid example, if in the biological case we want to reject ebola, should we reject "aging" by means of people stealing stones? That sounds like a substantially less gradual form of aging relative to erosion by sand and wind, just like ebola is a sudden onslaught of damage accumulation.

This could be addressed by defining some measure of "normal environment" where the system is supposed to be healthy; an obese person may be unhealthy but if a famine is about to strike maybe the obese person survives and the fit athlete dies, so perhaps in such environments we would think of being somewhat fat as a good thing, just in case. This notion (that health, and thus aging are relative to an environment) is similar to what François Chollet (2019) has said about intelligence, that we may talk about general intelligence, but that superior intelligence is so relative to a given (complex) environment.

An example of something that is external to us, and not necessarily present in everyone that we may want to consider as part of aging is infection by CMV. Some have argued that this plays some part in the weakening of the immune system over time. Even if not true, it could have been conceivable true. However, infected with CMV or not it seems that aging does not progress that differently so considering CMV infection as part of the natural aging process is not very relevant.

So far I have not talked about homeostasis, or mortality rates. I don't think these concepts are central to aging; one could have a very robust system that gradually fails or one could have a dynamically repaired one that ends up in the same aged state. Biological systems are a mixture of both, with most of it being relatively dynamically repaired but having some elements that are not (like our teeth).

What does seem key to the concept of aging is, again, systematicity, the idea that aging is something that affects a vast array of subcomponents of the system, it is generalized, perhaps even uniform degradation rather than a one hit kill. There are many ways a biological system could be taken out of the region of functions we call 'health', a limb could be amputated, or an organ removed, a lethal disease could be inoculated, or one could inject air into the arteries, one could ingest poison. Here the causes would be blatantly external, and focused on the dramatic failure of one system. Aging is more about the slow failure of an interrelated network of system that make up a living being.

Causality in biological systems is difficult to ascertain. During aging, senescent cells accumulate, generating inflammation. Inflammation may lead to senescent cells. Other things may lead to inflammation as well. Brain aging could lead to malfunction of the immune system. Is the brain causing senescent cell accumulation? Is senescent cell accumulation causing brain aging? These questions are not very clear, and one would need a perturbations approach to investigate them. What is clear is that everything is connected, and the ways in which this is true become more intriguing the more one learns biology.

Lastly, one more consideration to include is the root causes of aging. (Closed) Systems decay by default, everything tends towards bland mush unless there is a reason for it not to. That's the first explanation. The other two is that the system itself may by its design damage itself in a programmed, regulated, intentional way (e.g. see Ezcurra et al. 2018), and that the same can be true but externally. A car may rust in the absence of efforts to maintain it, but you could also imagine a car that when it registers 20k miles, a system in it causes a rapid planned disassembly of the engine. Similarly you could imagine someone smashing the car. This latter would be the stone robbers in the pyramid case, or a lifelong parasitic infection in the case of humans. All external forces that would consistently tend to drive the system out of the healthy area would be in this third class. These are the three potential root causes of aging: noise, and internal and external forces that directly push the system away from health. Because biological systems evolve to last for a while in their environment, the environment itself will rarely be the cause of aging, at least if an organism is sufficiently long lived. The environment may periodically consist of nasty diseases but an evolved organism will resist and eventually recover from most of these. This is not to say that this does not matter at all: life expectancies are shorter there where there are high parasite loads. But it means that one could live in a perfectly disease-free environment and still recapitulate regular aging. The environment can accelerate aging somewhat, but the process requires the first two causes to be present.

What is aging in biological systems?

Aging is the systemic drift of an organism away from (a state defined as ) healthiness.

Important here are:

  1. Systemic: It's not just one thing that breaks down. Aging implies that everything (or most things) break down. In particular, mechanisms that are common to many systems break down. The breakdown of something that is not broadly connected to the rest wouldn't strongly count as aging.
  2. Relative to a state of health: We define some notion of health for an organism. This value judgement enables us to define something as aging, rather than as just the value-neutral evolution of the system. Development would be the process by which an organism gets to the state defined as healthy.

What are the root causes of aging? 1. Noise, or randomness, that leads to substances and changes that the system cannot fully reverse (misfolded proteins, epigenetic alterations, loss of stem cells). And 2. Quasi-programmed patterns (like thymic involution) or programmed ones, though these are not very relevant (like telomere shortening) that deliberately pushes an organism towards frailty and death. Both the hallmarks of aging and the SENS damage categories are downstream of these two causes. Once one moves past the root causes, causality gets complicated. One could add a minor cause 3. Which is low-grade chronic exposure to various forms of disease, but this is not required for aging. This may be a argued to be more relevant for example in the case of C. Elegans which die in 2 distinct ways, one of which being E. Coli taking over the worm's pharynxes (Zhao et al., 2017)

If we address all the SENS damage categories are we done? Probably, they are quite comprehensive. But the problem is that each category hides a lot of complexity inside, meaning that it won't be a matter of using one drug to target each. For example, both transthyretin amyloidosis in blood and the buildup of tau and beta amyloid in the brain are cases of extracellular aggregate formation, yet they are in different places and have different form. Immunoclearance may work for both, but for blood ones simple plasmapheresis may do instead.

Are the root causes of aging what has to be addressed to address aging? Not necessarily. In fact, it's not possible to fully eliminate at least the stochastic aspect of biological systems. We can hope to reduce but never eliminate noise; this is what mTOR down regulation does as well as say having the more efficient ribosomes (Azpurua et al. 2013) found in naked mole rats, and we can try to find ways to prevent thymic involution. But there are lots of points of intervention: DNA mutations may still occur but we can respond to that with better immunotherapies and gene therapies. Senescent cells may be inevitable in some amount but we can develop better senolytics. Unless we radically change our metabolism, cholesterol will get stuck in our arteries, but we can use cyclodextrins to get it out.

Is aging a disease? To me, it is. For most people, disease involves some degree of abnormality. Aging is universal, hence normal. But if one takes normal to be health and aging to be an alteration away from that then aging is an abnormality in search of a cure. Aging as a disease represents raising our expectations about what health (or healthy life) means whereas aging as not a disease is closer to accepting the view that the role of medicine ends at living within our evolutionarily-given constraints.

Does geroscience, or the study of "aging" as distinct from individual diseases, make sense? Again, yes. One can meaningfully study a single disease but the idea that all if not most diseases are connected is true, and very powerful. It has practical implications: With the geroscience glasses on one will go around looking for unexpected connections between different systems that break down, trying to find causally interesting points to intervene.

Why do all age-related diseases occur at once? Not because they have a single cause (Though that is a tempting thought), but because evolution has pushed them further in time to extend our lifespan to whatever point increased fitness. If say we "need" to last 50 years and we get lots of cancer at age 20 then evolution would add in cancer resistance up until cancer is pushed back to a later age; same reasoning applies to other diseases. The breakdown of the various systems that we can say comprise aging tends to occur at around the same time, but this correlation is driven by evolution, not a single master clock (as far as we know).

Is cancer caused by aging? Yes. The origin of cancer lies in a combination of DNA mutations, epigenetic alterations, and a weakened immune system that cannot fight cancer when it is nascent. These are all caused by the same two root causes of aging.

Are there diseases that are not caused by aging? Yes, childhood cancer, rare diseases like progeria, or infectious diseases. These are either sporadic, or programmed (If one was born with the mutation). Mortality due to infectious disease increases with aging because the immune system is weaker.

What is the role of free radicals, telomere shortening, or even disintegration of the 'self'/cooperation, etc? Over the years multiple theories of aging have been proposed. None of these are true in the sense that they point to a closet set of root causes. Various theories of aging can describe processes upstream of the two root causes of aging. What I've said here stops short of that. This is because whereas I think there is a lot of aging that is universal and I've talked about that here, the specific way in which aging progresses is very species-specific.

What should we think of progeroid syndromes? Progerias are not aging, they are diseases that recapitulates some of the features of aging at an accelerated pace. It still makes sense to use the lens and concept of aging when thinking about these, given that the causes of these diseases lie in the same set of core mechanisms present in all cells that break down with age (Like DNA repair, all cells engage in it).

Is aging research underfunded? Yes. But one has to be careful there: If you see for example this video, Kristen Fortney says only 1% of the NIH's budget is spent on aging research, which is what the NIA's Aging Biology division spends. Of course, the real situation is not that grim. If aging is an elephant, we may not be spending a lot of money looking at the elephant as a whole but we sure are looking at all the parts of the elephant separately. Research on cancer, Alzheimer, or immunosenescence separately is in part aging research.

In humans, is aging mostly noise, or more of a quasi-program? I don't know! I hope that once we get a better causal understanding of epigenetic clocks and examine whether aging can be slowed down, stopped, or reversed by directly targeting methylation sites, we'll also have an answer for this.

Can aging be slowed down? Yes. Caloric restriction is an example of a way to do exactly this.

Can aging be reversed in humans (or in mammals more broadly)? In principle yes, in practice it's not physically unfeasible but whether it can be done remains an open question. Luckily measuring the success of this is easier than say measuring whether string theory is correct: All it takes is an intervention and looking at the mortality rates of a cohort of mice, observing that they don't rise exponentially with age. Didn't Fahy (2019) show a reversal? No, they showed a reversal by a epigenetic clock. These are plugged to somewhere in the concept of aging but the extent to which the reversal is systemic vs just partial remains to be seen.

If it is possible to reliably reverse aging, have we ended all disease? No, because there will still be lots of causes why one may die that are not aging-related. I mentioned some earlier: Childhood cancers, or very nasty infectious diseases. Those need addressing as well.

What do anti-aging interventions look like? A grab-bag of things. Some will target processes that are very core to an organism. All of our cells need to make proteins and enhancing proteostasis will then have effects throughout an organism. But other interventions will be more targeted (Like cyclodextrins to reduce plaque in arteries). Anti-aging companies like Unity may have anti-aging interventions that are designed to target a single symptom of aging. For aging in general it is likely that a targeted, perhaps individually tailored, combination of interventions, ranging from small molecules to gene therapies, will be required.

What does aging look like? Aging looks like an entangled set of nodes that can be more or less decomposed into subsystems. Different nodes are tied to different individual diseases. Each node represents some part of an organism.