Bayesian Investor Blog

Book review: Ageless: The New Science of Getting Older Without Getting Old, by Andrew Steele.

The latest book on aging is a bit more ambitious than the previous two that I reviewed, but still rather modest compared to Aubrey de Grey’s book that heralded the start of serious attempts at fighting aging.

Ageless is relatively balanced, well-organized, and comprehensive.

Can Aging be Reversed?

Steele presents many hints that age-related health decline can be slowed, or maybe even reversed:

• the existence of organisms that don’t age (i.e. their mortality rate is stable or declines with age): naked mole rats, and many species of fish.
• A number of experiments have shown that small animals can be made to live longer than their maximum natural lifespan.
• the markers of age for individual cells get reset when they become a new embryo.
• cloned animals show the same resetting of cellular age, suggesting that the Yamanaka factors reset cellular age.

Hallmarks

Steele proposes 10 hallmarks of aging. These are closely related to the nine hallmarks on which many researchers seem to be converging.

How confident should we be that fixing those hallmarks will eliminate age-related health declines? I’m disappointed that Steele says little to answer this.

That contrasts with Aubrey de Grey’s argument that there are seven categories of damage that might need to be fixed in order to cure age-related health decline. Aubrey’s main argument for believing they’re the only categories that need fixing is that the last one was identified in 1972. If there were more, we’d expect scientists would still be identifying them, especially given the increase in funding for aging research.

One way that might be wrong is that 1972 was about when The Great Stagnation Strangulation slowed innovation in the US. A large centralization of funding may have caused a big increase in groupthink and risk aversion, suppressing many kinds of paradigm shifts that might be needed to find other categories of aging damage.

I don’t have any clear evidence of increasing groupthink among aging researchers around 1972. I have some evidence of groupthink, but I don’t think it has been increasing. So I’ll estimate there’s maybe a 10% chance that stagnating science is why Aubrey’s seven categories look complete.

A more likely reason for Steele to have avoided this argument is that Steele proposes three hallmarks of aging that don’t clearly correspond to any of Aubrey’s seven categories:

• Microbiome changes seem to have been one of the earliest proposed causes of aging. We have relatively good theories about how to fix the microbiome, but they’re a pain in the ass to evaluate. Aubrey might be guilty of underestimating the importance of microbiome changes, but I’m fairly sure that has few implications for the overall difficulty of fighting aging.
• Immunosenescence is hardly news. Aubrey treats it as downstream from other categories of damage. This seems like a fairly minor difference from Steele’s treatment.
• Epigenetic changes are a bigger challenge for Aubrey’s argument, and might explain why Steele avoids that argument.

Epigenetics seems not to have been proposed as a cause of aging until about 1990. How much should that weaken our confidence that we’ve identified all the factors that contribute to aging?

Part of Aubrey’s lack of interest in epigenetics is because his strategy for solving cancer ought to reverse the important epigenetic changes via supplying new stem cells.

Aubrey seems reluctant to classify epigenetic changes as damage. Maybe epigenetic changes are just symptoms of other kinds of damage, in which case Aubrey’s view will turn out to work fairly well.

But I see signs that epigenetic changes are programmed causes of aging, and that they qualify as a new category that weakens his claim that the last new category was found in 1972. Aubrey seems to be betting on a fairly pure model of aging as damage, and epigenetic changes as causes of aging don’t fit that paradigm very well. I prefer to weakly bet that some aging is programmed via epigenetic changes. So I’m glad that Steele hedges his bets on this subject.

My best guess is that if epigenetic changes are an important cause of aging, then progress will be a good deal easier than most experts predict.

Cancer

Cancer seems like it will be the hardest part of aging to solve.

Aubrey’s proposed cancer cure is a bit scary, since it would make patients dependent on getting new stem cell treatments every decade or so. That’s less scary than normal aging, but I want a good deal of research into other options.

Fortunately, Steele provides several ideas about treating cancer that are new to me, sound vaguely promising, and don’t appear to have unusual risks.

I’ve been intending, for the past two years, to take a deep dive (any month now) into cancer research, in order to better evaluate what I can do about my personal risk of cancer. A small part of my procrastination was due to not knowing where to start. Steele’s discussion provides some good starting points.

• We might increase expression of the p53 gene, combined with more telomerase. This might be how large animals such as elephants avoid the high cancer rate that theory says they should have (Peto’s Paradox)
• We might detect, and maybe treat, cancer at a much earlier stage (maybe when it’s just a millimeter in diameter).
• DNMT3A mutations appear to play an important role in cancer, while also creating important cardiovascular and diabetes risks. It’s unclear how this gets translated to treatments, but it presents an interesting target.

None of these individually are enough to get my hopes up much, and they seem to be further than I’d like from being safe treatments, but the variety of strategies is somewhat reassuring.

Why Now?

Research into doing something about aging didn’t start taking off until the 1990s.

Steele claims that was influenced by when proof became available that aging could be tractably altered in the lab in “scientifically interesting” ways.

Yet McCay’s 1935 paper on calorie restriction in rats didn’t generate interest in aging research.

There was a tiny reaction to Friedman and Johnson’s 1988 paper showing that a single mutation could slow C. elegans aging by 50%.

The trigger that, according to Steele, set off a clear movement toward aging research was Cynthia Kenyon’s 1993 discovery of another mutation that doubled C. elegans lifespan. That may have been the main advance inspired by the 1988 paper.

Unlike the first two discoveries, this one showed no trade-off between longer life and fertility. So it became hard to claim that natural aging was near-optimal for C. elegans.

That’s presumably part of how aging research became respectable, but it feels incomplete.

I suspect part of the timing was due to running low on other low-hanging fruit.

Up until the 1960s, fighting infectious disease was a more tractable way to save lives. After that, we had the war on cancer, which sure looked more winnable than a war on aging, back when the war was declared.

By the time aging research started its growth, the war on cancer had lost nearly all of its youthful vigor. The other leading deadly diseases looked almost as hard to tackle. As researchers became pessimistic about quick progress, it became harder to find research that was more promising than aging research.

Practical Advice

What can an ordinary person do now to live longer? Steele gives a mostly standard and unsurprising list of recommendations, which look about 85% correct. His top priorities are mostly things we all know we should do: exercise, eat more fruits and vegetables, brush our teeth, etc.

I got suspicious at the 7th entry in his list: wear sunscreen. I expect that will help people whose sun exposure is erratic. But I suspect the optimal advice is to, instead, get regular enough sun exposure to avoid the risk of sunburn. When combined with his advice to avoid supplements, sunscreen is likely to cause vitamin D deficiency. It’s hard to rule out other benefits of sunlight that might be impaired by sunscreen.

Another risk is that most sunscreens get absorbed into the body, and there’s some low-quality evidence that they disrupt sex hormones.

Root Causes of Aging

Steele provides a fairly standard explanation of the mainstream expert consensus, as it was several decades ago, on why aging evolved:

• antagonistic pleiotropy which is likely to be at least part of what’s going on.
• mutation accumulation used to be a leading hypothesis. Steele treats it seriously. It looks pretty unlikely to me, due to evidence that aging is promoted by genes that are conserved across a wide variety of species.
• the disposable soma theory clearly has some truth to it. Evolution does make some trade-offs between investing in reproduction versus in the health of adults. But there’s enough contrary evidence that it’s hard to consider this the main cause of aging (e.g. from experiments that increase lifespan at no apparent cost).

Steele rejects the idea that aging is programmed, treating it as if experts stopped taking it seriously long ago.

We now know that we don’t all have some ticking, internal clock, programmed to kill parents to make space for their children.

The earliest arguments for programmed aging were faulty enough that they don’t deserve to be taken seriously. Yet a recent survey (Lack of consensus on an aging biology paradigm?) suggests that about 40% of experts believe aging is programmed.

In other parts of the book, Steele describes features that seem fairly clock-like: telomeres, the programmed decline in thymus, and epigenetic clocks. When discussing these topics, he seems not to care whether he’s hinting at programmed clocks.

For each of these clock-like features, it’s not too hard to imagine explanations based on antagonistic pleiotropy or disposable soma. But that takes some effort, especially for the epigenetic clock, if we take seriously, as Steele does, the idea that those features might cause aging, rather than being just symptoms.

What about species that die too suddenly after reproducing for the death to look accidental (e.g. salmon)? Kin selection is at least close to an adequate explanation for programmed aging there. That’s probably of limited relevance to human aging, but it’s hard to fully rule out that possibility that some of our very distant ancestors evolved programmed aging, and that more recent selection pressures have not yet been sufficient to fully remove those effects.

Maybe Steele intended to dismiss only the crappy arguments for programmed aging that were somewhat popular a generation ago? It’s odd that Steele’s knowledge sounds firmly stuck in the 20th century when discussing the evolution of aging, but sounds quite up to date in the rest of the book.

Conclusion

Ageless is currently my top recommendation for a book on aging, although at the rate new books are being published on the topic, that recommendation is likely to age quickly.

It’s the most mainstream book I’ve seen on this subject. That’s a bit too mainstream for an optimal picture of aging research. But you shouldn’t expect a single book to provide an optimal picture in a field as immature as this.