LONGECITY

There are many theories of aging, a state of affairs that I would say is really due to past lack of resources put towards building means to treat aging in a targeted way, by addressing specific purported root causes. There has been a great deal of investigation of the biochemistry of aging, and will continue to be given its complexity, and all too little bold experimentation in means to extend healthy life spans. This was in large part cultural for the last generation of researchers, a way to reject any association with the fraud and self-deception of the "anti-aging" marketplace, and thus preserve reputations and the ability to raise funding for legitimate studies of aging. At the same time, however, this meant that greater progress towards longevity enhancement might have happened and did not: talk of treating aging became a threat to careers, an unfortunate state of affairs that has only recently abated.

Ways to increase longevity in laboratory species are the tools by which theories of aging can be winnowed and validated. Unfortunately all of the present means of slowing aging are far too general in their operation to serve as good tools in this sense. Take calorie restriction and its alleged mimetic drugs, for example: these approaches change near every measure of metabolism, to the point at which it remains an enormous puzzle to figure out how and why they act to slow aging. What is needed is a new generation of much more targeted therapies, things like clearance of senescent cells, or clearance of cross-links, or other proposed SENS biotechnologies based on the repair of one single type of tissue damage thought to cause aging. Build the treatment, run the experiment, and a lot will be learned from whether or not it does extend healthy life. If great progress is made by repairing forms of damage thought to cause aging, that tells us that theories painting aging as a process of damage accumulation are more robust and defensible. The types of damage being repaired and the results obtained will help separate out which theories on various types of damage are more robust and defensible.

Ultimately theories of aging matter today because they are used to steer investment in research. This will continue to be important until one group of theories wins out by weight of evidence obtained through extending life in laboratory animals. At the moment the division of greatest importance is between programmed aging theories and stochastic damage theories. Programmed aging theories would have us think that aging is the direct consequence of a set of evolved changes in (say) gene expression and protein levels and cellular operation, and these changes causes damage, dysfunction, and death. The right thing to do if this is true is to work to alter the operation of metabolism, change the gene expression levels, manipulate specific protein levels, to bring them back to a more youthful pattern. In contrast stochastic damage theories of aging tell us that aging is caused by what is effectively biochemical wear and tear, and our bodily systems react to the presence of that damage with altered gene expression and protein levels and cellular operation - but it is the damage that is the root cause of disease and dysfunction. The way forward if this is the case is to repair the damage.

Personally, based on my view of evidence to date, I'm in the latter camp: aging is stochastic damage accumulation. Repairing that damage if following the SENS proposals is very cost-effective, and producing full demonstrations of the various treatments in mice is a $1-2 billion, 10-20 year project at full scale funding. That's less than the cost of developing a single drug candidate in the Big Pharma world these days. Programmed aging on the other hand would direct us into a massive, unending project of trying to fully understand and safely alter swathes of our metabolism. That is a vast project. To give some idea of the scale, about a billion dollars has been swallowed up on research of sirtuins in aging over the past decade or so - just a couple of genes out of thousands worth looking at, and nowhere near a full understanding of their role yet, and no meaningful treatments or ways to alter metabolism resulted from all of that work. So from my perspective, I see programmed aging theories as the road into an endless swamp, a course that might be averted at a low cost by making enough progress on SENS rejuvenation treatments in the laboratory to demonstrate their worth in extending healthy life spans, and thereby showing that aging is mostly likely a process of stochastic damage.

Here is a better than average popular science article that covers many of the aspects of this debate over theories of aging, using a recent paper on programmed aging as a springboard. You should read the whole thing, given that the author took the time to gather opinions from various researchers with different takes on aging, who think that this particular line of research is flawed, and goes on to examine the point I make above, which is that all this theorizing is far from idle and unimportant. It in fact determines the prospects for the near future development of effective means of treatment for degenerative aging, with both sides believing that their road is the more effective one, but only one of them being right:

Are Limited Lifespans An Evolutionary Adaptation?

That aging is a deliberate function of our genetics remains a controversial idea, but it's an idea that's steadily acquiring adherents. One of these adherents is NECSI president Yaneer Bar-Yam, who contends that popular approaches to the aging problem fail to address a very important constraint, namely the ways lifespans are genetically controlled according to the resource limitations of a given environment. Without genetically programmed aging, he argues, animals wouldn't be able to leave sufficient resources for their offspring. And this holds true for all animals, whether they be rabbits, dolphins, or humans.

Bar-Yam and his team reached this conclusion by developing a simple model that analyzed how the lifespans of simulated organisms would change and evolve over time under spatially constrained conditions. Fascinatingly, group selection -- the idea that natural selection acts at the group level -- was never a consideration in the model. Yet the simulations consistently showed that a built-in life expectancy emerged among the simulated organisms to preserve the integrity of their species over time. This is surprising because a pro-group result was produced via an individualized selectional process.

"Beyond a certain point of living longer, you over-exploit local resources and leave reduced resources for your offspring that inhabit the same area," Bar-Yam said. "And because of that, it turns out that it's better to have a specific lifespan than a lifespan of arbitrary length. So, when it comes to the evolution of lifespans, the longest possible lifespans are not selected for."

Programed Death is Favored by Natural Selection in Spatial Systems

Standard evolutionary theories of aging and mortality, implicitly based on mean-field assumptions, hold that programed mortality is untenable, as it opposes direct individual benefit. We show that in spatial models with local reproduction, programed deaths instead robustly result in long-term benefit to a lineage, by reducing local environmental resource depletion via spatiotemporal patterns causing feedback over many generations. Results are robust to model variations, implying that direct selection for shorter life span may be quite widespread in nature.

This is the exact same simulation-based argument that Mitteldorf makes, and I don't buy it.  Do their simulations accurately model infection and predation?  In the wild, animals are at such high risk of death from these that I don't see how they would need any help from a programmed response to benefit the group.  In fact, the way evolution appears to have worked in the real world is the precise opposite:  Animals with high predation risk age quickly, and animals with low predation risk age slowly.  Both Mittledorf and Bar-Yam come out of a math/physics background.  Is this a case of guys from a very different discipline finally lifting the veil from the biologists' eyes, or is it a case of biological naivete?  Since the theory is so discordant with observations in nature, I fear it's the latter.

And what should be blatantly obvious and somewhat invalidate this type of simulation is that it doesn't account for intelligence. Malthus thought resources would run out as well. 200 years later some scientists are still running simulations with the same flaw. If they want to more accurately model what might happen to human lifespan, their simulations should factor in better use (more efficient use) of resources over time. One would think that would accommodate for much longer lifespans in the group. 

If we disregard the obvious things like disease and predation, I wonder how canibalism factors into their mathematical model.
Especially filial cannibalism.

And what should be blatantly obvious and somewhat invalidate this type of simulation is that it doesn't account for intelligence. Malthus thought resources would run out as well. 200 years later some scientists are still running simulations with the same flaw. If they want to more accurately model what might happen to human lifespan, their simulations should factor in better use (more efficient use) of resources over time. One would think that would accommodate for much longer lifespans in the group. 

I actually wouldn't simulate at all. We can't even predict what the unthinking atmosphere will do next week. It's a fool's game (but a great way to acquire funding). We need to look at the data and abide by what it tells us, no matter how unintuitive its implications might be. As quantum physicists say, "Shut up and calculate!"

To put it another way, I don't think the debate over programmed aging vs. stochastic damage actually matters from a clinical outcome perspective. I would just say that we need to identify the changes to the machinery which are morbid, and attempt to reverse them. Whether protein X is too elevated because I was exposed to pollution, or because I evolved to shut down in under a century, is of little consequence to me, except in the rare case that we can say "Don't expose yourself to chemical Y because it will cause morbid disregulation of protein X." I think it's much harder to prove such connections, than to just hack the machine so it restores normal levels of protein X.

I agree with Reason that mountains of time and money have been (are being) wasted on (epi)genetic optimization, when what we really need is targetted error correction for the each of the various common forms of damage. But in the US, it would literally take an act of Congress to force the FDA to allow even terminally ill people to experiment in ways which might help them and many more who come after them. (The proposed "Right to Try" legislation attempts to do this.) I'm confident that it will never work, and we'll simply continue to let the desperate die, while potential treatments are imprisoned in academia and drug research companies. The answer, quite frankly, is move research to more forward thinking shores. BioViva, for one, is doing this, hypey though they may be. Hopefully other serious research firms will follow. OTOH, it's sad to see fundamentally ill-conceived startups like Calico capturing massive amounts of funding for the sake of pursuing epigenetic tweaks that won't likely outperform existing cheap supplements, as opposed to directly pursuing repair technologies. Then again, from an investment standpoint, Calico is compelling: stable disease pays more than cured disease.