Can Aging Be Hacked? | Matt Kaeberlein | Escaped Sapiens #90

0:00What would you give or not give for an extra 10 or 20 years of healthy life? Because realistically that's what might be on the line for the average American. In this conversation I speak with Professor Matt Kaplan, one of the world's leading experts on the biology of aging and lifespan extension. Right now there's an enormous amount of hype online surrounding longevity which is driven mostly by influences. My goal here in this conversation was to separate out the science from the hype when it comes to living better and longer. We discuss what aging is and the lifestyle choices and medications that can slow it down, as well as some of the most exciting research happening right now at the forefront. This is the Escape Sapiens podcast. If you enjoy deep dive conversations with some of the world's brightest minds, then please support the show by subscribing. It costs you nothing and keeps the podcast alive. And now, his Matt Kaplan, I hope you enjoy. Escape Sapiens.

1:02Matt Kaplan, welcome on the podcast. Thanks. Pleasure to be here. For the average listener, for the person who's listening to this podcast, a representative person, let's say, how much of their life are they missing, are they losing due to lifestyle factors? Yeah, so I think my best estimates, and this is a hard question to answer precisely, so I'm a scientist by training first. I'll just start that so everybody understands kind of where I'm coming from. So I try to be precise. So I think I have to answer the question and say, I don't know the answer to that. But my best guess is the average American is probably losing 15 years of what I call health spans. So the period of life that is spent high function free from chronic disease and disability. That doesn't always translate to lifespan. So for some people, it's going to be more than 15 years. They're going to die more than 15 years earlier than they had to. For other people, they're going to live a long life. There's just going to be this two decades or so towards the end. Sometimes more that is a lower quality life than they needed to live.

2:05So that's really where I focus a lot of my attention these days. Not so much how long can we help people live, although that's certainly part of it. But how can we help people live as much of their life in high quality, high function, good health as possible? But I think 15 years is about right. I'm surprised I was expecting to say five years or something, but 15 years is quite large. So then for that same representative person, where is the best use of their marginal effort? If they had to, is it how they eat, is it being active? What should they be doing? That same representative person? Yeah, so it depends on the individual for sure. And I kind of bucket life. So lifestyle is the answer. That is where you're going to get the biggest bang for the buck. When I put lifestyle into four pillars, everybody's got their pillars of health. I think of four pillars of health span, eat, move, sleep, connect. I like active words. So it really depends. Those are all roughly equal, right? So eat is nutrition, sleep is obviously sleep, move is exercise, physical activity.

3:08Connect includes both personal relationships and sort of sense of purpose and mindfulness in my mind. So it depends on the person where they're weakest, where they're going to get the biggest bang for the buck. What I like to say is start there, kind of ask yourself, look at yourself, try to give an honest assessment. You know, optimally you'll have access to good quality health care, proactive health care. And you can use biomarkers and things like that to also help answer that question. But really try to take a close look at yourself, figure out where you have the most opportunity to make improvements and start there. But which of those is going to be most important, really depends on the individual. So before we go on, it might be useful. So you were surprised by that 15 years, I think a lot of people are, it might be useful just to give a couple of statistics that I think really illustrate how bad the problem is in the United States. So first let's start from what is the average life expectancy? It's in the mid 70s, high 70s, 78 or so average life expectancy.

4:09The CDC estimates that about 60% of Americans have chronic disease, okay, and that could be any form of chronic disease. Every major chronic disease has ages, it's greatest risk factor. The median age in the United States is about 30, mid 30s, maybe 38, let's say. So if 60% of people have chronic disease, it means more than half of the people, the median age is 38. So there's a lot of people younger than 38 who have chronic disease. And the average life expectancy is let's say 78. That's three or four decades that a lot of people are living with chronic disease. Now I wouldn't want to suggest that if you have a chronic disease, you have no health span. That's not, that's not right, that's not helpful. But certainly your health is lower than it could be if you're living with chronic disease. And especially if you're living with chronic disease for 10, 20, 30 years. So I mean, I say 15 years in my heart, I actually think it's a higher number than that if you look across the entire population. You separated out health span and lifespan, but I would have thought that if you solve

5:11chronic disease and you have someone who's healthy, that that would also mean they live longer. Is that not the case in general? Are they really... Certainly on average, yeah, certainly on average, that's going to be true. So they are partially separate. So let me put it this way. What we understand about the biology of aging is that that biology contributes to, and I would say, causally contributes to every major chronic disease that humans experience at least in developed countries. So if you slow biological aging, you're going to increase both lifespan and health span. I think there are things we can do to reduce health span that don't have anything to do with biological aging. And so they're not really separable when we're talking about increasing them, but they are separable when we're talking about decreasing them. And I think, you know, with the things that limit health in the United States, I think many of them do affect the biology of aging. In other words, we are actually accelerating our aging process when we are eating a low quality diet, not exercising, not sleeping well, things like that.

6:14But there is also this age independent component to things that limit health in the United States. So for example, I think, you know, some of the addictions that people experience don't necessarily accelerate biological aging, but certainly limit quality of life. So then for someone who... They listen to what you have to say, and then they change their lifestyle so they're sleeping well, they have good nutrition, they're moving, they have good social connections. When it comes to medications, so I've got my 15 years back and then I come to you for medications. How much more can you add on top? Yeah, that again is going to be very individual and so I think it's important to recognize that for a significant number of people, I don't know what that fraction is, but a significant fraction of people, first of all, it's not going to be possible to have a perfect lifestyle. In fact, I'm not sure that trying to have a perfect lifestyle is even a good idea for most people because that creates anxiety and, you know, self-loathing and all this other

7:16stuff that... I mean, humans are... We're funny animals. We're really complicated animals. So in any case, I think lifestyle isn't going to cure aging for anybody. And I think there are many people who are still going to experience either pre-chronic disease or chronic disease, even if they really get their lifestyle dialed in, right? And I'm happy to give personal sort of anecdotes here. So, you know, in my case, I'm certainly not going to suggest I lived a perfect lifestyle. I don't today, but I'm a lot better today than I was, you know, for most of my adult years. But I have made a very significant shift towards much healthier lifestyle and I think compared to the average person, I certainly lead a very healthy lifestyle. And yet, I still have testosterone deficiency, so I take medication for that. I am almost pre-diabetic or I was before I started taking medication to improve my metabolic homeostasis, even after I got my lifestyle dialed in.

8:17I still have high APOB, which is a biomarker, potential biomarker, of cardiovascular disease. So I think that it's, for some people, lifestyle is going to really move the biomarkers into the optimal range and they don't need to take medications to maximize health span. For other people like me, I think there is a place for proactive use of FDA-approved medications to optimize your biomarkers, to give yourself the best opportunity to avoid disease for as long as possible. So I don't know that I can give a number like how many more years are you going to get? For some people like me, my guess is if I didn't take an SGLT2 inhibitor, so that's a drug that basically causes your kidneys not to uptake as much glucose and so you pee out glucose. So it's an anti-diabetes drug primarily, also used for kidney disease. But if I didn't take that, there's a good chance I would become pre-diabetic in the next decade in diabetic within five to ten years after that. Now I almost certainly won't.

9:18Now I can't guarantee that I won't, but I almost certainly won't because my metabolic biomarkers are optimal, right? So I think for me, taking a few medications proactively is probably going to give me an extra of ten years. So then- It's also worth saying. You'll notice I said, I think this is all probabilistic, right? So people need to understand there's very little when it comes to projecting your future health span that is 100 percent certain. What we're really doing is saying, given what we know today, and we know a lot, what is the best trajectory we can get you on to give you the best chance of living a long and healthy life? I was going to ask, I noticed that in the longevity, sort of public facing space, different people have different medication stacks, and I wanted to ask whether that was because we didn't really understand what we were doing, or it's more that it's just very personalized I mentioned. I think it's both.

10:20I think- So there are really only a small number of medications. I would say, you know, less than ten, that- and I'm going to put supplements in a different bucket because supplements are- I, in my mind, completely different than FDA-approved prescription medications where we have good safety and efficacy data in humans. So when it comes to medications, there are less than ten, I would say, or maybe about ten, that I at least think of in a- in a- the context of proactive health maintenance or optimization. It's not a large set, okay? Now, of course, there are people who have specific conditions that need to take prescription medications for those existing conditions, but I'm talking about using medications proactively to prevent chronic age-related disease. And so I don't think there are a ton of different stacks when you look among the people who are really thinking about this from a rigorous medical perspective, and where they are, it's among a small set of interventions that people are using in an individualized way.

11:24You know, if somebody doesn't have early metabolic dysfunction, it doesn't make sense for them to take an SGLT2 inhibitor. If- for men now we're talking, if- 55-year-old man, I'm 55, by the way, if a 55-year-old man has perfectly great testosterone, he doesn't need to be taking testosterone therapy, right? So that's where it becomes individualized. I think what you're picking up on in the- the longevity space are largely people who are not scientists, not scientific, talking about their stacks of 10, 20, 100 different supplements. In my view, that is- irresponsible, they can do it for themselves. It's irresponsible to be suggesting that- that everybody should be taking these stacks of supplements. That's the majority of which don't have much in the way of real data to suggest benefits in humans. And the downside- I think there's this mindset that, well, if it might work, we should just take it. The downside, of course, to supplements is they- first of all, the active ingredients may have unanticipated consequences in a complicated biological system like our bodies, and they

12:26always come with impurities. So even if the active ingredient is completely inert, if you're taking 100 different supplements, you're getting impurities from all 100 different supplements in your body, which is putting a burden on your liver and your kidney, and who knows what else. So I think there is, unfortunately, a lot of marketing right now in the longevity space that is separated from the real data. So before continuing with the conversation, I have to ask you what aging is. I know you've answered this 7,000 times, but I have to- You can get a different answer every time. Well, then let me ask you in the following way that hopefully we'll get something a little bit different. Yeah. So most people when they think of aging, it might be to start with aesthetics. So gray hairs, wrinkles, that sort of thing, but then ultimately there's disease and death. So are those things what aging is, or are those downstream of what you would define aging to be? Yeah.

13:27Definitely downstream. So everything you describe, I would characterize as a phenotype of aging or a consequence of aging, right? And there are thousands of things that change in our bodies with age. Some of them are on the surface, wrinkles, gray hair, some of them are inside our bodies, but there are all phenotypes of aging, things that change with age, consequences of aging. I think one thing that's important to get out up front is there's chronological aging, which is just the passage of time, but that's very different in- Well, it's different from biological aging. Those two things don't have to match up completely. And what this is really saying is that some individuals, some animals age biologically faster or slower than other individuals or animals. And I think dogs are a great example of this because everybody gets the idea that one human year is about seven dog years. That's just another way of saying that dogs age biologically about seven times faster than people do. Chronologically it's exactly the same, right?

14:29German Shepherd is five years old, at the same time your kid is five years old, if they were born at the same time, right? But biologically that German Shepherd is at a different life trajectory or stage than your five year old kid because it is aged biologically faster. But you're asking, I think, what is biological aging? And there, one way of answering that question is we don't know. We know a little bit, but we really don't know. But I think a better answer is aging biological aging is to some extent the accumulation of different types of damage that lead to a loss of resilience or a loss of homeostasis. Those are related terms, which really just mean when a biological system is perturbed, the ability to go back to steady state, right? And so that ability to restore steady state or optimal function in response to a perturbation goes down with age as a result of accumulated damage.

15:30And we know many of the types of damage that accumulate with aging. If people have heard of the hallmarks of aging, those are to a large extent different types of molecular or cellular tissue organ damage that accumulate with age and contribute to the phenotypes of aging. So the way I think about this, and it's overly simplistic, but that's the way our human minds have to think about biology, is there's a network of proteins and metabolites and RNAs and biomolecules that interact with each other. And as those interactions become dysregulated, that leads to the damage. That's the hallmarks of aging. So it's kind of like, if you think about the core of the earth is the network, the longevity network. That includes things like mTOR and other proteins or metabolites people may have heard of. Outside of the core, the next layer out is the hallmarks. That's the damage that accumulates. And then the surface of the earth are the cosmetic changes that we see in our bodies or the organ and tissue level dysfunction, cardiovascular disease, cancer, the diseases we associate

16:34with aging. And so we can think about aging at any one of those levels, but if we really want to try to understand it, we have to think about it as a whole package. I was hoping you were going to say that it was more a biological program. And the reason why I was hoping you were going to say that is because when I think about reversing aging, maybe in a thousand years, in a distant future where we can really mechanistically understand aging, maybe we can turn it around, maybe we can pause it. And turning it around seems less possible if it's just the accumulation of damage. It seems like something we could just halt. Yeah. So I don't think they're mutually exclusive, although I would say I certainly have come to believe. And I think many of the thoughtful people are thinking about this in the field have come to believe that truly reversing aging is going to be really, really difficult, if not impossible. In part, because there is a stochastic component, right, which a randomness that is going to be hard to reverse.

17:35But that doesn't mean you can't reverse many features of aging and get to maybe not the exact state you were in as a 20-year-old, but a different state that functionally is equivalent or similar to the state you were in as a 20-year-old. So I don't want to be pessimistic and say that we don't have the opportunity to have very large effects on reversing functional declines that go along with aging. In fact, I'm pretty optimistic because we know in laboratory animals that many of the interventions that seem to slow biological aging, if you start treatment late in life, do actually reverse some of the functional declines that have already accumulated. I don't know that we can go all the way back to a 20-year-old, but I think we can have significant positive impacts on function across many different organs and tissues by impacting the biology of aging. To some extent, this really shouldn't be surprising. We've known for a long time that if you take a sedentary 55-year-old who's been eating garbage for the last 20 years and drinking alcohol at night and not sleeping well, it

18:39doesn't have any friends, and you get that person to kind of work on those pillars. Their health is going to improve. They're going to show functional improvements in liver, kidney, muscle, brain. In some ways, you're reversing the functional declines that go along with aging by adopting these healthy lifestyles that impact the biology of aging. I think it is possible to reverse many aspects of aging, but whether we'll be able to completely take a 70-year-old and turn them into a 20-year-old, I haven't seen any evidence in either human data or pre-clinical literature that that's feasible at this point. The closest is probably epigenetic reprogramming, which so far is promising, but has not led to very large gains in lifespan or health span in laboratory animals. Are you talking here about the Yamannakifactors? Yes. I will ask you about them later on, but before that, I wanted to just say that one thing

19:41that gives me hope is the fact that old people can have young babies. It seems, at least biologically, there is a solution to reversing age, in some means. Yes, and again, that goes back to the reprogramming. A big part of that immortality of the germline, so to speak, a big part of that is this epigenetic restoration where, in a subset of germline cells, you can take the epigenetic state and bring it back to a youthful state. The challenge that we run into trying to do that in a whole person or a whole animal, of course, is that we're made up of hundreds of millions of cells, and so there's two components to this that are really problematic. One is, many of those cells have accumulated damage to the genome, to the DNA. Epigenetic reprogramming doesn't fix that, so epigenetic reprogramming changes the regulatory

20:42state on top of the DNA, the methylation and the sedalation and things like that, that control which genes are turned on and which genes are turned off and turned up and turned down. So, the cells that already have significant damage to the DNA, for example, that are close to becoming cancerous, epigenetic reprogramming isn't going to fix that. And then our tool set for even enabling epigenetic reprogramming is very crude right now. When people do this in the laboratory, the majority of cells die when they try to do the epigenetic reprogramming. A few survive, those few cells do, in fact, seem to go back to a more youthful state. The problem is you've lost a large fraction of the cells along the way. If you do that in a whole person or a whole animal, that obviously is going to also create problems. And then the epigenetic reprogramming process itself is tumorogenic, right? So there is this risk that you're going to cause cancers by going through that reprogramming process.

21:42The latter two challenges are potentially solvable through improved technology. And there is reason to be hopeful. So I've given the sort of negative case for this. There's reason to be hopeful people have been successful in whole mice at doing what's called partial epigenetic reprogramming in certain tissues and organs and seeing improvements in function. So an even increasing lifespan a little bit, not yet to the level of something like rapamycin, but being able to have at least modest effects on lifespan. So it is possible to kind of partially turn back the epigenome and get positive effects on function, health span, potentially lifespan. So we're kind of in this middle ground where it seems promising, but there are real challenges to doing this in whole animals, in particularly in people. There are companies working on this, so ALTOs Labs is probably the best known. That was funded by Jeff Bezos and several other people, really, really top scientists in the field working at ALTOs Labs on this problem.

22:46There's a company that Brian Armstrong from Coinbase started also working on reprogramming. And then there's a company life biosciences that David Sinclair co-founded that is going to be doing the first human clinical trials using an epigenetic reprogramming technology. They say they're going to start those sometime this year. So there is progress, we'll just have to wait and see how difficult it is to actually move this into clinical practice. Is that later work, the retinal work that David Sinclair was doing where he's able to do? That's right, so this is... Yeah, you're going, sorry. Yeah, yeah. So what they published was that in a specific mouse model of retinopathy, so in the eye, that they could reverse some of the functional declines. I mean, you know, as is always, often the case, the hype got way ahead of the actual science. And we've cured blindness. No, that's not really true. But they did have a mouse model where the mice were going blind and they were able to show that using an epigenetic reprogramming technology, they could improve vision in those

23:48mice and reverse many of the molecular changes that went along with the disease progression. So yeah, so life biosciences is moving forward into a human disease condition of the eye. And testing their reprogramming technology in people. I don't know all the details, so I think this first study is just going to be kind of a phase one safety study, and then they'll probably move on and look at efficacy, you know, two, three years down the road. So you mentioned at the beginning the difference between chronological and biological age. And I know that we have tests for these, you know, tests that look at methylation and this sort of thing. I want to ask how good are these tests and yeah, how do they work? What are we looking at here? What do we have? Sure. So I think the first thing to say is nothing measures biological age. And again, in fact, the thoughtful people in the field, at least in my opinion, the thoughtful people, have sort of coalesced around the idea that it is probably not possible to measure biological age because biological age is a latent variable, right?

24:52So we, not only do we not have the technology to do it, it may not actually be fundamentally possible to measure biological age because we estimate it. When you say latent, do you mean that you have to set a baseline, like what do you mean exactly? No, what I mean is it's a feature that arises from a multidimensional process, right? So it's not one thing, maybe that's the easiest way to say it, right? So my liver could be a different age to my brain and so on potentially, or... Well, certainly, yes, at least at the functional level. See, this is where the definition becomes problematic because you're defining or other people have defined liver age based on how well the liver's function, right? That's only one dimension of biological ageing and that's fundamentally the problem. You would have to be able to measure every dimension, but even then, the unique features of each of those components that make up this biological age vector, so to speak, the

25:52way those are arranged influences the properties of that biological ageing state. So maybe it's... I don't want to get into the semantics of whether we can measure biological age or not. I don't know the answer. Many people believe it's not possible, but what I can tell you with certainty is none of the tests that we have today measure biological age. They're all estimators of some feature of biological age. And so one way to think about this is, as we already talked about, there are thousands of obvious changes that go along with aging, phenotypes of aging. When we get to the molecular level, there are millions, billions of things that change with age. And if you take enough of those features, you can create a linear equation that will predict mortality very well, right? And so what people have done is take high-dimensional data types. And so that's where epigenetics, metabolomics, transcriptomic clocks come in.

26:54So with the epigenetics, this is going a little bit in the weeds, but I'll try not to get too deep. With the epigenetics, one of the most common types of epigenetic marks that people study is called methylation, DNA methylation. So this is where a methyl group is added to specific regions of the genome, right? CPG islands. So there are hundreds of thousands of sites in the genome that can be methylated and sort of binary. So you can think of it as, if you look across the entire genome, you've got hundreds of thousands of ones and zeros, whether that site is methylated or not. And that changes with age. Those changes are some caused by aging, some random, some we don't really understand, but it changes with age. So what people have done is use machine learning algorithms to say, give me the 120 sites that are most predictive for whether somebody's going to die in the next 10 years. That's how the quote unquote biological aging clocks are built. It's those 120 sites that are most correlated to mortality, or you can do this for any other

27:59age related condition you're interested in if you have the right data set. Tell me which 120 sites are going to be most predictive for whether you're going to get prostate cancer. That becomes a prostate cancer clock, right? So this is useful, I think, because now you can ask the question, what's actually being measured? What's being measured is methylation at 120 sites in the genome, right? It's being estimated is mortality risk in 10 years. How does that correlate to biological age? Or mortality risk is one thing that changes with age. So it's a correlation to a correlation to biological age. It's not measuring biological age. It's an estimation of biological age that requires two correlations. So that's where I think people just have to appreciate. There's a very large error rate here, which we sort of understand what that error rate is between methylation clocks and mortality risk. We don't understand what the error rate is between mortality risk and biological age because we can't measure biological age.

28:59We don't have a way to understand what that error rate is. So we really don't know how good these clocks are at estimating biological age. They're okay. Are they better than our human eyes? Sometimes? Sometimes not. So that's one of the challenges, I think, with the clocks. But they are pretty good mortality estimators, at least at the population level. So that's useful in research, and it's potentially useful in population level studies, right? Or if we want to understand how does, I don't know, eating more protein affect mortality risk. It doesn't prove that eating more protein changes mortality risk, but if you see a change in the epigenetic clock, it's suggestive, right? And if you see a change in the epigenetic clock that matches epidemiology, that's even more suggestive. That's, I think, where we're at with the epigenetic clocks right now. The problem is, of course, people have started selling these things dishonestly to consumers and marketing them dishonestly, making people believe that at the individual level, these things actually measure something relevant for their health.

30:00They may or may not. And we can talk more about the consumer facing clocks if you want to. My take home messages don't use them. They are not useful for anything right now, except wasting your money. They might be good for entertainment value, for some people it might motivate you. And for other people who creates anxiety, you just have to realize they are statistically indefensible. You have no way of knowing whether two measurements that are different are actually different from each other. Okay, so let's go, sorry, I know I'm going on for a while, but let's go back now. If I talked about epigenetics, the useful thing to appreciate here, you can do exactly the same thing with any high dimensional data type. You can do it with metabolites. You can do it with mRNA, that's the transcriptome. You can do it with the microbiome. All of them have different strengths and weaknesses. All of them have different variability across individuals. We don't know right now which are the best at predicting mortality or disease risk or whatever you want to predict, because we just don't have the data set. Methylation has been used the most, and that's what people talk about the most, but it's

31:01been largely used the most because from a technology perspective, it's easier and cheaper. Not because it's necessarily the best type of data for estimating mortality or estimating disease risk. You can also do it with facial images, right? So there have been several studies now showing that if you take standardized headshots of people and you know something about their current health status, future health risk, whether they're going to die in the next 10 years, the AI can pick out specific features from those facial images, generate a linear equation of the 100 most relevant features to predict mortality. So there's nothing special about epigenetics or transcriptomics or molecular measures. Again, you can do this with any aging phenotype that changes in a predictable way across the population with age. If I soften you for a second, just go a bit deeper into the epigenetic clock there.

32:01When it comes to these methylation sites, I asked you earlier if aging was a program. When those sites turn on or off, is that turning part of the genome on or off? What is the function of those sites? Right. So that is certainly one of the main functions of methylation and the hope with the epigenetic clocks was, let's say you identify the 100 most important sites for predicting mortality. The hope was, then you would be able to go look at where those sites are, what genes are they regulating and you'd have a smoking gun, right? The gene that's getting turned off is affecting mortality. That hasn't happened. So this is also one of the big problems with all of the clocks is that we don't yet have a mechanistic connection between the change that we're seeing, the methylation change, the metabolite change, the transcript change, and what's actually driving mortality risk or disease risk or potentially biological aging. So that was the hope that hasn't panned out. It is also worth double clicking on this phrase program because I think this in the biology

33:06community, that's concept gets conflated quite a bit. So what you mean by program is not what at least evolutionarily evolutionary biologists mean by program. So when people talk about programed aging, what that really means is that natural selection acted on specific genes to regulate aging, right? That's different from saying that aging is genetically regulated. So hopefully that makes sense. Genes can still regulate aging and genes do regulate aging, but that doesn't mean that natural selection and evolution acted on those genes to regulate aging. But what most people think is natural selection acts on reproductive capacity, reproductive fitness, rate of development, optimizing to your niche so that you give your offspring the best chance of surviving, right? You have the best opportunity to pass your genes on to the next generation. That's fundamentally what evolution by natural selection says. It doesn't say we're going to evolve so that you live as long as possible.

34:10In fact, once you've done your job, so you told me before we started that you've got young children. I hate to tell you this, but you're within about a four or five year window where evolution doesn't really care about you anymore. You've done your job, right? So a lot of people think that aging is a byproduct of the absence of natural selection. There's no reason to keep you healthy and fit anymore because you've already passed your genes on to the next generation. And so I think that fits mostly with what we observe in the natural environment. So there isn't a lot of evidence in mammals, at least, for programed aging where evolution selected genes to regulate aging. But aging is absolutely genetically controlled, genetically regulated. Every phenotype is a combination of genes and environment. The relative contribution of genes and environment to that phenotype depends on the phenotype. Longgevity is a phenotype, right? Lifespan is a phenotype. It's got genetic components to control it, and it's got environmental components to control it. This actually leads to a interesting recent development in the field, which is that for

35:11a long time, you know, the dogma was that about 80% of longevity is environmentally controlled and 20% is genetic. A reanalysis, which takes into account the fact that, at least through historical periods, a lot of people died from things that didn't have anything to do with aging. So this is called extrinsic mortality. So war, famine, things like that, childhood mortality are age independent causes of death. That was influencing those estimations. You take out those age independent causes of death, seems to be about 50-50, half of your longevity is genetic, half of it is environmental, ballpark, at least in humans. So I think that should be both reassuring to people and also make the point, you should know what your genetics are. So you can do a lot to influence your longevity. You should also understand, as best you can, what your genetic risk factors are so that you can watch for those things. You can try to mitigate those genetic risk factors as much as possible. So all of that's to say, aging is controlled by genes.

36:13It doesn't seem to be an evolutionary program. And then I think maybe the last point is, we don't know yet in humans whether we can get very large effects on biological aging rate or longevity by manipulating single genes. And there's a little bit of nuance here. So we know in every animal where we've done this in the laboratory, there are single gene mutations that we can make that will increase lifespan by 30 plus percent. So people ask, why don't those genes exist in humans? Why don't we have people walking around that live to be 150, 160 years old? I think what's important to appreciate is when we make these mutations in laboratory animals, the ones that have big effects on lifespan, they always have big costs associated with them. Where if that animal was in the wild, it wouldn't survive past development. So an example here are what people call the dwarf mice. These are mice that have mutations that cause dwarfism.

37:13They're usually in the growth hormone pathway, so low growth hormone signaling. Those mice develop slowly, they're small as you might guess from the name, and they live 50 percent longer than wild type mice that don't have that mutation. The problem is, they've got all sorts of deficits where they wouldn't survive if they were in a barn, right, where most mice are going to live. In fact, they can't even regulate their body temperature. They have to be housed with wild type litter mates and sleep next to them, or they're going to die because they can't stay warm enough, even in a room that's 70 degrees Fahrenheit, right? So if those mutations happened in people, the very high likelihood is those people would never make it through childbirth, and if they did, they would never make it through childhood. Those genes are immediately selected out of the population. So I think it's still an unknown whether or not there are mutations in humans that could slow biological aging by 50 percent because if those mutations arose, we would never see them in the general population. It might be the case that there's a way to genetically modify a human which makes them

38:16live longer as long as you can stabilize the environment. I mean, we exist in highly variable environments, but maybe if you can set the temperature always at 25 degrees, and there's some genetic modification you can do that as long as you're... Yep. That might be... That's possible. Yeah. Can I ask? I know that you're interested in companion animals and longevity project. You've got this longevity project with dogs. I will get into that in a second, but I had one quick question on this topic that I want to ask straight up, and that is our test, our biological tests of biological age. Do they work across species? So say, for example, I give you a blood sample of my dog, I tell you it's my blood, and then you go and check their methylation sites, and say my dog is 10 years old. Would you guess that that dog is now 70, or how does that work? Yeah. So the answer to that is, yes, to some extent. So Steve Horvath, who is the person who actually started the epigenetic aging field, really,

39:18he's the one to develop the first epigenetic aging clock, has also developed pan mammalian epigenetic patterns that can do what you just suggested across animals. They don't work with the same level of accuracy at predicting chronological age as the species specific clocks, but there are a subset of epigenetic sites that across at least vertebrates seem to work pretty well, okay? So the answer to that is yes, to some extent. And within each species, you can then develop, if you have the right data set, you can then develop more precise epigenetic aging algorithms that can estimate mortality or health status, future health status. So the answer to your question is both. You can do both. You can do it in the species level, or you can do it across species. The interesting thing, of course in dogs, is that there is this body size effect that is very prominent, where big dogs age faster than small dogs.

40:20And so I don't know that people have developed breed specific clocks yet, but you can do a pan dog clock that looks at all breeds, and you can pick up that body size effect in the epigenetic, the rate of epigenetic changes. So it's consistent with the idea that big dogs are aging biologically faster than small dogs. And why is that? So you mentioned that dwarf mice live longer in this specific case. And I suppose there are whale species that live a long time. And I suppose generally there's these trends that larger animals live longer. Why is that? What is the relationship between growth and longevity? Right. So you just made a very common mistake. So let me clarify, and I mean everybody does it. So that's not meant to be critical of you, but there's this misperception. I mean, what you said is correct. When you look across species, big animals age more slowly on average than small animals do. But within species, it's inverted, right? Small individuals seem to age more slowly than big individuals do.

41:23And those are probably very different mechanisms, right? We don't know for sure, because we don't. We understand, I think, you know, fairly well, what the network is that influences the difference between big individuals and small individuals. So the within species network, that network, by the way, is very well-conserved. So the same factors that influence biological aging and mice influence biological aging in dogs, influence biological aging in humans. But that network is largely what determines the differences between you and me, or the differences between individuals of the same species. We really don't understand at all why it is that a whale, or an elephant, or a giant tortoise, lives so much longer than a mouse or a dog. I shouldn't say at all, but we really don't understand that nearly to the same extent as we do the within species mechanisms. So within species, this body-size effect is conserved. So in mice, smaller individuals tend to live longer. In dogs, that's very much clear, and in humans, there's pretty good data, that's smaller

42:26individuals. And by that, I don't mean body weight, because obesity sort of confounds things a little bit. Talking stature. So height is probably the easiest metric to think about, tend to smaller individuals tend to live longer. This is probably, because smaller people go through development with low growth hormone signaling, low IGF-1 signaling, that sort of sets a biological state that persists throughout life of improved metabolic homeostasis, slower biological aging, reduced m-tore signaling. So things like that. And we've learned that you can kind of flip that switch later in life and still get some of the benefits. That's where rapamycin comes in. We can start giving rapamycin to a 20-month-old mouse, which is about the same as a 60-year-old person, still get a 30% increase in lifespan. So you can't get as big of an effect without going through development in this low growth state, but you can get some of the effect by initiating that state later in life.

43:26I don't know if you've heard of the company Loyal, they're working on a therapy for dogs that is based on this principle, right, that you can reduce the growth signaling later in life and still potentially see benefits for health spend and lifespan in dogs. That hasn't been proven yet, but I think it's very plausible that that will work. So if you have a great day and then you should be very interested in that sort of a study. That's right, yeah. And it probably will work better in big dogs than it does in small dogs, although in principle it should work in both. But let's talk about rapamycin, because as far as I understand, this is one of the better medications that we have, or the more effective medications that we have, at least in animal models. Can you explain? I sort of have some coupled questions where I want to get to is I want to understand ultimately where we find these things or how we find these things, whether it's, we understand some mechanistic pathway and then we target that pathway, or whether it's just dumb luck. So just to start with, what is rapamycin and what's its motive action? Sure.

44:27So rapamycin is a small molecule. It was first found on Easter Island, which is also called rapanuni, that's where the drug gets its name from, rapamycin is produced by bacteria there. And it was first studied, because what was observed was that this molecule, if you put it on human cells, or on fungal cells, would slow cell division, would slow the cell cycle. So it was actually first studied as an antifungal, and then it was studied as an anti-cancer. It was FDA approved more than 20 years ago, though, as an immune suppressant. So for use in organ transplant patients, because again, in response to immune cell, or immune activation, right, you have a lot of cells that are rapidly dividing, and rapamycin is effective at turning that process down. So it's an anti-inflammatory in that context. It's not clear if it's a true immune suppressant, but it definitely modulates immune function. So that's how it's used clinically, primarily, as in organ transplant patients. So you're asking about, you know, how do we learn about these things affecting aging?

45:28It's different, depending on the situation, but with rapamycin, it was a complete accident. In the sense that there were, this all happened in the early 2000s, and I feel very fortunate to have been involved in some of this work. There were, I think, four labs that sort of independently were doing unbiased screens, looking for things that affected lifespan across three different organisms. So budding yeast, the immature worms, and fruit flies. And in our case, we were working in yeast. This was work that Brian Kennedy and I were doing. And we found that the tour, M-tour, when we turned it down, increased lifespan. And so we weren't looking for it. Do you mind if I just stop you just quickly? Sure, yeah. You mentioned M-tour a few times, but people won't be familiar with that. Could you give the just the quick summary? Yeah, yeah. Right. So M-tour is the mechanistic target of rapamycin. That's what M-t-o-r stands for. It's the protein that rapamycin inhibits. So rapamycin is a drug that turns down M-tour. M-tour is a growth-regulating factor, right?

46:29So the way I think the easiest way to think about it is, if you look through evolutionary time, every cell or animal has had to make a fundamental decision, is now the right time to grow and have babies, right? And if you make the wrong decision there from a natural selection evolutionary perspective, you're done, right? You're going to be selected out of the gene pool. So M-tour is the protein that helps cells and animals make that decision, right? And it does that by sensing the nutrient availability in the environment. So when there's lots of food around, that's a good time to have babies because you have something to feed them and they can grow. When there's not very much food around, it's a really bad idea to have babies because you don't have anything to feed them. M-tour senses nutrient levels in the environment. When nutrient levels are high, it says grow, grow and reproduce, do it now. When nutrient levels are low, M-tour gets turned down and that turns on a slow growth, turns off reproduction, promotes stress resistance. That slow growth, stress resistance state is also equivalent to slower aging.

47:33So people, animals, age more slowly when they're in this slow growth, stress resistance state. And we kind of understand how that works at the molecular level, but we don't fully understand it. So again, M-tour is one of the key gatekeepers that help cells and organisms make that decision. Rapamycin, because it inhibits M-tour, tricks the cell into thinking there's not much food around, even if there's lots of food around. And it induces the slower growth, stress resistance state, even when there's lots of nutrients in the environment. And so Brian and I found accidentally, because we were doing an unbiased screen, meaning we were just looking, we didn't know what we were going to find. We found that TOUR, when we knocked it down, genetically increased lifespan. And so the first thing I'd never heard of Rapamycin at this point, this was probably 2003. First thing I did was went to the literature and say, what do I know about TOUR? And found out there was this drug, Rapamycin, that was an inhibitor of M-tour. So we'd done the genetic experiment, we knocked down the TOUR gene, and we wanted to use this drug to sort of verify our observation with the genetic experiment.

48:37So we treated the cells with Rapamycin, they lived longer. That was one of four papers that got published with all within about an 18-month period, all showing that either turning down TOUR, or using Rapamycin, was sufficient to slow ageing and increase lifespan across these three different animal models, that were separated by a couple of billion years of evolution. So this really I think suggested that M-tour is an evolutionarily conserved regulator of aging, and it does so by affecting the same growth signaling, growth promoting pathways that the dwarf mice are deficient to. And so in that case it was an accident. Well, then is it downstream of calorie restriction then, because you're sort of simulating not eating enough, right? It's overlapping, but distinct. So certainly one of the things that caloric restriction does is turns down M-tour, definitely. Caloric restriction also does 1,000 other things in a cell, or in an organism.

49:39So I think you can think of it as like, you know, Rapamycin is like a precision tool, and caloric restriction is like a sledgehammer. You're doing a lot more with caloric restriction. I would say it's still a little bit of an open question, how much of the longevity benefit that we see in laboratory animals from caloric restriction is due to M-tour inhibition versus other things. Also worth noting, Rapamycin treatment in an animal that is eating a normal amount of food is going to be somewhat different in terms of the downstream molecular consequences to turning down M-tour in an animal that is caloricly restricted. Because against a network, right? Caloric restriction is hitting, you know, dozens of nodes in the network. Rapamycin is needing one node in the network. And so you said it was in the last 20 or so years that you did this research. Does that mean we don't know yet what the effect in humans will be ultimately or?

50:39We know some of the effects, but if you're talking about longevity, we don't know. And this, I mean, I think we have to be honest, right? There's very little that we have enough data over a long enough time to have confidence about in terms of longevity, or even really health span, right? So again, you know, the solid data, and again, even people will even nitpick about exercise, right? So proving that something has a positive effect on longevity in humans with 100% confidence, it's a fool's game. You're never going to get there, right? Other than things that we know shorten life span, like, you know, I could go outside and walk across the freeway, you know, with headphones on and sunglasses, and that will shorten my life span. But in terms of extending life span in people, it's almost impossible to prove with 100% certainty something works. But there are lots of, not lots, there are a few things that I think reasonable people can look at the data and say, yeah, that works, right? It's almost certainly regular exercise, eating a high quality diet, getting good sleep, having relationships, positive relationships, are good things for your health and longevity.

51:42There's enough data to back that up. But once we get outside of that, I think the criticism that you hear, well, we don't know if, you know, X slows agent. We don't know if apple mice and slow is aging in people. Yeah, that's true, but it's a false argument because the bar is so high. The only way we could possibly know that is that people had been taking rapamycin at the right dose for decades, right? We just don't have the data. So we have to say, given what we do know, how plausible is it that rapamycin has benefits for health-bending people? What do we know about the side effects? And do we know enough to make a rational, informed decision about whether the benefits are likely to outweigh the side effects at the individual level? I think with rapamycin we're almost there. I think reasonable people can have a disagreement over whether we know enough to make that decision. I think with things like SGLT2 inhibitors, which I mentioned before, we know enough. We know that those drugs are very effective at attenuating metabolic dysfunction, which

52:45is associated with all sorts of bad outcomes. If we attenuate metabolic dysfunction, that's going to have benefits for people who have metabolic dysfunction almost certainly. I can't prove it. People haven't been taking these drugs for decades for us to know epidemiologically if that's true, but it's a pretty reasonable expectation. So this is where I think getting comfortable with uncertainty is necessary if you're going to make rational, informed decisions when it comes to things like rapamycin or other medications that could improve health span or longevity or supplements. And it's really hard because in many cases we just don't have the data set we would like to have a high degree of confidence in some of these things. On making choices then, from the evidence when you look at animal models, so let's say mice, did you find that if you gave rapamycin to a mouse later on in life, you would have a similar effect to early on? And the reason why I ask this is because let's say for example there's a 20 or 30-year-old person who's trying to decide should I go on rapamycin, does it make sense for that

53:48person to wait 20 years for the results to come in? You see where I'm trying to get out here? Yep, absolutely. Yeah, so again, I mean I always have to say, be careful going from mice to humans, but mice are the best data we've got, so that's kind of where you start. So the answer is yes, if you give rapamycin to a middle-aged mouse, you get close to, and nobody's ever done the perfect experiment to say, is it exactly the same? You get close to the same benefit for lifespan, and you get many, many health-span benefits across multiple tissues and organs in a mouse that is roughly equivalent to a 60-year-old person. We also have data in people suggesting that at least in many individuals, rapamycin treatment at the right dose can improve function in certain organs and tissues. So the best data set is work from Joan Manek where they use the derivative of rapamycin, it's called ever-alimus, but it works exactly the same way. And they showed that in 65 and above healthy adults, they could improve immune function

54:50as measured by a vaccine response with six weeks of treatment with the drug. So, and there's other data showing in a bunch of different inflammatory conditions that rapamycin can improve health outcomes in humans who are not organ transplant patients. So it doesn't prove that it slows aging, doesn't prove that it increases lifespan, doesn't necessarily even suggest that everybody should be taking it, but at least in certain individuals, it seems as though some, maybe many of the effects that are seen in mice at the molecular and broader phenotypic level, are shared in older human beings. Let me switch gears a little bit. You mentioned at the start, testosterone replacement for older individuals, older males. It's just a very quick question. Why do women live longer than men? So if you account for lifestyle and the sort of thing, all things equal, why is it the case that women live longer?

55:50I don't think we know. I really think that's one of the mysteries in the field right now, why it is that across, it seems like the entire human population, so different ethnicities, different geographic locations, women live a few years longer than men. I don't think we know. There are hypotheses out there. One hypothesis going back to the evolutionary question is that there has been some level of evolutionary selection for mothers and grandmothers to live longer, to support the care of multiple generations. That's certainly possible. At the biological level, I don't know. What's interesting though is there's also evidence, and again, I don't think this has been completely demonstrated, but there's certainly at least a school of thought that while women have longer lifespan, they actually have shorter health span on average than men, and that that's driven, at least to some extent, by the hormonal changes that accompany

56:50menopause. It is the case that for some age-related diseases, women seem to have earlier onset across the population for chronic diseases, even though they live longer. Again, we don't really know why that's the case, but that's sort of what the data is telling us. Coming back to hormones, we certainly talk about testosterone for men. I think hormone therapy for women, though, is another really important area when we're talking about health spans certainly in humans and potentially longevity. Is there enough data coming in from, say, Unix? I guess that's a long time ago in the Chinese. Is there enough information there to say that testosterone is a primary factor there? There's some information, although it goes the other direction. It's very weak data, but there are a couple of papers. I think they were Korean Unix reporting that Korean Unix lived longer than intact men of the same sort of geographic representation.

57:51That's, again, human data is always messy, and so you always have to think, well, what else might be different about the Unix? You can probably imagine many things that would be different about the Unix compared to the men in that population. One being, they almost certainly had better nutrition, better quality in terms of their environment compared to the average person at that time. That's the data, right? Unix live longer than men, and you'll hear people say testosterone therapy accelerates aging because Unix live longer, and I mean, again, I'm not going to claim one way or the other what testosterone does to aging. I think it's a really important open question that we should be studying. I've tried hard to get those experiments going, and nobody has done them yet. So we don't know. Nobody's ever done a testosterone supplementation experiment, even in mice, to see what the effect is on biological aging. So we don't know. I can say for quality of life, for men who are deficient in testosterone, it has a huge impact. So maybe you're aging a little bit more rapidly, maybe you're aging more slowly, there's good reason to believe that it may be affecting biological aging in a positive way, but quality

58:56of life, I think, certainly is hard to argue with. And my other answer is, look, if you want to go become a Unix by all means, go do that if you think it's going to help you live longer, I don't think it's a great idea. That's not what I'm promoting on this podcast. One of the things that I'm interested in there when it comes to testosterone, and also caloric restriction and protein restriction, is I know that muscle mass is protective later in life, right? So I want to understand, I don't really know exactly how to ask this, but when it comes to the trade-offs, right, if you're eating less, it's harder to build muscle, if you have lower testosterone, it's harder to build muscle. So do we know much about the trade-offs there? What should I be aiming for, big muscles or low calories? Right. Yeah, so this is a really important question, and this is where I think, you know, we have to be particularly cognizant of the ways that mice in the laboratory differ from humans in the real world, right? So no doubt severe caloric restriction, up to about 60% reduction in calories in mice

1:00:03can increase lifespan almost linearly. So a 60% reduction in calories increases average lifespan in mice by about 50 or 60%. So it's a big effect. It's the biggest non-genetic effect that's ever been reported from lifespan in mice. But those mice are living in a highly controlled environment. It's not completely pathogen-free, but very, very low pathogen load. So they are almost certainly immunocompromised, but they aren't experiencing the pathogen load that amounts out in the wild or human in the wild is going to be experiencing. They still typically die from or experience, you know, debilitating fractures or frailty as they get older. Unlike people where we know that loss of muscle, loss of bone density, fractures are major quality of life and quantity of life determinants in older people. So I think in this context, I am very cautious about suggesting we should extrapolate the caloric restriction data from mice to people.

1:01:03So I'll tell you what I believe, and there's data to back this up, but this is my belief because it's hard to prove. I do believe that caloric restriction slows biological aging in humans, and you can see this by epigenetic markers, you can see it by blood-based biomarkers, I also believe that there are going to be negative consequences associated with caloric restriction in humans that will more than offset any benefit that you get from aging more slowly. That includes frailty, loss of muscle mass, weakness, just the overall crappy quality of life, like who wants to be hungry all the time, low sex drive, like all of these things that people practice in caloric restriction experience, I think are diminishing on quality of life and may also diminish quantity of life. We don't really know what the immune effects are going to be. That would be the other thing I would be worried about is if you are, you know, chronically in an immune compromised state, your risk of dying, say, from flu infection probably go up as you get older. So I would be careful with that, but I do believe caloric restriction slows at least aspects

1:02:06of biological aging in humans, and there's data to back it up. So then we can talk about, you know, would mild caloric restriction be good maybe? I sort of gravitate to, you want to be in the normal range. That's pretty solid. Like if you're in the normal range for body composition, and I would go towards the higher range of muscle to fat, you're probably, that's probably about the best you can do, given what we know right now. Should you be, you know, Arnold Schwarzenegger in his prime, probably not, especially if you have to take all sorts of synthetic molecules to get there, right? But should you shoot for, you know, for men, I don't know, maybe 15, 16, 18% body fat? Yeah, that's probably about right. Should you be muscled enough that you can do everything you want to do in your daily life? Yeah, you should. And I think when you're younger, you probably want to have more muscle because it gets harder to maintain your muscle mass as you get older. Protein is a really a tough one, and it's obviously a hot topic, you know, it has been

1:03:07for a while. Again, the animal data would suggest that low protein diet slows aging, at least in some genetic backgrounds. Let me come back to that genetic background in a minute. But at least in some genetic backgrounds, low protein extends lifespan. In humans, the data is much less clear, and in fact, if anything, it suggests that after people get above about age 60, having a higher protein diet is associated with lower all-cause mortality and better quality of life. That's probably again because as we get older, our muscle mass declines and our ability to build muscle in response to dietary protein also declines. So it probably makes sense to look to get what people are calling adequate protein, and again, what does that mean? It's hard to say. I kind of shoot for, you know, and that this is going to be personal, but I kind of shoot for about 1.6 grams of protein per kilogram of body weight, that's towards the higher end

1:04:08of what people are considering adequate protein. It's got to be in the context per day, yeah, yeah, yeah, per day, right? It's got to be in the context of a high-quality diet. So the one thing we know for sure is that high protein, if you're eating a bunch of processed foods and a bunch of garbage, is going to increase IGF1 and probably increase your risk of developing cancer. And that's the concern people have, right, is that high protein causes cancer. When you look at the epidemiology, that almost completely, if not completely, goes away when you control for quality of diet. So I think the best evidence we have is that if you eat a high-quality diet, which you should be doing, then you want to shoot for towards the higher end of what people call adequate protein intake and be doing some sort of resistance training or strength training, right, to maintain and build muscle mass. That's kind of where I land on, you know, thinking about protein and diet quality and quantity of food. The other, maybe the last thing I'll say is I think people get hung up on, you know, when to eat, right?

1:05:08This whole idea of time restricted eating or intermittent fasting, like if that works for you great, if it helps you control your calories, fine, do it. There's really no evidence that's good, even in preclinical literature, that time restricted eating or intermittent fasting, slow-aging, independent of caloric consumption. If the animals are caloricly restricted, sure, we already know caloric restriction works. But if they're not caloricly restricted, there's really no evidence those things have any real benefits in terms of longevity. So the way I think about it is, you know, you can think of nutrition as what you eat, how much you eat and when you eat, right? And I think of what and how much are really important and when is not very important at all. But if it works for you, use it. And I focus on what you eat, meaning high quality diet first, because I think if you eat a high quality diet, in many people, that more or less takes care of how much you eat. You're less likely to overeat if you're eating a lot of vegetables, right? A lot of whole foods, minimizing processed foods, getting adequate protein.

1:06:10You're going to be satiated, you're not going to overeat most of the time. So that's kind of the way I approach diet. So then I suppose maybe we'll find that the best approach then is to be adequately muscled and then use say, rapamysin to target certain pathways rather than to go for caloric restriction. I would agree with that. I don't, I, again, I think rapamysin, I'm certainly not somebody who says everybody should be using rapamysin, but maybe, maybe, maybe, once we have more data, we'll get there. Certainly, I would say, though, your first part of that is adequately muscled, eating a high quality diet, don't worry about caloric restriction, pay attention to your body composition and your other, you know, blood-based biomarkers and adjust based on those, as opposed to trying to adjust exactly the number of calories you're eating. One thing that I find quite interesting about rapamysin is, as we already discussed, it sort of seems to mimic a little bit of what you would get from caloric restriction. And so, in my mind, we shouldn't expect rapamysin to get us to 800 years.

1:07:14You know, there's probably some biological maximum that humans can reach, say, 120, and we have certain lifestyle approaches and medications that may increase the chances or modulate, you know, within that bound. If we want to get beyond 120, are we going to have to look for, you know, therapies that just qualitatively different to what we currently have and what we can currently do? I think so. So, you know, there is a growing, I think, body of evidence that this maximum lifespan of about 120 years that you alluded to in humans may really be a true maximum in the sense that the mechanisms that are regulating that are fundamentally different from the biological mechanisms of aging that we've been studying in the laboratory so far. And largely that's a mathematical argument is based on physics principles. We don't understand the biology, so it's really hard to evaluate from a biological perspective. I have become more of a believer in that than I was five years ago.

1:08:21But the honest answer is we don't know. What I will tell you is I definitely believe there is, we have more to learn about the biology of aging than we currently understand, right? There are people who will tell you, we're close to solving aging. They're full of it. We're not. I think we're close to like 5% understanding aging and there's 95% we still have to learn. And I do think we need more people in the field taking discovery science approaches to look for that 95% and try to understand it. I think this is very common in science and fields as they mature. In some ways it goes from being a very broad based area of study to very narrow as people think they start to learn more and more and more about what the truth is. They start to study what they think the truth is and stop studying what they don't know about yet. And that definitely has happened in the field of aging biology. I think the hallmarks of aging contributed to that. That's not the only reason why that's happened. I think we need to return, some people need to return to looking outside of the box and trying to understand what don't we know about aging.

1:09:23We alluded to earlier the fact that we really don't understand what's different about the mechanisms of aging in a well or in a giant tortoise or in a naked mole rat. These animals that live much longer than we would predict based on their body size or other features that we understand about those animals. We don't understand that. And so we should be studying that and trying to understand it because those might be the principles that allow us to get past this maximum lifespan barrier as it seems to exist today. Yeah, I always thought that answering what it is that gives rise to or controls the speed of aging would come down to large data. So for instance, if we could take blood samples of all the people that have lived beyond a hundred or if we compare blood samples of across species and we just have an AI that crunches the data, is do you think that eventually we're going to get on to the wormbot and some of your large data science at some point. And do you think ultimately that's the direction that we have to go?

1:10:25I think it's important for sure. Yes, I do think we need to gather more with the caveat they've got to be high quality data sets, right? AI is like anything else garbage in garbage out. But with high quality data sets, absolutely, I think that's necessary. It will be informative. If we just do that in humans, it's not going to tell us I don't think a lot about this maximum species lifespan barrier that we were talking about because nobody's getting past that, right? We can learn the things that influence variation among individuals. And some of those are going to be causal, which is the other piece. You can find lots of correlations you've got to get to causality. But we're not going to figure out how do we, how do we breach that 120 or 130 year maximum lifespan. Potentially by looking across species, we can get information about that. The hard part there, so we should do it. Don't get me wrong. The hard part there is how do you then go from predictions about correlation, right? Because we can find lots of things that are different to testing causality in people.

1:11:27That's harder because people age so much more slowly. I think the approach that we start with is we start testing causality of these principles in shorter-lit species, nematodes, fruit flies, mice, see if they are influencing maximum lifespan in those species. And then if they do, we can start to look in people. There are also potentially biomarker approaches we can take in people to give us confidence. But it's going to be really hard to do an experiment that proves that intervention X is going to increase maximum lifespan to 160 years or 200 years. That's going to be a 100 year experiment, right, to do that at the population level. There are two interventions that I wanted to, when we're talking about getting beyond 120, I've sort of picked out two of the more crazy sounding interventions that I wanted to just run by. Now, I'm going to mispronounce this. So you'll have to repeat it when I butcher it. But heterochronic probiosis, right, so this idea, what is it, and how does it work? What did people find? Yeah. So it's not crazy. It sounds crazy. It's not crazy.

1:12:28So the observation is, if you take a map of two mice of different ages, you take a young mouse and an old mouse, and you do a surgery that connects their circulatory system. So they share the same blood, right, so they're sewn together, essentially. The old mouse will live longer and see functional rejuvenations across some tissues and organs, and the young mouse will live shorter and apparently age at an accelerated rate. This is suggestive that there are factors in the bloodstream that are modulating the rate of individual aging, okay? It's a wacky experiment, but what I just said actually makes a lot of sense. There are factors in circulation that modulate the rate of aging. Both directions. There are factors in young animals that can slow aging in old animals, and there are factors in old animals that can accelerate aging in young animals. So how do we do this in people? Okay, so one possibility, which people have probably seen, you know, on these TV shows about Silicon Valley, would be you take blood transfusions from young people and you get them to

1:13:29old people, or, you know, people taking blood from their 18-year-old kid and putting it into their 50-year-old bodies. The probably more realistic therapeutic strategy is something called therapeutic plasma exchange, where you basically circulate your own plasma and take out the factors that are influencing aging in a negative way. The hope would be, at some day, we could identify specifically what those factors are. It's this particular molecule, this particular cytokine, whatever, and use those factors directly. People have been working on that for 30 years with pretty disappointing results. We still don't know what the specific factors are. The therapeutic plasma exchange is now being studied for potential effects on health spend, very early data, but, you know, it looks reasonable. Like, again, the data is nowhere near rapamycin level of quality or quantity, but it looks reasonable that for, I say, the best evidence is for mild cognitive impairment in dementia,

1:14:32that there may be some benefits to doing regular plasma exchange, pulling out the negative factors, and then putting your own plasma back into your body, or plasma supplemented with other factors. I don't think it's going to be, so you were talking about this in the context of a maximum lifespan breaker. There's no evidence that parabiosis or therapeutic plasma exchange can come anywhere near doing that in terms of magnitude of effect. So in mice, parabiosis isn't as good as rapamycin. So if rapamycin isn't going to break through the barrier, I don't think there's any reason to think that parabiosis or therapeutic plasma exchange will. Can I just stop you there for a second? Is that, you know, there's the trauma of the surgery, right? You've got an old mouse, coupled to a young man. Could that be the reason, or? It could be. Again, you're right. That could be the reason. And it is the case that the parabionce, the two mice that are surgically connected to each other, even in a mock experiment.

1:15:33So the controls, if you think about what would be the right control for this experiment, you can take two young mouse and connect them to each other. You can take two old mice and connect them to each other. It is true that the surgery itself depresses lifespan. But even when we look at the magnitude of effect, it's not going to be substantially better than rapamycin, even starting from that shorter lifespan. So I don't think, even when we take that into consideration, I don't think there's any reason to believe that parabiosis is fundamentally more effective than rapamycin. Maybe if we didn't have to do the surgery, we could understand what the specific factors are. You could get close or slightly exceed the benefit of rapamycin. But it's not, certainly, well, I shouldn't say certainly, very unlikely to get to that 60% effect that I talked about with severe caloric restriction in mice. So there's a benefit for the older mouse and it's terrible for the younger mouse in that experiment. That makes it sound as though there's sort of bad factors and there are good factors. Could it be the case that is donating blood beneficial?

1:16:34So should people go and donate plasma or whatever it is that it's the plasma where the factors are? Is it? Again, I don't think, I don't know that we really know because we don't really know what the factors are. So it could, some of it could be in cells, some of it, a lot of it's going to be soluble and metabolites. But yes, plasma has at least some of the factors and people have done experiments in mice just with plasma exchange and showing that you can get similar types of benefits at least directionally to parabiosis. So your question about blood donation, plasma donation. So first of all, you know, you should do it just because it's a good thing to do. It'll make you feel good about yourself and that's a good thing. Will it, you know, be effective at reducing the concentration of these negative factors that may be in circulation? Certainly to some extent, it will, I think the quantity of plasma that you actually donate is, I don't remember the numbers, but it's something like 10% of the volume that you get from a single round of therapeutic plasma exchange.

1:17:37So I mean, the honest answer is we don't know, but just from a mass effect, it's probably not going to be enough to give you a large benefit. That's even assuming that therapeutic plasma exchange itself has a large benefit in people, which we don't know. One other thing that's worth mentioning about therapeutic plasma exchange is independent of the aging component, the biological aging component, you know, another reason that's plausible to think that therapeutic plasma exchange could be beneficial is diluting out toxins that may be present in the blood, environmental toxins. So, you know, heavy metals, microplastics, things like that. There's a growing body of literature, still small, but a growing body of literature supporting the idea that therapeutic plasma exchange does, in fact, dilute out some of these factors that we associate as being negative in people. The other, so the second therapy, it's not really a therapy, but the second thing I wanted to mention, we mentioned Yamanaka factors, right, so this age reversal and say individual

1:18:38cells that people can do in the lab, could you say a bit more about this? So, I know we covered it a little bit, but is there any chance of, so first of all, what are the Yamanaka factors, and is there any chance beyond the retina that we'll be able to do sort of full body or targeted organ therapies using those factors? Sure. Yeah, so the Yamanaka factors are transcription factors that when you express them in human cells, you can do this in other animals as well, they will revert the epigenetic state, so the methylation pattern across the entire genome, back to what's called the pleuripotent state, right? So, in other words, the very, very early cells during embryogenesis that can differentiate into any cell in the body and make an entire human or an entire animal. So just by putting these spore factors in, you can basically revert the epigenome back to that primordial epigenetic state. When that's done, as I sort of alluded to earlier, in a population of cells in a dish, many

1:19:42of the cells will die. They just can't do it, we don't really understand why, they can't get back to that state. The ones that do seem to clear many of the other types of molecular damage that go along with aging, so damage mitochondria, misfolded proteins, those kinds of things also seem to clear out at the same time that the epigenome is reverted to this primordial state. So the hope would be, if you could learn how to control that in a whole animal, you could do that in an organ or a tissue or maybe in a whole person and shift back to an earlier aging state. You don't want to go all the way back to the primordial state, like nobody wants to become a vaguestem cells, but maybe you could go back to being this 20-year-old, right, in peak of your physical health. I think that's asking a lot, obviously, it's still somewhat science fiction, but what is real is, people have shown you can do this partial epigenetic reprogramming or transient epigenetic reprogramming, so meaning you don't go all the way back and you only do it for

1:20:45a short period of time. So you only express these factors for a short period of time and partially revert. That in mice does lead to improved functions in different tissues in organs and a little bit of an increase in lifespan, again, not even to the level of rapamycin, certainly not to the level of chloric restriction. It doesn't mean it's not possible to get to those levels because nobody has really perfected the technology yet, right? So maybe if we perfected the technology, you could do as good as chloric restriction, maybe you could do better. Just no evidence to support that right now. Okay, so what we know from animals, so we mentioned David St. Clair's work, there were other people who were doing this way before David jumped on the epigenetic bandwagon in other tissues in organs. To answer your question, yes, absolutely, there is reason to believe that the reprogramming technology and therapies will go far beyond the eye if we can actually deliver these things effectively and safely in humans. I think fundamentally that's the question right now. It's one big question. Can this be done safely in humans and do we have the technology to control it well enough

1:21:49to get high efficacy without pushing over into increased cancer risk or other things that are going to go wrong if a bunch of cells start dying and don't do this partial epigenetic reprogramming? We just don't know. So I think there's reason to be optimistic, but that's why we do clinical trials and safety trials and go through the process. I have no doubt that we will also see offshore clinics getting way ahead of the regulatory environment delivering epigenetic therapies to people. I know that's happening already. So we'll see if we learn anything from that, we'll see if either way. I could see it going really, really wrong, but maybe we'll also get some anecdotal stories of benefits in those people. But it's going to be a process for sure. So I'm optimistic that 10 years from now, we may have early epigenetic therapies in humans for specific conditions. I don't think, I hope I'm wrong. I don't think we're going to have widespread whole body epigenetic reprogramming in humans

1:22:51to reverse aspects of biological aging. It's not clear to me whether wealthy people have a big advantage over the average person or whether they're just going to be turning themselves into guinea pigs before the data comes in. Yeah. Well, yeah. I mean, I hear what you're saying and it's both, I mean, again, there are lots of wealthy people who don't go to offshore clinics, so I don't want to say all wealthy people are doing this stuff. And there are people who are in the quote unquote longevity community who are doing a lot of self-experimentation, not even an offshore clinics, but certainly there's a lot of self-experimentation going on right now. There are a couple of problems with that. One, I alluded to, we don't get high quality data. So we don't actually learn much from the self-experimentation that's happening out there or the services that are being offered in these offshore clinics. Unless someone looks to 140, right? Then you have a clear. We're not, yeah, we're not going to know that for 70 years, right? So that's the problem.

1:23:53But we could learn a lot about safety and efficacy if this was being done in a way where that data could be collected and be believable. The problem is a lot of these people who run offshore clinics, and again, this isn't a knock on them, but nobody's going to trust the data that they share. Why would you believe that if they show you all of these great outcomes, why would you believe they're not also hiding all of the negative outcomes, right? So we need some third party who is disinterested, or at least doesn't have a financial motivation to be gathering and collecting and vetting this data. That would be super useful because then we could potentially start to learn from all the self-experimentation. We may get there because there have been these new right-to-try efforts in certain Montana and New Hampshire and other states. And I think if those are done well and regulated at the state level, we might actually be able to collect some data from these experimental therapies that are happening in hopefully consenting individuals.

1:24:53One of the things I think is really unfortunate, and I won't name any names, but people who know, no, there are high-profile people in the longevity space who have gone offshore, got an experimental therapies, and had terrible side effects, and they didn't share that. And I think that is just disgusting behavior because they are continuing to set an example for people that is bad, because they are not sharing what they did and what happened to them. But wait, if you are listening, you know who you are. Is this publicly? I suppose I can't ask you because this is confidential information. It's well known within the community, I'll put it that way, within certain aspects of the community. Okay, but so this is sort of information that could save people's lives, right? That's right. It's not accessible to the average person. That's right. And in fact, these people continue to pedal therapies that are questionable, not necessarily

1:25:55the ones that they had that they got harmed from, but other stuff that could equally harm people. And I just think, you know, if you're going to put yourself out there as an example for people, you should be transparent and honest, and if you have information that could help people understand what the real risks are, you should share that with people. Let's show my view. Okay, go on. Sorry. Well, where I would just leave it is, people should understand that there are real risks that go along with these experimental therapies. From the therapies themselves, sure, I would say more often though, when sort of very bad or catastrophic side effects arise, is because these things are being delivered outside of the regulatory environment. So there's manufacturing problems, there's low quality of medical care, there's problems in the clinics themselves, people are doing things properly. That's where a lot of the really dangerous side effects arise from. But because you're operating in this system that's in the shadows, you have a much higher

1:26:57risk of experiencing the side effects independent of whether the therapy itself works or has risks associated with it. I don't think most people understand that. Yeah, this really muddies the data and makes it much less useful. Let's jump onto some of your research and two projects that I think are particularly interesting. So I asked you earlier on, how do we find these molecules that extend life, like rapamycin or potentially extend life? You have the million molecule challenge that you're working on. So I just want to ask you, just straight out of the gate, what is it and what are you attempting to do? Sure. Yeah, so the million molecule challenge, as its name would suggest, is an effort to test a million interventions for their effects on longevity, right? And this really stems from something we were talking about earlier, which is my belief that we understand about 5% of biological aging and 95% is still dark matter. We don't really know what it means or how it works. And so this was several years ago now, probably seven or eight, when I was really thinking

1:28:00about this, that the field had become more narrow. Everybody was doing sort of deep dive mechanistic stuff that we needed to return to discovery science. And I was thinking 100,000, not a million at that point. So I wasn't thinking big enough. And so I was trying to say, how would you test 100,000 drugs or whatever? Could be, could be genetic interventions. But how would you test that for lifespan? You could do it in mice. You could do 100,000 lifespan experiments in mice. You'd need 5 million mice, so that seemed not pragmatic to me. And if I'm anything, I tend to be pretty pragmatic. So I figured, thinking through the option, sea elegans would be a useful animal to start with. Because they age rapidly. They live about a month. You can do the experiments quickly. And we felt like we could build technology that would allow us to do these experiments in an automated way at that kind of a scale. So we built this thing called the wormbot. It's a robot that does worm lifespans. So it is basically an imaging system.

1:29:01So a camera system coupled with a robot that takes pictures of worms throughout their lives. It can be as frequent as you want, but you don't need to have pictures every 15 minutes. So we've got it set up now where you get pictures every couple of hours of individuals. And then using AI, you can identify the worms, learn about them through their morphology, but also see if they're still moving, if they're still alive over time. So we make time lapse series of the images and we do automated lifespan studies. So human beings still have to set the experiments up, but then it's all, you know, let the, let the system do its job. You come back a month later and you're going to get your lifespan experiment. And so this is, this is really pretty infinitely scalable, but it makes doing a million experiments possible. The devices aren't super expensive. We actually started with a make block, make block kit. We bought off Amazon for a hundred bucks, got a little bit more sophisticated since then, but they're not super complicated devices. As long as you have a temperature controlled environment, you could scale this as big

1:30:05as you want. So we tried, I tried to get academic funding for this. People in academia, again, this gets to it into a more complicated discussion about, about federal funding, but academic reviewers have been trained to focus on deep dive mechanistic work, not what people call fishing expeditions, right? If they say your grant is a fishing expedition, you are dead in the water. And so my, my sort of response is, you're never going to catch any fish if you don't go fishing, guys. Come on, figure it out. But that's the reality. So I wasn't able to get it funded in academia. So we spun a company out or a biomedical hoping that we could get funding in the private sector. And then we also started this community project, which is the Million Molecule Challenge, where individuals can sponsor molecules. They can pick from a list or they can sponsor their own for lifespan studies. And then we put that in an open access database that we make available to the community. I think this open access database has a ton of value once the data seconds large enough, because then the AI tools for predicting interventions or combination of interventions

1:31:08actually become useful. Right now, AI for longevity drug discovery, I know there are people who are going to get upset with me because they're trying to build companies around this. It doesn't work. Or what it gives you are rapamycin derivatives, metformin-like compounds, the things that we already know about. But it's not going to be fun. It's not going to be fun. I don't want to say rapamycin and metformin are garbage, but it's going to tell you what we already know, not help you find the new things, which is what I'm really interested in. And it's certainly not going to tell us anything about which combinations are beneficial, which are anti-synergistic, because we don't have a data set to inform on that. So my hope is, if we can make this open access data set large enough, then people who want to use these AI tools and do that sort of discovery science using the En silico models will be able to actually get out some important and useful information. It's a shame actually that funders won't go for fishing or treasure hunting expeditions. It is. Because I sort of understand the reason why, but now with AI, you really can have access

1:32:13to big data in ways that you previously couldn't. So this hopefully will change over time. I want it to ask you specifically about wormbond and how it functions, because the way I imagine it is that you have lots of jars that are filled with these sea elegans worms that are swimming around. There are two questions I have. A sea elegans, do they live in a fluid, naturally? How do you administer the molecule or the drug in a way that doesn't? If this animal doesn't usually live in water, then, or a liquid, you may get very different results than if they were in soil or, you know, I'm more natural. So I just, what does it look like in practice? Yeah, so that's a great question. So the vast majority of aging literature in sea elegans is done on solid media. So that's entering the liquid solids. You can do both. And people have done both. The vast majority is done on solid media. So this is an auger surface with E. coli bacteria, and that's what the worms eat for

1:33:16their food. So the worms are basically moving around on the surface of this auger plate through bacteria when they want to eat outside the bacteria when they're not eating, right? And so what we do, you can deliver, you can administer the drug in different ways. You can, you can administer it in the auger. So it's sort of, you know, dissolved throughout the auger. You can add it topically, right before you put the worms on the plate. That's what we do at Aura. We add it topically to a to a desire concentration. It's not perfect. So the one thing I will say is, and I think the question you were asking is the right question, which is that that environment certainly doesn't perfectly match the worms natural environment when they're out, you know, in the soil or on rotting fruit. And it's different from a liquid culture environment in the lab. We don't know what's what's going to be the best way to do this experiment if our goal is to predict which drugs are going to work in people, right? We have no idea. I think what we can say is there are 40 years of literature aging sealigans under these

1:34:17conditions in the laboratory. We know under those conditions, at the molecular level, the aging process is broadly shared with mice and people. And some of the genes that affect aging in mice also affect aging in sealigans under these conditions. So there's at least a data set that gives us some confidence that aging under these conditions is going to be to some degree similar to mice aging in the laboratory or humans aging in the real world. That's about the best we can do. So the way that the system works is we have 12 well plates and every well has a different experiment, about 30 worms on the surface of auger with the coli. And then we have 12 well plates for each robot, each robot. So it's 144 independent experiments on each device. And as I mentioned, the robot moves the camera over each well, takes pictures, you know, every couple of hours. We use AI to identify the worms and then, you know, time, time lapse analysis to see how

1:35:19the worms are changing morphologically, whether they're still moving. And then at some point they stop moving and if they stop moving for long enough, we call them dead. That's not a perfect measurement of when they die, but it's close enough that for these kinds of studies where we're really looking for drugs optimally that increase both lifespan and health span. If they keep moving for three weeks longer than the control experiment, we can feel pretty confident that that drug had a positive effect on lifespan in worms under that condition. I see. So it's sort of mimicking what you could do by hand, but at scale because the AI takes care of, you don't have to sit there watching each one of the worms. And so you're really looking at their entire life cycle and saying this, this worm, on average in this beaker, they're living 20% longer than the control. It's quite nice, right? Because I imagine you don't need to have, I imagine you might get false positives, but when you're cutting down from a million molecules to, you know, 10, that's not so bad because

1:36:24then you go, you take those 10 molecules and you do a second experiment. It's just, it's reducing the size of the parameter space. Can I ask a question that I, this was saying critical, but it's not really, I think this is a brilliant idea. There are lots of molecules, right? There are many more than a million molecules. So despite how great this is and it scales well, are you still just getting a drop in the bucket? Like, you know, how do you, how do you pick the a million molecules? Yeah. Yeah, there's no right answer to that. So let me, let me frame in a couple of ways. One, my goal going in is not to measure everything, right? First of all, you're never going to do it. That wasn't the goal. The goal is to twofold. One, find better things, things that are way better than what we've got right now and learn how much of, by aging biology do we not understand, right? What else is out there? Those were really the two things that I was after and so I didn't worry so much about, you know, how do we, how do we span all of intervention space, you know, how do you

1:37:28pick the molecules? That is, there's a few approaches we've taken, right? So one is, one of the first things we did when we were just, you know, just starting this was we thought it makes sense to test every FDA approved drug. So you can take sort of smaller pieces of intervention space and still learn interesting biology by doing that. So every FDA approved drug. We also did every FDA approved drug plus metformin because we wanted to understand what happens when you start combining interventions. There are maybe a dozen high quality experiments in the literature with multiple intervention, drug interventions being tested in worms. This is a completely unexplored space, not completely, but almost. Can I stop you? Is metformin, I thought metformin was one of the discredited drugs for long time. So yeah. So again, I wouldn't use the word discredited. Metformin works in sea elegance. Metformin works in people who are metabolically compromised. I think what's less clear is whether metformin affects lifespan in animals or individuals

1:38:34that are not metabolically compromised. So in wild type mice, eating a diet that does not induce obesity, right, or at least induce diabetes, the effects of metformin on lifespan are tiny. They're not anywhere near the effect of rapamycin. And people, I think where you're getting this discredited piece is that in people, there was a study that reported that diabetics taking metformin live longer, certainly the diabetics not taking metformin. That part is true. That study also reported that diabetics taking metformin live longer than non-diabetics. That has been discredited. So it does not seem to be the case that diabetics taking metformin live longer than healthy people are not taking metformin. There's really no good evidence, in my view, that in humans who are not metabolically perturbed, pre-diabetic, diabetic, that metformin has any benefits for healthspan. It doesn't mean that it doesn't just know evidence to support that.

1:39:36And there are some reasons to think that taking metformin may have some negative consequences that healthy people want to avoid. So personally, I don't think metformin is a good choice if you want to look for something where there's solid evidence to the slow-aging and that you want to take it from a preventative or proactive approach. But in worms, metformin works well. And so it was a nice test case where we could do this. It wasn't so much to understand metformin. It was really to say, if we look across lots of different combinations, what do the patterns look like? That was a really interesting experiment because we found things that synergized with metformin, that gave effects that were larger or positive on lifespan than either intervention alone. We also found things that maybe both interventions extended lifespan a little bit, but when we combined them, it shortened lifespan by a lot. So the take home is, I don't think we know enough to predict what these interactions are going to look like for lifespan. And we need more data on combinations to be able to do sort of rational predictions.

1:40:38This is also one of the additional reasons why I don't personally recommend taking a bunch of different things on the hope that they're all going to incrementally have a positive effect. See, this is information that should be out in the public sphere, for sure. Obviously, it's published research, but I mean, for the average non-scientist. Yeah, I agree. And so have you found anything that is better than Rappermycin? We have, yes. Okay. Are you able to talk about that? So what I can tell you, one thing I can tell you is there are other MTOR inhibitors out there that are much better than Rappermycin at extending lifespan, at least in worms. And so you asked about what are the things we want to test? So we talked about FDA approved drugs. We've done a screen of natural products that are generally recognized as safe supplements. But we also thought nobody has really tested other MTOR inhibitors for effects on lifespan outside of a few one-off cases. There are, you know, hundreds of drugs that have effects on MTOR.

1:41:41Why don't we just create a library of these things and look and see what happens. And not surprisingly, we found a few that have bigger effects than Rappermycin. What's also interesting about MTOR inhibitors is there are different classes of these drugs. So again, this is getting a little bit in the weeds, but I'll try to not get too complicated. So Rappermycin is what's called an allosteric inhibitor of MTOR. Meaning, it doesn't bind MTOR directly. It binds another protein called FKBP12, that complex, then goes and disrupts the MTOR complex or specifically MTOR complex one. There are drugs that directly bind to MTOR, that are what are called catalytic site inhibitors or ATP competitive inhibitors. So this library included both. And the biggest hit, the best at extending lifespan in worms, is not a Rappermycin-like inhibitor. It has a different mechanism of action. So again, this just gets to, I think, the larger message, which is that there's a lot to learn. And it doesn't take that much effort or that much resources to do it.

1:42:44And again, I think, I don't know exactly what it would cost. I think if it was done efficiently and at scale, the million molecule challenge would cost ballpark five million dollars, which I know sounds like a lot of money. But if you look at the crazy ship that we spend money on, not just in society in general, but in biomedical research, that is like a tiny, tiny drop in the bucket for an immensely valuable data set. So to me, this just seems like a no-brainer, just from a resource allocation perspective, what you're going to learn. I'm 100% on board, but I have to say I have to play devil's advocate. I speak to a lot of researchers and a lot of researchers say, I just want this, this lot of the cost for my personal product. But actually, I think this is, one of the things I like about your research, I must say, or at least these big, big projects, is that you've found clever ways to stretch resources in my mind, anyway. And so one way that you've done that is having this scalable wormbot, which is cheap to build

1:43:49and you can do mass, mass measurements, so big data for not many resources. On the other hand, you have this dog aging project. And in my mind, the genius there is you're doing tests on animals that aren't in a lab. So someone else is feeding the dog, someone else is looking after the dog, and I imagine it brings down the costs substantially. So could you, could you maybe explain what the dog aging project is and sort of your rationale there? Sure. Yeah. So the dog aging project is a community science project of companion dogs, so pet dogs living with their owners. And we're studying the genetic and environmental factors that influence health outcomes during aging in those dogs. So it's mostly an observational, what we call longitudinal study of aging. We're collecting data on the dogs over time to try to correlate genes and factors in that dog's environment that influence how they, how they age.

1:44:50And as you said, you know, these are all dogs that are living with their, with their owners. And the owners are providing a lot of the data that we're collecting through survey instruments. So our primary data collection tool is what we call the health and life experiences survey. The owner's complete that full survey when they nominate their dog to participate in the study and then annually after that. So that gives us information on the dog's health history, its home and lots of features about the home environment. So, you know, how many humans are there, how many dogs are there, even things like how much time does the dog spend outside? Is it on a concrete surface or a grass surface? So we try to get very granular in the type of environmental data that we collect, diet information, exercise, and health history. And then the owners are asked to provide an electronic veterinary medical record. So they get that from their veterinarian, uploaded to the portal. Owners who can provide that is about half of owners that successfully do that become eligible for what we call the sampled cohorts. So we have genome sequencing, full genome sequencing on 10,000 dogs.

1:45:55And then there are about 1,000 dogs, a little bit more, that are in more detailed cohorts where we send them a kit. And then they take that kit to their veterinarian when they go annually for their veterinary visit. The vet collects blood, urine, hair, and we do these sorts of molecular studies that we talked about before, epigenome, metabolome, microbiome, things like that. So that's the vast majority of the dog aging project. There are, I think, almost 55,000 dogs now around the United States and their owners participating in this study. So it almost certainly is the largest longitudinal study of aging ever, which is great. And then we have a smaller piece, which is a clinical trial. So the clinical trial, we only have one clinical trial, that's the only part of the study where owners are asked to do anything different with their dogs. So the observational piece, we don't ask them to change anything. In the clinical trial, it's a clinical trial of rapamycin, so randomized, double-blind, placebo-controlled clinical trial, really asking, does rapamycin slow aging in dogs? So half the dogs get placebo, half the dogs get rapamycin.

1:46:58Lifespan is the primary endpoint. So we're really looking at lifespan, does rapamycin increase lifespan. But of course, we're also collecting multiple health span metrics, so cognitive function, neurological function, disease incidents, things like that, to try to understand whether rapamycin can also impact health span metrics. I suppose this is what you mean when you said that the daughter is getting, it's coming in regarding rapamycin. And do you have the early results you can share? Is there any big take-home message? Yeah, I mean, so I can't tell you what the results are of the current clinical trial because it's still blinded, right? But what I can't say is leading up to this large clinical trial, the current clinical trial is called Tri-a test of rapamycin in aging dogs. It will eventually have 580 dogs in it when it's fully enrolled. We didn't do two clinical trials before this, primarily safety trials. Again, when you're thinking about doing a study like this in companion dogs, safety has to be your first priority, right? It's very much like a pediatric clinical trial if you were asking people to enroll their

1:48:00children in a clinical trial. So we did two safety studies. We were lucky in the sense that we had data from cancer studies in dogs on rapamycin. So we could go into this being very confident that the doses we were picking wouldn't have any significant side effects would be very unlikely that they would. So in those two safety studies, we certainly assessed side effects, but we also looked at cardiac function using echocardiogram, right, which is an ultrasound for the heart. We did that because three different labs at that point had shown in mice that by ultrasound, you could see improvements in age-related declines in heart function. And so in our first safety trial, it was a 10-week study, dogs got ultrasounds to beginning and end of that treatment period, and we measured three different features of heart function that declined with age in two of the three, even though we weren't powered for efficacy.

1:49:01We saw statistically significant improvements in the dogs that were getting rapamycin. And in the third, it was going in the right direction. It just didn't meet the statistical cutoff. So tiny study, I think there were 24 dogs total in that trial. So I don't want to make anything of it other than it's suggestive, and it's at least going in the right direction, suggesting that rapamycin in dogs may improve age-related declines in heart function. In that same trial, and in the second one, owners got a survey asking about side effects, but also asking about any changes in their dog's behavior that they would consider to be improvements. In both studies, there was a statistically significant difference where owners whose dogs got rapamycin reported that their dogs were more active. And the owners didn't know. So again, suggestive, what does more activity mean? I mean, it could mean rapamycin causes dogs to be hyperactive. No reason to think rapamycin does that. So it could also mean maybe the dogs have less pain, they're moving around more.

1:50:04That would certainly fit with rapamycin's anti-inflammatory activity. So I would say suggestive. So we didn't see any evidence for side effects. And suggestive that rapamycin may have benefits for age-related heart function, muscle function pain, things like that. And when does the blinded period end? When can we see results? Yeah, I wish I could tell you. So I have to be a little embarrassed because we started this study four years ago, before the pandemic, so longer than that, even. And so I was like, oh, we'll be done in five years. And it's five years now. Now, I will say there was a global pandemic. And there was a period of about a year and a half where the project lost its funding. And I don't want to get into the stupidity of that choice by NIA, other than in my opinion, it was a very stupid choice. Nonetheless, the project has been refunded. So between the pandemic and that, that set us back by about four years. So my hope is the project will be fully enrolled in the next 12 months, enrollment is proceeding.

1:51:08And so if people are interested, you can go to dogagingproject.org, nominate your dog to participate in either the longitudinal study or the clinical trial. Yeah, it's only in America for now, unfortunately, yes. That's another thing I wish I could change, but limited resources, limited time for me. Hopefully the project will be full, the trial will be fully enrolled in a year. It's a one-year treatment period, two-year follow-up. So we will do an interim analysis and the statisticians will be unblinded when all of the dogs have completed the treatment period. So maybe two years from now. Probably won't have statistical power for lifespan at that point because the power calculations for lifespan require the full three year, the two-year follow-up period. But we'll see once we get to that point. Okay, so then in the next five years or so, the full results will be powerful enough. I sure as hell helps out. This sounds like it should be privately fundable because especially in, it's sort of synergistic

1:52:11with the wormbog. I imagine a situation where you get funding from some private fund. You do some screen of FDA-approved drugs, you find some candidates, then you run them into the dog aging project, you run the project for five years, you find the best candidates, even as drugs for dogs, you know, these things must be mildable. So to me, it's sort of a no-brainer. You know, I have to wrap up the discussion here, unfortunately, because I love what you're doing, but I want to end in the following way. I know there's a lot of hype, at least in the public facing side of, I'm not going to say longevity research, but there's in the longevity sphere, let's say. Yeah. So when we're not talking about living forever or longevity escape velocity and these sorts of things, when we're looking at down to earth at the research that's being done, at the researchers that are really grinding at the wheel, why should we be excited?

1:53:12So for the listener who's an intelligent listener, who's interested in the science and what's actually going on, why should they care? Why should they be excited? Yeah. Well, so I think there's a few ways to think about this. One reason to be excited is that I think, as we alluded to earlier, I think we know enough today for the average person to get close to two decades of healthy life, right? That's a big deal. So I think, unfortunately, sometimes what happens is you get these people who want to talk about longevity escape velocity or extreme lifespan extension, where they're talking about 100, 200, 500 years, all of a sudden 20 years is like, oh, that's nothing. But the reality is, for most of us, really, just take a second and think to yourself, where were you 20 years ago? And what could you do with 20 years of high quality life, right, if you could get that? That has a lot of value. And so I think people should appreciate, we probably know enough today to come close

1:54:12to that. And the science, our understanding of aging biology, is continuing to progress, where I think the opportunities to do better than that are real, and they will be there. I don't know how fast that's going to happen, and I'm not suggesting that we're going to get to the point where there's going to be a pill that you can take that means you can do whatever you want in terms of nutrition and exercise and sleep. But I do think there's lots of reason to be optimistic that if you practice a moderately healthy lifestyle, that the discoveries that are being made today will translate into additional 5, 10, 20 year increments for health span and probably lifespan. Like I don't think it's beyond the realm of possibility that in the next 20 years, we're going to know enough and have enough confidence that we can get most people to 100, 105 years, with a lot of that being in relatively good health. Again, I don't want to suggest that we're going to have 105 year olds that are, you know, physically 21 year olds.

1:55:14But is it possible that, you know, 10, 20 years from now, the typical 100 year old will have the function of a typical 50 year old today? Yeah, I don't think that's outside the realm of possibility. So that's a big deal, and so people should really think about that. And maybe one of the lessons is do what you can do now to give yourself the best opportunity to take advantage of the discoveries that are coming down the pipeline. I love the vision. Matt Cableign. Thanks for going on the podcast. Thank you.