The Peter Attia Drive - August 04, 2025


#359 ‒ How metabolic and immune system dysfunction drive the aging process, the role of NAD, promising interventions, aging clocks, and more | Eric Verdin, M.D.

Topics
Hosted by
Guests
Mentioned
#358 ‒ Peter's takeaways on navigating HRT, rejuvenating the face, understanding the biology of aging, optimizing fertility, and learning to live well from the dying | Quarterly Podcast Summary #6 #360 ‒ How to change your habits: why they form and how to build or break them | Charles Duhigg, M.B.A

Episode Stats


Length

2 hours and 11 minutes

Words per minute

178.86

Word count

23,447

Sentence count

1,775

Harmful content

Misogyny

2

sentences flagged

Toxicity

6

sentences flagged

Hate speech

6

sentences flagged


Summary

Summaries generated with gmurro/bart-large-finetuned-filtered-spotify-podcast-summ .

Dr. Eric Verdon is a physician scientist who spent two decades uncovering how epigenetics, metabolism, and the immune system drive aging, and now serves as the President and CEO of the Buck Institute for Research on Aging. In this episode, we discuss Eric s path from studying viruses and HDACs to leading The Buck Institute and focusing on aging research.

Transcript

Transcript generated with Whisper (turbo).
Misogyny classifications generated with MilaNLProc/bert-base-uncased-ear-misogyny .
Toxicity classifications generated with s-nlp/roberta_toxicity_classifier .
Hate speech classifications generated with facebook/roberta-hate-speech-dynabench-r4-target .
Topics generated with Qwen2.5-3B-Instruct.
Hosts, guests, and mentioned names generated with spaCy (en_core_web_sm), reconciled against Wikidata.
00:00:00 Hey, everyone. Welcome to the Drive podcast. I'm your host, Peter Atiyah. This podcast,
00:00:16 my website, and my weekly newsletter all focus on the goal of translating the science of longevity
00:00:21 into something accessible for everyone. Our goal is to provide the best content in health and
00:00:26 wellness, and we've established a great team of analysts to make this happen. It is extremely
00:00:31 important to me to provide all of this content without relying on paid ads. To do this, our work
00:00:36 is made entirely possible by our members, and in return, we offer exclusive member-only content
00:00:42 and benefits above and beyond what is available for free. If you want to take your knowledge of
00:00:47 this space to the next level, it's our goal to ensure members get back much more than the price
00:00:53 of a subscription. If you want to learn more about the benefits of our premium membership,
00:00:58 head over to peteratiyahmd.com forward slash subscribe. My guest this week is Dr. Eric
00:01:06 Verdon. Eric is a physician scientist who spent two decades uncovering how epigenetics, metabolism,
00:01:12 and the immune system drive aging, and now serves as the president and CEO of the Buck Institute for
00:01:18 Research on Aging. In this episode, we discuss Eric's path from studying viruses and HDACs to
00:01:24 leading the Buck Institute and focusing on aging research, how aging changes the immune and nervous
00:01:31 system, thymus shrinkage, for example, loss of T-cell diversity, chronic inflammation, and weaker
00:01:37 vaccine response, and why these changes can ultimately shorten lifespan, metabolic drivers of aging, oxidative
00:01:43 stress, fuel choice, insulin, and IGF-1 signaling, and practical tips on zone 2 cardio, ketogenic nutrition,
00:01:52 and GLP-1 drugs, why NAD levels fall with age, the role of sirtuins and CD38, what NMN, NR, IV NAD can and
00:02:03 can't do, and the importance of stopping NAD loss, drugs that have the potential to slow aging, including
00:02:10 optimal rapamycin dosing, growth hormone-based thymus regrowth, blocking IL-11 or IL-1, and how these
00:02:18 things might compare with, say, exercise, current ways to measure biologic age, and the limits of
00:02:24 today's epigenetic clocks, new proteomic and organ-specific tests, and how combining multiple
00:02:30 metrics with wearables may guide personalized longevity care. So, without further delay, I hope you enjoy
00:02:36 my conversation with Dr. Eric Ferdin. Eric, thank you so much for coming to Austin. I know it wasn't
00:02:47 just to talk to me. I know that half of it was getting you to drive on the track at Coda tomorrow
00:02:52 with me, so we're going to have some fun there. My pleasure. But as much as I think the two of us
00:02:57 could sit here and talk about race cars for the next three hours, I don't think the audience would
00:03:02 appreciate it or care for it as much as they will care for what we will talk about, which is
00:03:07 your work in gyroscience. So, maybe give folks a little bit of a sense of what attracted you to
00:03:13 this field and how your journey and background brought you where you are. It's a bit of a
00:03:18 serendipitous type of story in a way that I'm an MD by training from Belgium, did my last year of medical
00:03:24 school at Harvard, and this just sort of opened my eyes to a whole world. I was the first person in my
00:03:30 family to go to college. Ending up at Harvard with some of the best teachers, some of the best
00:03:34 students was just mind-blowing. And I went to medical school wanting to do research. Never had
00:03:40 that sort of a doctor fiber, I call it, so really wanted to research. And so, after this, finished
00:03:45 medical school and came back for directly a postdoc at the Charleston Clinic working on diabetes and
00:03:51 metabolism. So, this is where the story gets circuitous. Ended up becoming interested in the reason for
00:03:57 the etiology of type 1 diabetes and worked on viruses and autoimmunity. This eventually led
00:04:04 me to mostly a career in virology, which confuses people. So, I spent many years working on a variety
00:04:11 of viruses, including HIV and herpes viruses and so on. And through that work, we ended up cloning a
00:04:17 family of protein called some of the first epigenetic regulators, the HDACs. And the HDACs at the time,
00:04:24 there was 1996, we were responsible for the cloning of a whole family of these epigenetic regulators,
00:04:30 ended up being important in aging. And starting in around 1995, 1996, my lab slowly shifted towards
00:04:37 the study of aging. And to this point today, actually, I only have one last postdoc in the lab
00:04:43 who's working on HIV. The whole lab is actually focused on epigenetics, immunology, and metabolism,
00:04:50 so that the interface between these variables. So, in some ways, it's the beauty of an academic
00:04:55 career, which I've just followed my interests, sometimes followed the money a little bit in
00:04:59 terms of funding. Now, I mean, I have another additional responsibility, which is to lead the
00:05:05 Buck Institute for Research on Aging. I have split my time between the lab and some more leadership type
00:05:11 of activities.
00:05:11 So, you mentioned two things there, metabolism and immunology. Talk a little bit more about how
00:05:18 each of those individually contributes to aging. I think most people will intuitively understand it,
00:05:24 but talk maybe a little deeper about it.
00:05:26 Well, first, immunology is central to aging in many respects. I hope we can talk about this later.
00:05:34 There is data showing that there are two organs that are rate-limiting in terms of your aging,
00:05:39 and it's the central nervous system and the immune system. And the reason for this is actually,
00:05:45 one could have predicted this based on the fact that both organs are distributed organs. If you
00:05:51 think of your immune system, it's located pretty much throughout the whole organism. And so,
00:05:56 its activity can influence the well-being or the functioning of every single organ.
00:06:01 The same goes for the central nervous system. And there's a recent study coming out,
00:06:06 actually, from the lab of Tony Whiskere showing that those biomarkers that measure aging in those
00:06:13 organs appears to be the most predictive of your lifespan. There's also incredible data showing
00:06:19 that if you induce a specific lesion in the immune system, for example, in mice model, if you knock out
00:06:26 ERCC1, DNA damage repair, only in the bone marrow so that the whole immune system is affected,
00:06:33 you actually induce accelerated aging in the whole organism and senescence in every single organ.
00:06:39 In what model?
00:06:40 It's been done in two different models. In mice, it's been done with the ERCC1 mutation. It's also
00:06:45 been done by knocking down the major TFAM, the major transcription factor for mitochondria. So,
00:06:50 if you induce mitochondrial dysfunction only in the immune system, you induce secondary senescence in
00:06:55 the whole organism.
00:06:56 Do you think that would be true in humans?
00:06:58 It's a million-dollar question. In some way, it's been shown in two different models in mice.
00:07:03 B6.
00:07:04 I don't remember the exact strand of the mouse, but there's no reason why it should be different,
00:07:08 frankly. And it speaks to the importance of the immune system. The second way for the immune system
00:07:12 is through chronic inflammation, which is tied cause and effect in the whole aging process.
00:07:19 And we can talk about this later as well. I find it fascinating, the whole idea of chronic
00:07:23 inflammation, which is induced by the aging process, but itself actually further accelerates
00:07:29 aging. So, there's really a lot of work that's being conducted in this area. The other one that
00:07:33 you were asking is metabolism.
00:07:36 That's a very interesting idea that two organ systems that are going to be rate-limiting in age
00:07:40 are the central nervous system and the immune system, both of which are distributed.
00:07:44 Where would you put the endothelium in that list as well? The endothelium is also quite distributed
00:07:51 across the organism. And do you think that there's an inevitability to basically endothelial damage
00:07:57 as a process of aging, which of course results in the leading cause of death, the atherosclerotic
00:08:02 diseases? Do you think of it the same way, or do you think of it as different?
00:08:05 It's not sort of defined as an organ by itself. It's a cell type. I agree with you,
00:08:10 has incredible importance, especially as it affects the heart and the cardiovascular system.
00:08:16 And the brain.
00:08:16 And the brain. But I think of it as not so much as an organ, but more as a principle that
00:08:22 maintenance of barrier function, not only in the endothelium, but also in the skin,
00:08:29 in the blood-brain barrier, are emerging as key areas to focus on if you want to maximize your
00:08:34 longevity.
00:08:36 Yeah. I want to come back to this in great detail, Eric, but let's, for the sake of summary and
00:08:40 synthesis, turn over to where you wanted to around metabolism. 0.73
00:08:44 So metabolism is essential to life expectancy for a number of reasons. One of them, I'm convinced,
00:08:53 even though that theory has been somewhat discredited, the whole oxidative stress theory
00:08:57 of aging, I still think oxygen is one of the major problems associated with the aging process.
00:09:04 We have not been able to target the oxidative stress using antioxidant. That has failed.
00:09:10 It doesn't mean that the whole oxidative stress theory of aging is not valuable. I think living
00:09:16 in an oxidative environment is one of the mechanisms that leads to aging. Not the only
00:09:20 one. Aging is pleiomorphic.
00:09:22 But just to make sure folks understand what you're saying, Eric, you're saying that the generation of
00:09:28 free radicals through oxygen. So I don't know how technical we want to get for people, but I think
00:09:32 unfortunately we might need to get a little more technical and apologies to those who don't want
00:09:36 to go this deep, but we have to talk about kind of what the role of the electrons are in oxygen
00:09:40 and why free radicals form and what they do. So maybe we do go a little deeper here and explain
00:09:46 what you're saying. It's a very important concept and I think we should probe it.
00:09:50 I don't know how much, I mean, maybe you do a better job at explaining this for the lay person.
00:09:54 I mean, oxidative stress is the fact that pretty much the main metabolic reaction are dependent on
00:10:01 oxygen, which gives its electron. It's in the so-called respiratory chain. There is leakage of
00:10:08 these electrons that are traveling down this respiratory chain, leakage at specific places. You know,
00:10:13 if the process was a hundred percent efficient, the whole energy would be transferred from metabolites
00:10:19 such as fatty acid, glucose, and so on. But it turns out the mechanism is actually leaky. These
00:10:25 electrons reacting with oxygen can generate these byproducts called radical oxygen species,
00:10:31 which are highly reactive.
00:10:33 Right. So they're not chemically stable the way we think of a normal atom of oxygen.
00:10:38 No. And so they tend to react with proteins, with fatty acid, and they induce lesions.
00:10:43 The importance of this system in terms of protection against it is highlighted by the
00:10:48 number of molecular systems that we have that are actually protecting against this.
00:10:53 And we know that as we age, that leakage increases.
00:10:57 Exactly.
00:10:58 So something about the integrity of the mitochondria and the respiratory electron transport chain
00:11:05 degrades as we age, and therefore we see more and more of this leakage, yeah?
00:11:09 Yes, absolutely. And so out of this came the whole idea, well, let's just suppress oxidative stress.
00:11:16 And there are chemicals, even some as simple as vitamin E, vitamin C, that you could imagine that
00:11:23 by chemical knowledge would be predictive to be able to quench these radical oxygen species.
00:11:31 Sorry to just keep interrupting you. We'll play off to each other to do this.
00:11:34 No, no, this is great.
00:11:34 So you eat, for example, an antioxidant, and as you said, it neutralizes that reactive oxygen species
00:11:41 with its unstable electrons, kind of like you would throw a blanket on a fire that's simmering.
00:11:47 Exactly. And that was the hope. So when the theory was proposed, a whole industry actually grew up
00:11:53 out of this, the whole antioxidant, and the antioxidant diet, and the vitamins, and so on.
00:12:00 You can still, by the way, that whole industry is still existing today.
00:12:03 Sure does.
00:12:04 Now, what happened is that when clinical trials were conducted in this area, they failed.
00:12:09 And so people who think relatively simply decided, well, the antioxidant failed, therefore the theory
00:12:15 has no validity. I would say not so fast, because it turns out that these radical oxygen species
00:12:21 also have important roles. They actually are inducing an inflammatory response, which can be
00:12:27 protective. And a good example is during exercise. There is some evidence of activation of oxidative
00:12:34 stress during exercise. And if you neutered this, for example, with anti-inflammatory, you probably
00:12:40 remember the data showing that anti-inflammatory drug tend to suppress some of the beneficial effect
00:12:46 of exercise. It's the same whole idea. And so this is one case in which these
00:12:50 radical oxygen species can have a protective role and actually a signaling role. So when you
00:12:57 suppress it completely with these global nonspecific antioxidants, essentially you're not only killing
00:13:03 the bad guys, but you're also suppressing an important signaling mechanism.
00:13:07 There's another hypothesis that I would offer, which is, is it possible that there's still a net
00:13:13 negative to the free radicals? So there might be some benefits, but more negatives. But it could be
00:13:19 that the trials were using agents that were simply ineffective, because the problem is we don't
00:13:25 have a great biomarker for the state of free radicals. So it's sort of like saying, I have a
00:13:31 hypothesis that this biological process is bad. I can't measure it really, but I think it's bad.
00:13:38 I have a drug that I think will tamp it down. Let's give the drug, the trial failed. Well, do you
00:13:43 actually know if it tamped the thing down? We don't even know if we tested the hypothesis, correct? And so
00:13:48 those would be kind of two distinct plausibilities.
00:13:51 I completely agree. And it's quite often the case. I mean, the whole story of vitamin D is a
00:13:55 good example. Where people will tell you, you know, vitamin D doesn't work because they conducted
00:14:00 clinical trials, but they didn't fix, they didn't adjust the dose, they didn't measure the level. So
00:14:05 it's a bit the same story. There are markers that you can actually do use in research environment,
00:14:11 like 5-hydroxynoninol or protein carbonylation, which are indirect markers of lipids or protein
00:14:19 oxidation.
00:14:21 How efficacious or beneficial, or I guess the word is, how complete are they in the scope
00:14:26 of understanding? And have we demonstrated that mega doses of vitamin E or vitamin C will
00:14:31 indeed suppress those markers in humans?
00:14:35 They're not great. They're not great. I had a colleague at the Buck Institute, Martin Brand,
00:14:39 who is one of the leaders of the whole mitochondrial field called bioenergetics, which is the study
00:14:44 of how the respiratory chain and energy metabolism happens in mitochondria. And he came up with
00:14:49 the idea that he identified many of the sites where these unique radical oxygen species are
00:14:56 generated. And he was able to generate specific inhibitors for each of the sites and was able
00:15:02 to show that actually inhibition at some sites was beneficial while inhibition at other sites
00:15:08 was not beneficial. So this project was actually supported by a pharma company, which eventually
00:15:14 decided to drop the program. And he's retired, which I think is a great loss because it is a
00:15:19 whole program that still needs to be pursued.
00:15:21 So if I'm understanding what you said correctly, Eric, it sounds like there's a much more nuanced
00:15:27 view. It's not that free radicals are bad and it's not that free radicals are good. It's like everything
00:15:33 in biology. It's the Goldilocks rule. You might need more of it during this circumstance in this part
00:15:39 of the body. You might need less of it in this circumstance at this totally different part of the body.
00:15:45 And as a result, any strategy that would try to globally suppress it could, even if successful
00:15:51 in doing it, which we haven't been able to measure, might actually not yield to a favorable outcome.
00:15:55 Totally correct. I get frustrated by the way that people sort of love to oversimplify
00:16:02 or sort of erase whole fields. I suspect we will get to talk about sirtuins because the same thing has
00:16:08 happened in the sirtuins. There's a lot of amazing work done. And then a few negative results or things
00:16:14 not working out. NAD metabolism, same thing. I always tell people, you know, once you get into
00:16:19 any field of study and you go deep and you start testing in humans, put on your seat belt because
00:16:25 it's not easy and there are no magic bullets. But I think stopping the study and saying the whole field
00:16:31 is BS is really, for me, not the way to go. We got to dig deeper. And eventually, you know, we'll get to that.
00:16:37 Tell me what else within metabolism you think is kind of a hallmark of aging. So we obviously talked
00:16:45 about the central part of metabolism, which is respiration and ATP generation and the leakage
00:16:50 that occurs there. And basically, unfortunately, that just appears to be inevitable.
00:16:55 Yes. We will never stop the oxygen in our environment.
00:16:59 I do like to tell my patients that this is
00:17:02 why I kind of harp on them to do a lot of zone two cardio training. So zone two, very specifically,
00:17:09 by definition, is the canonical exercise you would do to maximize fat oxidation, which of course,
00:17:18 implies the most efficient use of the mitochondria. And the hypothesis, because I don't think,
00:17:24 you know, we don't have proof of this, but the hypothesis is training at that level for specific
00:17:30 periods of time throughout the week is a way to improve the health and function of your mitochondria,
00:17:35 which would hopefully imply that you're reducing that degradation of function. Do you think there's
00:17:41 validity to that, at least first order logic? Yeah. I mean, the proof is in the pudding in the
00:17:46 way that we know exercising and a combination of exercise is the best anti-aging intervention we have.
00:17:53 But do you think part of it is through that exact mechanism? Yes.
00:17:56 Yeah. I mean, that's been my hypothesis. But again, we can't fully glean that in any human
00:18:00 clinical trial. No, hard to study. And I think your point allows me to sort of address your question.
00:18:05 What is it about metabolism that really is so important? I think I'm convinced that it is fuel
00:18:10 utilization. You mentioned fatty oxidation versus glycolysis, and I'll add ketosis to this. I think
00:18:17 if you think about your metabolism is able to oxidize a number of different substrates, amino acids,
00:18:24 fatty acids, glucose, and ketones. And lactate.
00:18:28 And lactate. And every one of those actually burns with different efficiency, both being car
00:18:34 aficionado. I think your audience probably knows also that they're different to burn diesel or to
00:18:40 burn 100 octane gas. And if you look at that hierarchy, I think ketones are probably the cleanest
00:18:46 fuel to burn in terms of, again, byproducts, oxidative stress. They seem to be really unique.
00:18:53 Yeah. How would you rank order from cleanest to dirtiest, inclusive of lactate?
00:18:57 Lactate, I would not be able to put it. I think it's probably clean.
00:19:01 Yeah. I think my intuition is it is as well.
00:19:03 Yeah. The top would be-
00:19:04 Beta-hydroxybutyrate.
00:19:05 Beta-hydroxybutyrate. Acetoacetate is present at such low abundance,
00:19:08 it's probably not relevant as a fuel source. Then fatty acid. Next is the worst is actually glucose.
00:19:14 And when you think about metabolism and aging, for me, it goes to a lot of the data that has emerged
00:19:22 from the ITP, for example, intervention testing program.
00:19:26 Rich Miller has been on several times.
00:19:27 Yes. I watched your recent podcast with Rich and others. One of the remarkable thing,
00:19:32 when you look at the drugs that have seven or whatever, 10 drugs that have emerged out of 80,
00:19:37 they are really targeting glucose metabolism via a completely different mechanism. Think about
00:19:44 acarbose, which is blocking absorption of glucose. Think about the canaglifosin,
00:19:49 which is targeting a protein that has nothing to do with links to glucose reabsorption in the kidney.
00:19:55 Think about metformin, which is-
00:19:57 But metformin failed.
00:19:58 Yeah. It failed, but it seems to be having very powerful effect. Well, it did not fail,
00:20:02 actually.
00:20:03 It failed unless it was paired with rapamycin.
00:20:05 Yes. And in monkeys, there's a study coming out that showed that it actually had an effect on lifespan.
00:20:10 And do you think rapamycin has any impact on glucose metabolism favorably?
00:20:14 Generally, actually, rapamycin is the exception to this because it seems to be having,
00:20:19 it's not indifferent. It has been claimed to be having an effect on insulin sensitivity.
00:20:24 Although I'm not clear if that's true at the doses, but anyway, yeah, we can come back to that.
00:20:29 I've taken rapamycin. I have not seen any effect on my blood sugar.
00:20:33 Think about acarbose, canaglifosin, metformin, and now the GLP-1 agonists,
00:20:39 which I predict will emerge as geroprotectors in the future. So I think that really speaks to
00:20:47 an important aspect, which is fuel utilization and how, whether you're burning a clean fuel,
00:20:53 whether you're burning a dirty fuel. We've put, for example, mice on a pure fat diet. These mice
00:21:00 never saw a carbohydrate during their life and they lived longer, which I thought was actually
00:21:04 quite interesting.
00:21:05 It is interesting, Eric, because a lot of the mouse literature, I think people don't read the
00:21:11 fine print very closely. They don't notice that the typical thing you'll see is these mice were
00:21:18 fed a high fat diet to induce obesity so that we could test drug A, B, or C against obesity.
00:21:23 In those studies, it's not just a high fat diet. It's a high fat, high sugar diet.
00:21:28 So they're making some insanely hyper palatable. The closest I can come up with is they're making a
00:21:33 donut, right? It's a fried dough sugar food. So they're making basically donuts for these monkeys.
00:21:42 And that's different than saying it's a high fat thing. So yeah, I think that's important to point
00:21:45 out because high fat minus the sugar might not be the same issue, right?
00:21:50 I agree. At least in that model. So what do you think it is about glucose metabolism
00:21:55 that leads to this? Because for all intents and purposes, let's just go through the metabolic
00:22:00 pathways. So glucose, six carbons, it gets broken down into pyruvate. You get two pyruvates for one
00:22:09 glucose, right? And then pyruvate, let's just assume we're doing this under aerobic conditions so we're
00:22:15 not in a rush. We're going to take those pyruvates. Do they turn into acetyl-CoAs? I can't even
00:22:20 remember to then enter the- It's actually one pyruvate and that enters the mitochondria and
00:22:25 becomes acetyl-CoA. Acetyl-CoA, okay, okay. So what is it about that process that is not as efficient
00:22:32 as when you are cleaving off carbons from a free fatty acid and those carbons are turning directly
00:22:39 into, I think, just a straight acetyl-CoA and then entering the Krebs cycle?
00:22:44 I mean, it's a very subtle difference. Why is one so much more inefficient?
00:22:47 You mean why is there more calories per fatty acid?
00:22:50 No, no, no, no, no. That can be explained by the stoichiometry. Why is one quote-unquote dirtier?
00:22:55 Okay. Obviously, this is a really complicated question. So I don't know that I would be able
00:23:00 to really tell you purely as fuels whether there is a difference. I think the biggest difference is
00:23:06 in terms of the whole mechanism that they elicit. And when we think about glucose, I don't think
00:23:13 necessarily of it if you were to study it in a tissue culture dish that one would be more toxic
00:23:17 than the other. I don't think there's any evidence for this. But glucose, and particularly
00:23:22 the form of glucose that we have not evolved to actually be exposed to, which is all the wheat
00:23:28 products, this fast form of glucose elicits insulin secretion. And I think insulin and IGF-1,
00:23:36 particularly insulin, is the culprit in this whole process.
00:23:40 So you're not saying that one mole of glucose, one mole of free fatty acid, we know there's
00:23:47 a difference in ATP generation, but you're not saying that there's a different, assuming
00:23:52 they're both going through the mitochondria, you're not saying there's a difference in free
00:23:55 radical formation mole per mole? Or are you saying that it's this way? There's another way to explain
00:24:01 it, which is per mole of ATP, you need to run so much more glucose through that, of course,
00:24:08 you're going to get more leakage.
00:24:09 The key difference is that the glucose is generating ATP, not only via acetyl-CoA and pyruvate,
00:24:16 but is also generating ATP in the intracellular plasmic components. The fatty acids do not generate
00:24:23 any. So I suspect that there might be a difference in terms of the amount of free radical that are
00:24:28 generated. There is evidence, but I would not be able to cite you the paper, that one burns more
00:24:34 cleanly than the other. And I suspect it's partly the cytoplasmic component of glucose. It's also
00:24:38 less efficient in terms of the amount of energy that's being generated per gram of fatty acid,
00:24:44 or per mole of fatty acid versus per mole of glucose.
00:24:48 And then going back to the insulin IGF component here, what role do you think they're playing?
00:24:52 critical. Because epidemiologically and through studies, we know that the insulin response to
00:25:00 your glucose. So if you do a lot of sports, I'm not a proponent of the low carbohydrate or no
00:25:06 carbohydrate diet, because there's very little evidence that those diets are actually beneficial.
00:25:11 I gave you the example of a ketogenic diet, which we did experimentally, but these are not practical
00:25:16 diets for anyone.
00:25:17 Just because of the challenge in avoiding carbohydrates in the standard world we live
00:25:22 in?
00:25:23 Yes. But socially, palatably, I mean, there's so many reasons. I went on a ketogenic diet.
00:25:29 And from what I remember, I think you went too.
00:25:31 I did for three years. I was on a ketogenic diet. So we should compare notes. I want to hear your
00:25:36 experience. And then I want to ask you a couple of questions about it.
00:25:38 It was very hard.
00:25:39 How long did you do it?
00:25:40 For a couple of years. And I did not feel super healthy, which is really kind of interesting.
00:25:47 I found it socially isolating. We've worked actually to remedy this on, we can talk about
00:25:52 this later on, novel keto.
00:25:54 Like a ketone ester?
00:25:55 A ketone ester, exactly, of beta-hydroxybutyrate. So going back to the role of insulin, there is a
00:26:01 lot happening that's been documented that the intensity, first, your average glucose plays
00:26:07 a role. Average blood glucose, this is measured by hemoglobin A1c in a whole series of complications,
00:26:13 cardiovascular, as you know. But perhaps more important is the intensity of your peaks. And I
00:26:20 think the intensity of the peaks of insulin is a reflection of your glucose intake, fast-absorbing
00:26:27 glucose. And that's the reason why we advocate, I advocate people to go on a CGM, continuous glucose
00:26:33 monitor, and to really learn to understand what spikes them. The whole idea is to mitigate these
00:26:39 peaks of insulin secretion. I'm just going to play this for all of our patients. We have this
00:26:44 discussion with every one of our patients, so it'll be nice to just play this video and let you do the
00:26:48 talking. The whole idea there is to, again, mitigate these peaks and either dietary or, for example,
00:26:58 the GLP-1 agonists are playing a role in this as well. Yeah, well, let's talk about this because
00:27:02 there is, at least for me, a great deal of confusion around this point. Now, we understand today the role
00:27:09 that the gut plays in metabolism, and we understand that a lot of it is transduced through GLP-1.
00:27:17 So endogenous production of GLP-1, according to Ralph DeFranzo, the world's authority on this,
00:27:23 is what's driving 80% of beta cell activity with respect to insulin. And therefore, when we have
00:27:31 insulin resistance, the GLP-1 we're making is insufficient to generate the insulin that's
00:27:36 required to manage the glucose. Makes sense if that's the case, that giving exogenous GLP-1,
00:27:42 you take a shot of terzepatide or semaglutide, you're going to put more GLP-1 in the system,
00:27:48 you're going to overcome the resistance at the beta cell, you make more insulin, you now have better
00:27:53 glucose control. Everybody wins. Now, it's not clear that that has anything to do with the weight
00:27:58 side of it. That's a separate issue. And I want to actually talk about that because there are two
00:28:02 very interesting theories as to why these things cause weight loss. But point here is, wouldn't you
00:28:08 expect to see higher levels of insulin in someone taking a GLP-1 agonist to achieve that better
00:28:15 glycemic control? Yes. And that's not what you see. And I don't have an answer for this. I've seen the
00:28:21 same thing, including personally, I've been experimenting with terzepatide. My insulin is
00:28:26 five now, which is lowest that you can possibly get it. There was a part of me that was worried
00:28:31 that I was going to go against my own whole theory about... Have you checked postprandially?
00:28:38 Have you done an oral glucose tolerance test? Because that might be something to do to see
00:28:42 what is happening to postprandial insulin along with postprandial glucose, which of course will be
00:28:47 better. No, I haven't. That would be an interesting test to do. Yeah. I've won a CGM.
00:28:51 My A1C has gone from 5.4, 5.5 to 5.0. And my insulin is down to 5.0 as well. So...
00:29:01 And did you lose any weight? I lost a little bit of weight, not a huge amount,
00:29:05 six or seven pounds, which was never the goal to start with. And no loss of muscle mass,
00:29:11 which actually is the big boogaboo that people will have you fear. No loss of muscle mass if you
00:29:17 are exercising. So for me, it's an experiment. I haven't decided this is something I'm going to
00:29:21 continue, but I just wanted to really experiment for myself to try to see, okay, what is this drug
00:29:26 really doing? And it's been nothing short of remarkable, I think, in some way. One of the
00:29:30 most surprising has been for me this feeling of satiety. You hear about satiety. I was never in my
00:29:36 whole life the type of person that fell full. I could always eat more. And all of a sudden,
00:29:41 after about two weeks on this, I just looked at my plate and said, I'm full. And I heard myself
00:29:46 saying this, and I just felt like, well, this is really completely different. And for me,
00:29:51 you know, the reason why I'm excited about these drugs is, and by the way, this is not an endorsement.
00:29:57 This is something to...
00:29:58 Yeah, yeah. This is self-experimentation, curiosity.
00:30:00 Self-experimentation, which is a long part of the tradition of our field. The whole idea is really,
00:30:05 the thinking was, one of the biggest advances in longevity medicine is this idea that a range
00:30:12 is meaningless. And as a practicing physician, you know this. I went to medical school and we
00:30:18 were told that your blood pressure has to be 130 over 90. And that was still the normal range. So
00:30:23 you could be 128 over 88, and you were still considered normal. The same thing, I went to see
00:30:29 my personal physician and told him, my blood sugar is creeping up every year that I'm doing it,
00:30:35 and now it's 96, fasting blood sugar. And I'm worried because soon I'm going to be...
00:30:39 And he said, it's below 100. It's okay.
00:30:41 He told me you're normal. Don't worry. And I told him, I said, what is normal? And I think this
00:30:46 really is where I think longevity medicine is going to make an important impact. It's really
00:30:50 sort of revisiting.
00:30:51 I can't tell you how many times I've had this argument with people about glucose.
00:30:55 And here's the funny thing. We have the literature. In other words, we have literature in
00:31:00 non-diabetics that says the lower the A1C, the lower the all-cause mortality. It's a monotonic
00:31:08 reduction that knows no lower limit.
00:31:11 I'm with you.
00:31:12 So we say that up to 5.6 is normal. And if you're at 5.6, you're fine. But 5.5 is better
00:31:19 than 5.6. And 5.4 is better than 5.5. And 5 is better than 5.4. And 4.8 is better than 5.1.
00:31:25 But my point is, I also find it, I don't know what the word is, maybe sad. I find it sad that
00:31:31 we've simplified this problem in an effort to communicate, but have lost the essence of
00:31:38 where is lower better? Because it's not always true in biology. When you look at TSH, for example,
00:31:43 when you look at thyroid hormone, much more narrow band in which we would say there's optimal. If it's
00:31:48 too low or too high, it's problematic. But it turns out that when it comes to average blood glucose in a
00:31:54 non-type 1 diabetic or someone who's taking insulin, under natural physiologic circumstances,
00:31:58 it's just better to be lower. And as you age, it just keeps creeping up.
00:32:03 Same thing for blood pressure.
00:32:04 Yes.
00:32:05 They're revisiting the number every five years in terms of making it lower. I think if your
00:32:11 blood pressure is at 105 over 65, you're better off than if you're 115 over 75.
00:32:16 That's right. Provided you're not symptomatic, lower is always better.
00:32:19 You know, I'm frustrated, but I'm also excited by the fact that this is now becoming
00:32:23 norm in a whole new field of physicians who are more aware of actually what is health. And the
00:32:29 same for your weight. We know that that's the thing that is really interesting in the whole
00:32:34 aging field is this idea that everything is a J curve. So there is a sweet spot where you want
00:32:39 to be. And quite often it's broad enough that you can maneuver this in a way to optimize people's
00:32:44 health.
00:32:44 What do you think is the relationship between, I mean, body weight is so crude, but maybe we can
00:32:49 even talk about it through adiposity, body fat and longevity once correcting for metabolic health.
00:32:56 So it's obvious that so much of the relationship we see between body fat and poor health is really
00:33:04 just a proxy for something that's harder to measure, which is metabolic health. It's very
00:33:07 easy to measure body fat and we estimate body fat from BMI. And so that's why we have all these
00:33:11 population data from BMI. But if you have the luxury of working with actual patients, I couldn't tell
00:33:17 you the BMI of one person I take care of, but I know everybody's body fat, everybody's visceral fat,
00:33:22 and everybody's oral glucose tolerance tests. We know what we know and we know what matters.
00:33:26 Are you convinced that adiposity per se is problematic? Or do you believe that a person can have
00:33:34 excess body fat, but be metabolically healthy and confer the same longevity benefit as a metabolically
00:33:43 healthy lean person? We know there are people who are considered overweight who are metabolically
00:33:48 healthy. Yes. Easily 20% in my experience. Yes. And these are facts. No one can dispute them. You can
00:33:56 be overweight and metabolically healthy. What I worry about is the long-term effect. Do you mean from an
00:34:02 orthopedic perspective with the other complications that come from excess weight? Or are you saying that
00:34:07 they're basically increasing their probability of eventually going off the metabolic slide?
00:34:11 Both. Honestly, I don't know what the data says, but my worry would be that you might be
00:34:17 metabolically looking healthy when you're 40. But if you sustain this for 20 years, clearly visceral fat
00:34:25 is highly predictive of everything. The other thing I'll say also, the BMI itself is that my BMI is at
00:34:32 the border of being overweight. I am overweight by BMI, I think. I'm four pounds. If I lost four pounds,
00:34:37 I would get down to a BMI of 25. And I have 11% body fat.
00:34:41 So I don't worry about it because I know all in all, I'm metabolically healthy. My numbers are good
00:34:47 and all this. So in some way... It's not a particularly helpful... I mean, it serves its
00:34:51 purpose at the population level, but it can't be used to make a decision about an individual at all.
00:34:55 Exactly. But it can also sometimes become a confounding variable. And when people do studies
00:35:00 and they use these numbers and they make predictions or they draw conclusions that are really not based on
00:35:06 the fact that high BMI fraction of the population is heterogeneous in terms of metabolic health.
00:35:12 So my colleagues at the Buck, Nathan Price and Lee Hood, I've actually published a paper.
00:35:18 Wait a minute. I didn't realize they were at the Buck.
00:35:19 Yeah. Both of them.
00:35:20 They were up in Seattle before, weren't they?
00:35:22 Yes. We recruited both of them actually in the last two years.
00:35:25 Oh, congratulations.
00:35:25 Yeah. Thank you. I think this is transformative for us.
00:35:28 Fantastic, guys.
00:35:29 And really exciting. Lee is still partially in Seattle. So he's partially at the Buck.
00:35:34 We've established a collaboration with Phenome Health. And Nathan was at Thorne and still a CSO
00:35:40 at Thorne, but faculty member at the Buck. And they're really helping us to do something really
00:35:45 exciting. Along these lines, for example, they had a paper describing this BMI, but biochemical BMI,
00:35:51 based on biological markers that essentially assess your metabolic status. So I think that
00:35:57 those tools are available and it's a question of educating the physicians.
00:36:01 And do you know what makes up that biological BMI?
00:36:04 No. I'll give you the paper.
00:36:07 Okay. We spent a little more time on metabolism than we did immune health and the immune system
00:36:12 overall. I'd actually like to go back and talk about it a little bit more. I think,
00:36:16 again, the listeners of this podcast are very familiar with the metabolic stuff. We haven't
00:36:20 had as many discussions on the immune system. Talked about it at length with respect to cancer.
00:36:26 Had Steve Rosenberg on a few years ago. That was a fantastic discussion explaining the role of the
00:36:31 immune system in cancer, which I think we're going to have to talk about here because I certainly feel
00:36:36 convinced that a big part of why cancer incidence goes up exponentially with age is the declining
00:36:43 immune system, not just the accumulation of mutations, although I imagine they both play a
00:36:48 role. But I will tell you something else, Eric, which is, you know, I wrote a book a couple of
00:36:52 years ago about the space. And in the book, I talk about these things called the four horsemen.
00:36:56 And I describe them as the four things that are basically coming for us all. If you manage to
00:37:02 outlive youth, this is not to diminish the role of trauma and other things that are deadly. But for
00:37:07 many people living in OECD nations, it's going to come down to ASCVD, cancer, dementing and
00:37:14 neurodegenerative diseases and metabolic diseases. And people often say, Peter, is there anything you
00:37:19 wish you'd written in the book that if you go back in time, you would do? And I say, yeah, there are
00:37:22 probably many things if I thought about it. But the first thing that jumps out is I really should
00:37:27 have added a fifth horseman, and that is immune health and the types of infections that ravage
00:37:34 people in old age that a young person would laugh at. Thank you for bringing this up. Immunology and
00:37:40 aging have been not really mixing very well. One problem is that immunology is an extremely complex
00:37:45 and advanced field, along with neuroscience, one of the most complex. So when you go to an aging
00:37:51 meeting, there is no one talking about immunology. You go to immunology meeting, there are very few
00:37:56 people talking about aging. We try to navigate, even the nomenclature is being used differently.
00:38:02 People in immunology talk about immunosenescence, meaning aging of the immune system. They don't
00:38:07 mean senescence the way we talk about it in the aging field. So that yields all kinds of crazy
00:38:12 communication problems. Yeah, because if you're in the aging field, then you hear immunosenescence,
00:38:17 you think of SASPs and things that are being secreted by T cells. It just means aging of the immune
00:38:21 system. Now, the reason why I think this is a tragic failing for both fields is what happened
00:38:29 during COVID. It became obvious that your risk of infection was not linked to your age. The virus
00:38:35 infected everyone across, but the outcome could be completely different with 84-fold excess mortality
00:38:42 if you were above 75, 84-fold. Now, when this happened, and we can go in terms of trying to
00:38:49 understand why did this happen, what are the reasons for this? I went and started to look
00:38:53 at the literature. Influenza, it's exactly the same thing. RSV, same thing. So all of these viruses
00:39:00 that you can contract in later years will kill you with really significant rates. Influenza, I think
00:39:06 30,000 people die every year from influenza. The mortality in terms of COVID was really highly
00:39:13 segregated into the older part of the population or in that part of the population that showed
00:39:18 accelerated aging, obesity, and so on. Do you think that most of the mortality,
00:39:25 anytime we saw a gap in mortality, whether it was young versus old, whether it was obese versus non-obese,
00:39:31 diabetic versus non-diabetic, anytime you looked at that, you saw a difference in mortality.
00:39:35 Do you believe that it was always a difference in immune function? I mean, with young versus old,
00:39:40 it's very obvious, but do you think that was also true in the other comorbidities?
00:39:44 I would say so. And it comes from two reasons. One is there are two broad immune systems,
00:39:50 what we call the innate and the adaptive immune system. I don't know if you want me to-
00:39:54 I would. I actually was going to say, I think it is worth going full bore on this. I think it is time
00:40:00 for people to roll up their sleeves and understand arguably the most interesting system in the human
00:40:05 body. I am biased. I spent two years at the NCI doing immunology, but I think this is such an
00:40:11 interesting field.
00:40:12 Our immune system is built to recognize foreign elements. That really is why it evolved. It has
00:40:20 two lines of defense against microbes, bacteria, viruses, fungi, all of those. We are constantly
00:40:27 bombarded by those. It is actually amazing because, I mean, the evidence of this is if your immune
00:40:31 system doesn't function, the bubble-
00:40:33 It's incompatible with life.
00:40:34 It's incompatible with life. So we are colonized with bacteria in and out,
00:40:38 on our skins, everywhere. So we constantly respond to them in an appropriate manner.
00:40:43 And we survive everything, including disruptions to the barriers.
00:40:47 Absolutely. Absolutely. So we have two lines of defense in the immune system. First,
00:40:52 the so-called innate immune system, which is your macrophages, your dendritic cells,
00:40:56 but also pretty much every cell has a whole series of mechanisms that are not pathogen-spoken,
00:41:03 specific. That is, they will recognize an intruder, be it a virus, be it a fungi, be it bacteria,
00:41:11 and it will activate a first line of defense. Those line of defenses are nonspecific, and therefore
00:41:17 they're less effective. And they give time to the so-called adaptive immune system, which
00:41:23 is the second part, which is made up of T cells and B cells. And both of those cells have
00:41:30 highly selective defense mechanisms. The B cells make antibodies, which will go recognize a bacteria
00:41:37 or a fungus or virus, and the T cells, which are able to actually kill the infected cell itself.
00:41:44 So it will recognize when the cell is colonized by a foreign pathogen and will kill it. So the time
00:41:52 course of these is that once you encounter a pathogen, you will activate your innate immune response.
00:41:57 Typically, it can be fever, it can be all kinds of symptoms, but activation of this defense.
00:42:03 And this gives the whole organism a couple of weeks to actually build the defense
00:42:07 for the specifically recognized disorganism.
00:42:11 Let's talk a little bit about memory within that system.
00:42:13 So the innate immune system does not really have a true memory. It will always react in the same way,
00:42:19 no matter how many times.
00:42:20 If your kids are ping-ponging the same respiratory virus at you from school,
00:42:25 your innate immune system has the same playbook. Fever, you're going to get red.
00:42:31 Inflammation.
00:42:31 Yeah, you're going to get sore.
00:42:33 Yes.
00:42:33 All of those things are going to happen regardless.
00:42:35 Exactly. Yeah. And that's in contrast to the adaptive immune system, because once the initial
00:42:41 response has been generated, either via an infection or a vaccination, this is what a vaccination is,
00:42:47 presents you with a given fraction or the whole virus or a part of it. Your body will mount a
00:42:54 response, and this will lead to the amplification of a subset of cells that are selective. So think
00:42:59 about your T cells or your B cells. None of them are the same. We have a process by which we generate
00:43:06 so-called diversity, which is billions of different forms of antibodies or T cell receptors that are
00:43:13 recognizing, in principle, every chemical structure or every protein from a microorganism.
00:43:20 Now, what happens during the initial encounter, either being a vaccination or an infection,
00:43:26 is those B cells or those T cells that have a receptor that is able to recognize the pathogen
00:43:33 will become amplified, and they will turn out a large amount of the antibody or the T cell clones.
00:43:40 Once the job has been done, they will contract, but they will not contract back down to the same
00:43:48 level. They will become what we call memory T cells or memory B cells. So that if you encounter the
00:43:54 same antigen in the future, the reactivation process is shortened, the maturation happens faster. So
00:44:02 eventually, the whole idea of the vaccination is to sort of get yourself ready with a subset of memory
00:44:07 T cell clones or B cell clones that once the true virus will come, you will be able to mount a response
00:44:14 within a few days or up to a week. And so that's how vaccination works. Now, what's interesting during
00:44:20 aging is, and people are not aware of this, if you're above 70, most vaccinations do not work. So people then
00:44:28 will ask, actually, your immune system has aged and your vaccination rate really decreases very strongly.
00:44:35 From what I remember, this might be different in different populations, but vaccination rate
00:44:39 success is close to 30% if you're above 70.
00:44:43 During COVID, what was the risk reduction for a person over 75 who was vaccinated versus not vaccinated?
00:44:52 It was almost complete reversal of the effect in terms of the protection.
00:44:57 Meaning it was highly, highly protective.
00:44:59 Yeah, it was protective.
00:45:00 So how do we reconcile those two facts?
00:45:02 That's true. To be honest, I don't know how this has been studied. I would be happy to read about
00:45:08 this.
00:45:09 Because the COVID vaccine seems to have had a remarkable risk reduction in very old people.
00:45:16 Didn't seem to have an impressive risk reduction in younger people because the absolute risk was so
00:45:21 low. It didn't seem to matter that much. But boy, did it matter in older people.
00:45:25 But did it matter at the population level or at the individual level? This is what I'm not sure about.
00:45:31 I certainly don't want to go on record saying something. I think we can find the answer and
00:45:35 put it in the show notes. My recollection, which could be wrong, is that the older a person got,
00:45:42 the greater the benefit they got from COVID vaccines with respect to mortality. So I guess the question
00:45:49 is, let's maybe talk about other vaccines. Is that not the case with influenza? Is that not the case
00:45:56 with pneumococcus or any of the other vaccines that are used primarily in older adults?
00:46:01 In general, and I'm not a vaccine specialist, but the thinking is that there is a dramatic decrease
00:46:08 in the efficiency of vaccination against influenza, against RSV, against all of those as you age.
00:46:15 The thinking then is, how does it work at the population level? And this is where the whole
00:46:20 concept of herd immunity works, is that if you limit the spread of the infection in a family,
00:46:25 for example, you're much less likely to infect grandpa. So that's been my understanding of how
00:46:31 most of these viruses, these vaccines work.
00:46:33 Yeah, I'm asking a different question. That's an important question. I guess I was asking,
00:46:37 obviously they didn't probably do a randomized control trial, so you've got all these confounders in
00:46:42 it. But I wonder if they just looked at all comers to the hospital, vaccinated versus non,
00:46:49 let's try to control for all the confounders. If the hazard ratio is 1.2, it means nothing or 0.8.
00:46:56 But if the hazard ratio was 0.2 or 8, well, you'd say even with the confounders, there must be some
00:47:04 high degree of protection that came from that. So anyway, I'm sure someone listening to this knows
00:47:09 the answer to that. We'll try to find the answer and put it in the show notes. But let's go back
00:47:13 to the why. Why is it that as a person ages, they're less likely to respond to a vaccination?
00:47:19 Is it because A, their immune system, the adaptive immune system is less able to recognize the foreign
00:47:26 pathogen and build up a high enough reserve of T cells and B cells that will respond? Or is it B,
00:47:34 that they can do that, but the ability for those cells to stay in a memory state and be reactivated
00:47:42 is somehow impaired?
00:47:44 I think it's both. Isn't everything aging? But there's one aspect which is really unique,
00:47:50 at least in terms of T cell, which are really instrumental in terms of most vaccine response,
00:47:56 is the fact that these T cells are generated, the diversity of the T cells is generated by the
00:48:02 thymus. And the thymus, a small organ behind the sternum.
00:48:07 How big is your thymus and my thymus right now?
00:48:10 I'm 68. So it's probably very, very embryonic and it's probably not much left after age 50.
00:48:16 And most people, you find it very small.
00:48:19 Whereas when you're young, it's actually, you can see it on an imaging study. And I would imagine
00:48:24 if you and I had a CT scan of the chest, you'd barely be able to pick it up.
00:48:27 Exactly. And it's replaced by fat actually in most people as you age. Although there is some
00:48:32 somewhat controversial evidence that there might still be some clones that can be reactivated,
00:48:38 even in older people. And human growth hormone, as you know, is one of the interventions that has
00:48:43 been shown to actually re-induce thymogenesis.
00:48:47 So let's talk about that a little bit. Are you referring to that Fahey paper from about
00:48:51 seven or eight years ago that looked at growth hormone with metformin and DHEA or something
00:48:56 like that?
00:48:57 That's one. But that Fahey paper was actually inspired by work of a colleague of mine when
00:49:03 I was at the Gladstone Institute who did this, actually Mike McCune and colleagues did this
00:49:10 in patients with HIV who are chronically infected with HIV where they lose a lot of their CD4 T cells.
00:49:16 And there was an interest. So there's a big lesion initially in infection. And there was
00:49:23 an attempt to actually try to see if you could regenerate these populations to bring them back to
00:49:28 our normal. Because even though we had great drugs against HIV, they could not bring those patients
00:49:34 back to normal. There was a remaining original insult. So they did a trial with human growth
00:49:40 hormone that were able to show some degree of thymogenesis and increase in naivety cells in
00:49:47 these patients. And I believe the Fahey trial actually tried to reproduce this. I think there's
00:49:52 a second Fahey trial that is ongoing or, but I haven't seen the results.
00:49:57 Yeah. I mean, the first one was, I don't remember the results. The cocktail was a little suspect.
00:50:03 Right. So the GH made sense, if that's your hypothesis. I believe, I've never spoken with
00:50:08 Greg, but I believe reading the trial, the metformin, which was really given at a homeopathic
00:50:15 useless dose. I think it was only given at 500. So apologies if it wasn't, but I think it was only
00:50:20 given at 500, was meant to offset the glucose metabolism disturbances of GH. Do you remember why
00:50:26 the DHEA was given? No. There was some reason for it that made sense on paper, but didn't make sense
00:50:34 physiologically. Now, the more important question is, my take on that trial was it was a single
00:50:39 active agent, which was growth hormone. Like I don't think DHEA does anything. I don't think 500
00:50:43 metformin does anything. So the question is, and it was a very small trial and I think it was open
00:50:47 label. I have a significant problem with the readout of that trial. So that's what I wanted to ask you
00:50:52 about. Remind me of the readout. The readout was one of the clocks. Ah, that's right. This was,
00:50:58 Steve Horvath was the other author on that paper. Exactly. And actually this whole story sort of
00:51:03 pushed us into a whole project that we've published down on what we call entry and clock,
00:51:08 because there was in the experiment, in the patients, they'd indeed observed some increase
00:51:13 in the fraction of naive T-cells, which tells you and me that something worked. The fraction of naive
00:51:20 T-cells increased with respect to the memory T-cells. The naive T-cells are the ones that are
00:51:26 generated in the thymus. They're naive because they have never met their cognate antigen and they sit
00:51:31 there waiting for something to happen. So the whole idea of treating with human growth hormone was to
00:51:36 induce thymogenesis and to restore the pool of these naive T-cells. So I think to some degree it
00:51:42 worked at low level. Then they used the clock on the whole blood. And my worry when I saw the paper,
00:51:48 which is a worry that actually existed, predated this, and was also a worry when people were using
00:51:54 telomere length, is the idea when you sample the blood, as an immunologist, I know this is a highly
00:52:00 dynamic organ. Think about the blood as an organ. We enumerate at this point today with the best
00:52:06 technology, more than 500 different populations of cells in the blood. Suppose that these cells vary
00:52:13 in response to any intervention and that these cells individually have a different epigenetic age,
00:52:20 you would have the impression that you are rejuvenating, which was the claim of that Fahey
00:52:26 paper that they had rejuvenated people. But in effect, what you would do is simply change.
00:52:31 Yeah. It's like you're on a sine wave that goes like this and you take two sample points. They could
00:52:36 be here. They could be here. They could be here. And by the way, as you probably know, Matt Caberlin has
00:52:41 famously purchased, I think, four or five of the commercially available aging clocks. He bought
00:52:48 them in duplicate and did all of them, sampled them all simultaneously. Two of this, two of this,
00:52:53 two of this, two of this, simultaneously take 10 samples. And not only do all the clocks disagree
00:52:59 with each other, but even within the same clock, there was disagreement, significant disagreement.
00:53:05 So yeah, I mean, I want to actually come back and talk about clocks in some detail, but given that
00:53:09 that study was done years ago with an older clock, I think the clock part of it is not even remotely
00:53:15 interesting. I think the more interesting question is, was there genuine thymic regeneration? If so,
00:53:22 how do we reconcile a very pressing and vexing question within gyroscience, which is the role of
00:53:29 growth hormone? So I've never taken growth hormone. I've never, I shouldn't say I've never prescribed
00:53:33 it. I've prescribed it in very rare circumstances for injury healing, but I've never prescribed it for
00:53:38 longevity benefits, but a lot of people are out there doing so. And as such, I've had lots of
00:53:44 patients who come to my practice who have been taking or are on growth hormone. And I will say this
00:53:49 to a person, every single one of them has said, I feel so much better when I take growth hormone than
00:53:56 when I do not. I mean, across the board, 100%. And I can't actually point to evidence that tells them
00:54:04 it's bad to take. I can just say it doesn't make sense to take if our goal is to reduce the risk
00:54:10 of cancer. And if our goal is to slow the aging process. So what is your take on that? Just your
00:54:17 intuition, or is there any data you're aware of that would lead one to think that, well, maybe we
00:54:22 could pulse a little bit of growth hormone here and there. If we get some thymic regeneration, we don't
00:54:26 have to be on it all the time. I mean, how would you think about that?
00:54:28 I do worry about it. I'm not a specialist on growth hormone itself. It induces diabetes.
00:54:35 It induces glucose intolerance. So from that angle, I do worry about what it would do chronically,
00:54:43 especially in someone young. It's a bit like increasing your protein intake. There's clear
00:54:48 evidence that increasing your protein intake, especially as you age, becomes beneficial. And
00:54:53 the people who have higher protein intake actually do better in terms of muscle mass and so on.
00:54:58 So in someone who is 65 to 70, who is starting to feel the effect of manifest some form of
00:55:07 sarcopenia, there might be a benefit for that person to actually increase muscular mass and all the
00:55:13 benefits with this, especially if it's not done continuously.
00:55:16 But I mean, I would argue there's no doubt that there's benefits, but you're going to get far more
00:55:20 efficacy from testosterone or anabolic steroids when it comes to mitigating sarcopenia. Growth hormone
00:55:27 actually is not remarkable at inducing muscle mass. It's nowhere near as effective as testosterone.
00:55:34 It's more effective at eliciting fat loss. But I wonder if there's something that goes beyond that.
00:55:38 Because I think when people tell me they feel better on it, I think they're talking about less aches and
00:55:43 pains. Joints just feel better. I don't think anybody's saying they feel better because their thymus is
00:55:49 more plump. But I wonder, that to me would be a reason to potentially consider a schedule,
00:55:56 an intermittent schedule of something. If it's, again, going back to my macro thesis here, which is
00:56:01 I've been harping on these four horsemen, four horsemen. Well, if we introduce a fifth horseman,
00:56:06 what is the strategy? Because I can give you chapter and verse the strategy for how you will mitigate
00:56:13 heart disease, cancer, all of these other conditions. What is our strategy for mitigating
00:56:18 immune decline?
00:56:19 I would say the same as a strategy that would mitigate decline in every other organ. There's
00:56:25 clear evidence that the effect of exercise on immunology is the same as in every single...
00:56:29 So I'm not familiar with it. So tell me a little bit about that. I don't know
00:56:32 specifically how exercise impacts the immune system.
00:56:35 I cannot speak to specific papers. Clearly, there's evidence that people who exercise actually
00:56:41 respond to infection better, respond to vaccination better. So that's all been documented.
00:56:46 I cannot speak to specific studies.
00:56:49 Do you have a sense of mechanistically why that's the case?
00:56:51 It is so complex, I would say. I would not be able to tell you. But that being said,
00:56:57 I think the whole line of investigation to induce thymic rejuvenation, I think is an important
00:57:04 one area, especially if we're thinking about increasing lifespan further for what we are
00:57:09 doing now. That's in the future, it will become one of these rate limiting steps. It's a bit
00:57:14 the same situation as the ovary, where the ovary and the thymus, we call them the canary in the
00:57:20 coal mine. I mean, there really are specific organs that show accelerate aging way earlier than
00:57:26 other tissues. Now, the question is, why is the thymic involuting so early? I think it's probably
00:57:31 because evolutionary, we were never meant to live this old. And so that really is one of the
00:57:37 thinking that goes on. That's going to be, in the long term, one of the problems that we have to face.
00:57:42 And this is something we're actively studying. And the lab is trying to, we just completed a study
00:57:47 where we are looking for novel biomarkers that are predictive of whether you will respond to a
00:57:53 vaccination or not. And there's something done in collaboration with Mark Davis at Stanford,
00:57:57 using the 1000 Immunome Project, which is one of the largest studies studying aging and immune system
00:58:03 only in humans. So we've been able to, studying people to identify some metabolites that are
00:58:09 associated with poor response to vaccine. And so those are not only markers, but they could also become
00:58:15 tools that we include as adjuvant or as a pretreatment theory. I'm sure you're familiar with the work of
00:58:21 Joan Manick. Of course. Yeah. I was going to ask you about Manick and Clickstein in a moment. Before
00:58:26 we do, I want to go back to this point here, which is biomarkers are so important. When I think about
00:58:33 cardiovascular disease, and even though it's the leading cause of death, why I tell my patients,
00:58:38 it's the one you need to be least afraid of if you're willing to be proactive in management.
00:58:43 And it comes down to the fact that we just have such a clear understanding of how the disease works
00:58:48 and we have exceptional biomarkers. So we can measure the things that are causing the disease.
00:58:54 We can measure inflammation. We can measure ApoB. We can measure VLDL cholesterol, LP little a.
00:59:01 We can measure blood pressure. We can measure metabolic health. And we know how to address
00:59:06 those things. And we know that when we address those things, we can measure whether what we're
00:59:10 doing is working. Okay. So problem solved, basically. When it comes to the immune system,
00:59:16 we're going to talk about Manick and Clickstein in a moment, but as we saw from their paper 10 years
00:59:22 ago, they gave a rapamycin analog to people, people who were in their sixties, vaccinated them
00:59:28 and demonstrated that, oh boy, you got a much better immune response. Okay. They were able to
00:59:33 demonstrate that using laboratory techniques. I'm sure they used flow cytometry or something like that
00:59:40 to measure it. How close are we to being able to do that sort of thing commercially?
00:59:45 By commercially, I mean over the counter.
00:59:48 Not close. I think in that study, they actually measured antibody titers.
00:59:53 So even more complicated than flow cytometry.
00:59:55 Yes.
00:59:56 Okay.
00:59:56 In that case, they definitely showed an enhancing effect with a known gyroprotector.
01:00:02 A suspected gyroprotector, at least in humans. 1.00
01:00:04 Exactly. A rapalog. And they showed not only increased titers, but also protection, 0.60
01:00:10 increased protection. Eventually the clinical trial failed for a whole series of other reasons,
01:00:14 which were in part due to the way that the FDA imposed the trial to be generated. I think it
01:00:21 just complicated the whole picture. Yeah. By the way, for folks listening to us who are confused by
01:00:25 that, Matt Caberlin and I had a specific discussion because it wasn't the 2014 trial. It was a later
01:00:33 trial. It wasn't the RAD001 trial. It was the other trial that failed. And I actually don't remember
01:00:39 the reason, but Matt explained it. It was very clear that it was a tragedy of bureaucracy.
01:00:47 It is.
01:00:47 And it shouldn't be viewed as a black eye on that molecule.
01:00:51 Yeah. Matt is more of a specialist in the whole rapamycin. So I will defer to what he said.
01:00:55 We'll link in the show notes to where Matt and I had that discussion.
01:00:58 Yeah. What I've heard from anyone that I've talked to, including Joan, is that
01:01:01 this was in some way bungled, which is sad because sometimes things like this can put a field back
01:01:07 for a number of years and discourage investors. We have a startup that originated at the Buck
01:01:13 called Eovian, which has raised $50 million. Again, coming up with rapalogs, novel rapalogs that
01:01:19 are going to be, I think, revisiting that whole picture. So we're quite excited. The field is far
01:01:24 from being dead. Will we ever be able to measure this in people the way we measure hemoglobin A1c or
01:01:30 things like that? Or is it going to be one of those things where it's a bit of a leap of faith
01:01:33 and you're going to have to look at the clinical trial where the outcome was there. And then you're
01:01:37 just going to have to say, well, even though there was probably massive heterogeneity amongst the
01:01:43 participants in the trial, we're going to dose this thing individually. I mean, it's a little bit like
01:01:47 you brought up vitamin D earlier. I mean, one of the problems with the vitamin D trials
01:01:51 is that they're all garbage because they all just give people a given dose. They don't measure the 0.92
01:01:57 response. They don't measure compliance. A vitamin D trial should be done based on target level,
01:02:03 not target dose. And we run the risk here of the same thing in a much more complicated system.
01:02:10 Agree. That being said, measuring pathogen-specific titers is done routinely in the clinic. I don't know
01:02:18 if you did this, but I just had my measles titer measured. I was born in 1957, which is right the age
01:02:25 before 1957. Everyone was exposed to measles. So you're typically safe, but you should measure
01:02:34 your titer to determine whether you need to be revaccinated. I found out that, yeah, you can do 0.98
01:02:39 this very easily. You get a titer. But would the titers by themselves tell you, so what would you
01:02:44 predict? If I measured every titer right now? Yes. If I measured polio, shingles, did a pan titer on you
01:02:52 and then started you on rapamycin for eight weeks and then stopped it and then remeasured your titers
01:02:59 without vaccinating you, what would you expect to see? I would not expect them to change.
01:03:03 Yeah, exactly. So how do we know we're improving your immune system if indeed we have?
01:03:08 Oh, I see what you're saying. So in terms of if we were to start you on rapamycin,
01:03:13 what would happen? How could we measure the improvement in immune function?
01:03:16 By the way, the manic trial showed that first they did a one-month treatment with the rapalogue
01:03:23 before vaccination. They demonstrated not an effect on existing vaccinations, but only demonstrated on
01:03:30 de novo vaccination. That's right. Yeah.
01:03:32 And I think what would be the effect on existing titers against all of the other pathogen? I don't
01:03:37 know. I don't think this has ever been done.
01:03:39 Yeah. Interesting. You want to just say a little bit more about that trial? So that was at least
01:03:44 for me, a pivotal moment in my journey in this space and in understanding this world. So that
01:03:49 was December of 2014, that paper came out. And if I recall, roughly 300 plus participants divided
01:03:55 into four groups. So placebo group, a group that got one milligram every day, a group that got five
01:04:01 milligrams once a week, and a group that got 20 milligrams once a week. Two people were pulsed,
01:04:06 one much higher than the other, and then one given daily, and then a placebo.
01:04:09 I believe they were all over 65. I think the study was done in Australia. And as you said,
01:04:14 they were put on their, whatever treatment was for four weeks, immunized. I think it was another
01:04:21 four weeks and then a six week washout. And then the titers were checked. The best response I think
01:04:27 was in the five milligram pulse and the 20 milligram pulse. The one milligram daily still had a better
01:04:33 response than the placebo, but not as strong as the two pulse doses. But the five and the 20 weekly
01:04:38 were nearly identical, but the 20 had much more side effects. I don't remember perfectly. So
01:04:44 correct me if I'm wrong. You remember pretty well.
01:04:46 The takeaway was basically five milligram pulse was the sweet spot. You get all the benefit without
01:04:51 the side effects. That's how I remember that trial as well. Although I'm always impressed by
01:04:55 how you remember all of the details of these clinical trials. What was remarkable about that data
01:05:01 was the fact that this is from a drug that is supposed to be an immunosuppressant and it's been
01:05:08 a long road for the longevity field to try to get our colleagues who are actually using a rapamycin as
01:05:15 an immunosuppressant to have them believe that this actually has an effect on immunity and not only
01:05:20 not immunosuppressive, but actually a promoting immunity. How do you reconcile that? Not their
01:05:25 disbelief, which is warranted, but how do you reconcile that one molecule? So if you think
01:05:31 about the doses we used to give rapamycin, it's not actually used that much, by the way,
01:05:35 today in the transplant clinic. So FK-506, I'm blanking on what FK-506's real name is,
01:05:42 but anyway, whatever. It's largely displaced, sirolimus, which is rapa. But that said, when we
01:05:47 used to give it out, we were giving two to four milligrams a day. Now let's just assume that it was
01:05:54 indeed contributing to prevention of organ rejection. Do you think it was doing so because
01:05:59 that's a high enough dose of constitutively giving a drug that it suppresses the immune system? Or do
01:06:04 you think it was only suppressing the immune system because it was being given in combination with two
01:06:09 other drugs and it was only as part of that sea of other drugs that it has the immunosuppressive effects?
01:06:16 I think there's clear evidence it is immunosuppressive by itself. I can tell you that for the period when I
01:06:22 was on rapamycin, I would take either four or six milligrams a week, every morning, once a week.
01:06:29 The biggest difference between the immunosuppressive and the geoprotective effect is really the amount,
01:06:35 the frequency and the amount. The reason why people adopted this once weekly dose is to
01:06:40 first not have any immunosuppression and second, to mitigate the secondary effect,
01:06:46 which are thought to be caused by inhibition of mTORC2, which is the second complex.
01:06:51 Yeah, the glucose effect. And that seems to be working largely. What was, in my case,
01:06:56 remarkable is that every time I took my dose, not two, I only did two for a couple of weeks,
01:07:02 but either four or six, the next morning I would have a pimple on my nose. So I was immunosuppressed,
01:07:08 clearly, every single time.
01:07:10 For a day.
01:07:10 Yeah, for a day, for a day or two. I sort of made peace with it in the fact that if I had a really
01:07:16 heavy workout, I would have exactly the same thing. Exercise is immunosuppressive. If you go all out,
01:07:21 you can get a cold. You're temporarily fragilized after a really heavy exercise. So I think
01:07:26 the difference really between these two worlds, the immunosuppression, which clearly has been
01:07:32 documented by clinical trials, it is immunosuppressive by itself, versus the beneficial effect on the
01:07:37 immune system. To me, it's a question of dosage and frequency.
01:07:40 And yet, I cannot reconcile the unambiguous success of the interventions testing program,
01:07:48 where those mice were eating rapamycin in every single bite of food they took. In fact,
01:07:56 they were consuming it more continuously than even the most immune-compromised patient.
01:08:02 And without exception, every single ITP study of rapamycin, whether they started in old mice
01:08:10 or young mice, rapa alone, rapa with another drug, it just doesn't matter. It always worked.
01:08:16 How do we reconcile that?
01:08:19 Well, I don't have the answer, but I can sort of talk about it. There's something that worries me about
01:08:25 our reliance on the mouse as a model system for aging, for studying aging, and how relevant it is to
01:08:32 us as species.
01:08:34 Even mice, because we would all admit that the ITP mice are the best, they're the Ferrari of mice.
01:08:38 Exactly. The ITP is the best way to address this question, because they're using mice that are
01:08:45 crossed. So it's not-
01:08:47 Not inbred, yep.
01:08:47 Not inbred. When you're using Black 6, you're essentially doing the experiment on N of 1. 1.00
01:08:52 And the whole world, I mean, 80% of the work that's being done in mice is done on Black 6. 0.86
01:08:58 We're all studying the same person. So obviously, when you go and try to transfer this to a human
01:09:03 population with all of its variation. So the ITP did the right thing. That being said, and this is
01:09:08 not an attack on ITP. I think ITP is a great program and should be funded, and should continue
01:09:13 to study this. I just worry about the over-reliance on ITP alone. And I think we should have another
01:09:21 system that studies primate interventions with drugs. There are a number of primates,
01:09:26 non-human primates that are actually much closer to us. The reason I worry about mouse is something
01:09:32 that actually Steve Ostad, you've had on this podcast as well. Steve is a good friend, and he
01:09:37 came up with something called the longevity quotient, which I think is something that people do not pay
01:09:42 attention enough. So the longevity quotient is this idea that if you look across the animal kingdom,
01:09:48 the larger you are, the longer you live. Okay? So you can take 1,000 species and you can,
01:09:55 on the x-axis, you have their size, on the y-axis, their life expectancy.
01:09:59 It largely rises to the right.
01:10:01 And you can see a monotonous curve. Now, there are exceptions to this. One of them is
01:10:07 naked mole rats, for example.
01:10:10 They punch above their weight. Dogs tend to punch below their weight.
01:10:12 Exactly. Although in dogs, again, this is between species. Then when you look
01:10:18 intra-species, it gets even more complicated, which is the larger dog lives shorter than the
01:10:24 smaller dogs, the Great Dane versus the Chihuahua. And that is down, actually, that's driven mostly
01:10:29 by growth hormone, which is, again, another reason why we should look at taking growth hormone as an
01:10:35 anti-aging drug with some degree of circumspection. Because in dogs, the more growth hormone you have,
01:10:41 the larger you are, and the shorter you live. We know also in humans, the larger you are, the taller
01:10:46 you are, the shorter you live. So are these effective growth hormone? Yes. Are they only
01:10:52 important while during the growth phase? That's a possibility, but it's something that really gives
01:10:57 me pause to go back to our discussion about growth hormone. So going back to the longevity quotient,
01:11:03 mice are also an exception. They punch below their weight. So they live shorter than they should based
01:11:09 on their size. And humans is the biggest exception. We live about five to six times longer than we
01:11:15 should based on our size, which tells me that we aren't a naked mole rat of primates. We do
01:11:22 incredibly well, which means that we already have optimized a lot of these pathways that are
01:11:28 promoting aging. I suspect the mice is exactly the opposite. I don't know that someone has really
01:11:33 compared sort of intrinsic TOR activity in mice. Are they, for example, living? Mice are, especially
01:11:39 laboratory mice, are engineered to reproduce and grow as quickly as possible. They have large litter
01:11:45 size. They do everything very quickly. Now we know all of these activities are requiring a lot of
01:11:52 anabolic strength, which is driven by TOR. So the question is, are the mice examples of animals that are
01:11:59 maximizing TOR activity to do everything they do very quickly? And we are maybe at the other end
01:12:05 of the spectrum where we have low basal TOR activity. So that's where I worry when people just
01:12:11 transfer everything we know from TOR from mice into humans and saying it's going to show and work in
01:12:16 humans. I don't know if you heard- Such an interesting point.
01:12:19 Yeah. And this is frankly why I stopped taking rapamycin. I thought I did not really see anything
01:12:24 in terms of anything. Metabolically, physically, muscle strength. I could not. In contrast to GLP-1
01:12:31 agonist, where I saw all of my numbers get better and functionally strength, all of this, I saw
01:12:36 everything getting better on GLP-1 agonist. With rapamycin, I never could tell whether I was taking
01:12:41 it or not. Yeah. Although it's not just that. I would say where rapamycin acts, I don't know that
01:12:49 we would see anything getting significantly better. Because if we think that the main places that
01:12:55 rapamycin is going to act would be on autophagy, well, there's no way you're going to measure
01:13:00 autophagy. You're not going to feel autophagy. You're not going to see it or measure it. Does it
01:13:05 tamp down on certain subsets of senescent cells? That's certainly plausible. Again, I don't know how
01:13:11 we're going to see or measure or necessarily even feel that. Does it reduce some of the tonic,
01:13:16 low-grade, unhelpful inflammation? Probably. But again, if a person doesn't have much to begin
01:13:24 with, it's going to be tough to measure. Conversely, GLP-1 agonists act directly on a
01:13:30 thing that is so easy to measure, which is glucose metabolism and body weight for those who are losing
01:13:36 weight as well. So it might not be a fair comparison. I guess the other thing I would add
01:13:39 to this interesting observation is that, of course, the mice and the ITP are still in a relatively
01:13:44 sterile environment. And it might be that even if they incur some immunosuppression, it's not going
01:13:51 to be as maladaptive as it would be if they were wild animals as we are.
01:13:55 They live in a sterile environment. They live grouped in a cage with no ability to move,
01:14:03 to exercise. They eat a diet which makes the American diet look like the most healthy thing ever.
01:14:11 I mean, have you ever seen the pellets that these mice are eating?
01:14:15 Do the ITP mice eat the crappy pellets as well? 0.82
01:14:18 I suspect they're eating...
01:14:19 Okay. I don't actually know what their diet is.
01:14:21 Yeah. I don't know what their diet is. I can guarantee you they're not eating salad
01:14:24 and fruits and vegetables. So in some way, they are an incredibly artificially bad sort of
01:14:32 environment. These mice are actually doing everything that is conducive to a poor health.
01:14:37 And so the fact that we see something that works in that system might have some value for
01:14:42 a fraction of the population that has a very poor lifestyle. I do worry about transferring this to
01:14:48 someone like you and I who are exercising or trying to eat well or trying to sleep and all of this.
01:14:54 I take these observations with some degree of caution. And frankly, when people ask me,
01:14:59 should I go on rapamycin? I do worry. Now, this is a different story if someone,
01:15:04 a patient comes and sees you at 40 years old and tells you, I think, I want to go on rapamycin.
01:15:09 I would strongly argue that you should not do this because even in the studies that have been
01:15:13 conducted, they still saw an effect in mice that were the equivalent of 65 to 70 years old.
01:15:18 Now, if you're 75 years old and you have the feeling you're chronically inflamed and you have
01:15:22 the feeling that things are not doing well, there are a number of anecdotal cases where people have
01:15:27 described really feeling a lot better and a lot stronger very quickly.
01:15:31 On rapamycin.
01:15:32 On rapamycin. But I would predict it would be the same thing with a growth hormone or
01:15:36 some of these interventions. So I've really put those in different categories. My argument to
01:15:40 people is today we have one intervention that is very profoundly anti-aging and it is physical
01:15:47 activity exercise in all of its forms. Once you have optimized this, I think let's talk about doing
01:15:53 something else on top of that.
01:15:54 Yeah. Earlier, you brought up the Sirtuin story and NAD. I'd love to spend a little bit of time
01:16:00 there. I was at a talk recently and as always, I get asked questions about stuff like that. And I
01:16:06 got asked the question about NAD and I said, look, this is one of those things where if I tell you the
01:16:12 following facts, I'm going to tell you three facts, NAD is completely ubiquitous throughout the body
01:16:19 and it is absolutely essential for the most important chemical reactions that happen in the
01:16:26 body. You cannot undergo redox reactions, metabolic reactions without NAD. That is point one.
01:16:32 Point two is a class of proteins called Sirtuins rely heavily on NAD as the substrate in the process
01:16:43 of repairing DNA. That is fact two. Fact three is as you age, NAD levels decline precipitously.
01:16:52 Okay. Those are three facts and I don't believe, is there any dispute to any of those facts?
01:16:56 No controversy.
01:16:57 Okay. Armed with those three facts, how could it be that supplementing NAD does not lead to
01:17:06 a longer, better life or some health benefit? That's a logical conclusion, right?
01:17:11 Well, not completely because it depends also what is the reason why NAD levels decrease and it depends
01:17:18 also what supplements, how you remedy and decrease. And this is something I'd love to talk about CD38.
01:17:24 So it could be that NAD levels go down because their consumption goes up. As we age, there's
01:17:30 more DNA damage, there's more consumption, the Sirtuins need more of it and it goes up. And then
01:17:34 of course the question would become, is the current level of NAD that we have rate limiting to that
01:17:38 reaction? If not, then all the extra NAD in the world should have no benefit because you're just
01:17:45 adding more substrate to a reaction where it's not needed. Conversely, if NAD levels are going down
01:17:51 because there's a production issue and if you provided more of it, you could actually do more
01:17:57 good, well then it could be the exact opposite story. So let me pause there for a moment and have
01:18:02 you fill in the edges of everything I just said so that we can go deeper into this discussion. So
01:18:06 maybe explain a little bit what NAD is, explain what it means in Redox. And obviously let's talk
01:18:11 about Sirtuins and the role that NAD plays there.
01:18:13 Lots to unpack there. It could be a two-hour podcast. It's one area that we've worked on for
01:18:19 the last 25 years. We were responsible for cloning the human Sirtuins actually after Lenny Guarante 0.99
01:18:24 published his paper on Sirtu and yeast. We were the graduate student-
01:18:29 Matt was the first to publish this, wasn't he? Cable and-
01:18:31 Yeah, actually Matt and Brian, I mean David, I mean that whole gang was the original gang along
01:18:37 with Lenny Guarante were paved the way for a lot of what we know. One thing that I would just start
01:18:42 by saying is that it pains me in some way in a field that is so rich and has generated so much
01:18:49 data that there's a whole cloud lying on top of Sirtuins and NAD. There's nothing there. I just tell
01:18:56 people it is an incredibly studied system. We are still juggling the complexity and I would argue that
01:19:04 any field where the same degree of investigation will be conducted will have the same controversy.
01:19:09 This is the nature of science. The beauty of science is that it's incredibly messy on the way
01:19:14 up, but eventually things are getting clarified. And I think in the terms of the Sirtuins,
01:19:19 we're still right in the middle of it. So there's some complete garbage. 0.98
01:19:23 Yeah. And by the way, I'm not completely dismissive. What I will say has made this field complicated 0.98
01:19:29 is that the leading proponents of it have all opted for a commercial pathway.
01:19:34 And therefore they have opted not to study this in a rigorous way, but to study it in
01:19:40 a commercial way. And I mean, I understand why you would do that. Like that's the nature
01:19:44 of it. And this is not a molecule that you're not going to generate intellectual property in
01:19:48 the same way that you would around a novel drug. And so it poses a limitation to how these
01:19:53 things can be studied. But unfortunately that coupled with the resveratrol fiasco, unless you
01:20:01 think otherwise, Eric, we don't need to talk about resveratrol. I remain completely convinced
01:20:05 that resveratrol had zero benefit whatsoever. I think it is an absolutely useless molecule. 0.87
01:20:10 So I think that resveratrol debacle, the overhype complete debacle of that, coupled with the fact
01:20:17 that all of the participants in the NAD landscape are doing it through their own commercial enterprise
01:20:22 with their own proprietary blend, has resulted in this inability to drive forward in this field.
01:20:28 I agree. You've identified the problems, the hype and the commercialization. I mean, commercialization
01:20:34 can be helpful if the companies are actually willing to invest in clinical trials and so on. I always
01:20:40 use the example of timeline, urolithin A. I mean, we've worked with them. They do clinical trials,
01:20:45 rigorous. They publish them in the best journals. And at the end, you know what you're measuring.
01:20:50 That being said for the sotuin, so let's try to maybe step back. And given the controversy,
01:20:55 I would say I would encourage your listener not to just discard it all. We're still in the middle
01:21:00 of it. And I think there's something interesting will emerge out of it. So NAD is a critical
01:21:05 intermediary metabolite. It has two big roles. First is it plays a key role in redox reactions. Again,
01:21:12 we've talked about these reactions, reduction oxidation.
01:21:15 Anytime electrons need to move around.
01:21:16 Exactly. So it exists in two forms, NAD and NADH. And it's critical to intermediary metabolism.
01:21:24 There are more than 600 of these enzymes that use NAD in the whole metabolism. So it stands to
01:21:30 reason that if you're losing NAD levels, you go below a certain critical level, these enzymes are
01:21:36 going to suffer. Your whole metabolism is going to go down. And we know, by the way, that decreasing
01:21:41 metabolic efficiency at all levels is one of the hallmarks of aging. So in addition to these
01:21:46 enzymes that utilize the NAD and NADH couple, there's a whole series of other enzymes that
01:21:52 actually are digesting, cleaving NAD. And these would be the PARPs, polyADP, ribos, polymerase.
01:21:58 So these are enzymes mostly involved in DNA repair. So it plays a critical role.
01:22:03 Sirtuins, seven sirtuins, all doing different things within the cell. And we can go back and dig
01:22:09 into this a little bit in terms of what are the sirtuins doing. There's also another two enzymes
01:22:14 called CD38 and CD157. These are also NAD hydrolases, and we are studying them a lot.
01:22:22 So that's, I guess, the background of what these enzymes are doing. One thing that your listeners
01:22:29 should know about NAD levels and why the decrease in NAD levels are relevant to aging with respect to
01:22:36 the sirtuins. Because sirtuins have a relatively narrow range of KD for NAD. So if the NAD levels
01:22:45 change, as we know they do during aging, it will lead to a change in the activity of the sirtuins.
01:22:52 And I think this is something that was proposed by Lenny Garante back in the days and showed,
01:22:57 for example, even during fasting, your NAD levels will increase and this will activate sirtuins.
01:23:03 So I think this is something that's really unique to the sirtuins. And we know this in a really acute
01:23:10 way because there are sirtuins that are present in the cytoplasm versus in the mitochondria versus
01:23:15 in the nucleus. And the NAD levels in each of these organs are very different. For example,
01:23:19 much higher in mitochondria. It turns out that SIRT3 has a KD for NAD, which is much higher than SIRT1.
01:23:27 And so variations, that really is the indication that they are sensors of NAD levels, which goes back
01:23:35 to the initial model that you mentioned, NAD levels change during aging. Therefore, we can expect
01:23:40 the activity of the sirtuins to change. Now, what else have I not addressed in your initial batch of
01:23:47 questions? I think we're now ready to then move on to, if we believe with some conviction that
01:23:57 restoring NAD levels in an aging individual is beneficial, we now have to deal with the same
01:24:04 problem you deal with any small molecule or large molecule for that matter. How do you get it in the
01:24:10 body? So what are the ways in which you could get NAD into the body directly or indirectly?
01:24:15 So that brings me to maybe another element of the biochemistry. So one thing that has emerged
01:24:20 is this idea that the question as to why do NAD level decrease? And there's been lots of theories.
01:24:28 Activation of the PARPs, that seems to be happening in C. elegans. In mammals, this is the work of
01:24:35 Eduardo Cini, was the first one to show that CD38 appears to be the major driver of the decrease
01:24:42 of NAD during aging. And the way that he's demonstrated this, and we've actually repeated
01:24:47 some of his results and published on this as well, if you study a mouse that's knocked out for CD38,
01:24:54 you find that NAD levels actually do not decrease during aging. And that's pretty much across all
01:24:59 organs. And I think what this does is really brings the whole question in terms of what should
01:25:04 we be targeting?
01:25:05 Did we talk about what CD38 is doing specifically?
01:25:08 Yeah. So CD38 has a membrane-anchored protein. Some of it is facing outward on the outside of
01:25:16 the cell. Some of it is facing inward. For example, in T-cells, it's mostly facing outward. In macrophage,
01:25:22 it's mostly facing inward. And it is in a dehydrolase. Now, what is it doing in the immune system? Why do we
01:25:29 have it? Not entirely clear. One idea is that because it is present in T-cells...
01:25:34 Is it on non-immune cells?
01:25:37 Endothelial cells as well. One idea, at least for the immune system, is that it might come up and
01:25:44 eat up all the NAD that's local in the extracellular fluid, although there's not much of it, and limit
01:25:50 the abilities as part of the innate immune response and limit the ability of bacteria and other organisms
01:25:55 to actually access these micronutrients for their own growth. That's one thinking. But I think it's
01:26:01 a lot more complicated than this. And we're really in the middle of it. I have a good part of my lab
01:26:06 actually studying the role of CD38 in the immune system and in endothelial cells and in the brain
01:26:11 as well.
01:26:12 Now, you mentioned a moment ago that the CD38 knockouts do not see a decline in NAD with aging.
01:26:18 Zero. And they live longer.
01:26:19 I was just about to say, what is the phenotype of a CD38 knockout? What deficiency do they have?
01:26:23 Nothing that we can tell. And they live longer.
01:26:26 How much longer?
01:26:28 15%.
01:26:28 That's comparable to rapamycin.
01:26:30 Yeah. It's pretty significant. This has been published by Eduardo Cheney.
01:26:34 Do you think that that is true, true, and unrelated to the increased pool of NAD?
01:26:40 Now, CD38, that's a key question. And that's one that's not been answered. And I would say,
01:26:46 if I had to go on a limb, I would say it's not linked to the NAD decrease.
01:26:50 Just to make sure everybody understands what you're saying, your belief is that the CD38
01:26:55 mouse, the knockout, does not live longer because he has more NAD. That's just another
01:27:00 issue we're seeing and that there's something else about that mouse.
01:27:03 Yes. Or it might be partially the NAD and partially the other mechanism I'm about to discuss.
01:27:09 One thing that's remarkable is that as we age, us and mice, we see an increase in CD38 level
01:27:16 across the organism, especially in the immune system. We've published a paper showing that
01:27:20 the SASP from senescent cells is a very powerful inducer of CD38 expression in macrophages. So
01:27:27 that's one mechanism by which we're linking senescence and the SASP to increase CD38,
01:27:34 leading to a depletion of NAD and other effects.
01:27:36 And yet we have no idea what it's doing other than hydrolyzing NAD.
01:27:40 It has a cognate receptor on other cells. They don't seem to be immune deficient. It's really
01:27:46 one of these players that people, there are hundreds of papers.
01:27:49 Do you think it plays a role in inflammation, a negative role in inflammation?
01:27:52 One idea about it is that it plays a suppressive role in the immune response because we see it being
01:27:58 induced latish in the immune response and the idea it comes down to tampen down.
01:28:04 So it's actually the exact opposite. It's so pro-anti-inflammatory that it can be
01:28:09 harmful in that way, as opposed to contributing to sterile inflammation,
01:28:13 which is the more typical problem we see in aging.
01:28:16 Yeah. We did not discuss this. The other side of the immune system is that
01:28:20 it has to be incredibly balanced between reacting appropriately towards exogenous pathogen,
01:28:27 but not reacting against the self. As you know, as a physician, there are so many conditions
01:28:33 that are a manifestation of an excess of immune response against the individual. All of the
01:28:39 autoimmune diseases, which by the way, increase during aging. So-
01:28:43 Except for type 1 diabetes.
01:28:44 Yes. That's the one young.
01:28:46 I know that you used to study that. Why do you think that is?
01:28:49 It's really an interesting question. Although there's something called LADA that I'm sure you've
01:28:54 heard about that is emerging that we might have been diagnosing some type 2 that were actually late
01:29:00 type 1. The thing that is really unique about type 1 is the fact that I remember a number of papers that
01:29:08 highlight the fact that there might be something happening during development that exposes the
01:29:14 immune system to the developing beta cells, and that might trigger more autoimmunity at that time.
01:29:21 It could also be linked to the fact that there's been for many years a discussion of the role of
01:29:25 viruses, infection, and molecular mimicry between some of these viruses and beta cells. So it could
01:29:32 be that a subset of infections that happen during childhood actually puts you at risk of activating
01:29:38 your immune system inappropriately. That's the whole idea. But there's clearly an increase with
01:29:43 autoimmunity throughout life.
01:29:45 By the way, CD157, do we see the same effects? Do we have a CD157 knockout?
01:29:50 No. Much less studied. There is some interesting effect also, but CD38 is garnered.
01:29:55 Most of the attention. If you think about it, I'll go back about CD38 in terms of what it's doing.
01:30:02 It's taking an AD and cleaving it into ADP ribose, which is sugar, and nucleotide, and nicotinamide.
01:30:10 Nicotinamide is a precursor to an AD. And so this nicotinamide, which is generated by CD38,
01:30:16 but by the sirtuins, by the PARPs, normally gets recycled in a two-step reaction all the way back to an AD.
01:30:23 Now, what's really interesting is if you block this, it's called the salvage pathway for nicotinamide.
01:30:30 If you block it, within a few hours, your NAD levels go down to zero. So the system...
01:30:36 Wow. Heavily dependent on recycling.
01:30:38 Incredible, actually. And there's a specific inhibitor of this enzyme called NAMPT. You can add it to cells.
01:30:44 We've done the experiment. Within four to six hours, NAD levels down to zero, the cell dies. So
01:30:50 there's an incredible churning through that whole pathway, which is a reflection of the activity of
01:30:55 sirtuins, CD38, CD157, the PARPs, and so on. So you have a situation during aging where CD38
01:31:03 increases. You increase the degradation. So you decrease the pool of NAD.
01:31:09 But you're increasing, in theory, the metabolites.
01:31:12 You're increasing the metabolites, nicotinamide. Now, one important thing is nicotinamide
01:31:16 metabolism is either salvage, back to NAD, or methylation by an enzyme. And this is important
01:31:24 for supplementation, because it turns out CD38 not only cleaves NAD, but it also cleaves NMN,
01:31:32 which is one of the two precursors, NMN and NR. So when you actually have increased CD38 activity,
01:31:37 and you take NMN, you churn through this pathway, and actually you increase your nicotinamide,
01:31:43 and you increase its methylation.
01:31:45 So NMN is also cleaved by CD38?
01:31:47 Yes.
01:31:47 Into what?
01:31:48 Into what?
01:31:49 Nicotinamide plus something that's not ADP ribose.
01:31:51 Yes, exactly. Yes.
01:31:52 Okay.
01:31:53 And so when you do this, you're increasing your level of nicotinamide to the point that
01:31:58 it's shunting to methyl nicotinamide starts depleting your one carbon cycle. So what you see in a number
01:32:05 of people actually on NMN is their homocysteine level going up, including me. I stopped taking
01:32:11 it when I saw this. I thought this is not, I was taking about a gram of NMN for a while,
01:32:16 and then I saw my homocysteine level going up, I think as a reflection of this pathway,
01:32:22 and basically stopped it.
01:32:24 Now, why is it, Eric, that the increased pool of nicotinamide preferentially goes down
01:32:30 a methylation pathway as opposed to the salvage pathway to give you more NAD?
01:32:34 Yeah. I don't think it goes preferentially. It just depends how much.
01:32:37 It's just the more you put in, even if it splits stoichiometrically or stochastically even,
01:32:43 you're going to take away one carbon.
01:32:45 Yeah. And I do not know what the relative proportion is, but clearly the more you drive
01:32:50 the system with NMN, the more you're going to yield these.
01:32:54 How much did your homocysteine go up, by the way?
01:32:56 Up to 15.
01:32:57 From?
01:32:58 From typical seven.
01:33:01 Wow. That's a big jump.
01:33:03 And then how long did it take to resolve once you stopped the NMN?
01:33:06 I measured typically every three months. After three to six months, it had gone back to normal,
01:33:11 seven to eight.
01:33:12 What about NR, nicotinamide riboside? How is that treated by CD38?
01:33:16 It's not metabolized by itself. NR eventually, in the cell, has to make it back to NMN,
01:33:23 which is on the salvage pathway that we talked about. So NR is less bulky, less big than NMN,
01:33:30 so it is able to get into the cell, but eventually it makes it into nicotinamide NMN and then goes
01:33:36 back into the same pathway. So eventually they all come back to the same.
01:33:40 So do you think that there's no difference between the same amount of NR and NMN?
01:33:44 No, there clearly are some differences, especially in all the really complex biochemistry that happens
01:33:51 in terms of getting them into cells. The problem with NR and NMN is that if you think about what
01:33:56 the approach is, you essentially, you have a pool of NAD, which is a sink. Think about a sink full
01:34:03 of water. It's leaking. That's your CD38. That's the leak at the bottom of the sink. And you keep
01:34:09 pouring more NMN, more NR inside of it. You're just going to accelerate the leak. You're not going to
01:34:16 solve the problem, basically. Maybe you'll reestablish the level at a normal, semi-normal,
01:34:20 but the churning through is problematic. Now, why is the churning through problematic? Because
01:34:25 some of the byproducts of CD38, for example, are cyclic ADP ribose. So there's two forms of ADP
01:34:32 ribose, non-cyclized and cyclized. Cyclized activates calcium signaling. And so there's a whole
01:34:41 aspect of the biology of CD38 that's linked to calcium signaling. So I think I do worry about the
01:34:48 supplementation with the NR and NMN. I do worry about it. I'm not discounting them. I think
01:34:55 clinical trials are ongoing. There's dozens of clinical trials. So we will soon identify
01:35:00 something in which it has a benefit. Again, if you think about the metabolism of these metabolites,
01:35:08 it's incredibly complicated. There are effects on the microbiome. There are effects on different
01:35:13 absorption by different cells, just literature, hundreds and hundreds of papers, I think way
01:35:18 beyond what your audience probably wants to hear. But I would say at this point, most of what you can
01:35:24 buy a supplement have doses that are so low. This is where there's an important discordance also.
01:35:30 When we do experiment, we've seen amazing things in laboratory animals in terms of supplementing with
01:35:36 an RN and NMN. This is where the excitement comes from. But typically, these animals are getting
01:35:41 10 times more than what you're buying as a supplement. And the reason is I think grass
01:35:46 status is given to these companies to give a small amount.
01:35:50 Grass meaning generally regarded as safe, the FDA's criteria for giving something that is naturally
01:35:55 occurring, but doesn't require it to go down the IND pharmacologic pathway. Eric, if you were taking
01:36:02 one gram a day of NMN and your homocysteine went from seven to 15, I guess two questions would be,
01:36:08 has that been reported elsewhere? Is that a known phenomenon in the trials that are testing NMN? Are
01:36:13 they measuring homocysteine to see if that's... I must have read it somewhere because I was...
01:36:18 You were looking for it. I was looking for it, yes. And then the second question is,
01:36:22 how would you then tolerate 10 grams of NMN? I mean, if one gram is doing that,
01:36:26 you would deplete all one carbon. Does that mean you wouldn't even be able to alter your epigenome
01:36:32 in ways that might be favorable? You run all kinds of risk. And a number of people that I've seen the
01:36:37 same thing and start taking trimethylglycine to try to supplement this, I do worry about this.
01:36:42 I think for me, I want to reiterate the fact that I think the data and non-animal models of some of
01:36:48 the things that we've seen with some of these precursors is really interesting. And this is why
01:36:52 there's so much interest. Did you ever try using TMG to see if it would offset the...
01:36:56 No, I didn't.
01:36:57 Okay. That would be an interesting little self-experiment.
01:36:59 Okay. What about intravenous NAD?
01:37:03 So that is one of my pet peeves. I try in everything to remain open-minded to things
01:37:09 that I don't know and don't understand. My prediction is that first, NAD is not
01:37:16 an extracellular molecule. NAD does not exist, almost does not exist at all in your plasma.
01:37:24 It is an intracellular. And as I mentioned, high concentration in mitochondria, much lower
01:37:30 in the cytoplasm in the nucleus. So the whole idea of injecting intravenous NADs, first, it's
01:37:37 too big to be absorbed by cells. So what is the body doing with it? There is a famous paper
01:37:41 by Joshua Rabinovitz that showed that if you inject it actually intravenously, you actually
01:37:47 get it mostly catabolized by the liver into nicotinamide. Nicotinamide is one of the fraction
01:37:54 of niacin. You can buy this at the pharmacy for very cheap. You can go into an IV clinic
01:38:01 and get a $700 injection of NAD. Very few studies, one or two. I've read both of them. They're
01:38:09 interesting. My opinion is that NAD intravenously is not something that should be done. The same
01:38:16 thing for subcutaneously. I've seen another company that sells it subcutaneously. Really
01:38:21 no evidence for doing this. Now, that being said, I've heard, and this is where I try to
01:38:27 remain open-minded. Obviously, we don't know everything. I have heard anecdotal evidence of
01:38:34 dramatic effect in some patients with Parkinson's. People really describing not a miraculous but near
01:38:40 miraculous effect right after the infusion having increase in motor performance that you can really
01:38:46 assess in someone who's a severe Parkinson's. So, not studied systematically.
01:38:52 Mechanistically, is there a reason you could explain that through dopamine or something else?
01:38:56 There's a whole literature on the effect of NAD precursors and so on on Parkinson's,
01:39:02 mostly animal models. And I think there are lots of clinical trials going on in Parkinson's as well,
01:39:08 but more using the standard NR and NMN. The thing that I've described is more as a couple of friends
01:39:14 who've told me, I've seen this. Not enough to make a product, but enough maybe to question. Maybe
01:39:19 there's something more to it than what we truly know. But the proliferation of these intravenous
01:39:25 clinics, frankly, is- How complicated is it to produce a bag of intravenous NAD?
01:39:29 I don't think it's very complicated. I mean, I've never made it. I know making NMN purity took some
01:39:35 effort to scale it up. And for some of the companies that have been doing this right now, there's like
01:39:39 one major supplier out of China that pretty much everybody uses. But in terms of NAD, I don't
01:39:45 think this is an industrial process. I can tell you it's not $700.
01:39:49 Yeah, I'm sure it's not. So, if a person was going to supplement with one orally,
01:39:54 do you think there's a case for NR being superior to NMN?
01:39:58 I would say no. I would say take them both. If you're going to do something and you want to
01:40:03 a bit of an insurance, I did this for a while. I'm not doing it right now. Take 250 milligrams of
01:40:10 each and you'll have a half a gram. You are in a relatively safe dose. Follow your homocysteine.
01:40:16 If you are 60 or above, you could make the case this could be part of a stack. Although this is
01:40:22 the same thing that we see with so many of these supplements right now. Which one do you take?
01:40:27 Which ones are beneficial? There's one little bit of a dark cloud linked to NAD supplementation. It's
01:40:33 the demonstration that the SASP is actually dependent on NAD levels. And so, when you are
01:40:40 actually increasing NAD levels, you might be increasing these pro-inflammatory markers.
01:40:45 Let me make sure I understand why. Because the SASPs, which for folks listening, these are the
01:40:50 soluble products made by the senescent cells that effectively are doing all the bad things that we
01:40:56 don't want to see senescent cells doing. So, now they are dependent on CD38 to some extent.
01:41:02 So, as CD38 goes up, they go up. And are you saying as you give more NR and more NMN?
01:41:10 You might churn it up.
01:41:11 You might be churning up the SASPs.
01:41:13 Yeah.
01:41:14 You finish your point and then I want to make a broader point.
01:41:16 There's also some worry about the fact that supplementing with some of these precursors might
01:41:21 also accelerate tumor growth. So, this would not have an effect in you and I who don't have a
01:41:26 cancer. But if it's someone out there who has an early form of a cancer, this could lead to an
01:41:31 acceleration. This is something that's been shown in animal models that giving some powers to some
01:41:36 people in terms of recommending this to be taken over by everyone.
01:41:40 The folks who make this have strenuously denied that there is any validity to those animal models
01:41:47 that have suggested that. And some of this has been done in vitro as well, correct? I'm not very
01:41:52 familiar with that literature. I remember seeing one study. It was very small. My take on it was,
01:41:59 I guess if you had cancer, this might be a bad idea to take. But I didn't find it that convincing.
01:42:03 I agree with you. It is a general consideration for our whole field of longevity research. It's
01:42:10 better as the enemy of good. It's something that was sort of drilled into me as I went through
01:42:14 medical school. We have a term in French which is sort of like therapeutic, overdoing it,
01:42:21 doing too much. There's such a thing as overdoing it for your patients. In this case,
01:42:26 the whole longevity field is embracing a whole series of these interventions. I mean,
01:42:31 there's not a week that doesn't go by that I don't see a new supplement being touted online and so
01:42:36 on. And I read about all of them. The question is, which ones should you be taking? Which ones are
01:42:42 actually risky? Which ones are not? And to me, this is part of the whole balance of the equilibrium
01:42:47 that I'm trying to reach. There's something that really has a beneficial effect. You want to be on it
01:42:51 as soon as possible. If not, why take the chance?
01:42:55 Yeah. That's the point I was going to make at the outset. You've said it so much better.
01:42:58 Let's pivot a little bit to a couple of the things I want to chat about quickly. Let's talk
01:43:02 about interleukin-11. Big trial last fall that looked at blocking interleukin-11, which is a
01:43:10 molecule that's made by immune cells, plays an important role in inflammation. And this was done
01:43:15 in mice and those mice lived longer. What do you make of the study?
01:43:19 I read the paper like you. I don't have sort of an inside knowledge about, of course, when the paper
01:43:24 came out, it was like interleukin-11. I mean, as an immunologist, you talk about one, interleukin-1,
01:43:30 two, four, six, seven, but 11, never heard about it. It goes up to like 30. So I went and read the
01:43:38 paper. It's an inflammatory marker. So again, it could be-
01:43:41 No, I would say probably not. So it came out of the left field, but it sort of makes sense in the
01:43:50 context of what we know about the inflammatory response linked to aging. And maybe this is
01:43:56 where I can add one point is when we think about the chronic inflammation of aging, sort of
01:44:01 inflammation, it is both cause and effect. We talked about how the immune response helps you to protect
01:44:08 yourself against the innate immune response, against pathogen as a first line of defense.
01:44:13 The innate immune response also has another important role is that it recognizes damage,
01:44:17 any kind of damage. If you cut yourself, if you have a wound inside of your organism-
01:44:24 Coronary artery.
01:44:25 Huh?
01:44:25 A coronary artery.
01:44:26 A coronary artery. This will act any kind of damage, unfolded proteins. There's all kinds of things.
01:44:32 So the innate immune response will be triggered and will activate itself. So as we age,
01:44:38 and as damage slowly accumulates, because aging is a slow, irreversible accumulation of damage,
01:44:44 eventually your immune system responds to this by becoming chronically activated. And so the problem
01:44:52 is that you might think, well, this is great because you're actually repairing all of this damage.
01:44:56 The problem is the activation of the immune response by itself becomes problematic because
01:45:02 these cells, macrophages, for example, are powerful tissue remodeler. The immune system in this
01:45:08 case is Dr. Jekyll and Mystic Hyde. It's helping, but it's facing an unsurmountable amount of damage.
01:45:15 And eventually its activation leads to an AD depletion. That's one of the things that it does,
01:45:19 but many other things, stem cell dysfunction and mitochondrial dysfunction.
01:45:23 So the whole idea here is that 11 might simply be, is one of the key markers of this chronically
01:45:30 activated immune system. So this is not something I imagine you're going to give to a 20-year-old,
01:45:35 but in someone who's getting really in the part where chronic immune activation is present could
01:45:41 really play an important role in the future. And what the paper showed was, again, in mice,
01:45:46 but from what I understand is already an existing molecule. And they actually recently was contacted
01:45:52 by a company that has another novel inhibitors of IL-11. You could imagine this to become part of the
01:45:58 whole armamentarium that we have against aging. And then how do you see playing that off something
01:46:04 on the other side of the spectrum? Because we're really trying to deal with two sides of this system.
01:46:08 We want to tamp down the part that's overactive, and we want to ramp up the part that's underactive.
01:46:14 So we've got basically the only example we have over here is rapamycin. This one does this. And then
01:46:20 we now have IL-11 inhibition, or use knockout mice, but block this. That's a good thing. So
01:46:27 is this one of those things where you need to do both? By the way, maybe you have growth hormone
01:46:30 over here as well, right? Yeah, IL-1 there also. Anti-IL-1 as well, which has been shown to-
01:46:35 Yeah, block IL-1, block IL-11, give growth hormone, give rapamycin. I mean, here's the problem.
01:46:41 You get into this reductionist state, which is like the whole NAD world of NR and NMN. Hey,
01:46:46 it sounds great, but what if there's unintended consequences we can't see? Like, even as much as
01:46:51 I love thinking about this and want to do all of these things, I start to think, man,
01:46:59 what is the probability we're going to get this right?
01:47:02 I agree. The immune system is an incredibly tenuous system, which is in really delicate
01:47:08 balance. So the balance is too much immunity. You might say, well, this is good protection against
01:47:14 cancer, protection against microbes. Right, but then you get autoimmunity.
01:47:20 But then you get autoimmunity. Not enough immunity while you run the risk of being
01:47:24 killed by a pneumonia or some kind of infection. But at least you don't have too much inflammation.
01:47:29 Yes. Yeah. So this is a very fine balance.
01:47:31 This is why I wish we had a dashboard. What are the biomarkers we can use for these things? Because
01:47:36 we don't have this problem with blood pressure. We don't have this problem with thyroid hormone.
01:47:41 We don't have this problem with so many things that we treat because we can measure what we care
01:47:47 about. That's a good point. And the question is, the immune system is so complex, there's not going
01:47:53 to be one single marker. My colleague, David Furman, is this thing called I-Age, which is an immune
01:47:59 aging set of tests that you can actually conduct. That was the first attempt at trying to measure
01:48:04 immune aging. What do they actually measure? Is it all serum biomarkers?
01:48:09 Yes. Serum biomarkers and mostly cytokines.
01:48:12 Validated how?
01:48:13 Been validated in clinical studies.
01:48:15 I-A, immunoaging?
01:48:17 Yeah. Yeah. I-Age.
01:48:19 Oh, I-Age.
01:48:20 Yes. So this was developed and pioneered by David Furman and Mark Davis. So David Furman is with us
01:48:26 at the BAC. Mark Davis is still at Stanford.
01:48:28 I guess this brings us to clocks.
01:48:31 Yes.
01:48:32 I don't even know if I have the energy to talk about this. Okay. Where do you want to begin?
01:48:35 There are so many of these things out there. Some of them are commercially available. Some
01:48:40 of them are just tools of research at the moment. Some of them aim to tell you an actual
01:48:47 age, an actual number that represents your biologic age as opposed to your chronologic
01:48:51 age. Some of them don't aim to tell you that at all. They just want to tell you a rate of
01:48:56 aging. Some of them look only at the epigenetic signature. In other words, they look directly
01:49:02 at the methylation sequence. Others look at a host of markers, including some very simple
01:49:09 serum biomarkers like glucose levels and vitamin D levels and things like that. So how do you
01:49:14 make sense of all of those tools?
01:49:17 Right now, we don't. First statement is they are not ready for prime time in terms of patient
01:49:23 management. There are research tools.
01:49:26 Which is interesting because they're far outside of research labs at this point.
01:49:29 Yes. They are available commercially. I've done the same thing. I don't remember who
01:49:34 told, you told me someone actually, was it Matt?
01:49:37 Matt Caberlin.
01:49:38 Measured his clocks. I do the same thing. Actually, I measure them every three months.
01:49:42 It's just a scatter plot.
01:49:43 It's a scatter plot in a way, you know, I'm between 25 and 68, which is, of course, I like
01:49:48 the clock that show me to be young. But that being said, we know that we're learning. So we
01:49:54 know that, for example, you alluded to the fact that they can vary the same clock. There's
01:49:59 circadian variation, for example, five years. So your age can vary by five years using some
01:50:05 of the clocks, depending on when you measure.
01:50:07 What time of day?
01:50:08 Yeah, what time of day. That's biology. That just tells you the epigenome is something
01:50:12 that's highly dynamic. And so that's something, as we learn, obviously, the companies will
01:50:17 encourage you to measure it, you know, to draw the blood always at the same time. Now, the
01:50:22 whole field right now is pretty much focused, almost completely focused on DNA methylation.
01:50:29 Steve Horvath has done beautiful work. I mean, it's really pioneering work identifying all
01:50:34 this. And Morgan Levine and others have gone on. Dan Belsky, I think, with denadine pace,
01:50:40 which is another epigenetic clock that measures the pace of aging. By the way, I think this is
01:50:45 probably my favorite because it really seems to be responding to interventions. If you change
01:50:50 your diet or if you do something, you will see your pace of aging changing. So I think that one,
01:50:55 to me, seems more promising. We don't know really how to use these tools clinically. That's the
01:51:01 problem. They're nice gadgets to buy. The companies are selling you supplements and then they're selling
01:51:06 you the tests with it. I don't know what to make of it. Personally, I think this is not ready for prime
01:51:12 time. It's something that should be done in the future. It might become in the future.
01:51:17 Would you agree with my stern words on this? Because I've made a lot of enemies by saying
01:51:24 that if as a consumer, you encounter a company that is selling you a test, especially a test that
01:51:34 is not validated in any clinically meaningful way, and then in the same breath, selling you a
01:51:39 supplement to fix the result of that test, you need to run. I agree.
01:51:45 I don't have the patience for that kind of behavior.
01:51:49 Someone told me actually recently that one of these tests actually that you can measure,
01:51:55 almost everyone who gets their result is low. And of course-
01:51:59 Low being good or bad in this test?
01:52:01 It's bad.
01:52:01 Bad, yeah.
01:52:02 It's insufficient. The next step is the recommendation. You have to buy this
01:52:06 supplement to solve the problem. So yes, again, it's the same thing with the sirtuins and the NAD.
01:52:11 Let's not throw out the baby with the bathwater. There is a whole series of these players. I'm not
01:52:17 disputing their honesty or their good intention. From what I've seen, I think it's too early.
01:52:23 And what do you think is the biggest problem? Is the biggest problem the biologic noise in the
01:52:29 system, which means even if you had the absolute perfect tool to measure and you knew exactly what
01:52:36 to measure the movement of that thing is so great that the probability that you're capturing a
01:52:42 meaningful value is irrelevant. In other words, imagine that there's a variable that moves like
01:52:47 this, but on the small level, it's moving like this. I'll give you an example. Imagine you were
01:52:53 measuring heart rate, but you could only sample it milliseconds at a time. And what you were actually
01:52:59 measuring was heart rate variability instead of heart rate. It would be useless. It's too noisy.
01:53:04 I agree. Do you think that's the problem? No, I don't think that's a problem. I'll speak
01:53:07 personal experience. I work with True Diagnostic. They use the Epic Array and you get not one clock,
01:53:13 you get dozens. So I get all of them. And they tend to be reproducible every three months unless I make
01:53:21 some interventions. But in general, there is some consistency. I'm not the only one who has seen
01:53:26 this. So my advice, if you really are determined to use them, use all of them. They all are different
01:53:33 mirrors of your reality. The problem of the methylation clocks is that there's a very
01:53:39 tenuous link between the change of methylation at any given site and the biology. Typically,
01:53:46 each clock would rely on about 500 different methylation sites, but they're not attached to
01:53:51 a specific gene. So you don't really know what it means. But how are they even doing that? They're not
01:53:56 measuring with point arrays. They do this with arrays.
01:53:59 They're doing this with an array?
01:54:00 Yeah, they have about 20 million methylation sites that they're assessed. But each clock
01:54:04 uses a subset, four or five hundred.
01:54:07 Sorry, just to be clear, you're saying they're actually measuring point of methylation?
01:54:12 Yes. Okay.
01:54:13 Yes. They're quantifying the level of methylation at each of these sites. The problem with the
01:54:17 clocks is also where do you obtain them from? Typically blood, as I mentioned, it's a heterogeneous
01:54:24 compartment. As you age, for example, you know that your fraction of naive T cells decreases
01:54:30 down to close to zero. If you're 80 years old, your memory T cells increase. So we did
01:54:35 a very simple experiment. We sorted all of these different T cell subsets, memory, naive, central
01:54:42 memory, TEM rather terminally differentiated, and measured their epigenetic age using several
01:54:49 of the clocks. 20 to 25-year difference between the naive and the central memory T cell.
01:54:57 In the right direction, the direction you would predict.
01:55:00 Yes, the naive and much longer.
01:55:01 Yeah. That's somewhat interesting.
01:55:03 That was really interesting for me because it means also any conditions where you see a
01:55:07 shift in the relative proportion of these cells. For example, you get an acute COVID infection.
01:55:13 What happens? You have a massive expansion of your memory T cells. So it looks like you're
01:55:18 going to, and then you sample. And given that these cells look much older than the other
01:55:23 ones, you're going to look like you're aging. And there's a whole literature that talks about
01:55:28 accelerated aging and rheumatoid arthritis and COVID and HIV, all of these conditions that
01:55:35 are all associated with chronic immune activation. So that's another confounding variable. So what
01:55:41 we did to do this with a student in a lab, we made a new clock in which we eliminated all of these
01:55:48 methylation sites that are linked to differentiation. Okay, so now this clock that we've done does not
01:55:55 vary, actually, as a function of the types of cells that are in the blood. As a T cell goes from being a
01:56:03 naive T cell to being a memory T cell to being a TEMRA, the methylation patterns change. That's part of the
01:56:08 epigenetic regulation. So we eliminated all of those sites, made a new clock called entrant clock,
01:56:14 which actually is impervious to your level of immune activation. And what's interesting is that
01:56:18 that clock doesn't change anymore during COVID. It doesn't change very little during HIV. It doesn't
01:56:25 change during a whole series of conditions where people have talked about aging acceleration,
01:56:30 including the story that we talked about earlier on growth hormone.
01:56:34 What does change it then?
01:56:36 What does change it? Cancer, senescence, which is really interesting.
01:56:42 What about short-term interventions that might be beneficial? So if you took an individual who
01:56:47 is insulin resistant, and you put them on a GLP-1 agonist, and three months later,
01:56:53 they're 20 pounds lighter and their insulin resistance has resolved, how does that change on the clock?
01:56:58 I would not be able to tell you specifically for individual clocks, but Dan Belsky's Dennett & Pace
01:57:05 clock is the one that repeatedly people have shown seems to be responding to interventions, which is
01:57:12 the two qualities that you want in a clock is one to be predictive, and the other one to be
01:57:18 predictive of ultimate income, sort of life expectancy or the occurrence of disease. But also,
01:57:23 you want it to be modulatable, responsive. And reproducible.
01:57:28 And reproducible, yes. Reproducible, I think, is more a question of the laboratory that's doing
01:57:34 it. So there are no- But also potentially the biologic noise still.
01:57:37 Exactly. So biologic noise and laboratory conditions speak to reproducibility. I agree with
01:57:42 what you said. I mean, I've often made this case when people ask me about clocks is my gripe with the
01:57:47 age clock. So again, the pace clock is different because it's just trying to give you a rate of aging,
01:57:51 and I agree with you. I think there might be more there. But these clocks that spit out,
01:57:55 hey, Eric, congratulations, you're 25. I say to someone who says, isn't that wonderful? I say,
01:58:02 maybe, but do you actually believe that you're 68? You're 68 and your clock said you're 25.
01:58:10 Should I expect you to live another 55 years? Yes.
01:58:14 In other words, is it a better predictor of future life than chronologic age? And the answer
01:58:21 is, to my knowledge, no. There is no clock that has a better ability to predict lifespan
01:58:27 than chronologic age does. And until that's the case, I worry that the biologic clocks are creating
01:58:35 a bit of a distraction, at least this subset of clocks, and that we maybe ought to focus better on
01:58:42 clocks where the readout state is more about, is this intervention good or bad? Or is this a net
01:58:50 positive intervention or a net negative intervention? I agree. As we mentioned earlier,
01:58:56 the field initially focused on the epigenetic clocks because this is Steve Horvath's pioneering
01:59:01 work, so it got everybody to start thinking we can generate these tools. But the field is now moving
01:59:06 into proteomics clock. So what makes up Dan's clock? Dan is a methylation. It's also methylation.
01:59:13 Why do you think it's doing a better job than maybe Horvath's clock at the moment? Typically,
01:59:18 it really depends on what the variable, what the cohort, what the question was. I don't know. I mean,
01:59:25 I think that Dan's is the only one that's doing it in this way. Why is it working better? They're just
01:59:30 looking at it in a completely different way. How much is AI facilitating this at this point?
01:59:35 Machine learning is the key instrument. Essentially, what these clocks are is a regression analysis on
01:59:40 to start with a variable, which is your age, and you regress each methylation site onto the age. You
01:59:47 do this on enough people of different ages, you find an average. I wonder if that's the wrong way to
01:59:53 do it. Wouldn't it be better to get biobank data and instead of mapping it onto age, map it onto number
02:00:01 of years remaining in life? Because if you'll know that in a biobank.
02:00:05 They've done this. They've done this. They've done this in terms of life expectancy. They've done
02:00:09 this in terms of morbidity. So this is like the third and fourth generation of these clocks now
02:00:15 are looking at regression. The initial one- Was just chronologic.
02:00:19 Yeah. And Steve used to go around saying my correlation coefficient is 99. And I was like,
02:00:24 well- That's because that's what you built it on.
02:00:26 Yeah, exactly. And I can look at a calendar. I don't need an epigenetic clock to tell me how
02:00:30 old I am. But the next generation clocks actually had a bigger spread. Of course,
02:00:35 you have an average- That's right. Because they started to build it on a different variable.
02:00:39 Exactly. So what really excites me right now is the whole idea that the field is moving on to the
02:00:44 next stage, which is non-epigenetic clocks. Because I'm still frustrated as a biologist trying to
02:00:50 understand what are these clocks- It should be everything. It doesn't make any sense to me that we
02:00:54 wouldn't look at the metabolome, the proteome, and the epigenome. There's no excuse today with the
02:01:00 compute power not to do that.
02:01:01 We have clocks based on fundus. We have clocks based on skin. We have clocks based on facial
02:01:08 recognition. So the clocks are going to be measured using dozens of different biological variables.
02:01:15 Any biome, a small company in the Bay Area is using the tongue, a tongue picture, the old doctor
02:01:21 looking at your tongue. So you can actually use machine learning to recognize patterns of
02:01:25 discoloration. And another exciting really story was, I don't know if you're familiar with a Tony
02:01:31 Wiscore's paper using proteomics. He has shown, for example, proteome in plasma changes throughout life,
02:01:38 pretty dramatic manner, which is really completely mind-boggling for me just to see that you can be
02:01:43 so different as you age in terms of your whole blood proteome.
02:01:48 Why? If the epigenome is changing, then gene expression is changing. If gene expression is
02:01:53 changing- No, no, you would.
02:01:55 Yeah, yeah, yeah. Okay.
02:01:56 But that it would change to such a degree. Tony has a beautiful slide which shows all of the proteome
02:02:02 in the blood and how the colors change across age.
02:02:06 And do you think that most of those changes are post-translational?
02:02:08 No, most of them are probably expression level.
02:02:11 It's expression.
02:02:12 Yeah, it's expression. People are building transcriptomics clock. And so Tony now has a
02:02:17 study that is, I believe, in press or coming out soon where they've gone back using this proteomics
02:02:24 clock and they've done this on a UK biobank, more than 40,000 different people. And this is the study
02:02:30 we started this discussion on identifying. What Tony did was actually remarkable. He looked at each of
02:02:37 these proteins that are in the blood and selected some that were predictive to be coming from unique
02:02:44 organs. Imagine what you know about how you measure a tropomyosin for heart attack. So they did this.
02:02:51 They went and looked in every single organ and said, okay, what proteins are specific of this organ
02:02:56 and which ones actually can be measured into the plasma? And using this, they were able to generate
02:03:02 what they call an organ-specific clock. Simply from a blood draw, they're able to really determine,
02:03:08 do you have a frailty point? When I look at you, is there like suffering happening?
02:03:13 And this is Tony's work through the proteome?
02:03:15 Tony Wiscore. It's a new startup called Vero. For disclosure, I've joined the board of this
02:03:20 company, but I only joined the board because I was really excited about what they're trying to do.
02:03:25 And I think it really brings a whole new dimension to these predictive biomarker, which is more aligned
02:03:32 to what you and I have seen as physicians. Kind of simple-minded. As the protein is released into
02:03:38 the blood, it shouldn't be there. It might be indicating some suffering. And I discussed with
02:03:42 some colleagues who have used this clocks and have identified some abnormal aging in a unique organ,
02:03:50 only to go back and find that there was indeed one problem without going into what the issues were.
02:03:57 The reason I tend to be a slow adopter of these things is, even if that's the case,
02:04:03 the question is how much noise is in the system. I go and do that test on a patient and it comes back
02:04:09 and says, oh my God, there's something wrong with your liver. Your kidney's a bit too old. You're this,
02:04:13 you're that, you're this, you're that. So I have two fundamental questions. The first is,
02:04:17 could I have figured that out another way? So if it's telling me your liver is angry,
02:04:20 or something's wrong with your liver, how do your transaminases look? If it's telling me
02:04:25 something's wrong with your kidney, could I have picked that up on a urinary analysis,
02:04:29 looking at creatinine clearance or cystatin C or something else? In other words,
02:04:34 is it giving me information that I can get elsewhere in a more reproducible, more validated
02:04:38 fashion? The second thing is, let's say it tells me seven things are not perfect. And by the way,
02:04:45 everything looks perfect. I have my standard assays. Everything looks awesome. This test says,
02:04:50 oh my God, these six or seven things are problematic. And I go poking around,
02:04:54 poking around, poking around, and I find out one of them is indeed not working,
02:04:58 but the other six were perfectly fine. So now we have this huge false positive situation.
02:05:03 That's a whole MRI.
02:05:05 Yeah, exactly. It's the same problem we have with cancer screening, which is buyer needs to beware
02:05:09 of the Pandora's box you open. And at least with MRI, you're dealing with imaging,
02:05:14 but this sounds like exciting. And yet it's a bit of a black box.
02:05:18 Agreed. 1.00
02:05:19 Where it's going to spit out, oh my God, there's something wrong with your left testicle. And what 1.00
02:05:24 do I need to do? You know? And...
02:05:25 That being said, I think the assays are generated in a way that there are multiple. It's not like
02:05:30 one single protein, like a tropomycin. We know that's a clear indicator. There's something cell
02:05:35 death in terms of your heart. In this case, the clocks are generated in a way that there are
02:05:40 multiple sentinels for each organ, many. The story I was talking about early days, okay? We totally
02:05:46 agree with this. It's a startup. I think they will deploy it. And obviously, it's going to take,
02:05:51 again, a group of physicians who are able to look at these tests. This is what research is. This is
02:05:56 what startup. I can't blame them for trying because I think it has a potential, for example,
02:06:01 to highlight a frailty point, which is, in aging research, to me, is really critical. You could have
02:06:07 the best mind and the best heart in the world if something else is going to fail that you are
02:06:12 completely unaware of. You want to know as soon as you can. My prediction is, I'll share the paper
02:06:18 with you if you're interested in looking. It's quite exciting in terms of where this is leading.
02:06:23 But I agree with you early days. Yeah. I will probably maintain a shockingly high degree of
02:06:28 skepticism and probably enjoy some experimentation with it. But again, my experience in the real world
02:06:35 is that that's just not how it works. There aren't people walking around that are insanely,
02:06:40 remarkably healthy, where everything looks amazing, but they have some time bomb they don't
02:06:45 know about. With the exception of a few things, I'm not sure it would pick up. For example,
02:06:50 cancer is always that thing. And of course, there's an entire field of medicine that's going around
02:06:55 with liquid biopsies that's exactly trying to solve that problem. You could reword the liquid
02:06:59 biopsy industry through the lens you said, which is, it's looking for that weakest link,
02:07:03 which in this case is the earliest signs of cancer. And it could be that a cancer will manifest itself
02:07:08 also in local organ suffering. And again, leaching, it might actually point along with a liquid biopsy
02:07:15 that tells you you have some cancer cell. It might tell you, it might point you to one place where
02:07:19 actually this is actually happening. Yeah. It's interesting. The case that I've made about MRI is
02:07:25 the same. I have a whole bunch of physician friends. I get a yearly MRI and they tell me, why do you do
02:07:31 this? I say, well, because I would rather know. He said, well, you're going to find all kinds of
02:07:35 things. I said, we did find something. I had a tumor behind my jaw and a mass. It was not a tumor,
02:07:42 but it took me six months of worrying about what it was and decided not to biopsy and anything.
02:07:48 My sense of all of this is that these are novel ways to practice medicine.
02:07:53 I'm criticized heavily for being too much on the forefront of doing that, but probably not nearly
02:07:59 as far as some. At the end of the day, I think about every time you do a test, one, you never do
02:08:05 a test unless you're willing to act on an outcome or you have a sense of how an outcome will change
02:08:10 your behavior. We don't order tests for the sake of information. We order tests to make decisions.
02:08:16 Therefore, you must at a minimum understand the full suite of outcomes that can come from the test
02:08:24 and how many of them will pose huge trouble for you. 20% of my patients opt not to do whole body MRI.
02:08:32 Yes.
02:08:33 And I fully endorse that decision. And I try to talk patients out of it. I really try to
02:08:37 highlight how many times we find thyroid nodules that we have to put needles into that ultimately
02:08:43 end up being nothing. And all we do is subject them to that risk and the anxiety that comes along
02:08:49 with it. So I'm eager to look at this because I do think that the proteome offers a lot, but I'm
02:08:54 always worried about going a little too far on the clinical implication of a test.
02:08:58 I'm with you in terms of being careful. I view this as another attempt. For example,
02:09:03 we talked about the data showing that the two organs that appear to be rate limiting in terms of aging,
02:09:10 the immune system and the brain. That came out of that story. That's actually the title of the
02:09:14 paper, essentially, that identifies the brain and the immune system. So they have a whole series of
02:09:19 immune markers that are predictive of some degree of immune activation and so on.
02:09:26 Well, Eric, there's a lot of other things I wanted to chat about, but I think what we'll do
02:09:29 is we'll have you come back out to Austin for another day of driving a COTA, and then we'll justify
02:09:35 it by doing another podcast where we dive deeper into some of these topics. We really take advantage
02:09:40 of the fact that we have the best race course in the country here in our backyard. So I think
02:09:44 you're going to have fun tomorrow and you'll be like, let's come right back and do it again every
02:09:48 month. I'm looking forward to this. Thank you, Eric. Thank you.
02:09:51 Thank you for listening to this week's episode of The Drive. Head over to peteratiamd.com forward slash
02:09:59 show notes. If you want to dig deeper into this episode, you can also find me on YouTube, Instagram,
02:10:05 and Twitter, all with the handle peteratiamd. You can also leave us review on Apple podcasts or
02:10:12 whatever podcast player you use. This podcast is for general informational purposes only and does not
02:10:18 constitute the practice of medicine, nursing, or other professional healthcare services, including
02:10:22 the giving of medical advice. No doctor patient relationship is formed. The use of this information
02:10:29 and the materials linked to this podcast is at the user's own risk. The content on this podcast is not
02:10:35 intended to be a substitute for professional medical advice, diagnosis, or treatment. Users should
02:10:40 not disregard or delay in obtaining medical advice from any medical condition they have, and they should
02:10:45 seek the assistance of their healthcare professionals for any such conditions. Finally, I take all conflicts
02:10:52 of interest very seriously. For all of my disclosures and the companies I invest in or advise, please
02:10:57 visit peteratiamd.com forward slash about where I keep an up-to-date and active list of all disclosures.