The Peter Attia Drive - #35 - Nir Barzilai, M.D.： How to tame aging

00:00:00.000 Hey, everyone. Welcome to the Peter Atiyah Drive. I'm your host, Peter Atiyah.

00:00:10.160 The drive is a result of my hunger for optimizing performance, health, longevity, critical thinking,

00:00:15.600 along with a few other obsessions along the way. I've spent the last several years working with

00:00:19.840 some of the most successful top performing individuals in the world. And this podcast

00:00:23.620 is my attempt to synthesize what I've learned along the way to help you live a higher quality,

00:00:28.360 more fulfilling life. If you enjoy this podcast, you can find more information on today's episode

00:00:33.020 and other topics at peteratiyahmd.com.

00:00:41.400 Hey, everybody. Welcome to this week's episode of The Drive and Happy New Year to everyone. We took a

00:00:45.920 week off, as you probably noticed. So hopefully everybody's ready to jump back into this fun

00:00:50.680 stuff. My guest this week is Nir Barzilai. And if you're at all into the space of longevity,

00:00:57.440 he'll be no stranger to you. Nir is the founding director of the Institute for Aging Research,

00:01:03.740 the Nathan Schock Center of Excellence in Basic Biology at Albert Einstein. He's completed two

00:01:09.440 fellowships, one in metabolism at Yale, the other in endocrinology and molecular biology at Cornell.

00:01:16.420 He directs the Longevity Genes Project. And in my estimate, Nir is probably the most knowledgeable

00:01:22.600 person ever on the genetics of longevity. And we talk a ton about that during this episode.

00:01:28.900 He also is leading the effort to test metformin in a prospective clinical trial for non-diabetics

00:01:36.800 with respect to aging. And this is referred to as the TAME trial. We get into that obviously in detail

00:01:43.100 here and also will use a lot of the data for that in the show notes. This is in many ways,

00:01:49.900 I think Nir is probably one of the most insightful people when it comes to understanding the clinical

00:01:53.580 benefits of metformin. And we talk about that a ton. We also talk about insulin resistance. I can't

00:01:58.920 resist the urge, no pun intended, I guess, to get into sort of a detailed discussion on what IR

00:02:04.100 is with sort of people who are really deep in this space. We talk a lot about IGF and growth hormone.

00:02:09.760 And I got to tell you, this is a topic that you've probably heard me waffle on a little bit

00:02:14.520 because I'm still really kind of on the fence about this relationship of IGF, GH. I think the

00:02:21.480 centenarian data point in one direction. I think the epidemiology outside of centenarians point in a

00:02:26.980 different direction. And of course, none of this really speaks to the question I get asked constantly,

00:02:31.580 which is, do you think administration of growth hormone is beneficial from a lifespan or health

00:02:36.360 span perspective, or do you think it's harmful? And truthfully, I've always leaned towards the harmful

00:02:41.820 side, but we get into this in detail and near offers some great insights. Certainly for me,

00:02:46.980 I was, I was really helped by this process. We do go back and talk about the centenarians.

00:02:51.440 And because I'm in the midst, as some of you know, of writing this book, I'm knee deep on all that

00:02:55.860 literature. And so it was really great to kind of clarify a few things that I think even if you're

00:03:00.840 not steeped in this stuff, what you'll find very interesting. And of course we talk about all of my

00:03:05.140 other favorite topics like autophagy, caloric restriction. We even get into a little bit of the stuff

00:03:10.980 around NAD and nicotinamide riboside and things like that. So overall, I think this is a bit of a

00:03:16.620 technical episode, but not that technical. We've certainly done more technical stuff. The show

00:03:21.660 notes will be valuable as always, especially for a show of this nature. So with nothing else to add,

00:03:27.720 please welcome to the show near bars alive.

00:03:33.220 Hey Nir, how are you?

00:03:34.440 I'm terrific. How are you?

00:03:35.820 I'm good. I can't believe not only did you make it down here on time, but you made it down here

00:03:40.400 ahead of me at my own place.

00:03:42.560 Nice to stay young.

00:03:45.380 Yeah, this is, I don't know. These days, this weather is so unpleasant, but yeah, you beat

00:03:50.540 me here. I was a bit embarrassed.

00:03:52.380 It's unpleasant for you because you're in California most of the time.

00:03:56.040 Yeah, yeah. I think that's part of it. But you're from Israel, so this has to be unpleasant

00:04:00.280 for you too.

00:04:00.940 Israel was too warm for me. I'm okay.

00:04:03.420 Got it. Well, speaking of which, so you were born in Israel. You spent how long there?

00:04:08.980 Did you serve military time there?

00:04:11.040 Yes, I served military time. I was a nurse and a medical student. I went to the Technion Medical

00:04:18.060 School. I went then to Hadassah Hospital for residency, which I finished. And only then I came

00:04:26.540 to the United States. I was at Yale with Ralph DeFronzo doing metabolism. Actually, I was looking

00:04:35.160 at the mechanism of action of metformin in 1987 before it was approved for use in the United

00:04:42.260 States. There's a serendipitous connection to that later. And then I went to Cornell for an

00:04:48.520 endocrine fellowship, and then I was recruited to Einstein. And so my first part of my life was

00:04:54.720 metabolism. But then I started doing what I really was interested in, and this is aging and the

00:05:01.120 biology of aging.

00:05:02.140 And I remember reading a quote once from you that said, maybe paraphrasing, but metformin is the

00:05:08.280 reason I came to the United States.

00:05:10.740 Well, in a way it was, but this is life in retrospect, right? I wasn't really expecting. I was done with

00:05:17.680 metformin in 1988. I was done with metformin until it started again about three, four years ago.

00:05:25.240 Well, I can't wait to talk about metformin because they're probably, I'm trying to think,

00:05:30.620 if I think of four or five exogenous molecules, which is just the terminology I use to describe

00:05:36.160 anything that you ingest or take that comes from outside the body. So I would include drugs,

00:05:41.300 supplements, hormones, anything in there. But when you lump all of these things together,

00:05:45.320 if I were to say, what are the three or four of these that I am asked about the most frequently

00:05:50.060 from patients or frankly, anybody, if I'm at a party and I let it slide, what I do for a living,

00:05:56.420 the first question is, should I be taking metformin? And there's usually a handful of

00:06:00.920 others that they want to know about. Should I be taking nicotinamide riboside? Although they

00:06:05.000 usually don't say that. They usually say, should I be taking NAD or something to that effect? So

00:06:08.620 it's wonderful that we will be able to speak today because few people can speak about metformin

00:06:13.800 the way you can. And so I'm really looking forward to that. But before we go down that path,

00:06:17.920 I do still want to kind of understand a little bit more about your journey and your interest

00:06:23.060 in endocrinology. Did you know from day one when you entered the field of medicine that this was

00:06:28.520 the area that you wanted to study? Yes. I was interested in aging from the time I was pretty

00:06:35.860 much 13 and spent weekends with my grandfather who was telling me his life story. And his life story

00:06:44.980 wasn't easy. And he did lots of physical things and he dried the swamp and he did this and that.

00:06:52.620 And I'm looking at the man who was then 68 years old that walks slowly, he's obese, white hair,

00:07:01.960 and he just didn't look like somebody who did everything he told me to do. And you know,

00:07:08.940 they say that children have imagination, but most kids don't see their grandparents as what they will

00:07:17.060 be, right? They see them as, I don't know how they got there. And this really stuck with me. And I

00:07:24.560 started to be interested in the biology of aging. When I was, for example, when I did my residency,

00:07:30.500 I always was interested in how, not what's the age of the patient, but does he look older

00:07:38.000 or younger than his age? I kind of realized intuitively that there's a chronological age

00:07:45.200 and a biological age. And what is this differences between the biological and chronological age?

00:07:51.440 And of course, as a physician, it looked like endocrine is a good place to start because you knew that

00:07:58.720 there's a lot of endocrinology in aging and you assumed that hormones are going down or some are

00:08:04.600 going up. But if you could fix that, maybe you could fix a lot of aging. Now, everything that

00:08:11.560 happened in my life was fascinating from the biology of aging. Replacing the hormones wasn't really a part

00:08:18.480 of it, but we'll get to it. Going back to the hormone thread, the most obvious example of changes

00:08:23.280 in hormones with aging, of course, occur in women where, you know, they have this very abrupt

00:08:28.560 change in one of their endocrine systems, this androgen system. Did you think about it even

00:08:34.220 more broadly than that? For example, like what was happening in thyroid hormone and what was

00:08:38.560 happening in fuel partitioning hormones and other things like that? Oh, absolutely. Absolutely. But

00:08:43.740 I'll tell you to the point when there was a discussion whether to do the Women Health Initiative.

00:08:49.760 You know, many people thought this is a waste of time. You know, we know that estrogen is good for

00:08:54.920 women. There's lots of studies like that. Of course, it's major in aging. And I didn't look at

00:09:01.040 this simply for maybe two reasons. One is men are also aging and in several ways similar to women.

00:09:11.160 And that's not an estrogen story. And it wasn't, the testosterone decrease wasn't really as dramatic

00:09:18.320 as the estrogen decrease. So I didn't, I thought that there's a lot of aging that's done without

00:09:23.780 estrogen. Let's start with that. And also you're replacing just one hormone with lots of hormones

00:09:29.300 are changing. It didn't look like a good study to me. Well, and especially at the time, and I try to

00:09:35.240 temper my criticism for that study by trying to have some empathy for the fact that the investigators were

00:09:40.800 dealing with what the treatment protocols were at the time, but to use oral estrogen that's,

00:09:48.160 you know, conjugated equine to use synthetic progestin and not even actual progesterone.

00:09:53.380 And there, you could come up with a list of 18 things that in retrospect are so obvious why that

00:09:58.660 study was a failure. Absolutely true. But the other part of that is that the way to show effect of

00:10:06.320 estrogen in animal models was to take their ovaries out and give them estrogen. And then you could show,

00:10:14.080 you could do lots of provocation. It was good. So estrogen is a good hormone on a young body,

00:10:19.980 but then there were studies that were hardly published, but I was aware of them of taking

00:10:25.860 old animal, you know, so you could take the young animals in just one center in Houston. They did it.

00:10:32.500 You took the young animal, you took their ovaries out, you gave them estrogen and you employed a stroke

00:10:38.480 model. Okay. And when you gave estrogen, the stroke was smaller and everything. When you did the same

00:10:45.180 in older animals, the stroke became worse. Okay. And they couldn't publish it for a while because

00:10:52.120 they said, well, it's old animals. So they have other diseases or other things, but every experiment

00:10:57.940 that had estrogen in old model wasn't, was the opposite. Not only that it didn't affect, it was the

00:11:04.800 opposite, which what happened, you know, to the WHI in a certain ways. Yeah. Although I still think

00:11:11.780 that the, it was probably the breast cancer that garnered the most headline in the WHI. And actually

00:11:17.380 that's an interesting topic, which I'll be going into in great depth on another episode. There's a

00:11:22.500 great book that just came out that tackles this, that goes through the history of the WHI. At what

00:11:27.500 point, well, let's go back to the Metformin thing actually. So it's 1987 and you're thinking like,

00:11:33.420 okay, Metformin is the next line of agent we will use to treat type 2 diabetes.

00:11:38.320 So Metformin was already in clinical practice in Europe. Is that correct?

00:11:42.500 Oh, correct. For many years.

00:11:44.300 How many years? Like 20 or 30 years?

00:11:45.800 So Metformin a few years ago was 60 years. So we're talking about, you know, 1950 or something,

00:11:53.940 1950, 1960. When I was in Israel, I prescribed Metformin. You know, that was the first line

00:12:00.320 for type 2 diabetes. What took so long for it to come to the U.S.?

00:12:04.160 That's a terrible story that still continues. You know, the FDA said, we don't know. We don't

00:12:12.700 know that Metformin is going to be effective in the population in the United States. So you have to do

00:12:17.500 the studies. And it was Bristol-Myers really that got to do the studies. And basically they had to do a

00:12:25.480 phase 3 study again to show that Metformin is effective as sulfonylurea as an indication for

00:12:35.040 type 2 diabetes. And if even then, the idea was that it will be good only in obese people,

00:12:43.820 which happened not to matter much because all the type 2 diabetes in the United States were obese,

00:12:51.100 but it was really a matter of regulation. And part of having regulation was that the FDA asked more

00:13:00.820 studies to understand the mechanism of aging and Metformin. And in a certain way, maybe they are

00:13:07.600 right because Metformin mechanism of action is really complicated. And I'm not sure that today

00:13:15.460 as a new drug, it would be approved. Interesting. Where do you think it would fail? In safety or

00:13:22.020 efficacy? No, it's not in mechanism. In the inability to elucidate the mechanism.

00:13:27.800 Right. When you come to the FDA and you say, you know, we've done that and we've done this in animal

00:13:31.740 and we've done this in cells, and you're trying to show what is the major mechanism of action,

00:13:38.100 you would have failed. So what I showed in humans is that Metformin targets hepatic glucose production

00:13:47.560 or the insulin sensitivity of the liver. And that's the major mechanism of action. Unlike,

00:13:53.680 for example, sulfonylurea that increased insulin secretion, right? Or TZDs that increase the insulin

00:13:58.740 sensitivity in the muscle more than in the liver. So this was a mechanism that you could pack

00:14:05.360 to the FDA, but it's really not the intracellular mechanism of Metformin.

00:14:11.380 So at the time, was it understood what Metformin's activity was on complex one of the mitochondria?

00:14:18.940 No, not really. This came later. I don't know what year, but it came later. It came when the

00:14:28.340 seahorse's assays were developed to show really which mitochondria in which pathway of mitochondria.

00:14:35.360 mitochondria, you have changes. Maybe for the listener, explain what a seahorse assay is.

00:14:40.960 Seahorse assay is an assay to look at mitochondrial action, whether you extract mitochondria or even

00:14:48.720 tissues. It really shows some relationship between oxygen consumption in relationship to polarity.

00:14:58.400 And it's very sensitive to look at some of the effects of drugs that are interfering with

00:15:06.720 mitochondria action or decreased mitochondrial activity.

00:15:10.080 It's almost like the indirect calorimetry of the mitochondria.

00:15:12.360 Right, right. Indirect because it's a provocative test, really. So yes.

00:15:17.240 Metformin was discovered from a plant as well, correct?

00:15:21.060 Right. It's a French lily.

00:15:22.640 And it was a French lily, obviously. And this was discovered in the 40s or 50s?

00:15:26.560 Correct. The first thing that happened, there was a drug by the name of metformin,

00:15:33.320 a cousin of metformin.

00:15:35.000 Which apparently is much more potent.

00:15:36.780 It's much more potent because unlike metformin, it doesn't need a transporter to get into the cells.

00:15:44.420 On the other hand, it was associated with a lot of lactic acidosis.

00:15:50.860 And it was considered unsafe eventually.

00:15:55.340 And then metformin was a safer part with much less of a lactic acidosis side effects.

00:16:02.920 What is the mechanism by which metformin caused lactic acidosis?

00:16:06.180 Because in all the use I've seen of metformin clinically, I've never seen a case of it.

00:16:11.420 Which is not to say it doesn't happen. I'm sure it does. And you could stack risks by taking someone

00:16:15.700 with renal insufficiency, giving them contrast and tons of metformin. But in medical school,

00:16:21.900 this was like the board question you got asked every single test. What do you have to worry about

00:16:27.460 with lactic acidosis? What's the putative mechanism by which that happens?

00:16:31.060 So, first of all, I would tell you that in my study in the 80s, every patient that we gave

00:16:38.020 metformin had an increase in lactic acid.

00:16:41.140 From what to what in millimolar?

00:16:43.080 Within the normal range. So, if it's two was the...

00:16:46.880 Yeah, the cutoff.

00:16:47.620 The cutoff. So, it was, you know, between... went from 1.5 towards the two. And sometimes

00:16:55.140 even went over the two. But it wasn't really anything associated with acidosis.

00:17:01.140 And there was no change in pH?

00:17:02.640 No. No. No onion gap. Not anything like that. And it probably has to do... I don't know that

00:17:09.260 I can tell you for sure, but it has to do with what happens when complex one, part of the metabolic

00:17:17.060 effect when complex one is inhibited. But there's lots of speculation and I don't really care to

00:17:25.100 comment on that. But I think what became clear, there is a whole... we call it endocrinologists,

00:17:32.060 call it MALA. It's metformin-associated lactic acidosis. In other words, we moved away

00:17:38.200 from metformin-causing lactic acidosis to the association because it's, if anywhere, it described

00:17:45.920 more in people that have kidney failures or have heart attacks or something and had lactic acidosis

00:17:54.080 and were on metformin too. And that was kind of the association. But it's not clear to me that

00:18:00.780 there are truly people who develop lactic acidosis from metformin that is just because of metformin

00:18:08.180 and not associated with something else. I'm glad to hear you say that. That's sort of generally

00:18:12.600 been my bias, but I'm happy to be corrected if that bias is incorrect. But yeah, I've always

00:18:19.960 thought that, frankly, so many of the drugs that we're really interested in now as we look back

00:18:25.260 from an aging perspective, whether it be metformin or rapamycin, so many of the negative side effects

00:18:31.020 that people typically associate with those drugs are very difficult to isolate from the patients

00:18:35.460 in whom those drugs have historically been given. And so it's nice to hear that Mala is now being

00:18:42.080 generally recognized as an alternative viewpoint to that. You know, metformin is to me an interesting

00:18:48.180 drug from a diabetic standpoint because, and I don't know that this was appreciated in the 80s,

00:18:53.300 in fact, I would suspect it was not, because there's really two macro strategies for improving type 2

00:19:00.580 diabetes. You can, obviously the highest goal is to control glucose levels, so to regulate the degree

00:19:07.960 of glucose in the blood. But you can do that through, at the highest level, two ways. You could

00:19:13.760 increase insulin, either exogenously or through increased insulin production pharmacologically,

00:19:19.100 or you could reduce glucose. And of course, metformin falls into the latter category of that,

00:19:24.880 or increasing muscle insulin sensitivity to enhance glucose disposal. Today, it's generally regarded that

00:19:32.420 while both strategies will have an equal benefit on the microvasculature, the glucose lowering by insulin

00:19:40.340 lowering, so the less glucose production strategy, has a superior effect on the macrovasculature. And therefore,

00:19:48.180 metformin would be a better alternative to, for example, a drug that's going to increase insulin production

00:19:53.320 out of the pancreas. Was it appreciated at the time, meaning when you were doing this in the late

00:19:58.360 80s, how potentially beneficial this drug was? Well, at that time in the United States, it was all about

00:20:05.040 insulin resistance. You know, people like Jerry Riven, Ralph DeFronzo, the people at the NIH, Ron Kahn,

00:20:12.320 were all, you know, discovering the insulin receptor, discovering insulin resistance,

00:20:17.560 discovering the association of insulin resistance with the metabolic syndrome.

00:20:21.080 So it was all, we all had the bias that the major problem with type 2 diabetes is insulin

00:20:27.820 resistance. And, you know, we know now that it's a total collaboration. Yeah, you increase

00:20:34.440 insulin resistance and the pancreas have to secrete more insulin. And we humans, at least in the condition

00:20:41.940 of our environment, obesity, everything, many of us, I would say 40%, cannot deal with it. And we become

00:20:50.080 diabetic. There is this starling curve of the pancreas, you know, like for the heart, that you

00:20:57.300 increase insulin secretion. And at some point, you cannot increase insulin secretion, you become

00:21:01.820 diabetic and then insulin secretion decrease. And so insulin resistance was a major way. And that's why

00:21:09.080 metformin was a good place to come with, you know, at least it affects mainly the hepatic glucose

00:21:19.160 production rather than the muscle. Although I have to tell you in vitro on muscle specimen,

00:21:25.380 it's an insulin sensitizer in the muscle too.

00:21:28.300 Let's talk about this idea of insulin resistance. I think there are a few terms that leave me scratching

00:21:33.840 my head more than that one. So this is an example of something where maybe five years ago, I thought

00:21:39.260 I really knew what insulin resistance was. And I think today I'm pretty sure I don't know what it is

00:21:43.840 in the sense that when you take the typical phenotype of someone who's insulin resistant,

00:21:48.600 what do they look like biochemically and anthropomorphologically, right? So they're

00:21:53.460 hyperinsulinemic, they have elevated levels of glucose, they probably have some degree of obesity or

00:21:59.360 adiposity. So what does that mean? Because clearly their fat cell is quite sensitive to insulin.

00:22:05.840 If the fat cell ever became resistant to insulin, you could not re-esterify fatty acids and you would

00:22:13.000 have an endless stream of lipolysis exiting fatty acids from the fat cell. So you'd actually be quite

00:22:19.280 lean. There's something going on in the muscle that clearly is resistant to the effect of insulin. But

00:22:24.060 of course there are both insulin dependent and insulin independent means by which we can dispose

00:22:29.060 of glucose. And this is where, coming back to what you said a moment ago, I actually wanted to ask

00:22:33.500 you that question, which is, but we'll park it, but the question was, does metformin participate in

00:22:40.140 the AMPK driven insulin independent modality of glucose disposal? You're nodding, so I think that's a yes

00:22:47.320 and we'll come back to it. And then there's the liver. And this is the one to me that is the most

00:22:51.940 complicated and the one I'd like to begin with. So before I ask you to elaborate on the specifically what is

00:22:58.400 meant by insulin resistance in the liver, am I a complete moron for not understanding this?

00:23:03.700 No, but can I make it a little bit more interesting even? Because I need to bring it to aging.

00:23:09.340 Yes.

00:23:09.680 I would love to talk about diabetes, but let me just put up front. I'm not sure that the diabetes

00:23:15.780 property of metformin are really the aging properties of metformin. Okay. This is my provocation.

00:23:23.720 But let me go back to insulin resistance because in 1997, a science paper appeared that made it

00:23:34.520 the best day of my life and the worst day of my life. Okay. Wow.

00:23:40.020 What was the paper? The paper was taking a nematode. Okay. So a primitive model decreasing

00:23:48.360 the insulin sensitivity. This is known as the DAF2 model. And when you do that, the nematode accumulates

00:23:57.500 fat in their intestinal cells. So they are becoming visceral obese and they live several times longer.

00:24:06.000 So why is it the best day in my life? Because you could, with one genetic manipulation,

00:24:12.740 extends lifespan significantly. That meant going from hope, actually from nothing.

00:24:21.300 Was this Cynthia's paper?

00:24:22.300 No, that's Gary Rufkin. But at the same time, Cynthia had their DAF16. Tom Johnson has had the

00:24:28.720 age one model. There are several papers that came and all of them said, hey, we can change

00:24:34.260 a lifespan. And the example was always insulin sensitivity.

00:24:37.120 And this mutation, to be clear, because between DAF2, DAF16 and the dietary manipulations,

00:24:44.280 there are several permutations of that C. elegans model. But just to make sure I know which paper

00:24:49.560 we're talking about, this was only attenuation of DAF2, nothing to DAF16 and nothing to dietary

00:24:56.800 change. Right, right.

00:24:58.500 Can you just, for the listener, define what is the analog of DAF2, DAF16 in us?

00:25:03.540 DAF2 is the insulin receptor. And DAF16 is the FOXO, the FOXO transcription factor. But the point

00:25:13.160 is, at that time... Yeah, you took a worm that normally lives two weeks and you turned down its

00:25:20.420 insulin receptor slightly. Not off, correct?

00:25:23.180 Right, right. But you made it insulin resistant and it was accumulating fat. And what was I bringing

00:25:30.360 in to the field? I was saying the major reason for aging is this insulin resistance syndrome. And the

00:25:38.280 main part of the insulin resistance is accumulation of visceral fat. And so the premise was good. The

00:25:49.160 example was disaster to me. I had the JCI paper at that time in press, really showing with MRI pictures,

00:25:57.740 how caloric restriction decreases the visceral fat. I was talking about the biology of those fat and stuff.

00:26:05.120 And it was a major paradox for us in the field and how we continue if we're saying insulin resistance is

00:26:12.320 good for longevity and everybody in diabetes knew that this is a disaster. Why am I telling you that? Because

00:26:19.780 I learned later, I was thinking later on experiments that I've done. Oh, so first of all, the experiment

00:26:25.680 that I've done that was very conclusive was I took a bunch of rats, 150 actually, and all of them

00:26:34.520 underwent surgery after puberty. In some of them, I removed their visceral fat depots by surgery.

00:26:42.440 And in the other, it was a sham procedure. So I just moved it, but didn't remove it. And we had three groups

00:26:48.920 in the experiment. One was ad libidum feeding. The second was caloric restriction. This is the control.

00:26:56.660 They would live 40% better. And the third group was ad libidum feeding of the rats that their visceral fat was

00:27:06.040 removed. And I said, you know, without the visceral fat, even with nutrients, they live longer. And

00:27:12.600 they actually lived significantly longer than the ad libidum by 20%, but not as much as the caloric

00:27:18.000 restriction. So just to make sure I understand, you had a two to one ratio of your animals because you

00:27:23.580 had... No, I had three groups. Three groups, but one of them was all ad lib feeding of the visceral fat

00:27:29.860 removed. That was... Right. And then of the ones that had sham surgery,

00:27:34.100 they were randomized to ad lib versus CR. Correct. It really showed that visceral fat,

00:27:39.000 the removal of visceral fat had major effect. And when the animals who had the visceral fat removed

00:27:44.980 were autopsied or whatever the word is for rat autopsies, had they figured out a way to reaccumulate

00:27:52.200 visceral fat or where had they accumulated fat? No. In fact, when you do this procedure after puberty,

00:27:58.500 if you do it before puberty, they accumulate visceral fat. Otherwise they don't accumulate much

00:28:03.800 visceral fat. Did they have changes in subcutaneous fat? That's a good question.

00:28:08.420 Mice have it when you do it and rats do not. What did they ultimately succumb to? They all die

00:28:13.800 from the same thing only at different times. Which is what? Cancer? Yeah. It was sprague

00:28:19.060 dolly. So kidney disease and cancers were the leading cause of death. So you basically got half the

00:28:24.880 benefit of CR by doing this operation. By eating ad libidum, but without visceral fat. Yeah.

00:28:32.000 So you understand it's very... Sorry, one other question, Nir. Was there any change in their

00:28:36.440 health span or their spryness? Yes, there was a health span. In fact, what happened to the ad libidum,

00:28:43.160 the ad libidum starts losing weight at the end of their life. And the ad libidum with visceral fat have

00:28:51.280 still gained weight after the others, you know, it was a significant change in weight. And then there's

00:28:57.240 a whole health span, their insulin levels and other things that we've done in parallel. So they were

00:29:02.380 healthy and they were healthier for longer, which is kind of the experiments that we do

00:29:07.220 with animal models. Is it possible to take one of those rats and either through caloric restriction or

00:29:14.860 other dietary restriction, get it to puberty to the point at where you did the surgery,

00:29:19.820 but without any visceral fat. And then ask the question, if they've gone through that period of

00:29:25.360 development without developing visceral fat, do they have somehow protection from the diseases that

00:29:31.360 come even if you feed them ad libidum thereafter? So we did this experiment in Zucker fat animal,

00:29:37.940 Zucker diabetes animal, and we took their visceral fat... Tell people what the Zucker rat looks like.

00:29:44.640 So the Zucker rats are rats that are really, really, really obese, hungry all the time.

00:29:51.980 So they have hyperphagia. They can't stop eating.

00:29:54.160 Right. Because they don't have the leptin receptor. So they are hungry all the time. And they're nasty

00:30:00.700 because you try to come close to them, they think your finger is food. So they're eating all the time.

00:30:06.780 And we did surgery on them before puberty because they became obese really much before puberty.

00:30:12.460 And what happened is by six months, they all become diabetics. Our animals up to four months

00:30:21.220 did not become diabetic. And then between four and six, they became diabetic, although we took

00:30:28.260 their visceral fat. But what we found out that 80% of the visceral fat grew back. So we actually had

00:30:35.780 the perfect experiment. As long as they didn't have visceral fat, they didn't develop diabetes.

00:30:40.280 Once they got the visceral fat, they developed diabetes. But I want to make...

00:30:44.560 And sorry, one other thing. If you took that group and who already developed diabetes with

00:30:49.880 visceral fat, and you just removed the visceral fat, can you reverse the diabetes?

00:30:54.500 No, we've never done this experiment. Don't know. Because we're interested in aging.

00:31:00.700 But I'm telling you all that to make one point, because so far, the paradox just got worse,

00:31:07.220 right? The nematodes with insulin-resistant visceral fat lives longer. And we showed that

00:31:12.280 mammalians live longer if you take this visceral fat and make them insulin-sensitive. Whenever we

00:31:17.720 take visceral fat, we make them more insulin-sensitive, okay? Until I realized one thing. It takes me

00:31:24.600 few hours to make rats insulin-resistant. I give them glucose. I give them free fatty acid. I give

00:31:32.260 them some other things. I can make them insulin-resistant very rapidly. So at first, I used

00:31:38.900 it as, you see, we can do the chronic insulin-resistant in a few hours, okay? You just load

00:31:44.920 them with nutrients, you become insulin-resistant. But then I thought of the aging part. When you get

00:31:51.040 glucose to the muscle, the glucose goes into the muscle, if you're not moving the muscle, it's not

00:31:57.180 going to turn more glucose to energy. It will switch from free fatty acid to glucose, but there's no

00:32:04.020 more glucose burning. So it goes to glycogen, right? And it goes to glycogen, but you keep on putting

00:32:11.360 glucose in the muscle, and there's just so much glycogen that the muscle can store. So the muscle has to

00:32:18.360 become insulin-resistant. Okay? It's a protective mechanism.

00:32:22.480 There's no evidence that the muscle can, in any way, shape, or form, undergo de novo lipogenesis

00:32:27.780 and create any of the fatty acid, is there?

00:32:30.200 To create any of the fatty acid?

00:32:32.500 From glucose? No, not.

00:32:33.900 Is there any evidence that muscle can carry out de novo lipogenesis?

00:32:37.500 I don't think so. I don't think so. I'll have to think, because I have a different...

00:32:42.940 Yeah, yeah.

00:32:43.560 I'll have to think about it. But the point is... Okay, but I understand your point. I don't think...

00:32:49.940 If anything, there's no much. The glucose, the muscle tells the glucose, you go somewhere else.

00:32:55.000 You go to fat, okay?

00:32:57.000 So the only organ that can take in a completely excess amount of glucose in the end has to be the liver,

00:33:01.580 because it at least has the capacity to turn excess glucose into fat, whereas all of the other organs

00:33:07.660 are going to be saturated.

00:33:08.860 That's true. And that's part of, I think, where the glucose is going, okay? To fat and to liver.

00:33:14.320 Okay, so what am I telling you? I'm saying that insulin resistance is a protective mechanism.

00:33:19.980 It's a modulator. It's a stress response. It's something.

00:33:23.420 So now I understand why a stress response mechanism in one animal caused them to live longer,

00:33:31.780 and in another animal, it's a pain in the butt.

00:33:37.100 So I went and I wrote this review with Luigi Ferrucci. I wrote this review where we took

00:33:42.580 all the animals that had problems with their insulin sensitivity, the animals that are insulin

00:33:49.660 resistance and live longer, and the animals with insulin sensitivity that live shorter.

00:33:58.340 And there's a huge list of it. Rapamycin is an example, right? Rapamycin causes insulin

00:34:03.120 resistance, and the animals is the best intervention in rodents. There's a PTB1, a transgenic

00:34:11.080 animals. It's very insulin sensitivity, and it has half the lifespan of a wild-type animal.

00:34:18.620 So when you say insulin resistant or insulin sensitive, are you always referring to muscle?

00:34:22.440 Are you referring to liver? How are you defining them?

00:34:24.880 Well, in this sense, it was everything. It was IRS, for example, IRS-1, IRS-2 knockout,

00:34:32.720 some of them only in the brain. So it was really insulin resistance everywhere.

00:34:38.740 Is IRS-1 found in the muscle and not the liver?

00:34:42.240 No, IRS-1 and IRS-2 are everywhere, but IRS-2 is more in liver and brain.

00:34:48.900 Yeah, they're selective. They sort of have selective expression.

00:34:52.260 Right, right.

00:34:52.780 Yeah, so I'll just translate. IRS-1, insulin receptor.

00:34:55.880 Substrate.

00:34:56.420 Substrate. So let's talk about exactly how the muscle takes in glucose. So we are walking around.

00:35:04.240 If I bolus you with glucose right now, and your blood glucose rises from wherever it is right now,

00:35:10.620 90 to 200. Let's say I give you an enormous bolus of glucose, I double it. So you're up to 180

00:35:16.880 milligrams per deciliter. This is a very high, screaming high level, which by the way, only

00:35:20.660 amounts to an extra few grams of glucose. But nevertheless, what is the chain of events that

00:35:25.480 leads to insulin being secreted and the muscle ultimately disposing of that glucose,

00:35:30.300 both actively or passively? Well, there'll be obviously a secretion to insulin that's relative

00:35:36.240 to your glucose level. And so the beta cell is the sensor. Right. And out comes insulin. And now

00:35:42.620 what is insulin doing to the muscle? And insulin through the insulin receptor will get translocation

00:35:49.660 of the glutphor. Those are the glucose transporters that are the major glucose transporters in muscle.

00:35:56.120 It will go from an intracellular pool into the plasma membrane, integrate it, and starts getting

00:36:02.980 glucose like crazy into the cells. And that happens quite passively, meaning once the GLUT4

00:36:07.560 transporter is translocated across the cell membrane, glucose can passively rush in, correct? It doesn't

00:36:13.460 require ATP to bring it in against a gradient of any sort. Correct. And in fact, it has its own

00:36:20.980 intrinsic activity that's not energy dependent. So it can go faster or slower by different ways.

00:36:28.140 Now, is there another method by which without insulin, we can get glucose into a cell that

00:36:33.740 somehow relies on AMP kinase? Not on AMP kinase, but there's non-insulin mediated glucose uptake. In

00:36:41.240 other words, there's a way for glucose to get up without the insulin. And we know that from hyperglycemic

00:36:47.220 clamp and from other things that when we can move the insulin or keep insulin level at basal,

00:36:54.020 and still there's a substantial glucose uptake that's happening. So yeah, we know that.

00:36:59.520 What's the mechanism of that?

00:37:00.720 I'm not sure I know.

00:37:03.180 It also seems to be enhanced by exercise.

00:37:06.440 That's true. That's true. This NIMJA is the non-insulin mediated glucose uptake

00:37:11.360 is something that he's exercise dependent to. That's why now I understand your question. Is AMP

00:37:17.940 kinase part of the thing? And I don't know that. Yeah. I only have one patient in my practice who

00:37:24.480 has type 1 diabetes, but he is such an interesting patient because of his incredibly strict dietary

00:37:32.240 control and his unbelievable appetite for exercise. And I've never seen a higher level of adiponectin in

00:37:42.660 a human. I've never seen a higher level of sex hormone binding globulin in a human, which are

00:37:48.420 basically, and we know how much insulin he has because he injects it and he injects so little

00:37:53.160 insulin into himself and yet can seem to, you know, he has a hemoglobin A1C below six using six to eight

00:38:00.780 units of insulin a day. And so what got me, it got, that's what got me very interested in this

00:38:05.540 non-insulin dependent glucose uptake. Yeah. I'm not doing this research anymore.

00:38:11.080 I used to be in front of that. I, glucose itself can modulate glucokinase activity in the liver.

00:38:16.960 There's a lot of things that happening. And of course there's coordination between the fat and

00:38:21.200 the liver. And then I started being interested because we can do things to the hypothalamus and

00:38:28.440 take over this insulin and muscle and fat and liver. We can do it all from the brain.

00:38:36.260 Isn't that amazing? Well, just because I'm so fascinated by this and I think it will help the

00:38:41.760 listener to understand the complexity of what you just said. If I took an animal and put a lesion

00:38:48.000 into the ventral part of the hypothalamus, a normal animal, what could that do? What would that change

00:38:53.940 about its metabolism? Well, that would increase basically food intake and change a lot of the

00:39:01.160 peripheral physiology, but we do basically the opposite. We give insulin to the hypothalamus or

00:39:08.040 glucose or IGF or leptin. Okay. And then we see what happens to the periphery, although in the periphery,

00:39:16.560 those hormone levels are not increasing at all. And not only that, the liver is a good target to follow

00:39:23.300 because what you can do with the liver, you can do selective hepatic vagotomy. And everything that

00:39:30.480 you did through the brain doesn't work anymore because it needs the nerves. Right. So I'll explain

00:39:35.700 to the listener what that means. So the vagus nerve connects the body through this parasympathetic

00:39:41.040 system. And of course, if a patient has a liver transplant, or if you do an operation where you sever the

00:39:46.580 vagus nerve, you sever that connection between the central nervous system and the periphery.

00:39:51.980 Correct. And so we can do it experimentally and it will help us see if it really-

00:39:57.400 How much of it is vaguely mediated versus not?

00:40:00.100 Or, yeah, or more nerve only. You know, it's the nerves because we do, it's not only vagotomy,

00:40:06.280 we do a little bit adrenergic too, but it's through the nerves and not through a chemical reaction on the

00:40:12.920 liver. You see, this is the problem with metabolism, Nir. The more I go into it, the less I know.

00:40:20.040 It drives me nuts. My absolute knowledge increases incrementally and my relative knowledge falls

00:40:26.320 precipitously. Well, I think I realized that you know what to ask. So you're undermining your own

00:40:31.960 abilities, but it's really, it's really the integrative metabolism is really very confusing

00:40:37.680 to explain. So we will define insulin resistance at the muscle, whether it be from an aged phenotype

00:40:46.420 or a diabetic phenotype, as a scenario under which a fixed amount of insulin hitting the insulin receptor

00:40:53.380 produces fewer GLUT4 transporters. Is that a fair definition?

00:40:57.960 Well, the definition of insulin resistance is really different and very simple. It's all about

00:41:07.740 the glucose uptake in the muscle, really. It's all about the ability of insulin to clear glucose.

00:41:16.860 Okay. That's the only way we define clinically insulin resistance, although that totally misses the

00:41:24.120 point because those insulin levels have different effects on different tissue. But for us, the

00:41:30.380 insulin sensitivity is totally related to the glucose uptake. So let me give you an example. So

00:41:36.520 when I use an oral glucose tolerance test with my patients, I'm a bit of a stickler. So if they take

00:41:42.620 their 75, you know, grams of glucola, I measure their glucose and insulin at baseline, administer the

00:41:49.040 glucola. 30 minutes later, 60 minutes later, 90 minutes later, and if I'm feeling aggressive, 120

00:41:55.460 minutes later, we remeasure the glucose, the insulin, and the FFA, and maybe even the C-peptide. But for

00:42:01.500 simplifying it, just the glucose and insulin. So let's assume you have two patients who start out with a

00:42:07.180 fasting glucose of 90 milligrams per deciliter and six insulin of IUs of six. They both get their 75 of

00:42:15.220 glucola. And let's, again, simplify this by just looking at one hour, what happens. Both of them at

00:42:20.560 one hour have a rise of glucose from 90 to 130 milligrams per deciliter. One of them did so with a

00:42:28.400 rise of insulin from six to 20. The other did it with a rise of insulin from six to 90. They've both

00:42:36.640 disposed of glucose with equal magnitude. Would you describe them both as equally insulin sensitive?

00:42:44.120 It's a little bit complicated. You know, first of all, so let me just say for aging, 120 is not

00:42:52.960 enough. In elderly, the glucose tolerance goes. So the rise in insulin is changing throughout normal

00:43:01.460 physiology and throughout individuals. And you might have missed some of the insulin peaks earlier on

00:43:09.300 or later on. Yeah. The third, looking at 30 versus 90 is a big insight. And you're right. I mean,

00:43:17.260 Joseph Kraft, who is a pathologist who has done a lot of work on this, he samples to five hours. Now,

00:43:24.600 again, clinically that's challenging in the regular day-to-day practice, but in the laboratory that's

00:43:29.860 feasible. And you're right. You can see so much happening beyond that. But I tried to pick an example

00:43:35.920 that was as egregious enough in its difference that what I'm trying to get is the difference in

00:43:41.140 hyperinsulinemia and whether that factors into how we think about insulin resistance.

00:43:45.480 Right. No, I don't think the clinical world is thinking like that. So I'm actually an active

00:43:52.860 endocrinologist. I still see on Thursdays if I'm in town teaching fellows in diabetes clinic in Montefiore

00:44:00.840 Hospital. So I am involved very much in this field. And I'll tell you that we rarely measure insulin in

00:44:09.560 any of our patients, but that's not really your patients. Okay. I'm talking about what's the use of

00:44:15.820 measuring insulin in type 2 diabetic patients. So it's a different issue. But yeah, this sounds like

00:44:23.420 there's a different insulin response to lower the glucose to the same extent.

00:44:29.540 Because the hypothesis would be that the patient with hyperinsulinemia is now demonstrating. I mean,

00:44:35.380 that's basically a harbinger of the first one who's going to struggle with glucose disposal.

00:44:40.100 That would be my hypothesis is the one who needed 90 units of insulin to dispose of glucose with the

00:44:46.200 same efficacy as the person who needed 30 is sooner rather than later going to have a harder time

00:44:52.140 disposing of glucose. Well, you know, there is another possibility that those are the people who

00:44:58.240 get other diseases first, like heart attacks or stroke. In other words, their insulin resistance

00:45:05.080 will affect a lot the way they do the atherosclerotic plaque and, you know, this hyperinsulinemia on

00:45:14.060 organs that might not be insulin resistant even, right? On cells that not come in insulin resistance.

00:45:19.360 So I would say that they could either the one who become diabetic or the one who are going to get

00:45:25.000 macrovascular disease much faster. Which brings it right back to our observation of glucose versus

00:45:30.740 insulin, micro versus macular vasculature. That's a very good point. So how does insulin signaling work

00:45:37.840 to get glucose into the liver? The liver has a different glucose transporter. It's a GLUT2

00:45:43.560 that is not translocated and that is the major way by glucose gets into the liver. So it's a little

00:45:53.320 bit different, but it's not stimulated the same. So it's constitutively expressed across the membrane?

00:45:59.720 It's not stimulated really by insulin as much. So it's more gradient driven to get the glucose from

00:46:07.240 the circulation into the liver. Yes. And that's where the portal blood is. So there it's very

00:46:13.180 sensitive to the increasing glucose concentrations. Well, that's very interesting because it certainly,

00:46:19.160 I mean, there's lots of evolutionary reasons why that would be a fail safe, right? Given the

00:46:22.820 importance of, you would never, ever want the liver to be denied glucose given that if you ever shut down

00:46:30.660 the liver's ability to make glucose, by my calculation, you could live about six minutes.

00:46:35.100 Yeah. Look, I was an intern of Sheila Sherlock, Dame Sheila Sherlock in England in a Royal Free

00:46:40.960 Hospital, very known pathologist. And she asked a group of us, what's the main role of the liver?

00:46:47.320 And I raised my hand and I said, it's to produce glucose. That would be my answer, by the way.

00:46:51.340 And everybody laughed at me and she threw shocks at them and stuff. And she said, just a minute,

00:46:55.740 why are you saying that? I said, because if you take the liver out and clamp the vessels,

00:47:00.540 the guy will die from hypoglycemia, which is absolutely true.

00:47:04.940 Yeah. Yeah. It's, um, I, you know, maybe it's just because I'm in the midst of writing this

00:47:10.780 chapter for my book about the liver, but I literally yesterday I sat here in my room and,

00:47:16.420 and penned out the calculation of 180 pound person with blood glucose of 180 milligrams per deciliter

00:47:24.020 clamp the liver. How many minutes until they die? It's about four under those, under those conditions.

00:47:30.540 Can you imagine that? Think about like what this beautiful organ has to do.

00:47:34.580 But you know, another thing to consider in your example, right? Uh, so for example, elderly

00:47:42.100 have enough insulin to suppress glucose production. Glucose production is easier to suppress

00:47:49.980 then it's suppressing lower level of insulin than stimulating glucose uptake in the muscle.

00:47:57.480 The liver is more sensitive than the muscle. And so elderly have the ability to suppress

00:48:05.720 hepatic glucose production because they have enough of insulin to do that. But when you start giving

00:48:11.960 them food during the day, they fail and they become glucose intolerant and diabetic. They might have

00:48:18.640 diabetes, although their basal level are well, just because of this, this enough insulin to suppress

00:48:27.840 glucose production and not enough to increase glucose uptake. So the sensitivity of the tissues is really

00:48:34.460 to insulin is really part of what you're describing. If you look at the basal, it's very different when

00:48:41.240 you look at the challenge. That's such, such a great point. Um, and I won't, I won't go down that

00:48:46.940 rabbit hole anymore because I, I, I would love to, but there's so many other things I want to talk

00:48:51.140 about in here. Let's, let's go back to metformin and let's do it now through the context of an

00:48:56.240 anti-aging drug. So there's going to be some people listening to this who already know everything

00:49:00.980 about you. If for no other reason than the efforts you've put into tame, obviously I want to have plenty

00:49:06.700 of time to talk about that, but let's back up for a moment. Tell me when you first realized,

00:49:12.140 Hey, this metformin drug that I worked on 30 years ago, it's not just a great drug for people

00:49:18.180 with diabetes. This could actually be a drug that helps someone without diabetes live longer.

00:49:23.440 What was that epiphany for you?

00:49:25.260 So first of all, there were studies on the biology of aging in rodents. And the first guy,

00:49:34.100 which I'll be embarrassed now to forget, he's a guy from Leningrad that was the first to say,

00:49:40.640 I've done a rodent studies and animals with metformin, Anatoly. I'll remember it maybe later.

00:49:49.780 And he showed that a life extension, extension increases in variety of animals to which he gave

00:49:56.520 metformin. That led to some studies.

00:49:59.480 What was his hypothesis? Do you remember? Why would he, what would he, what drove him to do that?

00:50:03.480 There was never any hypothesis.

00:50:06.220 This was just a pure fishing expedition.

00:50:07.960 It's even wasn't related to glucose metabolism necessarily. He thought that maybe there's an

00:50:14.120 anti-cancer effect and he looked at that, but then he got models that were hypertensive and they also

00:50:20.360 live longer. So he thought those are different effect. And then he considered to aging. But at that

00:50:25.860 time, this was in the eighties when he started that. And it bothers me that I don't remember his name.

00:50:34.220 That's okay. You'll remember.

00:50:35.320 I should take more metformin. And he was pushing, he was showing up in meeting and showing. And so

00:50:42.100 people have tried to do it in the United States and they showed that there's a significant effect

00:50:48.140 on insulin. By the way, the effect of insulin in animals is 10%, not like rapamycin is 24%.

00:50:56.620 So there's a milder effects, but in every study, they showed that health span is improved, you know,

00:51:05.460 by 50%. So the effect on longevity was less than the effect of health. By the way, that's perfect for

00:51:12.560 what I'm trying to say. I'm not for longevity. I'm just for health span, right? So if we can leave

00:51:18.360 healthy, healthy, healthy, die, that's fine with me.

00:51:21.160 Yeah. The so-called squaring of the longevity.

00:51:23.500 Squaring, right. Squaring of the longevity. So he started Asimov. His name is Asimov.

00:51:29.640 So that's what Asimov showed. People have started looking at that. There are several studies,

00:51:36.600 three studies that gave it also to nematodes. They also lived longer. And there's several effects

00:51:42.500 on health span in the biology of aging. Now, the reason there's a lot of needle threading here in

00:51:51.980 my story here, okay? But I want to tell you the moment that I knew we have to do this study

00:51:59.900 was a publication, not in a very high impact journal, but it was an amazing study that could

00:52:08.240 be done only in the UK. Because in the UK, you can go into pharmacies and look at prescription

00:52:19.140 that they gave. And you could follow the mortality of those patients.

00:52:28.040 Because the NHS system allows you to centralize everything from prescription to usage to mortality.

00:52:34.860 Exactly. And they're probably, they don't need to be de-identified after mortality. When you're

00:52:43.560 dead, you're not private anymore. So they did this study. So what they've done,

00:52:50.040 there are four arms to this study. They came and took patients who are on sulfonylurea, 12,000 patients,

00:52:58.240 okay? And they matched them to 12,000 people without diabetes. Okay? Same pharmacies, same doctors,

00:53:07.940 controlling for some other things. And of course, the people on sulfonylurea had higher mortality

00:53:14.240 than their control, you know?

00:53:16.140 Right. Because of course, they had a reason to be taking the sulfonylurea.

00:53:19.900 Right. They had diabetes. So that's okay. Then they took 78,000 people on metformin and 78,000

00:53:29.600 people who were non-diabetic. Again, matching them.

00:53:33.900 And the 78,000 on metformin was metformin monotherapy.

00:53:37.600 Monotherapy.

00:53:38.600 First line.

00:53:39.380 Right. And they watched their mortality. So just to underline again, the people on metformin

00:53:48.600 had diabetes. The control didn't have. They were more obese than the control. They also

00:53:55.100 have more diseases than the control. But they had significant less mortality, 17% less mortality

00:54:03.440 in the metformin group. And, you know, the sulfonylurea group was, okay? So if you get less mortality

00:54:15.880 with metformin, when the setup is diabetes, that really shows that metformin has a very important

00:54:26.220 effect in humans as far as aging.

00:54:29.040 What was the median dose of metformin in those populations?

00:54:32.240 They were over 1,000 milligrams.

00:54:35.620 That's it. Just 1,000 milligrams.

00:54:37.260 Well, they intended more, but the average was about 15 milligrams.

00:54:41.420 Yeah. You know, I don't-

00:54:42.960 The average was 1,500?

00:54:43.820 Yeah. No, 1,000 milligrams. I'm sorry. They, this is a discussion there. I don't think they

00:54:52.460 really could validate the dose.

00:54:56.380 Because an elegant thing to do if you have 78,000 in your database is now stratified by dose.

00:55:03.760 Absolutely.

00:55:04.020 Who's taking 500, 1,000, 1,500, 2,000?

00:55:07.380 They couldn't do that. And I really cannot answer you if you can get back to the data or not. I really

00:55:14.480 don't know. The dose we chose came from other studies. But, you know, that's not the only thing

00:55:21.260 we know about metformin.

00:55:22.460 What was the time course of that, by the way? Do you know how many-

00:55:25.580 Five years. Five years mortality.

00:55:27.060 So, five years exposure to metformin or five years of prospective mortality following?

00:55:33.860 No. Well, it's all, it was all a prospective study. Okay? It was all a prospective study. When

00:55:41.580 they started, okay, they went back, they looked at everybody on metformin.

00:55:46.900 Oh, I guess what I'm asking is, do we know how long they needed to be on metformin to achieve

00:55:51.100 the benefit?

00:55:51.720 No, no. We don't know.

00:55:52.460 Okay. So, we followed them for five years of mortality in a prospective cohort. But they

00:55:58.280 could have been on metformin for five years or 25 years. We don't know.

00:56:02.160 No, no. It's newly prescribed metformin. They were not more than five years.

00:56:07.100 Oh, so we have normalized by duration as well.

00:56:09.100 Absolutely. Absolutely. No, they haven't been on metformin before. Those are newly diagnosed

00:56:14.060 type 2 diabetes.

00:56:15.860 So, Nir, that's a pretty, to be blunt, goddamn staggering result.

00:56:21.260 So, my metformin moment, nowhere near as dramatic as yours, basically came out of a research project

00:56:31.080 I did with two analysts in 2013. And I don't even remember what prompted the question. But there

00:56:38.680 was a question we had internally about what was the benefit of metformin. Oh, no, no. Now I remember

00:56:46.460 the question. The question was, was there a relationship between hyperinsulinemia and breast

00:56:51.240 cancer? This was the question we wanted to ask. And as we dug and dug and dug and dug and dug,

00:56:56.480 something kept hitting us over the head over and over and over and over again, which was with or

00:57:02.520 without diabetes, with or without obesity, with or without hyperinsulinemia, any way you sliced and

00:57:09.180 diced the data of people with type 2 diabetes with and without metformin, they got much less breast

00:57:14.340 cancer, which then dug us down a rabbit hole if they seemed to get much less cancer, which was,

00:57:20.420 you know, that was a big aha moment.

00:57:23.860 Right. And there are hundreds of studies that show the association between metformin and less

00:57:30.740 cancers. And not only that, less all cancers, except prostate, by the way. Prostate is hanging there

00:57:37.080 maybe a little bit. And when we went, and I'll tell you later, but when we went to the NCI to make

00:57:43.480 them partners in this TAME study, it was really interesting. Can you tell people what TAME stands

00:57:48.120 for? TAME stands for taming or targeting aging with metformin. And it's a study that's designed to

00:57:54.820 prove the concept that aging can be targeted, but also, or mainly for me, is for the FDA to give an

00:58:02.760 approval to an indication that's like aging. The first time I heard about TAME was actually here

00:58:07.800 in this city. I was having dinner with Steve Osthead. It must've been in 2015, maybe 14, 15.

00:58:14.860 Steve is a gem of a human being. I consider him certainly one of the most important mentors in my

00:58:20.700 exploration of this field of aging. We're partners in many ways. He's a great guy.

00:58:26.080 He is, he is one of these guys that has, his knowledge is out of control. You can't ask him

00:58:32.120 a question that he doesn't know where to go for the answer. I have the highest respect for Steve

00:58:37.400 and can't wait to interview him here as well. But I remember him telling me about this. And my first

00:58:42.360 thought was, Steve, that's a crazy idea. The drug is free. Who the hell is going to pay for this study?

00:58:49.280 NIH can't pay for it because they don't consider aging a disease. Pharma can't pay for it because

00:58:53.720 there's no way to make money on it. I mean, I might pay like 10 cents a year for my supply of

00:58:58.440 metformin. It's a free drug. I said, you're doing the wrong study, man. You got to do this with

00:59:03.700 rapamycin. So talk to me about the challenges of this. This is, you're proposing something that is

00:59:10.660 really never been done before. Can I just ask you, are we going back to metformin? Because you asked

00:59:15.500 me about diabetes, but not metformin in action. Will you ask me later? Absolutely. I cannot wait to

00:59:21.680 dive into metformin. Yeah. We're not going to get off this topic for a while.

00:59:24.880 Yeah. So, so look, if I'm answering now, why metformin and not rapamycin?

00:59:31.320 Oh, you don't need to answer that question. I was just sort of teasing Steve that day.

00:59:34.400 Well, there are people who are asking, and of course, metformin, we have preliminary data.

00:59:40.040 We have the cancer. We have clinical studies, you know, clinical studies of metformin,

00:59:44.900 the diabetes prevention program, it prevents diabetes. The UKPDS, it prevents cardiovascular disease.

00:59:51.680 The Alzheimer literature, there are two clinical studies to suggest that in mild cognitive

00:59:58.780 impairment, it improves some domain, by the way, like name recalling. Asimov.

01:00:04.720 We'll, we'll up, we'll up your dose after dinner.

01:00:08.120 So there's lots of, and safety. Look, if we're going to have a new indication,

01:00:13.660 we don't want to kill anybody on the road. With rapamycin, we won't be sure. And it causes

01:00:19.860 diabetes and it causes a testicular atrophy and cataracts in animals. Okay. So we have to play

01:00:26.340 it safe. So I would only add one caveat in defense of rapamycin, which is with constitutive

01:00:31.100 dosing. Right. And, and you're, you're aware of the studies from John Manick, right? Yes,

01:00:37.180 of course. The last ones. Yes. Okay. So, so my view on rapamycin is it should not be dosed

01:00:42.800 every day. Exactly. It should be selectively dosed to target mTOR complex one, leave complex

01:00:48.480 two alone. Exactly. And I'll, we'll get back to that because I'll tell you what we've done

01:00:53.040 with metformin for the biology of aging part. So that's why the fact, you look, we are a bunch

01:00:59.780 of professors, including Steve Austin. We go to the FDA, by the way, beautiful movie by Ron Howard.

01:01:05.280 He went with us to the FDA, to the Senate. Wait, has it been released yet? This documentary?

01:01:10.400 A year and a half ago. It's called The Age of Aging. It's National Geographic. It's the best

01:01:16.960 movie that was done ever on aging. It's really quite incredible. Ron Howard is narrating it.

01:01:23.300 I remember, I'll tell you what, I was having dinner with Steve at some point during, he was

01:01:28.620 here in New York for that being filmed. Yeah. So, you know, so we're a bunch of professors

01:01:33.140 professors that are coming to the FDA and kind of telling them, you know, you know that aging

01:01:38.520 has biology, but this biology also can be targeted. And they said, so what? We said, well, if you slow

01:01:44.080 aging, then you prevent a bunch of aging-related diseases altogether. And I said, well, we're

01:01:52.160 interested to hear. And we went through this discussion with the FDA. I would say that the

01:01:57.760 most interesting part, and you'll appreciate it because you're diabetes centric. I see you're-

01:02:04.060 I'm actually not. I'm purely age centric.

01:02:07.540 You're metabolic-

01:02:08.780 But I'm diabetes. I think of diabetes as a great example of what happens when you don't fix things

01:02:14.940 earlier.

01:02:15.660 So we told them, basically we came and we said, we'll do a study. And what we're going to see is

01:02:21.580 cardiovascular disease and cancer and Alzheimer's mortality and diabetes. And I said, not diabetes.

01:02:27.040 You said, what do you mean not diabetes? I said, no, diabetes. Look, you diagnose diabetes. First

01:02:32.200 of all, you diagnose it on a chemical test. You know, you decide. And only 40% of them,

01:02:39.620 10 years later, will get complication. We're really not interested in diabetes.

01:02:43.420 Wait, wait. Did the FDA actually say this?

01:02:45.640 Absolutely. Absolutely said it. And by the way, I should add that there's another bunch of people

01:02:52.260 who came to tell the FDA that they should allow metformin for prediabetes because the DPP said

01:02:59.460 metformin prevents diabetes by 30%. And the FDA said, well, if you think that's important,

01:03:06.800 why don't you make the diagnosis of diabetes in 5.8 hemoglobin N1C? You know, it's not up to us

01:03:12.420 to do that. And they wouldn't give them an indication for prediabetes.

01:03:19.620 That's a very counterintuitive point of view when you consider the long-term, not short-term. The

01:03:27.740 short-term cost burden of diabetes is relatively trivial, but it's that long-term cost burden that

01:03:32.600 becomes quite elaborate.

01:03:34.360 You know, it's very boring with you because we have the same views. You're right.

01:03:41.860 So, yeah. So, okay. But so we had to adjust our study. We had to take diabetes as one of the

01:03:47.840 outcomes and put in and, you know, just increase the number and take diabetes out. It's not that

01:03:52.940 we're not going to follow diabetes, but the FDA is not interested in diabetes. So the, really the

01:03:58.340 challenge was how do we define aging in a clinical study? Because one view is, well, let's-

01:04:05.000 Because you can't use survival, overall mortality?

01:04:08.500 Well, there's no, there's no biomarkers to aging. All the diseases are, are, their major risk factor

01:04:15.920 is aging, you know? So how do you define it? And we basically did a lot of work and published a lot

01:04:21.620 of work and showed that, okay, let's, let's take somebody who has survived cancer and he's 65 to

01:04:30.020 80 years old. What's his chances of getting cardiovascular disease, cognitive decline,

01:04:38.160 mortality, right? The other things. And let's say it was 10. Okay. And okay. Now let's switch it

01:04:45.700 around. Let's take somebody who had cardiovascular disease. What's his chance of getting cancer and all

01:04:51.220 the others? 10. Okay. Let's take somebody with Alzheimer's. What's his chance? It really, with

01:04:56.720 aging, it doesn't matter. The disease you get first depends on, you know, your genetics and

01:05:04.740 environment. If you have a mother who's diabetic and you're obese, you'll get diabetes first. Okay.

01:05:10.160 But really, because, because we age biologically, you know, in different way, whatever it is, the next

01:05:18.020 disease, you're going to get the next disease. And we don't know what's the next disease, but that's

01:05:22.680 what we're going to prevent. That's sort of like, it's a really interesting idea you propose because

01:05:26.780 it's in lending, you would call that asset value correlation. So you'd say, well, let's say Nir comes

01:05:33.600 to my bank and he has a credit card and he has a car loan and he has a student loan and he has a

01:05:39.900 mortgage and he has a personal business loan. Which one is going to default first? Eh, I don't know.

01:05:45.700 I mean, I could sort of tell you typically people are going to stop paying their credit

01:05:49.340 card before they stop paying their car and their mortgage. But boy, once you default on

01:05:53.620 one, those dominoes start to fall very quickly. And so, so that would have what we would call

01:05:59.000 a very high asset value correlation of default. Yeah. So it's sort of like a, this is the same

01:06:03.500 for disease. Yeah. I think you're right. I'm saying it a little bit different. I'm saying

01:06:08.440 I'm agnostic to the disease. To which the first one is. Yeah. It's just. I'm agnostic. It doesn't

01:06:13.140 matter what you come and what, what you're going to have. I'm going to prevent whatever

01:06:17.200 it is. That's why we have a composite. Okay. A composite. And people get very prickly about

01:06:23.300 composite outcomes. Very. And I got it. I'm going to get on my soapbox for a minute here

01:06:27.840 because this pisses me off to no end when people get all bent out of shape about how you can't

01:06:33.540 have a composite outcome. And this is the bias against aging really. Because in the end,

01:06:40.120 what matters is how long are you alive and how long is your health span optimized?

01:06:45.800 So I'm sorry for, I made that rant as short as I normally could. Normally that would be a 10 minute

01:06:50.620 rant. Well, that would be a 10 minute rant for me too. Okay. Yeah, absolutely. And, and this is where

01:06:57.640 we get into trouble. I'll tell you what the other pieces, but we had our grant reviewed at the NIH and

01:07:04.220 unfortunately. And is this inside any of the groups except NIA? Like is NCI weighing in on this?

01:07:10.400 Yes. Yes. I'll tell you what we've done. I'll tell you what we've done in a second. Let me just

01:07:16.560 make sure that I say that. So our reviewers were not GRSI because we, everybody's involved and we have

01:07:26.500 14 centers. Okay. So everybody's involved. So the reviewers are from other place and they just,

01:07:32.680 they just, just a minute. You're saying that aging can be targeted and one drug can do it.

01:07:38.980 You're crazy. Okay. There's no such thing. Why don't you do three studies? One study is metformin

01:07:45.360 cardiovascular. One study is metformin cancer. One study. And we're saying no, because what are we doing?

01:07:52.020 We're trying to give aging an indication. Suppose we do it. It's just such an anthema to the way they

01:07:58.880 think. And I, I'm not saying that to be critical of them, right? But they're in a silo. I know.

01:08:03.680 Which is if you come to NCI, how can you care about anything but cancer? Right. I know the silos are

01:08:09.760 really, are really killers, but for us, so you, you could say, well, let's do all those studies. But for

01:08:16.080 us, we're calculating the power. I was just about to say the power analysis on that study

01:08:21.840 is going to quadruple your budget. No. Well, for what they do. Yes. For the silos. But for us,

01:08:29.480 it's not only, it's not only the budget. We don't want to stop this study. Let's say

01:08:34.380 that we show in two years that we, we delay cardiovascular disease. It's significant.

01:08:39.280 They'll stop the study. And they said, no, everybody now has to be on metformin,

01:08:42.480 but we wouldn't be able to show the FDA that we are aging. For us, aging is, stop telling us

01:08:51.560 diseases. We are going to have a cluster and we're just going to delay the aging. So we'll get the

01:08:57.600 health span extended by two, three years. That's what we're trying to show you. And the significance

01:09:02.560 is for the cluster, not for the individual disease. And if you tell me to do individual disease,

01:09:07.300 I need triple the, the budgets and I'm not going to get to the FDA with the indication for aging.

01:09:13.260 So it seems to me that the biggest challenge here is not the funding. It's not the study design.

01:09:19.460 It's the conceptual leap. It's, it's a completely different paradigm of how we think about delaying

01:09:26.420 death. Correct. Correct. Where, where do you stand today? So I'll tell you a few things. First of all,

01:09:32.440 we conceived it. I got my friends to what I call a prison, a French prison in Spain somewhere. We

01:09:41.340 were in a parador in the middle of nowhere in Spain for several days saying, what do we have to do in

01:09:48.100 order to open the field for aging, to allow us to get to treatments? And we decided several things,

01:09:55.200 but one of them is to go, to go ahead with time. By the way, I'm saying French prison because the food

01:10:00.280 was good, but it was in Spain. And we started, and I started getting those people, those clinical

01:10:06.940 people together. And we started really thinking through for what it takes, for what it takes.

01:10:12.900 And we had this whole academic and wrote papers in order to say how we're thinking, how we're thinking

01:10:18.760 of cluster of diseases, how we're doing power calculations for targets, aging and stuff like

01:10:24.900 that. We really spend a lot of time with the best people. Now, our friends came to us and said, just a

01:10:31.220 minute, your study is $70 million. If the National Institute of Aging is going to spend $70 million, that

01:10:38.860 means this, the National Institute of Aging is very small. It's just 3% of the budget, although we are 80% of

01:10:44.940 the diseases, right? Then we won't have money for the grants. And I said, well, what I'm going to do is

01:10:51.680 two things. First of all, half of the money is going to come from somewhere else. That's my

01:10:55.400 association with AFAR, where Steve and I are the scientific directors. And for AFAR, I got the

01:11:03.520 other half. So the $35 million is already raised, and that's mostly philanthropic. Right. And then we

01:11:10.240 went among the institutes. So NCI should chip in a bit. NCI, NHLBI, NIDDK. They didn't commit because

01:11:19.300 first they have to see the review, okay? Or they have to have the decision that the NDI wants.

01:11:25.500 I would tell you though, that the NCI director, who's a good friend of mine, is waiting for TAM to

01:11:30.820 come. You know, he's like, where is TAM? So the NCI has committed. It's basically a bunch of

01:11:36.320 contingent commitments. Yeah. I just want to say committed is the wrong word because they didn't

01:11:42.360 have a budget yet and the grant didn't come to them. But I know that they will give them money

01:11:47.100 because their head is, will do it. Yeah. So that's good to hear. So

01:11:53.000 when, realistically, when is the soonest you would begin enrolling for TAM?

01:11:57.740 You know, I'm waiting for something to know next week. And if next week goes well, so I just want

01:12:04.800 to let you know that we have an alternative. The alternative is non-for-profits. That the reason

01:12:11.440 they haven't joined us is that their charter is not to fund anything that the NIH funds. They want

01:12:19.240 something riskier. Of course. So if the NIH is not funding because they feel it's risky,

01:12:25.240 those will fund. So we can start anywhere. And it's, the answer is early 2019. Okay. We're basically,

01:12:33.960 all our centers are ready to go. How many centers? 14. How many subjects? 3,000. Age? 65 to 80. Any

01:12:43.440 limitation on comorbidities? No. In fact, well, there are a lot of limitations, but we're very

01:12:49.500 inclusive. But no existing diagnosis of cancer? No, unless they're cured. Okay. So you could be in

01:12:56.300 remission, but you can't have active cancer. Yes. Is it secondary prevention or primary of heart disease?

01:13:01.180 You could have cardiovascular disease, you know, in the past, if you're okay now. You could have

01:13:07.460 mild cognitive impairment and low speed walk will be a criteria. So in other words, we don't want the

01:13:15.520 people like in my study who will become centenarians because we're wasting time on them. We want people

01:13:23.700 who we know are kind of in the midst of aging. And we know that it's never too late to target aging.

01:13:30.040 So 65 years and up, five years studied, and you're going to randomize one-to-one or two-to-one?

01:13:37.920 One-to-one. And the placebo versus what dose of metformin? Two grams?

01:13:42.540 1,500. So, you know, the, so first of all, I should tell you that Merck company from Germany is

01:13:48.540 giving us the placebo and metformin. Right. So the drug is free. So when, when you say nobody's

01:13:53.520 interested, it's true, it's generic, but actually Merck has the worldwide license for metformin.

01:13:59.740 And they're actually giving supply to lots of studies because they kind of realize that this

01:14:06.140 is good. And do you think there's a difference between generic metformin and branded in terms

01:14:11.600 of efficacy? Is there any reason? No.

01:14:13.160 So they're really, this is kind of a, this isn't necessarily accretive to them because one can buy

01:14:18.980 metformin generically anywhere. They're not going to make more money from giving us the, well, you

01:14:25.020 know, I should say even now, a lot of the metformin suppliers have higher sales and a lot of them are

01:14:33.180 non-diabetic. So this thing, which is, which is bad for me because I don't want too many people to

01:14:40.860 know that, right? I want to do this study. I don't want, and I don't, I want people.

01:14:44.440 Well, you'll have, you know, I don't think hopefully it won't impair your, your recruiting

01:14:47.520 and the power analysis. I mean, it's funny, honestly, near, I would have guessed, but this

01:14:52.660 is, this speaks to the non-linearity and complexity of estimating power. You can't do it without the

01:14:57.500 tables. I would have guessed you needed a bigger end than 1500 in each group. You are using 80 or 90%

01:15:04.140 power.

01:15:04.820 90% power. But I should tell you, look for every disease, we have almost a 30%,

01:15:10.860 effect. And we, we chose 22% effect. Okay. For every disease, we have preliminary data for the

01:15:18.620 same.

01:15:18.760 Got it. So you lowered your threshold to call it 20% from 30%. And at 90% power, you've hopefully

01:15:25.960 powered this. You've got buffer both on your power and effect size.

01:15:29.860 Right. And, and again, we don't want to stop this study early either for one effect. So we carefully

01:15:37.580 thought about it. We asked for a grant for six years because we don't want to overdo it.

01:15:45.920 Okay. So it might finish in four or in six, but we have the flexibility to look at that.

01:15:54.100 Any non-U.S. centers? Are they all U.S.?

01:15:56.440 No, we decided to do only in the U.S. because of communication. And because look, outside of the

01:16:05.940 U.S., people came and said, you know, we'll fund Tain. But how, you know, if they say after two years,

01:16:12.920 we won't fund them, you know, it's very different. Because, because if it's a center, they need to be

01:16:18.560 part of part of the power and we cannot afford to lose them. So we, you know, those and other

01:16:24.240 consideration, we decided to stick with the United States. So let's go back to what we really glossed

01:16:31.560 over, but now I want to dive into. Why do you believe at the cellular mechanistic level, metformin

01:16:38.540 is an anti-aging drug? So let me take you through that in a, I don't want to say schematic way,

01:16:44.860 because we don't have this scheme here, but in a way that maybe I think will be easy to think.

01:16:50.320 So I want to start by saying that studies have shown that some of the effects of metformin are

01:16:58.520 through AMP kinase, and some of them are independent of AMP kinase. Okay. Is it important not for

01:17:07.020 metformin, but I don't know which are the ones who are relevant more for aging. Okay. And it's hard to

01:17:15.980 know, and it'll get complicated in a second more. So metformin gets into the cell through a transporter

01:17:24.180 that's called OCT1. Okay. This transporter is not equally, equal in every cell. So unlike

01:17:33.480 fenformin that goes with other transporters and affect everything, metformin is peculiar in this

01:17:39.980 way, but we have the preliminary data and it binds to complex one in the mitochondria.

01:17:47.520 Does it do so preferentially? Does it gain entry preferentially into the liver in humans?

01:17:53.220 Only in the sense that there's a nice concentration of OCT1 in the liver, but not, not, but it's not,

01:18:00.980 I don't think it's the top other organs have. So basically it's more about OCT distribution

01:18:06.340 and expression maybe than tissue specificity. Well, it's more that I want, I want metformin

01:18:13.540 everywhere. Okay. I want metformin everywhere. And I'll tell you how I kind of proved that.

01:18:21.260 So now metformin binds to complex one in the mitochondria. I'll tell you that metformin has

01:18:27.140 some action, epigenetic action that are independent of the mitochondria. You can do it with what's called

01:18:34.420 row zero cells. You can, you can deplete mitochondria and see and measure things and you can measure

01:18:39.800 things even without mitochondria. Okay. Again, do I need them? What type of epigenetic changes do you

01:18:46.680 see in the row cell? In histone deacetylation. Okay. So I cannot argue that this can be important.

01:18:54.180 It's not important if we use metformin, but it's important if you want to ask me,

01:18:58.480 are the mechanism of metformin AMP dependent, mitochondria dependent.

01:19:05.100 So there's an HDAC property that we may not have even thought of before.

01:19:09.140 Right. Okay.

01:19:10.300 So it gets to the mitochondria and in the mitochondria, it basically changes ADP, ATP ratio. And that's,

01:19:18.420 a lot of it is important in the liver, but not only to the extent that there is an activation of AMP

01:19:27.360 kinase. And Nir, is there a demonstrable reduction in NAD to NADH in that cell when complex one is

01:19:36.040 inhibited? Yes. Yeah, absolutely. Those studies have been, have been done. So there's a lot that's

01:19:43.420 going through AMP kinase. And I would tell you that downstream of AMP kinase, there is mTOR. Okay. And

01:19:51.280 it depletes, decrease mTOR activity and increase autophagy. And there's a whole pathway that will

01:19:57.280 go to the pillars of aging. You know, I asked David Sabatini this question a month ago and where he

01:20:04.080 doesn't know the answer, but I'm hopeful that maybe one of the postdocs in his lab will start to figure

01:20:08.760 this out. I'm very curious about what the dose equivalent is between, in terms of purely looking

01:20:16.700 at the readout of mTOR inhibition, what is the dose to dose equivalence of metformin via AMPK versus

01:20:26.580 RAPA directly? And I haven't, I mean, David didn't know the answer, which tells me nobody probably knows

01:20:31.420 the answer. Right. Look, it's, it's really difficult. First of all, for example, lots of the

01:20:36.900 non-AMP kinase activities are on much higher doses of metformin. Well, that's my point is,

01:20:42.760 can you clinically match them? No, because you, you really don't know at the end how much metformin

01:20:49.520 is in the cell. What we're doing now, we're doing isolated cells. In other words, we treated animals

01:20:56.660 with metformin and we're taking isolated cells to just see the variability of, of metformin in

01:21:05.300 individual cells. Even amongst homogeneous tissues? Yeah. Like we take hepatocytes and look at the

01:21:12.620 variability of expression. Yeah. Right. So it's all kind of a important question, but, but the story

01:21:20.420 is still, I'll tell you the story because at the end, I'm going to answer you what is that doing for

01:21:26.160 aging? Okay. And I don't want to rush this story because this is a great story. You take your sweet

01:21:30.920 time. So think about it in a scheme. You have the metformin in the middle, getting through the

01:21:36.580 plasma membrane, getting to the mitochondria. And on the left side, let's say there is the AMP kinase.

01:21:43.420 Okay. Now in the right side. Oh, sorry. Just for the listener, explain what AMP kinase does

01:21:49.240 at the high level. It's, it's a hormone of what it's a hormone of nutrient deprivation. And therefore

01:21:55.740 it tells the body to do what, like when you don't eat for a day, AMPK goes up. Why?

01:22:02.260 Well, but it's also an exercise mimetic pathway, right? Yes. That's what happens with, with exercise

01:22:10.060 really. That was the, the hope of what AMP kinase is doing, but it's a nutrient sensing that is in part

01:22:17.280 upstream of mTOR. Okay. But the other side of metformin is what happens to the mitochondria because

01:22:24.780 in a way, and I hope I won't regret saying it here, but metformin is a weak cyanide.

01:22:32.720 Okay. Look what cyanide acted at complex four. Right. But right. Oh, you just mean more broadly

01:22:40.180 speaking in inhibiting the ETC. Right. It inhibits. So there is less ROS production. Okay. And then

01:22:48.840 there's less inflammation and there's other things that's going just because the mitochondria

01:22:54.780 is less oxidative pathway going down. So this is a great example of why binary thinking doesn't very,

01:23:02.100 doesn't do very well in biology, right? It can't be all or none. Well, so now I can go on and on

01:23:10.420 and connect all those. Please do. To eventually the pillar of aging, but I want to insert another

01:23:18.720 thing because other things are happening. For example, insulin levels goes down, right? Why is

01:23:23.860 it just due to the reduction of hepatic glucose output? I'll tell you why I think. And then also

01:23:29.980 inflammatory factors are going down and there's a whole NF kappa B action of metformin that might be

01:23:38.880 part of the ROS and part could be independent even without mitochondria. Okay. It's that confusing.

01:23:48.180 But this is the point I want to make. I don't know which one of those is important for aging,

01:23:53.660 but parts of what you're measuring is the following. You fix the aging on a cellular level. Okay. So the

01:24:01.540 younger, the cells are younger. A lot of things are correcting themselves.

01:24:07.440 So I think at the end, the lower insulin levels, the low inflammatory are not necessarily a direct

01:24:15.680 effect. And that's why we're, we're fighting all the time about what metformin is doing, because

01:24:19.980 you can measure lots of things, but the things you're measuring are because aging was fixed.

01:24:25.080 And once aging is fixed, then there's whole hemodynamic readjustant or whatever you want to call it.

01:24:33.320 And then you're measuring that everything, and this is not only typical to metformin,

01:24:38.320 this is with rapamycin, with resveratrol, you also can see that there's a lot of things that are

01:24:45.420 improving themselves. And I think we have to get used to the fact that when you have a drug,

01:24:52.940 that we argue what it's doing, because everybody's measuring something, some of it is true,

01:24:58.740 but it's like secondary. It's, you fixed it. How did you fix it?

01:25:02.740 Yeah, it's true, true, and unrelated. And it's funny you mentioned those three,

01:25:05.980 which again, I think speaks to an advantage you have with TAME. So when you look at rapamycin,

01:25:12.800 resveratrol, and metformin, the big advantage of metformin is you already know the dose and the

01:25:20.640 frequency. With rapamycin, over dinner tonight, I'm going to give you my philosophy on the dose and

01:25:27.740 the frequency, and we can discuss it, but I can't demonstrate it to you with anywhere near the

01:25:32.340 validity that you could do the same with metformin. I had a wonderful discussion with David Sinclair,

01:25:38.320 and bioavailability came up on resveratrol, right? So maybe all the trials that say resveratrol is

01:25:43.220 meaningless are simply not looking at what happens if you get enough resveratrol in the right place.

01:25:48.100 And again, it's amazing to me how many times in biology we can make mistakes of the first order

01:25:58.020 because of a third or fourth order omission. And this, I guess I don't think I realized it until you

01:26:05.720 just gave that explanation, the value of effectively millions of patient years worth

01:26:14.020 of data on metformin, not just from a safety standpoint. By the way, billions, it's billions

01:26:19.240 of years. Is it billions? Yeah. So billions of patient year data on metformin. I know that from

01:26:24.900 an FDA standpoint, the highest priority is safety. That's a low bar to cross. The more interesting

01:26:30.620 one from the standpoint of this study is actual efficacy. Right. And it seems like if I'm hearing

01:26:35.680 you correctly, metformin is an amazing B student. Let's not be insulted by that. What do I mean?

01:26:41.120 Are there better ways to inhibit ROS if you want to hammer the ROS chain? Probably. There's clearly

01:26:47.320 better ways to inhibit mTOR if that's all you want to do. Are there better ways to modulate HDAC? Yes,

01:26:53.680 I'm sure there are. But it might be that maybe part of metformin's beauty is it does so many things

01:27:00.260 at a B plus level. Never it's an A student, but it's not a D student. And it's never, you know,

01:27:06.340 I mean, I'm sort of thinking about this sort of in a tongue in cheek way, but it's,

01:27:09.800 it seems to do so many things reasonably well. Well, I call metformin a tool from my perspective.

01:27:16.580 It's just a tool to show that we can target aging because I think that there'll be much better

01:27:23.880 drugs and combination drugs than other in the future. Human lifespan is 115 years. We argue.

01:27:30.400 Maximal lifespan. Yeah, maximal lifespan. We argue because there's somebody 122, but you know,

01:27:35.120 there's a statistic thing. So, okay, we die before the age of 80. So there are 35 years

01:27:40.940 that as a species without, not that we cannot change more, but just now without thinking too much

01:27:48.300 futuristic, long hanging fruit that we can do. And metformin is the perfect tool to start showing

01:27:56.100 the proof of principle here. But the future is much better. It's just, you have to,

01:28:01.900 to pave the road here. This is a beautiful pivot to something else I want to discuss with you,

01:28:07.020 because unfortunately I could talk about metformin for three hours, but there's so many other things

01:28:11.860 I want to discuss with you near based on your work. And you just alluded to one, which is the work that

01:28:17.900 you and your colleagues have done on centenarians has been very influential in my thinking. So as I think

01:28:24.640 about aging clinically, I ask the question, start with the existence principle. What does it mean

01:28:31.640 to live a long life? Is there proof of this having existed? And there is, we have centenarians,

01:28:39.320 they exist. My hypothesis, when I first started reading your literature, which was only about five

01:28:46.780 years ago, my hope, I should say, not my hypothesis. My hope was that whatever genetic benefit

01:28:54.160 they had, which it became very clear to me very quickly, that this wasn't about what they did.

01:29:00.460 In fact, they seem to have almost near immunity to the worst behaviors imaginable. My fear,

01:29:06.300 when I read your first paper, which was a review paper, I can still see it. And I still remember

01:29:11.400 where I was sitting in a Sheridan hotel reading it. My fear was whatever blessing they had,

01:29:20.620 it had nothing to do with chronic disease. And it had to do with something nebulous. And they had

01:29:26.800 complete immunity from chronic disease. And they just died in car accidents. Like they eventually

01:29:31.260 just died because something else tripped them up. And what I took away from your work was no,

01:29:36.220 their genetic gift was a phase shift. And when they got chronic diseases, they still died of heart

01:29:41.840 disease. They still died of cancer. They still died of Alzheimer's disease. They simply got a 20 year

01:29:46.440 bonus, if not more, maybe 29 on average, you could calculate, call it a close to 30 year bonus.

01:29:54.020 Well, of health span. Yes. Yes. Because completely functional life.

01:29:58.060 Because they had compression of morbidity. So it's not only that they lived longer and they lived

01:30:03.880 healthier, but they died. They died quicker.

01:30:06.920 Quicker at the end of that. And you know, it's interesting, the CDC, the Center for Disease Control,

01:30:11.840 have looked at the last two years cost of life of people, you know, with different stuff. So

01:30:18.000 they looked at people who die at 70 and after a hundred. And the cost of dying after a hundred was

01:30:25.800 third of that of dying in 70. So it's not only in our study, there's actually evidence.

01:30:32.980 Wow. I didn't know that.

01:30:33.920 To suggest that the medical, and by the way, those, when they were 70, they didn't go to the doctor.

01:30:39.820 Okay. So the, and, and this is the base of this concept that's called the longevity dividend.

01:30:46.520 Okay. What, what will happen to society if we'll actually be healthier for two and a half years?

01:30:51.420 And, and the benefits are immense. You think only of a, you know, so we pay social security more and

01:30:57.500 stuff, but there are seven trillions dollars just in saving of medical costs. If you could just

01:31:02.940 live healthier. So I know you don't spend an enormous amount of time anymore thinking about

01:31:08.060 the centenarians, you know, your literature follow it very closely. You, you definitely are spending

01:31:12.680 much more time talking about and writing about the things that we just discussed, but if you could

01:31:16.940 go back in time a little bit for me.

01:31:18.780 By the way, that's not true.

01:31:20.720 Maybe it's just selectively where I'm reading.

01:31:22.840 Well, no, I think that's, what's interesting to people now. And I'm very happy about that,

01:31:27.780 but I just got the, you know, the longevity, the Ibsen longevity price. And my talk was actually

01:31:33.080 about centenarians with a little bit of my hypotalamus stuff that was related to something,

01:31:38.360 but the centenarians are, and we have lots of grants on the centenarians because, you

01:31:44.400 know, the...

01:31:44.800 Is that talk, by the way, is it something that people can watch?

01:31:48.420 No, no.

01:31:49.380 It wasn't recorded?

01:31:50.080 It wasn't recorded.

01:31:50.700 Oh, it's a shame.

01:31:51.480 It's a real, I thought about, about this talk. It was fun for me to give because it was

01:31:55.780 reflective a little bit to, you know, one of the things that's always funny for me is people

01:32:00.980 say, oh, you know, I have, I can measure this thing. I'm sure that centenarians, they have,

01:32:06.680 they have high level because high levels are good. And I had to put into rest the fact that

01:32:12.060 the genetics of centenarians is really great for me. But the phenotype is not because you have a

01:32:20.600 hundred years old and you can measure something that can reflect what brought you here or the fact

01:32:27.260 that 30% of them are going to die in the next year. So it could be a marker of,

01:32:31.820 of death and not of longevity. Yeah. Their phenotype makes, doesn't interest to me. That's

01:32:38.940 why I have their offspring because their offspring will have the phenotype or half of their offspring

01:32:43.320 will have the phenotype. But those things were something that was nice because so many people

01:32:48.340 are saying, you know, we should measure in centenarians. No, we are, we're doing a longitudinal

01:32:51.800 study on their offspring and their offspring are equal everywhere to our control group. They just

01:32:57.220 have half of the diseases. Although you, by the way, that's a really interesting point, Nir. I never

01:33:03.280 once thought of that. I never, what you just said that if I have a hundred year old centenarian in my

01:33:09.160 study and he or she has a CTEP mutation or an APOC3 mutation, and I look at the predictable

01:33:16.700 phenotype because I know what you should have if you have that mutation. Hypofunctioning APOC3 should

01:33:22.640 have low triglyceride, but technically without a little more longitudinal legwork, I don't actually

01:33:28.340 know if what I'm measuring is a result of that. But that said, there must still be some benefit in,

01:33:36.660 you know what, now that I'm saying it, I realize you can't do it. The dream state would be to have

01:33:41.600 health records and phenotypes for centenarians that go 30 years back. In other words, it would be

01:33:46.180 beautiful to follow a centenarian from 70 to a hundred with phenotypic data. So that's why we

01:33:52.380 have the longevity study. It's a longitudinal study of offspring of centenarians and people without

01:33:57.820 longevity. And we're going forward, we're already 10 years into the study and we're starting, you know,

01:34:04.920 to see the differences between those guys. It's different population. They are aging slower. Okay.

01:34:11.120 The children of centenarians. So we have some of these things in mind. What are the most important

01:34:17.680 genetic differences between those either centenarians or offsprings of centenarians

01:34:23.600 and the rest of us? So it's very interesting. And so this is a part that you haven't heard so

01:34:29.640 from, but it's actually the growth hormone IGF axis. We have probably 60% of our centenarians

01:34:39.600 have genomic reasons. So let's explain the normal path of that. So the pituitary gland

01:34:46.560 makes GH. It tells the liver primarily to make IGF. Right. It binds to a receptor, right? In the liver

01:34:53.240 and IGF. And so some of the effects of the growth hormone system are through growth hormone

01:34:58.840 itself. And some of it is through IGF one level. Both of them are decreasing with aging. So in this

01:35:07.940 system, we identify, you know, more than 60%. And I think I'm underestimating. Okay. It's more,

01:35:16.080 it's probably more, but I didn't really calculate who has several mutations. You know, I can calculate

01:35:22.220 how many mutations there are, but not the overlap of the mutation. So I'm saying 60%, but it's the

01:35:27.520 most common genomic alteration in our centenarians. And I have to tell you- And where is it specifically?

01:35:35.160 I'll tell you in a second. And I want to just to tell you, I never, I wrote grants on that

01:35:40.820 because I thought, because there's preliminary data in nature, right? The small dogs live longer

01:35:47.700 and the ponies live longer. And when you mutate growth hormone or, or they're born dwarf,

01:35:53.920 they live longer. And when you give more growth hormone, they live shorter. So there is a lot in

01:35:58.160 nature, but the human data was really confusing. Some diseases, you know, when you have high IGF,

01:36:04.320 you have more cancers, but you have- Less Alzheimer's disease, less cancer. I mean,

01:36:08.520 sorry, less heart disease and less diabetes, but more cancer. Right. And mortality, in mortality,

01:36:13.500 it's like in the middle, you know, a little bit tending to more from, you know, from cancer.

01:36:18.600 By our analysis, which I think pulls all of your data, we found that there was an,

01:36:23.820 a U-shaped mortality curve with the nadir being between 60th and 80th percentile of all-cause mortality.

01:36:31.220 So I want to tell you the problem with this association. On one hand, an individual that

01:36:38.780 age quickly. Okay. Their growth hormone IGF goes down quickly. Okay. So you would measure falsely low

01:36:51.660 IGF-1 level. If you catch them on the wrong side of their aging curve. Right. That wouldn't fit our

01:36:58.120 theory that low IGF is good because that's their, their accelerated aging. Yeah. So this is a problem

01:37:05.920 with an association study like that. We discovered, first of all, that when we take our centenarians,

01:37:15.880 so they're, they're in our study because they've been healthy at age 95, living independently. We take

01:37:21.700 them at any age, but that's what they have to do. And we just look at those with the highest IGF and the

01:37:28.420 lowest IGF-1 level. Those with the lowest IGF-1 level live twice as long. Those are already the

01:37:35.420 centenarians, but only women. Men do not. And every example that I gave you, there's a sex

01:37:43.160 difference. The growth hormone IGF is very sensitive for women or for females.

01:37:48.900 Sorry, Nir, let me ask you another question. It's very difficult clinically to measure GH because

01:37:53.780 it's pulsatile. Right.

01:37:55.260 So whenever you're saying growth hormone, are you really saying IGF, which is easier to measure?

01:38:00.720 No, because I'm going to talk sooner, but not what I'm measuring, but where the mutations are.

01:38:07.800 Okay. But to repeat what you just said, the difference between men and women, was that based

01:38:12.640 on IGF level or GH level?

01:38:14.840 IGF levels.

01:38:15.700 Okay.

01:38:16.240 But we discovered in aging, we did a huge faux pas, okay, that we are now correcting. We,

01:38:22.800 we studied in our labs, males only. We had excuses why females are menstruating. We don't know

01:38:30.720 their effects. And we're all discovering, look, even rapamycin that really works everywhere is

01:38:37.760 better in females than males. But ACABOS is in males, right? There are examples where we totally

01:38:46.000 miss the sex effect. And growth hormone IGF is a perfect example, except those mice that are the

01:38:52.780 snail mice or the Ames miles. They have a pituitary problem, which makes them, I think,

01:39:00.560 sex confused. So, you know, they have so many other alterations, but sex difference is important.

01:39:07.320 We also showed that those women with the lowest IGF one level has much less cognitive problems,

01:39:14.260 you know, third of the cognitive deficiencies of those with the highest IGF.

01:39:19.000 All right. Okay. So we had to grab our second round of Topo Chico there, which by the way,

01:39:23.140 I didn't even ask you up front. This is your first time having Topo Chico, right?

01:39:26.880 Yeah. I'm in love.

01:39:28.120 I mean, this is, there are a few things that bring me as much joy as introducing people to this bottled

01:39:34.400 water.

01:39:35.060 The Perrier is much milder. I like to feel the bubbles.

01:39:41.280 Well, I would, I think one thing we should consider for TAME is if we can have an arm that also

01:39:47.240 includes Topo Chico. So there's a placebo, there's a metformin, and then there's a metformin plus

01:39:53.060 Topo Chico. And the question, the hypothesis will be, does that somehow enhance lifespan? Because even

01:39:58.900 if it does not, I predict it will enhance happiness.

01:40:01.640 Well, I have to tell you, I'm an advisor to the prime minister of Singapore. Singapore is a place that

01:40:07.160 can, you know, plan the future. And before going, I said, why don't you ask me questions? And one of

01:40:15.880 the questions, shall we put metformin in the water or in sodas? Now, good to ask, you know, the answer

01:40:23.560 is no, obviously, but it's good that you're thinking that way. So metformin...

01:40:29.040 And should you change your mind, I think you now know which soda it should go into.

01:40:33.700 Right. Can you get it in Singapore?

01:40:36.080 Oh, I'm sure we can, yeah. So back to sort of what you were saying, this IGF male-female

01:40:42.460 disconnect, let's keep going down that rabbit hole.

01:40:44.820 So I was saying that one of the worries about low growth hormone in IGF is the effect on the muscle,

01:40:50.700 because people think that growth hormone affects muscle function, muscle size. And we found that in

01:40:58.140 females and males, the IGF-1 level is not associated with better or worse,

01:41:04.860 lots of muscle function, including grip and getting out of chair and, you know, other things. So maybe

01:41:11.400 the effect of low IGF and aging are just as good as having a higher effect of IGF on muscle. It just

01:41:20.000 weighs itself out. Okay. But then our study is genetics. Okay. That's what we try to do. And

01:41:26.820 several years ago, almost, I guess, a decade ago, we found clusters of mutations in the IGF

01:41:37.240 receptor. Those were new mutations that haven't been found before, but we found them in nine of our

01:41:44.360 centenarians, which was 2% of our centenarians. And that was kind of a proof of concept for us

01:41:51.440 that clusters of mutations that are functional is what we really need to look for, rather than

01:42:01.260 like everybody's doing, doing GWAS and find, you know, something intragenomic somewhere.

01:42:07.020 And this was also the first proof of concept that the growth hormone IGF could be relevant to human

01:42:13.160 aging. People that those 2% of our, those nine people had higher IGF-1 level because they were

01:42:21.940 resistant, right? It's the IGF receptors. They were resistant to IGF. So their level was a little bit

01:42:27.120 higher, but they were significantly shorter than others. Which is a great explanation of how that

01:42:35.220 phenotype would play out. You would say, well, how can they be short statured with high IGF? And the

01:42:41.260 answer is if the IGF isn't as effective at the receptor. Right. But remember that most of our

01:42:47.460 people with low IGF are doing better. By the way, we take the people with the IGF- Sorry, men also,

01:42:53.360 or just women? Two men in those nine people. And by the way, then there are dwarfs, human dwarfs,

01:42:59.100 that are called Laron dwarfs, that live in Ecuador. And Hassi Cohen and Volter Longo and some of the

01:43:07.540 people in Ecuador were looking, basically were trying to find if they live longer, too few of them

01:43:14.320 to really know. But they have less cancer and less diabetes, you know, significantly less. So

01:43:21.300 there's other evidence from humans came after hours that, that there is at least less age-related

01:43:28.040 diseases in those people. They certainly didn't live a better life in any way that we could

01:43:32.200 assess, correct? Right. I mean, we always joked that they weren't happy with the fact that they're

01:43:40.840 short. So they drank a lot, and that's the alcoholism. And then when they crossed the road,

01:43:46.960 nobody saw them. So then- They died of more trauma.

01:43:50.500 But I think it's really not true. So I met some of them. We were in a Vatican conference. Everybody

01:43:58.080 has to bring his patients. And Longo brought somebody from Ecuador, and I brought a centenarian

01:44:03.820 from Rome. And the guy that he brought told a different story. He wasn't unhappy. He was also

01:44:11.740 happily married with a big woman and had children. So I don't know.

01:44:17.700 Is the defect in them at the- Growth hormone receptor.

01:44:21.700 In the liver. In the liver.

01:44:22.900 So they make a normal amount of growth hormone. Their liver doesn't acknowledge it.

01:44:26.640 A high amount of growth hormone because it doesn't work.

01:44:28.420 The feedback loop. Yeah. So they have lots of growth hormone, but it's not being expressed

01:44:33.120 through the liver into IGF.

01:44:34.400 Right. And so they have no IGF. So we published last year a paper that took us 10 years to write.

01:44:41.400 And basically, we were looking at a relatively common mutation. I'm saying common because it's

01:44:49.580 like 3-4% in the population where they have a complete deletion of exome 3 in the growth hormone

01:44:56.980 receptor. Okay. So one of the exomes, you know, one of the things that are important for the

01:45:03.560 integrity of this hormone, right, is deleted. And of course, you would think that that means

01:45:11.840 that the growth hormone is less effective. And Giletsmon, who was a fellow with me, brought me

01:45:19.560 this data to suggest that while it's 3-4% in our control population, it's 12% in our centenarians.

01:45:28.480 So I asked, so what's the IGF-1 level? And he showed it's lower. And I said, and what's the

01:45:34.880 height of the people? He said, well, that's the problem. They're much taller. So I said,

01:45:40.260 I cannot do anything with this study. It just makes no sense. I don't know how to write it.

01:45:45.860 So the people with lower IGF, but the mutation was in the exon of...

01:45:50.600 Of the growth hormone receptor. Okay. So they had lower IGF.

01:45:54.120 So it was a dysfunction of the receptor in theory. So less IGF. And we know that it was fully

01:46:00.540 penetrant all their life. In other words, there's no chance that it didn't start showing up until

01:46:05.300 they were an adult. Well, we'll come back to that. But the mutation didn't change.

01:46:10.320 And by the way, just give me some numbers here. How low are their IGFs?

01:46:13.440 I don't remember. Directionally. Like, are we talking about like less than 100?

01:46:17.900 No, less than 200.

01:46:19.160 Okay. Yeah. 100 is about the median level of IGF for somebody over 65.

01:46:28.900 Yeah. I think in the studies that I told you, the average or the mean or whatever was 94,

01:46:35.140 something like that. I don't remember to answer about those people. Okay. Now? Okay. So he said,

01:46:41.800 what to do? I said, well, first of all, what you do with genetics, you do replication study.

01:46:46.340 You go to other populations and you see what's going on there. And second, you do a functional

01:46:51.860 study. Let's see if it's really a functional mutation. So it took almost 10 years, right? And

01:46:57.880 in 10 years, we replicated the data in three other populations, in Amish, French, centenarians in the

01:47:07.180 CHS here in the United States. And in all studies, the people that live the longest had much more of

01:47:15.040 those homozygosities in the growth hormone receptor. So that was a great validation.

01:47:21.420 Second, the functional. So Hassi Cohen, my partner in USC, we send him the lymphoblasts of our patients,

01:47:31.360 the affected and not affected. And he is a growth hormone IGF expert. So he incubated them serum-free,

01:47:39.460 so without stimulation, and with growth hormone, and looked both at activation and proliferation

01:47:45.380 by growth hormone. So in serum-free, without stimulation, they had less activation and less

01:47:53.120 proliferation, half of normal, suggesting, yeah, that's the mutation. That's the function.

01:47:59.960 When he incubated them with growth hormone, it totally switched.

01:48:04.440 I'm sorry, but going back, you talked about proliferation, but was there a functional

01:48:09.420 difference in the lymphocyte in addition to its increase in number?

01:48:13.820 No, there's, yeah, the first was activation by phospho-ERC and some of the, okay, some of the-

01:48:21.660 But non-specific activation.

01:48:23.360 No, no, specific to what growth hormone is doing, but ERC, proliferation, it's a, you know,

01:48:28.520 it's a cell data.

01:48:28.960 Yeah, that's an easy one, yeah.

01:48:29.840 But when he incubated it with growth hormone, it was the opposite.

01:48:35.440 Rather than this low proliferation, low activation, it was high activation, high proliferation.

01:48:42.340 So what's going on?

01:48:43.740 Well, what's going on?

01:48:45.400 When do we have high growth hormones?

01:48:46.920 Through puberty.

01:48:47.480 So with high growth hormone, something has changed in the activation.

01:48:54.020 From low activation, it jumped to high activation for reason with the, you know, the molecular

01:48:58.220 mechanism, the real mechanism we don't know really, but that's what happened when you have

01:49:04.840 high growth hormones.

01:49:06.000 So they all were taller.

01:49:08.220 And then when growth hormone got down for the rest of their life, they shut off.

01:49:12.620 Okay, so, okay, so this is 12% of our people.

01:49:18.380 Then Yuxing Su, who's a geneticist that's working on this, she was interested in microRNA.

01:49:24.360 That's another, you know, epigenetic thing that comes.

01:49:28.240 We haven't been thinking about it, but those are RNA that comes from certain region of the

01:49:33.440 gene and they bind specifically to active region of the gene and they modulate it.

01:49:39.940 Well, 30% of our centenarians have clusters of microRNA that are overexpressed by a lot, by

01:49:48.660 40 times, you know, just immensely activated.

01:49:52.380 One of them, for example, microRNA 142, that's increased by 35 fold.

01:50:00.240 When you incubate it with cells, it prevents the phosphorylation of the IGF receptor, decreases

01:50:10.780 some other signaling, FOXO signaling, and some other stuff dramatically.

01:50:15.480 So there's a microRNA targeting of the growth hormone IGF receptor in about 30% of our centenarians.

01:50:23.220 Then 22% of our centenarians have mutation in FOXO3A.

01:50:28.320 That's a common mutation in centenarians.

01:50:31.940 So you start 2% and 12%.

01:50:35.160 Yeah.

01:50:35.180 Tell people what FOXO3A does.

01:50:38.160 FOXO3A is another, how would I say, just housekeeper.

01:50:43.220 Homeostasis, yeah.

01:50:45.540 Housekeeper that lets in good stuff and change things and put things in stop.

01:50:53.440 It's too complicated to go more, but it is part of the insulin IGF.

01:50:57.400 It's very important, the transcription factor at regulating homeostasis, cellular homeostasis.

01:51:02.320 I mean, that's not a great explanation.

01:51:04.600 But isn't FOXO the most prevalent genetic difference?

01:51:10.720 If you were to just isolate them by genotype, if you look at CTEP and C3 and FOXO and GH and IGF,

01:51:18.120 I mean, my recollection is that FOXO might be the single most prevalent of the...

01:51:22.180 Well, FOXO, I'm not sure that it's true, but FOXO3A is common in all centenarians' population around the world.

01:51:32.100 Not every population has the same mutations.

01:51:36.300 In fact, what we're doing now, when we have the exome sequencing of all our subjects,

01:51:42.900 you know, almost 3,000 exome sequencing,

01:51:45.580 the important thing to do is to assign them to pathways.

01:51:50.700 Because, you know, we're doing something really silly with genetics.

01:51:54.560 When we had this GWAS, we said we have million SNPs around the genome.

01:51:58.560 We'll find the diseases, every disease in the world.

01:52:00.820 And one of the stupidest things we did, we took one SNP at a time.

01:52:05.700 Okay, let's see if this SNP is significant.

01:52:08.080 But we're not built of one SNP at a time.

01:52:10.840 And in certain population, down this pathway, there'll be another SNP that will change the function.

01:52:17.460 So you need to...

01:52:19.860 Our analysis now is totally different than how we started.

01:52:23.700 But this, you know, I mean, I agree with you completely,

01:52:26.580 but I don't think most of the world is listening near.

01:52:28.680 I mean, if I had a dollar for every time one of my patients came with their, you know, complete sequence

01:52:35.820 and, you know, they want me to interpret them, I have the same discussion so many times.

01:52:42.480 Okay, you have 20,000 genes.

01:52:45.160 To my last counting, maybe 78 of them have a deterministic relationship with a disease.

01:52:50.780 You have none of them.

01:52:52.140 And we know you have none of them.

01:52:53.660 Because if you had any of them, we would know by now.

01:52:56.920 There's no chance you got here with Huntington's disease and we sort of missed it.

01:53:01.180 Or there's no chance you have some inborn error of metabolism that somehow got missed.

01:53:05.180 And then we get into this whole GWAS morass.

01:53:08.680 And, you know, it's hard to explain to people how multifactorial these issues are.

01:53:16.860 And I find myself, maybe you do as well, somewhat frustrated by these discussions and the overemphasis

01:53:24.000 on this genetic, you know, it's a little bit of the drunk in the streetlight problem, right?

01:53:29.980 When the guy is standing below the streetlight and you say, what are you doing here?

01:53:33.320 And he says, I'm looking for my keys.

01:53:35.000 And you say, did you drop them here?

01:53:36.480 And he says, no, but this is where the light is.

01:53:38.360 So let me not increase the complexity, but give you another ammunition, okay?

01:53:44.120 So we did, you know, now several years ago, because of funding, we did our 44 best, you

01:53:52.380 know, first centenarians, their whole genome sequencing.

01:53:56.480 So think about it.

01:53:57.420 We have a study without control.

01:53:59.200 We just have 44 centenarians and we do the whole genome sequencing.

01:54:02.840 And our question was, do centenarians have the perfect genome?

01:54:08.100 I mean, maybe one out of 10,000, they just don't have all this crap from the GWAS, those

01:54:13.800 SNPs for heart and Alzheimer's stuff.

01:54:16.460 They're just great.

01:54:18.240 And we went to this data set that's called ClinVar.

01:54:22.320 ClinVar then had the 15,000 pathogenic mutation that most probably will cause a disease, okay?

01:54:30.460 Now they have about 30,000.

01:54:32.840 And our hypothesis is that our centenarians don't have any of those, okay?

01:54:38.100 We'll support the perfect genome.

01:54:41.320 44 centenarians had more than 230 mutations between them, another five to six mutations

01:54:50.100 that should have caused them diseases.

01:54:53.280 And none of them had this disease in 100 years of life.

01:54:57.620 And some of those mutations, I'll give you the best example there for Parkinson and cancer

01:55:02.180 and everything, but ApoE4.

01:55:04.640 ApoE4, yeah.

01:55:05.660 What's the E4?

01:55:06.280 We have two ApoE4 100 years old people that the textbook would say they're demented at

01:55:12.780 70 and dead at 80.

01:55:14.840 And they're not demented and not dead at 100.

01:55:17.980 Do you think this at all explains the LPA uptick we see?

01:55:22.200 Because, you know, one of my favorite topics is lipidology.

01:55:26.000 And of course, LP little a is a very virulent lipoprotein.

01:55:30.500 And yet centenarians seem to have more of it than the general population.

01:55:33.580 Exactly.

01:55:34.260 So we actually showed in a paper because we found a way.

01:55:42.560 So basically, obviously, we can have mutations that are protected because they have slow aging

01:55:54.420 or longevity genes that protect them, right?

01:55:57.280 So how do you prove that?

01:55:58.520 So we noticed that when you look cross-sectionally, I hope I can make it very simply.

01:56:06.560 When we put any genotype cross-sectionally, in other words, we have all the ages from 50 to 112 is our oldest.

01:56:17.400 We see patterns.

01:56:20.080 If we see pattern that the genotype is declining with age, we know it's killing people, okay?

01:56:27.480 If we see that it's kind of monotonically increased with age, like with the-

01:56:32.200 Yeah, it's concentrating.

01:56:33.540 Yeah.

01:56:34.140 Then means the people who are surviving are surviving with this mutation.

01:56:38.480 That's why we have it more in 100.

01:56:40.340 Like the example of growth hormone receptor going for 3 to 12 in a monotonic way.

01:56:48.180 So we looked at LP little a exactly.

01:56:51.540 And this is the most confusing of them all, because I don't see any compelling evidence

01:56:59.000 that LP little a is anything other than atherogenic.

01:57:02.540 And so it begs the question, why would it concentrate in centenarians as opposed to just rise commensurate

01:57:09.920 with the population?

01:57:11.040 Exactly.

01:57:11.540 So this is the answer, okay?

01:57:13.500 First of all, we show cross-sectionally that LPA drops until the age of 80 by half.

01:57:20.940 In other words, it kills half of the people.

01:57:22.300 It's killing you like crazy until it's not.

01:57:25.660 Right.

01:57:26.640 And then you look at that and you kind of try to understand, just a minute, now this thing

01:57:31.820 that killed, now all of a sudden centenarians have even more.

01:57:35.540 Right.

01:57:35.760 It should stay flat after 80.

01:57:37.660 Right?

01:57:38.080 It should be flat.

01:57:39.100 Why is it increasing?

01:57:40.300 Well, because it's a protected aging gene.

01:57:45.840 In other words-

01:57:46.700 It was, obviously, during evolutionary times.

01:57:50.320 Are you saying that the thing that LP little a did for us 500,000 years ago that is no longer

01:57:58.440 beneficial in this lousy environment, the centenarians have managed to tap into its properties of-

01:58:04.420 No, no, no.

01:58:05.440 So what we did, and it's all published in computational biology journals, we took every one of the

01:58:13.540 longevity gene that we have and we did gene-to-gene interaction.

01:58:19.900 In other words, we tried to see if there's interaction between the bad genotype and the

01:58:26.380 good genotype, the good longevity genotype.

01:58:28.460 And we did it to a variety of those U-shape, right?

01:58:32.460 I'm telling you there's a U-shape.

01:58:34.580 Yeah.

01:58:34.680 You know, it goes down with aging and then it's good in centenarians.

01:58:37.380 Mm-hmm.

01:58:38.500 And we really found statistically significant that LP little A is protected by people who

01:58:47.620 are homozygous to a CTP mutation that is a longevity mutation that we found before.

01:58:53.640 That's the VV mutation, isn't it?

01:58:55.200 The VV mutation.

01:58:55.920 In other words, most of the people with the LP little A that are centenarians were also

01:59:01.780 CTP mutation.

01:59:04.720 So can we tell people what that phenotype is?

01:59:07.220 So I've talked on this podcast a lot about CTEP and its role in reverse cholesterol transport,

01:59:13.120 et cetera.

01:59:14.000 Talk to me about what does the lipid panel look like for someone who is CTP VV plus or minus

01:59:20.480 LP little A?

01:59:21.160 So those with CTP VV, they have higher HDL levels.

01:59:27.020 They have larger lipoprotein particle size and they have lower CTP levels.

01:59:35.160 That's their phenotype.

01:59:37.240 And this CTP has been protective against several age-related disease.

01:59:41.760 The most dramatic is cognitive decline.

01:59:43.880 And by the way, the two ApoE4 are also CTP VV carriers.

01:59:49.740 So really, the question is, it sounds to me like if you are a CTEP VV, that's a very

01:59:57.120 protective phenotype.

01:59:58.500 And the difference, if I remember the data going from 70 to 80, was basically flat.

02:00:03.520 If you're a CTEP VV, it really goes up when you turn 80.

02:00:07.620 Meaning the concentration of CTEP VV...

02:00:11.480 The genotype.

02:00:12.400 Yes, the genotype goes up significantly once you hit 80.

02:00:15.820 It goes in 55, it's 8% of the population.

02:00:20.640 And at 100, it's 20% of the population.

02:00:24.680 So it's also important to know that not...

02:00:27.540 First of all, the 8% that have VV, I don't know how many of them will be centenarians,

02:00:33.740 right?

02:00:34.080 And also, not all centenarians are VV.

02:00:37.360 But it's really, it's still very impressive.

02:00:41.300 Because the problem with HDL cholesterol is it's just such a dumbass metric.

02:00:44.820 It's so useless.

02:00:45.800 It's so far downstream that it doesn't tell us much.

02:00:49.040 But the hypothesis here would have to be that these people have far more functional HDL

02:00:54.240 particles.

02:00:55.300 Because, I mean, we don't have to rehash this, but anytime you pharmacologically increase HDL

02:01:01.240 cholesterol by inhibiting CTEP, nothing good seems to happen.

02:01:04.920 Right.

02:01:05.400 Because you're actually impairing the reverse cholesterol transport.

02:01:10.680 Right.

02:01:11.020 So it has to be with effect.

02:01:12.680 Right.

02:01:12.920 That's why...

02:01:13.220 It has to be a functional...

02:01:14.800 A functional thing of the HDL probably or the lipoprotein particle side or something else.

02:01:20.040 So I understand all of that.

02:01:21.180 Here's the part I still don't understand, Nir.

02:01:23.340 Why wouldn't the LP little a phenotype or the LPA gene flatline after 80?

02:01:31.600 In other words, I understand why CTEP VV is protective.

02:01:35.620 I don't understand why LP little a should concentrate.

02:01:38.520 I think because the CTP is common and is very protective.

02:01:43.960 And that's why you do it.

02:01:44.960 Look, it's a cross-sectional, okay?

02:01:47.540 By the way, it's a good question.

02:01:49.100 That's what we ask, why it's not flat.

02:01:51.500 But for a person who's born and is going to be a centenarian, for them, it's really not

02:02:02.140 so...

02:02:02.940 So they don't die when everybody dies.

02:02:05.240 I mean, I guess I need to do the math.

02:02:06.640 And we could do this over dinner.

02:02:08.160 We'll sketch this out on a napkin.

02:02:09.600 You take 100 people who are going to be centenarians, 100 people who are not, 8% of each of them,

02:02:16.520 call it 10% of each of them, have LPA.

02:02:18.540 Okay, we should do the math on if it's, are we being fooled by the age?

02:02:23.520 Because I do have one alternative hypothesis, by the way.

02:02:25.860 Oh, I'll show you, I'll show you dinner.

02:02:28.380 I'll show you the actual calculation.

02:02:31.040 The computation.

02:02:31.460 Okay, so this will be good.

02:02:32.720 This will answer my question.

02:02:33.920 Because the other hypothesis I've often wondered is, depending on, so I'll start with my question

02:02:40.000 and then I'll go to my hypothesis.

02:02:41.220 How much phenotyping have you done on the LPA?

02:02:45.400 Do you know how much heterogeneity is with their Kringle repeats, for example?

02:02:51.020 No, and I don't have their LPA level.

02:02:54.400 I have just their LPA genotype.

02:02:56.500 So the things that we've done are, though we can do, actually we've done a little bit

02:03:02.520 on the CTP levels, or we've done.

02:03:06.120 But those calculations are on gene-to-gene interaction, okay?

02:03:10.860 Taking out the people without genotype and without genotype and seeing who are the people,

02:03:16.600 who are the centenarians that stayed with the bed LPA genotype.

02:03:20.920 They were only the ones who are CTP also.

02:03:23.260 I see it.

02:03:24.540 Do you have serum stored on these subjects?

02:03:27.560 Yeah, and I have a proteomics done, and maybe we can have some measurement.

02:03:34.440 I'm not sure.

02:03:35.740 I wonder if there's a clue to a very vexing question, which is, why is it that some people

02:03:42.060 with high LP little a do not go on to get premature heart disease?

02:03:47.000 Many do.

02:03:47.880 Most do.

02:03:48.780 A significant number do not.

02:03:50.340 And it appears completely uncoupled from the level of their LP little a, which makes

02:03:56.400 me wonder, are there virulent and non-virulent phenotypes that at the level of expression

02:04:03.320 by number or molecular weight aren't captured?

02:04:06.680 And my question then becomes, is the non-virulent one the one that is concentrating in the centenarians,

02:04:13.740 and are they the population in which we should understand that phenotype so we can better risk

02:04:19.280 stratify the rest of us schleps walking around today, 10% of whom have elevated LP little

02:04:24.920 a, but we don't know how aggressively to treat them.

02:04:27.340 Yeah, I didn't know that there's virulent or non-virulent LP little a.

02:04:31.040 My explanation is totally gene-to-gene interaction.

02:04:34.220 In other words, you do another phenotype.

02:04:36.660 I guess the answer is, is different, is that the people with the, what's the LP little

02:04:42.660 a, what's their CTP level?

02:04:44.460 What's their HDL?

02:04:45.400 What's their HDL particle size?

02:04:47.880 And maybe you can see that's generalized that if you inhibit CTP, the LP little a is less

02:04:55.060 interactive.

02:04:55.620 I mean, I think the problem we, I agree with what you're saying.

02:04:58.360 The challenge is we do not have an HDL functional assay.

02:05:01.280 So right now we can measure HDL cholesterol, we can measure HDL particle number, and we

02:05:05.680 can measure HDL size, but none of those come close to telling us how functional the particle

02:05:10.600 is.

02:05:10.860 Well, I have a collaboration with Dan Rader, who's doing HDL efflux and stuff like that.

02:05:16.740 Yeah, I mean, Dan is a god in this field.

02:05:19.440 So maybe, but I'm, you know, that's not my, this is a past life for me.

02:05:24.180 Yeah, yeah.

02:05:24.560 I moved on.

02:05:25.940 So I don't, I don't really know to answer you.

02:05:29.320 I, I, I didn't realize.

02:05:30.720 Let's go back to the IGFGH thing, because this one still creates tremendous, let's bring

02:05:35.900 it back to clinical stuff.

02:05:37.200 So I'm often asked by patients, Hey, should I be on growth hormone?

02:05:40.940 And my answer is, I don't think so.

02:05:43.880 But if I'm going to be brutally honest, I really don't know.

02:05:47.080 I really have no insight into whether growth hormone is exogenous administered growth hormone

02:05:51.660 is harmful, helpful, neutral, or what?

02:05:54.960 Because the clinical data certainly don't give us the answer, right?

02:05:58.260 So the epidemiology, if you're in sort of, you know, Valter's camp is, that would be

02:06:02.880 the worst thing you could ever do.

02:06:04.000 Growth hormone is bad because IGF is bad.

02:06:07.000 But again, the epidemiology tell a different story, tell a much more nuanced story that

02:06:11.520 has to do with how high is too high at what age and for what gender and with respect to

02:06:17.320 what disease.

02:06:18.220 So of course, how can you manipulate IGF?

02:06:21.020 Well, you can manipulate it, the two easiest ways to manipulate IGF, the three easiest ways

02:06:25.440 to manipulate IGF are to manipulate growth hormone exogenously, manipulate IGF dietarily,

02:06:31.920 predominantly through amino acids, and manipulate insulin to indirectly impair or enhance IGF-binding

02:06:39.260 proteins.

02:06:40.200 Well, let me tell you, we just published a paper in Nature a month ago, a study that

02:06:44.860 we're going also for a while, and what we did is we got from EmGen, EmGen, as much as

02:06:52.100 other pharmaceuticals, tried to develop antibodies against IGF receptor because IGF receptor is

02:06:57.720 expressed in many cancers.

02:06:59.240 Yeah, this was the disaster drug that they tested against metastatic pancreatic cancer.

02:07:04.360 It failed.

02:07:05.460 But if I recall, there was a very special little gift, which is no CNS penetration.

02:07:12.700 Am I thinking of the right one?

02:07:13.780 Yeah, right.

02:07:15.460 So it failed, but we're interested for aging, right?

02:07:19.900 Yes.

02:07:20.220 And not only we're interested for aging, our hypothesis was that IGF, you know, that you

02:07:26.520 need to decrease IGF action in the periphery, but increase it in the brain because of the

02:07:32.680 data.

02:07:33.320 Yep.

02:07:33.640 So we asked them to murinize their antibodies.

02:07:37.940 Murinize means to make it available for a mouse.

02:07:39.860 Right.

02:07:40.920 And we got it and we did a longevity study that we started at 22 months.

02:07:46.640 By the way, remind me in the human study, how much did it lower IGF in the periphery?

02:07:52.240 It increased IGF.

02:07:53.000 I'm sorry.

02:07:53.540 I know it.

02:07:54.000 I'm sorry.

02:07:54.300 How much did it lower IGF activity in the periphery?

02:07:56.740 Was it a reduction by what fold?

02:07:58.420 Like, was it a twofold reduction?

02:08:00.000 It's very hard to measure.

02:08:02.020 You don't measure it like that.

02:08:03.220 You measure it in, you know, in vitro.

02:08:05.540 It's very hard to measure how much.

02:08:07.860 Got it.

02:08:08.120 Okay.

02:08:08.560 And they rebounded a little bit with IGF, which is not the case with aging because with

02:08:12.700 aging, you don't have the growth hormone secretion.

02:08:16.000 So we gave it to 22 months old animals that are 70, 75 years of age equivalent.

02:08:23.320 We increased their health spend dramatically and we increased their lifespan by 10% at an

02:08:29.080 old age.

02:08:30.420 Okay.

02:08:30.860 So this is a drug that's already was-

02:08:33.420 And you were able to demonstrate.

02:08:34.560 I'm sorry I missed this paper.

02:08:35.940 It came out a month ago.

02:08:36.740 So I'll be inhaling it this time tomorrow.

02:08:39.920 You basically showed that you didn't change IGF levels in the brain or you actually increased

02:08:45.480 IGF levels in the brain.

02:08:46.520 We really don't know.

02:08:48.080 There is a little increase in IGF-1 level in the periphery.

02:08:52.020 And we're assuming that we didn't make it a big, I'm telling you what we thought.

02:08:57.460 We didn't make it such a big deal, but obviously the antibodies do not cross the blood brain

02:09:03.380 barrier, but the IGF wasn't lower if anything was high.

02:09:08.580 So if anything, IGF went up in the periphery because it was being blocked.

02:09:12.300 And therefore a little bit higher in the periphery.

02:09:14.140 So you could have had more CNS activity, less peripheral activity of IGF.

02:09:18.560 Now, again, those were aging.

02:09:20.480 By the way, it was in females.

02:09:21.900 In males, we started to do the studies in males.

02:09:24.740 And since we didn't see any major effects and we had relatively little, you know, those

02:09:32.360 are months of studies in many mice.

02:09:34.560 So we eventually did the longevity only in females.

02:09:37.720 We don't have data on males, but the extent of health span, and you'll see the paper.

02:09:43.160 It's really, it's a great paper with lots of studies.

02:09:46.420 And it's impressive how much they did better, the females with that.

02:09:51.080 So there you have an example.

02:09:53.240 We have all the paths where you ask, what do we do?

02:09:55.480 Here is, it's been in humans.

02:09:57.920 Okay.

02:09:58.540 The IGF receptor antibody is there.

02:10:01.020 The only problem for us is aging could be an indication what's before aging.

02:10:07.020 Like we need another indication in order to start selling it before.

02:10:10.400 But what's your hypothesis in those mice, in the female mice, that they, you said they

02:10:15.500 live 10% longer.

02:10:17.420 And much healthier.

02:10:18.380 And much healthier.

02:10:19.240 So they had compressed morbidity as well.

02:10:20.820 They had a cardio, you know, they had the cardiovascular protection.

02:10:24.940 They have cognitive advantage, functional advantage.

02:10:28.560 There's a whole thing.

02:10:29.380 The interesting thing between male and female, the difference is the mice, the female mice that

02:10:37.620 were treated with IGF receptor antibody had a lot of, of inflammatory markers, right?

02:10:45.120 When we started, they were 22 months.

02:10:47.700 And those inflammatory markers were really decreased with aging.

02:10:52.760 In males, we had a little bit more inflammatory markers, but they increased with the IGF treatment.

02:11:01.200 So maybe the fact.

02:11:03.660 And both the males and the females had an increase in peripheral IGF.

02:11:07.520 Presumably the antibody was working in both of them.

02:11:09.740 Right.

02:11:09.840 But somehow the inflammatory response in males was different.

02:11:14.300 So it's all driven.

02:11:16.000 So I'm answering you two things.

02:11:17.140 First of all, what do you do?

02:11:18.160 Here's a good example of a drug that was in females.

02:11:21.400 And we have preclinical proof of concept that it's good.

02:11:26.180 But the second is in males, all those have not been shown.

02:11:32.000 And in fact, you know, the functional that I told you functional in women, it looks like

02:11:37.320 the function was better for men with a higher IGF and not with lower, not statistically significant,

02:11:45.380 but you could see that it might happen.

02:11:47.900 So I'm telling you from human studies, from genetic studies, I'm telling you growth hormone

02:11:53.480 treatment is not beneficial.

02:11:56.280 It should be dangerous for elderly people, for elderly women.

02:12:00.920 With men, I'm ready to say that I'm not sure, you know, I'm not sure.

02:12:06.820 It's not that I've shown that low IGF or growth hormone is better in males, not in my rodent

02:12:13.820 studies and not in my human studies.

02:12:16.520 But it's hard to say that giving growth hormone is bad when, you know, by the way, the growth

02:12:23.640 hormone receptor story is a lot of them are males.

02:12:28.000 That receptor.

02:12:28.640 This is so interesting.

02:12:30.120 I mean, this is such a layer of sophistication to this question, which I've largely sort of

02:12:35.980 decided that I can't come up with a compelling reason for exogenous growth hormone.

02:12:41.480 But I also can't see much evidence clinically that it's killing people.

02:12:45.460 But of course, I think disproportionately the data are in men.

02:12:48.720 What do you think explains this sex difference between men and women?

02:12:51.940 I think I'm really, if you look at this inflammatory panel, lots of cytokines in lots of animals,

02:12:59.280 I'm not sure, we don't have the intellectual link to say what happens, but this is something

02:13:06.340 we want to actually go on and examine.

02:13:08.820 Because if for some paradoxical way that is sex dependent, low IGF, increased inflammation,

02:13:18.160 we want to know why and we want to know how we can affect it.

02:13:24.220 It would be great if you could repeat the study with younger mice, both male and female again,

02:13:30.500 because, I mean, the most obvious example is the biggest difference between the men and

02:13:35.360 women and women are the sex hormones, and the older they get, that difference is still there.

02:13:40.920 I mean, women will, you know, the female mice would go from having some testosterone to zero

02:13:46.700 testosterone.

02:13:48.300 But, gosh, the men would probably still have more estrogen than the, the males have more

02:13:56.040 estrogen than the females post-menopause.

02:13:58.080 Which makes me wonder if the estrogen, the progesterone, and the testosterone are somehow

02:14:02.440 protective of the inflammatory effects.

02:14:06.440 So there's a, you know, we're working on a grant, because the sex issue is so interesting,

02:14:14.360 we started working on there, lots of models where you can change X and Y chromosomes and stuff like

02:14:21.380 that, and, you know, we're looking at the way to look at the, because it's, I don't think it's

02:14:26.660 usually as simple as the sex hormones, you know, I think we're missing a whole biology,

02:14:33.160 a whole biology, and whatever is the sex hormones, I don't understand the inflammation in that either.

02:14:39.900 But by the way, your idea of doing it in young is good.

02:14:42.820 Mgen sold this IGF receptor and we cannot get, we cannot get more samples, but we're interested

02:14:49.800 to maybe do our own.

02:14:51.200 Wait, wait, Mgen sold the rights to the human antibody.

02:14:55.060 Right.

02:14:55.780 Their whole section, actually.

02:14:58.060 Their whole section.

02:14:59.200 To who?

02:14:59.580 Which company owns it now?

02:15:00.900 To like a private company that we're unable to get people to, though I have to say it's

02:15:06.840 not our major priority now, but, and we have some consultants that are trying to get it

02:15:13.080 together with those people.

02:15:14.500 So, you know, it's still an issue, but I'll tell you, if I had to develop a treatment,

02:15:18.840 I would rather use the micro RNA 142 because the IGF receptor antibody didn't work in cancer,

02:15:28.420 but maybe the micro RNA penetrates better in another way.

02:15:32.800 And maybe that's the way to decrease IGF receptor in, you know, in cancers.

02:15:38.320 So anyhow, but I really answered you that, yeah, there are things that we could do to affect

02:15:43.740 the pathway.

02:15:44.680 I mean, what do you think about sort of all the, to me, the most obvious way to manipulate

02:15:48.000 IGF is through fasting.

02:15:49.220 So about once a quarter, I fast for a week with just water only, and it has a profound

02:15:55.540 impact on my IGF level.

02:15:57.380 So for my age, I think I should know these numbers off by heart, but plus or minus two

02:16:03.820 standard deviations of IGF at my age is about 90 to 250.

02:16:10.460 And if I take my, if I measure my IGF level before a fast, right before a fast, it might

02:16:18.340 be 180 to 200.

02:16:22.680 And right after the fast, it's 80 or 90.

02:16:27.800 And six weeks later, it's maybe 140, 150.

02:16:30.840 So it sort of falls precipitously during the fast and then slowly rises and then falls

02:16:37.260 precipitously and slowly rises.

02:16:38.640 And there's part of me that just wonders if a cyclic approach to IGF is a healthier approach

02:16:44.560 than say, constitutively being calorically restricted and just, you know, because you

02:16:49.260 could starve yourself of methionine and eat no protein and live at a lower IGF level of

02:16:54.400 maybe 110 forever.

02:16:56.900 But I wonder if the real game is sort of figuring out this sort of cycle and periods of, just

02:17:02.360 as we think of autophagy, if you're always in a state of autophagy, that's a bad thing.

02:17:07.340 If you never have autophagy, that's a bad thing.

02:17:10.200 So you know about the calorie or the caloric restriction that the NIA funded in three centers?

02:17:17.040 Yeah, Eric Ravison at Pennington was one of the main PIs.

02:17:19.820 One of the things is that IGF level wasn't decreased in those people, right?

02:17:25.140 And we think IGF decrease is really an important part of longevity.

02:17:29.940 But we learned something else.

02:17:33.200 And maybe I'm digressing because I'm sure you talked about it more, that what we're doing

02:17:37.500 in rats wasn't really caloric restriction.

02:17:40.840 We're doing intermittent fasting because in our rodents, we would bring the food in the

02:17:46.280 morning, they were hungry, they would eat all the food, you know, the 60%, whatever we

02:17:52.980 gave to it, Libidon, and they were fasting for 23 hours.

02:17:58.020 And they had low, you know, low IGF-1 level.

02:18:02.700 But in Eric Ravison's studies, they didn't have low IGF because it wasn't caloric restriction

02:18:07.320 like that.

02:18:07.760 They gave them just less food throughout the day.

02:18:10.000 Okay.

02:18:10.420 And I think that was a big mistake that we realized we were doing.

02:18:14.580 And it doesn't answer the circulating.

02:18:16.640 I think what we're trying to do at Einstein is we're doing a study.

02:18:22.040 We know that autophagy is improved quite fast in fasting rodents.

02:18:28.620 We don't know the timeline for humans.

02:18:32.020 Okay.

02:18:32.940 So we're trying to figure it out.

02:18:35.780 And by the way, we can do with Anna Maria Cuervo, we can take T lymphocytes and look at the

02:18:41.520 autophagy on a time course in humans.

02:18:44.260 And they actually reflect also autophagy in the brain.

02:18:47.200 There's a lot of advantage in doing that.

02:18:49.280 What are you looking at specifically in the assay?

02:18:51.800 Well, there are several things because there are several autophagies, right?

02:18:55.980 So macro autophagy or chaperone mediated autophagy.

02:19:00.820 So there are several things.

02:19:02.500 Can you measure mitophagy also?

02:19:04.020 So yeah, we can do, we do that.

02:19:06.420 Yeah.

02:19:07.100 How sensitive is it to meal timing?

02:19:09.760 You know, like for example, if a rodent is fasting for 24 hours, especially a mouse,

02:19:17.480 I mean, that's, gosh, that's probably the equivalent of a human fasting for a week.

02:19:22.140 Presumably autophagy is highly, highly, highly upregulated.

02:19:25.620 If you give the mouse a little bit of chow and then measure that assay, does it obfuscate,

02:19:32.240 does it erase the fact that they've been, does it create the illusion, I guess I should

02:19:36.420 say, that it's erased, that they've had this high period of autophagy?

02:19:40.440 Did you transiently truncate it?

02:19:42.020 You know, Anna Maria Cuervo did, and I don't know the answer.

02:19:45.400 I don't know the time course.

02:19:47.760 And, but the important thing is to find the time course in humans so that we can really

02:19:53.160 say, for example, what I'm doing basically is I'm trying not to eat after dinner until

02:19:59.540 like lunchtime the next day.

02:20:02.440 So.

02:20:02.640 But do you really, I mean, so first of all, how hard is it to get an IRB at Einstein to

02:20:06.320 study this?

02:20:06.740 No, no, no, we have IRB, we just wrote a grand, we need money for that.

02:20:11.480 Okay.

02:20:12.000 Yeah, I guess, you know, I have a whole framework around nutrition, which I'm happy to kind

02:20:15.640 of walk through as we inhale our dinner tonight.

02:20:18.900 Have you decided, by the way, if you want Indian, Turkish, Persian, Greek, all of the

02:20:25.740 above?

02:20:26.180 Yeah, that sounds good.

02:20:27.100 You know, because it's raining out, so I guess one of the things we could do to the closest.

02:20:32.280 We'll figure it out.

02:20:33.200 Yeah.

02:20:33.380 So my concern with time-restricted feeding, which I practice quite liberally, is I'm worried

02:20:39.760 it is not a significant enough deprivation of nutrients in humans.

02:20:44.360 In other words, I think Sachin's data is so impressive in mice, you know, but I think that

02:20:50.420 for a mouse to go 16 hours without eating is an enormous task.

02:20:55.160 It might be the equivalent of us not eating for three days.

02:20:57.820 And now that said, I'm struggling to see a downside of time-restricted feeding.

02:21:03.340 And I know that, you know, Valter and others have said time-restricted feeding is somehow

02:21:06.900 bad.

02:21:07.700 I don't accept those arguments.

02:21:09.400 I'm not convinced by those data.

02:21:10.900 But this is a very interesting question you're asking.

02:21:13.460 I would argue it's the most interesting question of all.

02:21:15.700 Because if we understand the time course of that, all of a sudden we can program nutrient

02:21:22.520 exposure.

02:21:24.060 And have you looked at that assay in the presence of rapamycin or metformin?

02:21:29.380 I didn't.

02:21:30.140 But metformin, Ana Maria Cuervo is using metformin as the control for autophagy.

02:21:36.560 Okay.

02:21:36.820 It activates autophagy really well.

02:21:40.020 Okay.

02:21:40.240 So that's what I'm talking about in vitro assays.

02:21:43.620 Okay.

02:21:44.120 But we haven't measured it in patients.

02:21:47.800 The one thing I should say, I just don't want to forget anything.

02:21:51.840 Can I?

02:21:52.600 Absolutely.

02:21:53.700 And it's back to the metformin.

02:21:55.680 It just reminded me, it's back to the metformin studies.

02:21:58.200 So our reviewers were saying, you know, okay, all the data on the biology that you showed

02:22:06.100 was from rodents.

02:22:07.780 And we don't know that the biology is relevant to humans.

02:22:11.720 Okay.

02:22:12.440 Although, of course, the preliminary data in humans are much better than in rodents, right?

02:22:17.420 The association with diseases.

02:22:19.220 So we had a clinical study, a small clinical study, which we took 15 people that are 75

02:22:26.720 years old, and we gave some of them metformin for six weeks and other placebos, and then

02:22:34.600 we crossed over and it was blinded.

02:22:36.980 And we took it at the end of each period, we took biopsy from their muscle and biopsy from

02:22:42.860 their adipose tissue.

02:22:44.320 Okay.

02:22:45.100 So not liver.

02:22:46.940 And we looked at the transcript and the metabolomics, but mainly the transcripts in the tissues to

02:22:54.460 see effects on metformin.

02:22:55.960 And there are three interesting things.

02:22:59.560 First of all, when they brought me their clinical results, you know, insulin, HOMA, glucose, and

02:23:06.520 all, there were significant effects in many of them, statistically significant effects,

02:23:11.980 but I was expecting more.

02:23:13.720 I started to try and break it down.

02:23:15.840 And then we found out that the people who were last on metformin, although it was half of

02:23:21.960 the people, had much more significant results than the whole 15 people.

02:23:28.180 Wait, explain that again.

02:23:29.800 The people in the crossover who started, who went placebo to metformin.

02:23:34.020 Had better effects on metabolic parameters.

02:23:38.360 Relative to themselves or to their peers?

02:23:41.360 Relatively to themselves.

02:23:42.560 Because this was a person to person.

02:23:45.280 This was a person to person.

02:23:46.900 This was a paired T test.

02:23:48.620 And the reason is that probably the two weeks washover is not enough.

02:23:55.560 When you fix the aging, it lasts for a while.

02:23:58.700 Okay, so if you got metformin and then placebo, you didn't really, you went halfway back.

02:24:06.760 Also, by the way, because metformin is associated with weight loss and they lost a little weight,

02:24:11.380 that on itself really made them on the second part.

02:24:15.220 Yeah, it's funny you bring that up.

02:24:16.560 I'm glad you mentioned that.

02:24:17.860 What do you think explains the weight loss phenotype of metformin?

02:24:22.020 So when I started taking metformin, which was 2010, I didn't mess around.

02:24:27.620 I just went straight to 2.5 grams a day.

02:24:30.400 I didn't even mix it up.

02:24:31.820 I don't think I took it BID.

02:24:33.360 I mean, I took 2,500 milligrams first thing every morning.

02:24:37.960 And after a month, I lost quite a bit of weight.

02:24:40.660 But I was also nauseous 24-7.

02:24:43.280 I mean, the thought of eating was repulsive to me.

02:24:46.220 I still ate.

02:24:47.320 You know, I was exercising a lot and doing all my usual shenanigans.

02:24:50.380 But I clearly ate less just because of that low-grade feeling of nausea.

02:24:54.740 Today, when I prescribe metformin to patients,

02:24:57.620 I have them start at 500 QHS, then 500 BID,

02:25:01.480 then 500 in the morning, 1,500 before bed,

02:25:04.420 then one gram BID is sort of standard dose.

02:25:07.980 And I see far less of that nausea, and I see far less weight loss.

02:25:13.200 What other mechanisms do you think it could explain the weight loss?

02:25:15.860 So could it be lactic acid going up?

02:25:19.060 You know, just like ketones, you get a little bit-

02:25:21.340 An additional substrate, an additional substrate.

02:25:22.940 Yeah.

02:25:23.580 I don't know.

02:25:24.080 But look, I started metformin several years ago because I was pre-diabetic.

02:25:28.880 My doctor did it.

02:25:30.200 So it's before.

02:25:31.500 I'm glad that he did.

02:25:33.140 But it was before I thought much about it.

02:25:36.000 And I was just surprised after three months to realize that I lost, you know,

02:25:42.580 seven or eight pounds.

02:25:43.800 I was surprised because I couldn't-

02:25:46.320 I wasn't nauseated.

02:25:47.540 I wasn't anything.

02:25:48.640 I must have eaten less.

02:25:51.180 I was less hungry.

02:25:52.560 I reacted better.

02:25:55.160 I reacted better to my body, right?

02:25:57.360 To my leptin, maybe.

02:25:58.740 I reacted better.

02:26:00.260 And I probably didn't eat as much, totally not noticing it.

02:26:06.080 It was a surprise to me.

02:26:07.860 It's funny.

02:26:08.340 I tell patients not to expect that because I don't like the idea of patients thinking

02:26:12.480 of this as like a weight loss drug.

02:26:14.080 To me, that sort of is not the right way to think about it.

02:26:17.120 But I've never had a great explanation.

02:26:18.660 And by the same token, I have had, I should be clear, I've had also some patients who

02:26:23.140 have, they've come back and they've lost 10 pounds.

02:26:25.180 And these are patients who are not overweight.

02:26:27.120 This is not, you know, this is someone that by everybody's standard would be completely

02:26:31.520 normal weight.

02:26:32.620 You know, BMI is 24, body composition is reasonable, and comes back in, you know, in

02:26:37.980 three months and they've lost 10 pounds.

02:26:40.180 But I guess for in my, at least in my practice, that seems to be the exception and not the rule.

02:26:44.560 But very, very interesting to hear.

02:26:46.820 You know, I guess, of course, you wish that was more than a two by two, two week.

02:26:50.800 You wish that was a 12 week by 12 week.

02:26:53.040 We are doing the same thing with Acrabos now.

02:26:55.260 And we're doing longer study with longer washing period because we want to get away from

02:26:59.540 that.

02:27:00.220 But back to metformin, I said that's one thing is this metformin last issue.

02:27:05.440 The second thing that we showed that most of the transcript changes were relative to

02:27:10.560 this, to the tissue.

02:27:11.800 You know, in fat, it was more free fatty acid metabolism and the muscle was more pyruvate

02:27:16.460 metabolism.

02:27:17.220 So it was appropriate.

02:27:18.660 In other words, there wasn't the metformin, you know, the metformin part.

02:27:23.180 But in both tissues, there were genes that are not metabolic genes, you know, like BRCA1 or

02:27:30.880 myofibril genes and other things that are related to aging, but they are not metabolic.

02:27:37.480 They change by metformin.

02:27:38.860 In other words, the concept that metformin is not only metabolic, it's aging, okay?

02:27:44.120 BRCA1 changed?

02:27:45.580 Yeah.

02:27:46.060 You know, some other gene, genes that are associated with DNA repair changed significantly.

02:27:54.880 The second thing-

02:27:55.760 Did you see in the muscle, by the way, if you still have biopsies, I'd be very curious to

02:28:00.320 know if the lactate transporter out of the muscle, I'm blanking on it, it's MCT2.

02:28:07.920 In other words, as people make more and more lactate in the muscle, do they get more and

02:28:12.240 more efficient at shuttling it back to the liver?

02:28:15.200 And if so, you would expect MCT2 transport-

02:28:18.320 And is the gene expression changes by that?

02:28:20.960 Yeah.

02:28:21.280 You know, we have the raw data.

02:28:23.080 It didn't come up, I'll tell you, it didn't come up as a winner, but that doesn't mean it's

02:28:26.960 not changed.

02:28:28.340 And you looked at genome or you looked at messenger as well?

02:28:32.080 You looked at mRNA?

02:28:33.020 Only message.

02:28:33.340 It's transcript.

02:28:34.580 It's only message.

02:28:34.920 It's all mRNA.

02:28:35.560 So you should have seen it if it happened.

02:28:38.100 Well, if it was highly, I mean, there are hundreds of changes, okay?

02:28:42.500 So significant changes-

02:28:43.560 Yeah, maybe this was underpowered to see that difference.

02:28:46.180 Yeah, significant changes.

02:28:47.140 So I can still answer you, I can still answer you specifically.

02:28:51.620 And it's RNA-6, so it's really good.

02:28:53.980 It will be good.

02:28:55.000 You know, if I take that and see that it can be, without taking into account that there

02:29:00.600 are 400 other changes, it can be significant.

02:29:04.080 The other thing, though, what you do with transcript, you look at upper regulators of the system,

02:29:11.260 and there's a way to do that.

02:29:12.900 That's what everybody does now.

02:29:14.220 So if you look at the upper regulators, you get back to the same things that are affecting aging.

02:29:20.500 AMP kinase, mTOR, all those pathways were affected by metformin, except that what you measured below is the tissue-specific effects.

02:29:30.220 But it was all related back straight to aging.

02:29:33.600 You know, it breaks my heart.

02:29:36.660 And so we really connected not only the clinical data, but the biological data in humans to the TAMES study.

02:29:45.140 Yeah, it's wonderful.

02:29:46.540 You know, a few years ago when my father was diagnosed with cancer, I immediately put him on metformin.

02:29:54.100 And, you know, in his infinite wisdom, his primary care doctor took him off it because, you know, why would you put him on metformin?

02:30:01.820 And in the end, my father decided his primary care doctor knows much more than I do, and so remains off metformin.

02:30:09.300 But certainly this discussion, because I, you know, I probably know more about metformin than the average person,

02:30:15.040 but this discussion has been completely illuminating.

02:30:17.560 There's one other thing I want to ask you about, which is all of the discussion around NAD.

02:30:22.060 I mean, you know quite a bit about this.

02:30:24.260 What is your take right now based on the state of the science that we have,

02:30:28.560 which admittedly I think we're just scratching the surface of, as far as first and foremost,

02:30:35.000 just at the mechanistic level, what is your belief that orally administered nicotinamide riboside

02:30:41.040 can actually make it into a cell?

02:30:43.520 I don't know if you've seen it, but Josh Rabinowitz's paper over the summer would suggest,

02:30:47.460 no, most of this is going to the liver.

02:30:48.880 So wouldn't this sort of call into question the companies, there's two of them right now,

02:30:54.940 Chromadex and Elysium, that sell basically nicotinamide riboside plus or minus terastilbene.

02:31:03.620 What do you think explains the reported efficacy of those agents in light of Rabinowitz's NAD tracer study?

02:31:10.480 First of all, the Rabinowitz study was also criticized, okay?

02:31:16.640 Time course and other things, which, okay, I'm just saying that.

02:31:22.320 But I think really the biggest issue here, I want just to use this opportunity to say

02:31:28.680 that a lot of what we're doing is not based on clinical studies, okay?

02:31:35.700 And it really just underlines the fact that unless you do a clinical study, you can remain guessing.

02:31:45.840 Now, I'll tell you, I'm taking a good preparation of NMN, okay?

02:31:51.340 Tell the listener the difference between NR and NMN, because they're both precursors to NAD,

02:31:56.340 but there's a subtle difference.

02:31:58.620 Yeah, and I don't know.

02:32:00.180 I don't know to explain that really well.

02:32:02.480 I don't care that much about that.

02:32:05.920 Okay, but in theory, one is slightly more stable than the other.

02:32:08.440 Right, it's a stability issue, availability issue, but they'll fight.

02:32:12.460 And we don't know which is the better, and we don't know which gets to the tissue, right?

02:32:16.440 Yeah.

02:32:18.120 But what I wanted to say is the following.

02:32:22.100 So the answer is it could all be water, okay?

02:32:26.640 It could all be this.

02:32:29.160 You say that as you hold up a bottle of Topo Chico.

02:32:31.520 Right, sorry.

02:32:32.680 You didn't see that in the microphone?

02:32:35.060 Okay, but what I'm doing now is I'm opening my Fitbit, okay?

02:32:39.960 And one of the things that the Fitbit has is sleep, okay?

02:32:45.300 Do you use Fitbit?

02:32:46.660 No, I use something called Aura, which works much better.

02:32:49.660 Okay.

02:32:50.060 Yeah, this really measures sleep well.

02:32:51.740 Do you get also the deep sleep and the REM and all that?

02:32:56.120 Since I started taking NMN, and it might be a coincidence, so I have to stop it and start again, but my sleep has been much better.

02:33:08.100 That means I have, in the beginning of the night, I have much more deep sleep, and at the end of the night, I have much more REM.

02:33:14.400 And I thought, if there's anything I can say about it, is that just because of this association, okay?

02:33:22.820 But I don't believe it, right?

02:33:24.580 Because it's not a, it's me.

02:33:27.520 It's empirical.

02:33:28.440 Many things could have looked.

02:33:29.620 I was in vacation.

02:33:30.880 I don't know.

02:33:31.440 Many things could have happened.

02:33:32.940 Do you take the NMN by itself, or do you also pair it with terastilbene or another sirtuin activator?

02:33:38.780 Just NMN, but it's a good suggestion, maybe.

02:33:43.220 Okay, but then I run into Imai.

02:33:46.500 Do you know who's Shin Imai?

02:33:49.280 He's from St. Louis, and he has his own patents and studies in Japan on NMN.

02:33:58.040 And he tells me, unprovoked, he says, you know, I have 200 subjects, and one of the interesting things is their sleep patterns improve.

02:34:06.580 They get more deep sleep in the beginning of the night and more REM at the end of the night.

02:34:10.900 In other words, he tells me what I've noticed, and it makes me just think and maybe believe and maybe hope that something is getting into the cells and that there's a real effect, at least on sleep.

02:34:29.800 He didn't tell me any other things.

02:34:31.380 But really, my answer is, we don't really know enough.

02:34:36.600 It's very hard to measure effect because this NAD goes and goes away.

02:34:42.080 And, you know, what system can you do?

02:34:44.600 It's very hard to measure that.

02:34:46.800 And without clinical studies that are really well controlled, I think it's going to be hard to just support that.

02:34:59.000 And that's me as a conservative scientist and one who says that even metformin, that everybody says, you have enough data.

02:35:06.900 No, we have to do the clinical study to prove that.

02:35:09.600 And that's where I really stand.

02:35:11.380 And it seems to me that intravenous administration of NAD is also not particularly helpful, though it seems quite popular in the sense that I haven't seen compelling evidence that giving NAD intravenously makes its way into the cell either.

02:35:29.220 So I think the development, the interesting development will be to develop a drug and the drug will be a precursor that will get into the blood.

02:35:41.280 Outside of the liver.

02:35:42.600 Outside of the liver.

02:35:43.820 Yeah, exactly.

02:35:44.520 So if you could give NR or NMN sublingually or intravenously, that strikes me as very interesting.

02:35:51.640 Saying that there's a lot of data in rodents where also you look, you give it in different ways.

02:35:59.560 You give it some in water, but some are gavage.

02:36:02.560 And maybe that's different because one of the problems, maybe if you go deep enough.

02:36:06.960 You can bypass the portal circulation.

02:36:09.400 Right.

02:36:09.640 You can bypass something.

02:36:11.120 So, you know, but the in vivo studies are good studies.

02:36:15.700 Okay.

02:36:15.980 You just don't know what it means to humans.

02:36:18.320 Yeah.

02:36:18.900 Well, Nir, I'll do a time check here.

02:36:20.660 We've been going at this for a little while.

02:36:22.860 There's a lot more I want to sort of pick your brain on, but we could also just sit down and do this again sometime.

02:36:29.260 I want to make sure you've said at least as much as you want to say about metformin.

02:36:33.100 Because I think that to me is, you know, you're at this point arguably one of the world's experts on a drug that, just to put this in perspective,

02:36:41.020 I think Lou Cantley, who is obviously a close mutual friend of ours, and James Watson, who I don't know personally,

02:36:47.960 have both said quite publicly that metformin may have already saved more patients' lives from cancer than all other cancer drugs combined,

02:36:58.300 or something to that effect.

02:36:59.340 This is a drug that I think with each passing day, more and more people are beginning to learn about, beginning to ask questions about.

02:37:07.740 And my hope is that if nothing else, this podcast is a place that people can listen to this episode if they can't go and read 16 of your papers.

02:37:16.640 And they don't necessarily want to get, you know, that far down the rabbit hole.

02:37:20.940 But to hear it from you and not from me or some other schmuck who doesn't know much,

02:37:25.920 I want to make sure that if there's anything else you want to say about metformin that you say so.

02:37:30.600 And so is there anything else that you want to add?

02:37:32.340 Well, let me do one practical thing and one a little bit more philosophical, okay?

02:37:37.980 The practical thing that really surprised me, I was giving a talk somewhere for lay people.

02:37:45.420 So it was organized by a university and they invited people and to their surprise, there were 300 people in the audience.

02:37:54.860 And by the way, my talk wasn't about metformin.

02:37:57.980 My talk, the titles of my talk is usually how to die young in a very old age.

02:38:03.760 That's what my title is usually, if I can get away with it, that's a good title.

02:38:09.360 So people are coming and as they're entering, some of them come to me and say, so how much metformin should I take?

02:38:15.720 Okay, so I started talking and all of a sudden I said, let me ask you something.

02:38:21.520 Who here in the audience is non-diabetic and is on metformin?

02:38:29.000 And please, you know, you don't have to, just if you choose to.

02:38:32.180 So how many?

02:38:34.240 Half of the people.

02:38:35.580 Where was this talk?

02:38:36.260 What university?

02:38:37.140 I don't want to.

02:38:38.400 Okay.

02:38:38.900 What city?

02:38:39.440 Can you say what city?

02:38:40.620 Was it in the United States?

02:38:42.100 No.

02:38:42.600 Okay.

02:38:43.140 Fair enough.

02:38:43.780 So, you know, in other words, the prophecy is out, which makes me a little bit worried about the study.

02:38:53.020 When we first, Wall Street Journal is the first that picked on Tame, wrote an article, the next week we had 3,000 phone calls and emails of people volunteering to the study, which is what we need to do the study.

02:39:09.860 But just cut your, you trim that budget from 70 million down to 65, you cut all the recruiting costs out.

02:39:16.400 Exactly.

02:39:17.180 Except that then you realize.

02:39:18.540 Yeah, it's a bias sample.

02:39:19.540 Then you realize, just a minute, those people, first of all, we didn't advertise.

02:39:23.260 They had to find my phone number and email.

02:39:25.820 And second, they'll do, they're doing other things.

02:39:30.660 You know, it's like with the estrogen bias, they're probably exercising and stuff like that.

02:39:34.480 And then when I started getting emails, like I'm volunteering to the study as long as I'm not on the placebo, I thought, you know, if people care so much, if they'll figure out they're not on metformin, they'll just get metformin.

02:39:49.460 So from recruiting the 3,000, we, we outload those, those are not in my study, right?

02:39:58.560 So I want to say that still, we haven't done the study.

02:40:03.040 Okay.

02:40:03.640 We haven't done this study.

02:40:05.260 So you're between 65 and 80.

02:40:08.320 Okay.

02:40:09.020 Just know we haven't done this study.

02:40:10.760 We don't know if it's safe as it is in young people.

02:40:13.740 And so we don't have really a clinical study with evidence that this is good.

02:40:20.940 And I'm, I'm just worried now that part of the reason to do this study is because if it's not good, people should know also, right?

02:40:29.080 And not take another drug.

02:40:30.820 Okay.

02:40:31.280 Yeah.

02:40:31.480 Not that I believe that, but it is possible.

02:40:34.840 The second thing that is more philosophical.

02:40:37.160 So I said before that I was invited to the Vatican.

02:40:40.600 I actually was invited twice the second time.

02:40:43.060 And I'm, I'm very close to Cardinal Ravassi, who's a number four in the pontiffs.

02:40:50.840 He's, he's in charge of science and on arts that is including science.

02:40:55.560 And it's interesting because the Vatican basically says, we don't want to be in a Galileo Galilei situation again.

02:41:03.160 We don't want to be in a situation where the science is so right and we're so wrong that we don't know what to do with it.

02:41:09.080 Okay.

02:41:09.440 And so they called me and they said, you know, we have a meeting in the Vatican.

02:41:14.560 We'd like you to talk about aging.

02:41:16.600 Could you come?

02:41:17.520 And I said, sure, sure.

02:41:19.760 I'll come.

02:41:20.200 And then I'm going on and saying, am I, am I the keynote speaker?

02:41:23.480 And they said, the keynote speakers are the Pope and Joe Biden, who was vice president.

02:41:31.960 I said, okay, sorry, sorry.

02:41:33.540 I'm still coming.

02:41:35.160 So that's how it goes.

02:41:36.880 Joe Biden goes up and explains his cancer initiative and how difficult it is.

02:41:43.000 Because when you have cancer, you have in the cancer, maybe five other genomes.

02:41:48.120 And they're also different than every other cancer like that in the world.

02:41:51.920 It's really a mess.

02:41:53.960 Then comes the Pope and says, you know, I still hope that there'll be one little pill cheap for everyone in the world that will cure whatever cancer they wear.

02:42:08.860 Appropriate.

02:42:09.740 I hope so too.

02:42:10.780 But then I go up and I say, well, there is actually a cheap pill like that.

02:42:18.340 And I don't know if it cures any cancer in the world, but it can prevent a lot of the cancers in the world.

02:42:24.860 And actually, it has a side effect that it also can prevent a lot of other diseases in the world.

02:42:30.300 So it was like, you know, going back to aging and the risk of aging for age-related diseases and how impactful and cheap it could be compared to treating cancer or something like that.

02:42:48.900 So I think the prevention of aging is really a good place to be.

02:42:54.600 And I think because we went from hope to promise and we have to realize the promise, I think life is going to be very different in the next decade with our advance.

02:43:04.700 Last question, and not to end on a downer, but let's try to figure out where a blind spot could be.

02:43:11.940 Where could we be wrong?

02:43:13.580 Obviously, you and I share more in common in terms of philosophy and points of view than I even realized before we spoke.

02:43:21.520 But where could we be wrong?

02:43:22.960 So, you know, I'll tell you what's my optimism.

02:43:27.580 You know, we have, the European have nine pillars of aging and we have seven pillars of aging and they're all interconnected.

02:43:38.180 And when I say interconnected, I just spent the time with Anna Maria Cuervo this morning.

02:43:44.220 When she fixed autophagy, she fixes also metabolism.

02:43:50.980 Okay, there, when you start doing things in any one of those pillars, you start improving the others.

02:44:01.680 Now, where we could be wrong is we don't know what is the impact of each of those pathway in humans.

02:44:10.280 Okay, most of our data is kind of animal data.

02:44:13.900 But I think our fallback is that we all age, the advantage of in aging is that aging is universal.

02:44:27.440 Every animal has the skin, the hair, the skeletal, the frailty.

02:44:34.780 It's very universal.

02:44:37.480 And those treatments are metformin, rapamycin.

02:44:41.380 You give it to any animal, it almost delays the aging there.

02:44:45.600 So I don't think we're going to be wrong.

02:44:48.820 I think that maybe some pathways will not be so effective.

02:44:53.420 Some of them will be harder to treat maybe.

02:44:56.200 But I think we have to affect three, four, and we're right on our way to do better.

02:45:02.780 And that's just the beginning.

02:45:04.380 And in the TAME study, will it be one-to-one, male-to-female?

02:45:07.800 Yes.

02:45:08.800 So we'll also avoid that other...

02:45:10.500 Right.

02:45:10.760 And have you powered it such that men and women are different animals?

02:45:13.740 Well, no, because all our data on metformin, and we looked carefully to the DPP and other,

02:45:19.660 they didn't see any gender effects on any of the outcomes.

02:45:22.720 So we're assuming that...

02:45:24.620 So you're powered as though the genders don't have a significant...

02:45:28.380 Like we would do maybe rapamycin, you know?

02:45:31.020 Yeah.

02:45:31.400 Well, Nir, this has been fantastic.

02:45:33.500 You know, I was sick the last two days.

02:45:35.760 My voice is gone.

02:45:37.220 I feel like crap.

02:45:38.160 I was like, maybe we should postpone it.

02:45:40.620 But then when I woke up this morning, I was like, there is no way we are postponing this

02:45:44.740 discussion because I cannot wait to have this discussion.

02:45:47.940 So thank you so much for coming over.

02:45:51.000 It was great.

02:45:51.680 It was, you know, it was great because you asked really great questions.

02:45:56.580 And also, I'm now quite hungry and I'll be happy to have dinner with you now.

02:46:01.300 Oh, we are going to put together a seminar on how not to calorically restrict.

02:46:07.640 Tonight, only tonight.

02:46:09.320 All right.

02:46:09.940 Thank you, Nir.

02:46:10.540 You can find all of this information and more at peteratiamd.com forward slash podcast.

02:46:18.260 There you'll find the show notes, readings, and links related to this episode.

02:46:22.380 You can also find my blog and the Nerd Safari at peteratiamd.com.

02:46:27.000 What's a Nerd Safari, you ask?

02:46:28.480 Just click on the link at the top of the site to learn more.

02:46:31.360 Maybe the simplest thing to do is to sign up for my subjectively non-lame once a week email

02:46:35.680 where I'll update you on what I've been up to, the most interesting papers I've read,

02:46:39.500 and all things related to longevity, science, performance, sleep, etc.

02:46:43.200 On social, you can find me on Twitter, Instagram, and Facebook, all with the ID peteratiamd.

02:46:49.560 But usually, Twitter is the best way to reach me to share your questions and comments.

02:46:53.280 Now, for the obligatory disclaimer.

02:46:54.840 This podcast is for general informational purposes only and does not constitute the

02:46:58.660 practice of medicine, nursing, or other professional healthcare services, including

02:47:03.080 the giving of medical advice.

02:47:05.120 And note, no doctor-patient relationship is formed.

02:47:08.060 The use of this information and the materials linked to the podcast is at the user's own risk,

02:47:12.900 the content of this podcast is not intended to be a substitute for professional medical advice,

02:47:17.640 diagnoses, or treatment.

02:47:19.380 Users should not disregard or delay in obtaining medical advice for any medical condition they have

02:47:23.900 and should seek the assistance of their healthcare professionals for any such conditions.

02:47:29.020 Lastly, and perhaps most importantly, I take conflicts of interest very seriously.

02:47:33.220 For all of my disclosures, the companies I invest in and or advise,

02:47:37.060 please visit peteratiamd.com forward slash about.

02:47:42.900 For all of my disclosures, please visit peteratiamd.com.

The Peter Attia Drive - January 07, 2019

#35 - Nir Barzilai, M.D.： How to tame aging

Episode Stats

Length

Words per Minute

Word Count

Sentence Count

Misogynist Sentences

Hate Speech Sentences

Summary

Transcript