337- Insulin resistance masterclass: The full body impact of metabolic dysfunction, treatment & more

Talk a little bit about what insulin resistance is. Basically, every time you eat a meal, you're going to release insulin. There are many processes that insulin controls. So one of the challenges with the term insulin resistance is, as you said, it's a vague term and it's non-specific. There are two problems in diabetes: one is you don't make enough insulin; the other is your insulin resistant. Type two diabetes—you're knocking on death's door, correct? You're going to go blind; you're going to have your toes amputated; you're not ever going to have an erection again; and you're going to die of cardiovascular disease, or kidney disease, or Alzheimer's disease quickly. Some people say that Alzheimer's disease is, you know, Diabetes Type 3. I'm not sure. Brain diabetes? Yes. You are obese when you're 4 years of age, you're going to be obese when you're an adult, and your life expectancy will be significantly shorter, and your quality of life will be significantly reduced. What has changed so much in the last 30 years that has created this epidemic?

Hey everyone, welcome to the Drive podcast. I'm your host, Peter. [Music]

AA Ralph, thank you so much for coming down to—I guess up to—Austin from San Antonio. Um, very excited to sit down with you and talk about, uh, you know, potentially one of the most important subject matters in all of Health. Uh, people who listen to me all the time here and are familiar with me talking about these four horsemen, right? Cardio—cardiovascular disease and cerebrovascular disease; cancer; uh, neurodegenerative and dementing diseases; and then there's this fourth Horseman that I talk about, and it's in many ways the squishiest because it's not the one that shows up on the most death certificates, um, but in many ways it's the foundational one that is amplifying the risk of all of those other causes of death, and I refer to it as, you know, sort of metabolic disease spanning the spectrum from hyperinsulinemia to insulin resistance to fatty liver disease all the way out to type 2 diabetes. Um, so given how much I speak about that, it seems very important that we should have a really uh thorough discussion of that foundational metabolic disease, and no one better than you to have that discussion. So let's start a little bit with just kind of telling folks briefly about, you know, what you're doing at UT San Antonio and why you've spent the last 40 plus—almost 50 years now—working on this problem.

So for me, I actually have been sort of in this field of metabolic disease for a long time. Um, I think I'm the longest consecutively funded—53 years—NIDDK investigator, uh, and uh, I actually started even long before that, uh, when I was a medical student uh in at Harvard. I had this fantastic teacher, Professor Cahill, who gave us all of the lectures on intermediary metabolism, and I decided this is what I wanted to do, uh, and I worked each summer with Professor Cahill, and uh, you know, sometimes in life you meet the right person, the right opportunity, it changes everything you do. And basically what I do now, I contribute directly to George. And you know, when I gave the Banting lecture in 2008, people usually, you know, put a picture of their mother and father and children, and I love my mother and father and children, but I only showed one picture, and that was Professor Cahill, because he's really the person who's ended up sort of directing me to where I am today.

You know, people who are listening who are particularly astute might recall, um, I've referenced a number of Cahill's uh papers, but one of the more interesting studies he did, which it's possible he did while you were even a student there, was the 40-day starvation study. Um, now you might have not been quite at Harvard yet because this was, if I recall, in the mid-60s, maybe '66, '67, where—and it was probably a group of medical students that actually volunteered, if not medical students, undergrads—and they they they did a water-only fast for 40 days, and the study basically just followed all of the metabolites—what happened to glucose levels, obviously insulin, beta-hydroxybutyrate, acetacetate. Anyway, it was very fascinating stuff. One of the things that was most interesting to me in that study was even under a period of such extreme starvation, the brain never gave up its dependency on glucose. So even though ketone bodies began to um service the brain by about day 7 to 10 as the majority of the fuel, even at three and four weeks of starvation, glucose was, if my memory serves me correctly, still providing about a third of the brain's energy.

Your memory is very good. The brain did switch over to ketone metabolism. And believe it or not, I didn't do the 40-day fast, but I was one of the people who fasted for 5 to 7 days, uh, and if you fasted for 3 days, uh, you could get paid $50, and I thought I was the richest guy in the world, uh, from this study. So I can assure you that the physical specimens in this study were phenomenal. Uh, what did the 40-day fasting students get? Uh, I don't know, but I'm sure he paid them a lot of money. That's amazing in order to do that. And the interesting thing about that is you realize that we have so much energy stored in the human body. Who would have thought that, you know, you're a lean type person, you can, you know, fast for 40 days. But the real problem is at some point you start to break down muscle, and then if you start to break down cardiac muscle, then prolonged fasting at that point becomes a problem. But you have a lot of energy stored in fat, and uh, you can uh starve for a long time, and obese people easily can go for 3, 4 months, uh, with all the reserves that are in the body.

Well, let's maybe talk a little bit about um what insulin resistance is. Um, I we'll get into what causes it, um, but but let's just maybe define for people this term that gets thrown around constantly, and and let's, you know, explain what it is from a technical standpoint.

Yeah, so basically every time you eat a meal, uh, and your blood sugar level goes up, you're going to release insulin, and insulin is sort of a master regulator for all uh biochemical processes in the body. So one of the things that insulin is going to do is going to talk to your muscles and going to say take up glucose and uh burn that uh glucose. Uh, so what we need to know is in a normal person, when to infuse insulin, uh, how much uh of the glucose is taken up by the muscle, and then we could look at someone who is say overweight, or we could look at someone who's diabetic, and I actually developed the gold standard technique, which is the insulin clamp technique, to look at this. So we could take an obese person or a diabetic or a normal person, we raise the insulin, and then I'm using muscle as an example—how much uh glucose is taken up, disposed of by the muscle, and then I can compare if you're overweight compared to the lean person. Well, obese people are very insulin resistant in terms of muscle glucose uptake. I could look at the diabetic; they're even more insulin uh resistant. But there are many processes that insulin controls, so insulin regulates uh how much uh fat is uh released from your fat cells, and obese people unfortunately insulin keeps the fat in your fat cell, but in obese people insulin doesn't work so well. So instead of keeping the fat in the fat cell, even though your insulin is high, you're breaking down the fat. So you have to look at each individual process that insulin is controlling, and so for that process we know this is what a normal person should respond like, this is what a diabetic responds like, and the diabetic is much, much more insulin resistant; they're not responding. Uh, so it's in a certain way it's a general term because insulin controls so many uh things. Protein metabolism—insulin is very important in helping you to build protein. Uh, so I could infuse insulin, and we've done this using carbon-labeled leucine, and we can define how insulin promotes protein metabolism in a normal healthy person, and then I could do the same kind of study in an obese person, and we know that the obese people don't respond to the insulin as well in terms of aggregating protein metabolism. Uh, so it's kind of a general term.

Does that translate not just to structural proteins such as enzymes uh or cellular structural proteins, but also macro structural proteins such as muscle?

Absolutely. So, uh, I I can look at specific enzymes within the cell; I can look at certain genes within the cell that are turned on or off; or I can look at muscle in terms of muscle as a bulk. Uh, so uh there are many, as I said, many ways in which you could define insulin resistance, but basically whatever the particular process you're looking at, you're comparing what would be the normal response uh in a normal healthy person compared to what might happen in a diabetic person or an obese uh individual.

So one of the challenges with the term insulin resistance is, as you said, it's a vague term and it's nonspecific because the actions of insulin are so many. It has an action in the liver; it has an action in the muscles; it has an action with response to glucose; it has an action with response to amino acids; and it has an action with response to fat, both in the liberation of fat—lipolysis—and presumably in response to oxidation.

Absolutely. And we'll go through all of these, but let's maybe start with um how the euglycemic clamp test is done, and uh let's assume that I'm a healthy enough individual that we can use me as a proxy. I come into your clinic, um what are we going to do? How do you run this test?

Yeah, so let me bring you back in time when I was a fellow, because at that time we didn't really have a good measure of insulin sensitivity. So what people would do is you do an oral glucose tolerance test, yes, and the insulin level would go up, and so then some people would say I'll look at how much insulin comes out compared to the rise in glucose, and that's a measure of beta cell function, and then someone would just turn it around and say, look, I'm going to see how much the rise in glucose was per insulin, and that's a measure of insulin resistance, and it was very clear to me, well, this is insane; you can't take two variables and then just depending upon how you want to look at them, switch uh denominator, numerator. So I said, need to develop something uh that is really more specific. And just to be clear, Ralph, I mean, unfortunately we as clinicians are not able to do euglycemic clamps, correct? So we are still looking at oral glycemic uh tolerance tests; we are still giving people, you know, oral glucose and sampling glucose and insulin every 30 minutes and trying to impute what we can, which I'd love to come back and talk about interpretation, but carry on with the with the limitation. We could do that because we actually have done a lot of work on how you interpret that. So what we said is, why don't we develop a serious way? And so we developed the technique where I could take 100 people, and I would infuse insulin initially as a priming dose and then just clamp the insulin level. So I give a prime continuous insulin infusion. I can take a 100 people, and all 100 people, I can raise your insulin level uh by 100 micro units per ml, uh, and I can do that for two hours, and now I know that the stimulus, the insulin stimulus, whether you're lean, whether you're obese, or whether you're diabetic, the whatever particular process that I want to look at. So maybe I wanted to look at how insulin shut down hepatic glucose production, and actually we were the first people to ever use radioisotopes to trace this and show that in normal people insulin shut down glucose production by the liver very quickly, but obese people and diabetics were very, very resistant to the insulin. Then we said we wanted to know, look, everybody now has got the same insulin level, uh, how effectively does that insulin uh stimulate muscle glucose uptake? And again, what we showed—and these actually were the very first unequivocal demonstration that uh diabetic people, type two, were insulin uh resistant. Before this, there was a lot of controversy. You know, Dr. Ren, who's sort of the father of insulin resistance—I like to think I'm the the son of Dr. Ren—he's a a great idol of mine, uh, he really was one of the very first people uh to insinuate that diabetics were insulin resistant. Now with the insulin clamp we showed this very definitively, and we also know—we use the labeled uh glycerol and free fatty acids—and we could show the ability of insulin to shut down release of uh lipid from uh the fat cell was markedly uh impaired. Uh, so three of the major organs—all of this work originally was done by us when I was back at Yale.

So let's summarize those again. We're talking about this in an insulin-sensitive person, right out of the gate. Insulin is going to shut down hepatic glucose output.

Absolutely. So it's—which again, all of this kind of makes sense if you think through the pathway—our liver is constantly putting glucose into circulation because the muscles can't put glucose into circulation, so something has to feed the brain.

Yes, absolutely. If insulin is high, it suggests glucose is already sufficiently high, so let's not create more glucose toxicity; let's shut that down.

Yeah. The second thing it's going to do is it's going to take that excess glucose and put it in the place where we have the largest capacity to store it, which is muscle.

Absolutely. So point two is we increase muscle uptake of glucose, and then point three you said was it's going to shut down lipolysis.

Yes, it's going to shut down the release of triglycerides and/or free fatty acids from the adipose tissue.

Why? And that's very critical. And we also, when we did these studies, we would put a catheter in the hepatic vein and in a femoral artery and femoral vein, so we could look at the individual tissues, and what we showed is that when you infuse insulin, say 80 or 90% of the glucose is going to be taken up in muscle, only 10% is going to be taken up in the adipocyte and stored.

And how much in the liver? Uh, basically none under euglycemic conditions, and we were the first to show this conclusively as well. There's no glucose uptake in the liver by insulin.

Just explain to people what a euglycemic condition means. Euglycemic means your fasting glucose when you wake up in the morning is 80, uh, now you're euglycemic, uh, that means when we do the studies we keep you fasting glucose of 80; we don't let the glucose change; all we're going to do is raise the insulin. And that means you're giving glucose, of course, because we didn't give glucose, then your blood sugar level would drop, and then you'd release cortisol, you release epinephrine.

Yeah, so I just want to make sure people understand that. I was going to come back to that, but I wanted to not—I wanted you to finish that point. So let's let's make sure we go back to the test because it's very counterintuitive. So I've got a catheter in each arm; I walk in off the street; I've been fasting; uh, my my blood sugar is 80 or 90, whatever milligrams per deciliter it is. You are going to have to infuse both insulin and glucose into each of my arms, and the reason is, when you said a moment ago, you're going to steadily increase my insulin and take it to a steady state of 100 IU per micro per—per—that's a staggeringly high insulin level.

Not so high. In your eye, after a meal would be maybe 60. Obese people very commonly get to 100.

Sure. For a healthy person, they would never see an insulin level that high, and if you were not simultaneously running glucose into them, you would kill them within minutes.

Hopefully not. Yeah, but but but just to to get to the point of, they would become so profoundly hypoglycemic that they would cease to exist. And it should be obvious that if you're very sensitive to insulin, I have to infuse a lot of glucose. But the other beauty of it, as I said, uh, when when I was a young guy at Yale, there was a physician uh in New York, Dr. Al Shuler; he was the first one to use tritiated glucose to trace metabolic pathways, and I said, oh, this is astounding. So I actually went to visit Dr. Shuler and learned how he did it, and so all of the insulin clamp studies that we did, we were the first people to use tritiated glucose in humans uh and to show that the ability of insulin to shut down the release of glucose from the liver was markedly impaired. And this is—s—just again, sorry to interrupt, but just to make sure the people are following us, the reason you wanted to use tritiated glucose there was not to quantify the total amount of glucose disposal—you could do that on mass balance—you wanted to determine the ultimate fate of glucose—how much became hepatic glycogen, if any.

Sounds like the answer is none.

How much became muscle glycogen? Sounds like you said about 90%. And how much ultimately got converted through de novo lipogenesis into a to free fatty acid? Sounds like that's about 10% under the euglycemic condition. Is that correct?

Yeah, in general that's correct, except in the muscle, remember some of of the glucose is going to be oxidized. So if you look at the glucose once it gets into the cell, one-third would go through the glycolytic pathway and be oxidized right away.

Yes, and and the other two-thirds would be stored as uh glycogen.

Now what if the person is—I mean, presumably you're doing this test and a person is sedentary? Yes. So how is the—and muscle—is muscle—is it metabolically active at rest? I guess it is.

Yes, yeah. So that's really interesting. Does that mean you're increasing energy expenditure under these conditions?

I well, of course, uh, in a certain way you are, but it's not like when you go out and you exercise and you run a mile or two. Uh, so uh, you know, uh, there it's a—I I would say uh you are turning on a number of cycles which are, of course, going to increase energy expenditure; you're generating ATP. Uh, so there is a certain increase in energy expenditure, but if I really want to increase energy expenditure, I'd get you to go jog 5 miles or so, because exercise is really the thing that really increases energy expenditure.

And Ralph, just for a sense of amount, if you're doing this in say somebody my size who's insulin sensitive, how many actual grams of glucose would you be able to get into the person within the hour while whilst keeping insulin clamped?

Yeah, so uh I'm going to do it first in terms of rates, the way we express it, and then I'll translate that. So it's uh under basal conditions—you wake up in the morning—and your liver is producing and your tissues are taking up about 2 milligrams per kilogram body weight per minute. Okay. Uh, so I say that again—liver is producing—that's hepatic glucose output—that's a glucose out part—2 milligrams per kilogram body weight per minute, and we were the first to actually show this many years ago, and this is humans; mice are very, very different—totally different—and that's why extrapolating from mice to humans can be a problem.

Let's just reflect on that for a second. This—this people who listen to this podcast are probably sick of me saying this, but I'm sorry, I just can't stop saying it—the liver never ceases to amaze me, right? Like, I mean, it's an incredible organ; it's an unbelievable organ. And again, I come back to this idea—it's the only major organ for which we don't have extracorporeal support, right?

Interesting. Yeah. If your heart—if you went into cardiogenic shock and we felt we could, you know, reverse it in time, we could put an intra-aortic balloon pump in you; we could put an IABP in you; um, we could we could put a a left ventricular assist device in you to to to stem you over until we get you out of there. If your kidneys are destroyed, we can transiently dialyze you. Even if your brain is experiencing swelling, we can, you know, put enough steroids in you or decompress your skull to give you uh the time to recover and keep you alive; otherwise, you go through all the major organs. If your spleen is dinged, take it out, right? Um, even if you lost your small bowel, we could at least transiently keep you alive with, you know, TPN or something like that. None of this is true with the liver. You know, in the old days, um, they actually used to use pig liver perfusion. I know, uh, but that was in the old days; we don't do that uh anymore. Uh, so in baboons as well, in baboons. Yeah. So so again, the the fact that the liver can titrate this amount is remarkable. So 2 milligrams per kilogram per minute. So you take an individual who weighs 100 kilograms, you're putting 200 milligrams per minute of glucose into circulation, then you can multiply that by however minutes you want to look. So so that's a gram every 5 minutes—that's 12 grams of glucose every hour—that the liver is putting out. But now when I do an insulin clamp, depending on how much I raise the insulin, and over the years we've done the dose-response uh curve, and I can come back to this, because your fat is exquisitely sensitive to insulin. If I raise the insulin just by 10 micro units per ml, the uh fat stops producing uh free fatty acids and glycerol—you inhibit lipolysis literally uh uh completely.

The liver, you need to get uh the insulin up to about 50 micro units per ml to really get it shut down. So—and the fat you had to get how high? 10. A rise of 10. A rise of—and tell me these people when they come in and healthy, they're at what? They're at 5 to 10 faster—they're at 5 to 10. So I'm going to raise them from 5 to 10 to maybe 15 or 20, and that's going to in large part shut down lipolysis. Okay. Uh, and in fact, all of this sort of work was work that we originally did many, many years ago. Uh, now at the level of the liver, you really need to get up to about 50 micro units per ml. So maybe I'm at 10; I'm going to bring you up to 50, and that's in large part going to shut off glucose production by the liver. Now that's critical because, you know, you wake up in the morning and your liver is producing glucose. Now if you eat a meal, glucose is coming in from the gastrointestinal tract; you can't have glucose coming in from the liver at the same time; otherwise, you get very hypoglycemic. So when you eat a meal and that insulin comes out, it really needs to shut down hepatic glucose production. Now what's replacing the liver is what's coming from the meal, but then after you absorbed all of the meal, the liver needs to turn back on. So understanding how the liver is responding to insulin is really very important. And then if I want to look at what's going on in the muscle, the reason why we go to 100 micro per ml, which is—it's above physiologic, but it's still within the physiologic range, uh, and uh you really need to get—if you really want to stimulate muscle glucose uptake completely in a normal healthy person, you'd probably have to get the plasma insulin to about 200 micro units per ml.

But in reality—I'm sorry—at 200 what happens? Micro—you you have now maximized muscle glucose uptake.

But in reality, even in an insulin-sensitive person—

Yes.

And just to make sure I understand what you're saying, you're saying that if you took an insulin-sensitive individual at 100 units of insulin versus 200, you will actually drive more glucose uptake—you haven't saturated the GLUT4 transporter at 100.

Yeah, probably about 20 or 25% more uptake as you go from 100 to 200.

Wow. And these are all early studies that uh we did. Uh, so when we talk about insulin resistance, that's why I said you need to know which tissue you're talking about and which metabolic pathway, and if you want to talk about enzymes, you need to talk at what specific enzyme, because uh insulin resistance needs to be related to the tissue you're talking about in the process within the tissue that you're talking about. So insulin resistance is a very important concept, uh, but you all have to be a little bit more specific about what aspect you you want to address. So you can have insulin resistance in the fat cell; you can have insulin resistance in the liver; you can have insulin uh resistance in the muscle; and then something that's now pretty exciting—you may have insulin resistance in the brain. And the suggestions now—and there are many insulin receptors in the brain—Jesse Roth, you know, very famous diabetes person, you know, maybe 50 or 60 years ago was the first to describe insulin receptors in the brain, and this is an area that's now starting to unfold; it may have some relationship to neurodegenerative disease, Alzheimer's disease. You know, some people say that Alzheimer's disease is, you know, Diabetes Type 3. I'm not sure. I—diabetes? Yes. So that that insulin resistance is a very important concept, and unfortunate—and I would say in in in the—let's say we're going to talk about diabetes—the even though there's an ominous octet that I developed that's used everywhere in the world for the pathophysiology of type 2 diabetes, if we really wanted to take and solidify it and say what are the two big concepts, insulin resistance would be here. On the other hand, would be impaired beta cell function. Okay. So if you are insulin resistant and your beta cells work well, they know how to read the insulin resistance; they'll make enough insulin; you won't become diabetic. The hyperinsulinemia can damage you in other ways, but you won't become diabetic. But what happens is if you're insulin resistant, and particularly if you have a genetic predisposition, if your beta cells have to continuously pour out insulin, they start to exhaust, uh, and so insulin resistance is a disaster for someone who has a genetic predisposition; it's going to bring out uh the diabetes.

Insulin resistance, in my opinion, is intimately related to cardiovascular disease. That is why when you you see a diabetic patient—10% of them—you walk in, you have diabetes first time I see you—10%, 15% of the people already have clinically significant cardiovascular disease, and if you look carefully, virtually 100% of them do.

And sorry, Ralph, do you think that that is a result of the hyperinsulinemia or the untreated or poorly treated hyperglycemia?

All of the above, but more importantly, what we showed—and we were again the first people to show this—and the cardiologists, they're kind of hemodynamically oriented; they're looking at, you know, vessels, stenotically constricting. But if you look at the insulin signaling pathway, insulin has got to bind to its receptor, and then there's a signaling pathway—I can tell you all the molecules in there, which I'm not—and then glucose gets transported in the cell. We were the first people to show in humans that that pathway doesn't work normally. Insulin will bind to the receptor; it will activate the receptor, but the next molecule—IRS1, PI3 kinase—all those molecules don't get activated, so glucose doesn't get into the cell; that's diabetes. That same pathway activates nitric oxide synthase, and that generates nitric oxide. Nitric oxide is the most potent vasodilator in the human body; it's the most potent anti-atherogenic molecule in the human body. So this defect that's in muscle, uh, and it's in cardiac muscle, and it's in skeletal muscle—this all human data that I'm I'm talking about, not animal data—when you get a defect in that insulin signaling pathway, that's going to cause diabetes, and it's going to promote cardiovascular disease, and that is why you can never separate cardiovascular disease uh from diabetes.

Now, as you pointed out, right from—so you know, I believe that high levels of insulin are also atherogenic. Now I don't want people saying Dr. DeFano said you shouldn't be giving insulin to people who need it. Of course, uh, if people need insulin, you need to give them insulin. But our beta cells make 35 units of insulin per day if you're a type 1 diabetic. And again, true—what—what do you mean? Uh, so you—during the—you're going to have your breakfast; you're going to have your lunch; you're going to have your dinner; and if I were to add—sorry—35 units—I thought you meant micro units.

Got it. Yeah, of course—35 units of insulin. So we showed this many years ago when I actually was at Yale that if you were to take a type 1 patient and they were lean, they would only need 35 or 40 units of insulin to get their uh glucose controlled, assuming you gave the doses at the right time. But we have a lot of people who are taking a 100 units of insulin, both type ones and type twos.

So 3x physiologic.

Yes. That kind of hyperinsulinemia, I think there's evidence to support that's atherogenic. But now we have a problem. Okay. Can you have the glucose remain high? Yeah, it's a question of do you want to die quickly or or slowly? Because we have really good drugs—

Yes, yes. But if you were only doing this with insulin, be a problem. It's an awful trade-off. It's—you're going to die very quickly from hyperglycemia if you're left untreated, but if we overdo it with insulin to maintain normal glycemia, we're going to kill you slowly. You're stuck; you have to treat, but you also know that when you're giving these big doses of insulin, there may be some side effects. This is something, Ralph, I don't think that has been necessarily appreciated by the medical community, has it?

Oh, absolutely not. Yeah. There has generally been an ethos of—when I've talked to patients with type 2 diabetes—um, that what they've been told is—I'm told to cover with as much insulin as is necessary to maintain my glucose levels uh in this range, and it means I can eat whatever I want; it's okay if I have all the pasta and bread and sugar in the world, because as long as I'm covering it with insulin, I'm okay. And then you find out, wow, you're taking 150 units of insulin a day in all of its forms—the short-acting, the long-acting, etc. But but I didn't actually realize that that what we would consider physiologic is 35. I I may have known that at one point, and I've since forgotten, but that's a great reference.

So basically, if there's a person with type 2 diabetes listening to us today and they're taking 75 units of insulin, the one of the takeaways should be what do I need to do with my nutrition and other pharmacological activities plus exercise plus everything that's under my control to maybe get that down to 35, where I would be at a physiologic level?

Absolutely. There are things, as you already insinuated—weight loss—if you can get people to do it—exercise—and then we can add medications uh in combination with insulin—insulin sensitizers or some drugs to help you lose weight—that will all uh also allow you to get that dose uh of uh insulin uh, you know, uh reduced. The other thing we showed in this study—Dr. Del Prato, who's past president of the European Diabetes Association—we took normal healthy lean kids, 18, 25 years of age, and we uh gave them—we put them on the clinical research center uh for uh 3 days, and we gave them a very, very low dose uh of insulin infusion, and we raised their fasting insulin from eight, which is what a normal person would be, to 20, which is really quite low, and within 48 to 72 hours they were as insulin resistant as a type 2 diabetic patient. So hyperinsulinemia induces insulin resistance. This is—wait a second—why is that the case? So what insulin does is it downregulates the insulin signaling transduction system, so that insulin when it binds to its receptor and then it activates IRS1 and PI3 kinase and Akt, that system is downregulated by hyperinsulinemia. All of this that I'm telling you about, it's all published; these are all uh studies done in in in in humans, and this has also been shown in in rodent models uh as as well. So this is another reason why we don't want people to be hyperinsulinemic. You have to explain that to me again, Ralph; that is mind-boggling to me. I would never—I would never have predicted that.

So let me say it back to you because I'm—I feel like I missed it when I was writing something down. You took normal volunteers—y—who had a fasting insulin of eight—

Yep.

And they're lean, healthy.

Okay.

And you simply infused insulin in them, presumably with glucose.

Oh yes, of course. On the clinical research center where we can monitor—keep the glucose perfectly constant—we're not letting the glucose change. So person shows up, insulin eight, glucose is 90. You do a euglycemic clamp where you bring insulin up only by one and a half—per—one and a half X—much less than would be when you eat a meal.

Exactly. Not even a post-prandial bump, but now it's constitutively sitting there at 20.

Yeah. And you've obviously had to bring glucose—you had to infuse glucose to maintain euglycemia.

Correct.

Did you say that in 4 days—48 to 72 hours—these people are as insulin resistant as type 2 diabetics?

Okay. Again, very, very counterintuitive, because if our—if our model is that insulin resistance, which is the Hallmark factor contributing to type 2 diabetes in the combination of beta cell fatigue, is driven by lipotoxicity—which we're going to come to—that's an important one—these people didn't have any of that.

No, these people didn't have any of the intramyocellular lipid that we talked about with your colleague Jerry Ren as a predisposing factor. It's the direct effect of insulin downregulating the insulin signaling system and probably other distal metabolic within the cell uh as well. Uh, okay. So then when you turn the clamps off—see, let's just say we ran this for 72 hours—we've made them functionally diabetic—turn the clamps off—how many hours or days do that—what would you predict?

I would predict uh probably within 24 to 48 hours they would return to normal, 'cause we did this acutely. Now if we were able to do this for several months, uh, then I would anticipate that the insulin resistance would remain for a long period of time. Uh, and remember when we treat type 1 diabetics, we're always giving the insulin

Study you got this mixture of individual genes. What about what about the phenotype, Ralph?

So I've I've taken care of I've taken care of a couple of patients with type 2 diabetes who are very lean, including one patient whose body fat, by DEXA, was about 8%. So for people listening, that is insanely lean—very lean. So you take an individual whose body fat is 8%, and yet they have type 2 diabetes. The first thing that comes to my mind is a lipodystrophy. Is this an individual who's adipose tissue is the problem? In other words, they're not able to assimilate enough excess nutrient, i.e., glucose, into the fat cell, and so they're undergoing the toxicity associated with an insufficient reservoir? Is that what could be the the causal—not that I can tell you what's causing the lipodystrophy—but is the lipodystrophy that's driving the the diabetes here?

The answer to that is it it's very clear that lipodystrophy can cause diabetes. Okay, this is a I would say a very, very rare and unusual cause, but well established. But that's but you're saying that's not what would explain 1% of diabetics. That's two. Okay. And Jerry Shulman uh has done some beautiful work in this area, so it's unequivocal that lipodistrophic people, because their fat cells can't take up the fat, it ends up in your myocardium, cause heart disease, ends up end up in the beta cells, end your beta cell, the muscle. But that's a very, very, very small percentage.

Uh, so the the basic genetic ideology of the insulin resistance, the number the PPAR gamma gene has been associated. There are about seven or eight genes. There's a recent study uh I think it's in Nature Genetics by Brown where they've identified uh eight, and again they're associations. Except I would say the PPAR gamma gene that is is pretty clear. That's a—did Mitch Lazar do some of this work? Pardon? Mitch Lazar do some of this work? He's worked in this area. But again, the number of uh in my opinion, it's a long list of folks at this point.

Well, yes, the number of genes that have been described. So one of the and and the other—the other thing people said, well, maybe there are 20 genes involved, each giving a small component, and that's why it's so difficult. Well, all of these hypotheses have been difficult to approve prove, and the simple fact is we don't understand the genetic basis in part because diabetes, in my opinion, is a very poor phenotype. Diabetes is a very heterogeneous disease. So when we talk about diabetes, if that's your phenotype, it's not surprising to me that it's going to be difficult to define genes that are related to diabetes.

Uh, so what I'm going to tell you about—I don't want to take the credit uh for this. Okay, okay. Uh, so one of the people in my division, Dr. Luke Norton, uh working with Steve Parker at Michigan—I'm involved 'cause I'm doing the insulin clamp studies—we're taking as a phenotype muscle insulin resistance. This is a very, very specific phenotype. Okay, this is not diabetes, the ominous octet, my pathophysiology, that's eight problems. Okay, this is muscle insulin resistance. I'm going to do an insulin clamp now, and then I'm going to do a muscle biopsy before I do the insulin clamp, and I'm going to do a muscle biopsy at the end of the insulin clamp, and what happens? Okay. Well, during the insulin clamp, I know exactly how sensitive or resistant you are to insulin. I've got the most definitive phenotype in the world. No one gets this kind of phenotype, and now what do I see? An enormous amount of chromatin opens up. This is the epigenetic component. Genes in uh uh chromatin area that you're never ever going to see in the basal state. And that's why we think this is a hypothesis now, that why it's been so difficult uh to with all of these GWAS studies to identify genes that are associated with diabetes. And now we're starting to see in diabetic people, in non-diabetic people, we're starting to see some associations which we think now are causal, and we can relate to the insulin resistance with the clamp.

So the first paper along these lines—let's just pause there for a second, Ralph. I want to make sure everybody's following what you're saying. You're saying, look, one of the challenges of having a disease that isn't perfectly perfectly clearly defined, where every single member of the class that has the disease looks exactly the same, the word for that is heterogeneous, correct? Um, so let's take an example where the disease is very heterogeneous: sickle cell anemia. Correct? Every everybody who has sickle cell anemia from a pathophysiology standpoint is identical, correct? And guess what? There's a single mutation that defines the disease. So because you have a single gene that defines the disease, one gene mutated produces one change in one base pair that changes one amino acid that changes the uh property of the hemoglobin molecule, and everybody looks the same. But you're saying, Peter, it's totally different with type 2 diabetes. We have some people that are thin, some people that are fat, some people that have lots of insulin resistance in the muscle, some people that don't seem to have much, but it's all in the liver. We're going to go through—I want to hear—I want to make sure we define the octet, the ominous octet. Yeah. But if that's the case, why would you ever expect to find a simple genetic answer? Definition: it's going to be a mess. Absolutely. Yeah. And so if you don't have a very definitive phenotype, it's going to be difficult to—the implication, by the way, is any physician who approaches a patient with type 2 diabetes as a single entity is going to be providing suboptimal care.

Yes, and you know I've been fighting for 20 years to convince people you need to start with combination therapy from the beginning. Finally, 2022, the American Diabetes Association has made a comment and for the first time uh suggest that you should consider starting with combination therapy. We can talk about therapy later. We're going to talk about the therapies in detail, but yes, it's a—you have to take a precision medicine approach to type 2 diabetes, which begins by trying to identify which phenotype your patient is before we—

C I just want to make sure that everybody understands it's Luke Norton and and Steve Parker, and they're the you know the brainchild, you know, of course I'm involved. I understand the disease. We're doing the insulin clamps, we're giving them the phenotype, and they're doing single-cell, and it turns out like there are 10, 12 different types of cells within the muscle. So we we tend to think the muscle—oh, this is a myocyte, that's the problem—but it's probably cells also talking to each other, making it even more complex. So we're at an early stage in the development uh but we're we're enthusiastic. We think this is—I mean, we really have not discovered these genes, so we think that epigenetics are important, and and this is part of the epigenetic phenomena. We'll see where it where it takes us uh but we're pretty excited about these findings. Let's go back to the ominous octet, make sure I have that defined and all our listeners do.

Yeah, so in 2008 at the Banting Lecture at the American Diabetes Association, uh I the title of the Banting Lecture was From the Triumvirate to the Ominous Octet. So what was the triumvirate? I got the young investigator award, the Lilly Award from the ADA, 1987. So the triumvirate was very simple: the beta cell—it fails; insulin resistance in the muscle—when you uh ingested a meal, the muscle didn't take up the glucose 'cause you're insulin resistant; and insulin resistance in the liver—when you ate a meal, insulin didn't shut down uh the liver. So that was the triumvirate. So from the triumvirate to the ominous octet, we needed to add five more players. So who were the new five players? Uh, so number four on the list was the fat cell uh and a very deserving guy uh so the fat cell is your friend initially. You overeat, you take in excess calories, you store them in the fat cell. That can't hurt you there, but if you keep expanding those fat cells, the fat cells become very, very resistant to the anti-lipolytic effects of insulin, and now you start to pour fat out into the bloodstream. We've shown this as a big interested—very counterintuitive—counterintuitive, but very, very well—do do you have a—do you—not that we should mire ourselves in teic things—do you have a sense of why?

Yeah, so the insulin signaling system in multiple early steps uh become severely impaired uh and when you get insulin resistance in the glucose metabolic pathway, there are changes uh that alter the cell metabolism, so you become very resistant to insulin's anti-lipolytic effect, and so now if you look at people who are obese or people who have type 2 diabetes, their plasma FFA levels are very, very high, and those FFA levels uh and this is lipotoxicity, and we've got a long history of studying this. High uh FFA levels impair insulin secretion; High FFA levels cause insulin resistance in the muscle; High FFA uh levels cause insulin resistance uh in the liver; High FFA levels impair the insulin signaling transduction system; and in fact, one of my previous fellows who's now back with me here in UT, Dr. Belford, was the first author on this paper showing that uh just physiologic rises in the plasma FFA literally obliterate the insulin signal transduction system, which is the first step in glucose metabolism. I I always thought that the reason we saw high free fatty acids in people with type 2 diabetes was not because the fat cells were undergoing more lipolysis, but because the fat cells were themselves becoming resistant to insulin and not able to take up fat. So it's same net effect, but I was kind of drawing the arrow of causality in the other direction. No, the the the arrow is more in the other side—the fat is pouring out fat. Yeah, and you can show that the lipolytic enzymes are all resistant to insulin; they don't—we've shown this, other people shown. Yeah. Uh, so these elevated FFA levels are a disaster. So the fat cell initially—he's your friend, he's your friend—and goes to fo—then it becomes bad guy. Okay, so that's number four. Uh, number five is the gastrointestinal tract, uh and of course we'll I'm sure talk more about this when we talk about treatment, but when you eat a meal, you release two incretin hormones, uh GLP-1 and GIP, uh glucagon-like peptide one and glucose-dependent insulinotropic polypeptide. Those two incretin hormones, when you eat a meal, account for about 70% of the insulin that's released in response to the meal. So now what is the problem? Is the problem that you don't release enough GLP-1 and GIP, or is it that your beta cell is refractory to the GLP-1 and GIP? Well, it's the latter.

Just let's say that again, Ralph. I want to make sure people understand this, and the reason it's important is obviously everybody listening to us right now is very familiar with drugs like semaglutide and tepati, but I want people to understand why those drugs were were developed, and of course semaglutide is already probably what the third generation of it anyway. So when we go back in time, we'll understand why people try to develop these drugs, but just say that again. So you eat your meal, yep. GIP, GLP-1 are increased and they come out normally, yep. That's not the problem, and they're telling the beta cell, hey, make more insulin. Beta cell's deaf—not listening—is resistant to the GLP-1 and GIP, and he should be responding to 70% of his input should come from that signal—70% of the insulin that's going to come out is in response dependent on that GLP-1 and GIP. So you can imagine that that's a huge problem at the level of the beta cell in terms of the defect and insulin secretion. And tell me why is it mechanistically that that the beta cell becomes deaf to GLP-1 and GIP?

I don't know that we know the answer uh to that. So it's just another horrible piece of this puzzle where everything starts to work against the patient. So this is an area of of course intense investigation, but the clinical counterpart of this is you've already mentioned the the drugs that are out there, the GLP-1 receptor agonist. What I'm doing is I'm giving you a pharmacological dose of GLP-1, and I'm overcoming the resistance at the level of the beta cell. Now, I there's another component to this that we'll get to, and that's glucotoxicity, so uh and these were studies that were done by Yen Holtz and the group in Denmark that they took people and they uh infused GIP—we're talking about GIP—and you don't respond to the GIP. These are type 2 diabetics, and then they intensively treated them with insulin and lowered their glucose, and then when they come back with the GIP, you re-a normal amount of insulin. So this is a glucotoxic effect. Okay, so you ask me mechanism. Okay, so we know that at least for the GIP that glucotoxicity is impairing the ability, the beta cell, to response to the GIP, but not necessarily GLP-1. No, no, it and and that doesn't correct the GLP-1 problem uh so there's true resistance still, even though I normalized the glucose in terms of GLP uh one. So this incretin axis, the gut is a very important endocrine organ, and that's number five uh in the ominous octet. Number six uh in the ominous octet uh is uh the uh alpha cell and sort of the I would say the the father of you know hyperglycemia. This is Dr. Roger Unger in Dallas, and you know he was one of the very first people to show that diabetics had very high glucagon levels, and glucagon drives—tell people what glucagon does. Yeah, glucagon is—it drives hepatic glucose production. So if your glucose gets too low, your alpha cells will release glucagon. So the alpha cell can sense the glucose, and so if you're hypoglycemic, this is an important defense mechanism. You release glucagon that stimulates your liver and uh and the glucose production goes up; it returns your glucose uh to normal. But a diabetic already has a high glucose; we don't want high glucagon levels. So paradoxically, there's very high glucagon levels in uh the diabetic, and those high glucagon levels are a very important contributor to the hepatic insulin resistance because they're driving the liver to make glucose.

Sorry, just to make sure, because I'm embarrassed to say I forget this from biochemistry, is it driving the liver to make glucose out of, for example, glycerol, amino acids, orogenic pathway, in glycogenolysis? Y. But on a a more chronic—acutely, so if I acutely give you glucagon, the first thing that happens, you break down glycogen, but very quickly you get rid of all the glycogen that's in the liver, and so chronically now you're running on gluconeogenesis. But glucagon stimulates both pathways. And does it also drive hepatic glucose output, or does it just drive the creation of glucose? No, no, no. In absolute terms, it increases hepatic glucose output as well as gluconeogenesis. Yes, and that's an important reason why you have fasting hyperglycemia. So when you wake up in the morning, yeah, and your blood sugar is 110 milligrams per deciliter, that's the liver, and part of that is because your liver is intrinsically resistant to insulin; part of it is because the liver is now responding to the glucagon and producing an excess amount uh of uh glucose, both through gluconeogenesis and through glycogenolysis, although I would say the major contributor is the gluconeogenic pathway. Now uh that gluconeogenic pathway is also turned on because fat is coming from the fat cell. Remember I told you the FFA is—I yes, glycerol is coming from the fat cell, so—and then Jerry, you know, we talked about some of the work that Jerry did—this is Jerry's work—uh showing that glycerol coming from the fat cell is an important driver of gluconeogenesis, and then uh hepatic fatty acyl-CoA levels are up because you have all this fat pouring in, and that's activating the enzymes pyruvate carboxylase that are driving the gluconeogenic pathway. So the metabolic, the—the actual pathways, I think are very well worked out. So glucagon, alpha cell, bad guy. Okay. So is the alpha cell overproducing glucagon in this state? Yes, absolutely, absolutely. And this is really Roger Unger and Dallas's, and again, why is it overproducing it? Why is it doing something that doesn't make any sense in the context of what's happening? In a certain way, this is also insulin resistance because hyperinsulinemia shuts down uh glucagon, and we have very high fasting insulin levels in the diabetic. Okay. Now, what is it—what's the sensing mechanism within—it's counterintuitive. Usually when things go wrong, they get attenuated, right? Like it makes sense that the beta cell eventually fatigues because that's an attenuation of doing something that it's getting tired of doing. The the the alpha cell ramping up is a little less intuitive. You're going to see it gets even worse when we talk about the kidney, which is number seven on the list.

All right, let's go to number seven. Okay, so uh people don't know I'm uh you also board certified in nephrology. Uh, so in in in the old days, I trained as a micropuncture. I used to sit with a microscope. I I would draw my little pipettes out the night before, and and I would put the little micropipette in the tubules and I collect tubular fluid, and what I was interested in uh and this is when I was a a renal fellow at University of Pennsylvania, I was interested in glucose and phosphate transport, and I published a series—I'd say they're pretty elegant papers—in the JCI looking at how glucose and what regulated glucose in phosphate transport, and I knew that there was a molecule called phlorizin that blocked glucose transport uh in the kidney, and so I took this molecule called phlorizin, and it blocks the uh glucose transporters. There are two transporters in the kidney: SGLT2 and SGLT1. SGLT2 takes back 90% of the glucose if it does its job; SGLT1 takes back the other 10%, and then in your—even even though we filter 180 grams of glucose per day, no glucose appears in the urine. But what I showed is that phlorizin, it blocked both SGLT2 and SGLT1; it blocked glucose transport uh and it also blocked phosphate transport, and I showed that glucose and phosphate transport were co-coupled. But when I was doing these studies, even though I was a nephrology fellow, uh I had previously done my endocrinology fellowship at the NIH in Baltimore City Hospitals, so I I had an interest in diabetes, and I said this would be a great way to treat diabetes. So in the old days, you know we did things for science, and I published a series of four papers in the JCI, and uh I never even thought of, to be honest with you, of patenting this. Okay. And uh I have a significant other who said to me one day, she said, Ralph, you're one of the smartest guys I ever met, and I said, yeah, I know that, and she said, you know you're probably the stupidest guy I ever met. I said, why? She said, you could have patented this drug. And so I actually worked uh with Bristol Myers Squibb and then uh uh AstraZeneca, and that eventually led to the uh dapagliflozin coming to the market. But what we showed, and this is human bio—which was the first SGLT2 inhibitor—that's correct. Yes, brand name on that one—Forxiga, right? Yeah. And so dapagliflozin was—empagliflozin—yeah, then canagliflozin, empagliflozin, and we have a bunch of them, and they all—they're all very good—basically do the same thing. But what we showed was the SGLT2 transporter was markedly upregulated in the kidney. Okay. And uh the original studies that we did—hang on, let's just—let's just wrap our heads around that. This again, this is so counterintuitive, I know. Okay, this does not make any sense. So you take a—let's—I I want to just bring it back to people listening so they understand what we're talking about here. The kidney is this massive filtration—another remarkable organ—no offense to the nephrologists—not as remarkable as the liver, but every bit is remarkable in terms of—I think it's more remarkable than the than the liver guy. So that's okay. All right, all right. So so so everything that's floating through our plasma is you know our our our kidneys, by the way, represent—they take 25% of our cardiac output. Huge. Yeah. So it's massive—this organ weighs 2% of our weight and takes 25% of our cardiac output. Why? Because we have to take everything that is in our circulation and dump it out, and then the kidney has to selectively bring back in what's normal. This was explained to me—I still remember in medical school—as kind of a brilliant trick of evolution. Evolution was never going to be able to predict every toxic thing we might encounter, and therefore teaching the kidney how to spot toxic things and get rid of them would have been a failed mission. Rather, it was better to teach the kidney what was absolutely necessary and to discard all other things. So pretty simple way. Yep. So it's the—take everything out of your drawer and dump it out, and only bring back the socks and underwear that you need. So glucose, potassium, sodium, you name it—chloride, phosphate—all of these things get dumped along with everything else, and then it knows—I need this much glucose, I need this much sodium, I need this much potassium. So so SGLT2 does the lion's share of this. Yeah, it takes back 90% uh of the glucose. And now so here's a diabetic with a very high glucose, right? So my point was SGLT2, if it had a brain, would say, oh, you have too much glucose—turn turn it off—how about we just stop reabsorbing all this glucose? But you said it's the opposite. I told you earlier, it's going to get worse—it ramps up SGLT2. So as a doctor, I want the kidney to dump the glucose out in the urine. Yeah, but what is the kidney doing? It's doing the opposite—it's holding on to the glucose. And so even as the renal fellow, it became clear to me—this is such a simple way to treat diabetes, and the fact is it's so simple, no one thought about it. Okay. The only dumb thing that I did was I didn't patent it, which I should have done. I'd probably never have to write another NIH grant for the rest of my life. Uh, and then uh we went on to show—and in fact, this is the first definitive proof of the glucotoxicity hypothesis—so we did all of these studies initially in animals uh and this was all published in the JCI uh and Luciano Rossetti is is one of the fellows at this time—actually, Jerry Schulman was a fellow on the papers as well—and what what we showed uh was uh that uh you could take different types of diabetic animal models uh and you could show that they're reabsorbing excessive amounts of glucose, and then if I treated them with phlorizin, because that's what was available, they simply peed the glucose out in the urine, and now all of a sudden their beta cells started functioning normally; the muscle insulin sensitivity improved. So of course that's wonderful if you're a mouse or a rat. So we said, well, what about humans? And so uh the original studies actually were done—and there's kind of an interesting story behind this—but the initial studies were done with dapagliflozin, and we showed with just 14 days of treatment with dapagliflozin, we markedly lowered the fasting and postprandial glucose; we improved insulin sensitivity by 35%, and we made a major improvement in beta cell function. Now the beauty of this—SGLT2 inhibitors are only in the kidney; they're not in your muscle; they're not in your beta cell—and the only thing that the SGLT2 inhibitors do—makes you put glucose out in the urine—the only change in the plasma was the glucose came down, and now insulin sensitivity improved in muscle, in beta cell function improved, and this was the first—now in humans, even though the original studies were done in animals—first studies to show an improvement in uh uh the reality of glucotoxicity. What was interesting is that when we started to work on developing this with BMS and AstraZeneca, uh the company decided, well, we should get some nephrologists in to see about this story, and they they said, look, if you listen to what Dr. Defronzo says, this will be a disaster. And they said, why? Because if you put glucose in the urine uh it will glycosylate the proteins, then you'll cause kidney damage. And they actually held up the development of the SGLT2 inhibitors.

H and the way we finally convinced them to go ahead was that there's a disease called familial renal glucosuria—they from day one of their life to—they're bringing out tremendous amounts of glucose—they have perfectly normal kidney function. How many grams of glucose can be differentially or, you know, extra secreted basically in the presence of an SGLT2 inhibitor today? Upwards uh it it it kind of depends on what the level of your GFR is, but it could be anywhere from 40 to 60 grams, up to 120 grams of glucose uh so there's—and and the higher would be in somebody that with a higher gradient. Yeah, the higher the glucose, the higher the—yes, 'cause you filter more glucose, then there's more glucose to be uh blocked at the level of the kidney, and these drugs are very, very good. Now I I actually in in developing these drugs, as I said, I'm also a nephrologist. Based on the Barry Brenner hypothesis, I predicted these drugs would save your kidneys uh according to the Brenner hypothesis, and that's all turned out to be correct. These drugs are great for the kidney. What I never ever envisioned that these drugs were going to save your heart—that was like—that came—I mean, so I I want to come back to that because I'm making notes of other things I want to come back to, and and so I want to come back to—just so you can hear me say it now and we remember—I want to come back to combined inhibitors, the SGLT2, SGLT1 inhibitor. There's a single—I think there's a new drug that now—sotagliflozin—yeah, that it does both. We'll just touch on that, and then I want to also come back to the broader geroprotective nature of the SGLT2s as documented by the ITP in mice and then also in the human studies for for cardioprotection. But let's before we do that, let's—we need to finish the—exactly. Let's go back to number eight—the brain.

Okay, okay. So the brain plays a role in a somewhat indirect way. So every day, you know, you have your breakfast, your lunch—I actually eat only once a day—but at some point you eat a meal, and at sometime during the meal you'll say, okay, I'm hungry, I stop eating. Why? You ever think why does that happen? Well, because there are certain hormones that are released or inhibited that tell you, okay, you're satiated, stop eating. Well, one of the very important ones is GLP-1, uh that same thing that's increasing insulin secretion. Your brain has become very resistant to GLP-1. When you eat a meal, amylin comes out; it comes out in a 1:1 ratio with insulin. Your brain has become resistant to amylin; your brain is resistant to leptin. So there are a lot of these intercellular molecules uh that uh your brain has become resistant to, and uh these uh molecules—and it's another area of interest of mine—these they work in the hedonic areas in the brain, so in the putamen, in the prefrontal cortex, and they tell you to stop eating uh and unfortunately there's—and this is the big unknown—is what's going on in the brain. The neurocircuitry is clearly distorted. Not only is the neurocircuitry distorted, one of the big things that we are interested in—Dr. Peter Fox and myself at at UT—is if you look at the gray matter in these areas, in the areas that are critically important in regulating your appetite, there's shrinkage of the gray matter area. Okay. And in these areas, if you do an insulin clamp, okay, the brain is insensitive to insulin in your—in obese people, these areas in the brain with this abnormal—marked increase in glucose uptake—incredible finding—who would have thought?

I'm sorry, you're saying that these are the few areas in my brain and your brain that are actually default insulin insensitive? Yes—don't take up glucose? Correct—if I—in do an insulin clamp. So what is their fuel source? Lactate uh well, they're saturated—I'm sorry—that in response to insulin, they don't take up more glucose. Oh, oh, okay. I'm sorry, got it. 'Cause remember—as long—remember this is from the Kety studies—as long as your glucose is about 50, uh your brain is happy. So this is actually in in the evolution of the human being—this is phenomenal—because in the old days, you may not eat for—you you may slaughter one of these beasts, that—yeah, you're not eating for days; you're not eating for days. So your glucose would drop. So if your normal fasting was 80, uh if it dropped to 40, you were okay, because your brain saturated at 40. If you got below 40, you're in trouble. Uh, so uh you have a a big sort of buffer uh here, but now if I infuse insulin in your glucose as 80, your brain doesn't take up more uh glucose—it's quote insulin insensitive in a certain uh way. Uh, now of course if you take people with mild cognitive impairment, we we—there's been some experiments that actually suggest in these people insulin infusion can transiently improve glucose uptake, but presumably that's because they're insufficiently getting glucose in the disease state. Yes, this has been postulated. Yeah, and this also suggested that there's brain insulin resistance, yeah, uh which is I I'd say an interesting concept uh and may play some role in this neurocognitive dysfunction, Alzheimer's—whole kind of different story—that's sort of in uh evolution. But to come back to the ominous octet, now if you overeat, what happens? You gain weight, and when you gain weight, you become insulin resistant—severely insulin resistant—that's lipotoxicity. Okay. Uh, and we've done uh studies in both directions. I can uh put an IV, and I can infuse uh uh an emulsion of free fatty acids, and I can show within 20—within 4 hours—2 to 4 hours—I induced severe insulin resistance uh in uh the muscle uh uh in in in in the liver, and I markedly impaired beta cell uh function. And then—U—we don't have this drug in the United States uh but there's a drug that's available in Europe, and I have an IND to use it. It's called Acomplia. It inhibits lipolysis. It's like an SGLT2 inhibitor—the only thing to do is block glucose reabsorption in the kidney—Acomplia, all it does is block lipolysis; it lowers your FFA level, and we've done this uh we've done—does it result in any meaningful clinical increase in adiposity, or is it so subtle that you don't notice it? Over 12 days—no change in adiposity—huge improvement in insulin sensitivity in muscle. You can show all of these major—why is it not approved in the US? Uh, it it it it was uh I don't know that the company that developed it in Europe ever tried to get it approved in the US. It's I would say modestly effective in lowering triglycerides, and we have fibrates which are much more effective, so that may be the reason, although this drug's been around—but the triglyceride and the FFA are not the same thing. No. Uh, but that's the reason why it's approved in Europe, uh but uh if you lower the FFA—that's the precursor for triglyceride synthesis—so it it has an effect to lower the triglycerides, but the key thing is if you lower the FFA, and we did this for 12 days, we did in both obese people and in diabetic, you markedly improve insulin sensitivity in the muscle. If using MRI, you can measure muscle fat—goes down dramatically—and and correlates with the improvement in insulin sensitivity. Uh, we also measured ATP generation because there's this uh issue—is this clearly mitochondrial dysfunction if you're a diabetic—that that's unequivocal. The controversy is is the mitochondrial dysfunction causing the insulin resistance, or is the insulin resistance causing the mitochondrial dysfunction? So in this study uh that we did when we lowered the FFA uh and lowered the muscle uh lipid uh content, we saw uh about a 50% improvement in ATP generation—mitochondrial ATP generation. So at least this says that part of the mitochondrial dysfunction is secondary to the lipotoxicity and insulin resistance uh but this still remains I I would say a controversial topic. Clearly this mitochondrial dysfunction, if you can improve it, that's going to improve insulin sensitivity. Is there anything that improves mitochondrial function more than aerobic exercise training? Pioglitazone—the drug—how so? I can't get people to use—which is a phenomenal—it uh by activating PPAR gamma uh it it does a lot of good things, and one of the important things that it does—it has a huge effect to improve mitochondrial uh dysfunction uh and it has direct effects; it also works on the—it works directly through PPAR gamma to do this, and it also binds directly to the mitochondrial pyruvate carrier uh and that uh influences uh flux through the uh through the you know the the mitochondrial uh chain. And why don't people use this drug today? Huge misconceptions uh so we'll I guess we'll talk about therapy uh we'll come back to it. Okay, we'll park it until we come back to—part of my triple therapy regimen uh I use a GLP-1 receptor agonist; uh I use pioglitazone; and I use an SGLT2 inhibitor. That there's a fourth good drug, and that's metformin. Uh, and you might ask, well, why is metformin number four on my list of good drugs, since I single-handedly brought metformin to the United States in 1995? No other endocrinologist involved in this—1995—metformin was a revolutionary drug. Why? We had insulin, sulfonylureas. Okay. So now we had a drug that really could work. Okay. And uh it's still a very good drug, and of course it's very cheap—it's $5 a month in the state of Texas—but we have much better drugs. Pioglitazone causes weight gain. Okay. Now here's the problem, and it'll become very obvious—you know we we talk about these paradoxes—the more weight gain, the the greater the drop in A1C; the more weight gain, the greater the improvement in insulin sensitivity; and is it fat gain specifically? No. Oh, I'll I'll come back to that in a second. Okay. Uh, it is fat weight gain, and I also believe muscle weight gain. Okay. Uh, the more weight gain, the greater the improvement in beta cell function; the more weight gain, the greater the drop in blood pressure; the more weight you gain, the greater the drop in triglycerides; the more weight you

Of the lipotoxicity, I don't believe that this is the genetic basis, the genetic ideology. This is when you get fat and you start putting fat everywhere. This is very important, uh, critically important. And that was when he gave his Banting lecture. And I might say I'm delighted that I got to write his letter of nomination for the Banting lecture. He was incredibly deserving; he's done phenomenal work in this area. Uh, but that was his Banting lecture, and you're right, very, very, very important mechanism of insulin resistance.

And so, given that that's both a very important and very common pathway towards insulin resistance, bringing it back to PPAR gamma, PPAR gamma is part of the pathway; it's part of the IRS1, PPAR gamma, PI3K, GLUT4, bring the glucose in the cell. In other words, if people don't want to get mired down in this, which is totally understandable, insulin hits a receptor; that receptor kicks off a cascade that ultimately results in a little tube being like a little straw that goes into the cell surface that allows glucose to freely flow in against its gradient. But remember that same pathway also activates nitric oxide synthase. That's right; it generates nitric oxide, and that's why we see in patients with insulin resistance, even if glucose is controlled, cardiovascular disease is still up. A very important, yeah, very important point.

Um, so back to PPAR gamma. So, so, so now what does it do? Uh, it activates that signaling pathway; you generate nitric oxide; now you vasodilate. That's why the blood pressure drops when you vasodilate. So I'm a nephrologist; I understand this very clearly. Anytime you under-perfuse the kidney, you hold onto salt and water; you become edematous. Okay. Uh, and so people associate fluid retention and edema with heart failure. So we did the definitive study; it's published in *Diabetes Care* in 2017. People just don't read, okay. So we took people who had diabetes, uh, and we treated them with pioglitazone. And then, using NMR, very, very sophisticated techniques, what we showed is pioglitazone markedly improved myocardial blood flow. Now these numbers are going to blow your mind away: myocardial insulin sensitivity with pioglitazone and fluid deoxyglucose improved by 75%. Your heart, we showed this before, is severely insulin resistant. I came pretty damn close to normalizing insulin sensitivity in your heart. Now, since we're doing the insulin clamp with trace glucose, you can track it: 74% improvement in skeletal muscle insulin sensitivity, exactly the same. If you look at ejection fraction, it went up by 5 to 10%, not down; it went up. If you look at every measure of diastolic dysfunction, E/A, E/e', LV peak filling pressures, etc., cardiology people understand this. The point is, whether you're looking at systolic function or diastolic function, it all got better. So it's a victim of maybe not so nuanced thinking about the drug.

Yeah. Now the critic would push back and say, "Okay, Ralph, but don't we have better drugs?" Like, I mean, no drug that corrects insulin resistance. That pioglitazone is not an insulin sensitizer, and people keep going back to this. So I brought metformin to the U.S. in 1995. I know this; I did all the mechanism of action studies, and what we showed was the insulin clamp; the drug absolutely does not improve insulin sensitivity. So let's talk about metformin; everybody wants to know if metformin is geroprotective, but let's just remind people, um, we, so, so metformin inhibits complex one of the mitochondria, is that of the electron transport chain? Is that a given? Uh, yes. In, I'd say this is still controversial. Okay, in high doses for sure, yes. And the kind of doses you see with giving metformin, I would say somewhat equivocal in my opinion. So is it, is the belief that metformin's efficacy in diabetes is through reducing hepatic glucose output? That is 100% true. Okay. And what's the mechanism by which it reduces hepatic glucose output? Inhibiting the mitochondrial chain in inhibiting gluconeogenesis? Well, for sure it inhibits gluconeogenesis. Okay, okay. Uh, now, uh, metformin gets into cells through the organic cation transporter. The organic cation transporter doesn't exist in muscle; it can't possibly be an insulin sensitizer in muscle. You're asking the drug to do something that's impossible. And if you give labeled metformin, does it get into muscle mitochondria? No, it doesn't get into muscle at all. Why does lactate go up when people are taking metformin? Level of the liver; it's interfering, uh, with aerobic metabolism, and so there's a block. And this is very important: I have erroneously always believed that, so I'm really happy to be corrected; I love being proved wrong. I have always believed that the reason we saw an increase in fasting lactate, even in healthy people, if they took metformin, was because of the inhibition of the ETC in skeletal muscle, and you're saying not possible; it can't get into skeletal muscle. As absolutely not a single molecule in the world of metformin has ever gotten into any skeletal muscle anywhere. And tell me again why? What's the transporter? The organic cation transporter. That's the transporter that, uh, by which metformin enters cells. It does not exist in skeletal muscle; it does not exist in cardiac muscle. So metformin cannot get into these tissues. It's a huge major misconception. It can, if you have very, very high doses, that can occur when you have very low GFR, uh, because metformin is excreted via the kidney. If the metformin levels build up, you can get lactacidosis. That's a very, very rare complication. This, there's not a reason why you shouldn't be using metformin. And I'm not saying that metformin is not a good drug; it is a good drug. I don't think it's as good as the other three drugs we talked about, uh, but yes, it does, at high doses, increase the lactate level, all in effect on the liver. And the old drug that caused all the problem was fenfluramine. Ibuprofen as well, but it had a powerful effect. Yeah, fenfluramine was much more powerful, yes. And that's why people were afraid to, uh, bring metformin to the U.S. And when you say high dose, I mean, is two grams a day of metformin? That's the normal dose. That's the normal dose. Okay.

So metformin has the following going for it: it's free, very cheap. Yes, it's basically free. Yeah, it's free, absolutely. And it does a pretty good job at reducing hepatic glucose output. Why? And it has no myotoxicity; frankly, any toxicity, GI, yeah, the GI, but you can usually overcome that with a slow ramp-up. Yeah. And see, this is the reason why some people thought it's an insulin sensitizer: 15 to 20% of people have significant GI side effects, and they lose weight. And if you look at the studies, on average, there's about a 3-kilogram weight loss with metformin. And when you lose weight, you can improve insulin sensitivity. So I think this is what's confused some of the old literature to make people think that metformin was an insulin sensitizer. But when we developed metformin, uh, and I did all of the work that went to the FDA, if you look at the *New England Journal of Medicine* article, 1995, there are only two names on the paper: myself and a PhD oncology lady who was the person from Eli Lilly Pharmaceuticals. We did insulin clamps, many of them; we never could show metformin improve insulin sensitivity using the gold standard with radioisotopes.

Do you, do you think many people—I feel like I'm asking you this question a lot, and it's getting a little old—but do you, do you get the sense that most people are still thinking what I think, which is that metformin gets into the muscle? Metformin is an insulin sensitizer, absolutely. And it's an insulin sensitizer by getting into the muscle and inhibiting complex one, absolutely. People have done PET studies, so you can label metformin, and you give it, okay, and then where do you see it? It's all accumulating in the liver, okay, in the first 3, 4, 5, 10 minutes. And then what happens? You start to see it, uh, accumulating in the kidney, okay. Why? 'Cause that's where it's excreted. And then wait another 5 or 10 minutes, you see bladder, okay. And that's the only place where you see metformin; you never see it in the muscle, okay. And that's even more graphic demonstration, you know, uh, that metformin is not getting into muscle, and it is definitely not an insulin sensitizer.

Is there a downside to using metformin in combination with the other three drugs of which we know? The classic study, which we'll talk about, which to me should change the entire approach to treating diabetes, it's called the EDIC study. And in the EDIC study, what we did is we used triple therapy right from the beginning. And my, my, my point of the Banting lecture, the ominous octet: if you have eight problems, and there I'm sure going to be more to be found, and I can give you a few more if you want, but if you have eight problems, why in the world do you think one drug is going to correct eight problems? It ain't going to happen in our lifetime. So the point was, you need to use drugs in combination. So we said we're going to use what, and we, uh, what we think are the best drugs at the time. So we started with metformin, with exenatide, an old-time GLP-1. This is not the kingpin; this is the pre-glucagon-like peptide-1. Yeah, exactly, because that's what was available. That drug was useless, wasn't it? No, it's a good drug. Sure, it's not semaglutide, but you have to start somewhere. Yeah. Let's, let's pay it its dues as being the Gen 1 OG version of that drug, without which we might not have, we wouldn't have semaglutide. Yes. So it's kind of an old-timer, and pioglitazone. That was the triple therapy, okay. Uh, and then we said every diabetic patient—there are 315 people in this study—they're having insulin clamps, hypoglycemic clamps, muscle biopsies. No one in the world can do this study: 315 people followed for six years. Okay. So we said this is what we believe is the appropriate therapy, and then we said we'll use the ADA approach. The ADA approach is you start a metformin, okay, and when you fail—even though explicitly said—the next drug that's used is sulfonylurea, and then the third drug that's added is insulin. And we said, uh, that the goal of therapy was an A1C of 6.5%, okay. Uh, and that if your A1C, uh, rose above 6.5%, uh, either on our triple therapy or on the stepwise treat-to-fail approach that the ADA says—ADA says start metformin, and you fail, you add sulfonylurea, you fail, you add insulin—you titrate the insulin, basal insulin, up to 60 units. And we said 60 units is really, well, we'll cap it because, yeah, you're already at 2x physiologic, and now, now you have to, now split the dose of insulin; you have to be adding rapid-acting insulin. I think this is quite reasonable. Six years later, okay, 29% of the people with the ADA approach have failed; their A1C is above 6.5%, six years later. With our approach, 70% of the people have an A1C that's less than 6.5%. Why? Insulin clamp, huge improvement with our therapy. Okay. This is the EDIC study; the three-year data published, the six-year data we're writing it up. How much improvement in insulin sensitivity with the ADA approach? Zero. Beta cell function? You have almost a normal beta cell, Ralph. Why? Disconnect between where, between, between what you're seeing in the EDIC study and what the ADA is promoting? Uh, you have to ask the ADA what their answer is. If I'm a patient and I'm, and or if I'm a physician who's treating these patients, and I'm saying, "Guys, I'm confused; I'm looking at the literature; I'm seeing this; I'm looking at your," and by the way, I see this with the AHA and cardiovascular guidance, so I'm not singling out you or, or, but but like, is this simply a question of the pace at which medicine moves? It's so glacial. That's part of it. Plus, remember, if to do 315 people, follow them for 16 years, and do all the stuff we did, it's unequivocal. And why has there not been political pressure? Because the cost of insulin is enormous, and your approach is going to be less expensive. They finally said in 2022, there's a statement: the ADA approach is not based on pathophysiology. I, I view myself as a scientist, okay, as well as a clinician, as a good clinician. I've taken care of hundreds of thousands of patients, and I've done 850 publications. Uh, I do clinical research; I work on people. When I do an insulin clamp study and I see an improvement in insulin sensitivity, I do a hypoglycemic clamp, and I see in 315 people your beta cell function. I don't need 5,000 people; I don't need, I can't do this study in 5,000 people; no one can do this study. But the, the tools that we're using are so powerful. Look, if I normalize your insulin sensitivity and I give you a normal beta cell and your A1C is less than 6.5%, and it's well, it's half of the, half of the 315 people, why don't you think that's the best therapy? Okay. And now on the other side, I have this metformin, sulfonylurea, insulin, and that 71% of the people have failed; there's zero improvement in insulin sensitivity, zero improvement in beta cell function. Why do you think that's such a good regimen? And now, above and beyond all that, I didn't do this study; this is the GRADE study, it's sponsored by the National Institutes of Health. And what the GRADE study said is, uh, and I have to say this is the third study that's shown what I'm going to tell you: Dr. Robert Turner's United Kingdom Prospective Diabetes Study showed this, uh, in, uh, 1990; Steven Kahn showed this in the ADOPT study in 2005; and now we have the GRADE study, 2020. I call this the 15-year revolution, revelation. We saw what didn't work in 1990, oh, Steven Kahn did it again, oh, it didn't work in 2005, and now 2020, NIH did it; you know what? I'll show the same thing. And this was a sequential approach; you had to have failed on metformin to get into the study, okay. So you failed on metformin, then you enter the study, then we go single agent. I can, they, they want to know what's the best next drug to add to metformin. I can add a sulfonylurea; A1C went down in year one, up straight. I can add—tell folks how sulfonylureas work. Yeah, sulfonylureas, uh, are old-time drugs; they bind to the sulfonylurea receptor on the beta cell, and they kick out insulin, and they're very good drugs in the first year, and then they burn out the pancreas. Well, they stop working. Yeah, I mean, basically, they, they kick the can down the road without addressing the pathophysiology. I like that way. Yes. Then other drug: DPP-4 inhibitor. Tell people how those work. Yeah, so a DPP-4 inhibitor increases your GLP-1 and your GIP level, uh, endogenously; it makes your gastrointestinal cells, the K and the L cells that secrete the GLP-1 and GIP, makes them make more GLP-1 and GIP. But it, it doesn't increase the GLP-1 and GIP enough to really give you a knockout punch. I give you an injection—you all people are out there—of monjaro or semaglutide; that's the knockout punch. Okay. When I give you the DPP-4 inhibitors, they, yeah, they, they do increase GLP-1 and GIP a little bit, but not powerful enough to give you a long-lasting effect. So first year, A1C comes down; A1C goes up. Uh, third drug—this was very surprising to me—this was liraglutide; this is one of the earlier GLP-1 receptor. Yeah, I thought that was going to work the best; it failed. It worked in the first year and then failed. And then the fourth drug was insulin, and the docs just didn't titrate the insulin enough, so A1C down, and then they failed. So five years later, all four of those regimens added to metformin failed. Triple therapy: exenatide, an old-time GLP-1, pioglitazone, which people don't appreciate, the only true insulin sensitizer, and metformin. Six years later, you're 7% of the people have an A1C less than 7. And, and let's just go back: metformin is free; the Gen 1, uh, exenatide, basically free, is basically free now; pioglitazone is $5 a month. Okay. So we have three free drugs that work better, correct? Now it's interesting when you talk about today's triple therapy, which is way more efficacious; two of those three drugs are very expensive. Yes, the SGLT2 inhibitors are very expensive, and the modern-day Gen 3, Gen 4, and soon we'll have a Gen 5 GLP-1, they're very pricey; they're very $1,000 a month for the, yeah. Now are they great drugs? So the question is, yeah, is, but but I guess the question is, do they, do you need to be on those drugs if your old version of triple therapy—our old version is incredibly effective. The problem is, you can't get people to use pioglitazone, and the reason is the patients are frustrated with the fact that they're retaining water, no more than they, they gain weight. How much weight do they gain typically? How many kilos? Depends on the dose. I like, I don't go to the 45-milligram dose, so at the end of the year they may gain 2 or 2.5 kilos at the 15- and 30-milligram, uh, dose, okay. But their A1C is controlled. So you have to explain to people: here's another incredible thing, uh, so there was a cardiovascular—this truly incredible, by the way, I'm sorry to interrupt you again—if you give pioglitazone plus a modern-day GLP-1, offset the weight, you cannot believe—oh, you lose all the weight; you go lose with the GLP-1 receptor. So if a patient is willing to go down the path of a modern-day GLP-1, doesn't that completely eliminate—absolutely. And it also gets rid of the edema, yeah. And believe me, their A1C's are down in the normal range. You know, I, if you tied my hands, so, uh, let, let me tell you this first thing about pioglitazone and the PROACTIVE study. I'll come back, and we'll answer this. So in, in the PROACTIVE study, uh, this was done a long time ago; you have to show cardiovascular safety: 5,238 people to get in the study, you had to have an MI, stroke, or something bad, uh, half people on pioglitazone, half the people on placebo, okay. And, uh, the MACE endpoint, major adverse cardiovascular events, which is non-fatal MI, non-fatal stroke, cardiovascular mortality, you have to show the benefit to get approval by the FDA. The, the MACE endpoint was positive, okay. And so, uh, when I talk to cardiologists, I like to say, what was the one thing, uh, in the pioglitazone that predicted that you would not die? They don't know. You know what the one thing that predicted that you wouldn't die? Weight gain. So I jokingly say, look, you can either be a little fat and alive, or you can be lean and dead; which one are you going to pick? I think I go for being a little bit chubby. So again, but now we, that's not even a necessary comparison; you don't even need to make that tradeoff with a modern-day GLP-1 agonist. And we've done this, and we've published this, uh, if you tied my hands behind my back and said, "Ralph, you can only pick one drug," I would pick one of the newer GLP-1s; they're incredible drugs now. But, but that's not what I'm going to do. Even for a lean diabetic, a little bit different story, but the answer is basically yes. Okay. But let, let me, okay, narrow that down a little bit, uh, if I had to pick two drugs, I would pick pioglitazone with one of the newer drugs, and, uh, for sure, uh, if you had any kind of renal or cardiac disease, I'm going to pick an SGLT2 inhibitor. But I would say, although this study will never be done, if you're a newly diagnosed diabetic and you don't have any cardiac symptoms, why do you think that the SGLT2 inhibitor is not doing all of the beneficial things in that newly diagnosed diabetic that it's doing in the people who get into these studies who already have cardiac disease? So if you have a cardiac problem, I put you on the SGLT2 inhibitor; you're less likely to have MI, stroke, etc.; it's doing good things. It's doing, in my opinion, the exact same good thing in someone who I'm just diagnosing for the first time when I put them on the SGLT2 inhibitor, but no one is ever going to do a study; it's impossible. I'm going to take a thousand people, and you, you have to take five, you probably have to take 20,000 people newly diagnosed, uh, and then 10,000 go on SGLT2, and 10,000 on placebo. I'm going to follow them for 20 years to see who's going to have their heart attack. No one's going to do that study because they're going to get on all kinds of drugs. So no one is going to be able to show, but that's sort of like, yeah, that's never gonna happen. But I also don't think it needs to happen in the same way that, um, I agree with you, in the same way that we saw, for example, you know, PCSK9 inhibitors reduced MACE in people with secondary prevention—take people who had already suffered MACE, put them on a PCSK9 inhibitor, you secondary prevention, reduce subsequent—well, of course, everybody's using these for primary prevention now. That's effectively what you're saying: we already know the SGLT2 works for secondary prevention; that may never get approved for primary prevention, but it probably justifies—I agree with you 100%. So, so just to make sure I'm synthesizing what you're saying, Ralph, um, if you only get one drug and your price agnostic, GLP-1 agonist. Yeah. If you get to add a second drug, you're going to add pioglitazone. Yeah. If you get a third drug, especially if you care about your heart, SGLT2. And what's amazing is metformin didn't even make the top three in your list, but it's never far, so here's my question: given that metformin is free, yeah, should we just be adding it the second we put on the GLP-1? I don't have any problem with that. Why? And also, we have to be cognizant of the fact, you know, these newer GLP-1s, they, we're talking, yeah, but they're $1,000 a month. Yeah. So, you know, I want to ask you about that. So, so the, just again for the listeners, right, the semaglutide, Gen 3, uh, tepatitide is Gen 4, retatrutide is coming out, assuming the phase three goes according to plan, yeah, and KSEMA is the Novo Nordisk one. Yeah. So here, let's go back to retatrutide, yeah, GLP-1, GIP, and glucagon. Glucagon, can you explain that in the context of the octet, where glucagon is going up? Uh, I can, I think, okay, but it's not proven, okay. Uh, so, remember I told you that insulin knocks down glucagon. So if I give you it a GLP-1 receptor agonist and I kick out insulin and I get you, well, insulinization, okay. So that glucagon effect to drive, uh, glucose production will be totally blunted by the insulin secretory effect. And see, this, the other thing that bothers me about these GLP-1s: these are the best drugs in the world for losing weight; these are the best drugs in the world for saving your beta cells. I told you that when you, when you eat a meal, 70% of the insulin that's secreted is coming from the GLP-1 and the GIP. People have stopped talking about this effect on the beta cell. This, I told you, if you want to look at type 2 diabetes, big problems: beta cell failure, insulin resistance; these GLP-1s, they're saving your beta cells. I mean, we've forgotten about, we've become so enamored with the weight loss, and I don't want to downplay that at all, because the weight loss and the lipotoxicity, a huge problem that's causing the insulin resistance, but people have forgotten how powerful the drugs are on the beta cell. So when I give you this drug and they work on the beta cell and kick out insulin, any negative thing that glucagon's doing will be totally negated. Now you may see some good things that glucagon's doing that we couldn't appreciate before. So what are the good things? Some people have suggested that increases thermogenic energy expenditure. I don't believe that; there are animal data; I don't believe this in humans. I believe that it's exerting an anorectic effect in the central nervous system that is, I think, yet to be established. I'm pretty sure there are studies going on now at the Pennington Institute and maybe also in Orlando, uh, where they have these chambers where you can, yeah, yeah. So, uh, I think we'll get an answer about, you know, energy expenditure. I, yeah, I would be surprised if they're going to see a clinically meaningful increase in non-involuntary energy expenditure. I'm with you; I think it's all appetite. And here's another issue, okay. So it is very interesting if you look at all these big GLP-1 studies, cardiovascular: what's the reduction in cardiovascular events? Almost uniformly 20%, old dudes, exenatide, etc., liraglutide, new dudes, 20%, even though the weight loss with the newer ones is much greater, much greater. I suspect, in terms of cardiovascular benefit, there is a cap, uh, that once you've lost a certain amount of weight and you've gotten a certain amount of lipotoxicity and all the good things that these drugs are doing, you don't go beyond that, even though you're losing more weight. And also, if you look at the A1C, you know, yes, monjaro does drop the A1C a little bit more than semaglutide, but they're both pretty powerful. Retatrutide does a little bit more, and does kSEMA do a little bit more, but they don't do a lot more. So I also think there's also going to be somewhat of a cap on how much you drop the A1C. You know, you get a 2.5% drop; can you drop at 3? Do you need to drop at 3?

So you're saying if a person shows up with, um, hog-glob A1C of 9.5%, why—this is a person who hasn't come to medical attention soon enough—and I'm going to give you the answer definitively, but I'm going to let you ask the question. You're happy if they only go from 9.5% to 7%? If they only had a 2.5% drop, you wouldn't try to get them down to 6%? I would. And we've done the study, okay. Old-time guys, all right. So this called the QATAR study. So there's this, uh, concept that's out there, and again, people don't—what's always amazing to me, see what drives me is science. If you understand pathophysiology and there's an abnormality and you correct the abnormality, things get better, okay. So in the QATAR study, uh, and there are 220 people or so in the study; to get into the QATAR study, you had to be poorly controlled on metformin and sulfonylurea. So you had to have failed on this, and the average A1C was about 10, uh, and, uh, about a third of these people were symptomatic, meaning they had polyuria, polydipsia; they were losing weight. And so the current concept is, in those people, you would put them on a mixed split insulin regimen; you would get rid of the glucotoxicity, you get rid of the lipotoxicity, and, uh, you get their A1C down to 6.5%, and then now you could put them back on the oral medications or whatever, and now they respond 'cause you got rid of the glucotoxicity and lipotoxicity. We said, "Well, that may or may not be true." So we said, "Well, half of these people are starting with an A1C above 10, uh, will go on, uh, a mixed split insulin regimen, okay, with a glargine and a rapid-acting insulin, and the other half are going to go on that old dude, exenatide and pioglitazone, the one, one that people don't like to use." Okay. Three years later, the A1C in the group with the mixed split insulin regimen is 7.1%, uh, and we're very good at insulin; why couldn't we go lower? Because we got into trouble with hypoglycemia, okay. The A1C in the group treated with exenatide and pioglitazone is 6.1%. And so then we said, "Okay, look, uh, we'll, we'll look, we'll do a subgroup analysis." And so about one-third of the people were, we, we'll just look at the people who are symptomatic; the starting A1C is 12.2%; 3 years later, they're A1C is 6.1%. How long? 6.1%, 3 years later, 12%, yeah, 12.2%, symptomatic, on which combination? Exenatide and pioglitazone, without even metformin, without—they had failed on metformin to get into the study, okay. So what we're saying, look, if you have drugs that correct the insulin resistance—that's pioglitazone, GLP-1—again, Ralph, this is almost impossible for me to imagine. I can send you all the papers; it's all published, but I just, you know, I just, I hope every single family medicine internist, everyone who ever takes care of somebody with diabetes is listening. I hope so too, because you're basically saying we can take these two old cheap drugs and take someone from the most brittle type 2 diabetes—I mean, a hemoglobin A1C of 12, you're knocking on death's door, correct? You're going to go blind; you're going to have your toes amputated; you're not ever going to have an erection again, and you're going to die of cardiovascular disease or kidney disease or Alzheimer's disease quickly. These numbers that I'm telling you, they're right from the paper, and it's a large over 200 people, and in a couple of years on two old cheap drugs, you're normal. Yep. And what's better? What makes these studies so solid is we have very sophisticated pathophysiologic measurements; no one can do what we do. So, so the pushback is those patients are going to have to gain a couple of kilograms of, but, but of course, if, if you're willing to now spend a bit more money and switch them from Gen 1 to Gen 3 or Gen 4 GLP-1 agonist and GIP, then all of a sudden you ameliorate that, and you get all the benefits. This becomes a nonissue. Put cost aside. In fact, if you, I would wonder if you add metformin, you almost cancel out the weight gain a little bit because you might get a little bit of the GI improvement, and you get the 2 to 3 kilograms of weight loss there. Minus these drugs are so powerful; when you put them with pioglitazone, you don't get the, I mean, you lose almost the same amount of weight; they're huge in terms of getting you to lose, lose weight. I mean, so which, which was that study? Uh, this is called the QATAR study; it was done in, in Qatar, QA. Oh, like Qatar, the country. The country. So, uh, and I need to give credit to Dr. Bahmad AbuGhani, who's been sort of, you know, my, my coworker in all of these, uh, studies, and Mohammad's on the faculty at UT in our diabetes, uh, division. Can we at least assume that the Gulf States are paying attention to this? A, the study was done in Qatar; B, the Gulf States are disproportionately ravaged by type 2 diabetes. Yep. Is it, is it at least being heated there? They are. And I can tell you we have a big program, uh, that's going on there as well as in Kuwait, uh, and we actually have a formal, uh, cooperative agreement with the Qatari people, people, so at the DMIN Diabetes Institute, uh, we have, uh, trained them; my people have been there, trained them how to do these insulin clamps and, you know, sophisticated metabolic studies, and they take care of the patients, and we're comparing the genes. So here's another thing that's pretty exciting that we're doing. So, uh, and again, it's looking for genes that cause diabetes. So you eat a meal, okay, you eat a meal; your glucose goes up; that secretes insulin; there are amino acids in the meal that secretes insulin; GLP-1 goes up, and that secretes insulin. So now when you eat a meal, there are already three stimuli, and now you're looking for a gene or a set of genes that might be associated with beta cell failure when you have three stimuli. Now that's going to be pretty confusing. So what we said, what maybe what we should do is that we should do a three-step hyperglycemic clamp, uh, so we give you three steps of, uh, glucose, and we can get beta cell sensitivity to glucose from the slope. I give you a little rise in glucose, another rise in glucose, another rise in glucose; I see how much C-peptide

Do you have any concerns with long-term safety or anything other than simply the economics of the GLP-1's in this current generation? Again, huge, huge leap forward between liraglutide and semaglutide. And uh, I've discussed briefly elsewhere on the podcast what the roadmap looks like for how many of these drugs are in the pipeline.

Oh yeah, there seems to be no end in sight, and we're going to look back at semaglutide and say, "God, that thing was pedestrian." That's what's going to happen. So what, give us the bare case? What should we be concerned with? What should we at least be looking out for?

Yeah, I would say overall at the present time, uh, I would consider these drugs to be quite safe. The major issue is you have to go slow because of the GI toxicity. Where is the controversy involved? And it's something that I'm involved with myself. Uh, when you lose 20 or 30% of your body weight, you lose muscle mass. Okay. Uh, now, uh, I just gave a talk on this to one of the pharmaceutical companies that are involved in this area. I'm not going to name the name of the pharmaceutical company, uh, but I started off, uh, by saying, "Look, here is now a study with real data. This is a gastric bypass surgery study, RY bypass, and the people lost, I think it was 33% of their body weight, and their lean body mass came down quite significantly." One of the problems is people measure lean body mass, and that's not a real measure of muscle mass; in fact, it can be a very bad measure. You should measure muscle mass, but let's assume that the lean body mass largely reflects—it's a reasonable assumption—muscle mass. So muscle mass came down. Okay, so why is that so bad? But how much did it come down? Because if your total body mass came down by 33%, but 3/4 of that mass was fat and only one quarter of that was lean, we would consider that acceptable. And this is where the controversy is, cuz no one has really measured muscle mass.

That's right. We're doing it. Okay. We will have a definitive answer. And you're doing that with MRI? Yeah. Okay, so it's the gold standard. Okay. But now I said, "Look, in this study, they measured absolute strength." You can do grip strength or leg strength, and absolute strength went down a little bit, okay, maybe 25%. Were these patients exercising during the period of weight loss? No, no, no, no, no. Then they said, "Let's express strength for weight—loss up by 50% per appendicular." It goes up by 50%. And then they said, "How far can they walk?" They went from walking 200 yards to two miles. And then say one of the things is how many times can you get up out of a chair in a certain period of time? They increased like three or fourfold. And they measured your V02 max.

Yeah, of course, which is heavily dependent on weight as well. Yeah, it all got better. But in absolute terms, did V02 max get better? Not necessarily. Yeah, the total V02, not normalized per kilogram. No, everything got better. Okay. So, uh, so I said, "So why you guys—" That's counterintuitive, by the way. Normally when you lose weight, V02 max in liters per minute does not improve because you have less metabolic tissue. But, but uh, here, for whatever the reasons are, maybe all of the fat that's pushing on your lungs so you can't oxygenate, the epicardial fat, uh, that's not allowing your heart to contract, the fat that's in the heart that's causing myocardial lipotoxicity, which I believe is real, these things are all changing in a positive way. Uh, so again, it's a balance of yeah. So, uh, of course, they don't like this because if the people—and they had all kinds of psychological questionnaires—wait, why weren't they happy with these results? Well, because uh, they now the companies are all looking at developing drugs that will preserve the muscle mass or increase the muscle mass. But basically what I'm saying is that look there's a huge—it's lean body mass, but we have to say it's reflecting muscle mass—everything gets better, the patient feels better, they can walk better, they feel stronger, etc., etc.

Why are you so worried about muscle mass? Like I look at all these gloomy faces because they're all developing myostatin inhibitors or even—and so then the next slide comes up and says, "Retort: Here's a good thing." Okay, so now, uh, if you uh, lose all of this body weight and you improve insulin sensitivity in muscle, uh, and you improve it in the heart, uh, and there are cardiovascular benefits and you correct the Improvement in all of the cardiovascular risk factors, now even though uh, you've lost muscle mass, if you've improved insulin sensitivity, there may be an enormous benefit of seeing the Improvement in the muscle insulin sensitivity even though you've lost muscle mass. Uh, and they do have some concerns about these drugs, these myostatin inhibitors that actually may have some negative effects in the heart. And my suggestion is actually you may find a big Improvement in myocardium function uh from uh these—how—where are myostatin inhibitors in their development right now?

Phase two. Yeah. Yes. And um, of course, I think we've talked about myostatin before on the podcast. When you inhibit myostatin, you increase the expression of striated muscle, of which cardiac is striated, work through the Akt mTORC1 pathway.

To A and 2B system. Um, do you think that's a more promising pathway than the follistatin pathway?

Where follistatin? Yes, I do. Increasing follistatin inhibits myostatin, but this is a more direct way to go about it. This is a more direct way to do it. So you can either have their antibodies—by Guab—to uh, myostatin, or you can interfere with the signaling receptor itself. And we think that this can still be effective in a fully developed and mature adult. Clearly, this would be effective during development, and we see that in the animal work. Is it—how effective is it? Cuz a lot of the animal work is sort of a characterization stuff. It's knockouts, right? They take myostatin knockouts, and they look like bodybuilders. Yeah. But if you take a mature chicken or a mouse that's two years old and you give it a myostatin antibody, how robust is the response? Even more so, what about in a human?

Yeah, exactly. We don't know the answer to that. So what the phase two studies—obviously, the toxicity passed in phase one.

Yes. Yeah. I, I, I'm not concerned. There doesn't seem to be any adverse effect of these drugs, or they wouldn't have gotten through phase two. And they're actually some fairly large phase—what's the indication? The indication is sarcopenia. I, I don't know. I'm, you know, the uh, the FDA, if you have a sarcopenic disease, there are criteria that the FDA has established. If you want to develop a drug, you have to meet certain criteria. I'm not an expert in this, so I, I can't tell you exactly what these criteria are, but they are pretty well established. Okay. Uh, now for these kind of people, and I'm going to come back—you ask me about lean people, I'll come back to that in a second—because this is really an issue. Uh, what—let's say I put you on a GLP-1 receptor agonist and you lost 25% of your body weight, and I put you on a myostatin inhibitor and that prevented the muscle loss—didn't increase it, but just prevented it—so relatively—but that would be ridiculous. I mean, if you took a 200-pound individual, yeah, who's 30% body fat, yeah, so they've they've got 60 lbs of adipose tissue on them. Y—if you took 25% of their body weight off, you take them down to 150 pounds, but you're telling me potentially we prevent any deterioration of lean mass. Yeah, that means they're down to 10 lbs of fat mass on a 150-lb frame. So I'm making an assumption—this is pretty rad—this is remarkable, right? So let's say that happened. Okay, what would be the FDA's criteria? I'm going to give you approval for this drug.

I think the FDA would ask that you've also improved function in some way. Yes. And the function would probably have to be determined through absolute strength, not relative strength, would be my guess. So I, I don't know the answer to this question, cuz that's cuz the way I think about these drugs is less about that situation. Uh, it's more in the—it's more in the sarcopenic adult.

This is the—particularly the older person.

That's right. That's right. This—the elderly individual who's sarcopenic and whose fall risk is enormous. Yes. And their risk of fall and uh, morbidity and mortality is very high. Yeah. And in that individual, I don't think the FDA will be satisfied with simply an increase in lean body mass unless it is accompanied by strength. Now, I think that some of the tests that are used here are silly. I think the six-minute walk test should be folded up, discarded, put in the waste basket, and never discussed again. It is such a stupid test. They do it all the time. I know they do. Just makes me want to scream. We need much more rigorous tests than a six-minute walk test. We need—need a test that is actually uh, more of a submaximal test. So if we're testing cardiorespiratory fitness or some sort of peak aerobic fitness, we have to do more than walking. And if we're testing strength, I much prefer grip strength, leg extension, bench press, uh, yeah, again, these can be done with machines, they can be done very safely, but we really need to test strength.

Well, you see, you're raising very important and critical issues because there are many, many companies that are going ahead with these drugs that increase muscle mass, but to me, okay, increasing muscle mass—what does that mean? There needs to be some functional translation of that. And and by the way, there could be other functional benefits that exceed strength; for example, glucose disposal could be a functional sensitivity. That's the one I put at the top of the list—get rid of the insulin resistance.

That the FDA won't give them credit for that. I don't think. Yeah. But but I think that again it's harder to tease out because there's more moving pieces, and they might argue there are easier ways to increase insulin sensitivity and glucose disposal. But you know, one way to think about this is to go back to what if you did it the old-fashioned way? What if you got in the gym and lifted a bunch of weights?

That's been done. Yeah. And it increases insulin sensitivity and functional strength. And so the question is, can we replicate that pharmacologically? Uh, and that is actually exactly the way I ended my discussion to these people. I showed them what resistance training did, and if you could show what resistance training did uh with your muscle mass uh increase, then you'd have something, but you need to design the studies appropriately. And as I said, and as you said, I, I, I don't know what the criteria are going to be that the FDA uses to judge these things. They do have a sarcopenia set of criteria, but that's a very different group of people that we're talking about. But this comes uh and hits home to one of the things you asked me earlier: What about the lean person who's 80 years of age? Is this the right drug for that person? I don't know. Uh, maybe not. But now let's say you have a healthy 80-year-old person, and everybody in the family lives to be 105 and they have diabetes. Well, they're at risk to, you know, uh, the toxic effects of hypoglycemia. Uh, would it be reasonable to treat uh that person? We know this powerful effects on the beta cell. I would say it would be quite reasonable, but I think you need to monitor what's happening to their weight and other features.

Here's a bigger issue: childhood obesity. You are obese when you're four years of age; you're going to be obese when you're an adult, and and your life expectancy will be significantly shorter, and your quality of life will be significantly reduced. Now make it even better: adolescence. These young kids with diabetes, they don't respond to any of the drugs. What is the prevalence of type 2 diabetes in underage 18? Uh, it's uh increasing, but I would say maybe around 4-5%, something like that.

Five—one in 20 teenagers has type 2 diabetes. I guess I'm biased by San Antonio because we have more people with type 2 diabetes in our clinic. But you would—you would—and you could say potentially in San Antonio one out of 20 teenagers—it's going to be very high. Gosh, that is pre-diabetes. And I'll tell you about the pre-diabetes study that we did. So now let's say you're—and we know these studies are out there—these kids, uh, uh, in this big NIH-sponsored study, uh, they don't respond to metformin and sulfonylureas. They don't respond to any drugs very well, even the GLP-1 agonists. Uh, the first study has just come out; they respond better—it's a liraglutide study—they don't have any—but if you just clinically—if you're in the clinic and you're treating them without, you know, you're using the best drugs you have available, you're, you're you're in trouble. And we know—wait, why? Because you can't get them controlled. Why? There's—they're so insulin resistant, much more—more so than adults. Uh, these are well-published studies. Is this—not so much—is the causality here not—is this really a selection bias where for someone to develop type 2 diabetes as a 16-year-old, the underlying genetics and pathology are so severe that the current crop of drugs are the problem as opposed to when you take the current crop of drugs and you apply them to people who are young, they don't work uh all three? Cuz I'm going to add one more: genetic predisposition. So Hispanic population—huge problem—obesity—all of these kids are huge. Okay. So you don't have the lean diabetic phenotype in this age.

No, not in these people. And then the drugs don't work very well in this. So all three of these things. And what now has come—it's called the RISE study—and as these kids have been followed up, they're starting to develop kidney disease. They're even—I'm told a couple of people have had MI in their 20s. They're incredibly difficult uh to control. What do you think? I mean—not that—I mean, yes, we're going to argue that these kids are—that this is due to what they're eating, but what is it in the environment that is so obesogenic to these kids? I'll come back to this in a second, uh, but I want to raise the issue now. Let's say you're 16 and you are metformin—and these drugs are not—your A1C is nine. I—you—you can put someone on monjaro, and they're going to have to take this for the rest of their life because as soon as you stop the drug—okay, so this is uh what I treat the person. Of course, I can't let the A1C at nine, but that if you take that 16-year-old with a hemoglobin A1C of nine and you give them monjaro, where are they in a year? Uh, I think that if they can afford the drug and they stay on the drug—the three big ifs: if the doctor knows what to do—I know what to do—if the patient will cooperate with you—if you don't, they'll lose every time—and if they can afford it—if you can satisfy those three ifs—that person, we know from the studies, be pretty well controlled.

And insurance—what fraction of insured patients will have coverage on monjaro if their A1C is nine? I don't—I can't answer that. But uh, does CMS cover it? Does Medicaid cover that? Uh, yes, if you have diabetes. Okay. I don't know the Medicare coverage. I think it's pretty good if you have diabetes. Okay. If you have obesity, that's a different—if you have obesity without—that's a whole different issue. Why should you be treating these young kids who are—obesity is a disease. Okay, it's got all kinds of problems. Should you put these young kids on these newer drugs, and knowing that all I did is change you from food addiction to drug addiction? I didn't do anything else. Uh, it's almost like alcohol addiction. Okay. Uh, there are drugs that think—I can give you that can help you, but they tend to relapse. Uh, food addiction—I put you on the drug, you lose weight; you stop the drug, you regain the weight. This is a huge public health concern. I, I, I—is almost way beyond my capacity because finances are involved here. Can we afford to treat—42% of the people in the US are obese—or is there some way amongst the 42% we can define who are the people who are insulin resistant? Who are the people who have the metabolic syndrome that we know they're at risk that we can treat them? My guess is that's the great majority of that 42% of the people. Can we treat all of those people? And moreover, are they going to stay on the drug? We know on average what the data is saying: I put you on the drug—it—we don't know all the reasons why, but within a year, half of the people stop the drug. Yeah. And it's probably a combination of cost and side effect. Yeah. And uh, my patients very commonly tell me, you know, say, "Ralph, uh, I enjoy eating, and I can't eat anymore," and they—some people just tell me they, they just want to eat, and and so I'm going to get fat again, so I'm going to eat some.

Is GI side effects and some is cost. $1,000 a month is a lot of money for—it's insane. Yeah, for people. Um, so Ralph, again, I just kind of bring it back to this, this question, which I mean everybody wants to understand this, which is what has changed so much in the last 30 years that has created this epidemic. And you know, um, everybody has their favorite pet theory for what it is: it's the sugar, it's the carbs, it's the plastics, it's the video games, it's the internet, it's the whatever—perhaps suggesting that it's many, many things. What is—what is your best explanation for what's going on?

I would say all of the above: processed foods, uh, calorically dense foods, lack of exercise are critical, but these are—I would say the stimuli that has done something that's changed the neuros circuitry in the brain. So yes, there's a stimulus, and because now you've been oversubscribed to the stimuli, uh, that's now initiated a process in the brain which is going to be a self-fulfilling prophesy. This—something that I'm very interested in—Dr. Peter Fox and I at the at the Health Science Center—but if you go through the literature, and we've published on this uh as well, in the areas of the brain that control food intake, and I'm not talking about the hypothalamus that kind of regulates your your basal energy intake—what you need to be—keep your BMI of 25—do what you do during the day—but what is it that makes your BMI go to 35? Uh, those—that's all related to the hedonic areas in the brain: the amygdala, the prefrontal cortex, etc. And then uh, when you do structural MRI, what you can show is that those areas in the brain—the gray matter—is shrunk down. Uh, and if you now map the neuros circuitry, which Peter Fox uh has been involved with, uh, you can see that there's clear disruption using functional MRI of the neuros circuitry in the brain. Uh, we have a particular interest in defining where this dysfunction occurs, and we have some ideas, which I'm not going to go into, but how we might be able to sort of reprogram uh the brain. And sort of in concert with this—these are not—um—these are data that are published in the literature, and I think I mentioned this earlier: if you do an insulin clamp, okay, uh, I told you that in—your brain doesn't respond by taking up glucose, but in people who are obese, actually almost in proportion to how obese you are, in these areas in—in the brain—hedonic areas—there's a marked increase in uh, it's called fluorodeoxyglucose, which is the PET radiotracer we use in these areas, and that correlates inversely with the muscle insulin resistance. The more muscle—the more insulin resistant they are in the muscle—the more FDG glucose uptake there is in the brain. Now this is very interesting because what it says—there's a connection—that somehow or another we believe that the brain is talking to the muscle or the muscle is talking to the brain, uh, and that uh, that uh, somehow or other the brain is playing a very important role in the development of the insulin resistance, and that in large part that this deranged neuros circuitry which is related to food intake is now making you overeat. And as you overeat, then all of the things that we know that we studied that other people have studied that go with lipotoxicity—you put fat in the muscle, you're insulin resistant; you put fat in the liver, you got NASH; and—what people have totally overlooked—you put fat in the kidney, you get kidney uh disease; you put fat in your—

Yeah, yeah. So, so you've been in San Antonio since the late 80s. When did you really start to notice this was a problem, at least in your community?

Almost instantaneously, uh, because uh, I run the Texas—even in kids—even in kids—and uh, it was very—so—can't—we can't blame video games; we can't blame social media because that wasn't going on in the late 80s. So I never saw fat kids at Yale, and I was on the faculty from '75 to '88, and I, I kind of thought back uh—now I would say New Haven's not a large Hispanic—but it's more African American—but I don't remember seeing, you know, 12-year-old kids with type 2 diabetes. And when I came here, uh, and I remember this very distinctly, they're saying, "Hey, you're crazy. We—you could—you don't see kids with type 2 diabetes." Believe me, I see them. And what did your colleagues at San Antonio tell you as far as when they started to notice that in the Hispanic kids? I think they just—I, I don't know that I can give you a specific time that they told me except they knew it. But so—okay, what about in non-Hispanic kids? Because if the Hispanic kids are are genetically predisposed to this, then the question becomes when did you begin to see this in African-American kids and Caucasian kids?

Yeah, so we don't have a large uh African-American population here, but like in Philadelphia—there's a lady—Sylvia—Silvian—uh, she sees the same thing, uh, and she sees—I think it's a significant African-American population. So I think that in certain ethnic minorities where the genes for diabetes are enriched, that those are the populations that are uh predisposed.

Do you think this is mostly an energy balance issue and therefore it's mostly a food environment issue?

No, I think it's both. So I told you I'd come back to the genetic study that we did. So uh, we, we uh—in the—it's an Italian fellow was with me—Giovanni—Guly—a long, long time ago here in San Antonio. We wanted to know what is the earliest defect that you can see in uh people going to develop type 2 diabetes. So we said, okay, in the Hispanic community, it's very common to see Mom and Dad with diabetes, and it's very common to see a lot of children in these families. So we said, "Why don't we go look at the children and let's see uh if we can define"—because they're at high risk—yeah—"and if you have Mom and Dad with diabetes, you probably have a 70-80% chance if you're Hispanic, if you're born in that family, developing diabetes." It was very easy to find the children; the problem was we couldn't find lean children. So it took us a while because if you're obese, then you got the lipotoxicity. So we finally found them—this is a JCI paper, I believe—and so we did an insulin clamp. There is resistance as their parents; they have normal glucose tolerance. Why? Because the insulin levels are astronomical. And then we do a muscle—how—how high—just for understanding—oh, they're like two—two—two times uh normal—even higher sometimes. We do a muscle biopsy; the same defect in the insulin signaling pathway. How many of the—how many of the tyrosine kinase defects do they have? They have—starts at IRS1. Insulin binding to the receptor is okay, just like their parents. The ability to activate the insulin signaling pathway—the IRS1—it's already well established, which ends—in particular—it starts at the level of IRS1. Starts at the first one. Oh, so it's not just one enzyme though. Well, it—no—it's IRS1—you can't tyrosine phosphorylate it; you cannot activate PI3K. Okay. So it starts with IRS1. Y. And then the other thing—Jerry and I have both done this in somewhat different ways—so uh, Jerry—I'm talking about Jerry Shulman—he uses NMR by looking at uh phosphate derivatives. Even though I believe the the primary defect is in the signaling pathway, uh, there's clearly severe impairments in glucose transport and phosphorylation. His work would suggest that the primary defect is at the level of glucose transport. We did—we developed a novel triple tracer technique uh using three isotopes infused into the brachial artery. We believe that the primary defect is at the level of hexokinase and phosphorylating uh glucose. So we kind of agree to disagree because we can't do the study, and we'd have to do the MRI study at the same time we're doing the triple tracer technique. In addition to the insulin signaling defect, there's a severe defect in glucose transport and phosphorylation.

Once—let's just make sure people understand this—is this—we're kind of getting into some biochemistry here. Yeah. When glucose enters the cell passively through the GLUT transporter—

Yep.

It gets free glucose in the cell. Yes. Then to metabolize it—

Yeah.

The first step to that is hexokinase—

Yeah.

Which takes a phosphate off ATP and puts it on the sixth position—if I'm not mistaken—it's a specific type of hexokinase.

So it's hexokinase II. Yeah, because there's a different one in the muscle and in the liver.

Correct. And that is correct. Yep. So uh, Jerry would say the primary defect is in GLUT4, the transporter.

Yes.

I would say yes, that is severely impaired, but even more—what does—remind me what Jerry believes is wrong with the GLUT4 transporter. Uh, that it doesn't work normally.

I thought it worked fine; it's just not getting the signal to work because of IRS1.

That's where the controversy is. How—how do—when you have a severe—we already—we were the first to show this defect in muscle—in fact, we're the only people I think that have shown this in human muscle. It's been shown in rats, etc. That to me, you know, metabolism in rats and mice is so different. This is all people data. Okay. So you're saying—but we—we—it's possible that just having the IRS1 problem is enough. It's also possible that even if IRS1 is functioning reasonably, if GLUT4 is not getting up, that's the problem. There is evidence to support that. And then it's also possible that even if all those things work, if you don't get hexokinase to phosphorylate glucose, you back up the whole system. And I can show you that is a primary defect in pyruvate dehydrogenase and and glycogen synthesis. This comes back—I have an ominous octet for the insulin resistance. Okay. This—this—why people don't understand—look, there are eight organ—sort of things that are a problem. There are eight problems. I can show you within the muscle why you think one drug is going to correct all these problems. We need drugs to work on the beta cell; we need insulin sensitizers; we probably need different types of insulin-sensitizing drugs; we need drugs that reverse the lipotoxicity; uh, and uh, you know, will we ever have a single magic bullet that corrects all of these uh, you know, probably not. I think until we discover the genetic basis, and remember I said that diabetes is a heterogeneous disease, and it—I, I, I—did—in Diabetes Metabolism Reviews—I would say 30 years ago, I wrote a review article that said, "I can put a defect in the muscle and reproduce diabetes; I can put a defect in the liver and reproduce diabetes; I can put a defect in the fat cell and reproduce diabetes; I can put a defect in the beta cell and reproduce diabetes," and I went back and read that, and I said, "I can put a defect that starts in the brain and reproduce diabetes." So we see someone with an array of eight or nine—they—all these defects we've been talking about—they're already there. So you put that defect in the fat cell—they can look lean—and there's a syndrome called ALCAM syndrome—there's a specific defect in the—this is white adipocytes—there's this—Phil Sherrill will love me for saying this—he's the like top guru in adipose metabolism up in Dallas—and uh, uh, it's ALCAM syndrome. There's a specific defect in the glucose transporter in white adipose tissue. You know what happens? You become diabetic; you know what happens? You gain weight; you know what happens? You get NASH. So here—here's a defect that I said like 30 years ago—I just postulated and said, "Here's a syndrome," and not only that—now that they defined this in people uh in this paper, they then went to the animal model, and they knocked out the gene that's causing the defect, and they reproduce diabetes in the normal rat model, uh, mouse model.

So Ralph, I want to close by bringing it back to something that people can do to help um understand if they're at risk, either lean or otherwise. We talked about it at the outset but didn't go into it in detail, which was the OGTT. Yeah, the oral glucose tolerance test. Now again, um, none of us have the privilege of being able to use a euglycemic clamp, both clinically as physicians or as experiencing it as patients. So we're going to have to kind of rely on other things. We're going to have to rely on body fat; we're going to have to rely on triglycerides; we're going to have to rely on hemoglobin A1C, although I find that to be a particularly useless uh metric—not that useless—but it—at the individual level, I find it very unhelpful. I think at the population level it's great, and in deltas it's great, but boy, the correlation between hemoglobin A1C and realized glucose levels—it's—it's pretty weak. But let's talk about the OGTT because this is not a test that is done frequently.

Yeah. I believe it should be. Y. And I'd love to have you walk us through the interpretation of the following—I'm going to give you a couple scenarios. So case one—I'm making this up as we go—so you got—you got a person who starts out—all these people are going to start out normal. Okay, they're going to start out with a glucose of 90 and an insulin of six. Yeah. Okay. At 30 minutes—this is after 75 grams of oral glucose—the um, the insulin rises to uh 90—I'm nervous—yep—the the the glucose rises to 130. At 60 minutes, the glucose is uh down to 100; the insulin is down to 60. And we'll just do one more check at two hours. Uh, the glucose at this point is uh 60. Yeah. And the insulin is 20. That is a pre-diabetic state. This is a very insulin-resistant person. Uh, and two hours later, hypoglycemia is a reflection of the beta cell's early insulin secretion. This is kind of a pre-diabetic state. Yeah. And this—I so agree with you completely. And we see this all the time, Ralph. This is a person, by the way, with a perfectly normal hemoglobin A1C. Yes. And this is a person who gets passed all the time as totally normal—the severely insulin resistant. Your glucose is 90; your hemoglobin A1C is normal; and your insulin is six—even if the doctor is checking insulin—but as you point out, the thing that trips you off is not their glucose—90 to 130 to 100—is amazing. It's—90—was how high the insulin was at 30 seconds. And of course they overshot, which is why they become hypoglycemic.

Yes. Okay. Well known. Yep. Okay. Let's go another one. This person also starts at 90 and six. At 30 minutes, they go to uh 180. Yeah. Insulin goes to 30. Yeah. At 60 minutes, they go to 200; insulin is 40. They're diabetic. We showed—but just be clear—this is a person—these are almost real cases, by the way. This person whose hemoglobin A1C is 5.6.

Got it.

We already published this. Yeah. The best predictor of who's going to get diabetes is a one-hour glucose greater than 155. Uh, and this is uh from prospective data from the San Antonio Heart Study, also from the Botnia Study, where these people have been followed up. We were the first people to publish this—oh, I, I'd say seven, eight, nine, 10 years ago. There have been—I'd say at least 15 to 20 studies that have reproduced what we showed uh 10 years ago. Can—one-hour glucose is more than 155—you're in trouble. And that's a great predictor of type 2 diabetes regardless of all the other metrics. Yes. And if you also happen to be hypoinsulinemic, that adds more to the predictive value. But just pick one: 155 without knowing the insulin—that's a huge predictor of whether you're going to develop diabetes or not. That's from the San Antonio uh Heart Study, and that's also from the Botnia Study, uh, and also from our Vegas study. Okay. Next case—I'm not even going to give you the numbers—I'll just describe it. This is a person who has a delayed onset of insulin. So in other words,