The Peter Attia Drive - #402 ‒ NMR blood analysis： how heart disease risk, insulin resistance, inflammation, and mortality risk can be assessed from a single blood sample ｜ Jim Otvos, Ph.D.

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

00:00:16.540 my website, and my weekly newsletter all focus on the goal of translating the science of longevity

00:00:21.520 into something accessible for everyone. Our goal is to provide the best content in health and

00:00:26.720 wellness. And we've established a great team of analysts to make this happen. It is extremely

00:00:31.660 important to me to provide all of this content without relying on paid ads to do this. Our work

00:00:36.960 is made entirely possible by our members. And in return, we offer exclusive member only content

00:00:42.720 and benefits above and beyond what is available for free. If you want to take your knowledge of

00:00:47.940 this space to the next level, it's our goal to ensure members get back much more than the price

00:00:53.200 of a subscription. If you want to learn more about the benefits of our premium membership,

00:00:58.020 head over to peteratiamd.com forward slash subscribe. My guest this week is Jim Otvos.

00:01:06.520 Jim is a biophysical chemist who pioneered the use of nuclear magnetic resonance or NMR

00:01:12.140 spectroscopy to measure lipoprotein particles from plasma. After more than two decades on the

00:01:18.460 faculty at North Carolina State University. He founded Liposcience in 1994, where he developed

00:01:23.940 and commercialized the first FDA-cleared method for direct LDL particle, or LDLP, quantification.

00:01:31.140 Liposcience was acquired by LabCorp in 2014, where Jim served as chief scientific officer

00:01:36.180 of the NMR Diagnostic Group. He's authored more than 200 peer-reviewed publications and holds

00:01:41.940 numerous patents related to NMR-based biomarker activity. I wanted to have Jim on because his

00:01:48.900 work sits at the foundation of a test many listeners have seen in their own blood work,

00:01:54.820 but probably don't realize traces back to him. Many of you have probably had an LDL particle

00:02:00.520 number, and you may even notice that it mentions that it's done by liposcience. But what makes

00:02:05.660 this conversation especially interesting is that the same NMR technology that began with lipoproteins

00:02:11.800 has evolved into a much broader way of looking at metabolic health, inflammation, insulin resistance,

00:02:17.660 and even mortality risk. And that's where we spend a lot of our time today. So in this episode,

00:02:23.060 we go back and talk a little bit about the history. We talk about the unlikely story of

00:02:26.820 how Jim turned a flawed cancer test into a new way of measuring lipoproteins, what standard

00:02:33.700 cholesterol tests can miss, and why LDL particle number can reveal risk that LDL cholesterol alone

00:02:39.440 does not, why this notion of large, fluffy LDLs being benign is misleading, how LDL-P

00:02:46.860 and ApoB help guide treatment decisions beyond LDL cholesterol alone, how NMR can reveal

00:02:52.880 signs of insulin resistance before blood sugar rises, glyc-A as a window into chronic low-grade

00:03:00.460 inflammation as an NMR biomarker, the metabolic vulnerability index, or MVX, and what it may

00:03:07.300 reveal about frailty, resilience, and short-term mortality risk, the surprising finding that MVX

00:03:14.540 in healthy young adults may predict risk decades later, and why NMR diagnostics remain underused

00:03:22.660 despite the amount of information they can extract from a single blood test.

00:03:26.900 So, without further delay, please enjoy my conversation with Jim Otvos.

00:03:30.960 Jim, so great to be with you again. We were just talking a minute ago that I had forgotten briefly

00:03:43.060 that we were together once in 2013, but in many ways, this is a pretty wonderful opportunity for

00:03:49.920 me to sit down with someone whose work I've been following for literally 15 years. It was May,

00:03:55.700 I still remember it was May of 2011 when Tom Dayspring introduced me to your work, and I began voraciously consuming everything you had written in my personal obsession to better understand the fields of lipidology.

00:04:11.680 Again, I think there can't be that many people listening to us that haven't at some point probably had an LDLP or HDLP test done. And yet most of them will never realize until now, presumably, that that test goes all the way back to you and you are the creator of that. So maybe give us a little bit of a story, your journey. You did a PhD in physical chemistry or?

00:04:37.020 In chemistry and biochemistry. And yeah, so I was in academia for 20 years doing the kinds of

00:04:45.560 things that people did in academia and do in academia using NMR spectroscopy as a structural

00:04:51.660 tool. So NMR is a very common structural tool. You can find NMR machines in every chemistry

00:04:59.140 department in the country. So I had appointments in chemistry at the University of Wisconsin-Milwaukee

00:05:04.840 and then moved in 1990 to North Carolina State University. So I was basically doing my thing,

00:05:12.540 minding my own business, using NMR for the usual purposes.

00:05:17.220 And for the listener, we are going to explain how NMR works because for people who didn't take

00:05:22.080 chemistry and might not remember it, we'll come back to it. But I don't want to interrupt

00:05:26.100 now to do that. Yeah, we'll come back.

00:05:28.520 Yeah, I'm not sure that that's terribly relevant, but we can talk about it. But

00:05:32.220 But anyway, the point is that NMR was and is very useful for a particular purpose, which

00:05:40.140 is helping organic chemists determine the structure of molecules that they synthesize,

00:05:44.260 for example.

00:05:44.740 And I was using it to study biomolecules, so it was more challenging than small organic

00:05:51.700 molecules, and trying to get it understanding what was going on at the active site of zinc

00:05:57.440 metalloenzymes.

00:05:59.500 But anyway, so I was funded to do that and had an NMR machine in Milwaukee to do that research.

00:06:08.620 And then in 1986, there was a paper published in the New England Journal with a lot of hoopla,

00:06:14.860 and in particular in the NMR field, people paid attention to this. I didn't normally read the

00:06:20.380 New England Journal of Medicine ever, but it claimed that a very simple NMR test could tell

00:06:27.840 whether somebody had cancer or not, plus or minus, irrespective of whether it was this cancer or that

00:06:34.240 cancer. Nothing really in the paper that laid a mechanistic foundation for why this relationship

00:06:41.020 should be. It simply measured a couple prominent signals in the NMR spectrum of blood plasma

00:06:46.060 and measured how wide the signal was halfway up the signal. If it was narrow, you had cancer.

00:06:52.780 if it was not narrow, you didn't have cancer. And, you know, normally this wouldn't be given

00:06:58.580 much attention, but it was published in the New England Journal. And so everybody who had NMR

00:07:04.540 machines was interested in seeing if they could replicate this. I was in a chemistry department,

00:07:10.840 not associated with a medical school. So I had no idea how to get my hands on plasma if I wanted to

00:07:16.120 play around with this. But I went across the street to a hospital and talked to people in the

00:07:20.220 lab to giving me six leftover plasma samples from healthy people and popped it into my NMR machine.

00:07:28.300 And sure enough, half of the signals were narrow and half were broad. Did half of these people

00:07:33.820 have cancer? No. These three people were women who had just given birth. So pregnancy was one

00:07:40.780 false positive that was given in this New England Journal paper. So if it wasn't for that sort of

00:07:47.160 linkage to something that seemed consistent with what was published, I probably never would have

00:07:53.400 taken another spectrum of plasma. But just out of scientific curiosity, we started measuring plasma

00:08:02.240 and noticed that the signal that was this supposed cancer diagnostic didn't look like a nice

00:08:08.800 symmetrical NMR signal. It had lumps and bumps in the shoulders. And so what was up with that?

00:08:14.800 And pretty quickly when we started to ask, well, where does the signal show up and what

00:08:22.140 are the molecules that are giving rise to these signals, it was clear that these were

00:08:26.740 signals from the lipids in lipoprotein particles.

00:08:30.220 And so we then very serendipitously got funding from Siemens Medical Systems, basically $100,000

00:08:37.640 after me giving a one-hour presentation for what I might learn with $100,000.

00:08:43.000 So that was a pretty cool opportunity for a professor who had to go through a lot more

00:08:48.440 hoops to get funding.

00:08:50.100 So we had the wherewithal to get samples from people with and without cancer and then separate

00:08:55.920 the major lipoproteins, VLDL, LDL, and HDL.

00:09:00.600 And we looked at those signals and noticed that the VLDL signals were always to the left

00:09:07.000 of the LDL signals.

00:09:08.720 The LDL signals were always to the left of the HDL signals, and it was the superposition

00:09:13.640 of these signals and their relative concentrations differing that gave rise to the different

00:09:18.620 shapes of this composite mixture signal that you would see in a plasma sample.

00:09:24.420 So it was obvious that this signal was coming from lipoproteins.

00:09:28.780 So what's up with the narrow signal meaning cancer?

00:09:31.980 Well, that turns out to be due to the fact that those signals from those people were from people

00:09:40.240 with higher triglycerides and lower HDL cholesterol. And the combination of those

00:09:44.540 two things made the signal narrower. So this was nothing to do with cancer per se. It was to do with

00:09:51.280 the lipoproteins, the lipid levels of people with cancer. 20 years before that, people had published

00:09:57.920 that people with cancer on the average have higher triglycerides and lower HDL cholesterol.

00:10:01.980 So anyway, we published in 1990 or 91 a paper in clinical chemistry that showed that the NMR

00:10:12.300 signals from isolated VLDL, LDL, and HDL from people with and without cancer didn't differ at

00:10:18.220 all. So there was nothing distinct about whether the sample came from a cancer patient or not.

00:10:23.860 But what was very reproducible, and we didn't understand why, was this phenomenology of the

00:10:30.140 the LDL signals not showing up in exactly the same place as the LDL signals or the HDL signals.

00:10:36.180 So we got the brilliant idea, which I didn't think was very brilliant. I thought it was obvious to

00:10:40.640 use this putative cancer signal as a source of information about the concentrations of

00:10:46.420 lipoproteins in the blood. So with a fairly simple, low-tech NMR spectrum that anybody

00:10:54.160 could do with any NMR machine, you could generate this signal whose shape and amplitude could be

00:11:01.600 used to deduce the concentrations of the VLDL, LDL, and HDL that were giving rise to that

00:11:08.300 composite signal. And so we had been funded by Siemens. We published a paper in, I think,

00:11:14.960 1991 saying that, yes, it was feasible that one could get VLDL, LDL, and HDL

00:11:22.200 simultaneously from this simple NMR spectrum. So that seemed to have some advantage

00:11:27.360 over the usual way of measuring triglycerides and LDL and HDL cholesterol via normal chemical

00:11:34.080 methods. So that's as far as we thought we could go. And if it wasn't for the fact that

00:11:40.100 Siemens people were advising me that I never would have filed for a patent on

00:11:44.300 how to do this, what did I know about patenting and what did I care about

00:11:49.180 commercialization. But I did file a patent and the patent was issued. So if this seemed to be

00:11:57.760 something useful and clinically useful, then it would have to be clinically translated and that

00:12:03.640 would have to be via some sort of commercial entity because there were no NMR machines in

00:12:09.100 clinical laboratories. In fact, there are no NMR machines to this day in clinical laboratories.

00:12:14.600 So, you know, we needed to figure out how to transition NMR spectroscopy into clinical laboratory

00:12:22.260 medicine, and we needed a commercial vehicle to do that. So that was an idea that evolved

00:12:29.320 over the early 1990s to about 1995, 96. Had a couple NIH grants to support the analytic

00:12:37.380 development. And what we didn't expect to be able to do but found that we could was not only to

00:12:46.280 differentiate VLDL, LDL, and HDL, but the different size subspecies that make up what we call total

00:12:54.180 VLDL, total LDL, total HDL, smaller, medium-sized, larger particles. Because these signal positions

00:13:01.240 are so close to each other, it just didn't seem feasible that you'd be able to work backwards

00:13:06.740 from the composite signal and accurately get the concentrations of the subspecies.

00:13:11.860 But by that time, I'd been reading the literature a bit, and Ron Krauss at UC Berkeley at Donner

00:13:17.180 Laboratory was showing that via a quite laborious separation method, gradient gel electrophoresis

00:13:23.440 and others, that you could differentiate LDL on the basis of size and found that people

00:13:31.000 with a prevalence of small, dense LDL had greater cardiovascular risk at a given LDL cholesterol

00:13:36.860 level than somebody with large LDL. So this was something that was very interesting and had been

00:13:43.660 replicated in literature and people were talking about, and yet it took a couple days from start

00:13:49.940 to finish to do this electrophoresis and get the result. So it wasn't really clinically translatable

00:13:55.600 and inefficient. And just to see if NMR could do this, I hooked up with Ron. He sent me about

00:14:02.280 45 samples along with the gradient gel electrophoresis tracing so I could see who had

00:14:07.620 pattern A, the large LDL, pattern B, the small LDL. And sure enough, when we applied our analysis

00:14:15.660 for decomposing the composite signal into its parts, we could definitely tell the difference

00:14:20.980 between large and small LDL. So aha, okay, now we spent a couple years seeing if we could refine

00:14:26.780 the methodology for quantifying small LDL, large LDL, small HDL, large HDL. And that was quite

00:14:34.320 successful. So it really was with this idea that there was something really clinically useful

00:14:41.460 about being able to differentiate the size of LDL particles that drove us to take the

00:14:47.860 step that I was very unqualified to take, which is commercialization of NMR testing technology.

00:14:57.200 It really did seem that, yes, you not only could generate the same information as a lipid panel by

00:15:02.960 NMR, but really what would drive the utility of NMR testing was if it could measure something

00:15:08.840 better and different. And so if we could measure small, dense LDL pattern A and B,

00:15:13.620 threefold greater risk associated with that at a given level of LDL cholesterol, well,

00:15:18.420 that would be a pretty useful thing clinically. So Jim, tell us how, we don't have to go into

00:15:25.540 great detail, but tell us how a plasma, so you go to the doctor, they draw your blood.

00:15:31.080 The last thing the patient sees is that tube of dark blood that's leaving them. Tell me what has

00:15:35.800 to happen from there until they get a basic lipid panel back, which says total cholesterol is this

00:15:41.100 many milligrams per deciliter, LDL cholesterol, HDL cholesterol, triglycerides, all in milligrams

00:15:45.280 per deciliter. How do they get that out of that tube? What is the basic chemistry?

00:15:49.160 How do they do that chemically?

00:15:50.480 How do they do that, yes, chemically and in a clinical laboratory?

00:15:54.940 Yeah. So these are standard chemistry-based assays that are like all such assays. You

00:16:02.860 add a reagent that reacts with what you're trying to measure, like triglycerides. Actually,

00:16:08.240 triglycerides is interesting because triglycerides are a fatty acid esterified to glycerol.

00:16:15.420 So what actually happens in that assay is the blood is exposed to a lipase that hydrolyzes,

00:16:24.180 that separates the fatty acid from the glycerol. It leaves the glycerol. And then the glycerol is

00:16:29.740 what something else is added to to make a color change in proportion to the amount of glycerol.

00:16:35.900 You're actually counting the glycerol and imputing how much triglyceride you had because

00:16:41.660 you know it was a three to one ratio. Yes. The clinical issue, it's not a common,

00:16:47.020 but there are situations where somebody has a lot of glycerol, not esterified. And so the assays for,

00:16:54.780 standard assays for triglyceride would say that this person has very high triglycerides.

00:16:58.540 It's overestimating.

00:16:59.260 They actually don't have high triglyceride. Anyway, so the same thing with cholesterol.

00:17:03.820 So you're adding chemicals, adding reagents that cause a color change or a change in the

00:17:08.940 UV spectrum that is monitored, and you have a standard curve that relates known amounts

00:17:15.780 of LDL cholesterol to the signal, to the color that's created, and you can work backwards

00:17:22.520 from that measurement.

00:17:24.060 So these can be completely automated.

00:17:25.420 You're using potentially optical density or something like that, trying to light through

00:17:28.920 it.

00:17:29.360 That's right.

00:17:29.860 UV detection or visible light detection.

00:17:32.080 So these are really standard, very, very efficient auto-analyzers do this.

00:17:37.260 The challenge with cholesterol, though, so HDL cholesterol, so the problem is you're measuring

00:17:42.580 the cholesterol inside VLDL, LDL, and HDL.

00:17:46.860 So when does that get broken?

00:17:48.200 When do the lipoproteins break open?

00:17:50.660 In what part of the assay so that you are just looking at the total amount of cholesterol

00:17:54.580 contained?

00:17:55.520 Right.

00:17:55.780 So the original interest in cholesterol and its relationship to cardiovascular disease risk was just total cholesterol.

00:18:03.720 So basically, the reagents find all the cholesterol inside all these particles, don't differentiate where it's coming from.

00:18:10.040 If you want LDL cholesterol, you have to separate the LDL particles.

00:18:14.700 These are spherical containers that contain the cholesterol.

00:18:18.320 And separate it from the HDL containers and the VLDL containers.

00:18:21.540 And then do a regular cholesterol assay on what you've separated.

00:18:25.780 So there's a separation step.

00:18:27.260 Same thing for HDL.

00:18:28.420 And for many years until fairly recently, HDL cholesterol was always measured by first

00:18:34.780 getting rid of the VLDL and LDL by precipitation, and then it left HDL that you then did a

00:18:40.900 cholesterol assay.

00:18:42.980 LDL cholesterol is more tricky because you would have to separate VLDL from LDL, and

00:18:49.900 that takes an ultracentrifuge step, which is laborious, and even clinical laboratories

00:18:55.020 today don't even have ultracentrifuges. So people devised a way of calculating LDL cholesterol by

00:19:03.260 measuring total cholesterol minus HDL cholesterol, which is VLDL plus LDL cholesterol, and then

00:19:09.340 estimating VLDL cholesterol by dividing triglyceride by five. Triglycerides are mostly in VLDL. So it

00:19:15.980 was an easy but not terribly accurate way of quantifying or estimating VLDL cholesterol from

00:19:22.680 which you could subtract the HDL cholesterol from total cholesterol and get LDL cholesterol.

00:19:28.200 Is it safe to say, Jim, that when a patient gets a lipid panel today, unless it says LDL

00:19:33.200 direct, which we'll talk about, and it just says LDL cholesterol equals 127 milligrams per deciliter,

00:19:40.360 is it a safe assumption that it may have been done using that exact same methodology you described?

00:19:45.720 Yes, absolutely. And the only thing that's changed fairly recently is recognition that

00:19:51.500 Dividing triglyceride by five doesn't give a very accurate VLDL cholesterol estimate,

00:19:55.980 and that impacts the accuracy of VLDL cholesterol estimate. And so now there are equations that

00:20:02.780 interrogate other things, non-HDL cholesterol, triglycerides. There's an NIH equation that

00:20:09.640 I help people develop that is used by LabCorp and other major laboratories.

00:20:15.440 And so you can do better than estimating it by a Friedewald formula. But most laboratories,

00:20:23.620 I think, to this day still use the Friedewald divide by five to get the LDL cholesterol.

00:20:30.660 So let's talk a little bit now about NMR. So again, maybe someone remembers back in an organic

00:20:36.440 chemistry class that one of the problems you would receive on an exam or something was you

00:20:42.060 would be shown a picture. And the picture was very much like how you were just describing

00:20:45.560 your experience in the 80s and 90s, where you had along an x-axis a line, and then it would have

00:20:50.840 these spikes, and they would sort of correspond. So tell us, what did the x-axis correspond to,

00:20:56.220 and what did the amplitude or y-axis correspond to? And of course, I want you to get to the point

00:21:01.480 of these are protons, but get there in your own way, of course. Yeah, so you're exactly right.

00:21:06.820 That's what the output of a normal NMR spectrum looks like, irrespective of whether you're

00:21:12.860 detecting protons, hydrogen, nuclei, or carbon-13, or nitrogen-15, et cetera.

00:21:20.360 So on the x-axis is frequency.

00:21:22.560 It's just frequency.

00:21:23.540 These are signals that have different frequency.

00:21:25.900 And their amplitude is proportional to how much of the molecules carrying the hydrogens.

00:21:33.360 Let's just talk about proton, hydrogen, NMR.

00:21:35.840 They show up in different places.

00:21:37.180 They have different frequencies depending on their chemical environment, and that's why

00:21:41.560 this is a useful structural tool for organic chemists.

00:21:44.040 So a CH2 group next to another CH2 group as opposed to a CH2 group in a double bond

00:21:51.760 or whatever it might be, they show up in very different places, and there's other differences

00:21:57.420 that are structure-dependent.

00:21:59.120 So you can work back from an NMR spectrum and deduce the structure of the compounds that are

00:22:05.520 giving rise to that. So this was never used, though, as a quantitative analysis tool. It was

00:22:16.640 used sort of as relative signal intensities would tell you how many protons are here in a molecule,

00:22:23.520 how many protons are there on the molecule, does that fit the structure that you were attempting

00:22:27.520 to synthesize, for example. And so I just want to make sure people understand what you're

00:22:31.340 getting at there. So you're trying to synthesize something. You know the structure of the thing

00:22:36.120 you're trying to synthesize. You think you know it. You think you know it. And therefore you

00:22:39.680 should know what its spectra looks like. And now you're basically trying to match the NMR of what

00:22:47.160 you've synthesized to say, look, this should have a carbon, a double bond, a carbon, a single bond

00:22:53.280 over here is going to be an oxygen that's going to produce. And it's a really fun game. Not to

00:23:00.360 be too nerdy about it, but it's a super fun puzzle to solve effectively. Yeah. And it's pretty

00:23:06.820 complicated and I haven't done that sort of thing in 50 years. So I moved to a different application,

00:23:12.920 which is using NMR to detect a single signal. So I didn't care about what was in the rest of the

00:23:21.360 spectrum. I cared about the signals that were these putative cancer signals from the terminal

00:23:28.560 methyl groups of the fatty acid chains that are carried on different types of lipids that are

00:23:32.940 carried in these lipoprotein particles. By the way, did the NMGA, New England Journal of Medicine

00:23:39.040 authors know that they were looking at lipoproteins? I think they said lipids. I can't

00:23:44.280 remember. And again, yes, they did. And they suggested that if you had cancer, there was

00:23:52.880 some structural alteration in the lipoproteins that gave rise to a different signal. And that's

00:23:58.020 what we disproved by showing that that wasn't true. So anyway, back to this issue of whether

00:24:06.760 we could use that so-called cancer signal as a source of quantitative information about

00:24:12.880 the lipids and lipoproteins or the lipoproteins themselves. So what I told you is true. It was

00:24:20.280 empirical observation that there was this very consistent relationship between where the frequency

00:24:27.340 of the signal from the very same methyl groups on the very same molecules so that the molecules

00:24:32.860 don't differ at all in terms of what's carried in VLDL, LDL, and HDL. It's the same lipids,

00:24:38.100 so you'd expect they would all show up in exactly the same place. But for some magical physical

00:24:43.160 chemical reason that is explained by complex equations that I don't even understand very well,

00:24:49.080 it's been shown that a larger particle will always give rise to a signal that has a slightly lower

00:24:56.260 frequency, and a smaller signal will have a slightly higher frequency. So the same lipids

00:25:02.260 in different size packages show up in slightly different places, but very reproducibly

00:25:07.120 different places. And so the idea is that if you understand exactly where the signal shows up from

00:25:14.500 a particular diameter lipoprotein particle, and also measure that because the shape of the signal

00:25:21.320 will differ, that's another complexity, but adds to the accuracy of what we're able to come up with.

00:25:28.100 With a complete understanding of what the different parts are that make up the mixture,

00:25:32.820 make up the whole. The whole idea is that you measure the whole and then you decompose it into

00:25:38.840 the parts. So the sum of the parts equals the whole. What's efficient about the methodology

00:25:44.640 is that you're measuring something with a really low-tech simple NMR spectrum that you can obtain

00:25:50.040 in 30 seconds. A computer then with a deconvolution model that has in it what the signals look like

00:25:59.280 from all the different size VLDL, LDL, and HDL subclasses, and then take that measured,

00:26:05.100 30-second measured composite signal and spit out how big the signals must be from all the

00:26:11.360 different constituent parts to, when they superimpose, they will recreate that shape

00:26:17.040 of the composite signal. So that's the idea that the concentration information comes from how big

00:26:22.320 the deduced NMR signal intensities are in this mixture, blood. Now, the problem is that-

00:26:31.720 Sorry, one point to just add to that, Jim, that a person who's looking at their own NMR result

00:26:37.120 in their blood test will notice that the units are reported as nanomoles per liter as opposed

00:26:43.180 to milligrams per deciliter. So it's not a mass concentration, it's a molar concentration. Maybe

00:26:47.440 explain to people what that distinction is. Yeah, that's just where I was going to go next,

00:26:49.880 because as I told you, the chemical constituents in these particles, and they're quite heterogeneous,

00:27:00.160 there's different size legs of fatty acid chains. Some are saturated, some are monounsaturated,

00:27:05.240 some are polyunsaturated. All of these things give rise to the lipid complexity of these particles.

00:27:11.660 But the NMR detection, at least of this signal, knows nothing about any of that.

00:27:17.700 It's basically a lipoprotein particle signal.

00:27:22.340 And so what should be the case, if that's true, is that how big that NMR signal is from

00:27:28.520 the particle should relate to the number of particles, irrespective of what the lipid

00:27:35.180 concentrations are.

00:27:36.800 And there are variable amounts of cholesterol and triglyceride in most lipoproteins.

00:27:40.660 So what we realized at the beginning was if this was a lipoprotein particle signal we

00:27:48.100 were interrogating, we could get lipoprotein particle concentration information, but we

00:27:54.060 could not and should not report lipoprotein cholesterol levels or lipoprotein triglyceride

00:28:00.000 levels because we actually weren't able to differentiate the signals from those different

00:28:04.780 chemical species. So that is what we started to produce when we did this commercialization thing,

00:28:14.400 which I may come back to a little bit. But as you said, what's reported LDL-P is the particle

00:28:21.580 concentration in nanomoles per liter. So there's 6.02 times 10 to the 23rd particles in a mole

00:28:29.520 of LDL. And so this is 10 to the minus ninth. Anyway, so it's a big number. It's still 10 to

00:28:36.520 the 16th particles. So there's a lot of these particles in your blood, but you're reporting

00:28:42.820 the concentration of the package, not the lipid molecules in the package. So then the question

00:28:49.220 is, or one question is, is there any advantage to that? I mean, because by then, so I'll go back a

00:28:57.260 little bit to the commercialization, because as I said, the commercialization was driven by the idea

00:29:02.280 that small, dense LDL is much more atherogenic, much more to worry about than large LDL.

00:29:09.340 In our own studies, when we started measuring small and large LDL by NMR and did it in large

00:29:16.780 population studies, we found exactly the same thing that Ron Krauss and others did.

00:29:20.380 And which population did you look at? First, Mesa or Framingham?

00:29:24.380 Framingham, for sure, back in the day.

00:29:26.340 Mesa came a little bit later.

00:29:28.500 But I can't remember what we initially looked at.

00:29:31.620 But the point is that when looked at through the lens of at a given quantity of LDL, if

00:29:39.800 the LDL is small versus large, does it make a difference in your cardiovascular risk?

00:29:44.500 When I said the concentration of the quantity of LDL, the way everybody thinks of the concentration

00:29:50.000 of LDL is LDL cholesterol.

00:29:52.100 and that's what Ron Krauss and that's what we did. We said, okay, at a given level of LDL

00:29:57.740 cholesterol, you take people and you stratify them according to low, medium, and high LDL

00:30:01.760 cholesterol, or you do a multilinear regression and put LDL cholesterol in the model. And now

00:30:07.840 you're asking, does the size of the LDL add anything to LDL cholesterol in cardiovascular

00:30:13.860 risk production? And sure enough, it does. And it's quite an impressive increment of risk.

00:30:19.360 So we reproduced what Ron Crest did. However, we realized because we were also in the business of

00:30:26.660 measuring LDL particles that by definition, if an LDL particle is smaller versus larger,

00:30:33.160 it's always full of lipid. You don't get a partially filled container of LDL. It's always

00:30:38.740 full. So a smaller LDL particle means by definition that it's carrying less cholesterol,

00:30:44.100 less lipid per particle than a large LDL particle. So the phenomenology is at a given level of LDL

00:30:51.100 cholesterol, people with small dense LDL have higher cardiovascular risk. The trouble is that

00:30:57.180 people with small dense LDL at a given level of LDL cholesterol have more LDL particles. Their LDL

00:31:03.620 P is higher than would have been imagined from the LDL cholesterol measurement. So an alternate

00:31:10.160 explanation for the extra risk that small dense LDLs seem to confer is that there are simply more

00:31:17.800 particles rather than the size of the particles being the determining characteristic of the

00:31:24.040 atherogenicity of these particles. Yeah. Let me just use a silly childlike

00:31:27.640 example to make this point. So imagine you had four lipoproteins that each contain three units

00:31:35.380 of cholesterol. They're fully saturated at three units of cholesterol. So you have four of them.

00:31:39.600 so you have 12 units of cholesterol. Now imagine you have three spherical lipoproteins that each

00:31:46.400 have four units of cholesterol. They're fully saturated, so they're obviously bigger. But they

00:31:51.280 also collectively have 12 units of cholesterol. So here you have two people. One has 12 units of

00:31:57.460 cholesterol, but it's being carried with four lipoproteins. The other has the same 12 units

00:32:02.120 of cholesterol carried by three. You're telling me all the data say the first person is at higher

00:32:07.740 risk. Question is, are they at higher risk because their spheres are smaller or are they at higher

00:32:14.240 risk because they have more spheres, which happen to be smaller? Correct. So you explained that very

00:32:19.620 well. And so this is basically how science works. So you have a different explanation for the

00:32:27.580 phenomenology of small dense LDL having this seemingly extra atherogenicity. So it's then

00:32:34.240 untestable. So it's quite straightforward at that point to ask the question at a given level

00:32:40.640 of LDL particles, if you stratify people according to LDL-P, and then say some people have small

00:32:48.720 LDL, some people have large, is there any difference in their risk? And the answer is not,

00:32:52.500 not a bit. And we did this in many states. Now the example is you have two patients that each

00:32:57.820 have 20 particles. One of them is 10 big, 10 small. The other one is 15 big, 5 small.

00:33:07.420 If the particle size matters, the first one should be at higher risk. If the particle size

00:33:12.500 is irrelevant, once you've corrected for total number, they should be at the same risk.

00:33:16.620 You're saying they're at the same risk. That's right. And why does this matter? I mean,

00:33:20.280 if you're only interested in assessing the risk of a person, you're equally well off

00:33:26.520 with the cholesterol information and the size information as the particle information and the

00:33:31.860 size of the particle information, the size doesn't add to that. The reason that's important is that

00:33:38.280 if you believe that small LDL is bad and you can make it less bad by making the particles bigger

00:33:44.940 therapeutically somehow, then you will be telling patients that at a given, let's say they get

00:33:54.020 treated with statins and they lower their LDL cholesterol or their LDL particle concentration

00:33:58.480 to an acceptably low level, but the particles are still small, somebody who believes small

00:34:04.980 dense LDL is particularly bad will then try to do something to make them bigger and imagine that

00:34:10.640 there's clinical benefit. And a lot of drugs actually have that effect. CETP inhibitors are

00:34:18.140 one of them, niacin, HDL drugs of different types. So triglyceride lowering automatically

00:34:25.020 will have this effect. So there really are a lot of clinical trials that imagined when they

00:34:32.660 set out to do the clinical trial that there would be a lot of efficacy, like niacin, for example,

00:34:38.520 because not only did that modestly raise HDL cholesterol and lower triglyceride,

00:34:44.400 which were two good things, seemingly. But it made the LDL particles bigger and it made the

00:34:52.240 HDL particles bigger because there's also a similar argument about the size being important

00:34:56.740 in HDL. And yet there was no efficacy when they did the outcome studies. And this has been

00:35:05.000 reproduced with the fibrates and other drugs. So Jim, what would you say to the person listening

00:35:12.680 who, because I hear this all the time, who says, hey guys, I do have a high LDL cholesterol and

00:35:20.120 even my LDL particle number is very high on my NMR test, but I'm very pattern A. You see,

00:35:26.640 all of my LDL are very large and they even use words like fluffy and buoyant. I have large,

00:35:33.660 fluffy, buoyant. Yes, LDL. So my LDL particle number is over 2000 nanomole per liter,

00:35:41.020 which probably places me at the 80th percentile or so, but I don't need to worry about it because

00:35:46.840 they're all big. I don't have that pattern B. I don't have that small. What would you say to

00:35:50.220 that person? I would say that's a fallacious idea and the data that I mentioned supports it

00:35:58.120 completely. And then you also have to think about the fact that one of the best known or best

00:36:05.040 accepted genetic reasons for cardiovascular risk is FH, familiar hypercholesterolemia.

00:36:13.800 People with FH have very high LDL cholesterol levels. They also have very high LDL particle

00:36:20.460 levels, but those particles are large. They're not small. And these are people that die when

00:36:26.720 they're 30 or 35 years old when they're homozygous FH. So this idea that somehow fluffy,

00:36:32.500 large LDL particles are not to worry about or benign is completely fallacious.

00:36:38.480 Let's talk about one more thing on this before we pivot away from this, which you alluded to

00:36:43.120 very briefly, but we went off on a different path, which is the potential discordance that

00:36:49.020 exists between LDL cholesterol and LDL particle number. In fact, when I first came to learn about

00:36:55.780 your work, Jim, this was actually one of the first papers I read of yours literally 15 years ago,

00:37:01.220 almost to the month. And it was in the MESA population, and it was showing four Kaplan-Meier

00:37:09.340 curves. So I'll let you explain what a Kaplan-Meier curve is, but it was basically the four scenarios,

00:37:15.280 which was discordance between LDLP and LDLC when one is higher than the other each time,

00:37:24.400 and then concordance between them. So maybe there were three curves then.

00:37:27.840 Exactly. There were three curves.

00:37:29.360 Okay. So concordance between LDL-C and LDL-P, and discordance in favor of LDL-P,

00:37:34.280 discordance in favor of LDL-P. Okay. So that was a very, very eye-opening paper to me.

00:37:40.120 And I'd love you to just kind of explain that finding and what the significance are,

00:37:44.200 because to this day, it's still a very important thing for clinicians and patients to understand.

00:37:49.720 No, it is. And prior to that, MESA paper in 2011 was a 2007 paper. So it was the first one that we

00:37:56.500 sort of introduced this idea of assessing the situation by whether the LDLP and the LDLC agreed

00:38:05.820 or not. And in that case, it was a Framingham study. And in that case, one question that always

00:38:12.280 comes up is, well, what defines discordance? And basically, it doesn't matter, and everybody

00:38:19.660 defines it differently. And it doesn't matter because we're not saying discordance is a risk

00:38:24.640 factor, right? We're simply trying to disentangle a situation where most people don't have a

00:38:33.140 discrepancy between LDL-C and LDL-P. So if you're trying to understand whether LDL-P is a better

00:38:41.040 measure of LDL-associated cardiovascular risk, and you do a whole study population,

00:38:46.980 and 80% of those people have concordant or in agreement LDL-C and LDL-P.

00:38:51.980 And just explain what that means, because people will say, well, the numbers are different. How

00:38:55.360 can they be in agreement? Okay. So one way to do it, which we did in the MESA paper,

00:39:00.400 is transform the milligram per deciliter cholesterol number into a percentile.

00:39:06.200 So that defines people according to their rank, low, high, intermediate. And if you do the same

00:39:12.720 thing to the LDLP, now you're comparing apples to apples, percentile of one, percentile of the other.

00:39:18.300 and you plot that, and then you see the data points all over the place. And the ones that

00:39:24.680 track on the diagonal, the ones that are close to the diagonal, are the ones for which a 50

00:39:29.680 percentile LDL-P is close to a 50 percentile LDL-C. So that's what we call concordance.

00:39:35.720 And in that case, we picked this weird number of a 12 percentile difference was our cut point,

00:39:42.020 plus or minus. So everybody within 12 percentile units of each other, we called concordant.

00:39:47.480 Why 12? Why not a round number? And it was because we were trying to make the concordant

00:39:53.160 population about equal in number to the discordant in one direction and the discordant in other

00:39:58.760 direction population. So you had 50% in one, 25, and 50% in the two discordant groups.

00:40:04.360 No, it's actually, we're trying to make it sort of 30, 30, you know.

00:40:07.240 Ah, I see. 30, 30, 30. Got it. Okay.

00:40:09.160 And so that's what we did. And then we simply took these three groups of people. And you talk

00:40:17.400 about Kaplan-Meier. All this is basically cumulative incidence of cardiovascular events.

00:40:21.880 So on the x-axis, you can see the number increasing. If it increases a lot, this is a

00:40:27.780 higher risk subpopulation than somebody whose risk is much shallower and doesn't increase very much

00:40:34.820 as a function of concentration. So that very clearly showed that for people whose LDL-P and

00:40:42.800 LDL-C agree, you can make no argument about why LDL-P is a better thing to measure than LDL-C.

00:40:49.040 Then it's a matter of, well, from first principles, would we expect the cholesterol

00:40:53.700 to be a better measure of risk or the particle number? And really, if you're honest about it,

00:40:59.580 there is no mechanistic explanation that you should really be too comfortable with for one

00:41:06.040 versus the other. And we explained that in that paper, that what has happened when Ron Krauss

00:41:11.440 showed that small dense LDL was more atherogenic and that this was based on epidemiologic

00:41:17.640 population studies, the question was, well, why is that? And then basically the speculation was

00:41:24.020 ultimately supported by various lines of evidence that, yes, smaller LDL particles are likely

00:41:29.560 to get into the arterial wall easier than a bigger particle. And then the difference in the shape of

00:41:35.380 the stuff on the surface of the particles can bind more avidly to molecules that are in the arterial

00:41:41.540 wall and retain it more. And then it's more readily oxidized. So all of these things,

00:41:47.100 I can't tell you how many papers I've read were in the discussion section is this obligatory

00:41:51.020 paragraph about why small dense LDL is so much more atherogenic. The flip side argument is that,

00:41:57.600 okay, large LDL particles also get into the artery wall. And when they get oxidized-

00:42:03.200 They're more retained.

00:42:05.140 No, no, they're not more retained.

00:42:06.700 I'm just saying you could make that argument, right?

00:42:08.640 Well, let's just say that they get to the end of the trail and they get taken up by macrophages

00:42:14.540 and deposit their cholesterol contents into the arterial wall. Well, bigger particles have more

00:42:20.920 cholesterol to deliver. More oxidative damage.

00:42:22.740 So you'd think the more cholesterol-rich particles should be more atherogenic.

00:42:25.760 But if that's counterbalanced by not getting to the end of the party as often as the small

00:42:30.780 particles, maybe it's a wash. How do you answer the question? You do the study. And so that's

00:42:37.280 what the discordance analyses allowed us to do. Whenever there's a discrepancy in one direction

00:42:41.740 or another, the cardiovascular risk tracks with the particle number, not the cholesterol.

00:42:47.620 Yeah. So the curve really had a beautiful separation of those three figures. You had

00:42:51.720 the concordance in the middle. And then above that line, which means these are people that

00:42:56.100 are dying quicker, that's when LDLP was above LDLC. And below the line, these people died much

00:43:04.700 slower than you would expect. It was the flip. Cholesterol was high, but their particles were

00:43:08.400 low. Exactly. Yeah. So yeah, that, and so, I mean, and I can't tell you how many debates

00:43:16.040 I had to have with people, I mean, usually very well-regarded cardiologists,

00:43:22.540 tell me about it, who just refused to acknowledge there was ever a need to look at anything beyond

00:43:28.980 LDL cholesterol. Right. Well, now let me sort of go somewhere about the clinical utility of this

00:43:37.720 because what people would imagine from having said all this is that LDLP should be a much

00:43:45.680 better thing to use to assess cardiovascular risk. And the reality is that the way cardiovascular

00:43:54.960 risk is assessed today is pretty much the same way it's always been assessed by equations that

00:44:03.000 take into account total cholesterol, HDL cholesterol, diabetes, present absent, smoking, present

00:44:10.360 absent, hypertension. So these are risk equations, and there's updated risk equations,

00:44:16.460 just fairly recently a new one. They all have total and HDL cholesterol in it.

00:44:21.720 When you ask the question, if I use LDL-P or I add LDL-P to a model that has those things in it,

00:44:30.000 is my risk assessment better? And the answer is no, it's not better. And this sort of kills you

00:44:35.740 if you're trying to make a commercial argument that you should be testing LDL-P instead of LDL-C

00:44:42.660 for risk assessment. The reason I think that that is true is because the HDL cholesterol is in the

00:44:49.520 model. And we still don't really understand whether HDL is bringing extra risk assessment

00:44:56.740 to the table above and beyond what the LDL and the total cholesterol brings. Or is it the

00:45:02.600 triglyceride-rich particles. Now people have swung away from the idea that HDL cholesterol

00:45:07.400 is important because the HDL cholesterol-raising trials were negative or weren't positive.

00:45:14.260 And then because there's an inverse correlation between triglyceride and HDL cholesterol,

00:45:18.840 when HDL cholesterol is low, triglycerides are high. Oh, maybe it's the triglycerides.

00:45:22.880 Well, it doesn't make sense that the triglycerides per se, the molecules triglyceride,

00:45:26.860 are atherogenic, but the triglyceride-rich particles that carry them could well be because

00:45:32.340 they also get into the arterial wall and deliver a lot of cholesterol.

00:45:36.140 Now, Jim, I don't know if this was one of your papers, but I think it was. And it was around

00:45:41.240 that time, probably 2012, 2013. And if I recall the paper, the figure specifically, it was a

00:45:47.820 histogram, right? So on the x-axis, you had zero, one, two, three, four, five, and you were looking

00:45:54.060 at number of criteria met for metabolic syndrome. So for the listener, just to remind everybody,

00:46:00.200 metabolic syndrome has these five criteria. And the more of them you have, the more likely you

00:46:06.100 are to be insulin resistant. So there's one about having high blood pressure, obesity, so the

00:46:11.520 truncal diameter, truncal girth, blood pressure, fasting glucose, and HDL cholesterol. I think

00:46:18.160 those are the five, right? It doesn't include, yeah. Okay. And what this figure showed was if

00:46:24.580 you had zero of them, the probability that you were discordant between your LDL-C and LDL-P was

00:46:32.320 very low. If you had one of them, the probability went up a little bit. Two of them, considerably.

00:46:38.360 Three, a lot. And it was a monotonic increasing relationship, Jim, such that by the time you had

00:46:43.580 five out of five MET-SYN criteria, you were virtually guaranteed to be discordant. And so

00:46:50.440 I bring that up to say earlier you mentioned there's nothing wrong with being discordant per

00:46:55.220 se, although you could also argue that the more discordant you are, the higher the probability

00:47:03.920 that you're discordant means the higher the probability that you probably have some underlying

00:47:07.120 metabolic insulin resistance. Extra risk. Yeah. Yeah. And so the question then becomes,

00:47:12.200 is the reason that very elaborate multi-parametric risk models can erase the LDLP is that they are

00:47:19.280 simply capturing all of the cardiometabolic risk indirectly and directly? Yeah, you basically just

00:47:26.620 explained what I was getting at, which is because HDL cholesterol is in the standard risk assessment

00:47:33.840 models, when you have low HDL cholesterol, you're likely to have more LDL-P than the LDL-cholesterol

00:47:40.920 indicates or suggests. And so your risk is higher because you have higher LDL-P, not because HDL

00:47:48.280 cholesterol is low. But you're not doing better in risk assessment because the HDL cholesterol,

00:47:54.980 for the wrong reason, if you will, not a causal reason, but an association reason,

00:47:59.420 is telling you the same thing. So this was really important to liposcience back in the day because,

00:48:07.700 and this is something that I still argue with, but Alan Snyderman and I will get to the question

00:48:13.720 of ApoB. But he insists, he keeps wanting to convince people that ApoB is important for risk

00:48:23.240 assessment. And what we pivoted to in light of the evidence that we found ourselves convinced

00:48:33.560 people that LDLP was a better thing to measure for risk assessment to a better thing to measure

00:48:38.600 for risk management because what you do when you have high risk is you manage it by lowering

00:48:45.080 your LDL cholesterol. This is the best tool we have in the toolbox. There are not actually very

00:48:51.080 many others. We have statins and then we have things that are doing the same job as statins,

00:48:55.880 but better. So we can lower the heck out of LDL. And the question for a patient with high

00:49:01.620 cardiovascular risk is how much LDL lowering do I need? So what you really would like is

00:49:08.520 a biomarker that tells you my LDL-associated risk is adequately controlled. I've gotten my LDL low

00:49:18.540 enough. If it's not, I should add, if I'm on high-dose statin and I'm not there, then I want

00:49:24.880 to add a PCSK9 inhibitor or something else to lower it even more. So you want the best objective

00:49:30.340 measure of LDL-related risk, not total risk, but LDL-related risk. In that case, it's very clear

00:49:39.020 that LDL-P or ApoB is a better biomarker because a lot of people who achieve very low LDL cholesterol

00:49:48.400 have not achieved equally very low LDL particles or ApoB, and they therefore, with visibility to

00:49:56.240 that would be candidates for more aggressive LDL lowering. And so we absolutely noticed that as

00:50:02.780 clear as day in our practice, Jim, because we are very aggressive at managing these things.

00:50:07.120 Is there a biologic reason for why the discordance really happens at low levels in that way? Or is

00:50:13.760 it a chemical assay? Is it a property of the assays that's causing that? No, it's like everything

00:50:20.280 that you ask. It's complicated. You can get into the weeds. But what's true is that the lower your

00:50:27.760 LDL level is, the more likely that the cholesterol in the LDL particle is replaced to some extent by

00:50:35.240 triglyceride. Because there's always this interchange, this swapping of cholesterol

00:50:39.720 ester and triglyceride in the core of the particle. And it's driven by the relative amounts of the

00:50:45.200 triglyceride-rich particles and the LDL particles, which are cholesterol-rich.

00:50:49.680 The triglyceride-rich particles, the bigger that gap is, put triglyceride into LDL in

00:50:54.700 exchange for cholesterol ester. And so as you lower LDL with statins or whatever,

00:51:01.240 the LDL particles are lower, but the LDL cholesterol is even lower because not only

00:51:06.980 has the particle number gone down, but the cholesterol in the particles independently

00:51:11.340 is going down. That's actually a great explanation. I did not know that. And that

00:51:15.940 makes sense because we see that as clear as day. Okay. One other thing I want to talk about on this

00:51:21.180 front before we pivot is at least to my knowledge, the first composite score that you then developed

00:51:27.000 out of that, which was the LPIR score. Was that indeed sort of your first foray into

00:51:31.180 pool composite scores? Yes. And it was still used to this day. So many people again listening to us

00:51:37.680 will have an LPIR score every time they get their blood test.

00:51:40.760 No, it's interesting because, as I said, we first got into this game because we could measure small

00:51:46.700 LDL. And so the ability to differentiate different size lipoprotein particles seemed like it was

00:51:53.860 clinically useful. But at the end of the day, as I've gone through, it turns out that it's really

00:51:59.140 the particle number that matters. And if you can convince people to not pay so much attention to

00:52:04.240 LDL cholesterol and pay more attention to ApoB or LDL-P, then you're better off. But that means

00:52:10.820 that measuring the size of LDL doesn't matter. Oh, that's a bummer because we have a great

00:52:16.100 efficient way of doing that. So then the question was, well, are the lipoprotein subclass distributions

00:52:21.800 useful for something else besides cardiovascular risk assessment and management?

00:52:26.160 And the answer is absolutely yes. There is a well-known association between higher triglycerides

00:52:33.480 in lower HDL cholesterol and insulin resistance and diabetes and insulin resistance leading

00:52:40.460 to diabetes risk or leading to diabetes.

00:52:44.220 So it's already known that there's a lipid signature for insulin resistance.

00:52:50.880 And the idea was if we could measure the different sizes of VLDL, LDL, and HDL, could that do

00:52:58.460 a better job than just the triglyceride over HDL cholesterol ratio?

00:53:02.040 So the poor man's insulin resistance measure from a lipid panel is triglyceride over HDL cholesterol

00:53:07.940 ratio. And Jerry Riven, who was really discovered and promoted the idea of insulin resistance being

00:53:14.440 extremely important, he advocated that it be used because people were getting lipid panels and

00:53:20.000 this information was not being used for anything. So with this LPIR score, which brews together six

00:53:27.140 the LDL and HDL size and subclass concentration parameters. It brews them into a score from zero

00:53:36.340 to 100, higher scores being more insulin resistant. Then the question was, does this LPIR score do a

00:53:44.300 better job in assessing whether somebody is likely to become diabetic? So the pathophysiology has to

00:53:52.840 be talked about a little bit here, because there are some interesting parallels to where we are

00:53:58.960 today with respect to primary prevention of cardiovascular disease versus primary prevention

00:54:04.560 of diabetes. So in the cardiovascular situation, as everybody pretty well understands,

00:54:11.560 the initiating causal factor there is elevated cholesterol or elevated LDL or elevated WB,

00:54:16.920 but it's acting over time. So it requires an integration of exposure over a long period of

00:54:22.880 time. And that gradually leads to cholesterol deposition in the artery wall. That's atherosclerosis.

00:54:30.180 Atherosclerosis over time starts little, more, more. But that doesn't trigger any clinical

00:54:38.520 concern. This is a subclinical manifestation of cholesterol doing its dirty deed over a long

00:54:44.700 period of time. And then at some point, you might have a myocardial infarction or a stroke.

00:54:50.220 And that's when the atherosclerosis has transformed into the clinical event.

00:54:56.060 The good news from a measurement biomarker standpoint is that the initiator, the causal

00:55:01.560 factor is cholesterol, which is easily measured. So now what about diabetes? Diabetes is very

00:55:07.640 similar. It's a time-integrated process where if you are insulin resistant, over time, your

00:55:14.380 beta cells have to spit out more insulin to keep your glucose under control. So you're making the

00:55:21.800 beta cells work harder, if you want to think about it that way, if you're insulin resistant versus

00:55:25.760 insulin sensitive. And so insulin resistance times time leads gradually to hyperglycemia,

00:55:33.940 elevated glucose. So if it's 90 or below, you're A-OK. But over time, if you're going to convert

00:55:41.800 to diabetes, you go through a transition of the glucose going higher and higher until it crosses

00:55:47.180 this magic 126 milligram per deciliter line that defines diabetes. Just for folks to know,

00:55:53.540 we're talking average. Absolutely. So the causal factor is insulin resistance.

00:56:00.940 The thing that's sort of equivalent to atherosclerosis on the CBD side is hyperglycemia,

00:56:08.220 less than 126. So not diabetes, but pre-diabetes. So when glucose gets over 100 before it goes from

00:56:15.840 100 to 126, you're pre-diabetic. Guess what's easy to measure? Glucose. So in that case,

00:56:22.120 the effect of the cause is measurable. The cause itself is not. And so what that means is that

00:56:30.600 from a prevention standpoint, what you really want to do is to keep insulin resistance from

00:56:35.940 transforming over time into hyperglycemia and ultimately diabetes. If you don't know that

00:56:42.680 you're insulin resistant, you're waiting for the easily measured glucose to become elevated.

00:56:48.880 And now, once that happens, you've lost about 50% of your beta cell function.

00:56:56.620 So the opportunity for real effective prevention, primary prevention, primordial prevention,

00:57:02.200 is to act on people whose glucose is okay and hasn't gone to this transition yet because the

00:57:09.720 beta cells have started dysfunctioning. So Jim, I want to pause you there because the way you've

00:57:14.480 laid that out is very elegant. And I like the, I've never thought of it the way you just explained

00:57:20.060 it. So I'm repeating it just as much for me as for others. People who listen to this podcast know we

00:57:25.460 constantly use the way you describe the CBD prevention thing, which is you have a causal

00:57:32.700 marker. You don't need to wait until disease is measurable to treat it. And the example I always

00:57:39.520 give is smoking and lung cancer. We have a causal marker, for lack of a better word, smoking.

00:57:45.640 We always have to specify causality doesn't mean one-to-one mapping. There are some smokers that

00:57:51.580 never get lung cancer. There are some never smokers who still get lung cancer. None of those facts

00:57:56.540 erase the causality of tobacco and lung cancer. Do we need to wait for a smoker to develop a small

00:58:04.440 cancer to tell them to stop smoking? Absolutely not. That would be malpractice. You always

00:58:11.240 eradicate causal drivers of disease the moment they appear. And that's why when you have elevated

00:58:17.860 LDL-C or ApoB or LDL-P, you treat it immediately, not once they have disease. Same with hypertension,

00:58:24.800 same with smoking as it pertains to cardiovascular disease. Now let's pivot to what you said,

00:58:29.800 which is, look, we know that insulin resistance is the cigarette to diabetes as cancer. Do we

00:58:39.060 want to wait until we actually see the glucose rise, which by the way is the biomarker that

00:58:45.500 defines diabetes, when in reality, by the time that's happening, there's potentially already

00:58:51.280 cellular damage at the level of the beta cell in the pancreas, and it's basically running out of

00:58:55.600 steam. And what else can we measure? Now, I want to come back to the idea of there are things that

00:59:03.060 we can do, but they're very laborious. So an oral glucose tolerance test is a fantastic way

00:59:07.860 to find out that canary in the coal mine years before it shows up, but it's so fallen out of

00:59:13.540 favor as a clinical test because it takes two hours. I mean, it's just so cumbersome to do

00:59:19.760 that outside of our practice and a few others, I just don't imagine many people want to do it.

00:59:24.780 So Jim comes along and says, what if somewhere in this NMR spectrum is a whole series of things

00:59:33.500 that turn out to be a fantastic marker for insulin resistance that we can use as causal

00:59:40.560 proof that you're on the wrong path before your glucose goes up. Yes. And it seemed to us a very

00:59:48.260 compelling case that you would want to act on the causal factor and not the effect or the downstream

00:59:58.220 effect of the action of insulin resistance. But it's interesting. I mean, back to convincing

01:00:06.060 people clinical translation. I mean, I use that word. I don't think I was familiar with that word

01:00:12.180 when I started this company called Liposcience to try to get NMR testing introduced into

01:00:17.640 clinical laboratories. But I thought the argument of LDLP versus LDL-C was quite compelling and

01:00:24.460 that people really should be using it as the biomarker to determine management of LDL.

01:00:29.640 tremendous resistance by the establishment. I could never understand why they were so

01:00:36.780 resistant. The messaging people learn about cholesterol in medical school, oh, we'd have

01:00:41.280 to tell a different story and people wouldn't get it and all sorts of reasons that seem pretty weak

01:00:47.780 to me. So on the diabetes side, it's the same thing. People imagine that hyperglycemia,

01:00:55.800 pre-diabetes is a risk factor for diabetes. It's not a risk factor. It is the disease. It's just

01:01:01.720 in a less manifest form. So why not address the cause? A lot of resistance to that. It sort of

01:01:11.320 blew my mind. And we did NMR analysis in a number of studies to show the efficacy of, or at least

01:01:19.240 the relationship between insulin resistance score and likelihood of future diabetes.

01:01:23.820 And by the way, transitioning to pre-diabetes by no means means that you're going to get

01:01:29.200 diabetes.

01:01:29.900 I mean, my wife has been pre-diabetic 105 or so milligram per deciliter for 25 years.

01:01:39.180 It doesn't change.

01:01:40.720 So the insulin resistance score can tell you whether you're more or less likely once you're

01:01:45.860 pre-diabetic to transition.

01:01:47.620 And then at that point, even though you'd like to have intervened earlier, it's not

01:01:51.980 too late to still do something about it.

01:01:53.820 And so, Jim, was the LPIR score validated on longitudinal data to predict the development

01:02:00.520 of type 2 diabetes?

01:02:02.600 Yes.

01:02:03.180 Okay. So in that sense, we'll talk about MVX and how similar that was in that regard. Has

01:02:09.220 there ever been a comparison that says, how well does it do versus an oral glucose tolerance test?

01:02:15.800 Yes. And so, well, not oral glucose tolerance test because in the real world,

01:02:20.660 too complicated. So you're really, fasting insulin is sort of a easier way to assess insulin

01:02:26.580 resistance. It has some analytic issues and practical issues. So nobody has really been

01:02:32.780 interested in using insulin for that purpose generally in clinical practice. But the LPIR

01:02:39.100 score has been compared to fasting insulin and is better. The study that shows it the best is one

01:02:45.660 that unfortunately hasn't been published yet. It got very close to having the manuscript be

01:02:51.180 written. It's in the Diabetes Prevention Program, so you couldn't ask for a better study because

01:02:56.860 this was a study that put intervention, lifestyle intervention, metformin on the map in terms of

01:03:03.720 being able to do something about transitioning to diabetes once you had hyperglycemia, once you

01:03:09.300 were pre-diabetic. So this study was done about 20 years ago. We have baseline samples and then

01:03:15.660 one-year samples post-treatment. Everybody in the study had an NMR analysis done. LPIR and

01:03:23.920 other things that we can measure by NMR that make FPIR better if you want it to be better

01:03:30.020 were shown to be independently predictive, better than insulin. But most importantly,

01:03:37.460 because that was an intervention study, you know, lifestyle change and weight loss was shown to

01:03:44.600 significantly reduce the incidence of diabetes in these people. Metformin less effective but

01:03:50.060 significant compared to placebo. So what we were able to show really nicely is that the LPIR score

01:03:58.620 was reduced significantly by lifestyle, less significantly by metformin. Brand protein amino

01:04:05.060 acids, which we haven't talked about yet, but higher branch chain amino acids are also related

01:04:09.620 to insulin resistance and can improve the LPIR score. And we actually have developed another

01:04:15.800 score called the diabetes risk index that integrates or adds branch chain amino acids to

01:04:21.120 LPIR. Okay. I was going to ask you that. So just to confirm the DRI, the diabetes risk index,

01:04:27.680 is the LPIR score inclusive of the three branch chain amino acids. And when do you recommend

01:04:34.580 using one of those versus the other. Well, you know, it really never has gotten

01:04:38.340 to be far enough along or to have the amount of acceptance that that discussion has even occurred.

01:04:46.260 Is DRI commercially available with LabCorp now? It is, it is at LabCorp.

01:04:50.100 Okay. We should also point out for folks who are wondering,

01:04:53.300 your company Liposcience that you founded in mid-90s was acquired by LabCorp 10 years ago,

01:05:00.020 a little more than 10 years ago. And so for those of us like me, the dinosaurs, like we used to

01:05:05.280 still order a liposcience test. Now it just all happens through LabCorp. So presumably that has

01:05:12.060 increased, I assume, some uptake of the test. So I wanted at some point to get into the question of

01:05:19.840 why hasn't broad clinical translation occurred? Because it hasn't. Right now you can only go to

01:05:25.780 lab court for this information. So I'll go there now, briefly, if you don't mind.

01:05:31.400 Please.

01:05:32.460 So liposcience began, as I told you, as a spinoff of the university. I left the university.

01:05:39.320 And we tried to convince people that size didn't matter and LDLP did matter. And we did that for

01:05:45.520 quite some time. And we were a laboratory testing company. And so samples were sent to liposcience

01:05:50.940 and you get the results back. You had beautiful results. And that color report. I loved it so much.

01:05:58.200 Why not use color printers? It was beautiful. But the business objective was never to be a lab

01:06:05.140 that got bigger and bigger and did more testing. It was to make the ability to do NMR testing

01:06:10.380 available to any laboratory in the world. So we started by what was available. I talked about

01:06:17.000 NMR machines being in every chemistry department. These are research NMR machines. They're

01:06:21.920 engineered for multifunctionality. You can do any weird NMR experiment. There's lots of

01:06:27.960 variations on the theme. We wanted NMR to do one thing very efficiently and as rapidly as possible.

01:06:35.920 And so ideally, 30 seconds or less, pop one sample in, automatically another sample comes in,

01:06:43.860 boom, boom, boom. So we realized that we couldn't use a research NMR spectrometer for that purpose.

01:06:50.220 We used them for the initial years at Liposcience because that's all there was.

01:06:54.840 But we wanted to transition from a lab testing company into an IVD company,

01:07:00.220 in vitro diagnostics company. All laboratories rely on IVD companies to supply them the machinery

01:07:06.560 and the reagents to do all these assays. So LabCorp does 3,000 assays. They rely on other people

01:07:14.260 Roche, other people, to provide them with the wherewithal to do the testing.

01:07:19.020 Those are IVD companies.

01:07:20.260 We wanted to be an IVD company at Liposcience, make an NMR analyzer that looked just like

01:07:25.780 a regular chemistry analyzer to a med tech that had no experience, no knowledge of NMR,

01:07:31.440 walk up to it with 200 samples, present the tray, push the green go button, and walk away.

01:07:38.860 So that's what we actually did.

01:07:40.320 and a lot of investment and a lot of time and effort was put into that.

01:07:44.840 And that is the Vantara NMR analyzer.

01:07:47.760 It's the only existing NMR analyzer in the world.

01:07:51.900 And we went to the FDA because these analyzers have to be FDA cleared

01:07:56.640 in order to go into different laboratories.

01:07:59.220 That was an adventure because what did FDA know about NMR spectroscopy?

01:08:03.720 And it was a new platform, a new way of testing that had to be understood.

01:08:08.980 When did you get the CLIA approval?

01:08:12.060 No, not CLIA.

01:08:12.900 We got FDA clearance for LDLP and the Vantara analyzer in 2011.

01:08:18.880 Oh, wow.

01:08:19.560 Okay.

01:08:19.880 Yeah, 2011.

01:08:21.940 And so these Vantara analyzers, a number of them were manufactured.

01:08:26.580 And then about the time that just before, a year or so before LipoScience was sold to LabCorp,

01:08:33.100 these analyzers started to be distributed to major laboratories. LabCorp was one of the

01:08:41.120 recipients of these analyzers. So they could do the testing in-house rather than having to send

01:08:46.000 the samples to liposcience. Clinical laboratories hate sending send-out samples. And when that's

01:08:53.720 for a rare cancer or something, when it's a fairly rare event, no big deal. If you're doing

01:08:58.760 hundreds of these a day, you don't want that hassle. So they were very happy to receive the

01:09:04.640 Vantara analyzer so they could do the testing in-house. And every time they did, they would

01:09:09.060 pay liposcience for each analysis. Okay, that was the business model. We also made Vantara

01:09:16.020 analyzers available to the Mayo Clinic, Cleveland Clinic, Scripps Clinic, Arup, a big reference

01:09:21.780 laboratory in Utah. So we were on the way to making it available broadly because we didn't

01:09:28.060 want to be the only people that could do NMR testing. And also we wanted to convince people

01:09:32.360 that this wasn't, you know, magic. It was real. And it was analytically in many respects much

01:09:40.340 better than reagent-based chemistry testing. Unfortunately, by that time, liposcience had

01:09:45.800 gone public. I was no longer on the board. I was basically the investment that we needed to start

01:09:52.660 liposcience, we basically took too long to get to the payout of the initial investors. So people,

01:09:59.960 the investors were not patient. They were very patient up until then, but you almost couldn't

01:10:05.460 blame them because they wanted to get their money back, venture capital in particular.

01:10:09.700 So the company went public and then LabCorp came along and said, we'd like to buy you because we

01:10:15.660 can make more money if we don't have to pay whatever number of dollars to liposcience every

01:10:20.540 time we do this test. So it was a purely financial decision. LabCorp really didn't care about

01:10:27.240 measuring things other than the NMR lipoprofile, the LDLP, et cetera. It was financially based.

01:10:34.840 That was too bad for the vision of having NMR analyzers in every laboratory because

01:10:40.300 LabCorp is a lab testing company. It's not an IVD company. So too bad. The IVD business model,

01:10:48.480 the IVD vision was ended in 2014. Did you join? Did you go and become a scientist there?

01:10:58.120 The research group largely was retained by LabCorp. And we had a pipeline of things like

01:11:02.820 LPIR and DRI, things that were coming down the pike that we'll talk about later. So really the

01:11:10.360 best was yet to come in terms of the clinical value and the things that could convince people

01:11:15.360 that having Vantara analyzers in their lab was a commercially useful thing, but also a clinical

01:11:21.180 useful thing. So the bad news is that LabCorp, understandably not sharing the IVD vision and

01:11:29.980 being a laboratory that wanted to have NMR be proprietary to themselves, took back these

01:11:35.740 analyzers that had been placed in these other reference laboratories. And then unlike liposcience

01:11:42.380 that knew that when it developed a new test, it needed to go to the trouble of creating awareness

01:11:49.180 and interest and doing the studies to prove the clinical efficacy of these tests. That's a very

01:11:55.400 important activity. LabCorp doesn't do that because they're not an IVD company. They are

01:12:02.240 reactive. They're not proactive. They're very good at being reactive. And in COVID, they ramped up

01:12:09.040 the COVID testing. And, you know, so I'm not, you know, saying that LabCorp doesn't do a really

01:12:14.320 good job at what they do, but they really didn't know what to do with new knowledge that they

01:12:20.800 generated in-house by acquisition of liposcience. And so there was no marketing, no awareness

01:12:27.600 creation. And basically liposcience went invisible and is still largely invisible.

01:12:33.300 most of the testing is still done by the people that were interested in the testing,

01:12:39.440 thanks to liposciences efforts. So these X number of analyzers that were produced 15 years ago

01:12:47.480 are what LabCorp is using to produce this information and this more exciting...

01:12:53.180 And what's the life of these analyzers?

01:12:56.140 Good question, because nobody's ever... So the magnets themselves, these superconducting

01:13:01.620 magnets last a long time and don't degrade, but they're using PCs, 10-year-old PCs. And anyway,

01:13:11.460 all the moving parts. So they're not going to last much longer. And this is what maybe I'm most

01:13:18.520 concerned about and most interested in people hearing about, because what needs to happen

01:13:26.620 for this to continue, this clinical translation to continue at LabCorp, but also ideally

01:13:34.020 broaden to the rest of the world, is for an IVD company to come in and basically acquire

01:13:41.300 the liposcience technology. And there's a lot of patents, intellectual property associated with

01:13:46.360 this. A lot of expertise that comes from in terms of service and keeping these machines functioning

01:13:52.260 well. And we've done many millions of tests and you learn by doing. So what people don't know is

01:14:01.240 that at some point in the future, maybe sooner rather than later, these Vantara analyzers,

01:14:06.460 first generation, only generation, are going to cease to function. And it would be a real shame,

01:14:13.520 especially with the things that we're going to talk about that are more exciting than LPIR and

01:14:19.200 in my mind, LDLP, et cetera. So there's an issue here. There's a problem, and hopefully it'll be

01:14:25.860 addressed by an IBD company taking over. I've tried to make the case to big multinational IBD

01:14:32.400 companies, but NMR is too exotic. I mean, I basically show them that this is a completely

01:14:38.740 de-risked proposition because we've already gone through the regulatory hurdles. We've already

01:14:44.880 made the analyzers. We already have the experience. So it's not like starting from scratch with

01:14:49.500 something that you're unfamiliar with, technology you're unfamiliar with. But, you know, these big

01:14:54.280 companies have a lot of inertia and talking to the right people or having these big companies

01:14:59.500 be entrepreneurial. So I think this is maybe, if it happens, it'll be possibly a smaller IVD company

01:15:07.560 or a new startup IVD company, basically taking the torch that liposcience-

01:15:12.740 And has liposcience said that they would be willing to sell those assets, the IP?

01:15:16.620 It's LabCorp that owns all those assets.

01:15:18.060 Sorry, that's what I mean.

01:15:18.640 Has LabCorp said that they're willing to sell that IP?

01:15:19.980 Yes.

01:15:20.660 And I think they realize, well, yes, they're interested in licensing the patents.

01:15:28.140 Most of the patents have to do with the assays, so LPIR and MBX that we'll talk about.

01:15:34.900 So absolutely.

01:15:36.540 But who's going to license them if there's no machinery to produce the information?

01:15:41.100 So what is your estimate of the cost per machine now, if you were going to make a Gen 2?

01:15:48.520 Well, so these machines cost on the order of $400,000 or $500,000.

01:15:53.580 But the beauty of it is, of course, that they last a long time, which we've shown,

01:15:58.000 but they have no consumables associated with the assay.

01:16:01.300 So it costs just as much to do 10 assays a day as 1,000 assays a day.

01:16:06.280 So in a high-volume setting, these tests are very cheap.

01:16:10.100 And what we've talked about, we talked about LDLP and the NMR lipoprofile was the report that gave people the LDLP information.

01:16:19.920 But that comes from this simple NMR spectrum that can then use to extract much more information than LDLP, LPIR, DRI.

01:16:30.520 It's just part of the story.

01:16:32.220 There's all these other things that we'll talk about later.

01:16:34.800 So in terms of efficiency, analytic efficiency, you couldn't ask for anything better because

01:16:41.060 if you want to use NMR to produce a lipid panel, and actually after telling you that

01:16:49.460 NMR only measured particles and not cholesterol, the basic information is encapsulated in all

01:16:56.320 these NMR signals. So you can actually feed NMR data to a machine learning algorithm, AI,

01:17:03.120 and train it to produce accurate lipid panel and ApoB information.

01:17:09.660 So we published this four or five years ago that you can actually use the NMR spectrum

01:17:14.280 to produce an extended lipid panel, ApoB plus lipids, no incremental cost to a lipid panel.

01:17:21.880 The ApoB doesn't add cost.

01:17:24.660 It more than doubles the cost of a lipid panel if you want to do it by chemistry

01:17:28.100 and use regular reimbursement in the U.S.

01:17:31.360 So it's extremely analytically efficient and cost effective.

01:17:36.480 The thing that people probably don't appreciate and I didn't appreciate as a naive professor,

01:17:42.520 I thought, well, surely this will be commercially attractive if it can cut the cost of doing these tests.

01:17:51.320 The cost of these tests is so much smaller than the price that is charged for these and the insurance pays for these tests

01:17:58.300 that cutting the price, the cost in half of doing the analysis, makes no difference whatsoever.

01:18:04.800 So this is, you know, we'll maybe get into this later because what is true is that NMR,

01:18:11.740 a single scan, can tell you a heck of a lot more than just your cardiovascular risk,

01:18:18.240 your inflammation level, your diabetes risk. Your overall mortality risk, as we're about to discuss.

01:18:23.820 This all comes in the same assay essentially. Well, I can think of no better way to introduce

01:18:31.300 the MVX assay. So I'll tee it up for you and then take away the story. So by the way,

01:18:38.760 I went to have my first MVX drawn. We've been doing it on our patients for about four months now

01:18:43.780 and I just haven't got around to doing a blood test on myself. So I went out to LabCorp two and

01:18:50.140 half weeks ago because I wanted to have the results back when we were sitting here. And

01:18:54.180 wouldn't you know it, Jim, they screwed up the assay. So I don't have it. They got everything

01:18:58.760 else. Every other test we ordered, they got, but they butchered this one. So I don't have my own

01:19:03.920 to talk about. But the MVX is a composite score that measures, if I recall, six things. So small

01:19:11.880 HDLP, glyc A, which we haven't introduced yet and we'll talk about, citrate, and the three branch

01:19:19.780 chain amino acids. So leucine, isoleucine, and valine. So why don't you tell us the story of

01:19:27.580 how you developed this score? And I'll just give the punchline so the listener knows why they should

01:19:33.520 be paying attention. This score seems to have remarkable, when it's normalized to a zero to

01:19:40.400 a hundred number, the higher the score, the worse it is. This score seems to have remarkable

01:19:46.180 predictive value of all sorts of things we want to avoid, starting with death,

01:19:51.640 cardiovascular death, liver disease. While the first study that I saw was done in a very,

01:19:58.080 very high-risk group of catheterized patients, and it would be easy to dismiss that it was only

01:20:02.320 valuable in that population, it's been demonstrated in healthy populations as well.

01:20:07.400 So tell us about the score. Yeah. So it's a really interesting story. And

01:20:13.380 And one question that gets asked is, well, why did you think of measuring those things? Or did

01:20:19.480 you have some mechanistic reason for focusing on these things that ended up contributing to this

01:20:26.300 MVX metabolic vulnerability index score? And the answer is no, that's not how it happened.

01:20:33.940 And I just, I do want to make this point because MVX and especially MVX raises more questions

01:20:42.800 than answers right now.

01:20:45.100 Okay.

01:20:45.480 So you really would like to understand if it predicts mortality so well, why does it

01:20:50.480 do that?

01:20:51.060 Is it causal?

01:20:51.940 Can we intervene?

01:20:53.220 You know, these are the things that really matter clinically.

01:20:57.500 So it wasn't starting out with some idea that branch gene amino acids are really important

01:21:03.500 mechanistically, even though they are. This was because NMR analysis has given us a very efficient

01:21:11.160 tool to measure baseline samples from studies that follow people over time, 10 years, 20 years,

01:21:20.640 30 years longer, to see what develops. And so you really would like the ultimate idea in a biomarker

01:21:29.480 is that it predicts the future. You want to know not if it's related to cardiovascular disease,

01:21:36.580 but is it related to getting cardiovascular disease in the future? And so we have availed

01:21:42.620 ourselves over the years of baseline samples for many very large clinical studies. You mentioned

01:21:49.960 MESA, Multi-Ethnic Study of Atherosclerosis. That's a NIH-funded study that started in about 2000.

01:21:57.240 It's had more than 20 years of follow-up now.

01:22:00.000 We were asked to measure the baseline samples 15, 20 years ago for free.

01:22:05.680 Now I've worked for a company that has to make some money, but I had enough freedom

01:22:10.360 to be able to offer that test.

01:22:12.940 Initially, I asked for money, but then they finally said, well, we just can't find the

01:22:16.420 money for that, but we'd really like to have the information.

01:22:19.360 So it was measured for free.

01:22:21.180 Same thing from other very large studies.

01:22:23.780 Women's health study, 26,000 women at baseline.

01:22:26.720 in a study that's been followed up for many, many years.

01:22:31.200 Many clinical trials, Framingham offspring study, all sorts of intervention trials,

01:22:35.760 Jupiter, et cetera, the diabetes prevention program. I would like to quickly go back to

01:22:42.080 that just for two seconds because we didn't finish the thought of what the diabetes prevention

01:22:48.640 program, even though it's not published, tell us. And the important thing that it told us

01:22:53.360 was not just that LPIR score goes down with lifestyle intervention and branched

01:22:58.720 amino acids go down, which is all good thing, but that those things going down,

01:23:06.080 the delta between where it started and where you got after the intervention,

01:23:10.640 very powerfully predicts incident diabetes or the lessened diabetes risk.

01:23:15.920 So that was sort of the missing link. If you really want to argue causality,

01:23:19.360 you really want to show that lowering LPIR translates to lower diabetes risk. And it does.

01:23:26.460 It's quite powerful. It hasn't been published yet in part because the primary author died

01:23:32.960 recently, unfortunately. And so it will be eventually. All right. But that's one study.

01:23:39.400 So we've gone out of our way over time to make it possible for people that didn't have funding

01:23:44.920 to get NMR data on baseline samples for observational studies and intervention studies

01:23:51.460 going forward. MESA has been particularly useful to us, 7,000 people, baseline NMR.

01:23:59.080 And because we supplied the assay for free, the NIH has given us access to the information about

01:24:06.460 who developed cancer, who developed cardiovascular disease, et cetera, dementia, et cetera,

01:24:11.460 all sorts of outcomes. And so the beauty is that once we have taken the NMR spectrum

01:24:18.280 and gotten from it what we initially wanted to get from it, let's say LDLP,

01:24:24.820 these spectra are sitting there in a computer stored on a disk. And when we develop

01:24:32.160 the wherewithal to extract new information from the NMR spectrum, we can go back literally in a

01:24:39.280 couple hours to interrogate a very large clinical trial and get prospective information as to

01:24:45.900 whether it's predicting a particular outcome. So back to MVX, that's exactly what we did in

01:24:53.060 this cardiac catheterization study at Duke University. 7,000 people over a period of years

01:24:59.160 recruited into this biorepository. When they came to the cath lab, when they came to the cardiologist

01:25:05.540 with chest pain or some issue that qualified them to have coronary angiography done, and

01:25:13.280 their blood was taken at baseline, and it's stored at minus 80, where it's perfectly stable

01:25:17.820 for NMR analysis.

01:25:19.340 So we got a relationship with Duke to obtain those samples, do NMR analysis, and we found

01:25:26.760 in published papers that small HDL particles were particularly powerful in relating to

01:25:34.660 the likelihood somebody would die during the roughly five or 10-year follow-up of this study.

01:25:40.980 And then there was another thing that we could measure by that time called glycate. And we might

01:25:47.240 as well talk about glycate now. It's another signal in the NMR spectrum that's not where the

01:25:53.980 signal is that we interrogate for LDLP. And so for 10 or 15 years, we didn't care a whit about

01:26:00.200 any of those other signals. But a funny thing happened with where the glyc-A signal is. It was

01:26:07.520 basically superimposed or in the way of another broad signal that we were trying to interrogate

01:26:13.880 to learn about the fatty acid composition of somebody's plasma, how much monopoly and saturated

01:26:19.640 fats were in the blood. And to interrogate that, the signal, the sharp signal was on top of that

01:26:25.820 and was getting in the way. So we basically worked out a way of quantifying how big that

01:26:33.180 signal was so we could subtract it from the other signal. But that was perfectly good

01:26:38.220 quantification information. So that signal, which I wouldn't have thought to do anything with,

01:26:45.140 the person who does all my epidemiologic analyses, Irina Shalrova is her name.

01:26:50.960 We had the MESA data sitting there. We could go and very quickly use the software we had developed to measure this sharp signal. And she just said, well, does it relate to anything that happens in the future in MESA? Oh my gosh, it relates to all sorts of things, including mortality very, very strongly.

01:27:12.000 So what's up with that signal? Well, it turns out, going to the literature,

01:27:15.680 this often happens. 20 years ago, somebody published a paper saying that the signal

01:27:20.340 was an inflammatory marker. It actually comes from the carbohydrate, the glycan

01:27:27.420 decoration that's on a lot of proteins in the blood. And most of these proteins are so-called

01:27:33.640 acute phase proteins that increase in concentration in inflammatory conditions.

01:27:38.620 And so even though this signal was not telling us which acute phase proteins were contributing to it, it was a composite.

01:27:50.020 And not only did it essentially quantify the most abundant four or five acute phase proteins that contributed to this signal,

01:28:01.400 But this carbohydrate decoration, this glycan decoration, is used for all sorts of purposes, signaling of different types, et cetera.

01:28:11.440 So there's very complex.

01:28:13.480 People worry about the glycome.

01:28:15.160 It's like the proteome in the genome.

01:28:17.300 There's a glycome.

01:28:18.740 I know nothing about any of this.

01:28:21.220 But one thing that we know where the signal comes from, and it comes from particular sugars on this carbohydrate.

01:28:29.340 And in inflammatory conditions, more of this decoration is put on some of these proteins.

01:28:35.540 So actually, this glycase signal is reflecting not only the levels of these acute phase proteins,

01:28:41.600 but how much glycan is on them, which is also connected to inflammation.

01:28:45.860 So it turns out, rather miraculously, that how big this signal is is a very useful

01:28:51.100 measure of your steady state, of your systemic inflammation level.

01:28:57.500 It's a very stable parameter because it's the integration of lots of different things.

01:29:02.240 So rather than CRP that you typically measure clinically to assess inflammation, it's very

01:29:08.540 volatile.

01:29:09.080 It goes up and down day to day.

01:29:11.220 So all clinical recommendations for the use of CRP information say that you should take

01:29:17.100 the average of two or three different measurements.

01:29:19.760 Nobody does that.

01:29:20.700 That's the recommendation to get around some of this biological variability.

01:29:26.540 GlycA doesn't suffer from that problem.

01:29:29.280 That's probably in large part the reason that when added to CRP in a prediction model,

01:29:37.040 it typically assesses the outcome more strongly than CRP does.

01:29:43.560 But CRP tends to independently add.

01:29:46.400 So inflammation is a very complex thing.

01:29:48.860 this is a very unspecific marker, but it's a very stable and useful clinically marker of

01:29:56.780 systemic inflammation. Does it include, do you think, or capture what we see in the various

01:30:01.160 interleukins? Yes. So it's correlated strongly with IL-6 and other interleukins. And these

01:30:08.940 correlations, but that's all we know. I mean, so yes. And you'd really like to be more specific.

01:30:14.960 and if there's local inflammation as opposed to systemic inflammation, this is not going to tell

01:30:20.500 you anything. But what it does do is offer you a simple and very cheap, because it comes along for

01:30:27.160 the ride with the other NMR information. If you quantify this glycase signal, you have a very

01:30:34.440 powerful marker of the things that systemic inflammation contributes to. Now, using HSCRP

01:30:42.660 as an example, Jim, which, as you pointed out, is going to rise with inflammation. One of the

01:30:48.560 things that clinically we pay attention to is how high is it? So if I see a brand new patient

01:30:55.920 and their HSCRP is two or two and a half, in many ways, that's more disconcerting to me than if it's

01:31:05.480 40. Because the 40 is so high that I know it's really in response to something acute.

01:31:12.480 They're probably getting over a cold. Maybe they got a vaccine two, you know, four days ago or

01:31:18.820 something like that. Now, of course, that doesn't obviate the point you made, which is I still want

01:31:23.500 to see longitudinal data. Like I can't make it, I can't assume the two is bad because I could also

01:31:27.720 be catching something on the way down or on the way up. It could be three or one depending on the

01:31:30.920 next day. But it's usually the case that when something is very, very high, it really speaks

01:31:37.240 to acute inflammation, which is less pathologically concerning. These low simmering ones that I see,

01:31:43.300 those are the ones that give me pause. Is the glycate the same as that?

01:31:47.860 No, it's very different than that. So CRP levels could go up a thousandfold on an infection.

01:31:53.800 Glycate levels, the response is much more muted.

01:31:56.960 Because it contains so much information in it, you think?

01:31:59.820 I'm not, you know, I'm not so, I can't really answer the because as well.

01:32:04.740 It's just the observation.

01:32:05.720 Just the observation.

01:32:06.920 And it's true that if somebody has an active infection and you're trying to relate the glycate level to mortality in people that didn't have an infection, you're going to be misled by that.

01:32:18.720 It'll be higher by twofold, not a thousandfold.

01:32:21.540 So it's not immune to those changes.

01:32:23.960 But you wouldn't want it to be because, you know, it's an inflammation marker.

01:32:27.360 Yep, yep.

01:32:27.980 So in people with inflammatory diseases, rheumatoid arthritis, psoriasis, et cetera,

01:32:33.580 glycine levels are significantly elevated.

01:32:36.600 And they are responsive to anti-inflammatory treatment.

01:32:39.480 And so it has all the characteristics of a useful biomarker to assess not only things

01:32:46.300 that you would like to have some visibility to, like systemic inflammation.

01:32:51.400 I think we're coming to understand that that's an awfully important contributor, even though

01:32:56.180 we don't understand the mechanistic fine points to so many things, including mortality risk.

01:33:02.960 So anyway, I just wanted to tell you that the way that we discover these biomarkers

01:33:10.800 is different than the way other people discover a lot of biomarkers. So it's more the top down.

01:33:16.700 We have a very efficient way of doing the epidemiology. So we know already that these

01:33:21.840 markers have a strong relationship to human health conditions. A lot of the work that's done

01:33:30.340 starting from bottom up with a mechanistic idea and a particular enzyme that you might maybe want

01:33:36.320 to target as a therapy, you have to then do animal models and then you go up to humans

01:33:41.860 and then you have to do expensive trials. And then you typically measuring these things is not as

01:33:47.600 easy as it is to measure these things by NMR. So it's a completely different approach to discovery,

01:33:53.920 sort of irrational discovery, because you're basically using these large databases of

01:33:59.840 population studies and then discovering things that you really didn't go after discovering in

01:34:06.080 the first place. So there are three components to the MVX. You've already talked about the

01:34:10.800 lipoprotein one, which is the small HDLP. You've just explained the inflammatory one, which is

01:34:16.040 glycae. The third one is sort of the metabolic one, which has the citrate and the three BCAAs.

01:34:22.380 So how did you come to figure those out? So you're right. So actually with the Duke

01:34:28.620 collaborators, there had been a paper published showing that there was a glycae paper predicting

01:34:36.120 mortality. The interesting thing about this cath gen population is they came to the cath lab

01:34:41.120 because they had some presumed cardiac issue. In the study that we did, for five-year mortality,

01:34:51.120 17% of the people died in five years, and they were about mean age of 60 coming in.

01:34:56.980 So that's a pretty high- How many? 15%?

01:34:59.840 17. 17% mortality, five years in people that were not elderly. Three or four times higher

01:35:09.160 for sure than a regular population. So the presumption was that these are people that

01:35:14.220 died of cardiac causes, but less than half of them died from cardiac issues. 60% died of non-cardiac

01:35:21.980 causes. So the major thing that happened- So 60% of the 17 people, 17% of people who were dead by

01:35:33.200 65 were non-cardiac causes. Non-cardiac causes, yeah. So anyway, mortality was the most prevalent

01:35:41.160 outcome. It wasn't a new myocardial infarction or a recurrent myocardial infarction. These people

01:35:47.560 died. And so glycate predicted it, small HDL particles predicted it. We might talk more about

01:35:53.380 that. And then we simply looked at all the other things that we had learned to measure. And we had

01:35:59.300 just kind of started this activity. So branched chain amino acids, isoleucine, leucine, and valine,

01:36:05.420 ketone bodies, plasma protein. I mentioned those two things because when you look individually at

01:36:12.400 those things, ketone bodies and plasma protein, they have significant associations with mortality.

01:36:19.360 So why didn't we use those as part of the MVX composite biomarker? Because we were looking

01:36:26.420 for things that contributed independently and additively to the other things that we've already

01:36:31.040 talked about. So the inflammatory part, glycate and small HDLP, we created a subscore called the

01:36:38.960 Inflammatory Vulnerability Index, IVX. The other four seem to relate. So we discovered that these

01:36:47.520 branched-chain amino acids and citrate, independent of glycate and small HDLP, added to the prediction

01:36:54.240 of mortality. Then it was, okay, why? And so then we went to literature. So we really approached

01:37:02.800 all- And you're doing all of this inside of LabCorp. So LabCorp, at least still at this

01:37:08.660 point in time, had the appetite for the R&D. Yes, although really, yes, they paid our salary,

01:37:14.480 but that was, we were sort of left alone in this little building to continue what we were doing

01:37:21.040 at liposcience. So thanks to LabCorp for not getting rid of everybody. So then it was going

01:37:28.660 to literature, trying to figure out, does this make sense biologically that these things might

01:37:33.120 be related to mortality? And that's where you come to the literature that very powerfully speaks to

01:37:40.240 the great mortality risk that people with several acute diseases, especially kidney disease and

01:37:48.180 dialysis patients, heart failure patients. Anything with cachexia, I could see increasing.

01:37:54.820 Anything with cachexia, so does sarcopenia, old people. So there's a lot of literature, 1.00

01:38:03.400 especially in the kidney disease literature, that describe this vulnerability as coming from

01:38:10.180 a so-called malnutrition inflammation syndrome. And it's called many other things,

01:38:15.980 protein energy wasting syndrome. So the cachexia, the wasting syndrome is part of this,

01:38:21.980 but inflammation is part of this. Inflammation is probably the context that allows these

01:38:29.140 dysregulated metabolism situations to exist. So the fact that malnutrition inflammation syndrome,

01:38:38.060 We had the inflammatory parts, we speculated, but the branched gene amino acids were related.

01:38:46.520 And they were related in the opposite direction that branched gene amino acids are related

01:38:51.200 to diabetes risk.

01:38:53.080 So high branched gene amino acids speak to insulin resistance, obesity, diabetes risk.

01:38:59.200 Low branched gene amino acid levels speak to mortality risk.

01:39:03.220 And we can talk more about, you know probably much more about why this makes sense in terms

01:39:12.500 of mechanism, because it's partly related to mTOR signaling and the whole skeletal muscle.

01:39:21.520 Yeah, turnover of amino acids and muscle protein synthesis.

01:39:24.180 And turnover and catabolism, et cetera.

01:39:25.820 But it was satisfying to find this literature and to say, oh, maybe these are just better

01:39:31.220 biomarkers of something that's already understood in the acute context, acute clinical context.

01:39:38.500 But in the Katchen population, nobody had described this in a cardiac, well, in a cardiovascular

01:39:47.040 cohort. And then, so we said, well, this exists in spades, apparently, in this presumed cardiovascular

01:39:56.620 cohort. But then we did subgroup analysis within this 7,000 people in the cath joint

01:40:02.360 study. And we asked, does this MVX association with mortality exist equally strongly in men

01:40:12.180 and women, in people with and without obesity, in people with and without diabetes, with and

01:40:18.520 without heart failure, with and without a previous myocardial infarction, with or without

01:40:23.360 coronary catheterization, occlusion of the coronary arteries with and without kidney

01:40:29.300 disease. And it's not affected by any of those things. It's equally strong, if not stronger,

01:40:34.600 in people without the chronic disease versus those that are disease-free.

01:40:41.020 So in fact, the strongest relationship of MVX to mortality, you mentioned how

01:40:45.840 exquisitely strong it is. This was looked at by, well, in different ways.

01:40:51.620 I thought that the hazard ratios were bigger in the cath study than in the MESA study of healthy

01:40:57.300 people. That's true. That's true. And the reason is that for whatever reason, if these people came

01:41:05.280 to the cath lab and they were enriched in people who ultimately suffered from these wasting

01:41:11.920 syndromes. Okay. And the fact that you see the same prediction, if not as strong, for sure,

01:41:21.720 in people with absolutely no evidence of any chronic disease. In fact, the studies that were

01:41:27.600 most recently published and the most interesting one that I'll mention is one that's not yet

01:41:33.320 published but is about to be submitted for publication, MBX in young people, 30-year-olds.

01:41:39.060 would we expect to see the relationship? And you do see the relationship, but it's weaker.

01:41:46.060 And I didn't sort of finish that. So the two subparts of MVX are the IVX inflammation part

01:41:55.100 and the other four parameters brewed together to form MMX, metabolic malnutrition index.

01:42:02.360 It's kind of an arbitrary thing to talk about these two parts of MVX because it's known that

01:42:08.320 there's synergy between these. It's a syndrome. It's interrelated, intertwined. But on the surface,

01:42:15.560 at least, it looks like it might be useful to take a high MVX score and it might be due more

01:42:22.060 to inflammatory reasons than the metabolic malnutrition wasting reasons, in which case

01:42:28.960 different therapies might be better suited for that person rather than somebody with the same

01:42:33.920 MVX score with a more malnutrition issue. So we thought it might be clinically useful. That's why

01:42:40.100 we did that. Are they both, I know the aggregate score is reported zero to a hundred. Do the two

01:42:46.180 subscores get reported that way as well? Yes, same way. Okay. So you could say Mr. Smith is a 75,

01:42:53.020 which is very high risk. But when I look at his aggregate score, his IVX is only 25. His MMX

01:43:01.200 is 80, this is the issue. It's the sarcopenia and the wasting that is really driving his risk.

01:43:07.260 It's not so much inflammation in this case. That's a possibility. I don't think that

01:43:12.740 that possibility will ever be found to be that different. So usually these are more,

01:43:18.620 because it's a syndrome. Got it. Yeah. The numbers I used are wrong,

01:43:23.600 but that's the spirit of what you mean. But that's the idea. Absolutely. That's the idea.

01:43:26.780 But, and it's really, you know, we've learned a lot since this 2023 first publication about MVX

01:43:34.080 in this cardiac catheterization cohort. The important thing that we did there, I was advised

01:43:41.040 that I really shouldn't try to publish this until we had replication because these hazard ratios

01:43:49.180 were so dramatically different from low and high MVX scores. So in that paper, we reported a

01:43:55.980 completely independent cardiac cath population in Utah, Salt Lake City, and it replicated very

01:44:02.900 well. And since then, we have been interested in seeing if it replicates in other disease

01:44:08.940 populations. And as we referred to, what about people with no overt disease whatsoever,

01:44:16.180 younger, older? Well, let's talk a little bit about that. The one that just came out,

01:44:20.980 of course, was the Maslow-D paper. Is that the one you wanted to chat about?

01:44:24.580 No, actually, we could. Because these papers are coming out now.

01:44:30.120 They're just coming out at a geometric rate.

01:44:31.900 Yeah.

01:44:32.700 Which is great. I mean, that's so exciting.

01:44:34.500 The reason is that we just have to go back to existing NMR data. So we're mining other

01:44:39.940 people's costly studies. We're piggybacking on their work, their funding, and getting very quick

01:44:47.720 gratification about whether MBX is good for this or that or the other thing.

01:44:52.180 Is there one in an intervention study? Because that would be the next step. Maybe you've already done it where you say, look, we've got an MVX at baseline. We're going to treat you guys placebo, non-placebo. You know where I'm going.

01:45:05.460 Absolutely. That has to happen. It isn't going to happen if nobody knows about MVX.

01:45:11.980 And the thing that has been a little disappointing to me is how little... I mean, nobody read that

01:45:18.860 2023 paper. I was very proud of that paper. I thought there was a lot of meat in that paper

01:45:25.420 and a lot of implications, clinically and otherwise. But people read papers that are

01:45:32.000 call to their attention. And this is part of the problem. If this was a liposcience,

01:45:36.740 if we were a liposcience, we would be promoting awareness of this paper in meetings, et cetera.

01:45:43.160 And this isn't happening at LabCorp. But now this will help.

01:45:49.380 Yeah. I mean, there's going to be a lot of people that are listening to us.

01:45:52.540 Right. And so, yes, that's the next step. What will be published this year will be,

01:46:00.480 I think, sufficient for anybody to see the replication of the phenomenology fairly quickly

01:46:09.860 after that first paper in heart failure patients, we could show this. The other thing that was kind

01:46:15.500 of nice about that heart failure population is they also had a lot of frailty information.

01:46:22.760 A lot of these were older people with heart failure in Minnesota. And so they were able to

01:46:28.420 calculate frailty scores and also biological age by, well, not in that study, but another study of

01:46:37.740 older people. So some of the other things that are talked about a lot about relating to longevity

01:46:43.780 and so on were measured in these studies that we've been able to look at MVX in.

01:46:49.400 And in terms of frailty, physical frailty, correlation, but fairly weak with MVX at a 0.2

01:46:56.920 correlation coefficient. So it isn't like you need to have physical frailty to see the MVX be high

01:47:02.900 or vice versa. And in terms of mortality risk, frailty, physical frailty on top of MVX score

01:47:10.920 adds considerably. But MVX very powerfully still in the presence of frailty score

01:47:17.200 predicts mortality. And is it always a five-year look forward?

01:47:21.660 No. And in many studies, it's longer than that in part because you want the statistical power

01:47:28.080 from more people having the outcome. But there's something powerful about

01:47:32.660 the short window. I mean, in many ways, that's actually a feature, not a bug if you can offer

01:47:36.980 that insight. Because we don't have many short-term predictive biomarkers.

01:47:41.200 Absolutely. And actually, the one study that I was sort of confusing with the heart failure study

01:47:46.960 is a study called EPIs. It's a study of older people with lots of things being measured,

01:47:55.380 including these biological age measures by chemistry assay, not the epigenetic flavors.

01:48:03.520 And a couple of years ago, I think it was 2022, they had NMR information and then they had all

01:48:09.140 this frailty information and 186 different variables. And it was done by some epidemiologists

01:48:17.260 in Minnesota using the most high-tech ways of trying to deduce whether the associations were

01:48:25.760 causal or not. I'm not really sure that that really, to me, demonstrated that. But

01:48:30.220 they use methodology that purports to assess causal relations. And out of those 186 things

01:48:38.620 that were looked at, small HDL particles were the most powerful at predicting two-year mortality

01:48:43.740 in these people. So what are the blind spots? Pardon? What would be a blind spot? Where does

01:48:50.120 it get fooled? We already gave one example, right? Which is if you're in the throes of a

01:48:54.460 brutal infection, you're getting over a cold, that could artificially elevate, although not

01:48:59.400 to the same extent as CRP, the glycate, that could offset it. Have you seen other false positives,

01:49:05.160 so to speak? No, and we haven't really looked in ways to possibly see those because we've looked

01:49:12.320 overall at the prediction in a population, and these people at baseline either have this, that,

01:49:18.860 or the other disease. Yes, it gets kind of canceled out in the wash at the population level.

01:49:22.080 Right, exactly. The question of whether it's causal or whether... So there's two questions.

01:49:33.160 One, is it modifiable? I've been speaking to the intervening three years since the paper was

01:49:41.040 published. We now have really good data, some coming very soon in different disease populations

01:49:48.780 that this replicates and is seen. It doesn't matter who you are, this relates to mortality

01:49:54.600 risk. I want to come back to this later because you said something earlier about how MVX

01:50:01.100 remarkably relates to not just mortality, but diseases that reduce mortality.

01:50:09.580 So I want to quibble with that idea a bit later. But just to get back to the question of,

01:50:18.460 are there interventions which lower MVX? And then could we do the study of that intervention

01:50:25.860 to show that that's connected to a reduction in mortality risk?

01:50:29.000 Before you do that, Jim, I want to go back to the 30-year-olds because we didn't really finish

01:50:33.800 the swing on that. No, we didn't. Are you able to talk about that or is that not published?

01:50:39.000 I'm going to talk about it and I'm not going to say the name of the study, but the draft of the

01:50:45.880 paper is about to be submitted. It's just hard for me to wrap my head around the fact that any

01:50:50.600 biomarker in 30-year-olds could predict anything. This is very true and that's why this is so

01:50:57.120 interesting and novel. So the paper that appeared a couple months ago was from MESA. So we've been

01:51:04.900 talking about MESA, 60-year-old people at entry. They weeded out all the people in MESA that had

01:51:11.760 any self-reported or otherwise diseases, so restricted to healthy, and average age 60 or so.

01:51:20.880 And MVX by quartile had this stepping stone relationship. Not as strong, the hazard ratios

01:51:26.760 weren't as different as in cath gen, but very significant. So that was the first.

01:51:33.740 And do you recall in that study, Jim, what the difference was between the first and the

01:51:39.420 fourth quartile in hazard ratio? Were you talking about like a 1.6?

01:51:44.060 Unadjusted, it was about, maybe I might be confusing other studies, but adjusted,

01:51:50.620 it was like 1.5 to 2.

01:51:52.820 Okay. And we'll link to every one of these studies in the show notes.

01:51:56.380 But basically, in otherwise healthy 60-year-olds, the difference between being in the worst quartile, 75 to 100 score, versus the bottom quartile, 0 to 25, would be about a 50 to...

01:52:10.020 No, no, that's per standard deviation. So it's actually greater than that. It's maybe two to threefold and greater.

01:52:16.640 By quartile, by top to bottom quartile. Got it. Wow. Okay, so big, big difference.

01:52:20.440 Yeah. So, okay. So those are 60-year-olds. So these are people. So my idea about MVX,

01:52:28.260 people die when they're older. And so this thing, MVX comes into play when you're older

01:52:33.920 and maybe MVX scores go up with age. MVX scores are virtually unassociated with age.

01:52:41.400 What?

01:52:41.960 Yes. Unassociated with age. And the major proof of that is this 30-year-old study.

01:52:47.240 So here we've got 3,000 plus people who were entered into the study between the ages of 25

01:52:56.760 and 30. So we have an NMR analysis that was done when the average age of these people was about 30.

01:53:05.640 And this has got to be 30, 40 years ago, because otherwise you wouldn't be able to do anything.

01:53:09.320 was 35 years ago. And there were blood samples taken at time intervals, more frequent than five

01:53:18.520 years for the first few years and then five years after it. So we have NMR data at year 10, 15, 20,

01:53:24.920 25, 30. So we know how stable the MVX score is over time. That isn't actually reported in this

01:53:35.460 particular paper, but it's very stable. But the really interesting thing is that the distribution

01:53:41.000 of MVX scores when these people were 30 years old is just as wide, almost identical to the 60-year-old

01:53:48.940 people. There are people with low scores and high scores. And as you said, this is 30 plus years

01:53:55.720 follow-up. So this is definitely premature mortality we're talking about. And these are

01:54:02.220 all people that at baseline also were excluded from having any pre-existing comorbidities.

01:54:09.160 Okay. So there's not only young, but they're healthy. We need to just stay on this for a

01:54:13.040 moment, Jim. This is so counterintuitive. I just want to make sure not a single listener

01:54:17.960 is failing to appreciate what you are saying. So I'm going to repeat it back and I want you

01:54:23.360 to correct me because there might be errors where I'm oversimplifying. 35 years ago, we had a whole

01:54:29.240 bunch of people that were aged 25 to 30, and we excluded all the people that had known issues. So

01:54:33.860 if you had type 1 diabetes or you had some childhood cancer or whatever else, we didn't

01:54:38.680 include you. We really looked at boilerplate healthy 25 to 30-year-olds. We draw their blood.

01:54:44.560 Every 5 to 10 years, we draw their blood again, and we run the MVX assay on them.

01:54:51.380 The first and most surprising potentially feature, certainly the first surprising feature

01:54:55.940 when you're doing a bunch of MVX scores on healthy 30-year-olds is that any of them had

01:55:01.720 elevated levels. Because the most obvious thing is MVX must at some level be a correlate with age,

01:55:08.460 which is the single greatest predictor we have of mortality. And so big surprise number one is

01:55:13.700 you could be 25 or 30 years old and have an MVX score of 75, which is very high.

01:55:18.960 Okay. The upper quartile, the average was about 50, 51 score. The bottom quartile was about 27.

01:55:28.900 So that's the range. So the distribution looks a little different than it looks in a 65-year-old

01:55:34.360 population. No, but that distribution is identical to Mesa. I mean, it's identical.

01:55:37.180 And Mesa was in 60, 65-year-olds. Okay. And then you're saying not only do we have this

01:55:44.020 distribution that mirrors that of people 30 years older, as we followed these people for

01:55:49.440 35 years, it predicted mortality. I am not aware. I'd have to think, Jim, but I don't think I can

01:55:58.200 imagine a biomarker outside of a very extreme state. So you mentioned FH. Okay. If I know that

01:56:06.540 I have two 30-year-olds and one has FH and one doesn't, and the FH1 is not treated, I can tell

01:56:12.940 with a very high certainty that person's going to be dead in 30 years. This one will not,

01:56:17.180 or very unlikely to. But outside of edge cases like that...

01:56:21.660 Well, so that introduces the question, how did you get a higher or low MVX score when you're 30

01:56:30.300 years old? You didn't acquire it because of some disease vulnerability. You acquired it

01:56:37.580 at birth. We don't know. We have to look now at younger people.

01:56:41.420 Should we look at Framingham and Framingham offspring and try to tie that together?

01:56:45.160 We have plenty of MVX data from studies where genomic information is available,

01:56:52.480 epigenomic information is available. I mean, it's a great question. I mean, again, like I said,

01:56:57.820 I've got questions more than answers, but it's really fascinating and novel and important,

01:57:04.860 I think, for sure. And because of what we've just talked about, I want to go back to this issue of

01:57:11.920 whether MVX has anything to do with whether you're likely to develop cardiovascular disease

01:57:19.040 or diabetes or dementia or whatever. And what you will find already in the literature

01:57:30.380 are papers that indicate or suggest that MVX does have those associations with the diseases,

01:57:38.660 many diseases that cause mortality. But I think all of these are artifacts of the way the analysis

01:57:45.020 was done. Because as you appreciate, almost all cardiovascular endpoint trials, as well as many

01:57:52.680 other types of disease, endpoint trials, cancer, whatever, kidney disease, they typically combine

01:58:01.020 fatal and non-fatal events. So if you die of a heart attack as the first consequence of having

01:58:10.120 cardiovascular disease, or if you have a myocardial infarction and survive it,

01:58:14.540 these are grouped together in a composite endpoint called CBD. And when you look at

01:58:22.580 cholesterol, it makes perfect sense because of the etiology, because of how the cholesterol is

01:58:30.680 connected to cardiovascular disease and events mechanistically, that there's nothing wrong with

01:58:36.960 using a composite endpoint. The etiology is the same. You get cardiovascular disease,

01:58:41.740 you die from it. You get cancer, you die from it. So you want to have a biomarker that predicts

01:58:47.600 whether you're going to get the disease, and then that automatically tells you what your risk is

01:58:52.000 for dying of that. What this says is that maybe there is a separate influence on whether you're

01:59:01.440 going to die from the cardiovascular disease or the cancer or whatever. And that's your metabolic

01:59:08.480 vulnerability, your metabolic frailty, if you will. I like the idea of metabolic frailty because

01:59:14.460 frailty connotes susceptibility to dying. And even though people with high MVX score,

01:59:21.640 like these 30-year-olds who have high MVX score, you look at them, they don't look any different

01:59:25.920 than the people with low MVX score. So you don't see the frailty. But metabolically, it's there.

01:59:31.060 It's basically setting you up to be more susceptible to dying from whatever disease or event, old age, that is going to contribute to your death.

01:59:44.040 So dying sooner versus later is what MVX seems to influence, as opposed to getting the diseases that, quote, cause this.

01:59:53.560 I tried to say this in the paper, but it's becoming much more clear now, especially with

01:59:59.960 these 30-year-olds, that this is something that has to do with dying, not getting the diseases

02:00:08.080 that cause the death. So listening to you say this gives me an idea for a study that I'd love

02:00:14.840 to see you do, Jim. So I don't know how much time you spend in the oncology world, but I'm sure

02:00:19.820 you're familiar with Keytruda. To my knowledge, Keytruda is the single best-selling drug of all

02:00:26.900 time. And in many ways, it's been a miracle drug in oncology, the single most exciting

02:00:32.200 development in cancer in the last 25 years. For folks unfamiliar with it, this is a checkpoint

02:00:38.020 inhibitor. So people that have a PD-1 mutation that take this drug, regardless of what kind

02:00:43.280 of cancer they have. This could be pancreatic adenocarcinoma, lethal cancer. If you have this

02:00:49.540 mutation, this drug basically takes the brakes off the immune system and your immune system

02:00:56.560 eradicates the cancer. But here's the question. Why could you take two people that have the exact

02:01:02.720 same PD-1 mutation, the exact same cancer by all intents and purposes, and you give them both

02:01:09.980 Keytruda and one of them responds and one of them doesn't. Like, we don't know. We do not

02:01:15.680 understand what's happening at the immune level to understand why that's happening. It would be

02:01:20.700 very interesting for me to understand, and using Keytruda as an example, but you could do this with

02:01:26.140 any therapeutic intervention where mortality is very quick, right? And you could ask the question,

02:01:32.380 does this become a prognostic indicator of not just mortality, but probability of success of

02:01:40.180 an intervention? Yes. That's precisely what possibilities exist. When I first talked about

02:01:48.600 this to people at Duke, the collaborators of the CathGen study, the people around the table,

02:01:54.760 the first thing they said was, wow, this would be a great test for surgeons who are

02:02:02.460 asked to operate on people who are frail or are less likely to survive the surgery or to

02:02:11.660 benefit from the surgery. You'd like to be able to screen them for resilience somehow,

02:02:19.040 but there are no biomarkers, good objective biomarkers to do that. Malnutrition, metabolic

02:02:25.200 malnutrition, there are people, I've read papers where people are, surgeons are suggesting that

02:02:29.760 people really should avail themselves of these, you know, interrogating whether somebody is sort

02:02:36.560 of metabolically or physically frail and has evidence of wasting. But if there's a metabolic

02:02:44.660 component to that that is accessible via MVX, it could be very useful. And this is one of probably

02:02:52.460 all sorts of possible applications. You mentioned this paper that was just published two days ago

02:02:59.060 on MASLD. The artist formerly known as NAFLD. Formerly known as NAFLD. So liver disease. And

02:03:08.760 The paper speaks very forthrightly about the possibility of using MVX for entry into clinical trials.

02:03:17.780 Am I correct in remembering this paper, which I skimmed, so I'm ashamed to admit I didn't read the paper, but I could have sworn it said that MVX added predictive value to FibroScan in predicting subsequent fibrosis in the Masl-D patients.

02:03:36.320 Does that ring a bell?

02:03:37.420 I think that's true.

02:03:37.680 I only skipped it as well. I was not a co-author on that paper. But as you well understand, so many clinical trials are expensive and are not done. Many are not done because the events are too rare.

02:03:55.340 So you need a huge population or too far away. And so being able to juice up your likelihood of

02:04:04.820 people dying, for example. But mortality is probably the endpoint that people care most

02:04:13.200 about, right? And so it has sort of been a hierarchy of events. People care more about

02:04:18.820 dying than they do about getting an MI or getting diabetes or whatever. So anyway,

02:04:23.680 there's all sorts of possibilities, but we're just at the beginning of the trail of answering

02:04:29.920 the questions that, and I don't even know all the questions that could be posed, but this really is

02:04:35.340 very fascinating. And the fact that it was discovered fairly serendipitously by interrogating

02:04:41.140 these epidemiologic data sets, and then the relationships seem to make sense in terms of

02:04:48.960 what's been published about the detailed cell biological mechanisms, which I don't understand.

02:04:56.320 And it's an inexpensive test. It's a...

02:04:58.420 Well, it's basically free if you think about it. So, I mean, this is what...

02:05:02.640 How many tubes of blood do you need to run it?

02:05:05.040 No. So you mentioned that you got your MVX score and they probably drew an extra tube of blood

02:05:11.180 for that. They don't need to do that. The same specimen, actually 150 microliters of plasma,

02:05:18.960 produces the NMR spectrum that produces the NMR lipoprofile, produces glucose, produces

02:05:27.360 LPIR, produces glycate, produces MVX. All of that comes from the same analysis.

02:05:37.240 And when done in high volume settings, these are tests that literally cost a dollar or less.

02:05:43.660 okay but this is the problem commercially and this is the problem with our health care system

02:05:50.900 and the way things are set up that there's sort of no there's almost a disincentive

02:05:57.620 to provide analytically free information if you can't charge incrementally for it

02:06:03.880 and what you'd like to do if you're a company is charge a whole lot more for it and then you have

02:06:08.980 the tension between what you'd like to charge and what the insurance wants to pay for. And then

02:06:14.120 convincing the insurance company that it's worth paying for is what keeps you from being successful

02:06:19.240 commercially in producing this test globally, broadly. I had this experience with LDLP.

02:06:26.960 Tremendous resistance to paying. So, you know, I think the way around that, one way around that is

02:06:32.480 to not try to get paid incrementally for it and just do something that...

02:06:40.000 So the analogy is the comprehensive metabolic panel that you get done every time you go

02:06:46.340 for it.

02:06:47.240 There's 14 things that are measured there.

02:06:49.080 If you add up what the CMS reimbursement rate is for those 14, it comes to $60 and change.

02:06:57.020 CMS pays $12 for that.

02:06:58.980 And it's because these are all done at the same time. They have some clinical reason to be done

02:07:05.080 at the same time. And economies of scale make it efficient enough that you could make money

02:07:10.740 and people aren't going to starve producing this test, getting paid $12, as much as they'd like

02:07:16.220 to get paid a lot more. This is a situation just like that where the information is essentially

02:07:25.080 free. And we made that point in the paper we wrote about the lipid panel, the extended lipid

02:07:30.000 panel that includes ApoB. We didn't really get to the ApoB. It was actually the last thing I

02:07:36.500 wanted to get back to, which was the discordance between ApoB and LDLP. Let's get back to that

02:07:40.300 because the ApoB story is the same as the LDLP story. And the reason that I have partnered rather

02:07:49.060 than competed against Alan Snyderman, who's the biggest proponent of ApoB, is that it would be

02:07:55.040 disingenuous to say that one is really better than the other. I could make the case that the

02:08:01.240 NMR analysis tells you LDL-P, but also TRL-P, triglyceride-rich particles. Subspecies might

02:08:08.060 be differentially related. There are people publishing papers that suggest that's true.

02:08:12.560 So you could definitely be ahead of the game with more information than ApoB provides. ApoB is just

02:08:17.940 a single measure of all the ApoB on LDL and VLDL particles. But the challenge is convincing people

02:08:26.160 that you should do something other than measure cholesterol. And so you need as many people in

02:08:34.320 that fight as possible. So Alan and I are both telling the same story. And that's why we

02:08:40.800 transitioned. I mean, I advocated that we use NMR to produce ApoB, which actually we got

02:08:47.820 FDA clearance for the quality of the ApoB information that comes from the NMR spectrum

02:08:52.700 via the machine learning approach. So the idea was the extended lipid panel would have no

02:08:58.880 analytic cost associated with adding ApoB to a lipid panel. Now you have a better lipid panel.

02:09:04.860 The way that Medicare reimbursement is set up now, ApoB gets paid 20 bucks, lipid panel about 13

02:09:13.900 bucks. Last I looked, it might have changed a little. So if you want to add ApoB to make the

02:09:19.320 lipid panel better, it's more than double the cost to the payers. The payers aren't going to

02:09:26.160 want to do that. And what's the CMS reimbursement on the NMR of lipids? It's about $30 and change.

02:09:33.540 and that's for the NMR lipoprofile. The NMR lipoprofile comes with LPIR.

02:09:42.520 Okay. Yep. Okay. We couldn't get that FDA cleared at the time, but you might know that

02:09:48.440 a lot of laboratories can offer tests that are not FDA cleared through sort of a loophole in the,

02:09:55.440 the FDA has decided to exercise discretion about whether they will enforce this or not.

02:10:04.560 So laboratory-developed tests, LDTs, individual laboratories can develop their own tests,

02:10:10.540 go through a, you know, get CLIA certification. So this is more laboratory certification for

02:10:15.260 how well they perform the test. But the actual demonstration of the clinical utility of the

02:10:20.480 test is something that FDA cares about, but CLIA doesn't care about. And so it's an easier path to

02:10:29.100 offering commercially a test that doesn't have to go through FDA clearance. And so LPIR was added

02:10:38.360 to the NMR lipoprofile as an LDT, not part of what was cleared. LDLP was cleared, but not without

02:10:46.420 great difficulty. So anyway, a lot of the reason that NMR wasn't commercially successful has to do

02:10:57.180 with what I just explained about the resistance of payers to pay any increment to what they're

02:11:02.640 paying for now. And you really need to demonstrate. The path to getting insurance to pay is to get

02:11:10.380 some advisory panels to some clinical guideline group to bless it. And Alan Snyderman can speak

02:11:18.560 to the difficulty of having ApoB blessed by the cholesterol guidelines.

02:11:25.080 Although the European guidelines have.

02:11:26.520 The European guidelines and now the US guidelines are getting, but it's still, 0.59

02:11:31.520 it's just ridiculous. But part of it is because the guideline writers

02:11:34.640 are trying to protect the payers, which that shouldn't be their job. They should be assessing

02:11:42.180 the clinical utility only and let capitalism worry about. The actual Medicare reimbursement

02:11:49.420 cost for ApoB, which was set many, many years ago, has nothing to do with what it costs to do

02:11:55.080 these immunoassays on these modern analyzers. So again, there's this complete disconnect between

02:12:00.100 what's charged and what's paid for and what it costs to measure. It's the same thing that drug

02:12:05.340 companies are defending the prices that they pay to support the research, et cetera. So you can make

02:12:11.540 the argument that you need to stay in business, you need to make more money. But anyway, I succeeded

02:12:17.620 more as a scientist than as an entrepreneur in what happened to the liposcience because

02:12:25.740 I really, we really had gotten quite far down the road of making NMR testing broadly available

02:12:32.100 to the benefit of so many people internationally. And that just got, you know, tanked when it was

02:12:41.720 purchased by a lab testing company instead of an IVD company. So I hope anybody listening who has

02:12:47.220 Well, I mean, yeah, I don't know that door is closed indefinitely. I think that the MVX test

02:12:53.060 offers a very compelling reason why another company might want to come along and purchase

02:12:59.960 those assets, especially given the prognostic utility of that test. So let's talk now about

02:13:06.360 this edge case of CTAP inhibition. And one of the first things that stood out to me looking at the

02:13:16.540 Broadway and Brooklyn trials, which were the phase three trials of Obisetrapib,

02:13:23.620 were that the reductions in LDLP and LDLC were greater than the reductions in ApoB if memory

02:13:32.000 serves correctly. What do you think is happening there? I know what's happening.

02:13:38.580 So the NMR analysis, first of all, was using an older algorithm than the

02:13:45.640 one that we've been using for the last five years. But even in the older algorithm,

02:13:52.380 the issue of whether NMR can reliably quantify these very abnormal HDL particles that are

02:14:02.820 produced by CTEP inhibition. So HDL cholesterol doubles or more than doubles, not because the

02:14:09.960 number of HDL particles doubles. In fact, the number of HDL particles actually goes down a bit

02:14:16.560 overall. So what happens with C-tip inhibition is smaller particles are made into larger particles.

02:14:22.860 So the number of small particles goes down, number of large particles goes up.

02:14:27.060 These contain, the large particles contain five or 10 times more cholesterol per particle than

02:14:31.920 the small ones. So HDL cholesterol goes way up and HDL particle number does not.

02:14:38.180 But the problem in terms of the analysis is that there's a natural...

02:14:45.080 So again, we're taking advantage of NMR signals from the different size lipoproteins being

02:14:52.220 detectable and differentiated from their neighbors, right?

02:14:56.360 And so at the interface of small LDL, the LDL gets so small, and then the largest HDL

02:15:06.220 is its nearest neighbor. And there's a decent gap between the diameters of those particles,

02:15:12.180 so they don't get confused normally. But when you've got C-type inhibition creating human

02:15:17.520 beings that don't exist naturally and have HDL cholesterol of 120, 30, 50, your HDL particles

02:15:24.460 get perilously close to the size of small LDL particles. And now NMR has the possibility of

02:15:32.000 confusing the two. So that's incredible just given the size difference between these particles

02:15:38.400 normally, like the ApoB and the ApoA1 particles. I thought they were like a mile apart on that

02:15:43.600 spectrum. So the good news though is that when you have metabolic situations that cause you to

02:15:52.520 have large HDL, you also have the LDL size distribution skewed to the large LDL. So there's

02:15:59.540 fewer or no small LDL particles. What you can get away with in these extreme cases where somebody

02:16:07.360 has really large HDL, and the NMR can tell when you encounter that situation, that you basically

02:16:14.900 take away from the deconvolution model the smallest LDL particles so it doesn't have the

02:16:20.340 opportunity to say this large HDL is partly small LDL. What happened with the algorithm

02:16:29.020 that was used in that study is you didn't have the opportunity to have small LDL at all,

02:16:38.400 even though some small LDL was probably there. And so you saw this big decrease in LDL-P,

02:16:47.720 but not ApoB, because the NMR model was not allowing you. So it's really an artifact of

02:16:56.120 the difficulty NMR has with this situation. So is the implication that, in this case,

02:17:02.020 obacetrapib produces a disproportionate reduction in cholesterol content of particles

02:17:08.760 relative to number of particles? So I know you've talked about obacetrapib,

02:17:15.920 and you and most people are very optimistic about the prospects of obacetrapib, despite

02:17:22.620 CETP inhibition, not panning out for many other drugs. And of course, these were all initially

02:17:29.900 investigated because of the potential to create higher HDL cholesterol.

02:17:34.760 And now we certainly know that HDL cholesterol is not the HDL biomarker of interest,

02:17:39.380 and people were being fooled into thinking that would have benefit.

02:17:45.380 Part of the reason I think, and this is pure speculation, but it comes from somewhere,

02:17:51.800 that even in the face of the CETP inhibitors that came closest to being efficacious,

02:17:59.980 20, 30% LDL reduction, but no benefit. Maybe something bad was happening on the HDL side

02:18:07.480 to counteract what was good happening on the LDL side. What was bad on the HDL side,

02:18:13.040 given the understanding now that just having a lot of large cholesterol-rich HDL particles

02:18:17.720 It doesn't put you ahead of the game in terms of cardiovascular risk.

02:18:22.060 What we now know is that small HDLP is powerfully related to mortality, all-cause mortality.

02:18:29.900 So what I told you is true, especially with the most powerful CETP inhibitors.

02:18:34.620 They reduce small HDLP by 10 or 20%.

02:18:37.920 If you ignore what's going on in HDL and you only look at what's happening with ApoB and LDL,

02:18:44.580 You think obocetrapid is a no-brainer. It's going to be positive. But what if people are

02:18:50.580 actually being hurt, maybe in terms of mortality risk, by the small HDLP going down if there is

02:18:56.200 a causal relationship there? Don't know that there is yet. We haven't proved that. There's

02:19:01.440 biological plausibility because of the proteins that hang on to us on small HDL particles,

02:19:09.480 which is partly why we think it makes sense in terms of anti-oxidation, anti-inflammation,

02:19:15.480 that small HDL particles might have this inverse association with mortality risk.

02:19:19.980 So anyway.

02:19:20.780 But are you saying that you think that it's possible that, because again, we still don't

02:19:25.000 have the hard outcome trial, but your thinking is that if the hard outcome trial demonstrates

02:19:31.560 utility, it might be, you're saying it could be just due to the reduction in small HDLP

02:19:36.620 more than the reduction in LDL.

02:19:38.660 I'm suggesting that people's predictions about how much efficacy there's going to be

02:19:44.160 could be wrong if they're based on the LDL production.

02:19:47.700 It may still be that the trial overall is positive or positive enough to have the drug

02:19:53.520 go forward. But the most dramatic demonstration of something bad happening while something good

02:19:59.680 is happening, and the two counteracting each other, is if the trial doesn't succeed like the

02:20:06.000 other ones succeeded. So I'm just, you know, we have to wait for the trial. And then if the trial

02:20:12.980 doesn't succeed, then I'll say, yay, I was right. But it's pure speculation.

02:20:18.720 Yeah. Yeah. Well, very, very interesting. And I'll have to go back and look and see what the

02:20:24.100 magnitude of the ApoB reduction was. But I just remember that it was less than-

02:20:28.960 It was less, but it's still very significant.

02:20:31.780 But you're heartened by the fact that small HDLP was decreased.

02:20:37.700 Yeah. And we've actually reanalyzed that data set with the more recent. We do a better job

02:20:44.240 in differentiating large HDL from small LDL. And so those LDLP results are much more in line

02:20:50.080 with the ApoB reductions than that paper indicated. Well, Jim, this is such a fascinating

02:20:58.960 space. And this is a discussion that has been long overdue. Again, I don't think there's many

02:21:05.700 people listening to us that haven't at least heard of LDLP, HDLP. They might not know what

02:21:11.800 liposcience is. They might not understand in vitro diagnostics and any of the other things

02:21:17.380 that go around to it. Probably a lot of people are not familiar with MVX, but my hope is that

02:21:23.680 that that starts to change after this. So regardless of how you think you've fared as

02:21:29.940 an entrepreneur, you've fared remarkably well as a scientist. And I think that's the most

02:21:35.400 important thing because without the scientific foundation, I don't think any of the entrepreneurial

02:21:40.300 stuff matters, but we do typically want them to be aligned. But put it this way, at the risk of

02:21:46.780 insulting an entrepreneur, I would argue that it's easier to find a good entrepreneur than it is to

02:21:51.680 find a good scientist. Yeah, that's true. And that's sort of the frustration that the science

02:21:57.500 is so solid, so much more solid than many startup companies are investing in. But at the end of the

02:22:06.020 game, it's commercial. It's financial. I suspect it's the sector, right? I suspect you wouldn't

02:22:13.360 have this difficulty getting people interested if we were talking about therapeutics. I just think

02:22:20.140 that the diagnostic space and the reimbursement environment in the United States is one that is

02:22:28.120 not especially attractive to investors. That's my suspicion as to the issue. And that's why

02:22:33.980 the MVX, I think, offers more than just a diagnostic. If it could be paired to a therapy,

02:22:41.100 if it has the ability to save enormous cost on the back end with respect to therapeutic selections,

02:22:49.300 You know, there are a few trials that need to be done to demonstrate that.

02:22:52.340 But to me, that's the interesting area.

02:22:55.040 No, you're right.

02:22:55.720 And that's what will make it successful commercially.

02:22:58.720 My vision, though, that wasn't realized is how cool would it be to go to your yearly

02:23:08.640 physical and get a lipid panel that had glucose, LPIR, glyc-A, MVX, at no incremental cost.

02:23:17.880 In the rest of the world, not the US, where you have national healthcare, there is a premium

02:23:24.280 put on how efficient a diagnostic is.

02:23:28.180 And if you can get a lot of information for less work and money, that's worth something

02:23:35.040 in the rest of the world.

02:23:35.940 It's just, you know, we tried to skin that cat in the US.

02:23:39.680 So it's, you know, I haven't lost hope, but I really left LabCorp because I didn't want

02:23:44.920 to beat my head against that wall any longer and wanted to spend my remaining years doing

02:23:49.640 the science.

02:23:50.360 So that's what I'm continuing to do.

02:23:53.020 Well, thank you, Jim.

02:23:53.760 And thanks for taking the time to come out here today.

02:23:55.700 You bet.

02:23:56.880 Thank you for listening to this week's episode of The Drive.

02:23:59.920 Head over to peteratiamd.com forward slash show notes.

02:24:04.760 If you want to dig deeper into this episode, you can also find me on YouTube, Instagram,

02:24:10.180 and Twitter, all with the handle Peter AtiyahMD.

02:24:13.380 You can also leave us review on Apple podcasts or whatever podcast player you use.

02:24:18.760 This podcast is for general informational purposes only and does not constitute the

02:24:23.420 practice of medicine, nursing, or other professional healthcare services, including the giving

02:24:27.740 of medical advice.

02:24:29.380 No doctor patient relationship is formed.

02:24:31.540 The use of this information and the materials linked to this podcast is at the user's own risk.

02:24:37.940 The content on this podcast is not intended to be a substitute for professional medical advice, diagnosis, or treatment.

02:24:43.780 Users should not disregard or delay in obtaining medical advice from any medical condition they have, and they should seek the assistance of their healthcare professionals for any such conditions.

02:24:53.920 Finally, I take all conflicts of interest very seriously. For all of my disclosures and the

02:24:59.500 companies I invest in or advise, please visit peteratiamd.com forward slash about where I keep

02:25:06.760 an up-to-date and active list of all disclosures.

The Peter Attia Drive - August 03, 2026

#402 ‒ NMR blood analysis： how heart disease risk, insulin resistance, inflammation, and mortality risk can be assessed from a single blood sample ｜ Jim Otvos, Ph.D.

Episode Stats

Transcript