The Peter Attia Drive - #07 - Deep Dive： Lp(a) — what every doctor, and the 10-20% of the population at risk, needs to know

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

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

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

00:00:19.840 some of the most successful, top-performing individuals in the world, and this podcast

00:00:23.600 is my attempt to synthesize what I've learned along the way to help you live a higher quality,

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

00:00:33.000 and other topics at peteratiyahmd.com.

00:00:41.280 In this podcast, I'm going to be discussing LPLittleA. A little while ago, we put up a little

00:00:47.540 questionnaire on Twitter that said, if I'm going to do a solo podcast interviewed by Bob, what topic

00:00:53.900 would you want to hear? And we put up two options. The first was LPLittleA, the second was

00:00:58.360 hormone replacement therapy for postmenopausal women. The survey ran for about a day and the

00:01:04.460 results were unambiguous. 80% of you wanted to hear about LPLittleA, though many of you did want

00:01:09.100 to hear about HRT and we will absolutely get to that. So on this podcast, we structured it as an

00:01:15.500 interview. Originally, I thought I would just do it as a quote-unquote lecture, but I realized that

00:01:19.840 would just be way too boring and it would be more fun to play patty cakes with Bob and have him

00:01:23.820 interview me. So that's what we did. So Bob put together an interview and he just asked me a

00:01:30.420 bunch of questions about LPLittleA. We're going to talk about what the heck it is, why you should

00:01:35.220 care, why it's problematic lipoprotein, what some of the potential treatment options are, and what's

00:01:41.380 on the horizon. Now, I got to admit, this is a bit of a technical podcast, but I also know that this is

00:01:47.260 kind of a technical loving audience. So don't be discouraged. I also think this is one of the

00:01:52.360 podcasts where you really have to be able to look at the show notes. I find some of this stuff really

00:01:57.900 complicated myself and I find a picture is sometimes worth a thousand words. So especially

00:02:04.040 when I get into stuff like Kringle repeats and Kringle four subsection two zone five, like that kind

00:02:11.080 of stuff, you've just got to be looking at a picture to understand it. So if you can't be able to look at

00:02:15.820 something while you're listening to it, that's fine, but maybe go back after the fact and look

00:02:19.440 at it or look first and then listen, something like that. But the show notes here will be very

00:02:23.520 helpful. And hopefully this answers a lot of the questions that people have been asking me over the

00:02:28.080 past year about LPLittleA. And again, if this format is helpful, let us know, because we're really

00:02:34.140 happy to kind of do one of these every couple of months where we just put up a general topic and Bob

00:02:39.440 grills me on it. So without further delay, here's the discussion with Bob Kaplan on LPLittleA.

00:02:45.820 Hey Bob. Peter. How are you? I'm doing well. I noticed you have a coffee there. Of course.

00:02:55.180 What number is that today? Seven. At least. I've watched four. Well, seven doubles, probably seven

00:03:03.100 double espressos. That's true. Yeah. Yeah. That's impressive. It's a low day. So this is the first of

00:03:12.860 what I suspect and assume we might do more of where we threw a question out to people and said,

00:03:20.540 pick one, have a vote. I don't know. We gave them a day or so. And the choice was, do you want to know

00:03:26.760 about LPLittleA or do you want to know about hormone replacement therapy in post-menopausal

00:03:33.320 women, which by the way has a more politically correct name now that I can't remember,

00:03:37.800 endocrine modulating therapies for women in menopause or something? I don't know. I was going

00:03:42.560 to go with golden years or something, something to that effect. Yeah. The point is HRT versus this.

00:03:47.320 And it was about 80-20 in favor of LPLittleA, which kind of bums me out because I actually really

00:03:51.840 wanted to talk about HRT. But next time we throw HRT in, we're going to put it up against something

00:03:58.280 like bocce ball. And hopefully HRT comes out ahead and we can talk about it. I think you have been

00:04:07.440 accumulating a bunch of questions that people have also started sending in about LPLittleA.

00:04:13.900 And I think that's what we're going to talk about. Absolutely. So a lot of questions are around

00:04:19.900 what is LPLittleA? And I thought in order to explain that, maybe we might need a quick primer

00:04:26.700 on lipoproteins to kick things off. Do you know anyone who can do that? I think I'm looking at them.

00:04:31.300 Oh, okay. I was afraid you were going to say that. Starting from the basics, if you go to your doctor

00:04:35.700 and you get a cholesterol blood test, they're going to probably show you a couple of numbers,

00:04:39.400 total cholesterol, LDL cholesterol. And if you're really lucky, they'll put bad next to it.

00:04:44.740 HDL cholesterol. And if you're extra special, they'll put good next to it.

00:04:48.520 Triglycerides and non-HDL cholesterol. That is a standard lipid panel. Those numbers are largely

00:04:56.080 unhelpful, but more importantly, they're largely misunderstood. So when people look at LDL and think

00:05:02.940 it's bad cholesterol, that immediately tells you that they're missing what the L and the D and the L

00:05:08.400 stand for. The LDL stands for low density lipoprotein. And admittedly, if you don't have a background in

00:05:16.500 biochemistry or something, you might not understand in looking at that, that that implies that it's a

00:05:21.820 macrostructure. So cholesterol, which is the principal molecule that is carried by these lipoproteins

00:05:30.080 is something that is made by the body. So every cell in the body makes cholesterol and most cells in the

00:05:36.380 body make enough cholesterol to meet their own needs at the cellular level. And the single and most 0.89

00:05:42.160 important need of cholesterol we have is cellular membranes. So cell membranes must be fluid. They

00:05:48.300 must be able to move. They must be able to facilitate the attachment of one cell to another.

00:05:53.520 They must be able to hold transporters across their membranes and things like that. And of course,

00:05:58.160 cholesterol makes up the bulk of those membranes. So in addition, you turn cholesterol. When I say you,

00:06:07.720 I'm referring specifically to certain organs like the adrenal glands, the ovaries, the testes turn

00:06:14.280 cholesterol into hormones that are either sex hormones, glucocorticoids, gonadotropins, these

00:06:21.580 things. So if for no other reason than just being able to have cells that work and have hormones,

00:06:27.720 cholesterol is pretty important because we're not going to get too far into that. The point I want to

00:06:32.280 make is that you can't traffic or move around cholesterol in the bloodstream because blood

00:06:39.500 approximates water. And so the things that move freely in the blood have to be things that are

00:06:44.880 what we call hydrophilic or things that would be soluble in water. So something like glucose can move

00:06:51.320 around the bloodstream very easily, but cholesterol cannot. And therefore it needs to be packaged in

00:06:58.960 something that is itself water soluble. And that something is a lipoprotein. And the two dominant

00:07:05.180 lipoproteins that are found in the bloodstream are the high density lipoprotein and the low density

00:07:10.460 lipoprotein. And their names are referring to their densities in a type of assay called gel

00:07:17.520 electrophoresis, which has to do with how far these things move on an ion gradient. There are other lipoproteins

00:07:24.140 that don't stick around that long. So VLDL, very low density lipoprotein and IDL or intermediate

00:07:31.300 density lipoprotein, which is almost non-existent. It has such a short half-life and the longer

00:07:37.160 residence of the LDL is probably what explains its atherogenicity. And that's why LDL is considered

00:07:43.620 the most atherogenic particle after LP little a, which we're going to talk about today.

00:07:47.680 Okay. So when you're looking at your blood test, what you're seeing is the cholesterol concentration

00:07:55.060 within the various particles. So when it says total cholesterol, it says, well, if you break

00:08:01.560 apart the HDL particle and the LDL particle and the VLDL particle, and if you can find it, the IDL

00:08:08.640 particle, how much total cholesterol do you have? And that's a number, call it 200 milligrams per

00:08:14.200 deciliter. Okay. When it says LDL-C is 120 milligrams per deciliter, that means if you break

00:08:21.300 apart the LDL cholesterols, that's the concentration of cholesterol contained within them, et cetera.

00:08:28.240 Now, in the past, we've talked about the importance of knowing the number of particles you have and how

00:08:33.920 that is a more accurate predictor of your atherosclerotic risk. And so the LDL-P, which is similar to the

00:08:42.500 APO-B in terms of its predictive power, which is the number of particles. And the reason you can use

00:08:47.640 APO-B as a surrogate for that is that each LDL particle has an APO-B, which is an apolipoprotein that

00:08:57.660 wraps around the spherical lipoprotein. It's APO-B 100 specifically. So by counting those, since each LDL

00:09:07.740 has one and only one APO-B, you can quantify the number of LDL particles. And again, we care about

00:09:13.760 that because it tracks more with risk. The VLDL, the IDL and the LDL all have the APO-B 100. The HDL does

00:09:23.460 not. The HDL has something called APO-A1. It's a different lipoprotein. And it probably explains in

00:09:30.560 large part why HDL is not atherogenic and LDL is atherogenic. The pathogenesis of atherosclerosis

00:09:39.860 is one that's predicated on, and we should probably attach in a link to this, the post on heart disease,

00:09:45.980 where I go through this in great, great and gory detail, the process by which the lipoproteins get

00:09:51.900 through the endothelial space between cells, which is actually not that hard to do. You know,

00:09:57.020 an LDL particle is somewhere between 20, 21, 22 nanometers. It's probably not an order of

00:10:03.320 magnitude, but several multiples of that is the space between the endothelial cells. So it's

00:10:09.040 the size of the LDL particle really doesn't determine the ease with which it gets through

00:10:13.320 the cell or not, or between the cells. What's much more important is because most of the LDL that gets

00:10:18.820 into the subendothelial space gets right back out and doesn't cause any trouble. Where the trouble

00:10:23.100 comes is when they get retained and when they get oxidized and when they kick off an inflammatory

00:10:29.620 response. So it is certainly the case theoretically that you could have a very high LDL, but by hook

00:10:36.240 or by crook, if your LDL particles don't get retained in the subendothelial space and don't kick off an

00:10:41.480 inflammatory cascade, you're not going to suffer the effects that you otherwise would. But all things

00:10:47.640 equal, we would love to see a lower LDL particle number because the process by which those particles enter the

00:10:52.980 space seems relatively stochastic. So in a few minutes, that's kind of the overview of these

00:10:59.740 lipoproteins. Okay. So I think one of the reasons why we had so much interest in LP little a is a New

00:11:07.300 York times article by Anahad O'Connor. I think it was January this year that was entitled a heart risk

00:11:16.120 factor, even doctors know little about. And he tells the story of Bob Harper, who was one of the

00:11:22.560 biggest loser OGs. I think it was him and Jillian were the two trainers. And so Bob had a heart attack

00:11:28.860 at a gym at age 52. And according to his annual checkups, he always checked out very healthy.

00:11:37.000 And as it turned out, according to the article, Bob has, quote, perilously high levels, end quote,

00:11:44.560 of LP little a in his blood, something that was, I don't think was ever measured prior to his heart

00:11:50.040 attack. So I think this article was an introduction to this particle for many people who read it.

00:11:54.780 Not only that, it's reported that a small percentage of physicians actually know about it.

00:11:59.580 So kind of going back to the original question, what is LP little a?

00:12:04.420 Well, in full disclosure, Anahad is a really good friend of mine. And I know he'd been working on that

00:12:09.160 story for about two years, actually. I guess I'll take a little bit of credit for getting him

00:12:14.900 interested in Apo A and LP little a. And Anahad, because he's just such a curious dude,

00:12:21.140 was sort of like blown away at this. He's like, wait, wait, wait, wait a minute. Tell me about,

00:12:25.220 you know, we walked through everything that we're about to talk about today. And

00:12:28.520 he just couldn't believe that something that was so ubiquitous, probably somewhere between one in five

00:12:33.680 and one in 10 people walking around with this elevation. And of course, it's a long tail to the

00:12:38.700 right distribution. So where you define the cutoff as perilously high is a function of how many people

00:12:45.540 will be perilously high. But he just couldn't believe it. And then I, you know, introduced him

00:12:50.740 to many of my mentors and he did his own research. And the result of that was a story that I thought

00:12:56.320 was excellent because I can't count the number of patients that sent it to me saying, oh my God,

00:13:01.480 this is that thing you're always talking about. Yeah. So what is this thing? So we talked about the

00:13:08.000 LDL particle number. So it's this spherical thing, call it 20 nanometers in diameter.

00:13:17.160 And it has a outer spherical structure that is made of lipid, cholesterol, phospholipid inside. It has a

00:13:28.580 core that consists of cholesterol ester. So this is non, this is like the cholesterol without its bulky

00:13:36.340 side chain and the triglyceride. And on the outside, as I mentioned, it has this one

00:13:41.860 apolipoprotein called ApoB100. So we'll just refer to that from now on as the garden variety LDL.

00:13:50.160 Now a subset of these, and it's mostly, as we'll probably discuss genetically determined

00:13:55.340 and inherited in a codominant fashion, a subset of these have something else attached to that ApoB.

00:14:03.320 And it's attached covalently. So that means that it's not an ionic bond. It's an actual,

00:14:09.800 in other words, it's a much stronger bond. It's a disulfide bond, which in amino acids and in

00:14:15.480 biochemistry tends to be a pretty strong bond. So the ApoB has this disulfide bond that attaches it

00:14:22.320 to a totally different lipoprotein. And it's called Apo little a. And this lipoprotein is made in the

00:14:29.840 liver. And it has a property that it resembles another molecule in the body called plasminogen.

00:14:39.040 Now, I suspect that everything I'm about to say is going to not make that much sense until you look

00:14:44.700 at the pictures. This is one of those things where a picture says a thousand words. So

00:14:48.500 what we'll probably do is, and that defeats the purpose of a podcast, I realize, because people

00:14:53.920 want to listen to this, but they don't want to have to miss the picture. But I think this is one of

00:14:57.040 those things where it's worth looking up the picture, but this Apo lipoprotein A has a repeated

00:15:01.040 folding structure. These domains are referred to as Kringle domains. So we're sort of lost in a

00:15:09.360 nomenclature of Apo little a and Kringle potato chip folds and all this stuff. And it's just like,

00:15:16.060 it's super complicated, but these repeating structures are organized by Kringle domains.

00:15:21.760 And there are five of them. Plasminogen, that is, has five of them. Apo A does not have the Kringle

00:15:29.740 one, the Kringle two, the Kringle three. It does have a Kringle four that very much resembles

00:15:35.560 plasminogen. And it has the exact same Kringle five that comes from the plasminogen. So to distill

00:15:42.240 that again, Apo A looks like plasminogen in that it has Kringle domain five and a Kringle domain four

00:15:49.500 that is similar, but it's the Kringle domain four that has 10 sub-segments. So you have Kringle four

00:15:56.360 one, Kringle four two, Kringle four three, all the way up to Kringle four nine and Kringle four

00:16:01.160 10. And if that doesn't have, you're looking at me and you're laughing, it's like, it's hard to

00:16:07.940 believe we're talking about it at this level of detail, but the Kringle four two is where you see

00:16:15.800 the greatest variability. And you can have a Kringle four two with just a couple of folds in

00:16:20.840 it. You could have a Kringle four two with 40 segments that are repeating. And that determines

00:16:27.160 the mass of LP little a. And that's going to become, I wouldn't be telling this story if it

00:16:31.620 weren't for some reason in anticipation of talking about something else. And with the plasminogen

00:16:36.320 and the Kringles in the homology, in other words, how similar are they? You're basically saying that if

00:16:43.340 you were to look up, like if you were to look at the structures of both of those, you could very

00:16:46.580 easily confuse one for the other. They look very similar. It probably depends on the similarity

00:16:51.780 during the Kringle four because the Kringle four tends to dominate it. So I don't want to give an

00:16:56.860 answer that could be incorrect because I suspect it depends on the individual. I think there are some

00:17:01.680 individuals whose APOA looks more like plasminogen because everyone's plasminogen looks the same,

00:17:07.000 but the APOA is where we see the difference. So we're really dealing with two things, which is

00:17:11.560 how many of your LDLs have those APOAs attached to them. And then what do your APOAs look like?

00:17:20.620 And the, what they look like is basically what do your Kringle segment four subsegment twos look like?

00:17:28.640 Now it turns out that between those two factors, the one that probably matters most is the number of

00:17:35.240 your LDL particles that also have this covalent bond to the APO little a. In other words,

00:17:41.180 it's probably the number of the LDL, sorry, of the APO little a particles bound to the LDL

00:17:49.300 particles through the APOB or the number of LP little a's that matters more than the mass of the LP little a.

00:17:55.320 So on that note, I was thinking about APOB and that there's one APOB per LDL. And then with APOA,

00:18:04.560 there's one APOA per LP little a, but not necessarily not every LDL, but every APOA is on an LDL particle.

00:18:13.700 Every APOA is on an LDL, but not every LDL has an APOA. And that's the difference between

00:18:21.400 individuals. When you look out at a population is the, how many of their LDLs are rolling around

00:18:28.980 with APOAs. Now, I don't think we'll get into it today, but if, you know, if there's ever an appetite

00:18:35.200 to go ultra deep on LP little a, we could probably talk about the relationship between APOE and APOA.

00:18:43.620 So it turns out that as people may know, you have three different variants of APOE. You have APOE2,

00:18:49.920 APOE3, APOE4. Of course, they combine in all six combinations that I'm sure everybody's familiar with.

00:18:55.980 But as you move from the two to the three to the four, you see LP little a go up.

00:19:01.800 You see APOB go up and you see triglyceride go down. And this is a pattern that has been

00:19:07.540 demonstrated over and over and over again. And what's interesting is why that's happening

00:19:11.300 with respect to the APOA. But again, I think that's, that's probably more the seniors course

00:19:16.280 rather than the freshman course. Yeah. On the note of the freshman course,

00:19:19.520 just looking at it and thinking, so if we're asking what is LP little a, and if you're to look

00:19:25.460 at how it's spelled out, it's capital L, lowercase p, parenthetical, lowercase a,

00:19:32.320 and close parentheses. And it's basically saying it's a lipoprotein with an APOA attached to it.

00:19:40.160 Yeah. And if you were going to like come up with an equivalency, you'd do like three little like

00:19:45.260 parallel lines as an equal line and say, that's equal to LDL hyphen little a or little APOA.

00:19:54.160 It would be like the longhand way to write that. I don't know that anybody's ever written it that

00:19:57.740 way in the literature, but that's just another way to think about it.

00:20:00.180 Okay. So piggybacking on that, can you explain the difference between LP little a mass,

00:20:07.200 then there's LP little a cholesterol, and then there's the LP little a particle content. And then,

00:20:12.920 as you mentioned, there's the Kringle domains, the number of Kringles could also be called the

00:20:17.840 different APOA isoforms. Can we quantify those? How are those measured? Yeah. I believe the first

00:20:24.620 way this was quantified, and again, this will be the type of stuff that I think would be really fun

00:20:29.680 to explore with a guy like Sam Tamikis, who's probably the world's expert on this topic. And

00:20:34.540 we should definitely make sure we get Sam on the show. I believe LP little a mass was the first way

00:20:40.640 that this was quantified and whether it was the first or not, I don't know. But what I can certainly

00:20:44.620 say is it's by far the most ubiquitous. I'm sure that 19 out of 20 times when a patient is having

00:20:50.360 their LP little a checked, it is the mass that is being checked. Certainly if a patient comes to me

00:20:55.520 and they've been at least fortunate enough to have had their LP little a checked, it's a mass.

00:21:00.120 I almost never see the cholesterol checked anymore. I think there used to be a company,

00:21:04.740 I believe called Athrotech that did the test. I think they got bought,

00:21:07.360 or at least that assay got bought by VAP and now VAP does it. But I'll explain later why I don't

00:21:12.600 think that's such a great test. And then of course there's the LP little a particle number,

00:21:16.160 which is just the counting of it. So the LP little a mass is directionally a reasonable test,

00:21:22.520 but it's not a great test. And the reason is it's measuring for particles that carry ApoA,

00:21:31.960 it's measuring the mass of everything, which is the ApoA, the ApoB, the phospholipids,

00:21:40.500 the cholesterol, the triglycerides, the dogs, the cats, whatever. It's measuring the mass of the

00:21:47.460 entire structure. Now, the larger the Kringle section four, subsection two, the more that mass

00:21:57.280 is dominated by ApoA. But you can see very quickly how you could be misled. You could take

00:22:04.260 two people that have the exact same LP little a mass, but if one of them has a very long segment

00:22:12.080 four, subsegment two repeat binding domain, guess what? He's going to have a much fewer particle

00:22:20.380 number or more to the point, the person that has the smaller segment four, subsegment two is going to

00:22:26.180 actually have more particles. And so those people are not at equal risk. It turns out that the guy

00:22:30.680 that's got more particles is at higher risk. But when a patient shows up and their LP little a mass

00:22:37.440 is really, really low, like less than five milligrams per deciliter, the likelihood that

00:22:43.260 their LP little a particle number is very high is really, really low. And back in the olden days,

00:22:49.860 and by the olden days, I mean like four years ago before LP little a particle number was measured,

00:22:54.760 I used to actually look at both. I'd look at LP little a mass and LP little a cholesterol,

00:23:00.340 acknowledging that neither was perfect, but basically coming up with a two by two, which was

00:23:05.220 if both were high, I knew you had a ton of particles, case closed. If both were low, we were

00:23:10.760 off to the races and high-fiving. And then when one was high, one was low, we would just sort of follow

00:23:15.620 up and test for reasons that we'll probably discuss later around things that could actually change

00:23:20.260 LP little a as you march down the field. But luckily, most patients were either double positive

00:23:25.740 or double negative. And therefore, you had a pretty good sense of where their risk was.

00:23:30.560 You then asked about LP little a cholesterol. So that basically is analogous to measuring the

00:23:36.360 cholesterol content of an LDL particle number. Except here, it's measuring the cholesterol concentration

00:23:42.860 of an LP little a particle number. And that, again, in isolation is not very helpful. I am not an

00:23:50.320 expert in clinical chemistry, but I've spoken with people who are. And it turns out there are some other

00:23:54.460 technical issues with that test that renders it not entirely helpful and also misleading in its own

00:24:00.800 way under certain circumstances. And it's for that reason that I really prefer looking at

00:24:05.340 the LP little a particle number, which even though it's reported in animal per liter, to my knowledge,

00:24:13.240 is not actually measured via NMR the way LDL-P and HDL-P were pioneered by liposcience. It's a

00:24:21.700 different assay, but nevertheless, it is counting the APO little a's that are attached to little APO-Bs.

00:24:28.740 And so you're getting a number of those. And for that test, we like to see people less than 50 0.98

00:24:35.140 nanomol per liter. When people are sort of 50 to 100, I put them in kind of a gray area. When people

00:24:42.440 are over 100 or certainly over 125 nanomol per liter, that's when I start to get worried because

00:24:49.080 I often get asked this question. The highest number I've ever seen on a patient is about 650 to 700

00:24:55.220 nanomol per liter. And I've got a few patients that walk around at 400, 500 nanomol per liter.

00:25:01.760 So the LP little a cholesterol, it sounds a lot like when we're measuring LDL-C. So LP little a,

00:25:09.080 if you're measuring LP little a cholesterol, you're measuring the amount of cholesterol that's

00:25:13.440 carried within the LP little a particles. And similarly with LDL-C, you're measuring the amount

00:25:19.420 of cholesterol that's carried within the LDL particle. Similarly, you would rather know the LDL

00:25:23.980 particle. Yeah. And this is, this is actually what makes it so problematic is it's even worse than

00:25:29.020 the discordance between LDL-P and LDL-C because at least when you're dealing with LDL, you know the

00:25:35.500 molecular weight. You don't know the molecular weight of LP little a. This is actually the point 0.53

00:25:39.660 I forgot that I wanted to make a moment ago. I remember once having a patient come to me with

00:25:45.080 everything but the LP little a P. And I remember thinking, well, I used to be a smart little organic

00:25:51.860 chemistry whippersnapper. I should be able to convert this from milligrams per deciliter into

00:25:58.940 nanomol per liter. And of course, anyone listening to this who knows more about chemistry than me

00:26:03.880 will remember that all you need to know is like Avogato's number and the molecular weight and you're

00:26:09.980 ready to go. But of course, you don't know the molecular weight. That's the problem. Because the

00:26:14.480 APOAs don't look the same, the whole calculation goes to hell in a handbasket. You can't actually 0.95

00:26:20.320 calculate the molecular. You don't know the molecular weight because of that Kringle sub

00:26:24.520 segment, the Kringle 4 segment too. Because it's got such variability, it's not like you can say the

00:26:29.760 molecular weight of sodium is X or the molecular weight of testosterone is Y. So with the APOA and

00:26:35.320 how it comes in different isoforms, can you measure the APOA? Yes, but it would be for a given

00:26:40.860 individual. That's the point. Yes, you could absolutely measure the molecular weight of an APOA,

00:26:44.600 but it would be like yours and mine would probably be different. So therefore, at least to my knowledge,

00:26:49.500 and again, I don't want to speak out of turn because I'm sure someone listening to this is 0.95

00:26:52.200 going to go, no, no, no, you knucklehead, this is being done. But to my knowledge, this is not

00:26:55.600 something that's done clinically. Whether it will be or not, again, that's probably a great question

00:27:00.000 for someone like Tom Dayspring or Sam Tamikas or one of those guys. But I think that today the best

00:27:05.660 test we have is the LP little a particle number. And there's a proxy for it, but I'm guessing we'll talk

00:27:11.460 about that later, which is you can also measure the amount of oxidized phospholipid. If you normalize

00:27:16.940 that for APOB, you're getting almost a one-to-one mapping of that because there's another interesting

00:27:23.960 little, it's not a trivia point because trivia tends to be irrelevant. This is actually quite

00:27:28.340 relevant. APO little a has lots of lysine, the amino acid lysine, and lysine really binds oxidized

00:27:38.380 moieties. Now, APOB does not contain much lysine at all. And therefore, APOB is not particularly

00:27:46.540 effective oxidized moiety scavenger. But APOA is. And so if you think about you're rolling around as

00:27:55.400 an LDL, now you've got a tail thrown on you, which is called an APOA, right? You've got your

00:28:00.340 little disulfide bridge attaches to APOB. You've got your APOA tail. Only you can see what my hands are

00:28:06.060 doing right now, Bob. This is great radio. Yeah, this is awesome radio. We're smiling.

00:28:09.800 We'll have pictures. The pictures are definitely worth a thousand words because you can see the

00:28:14.980 Kringle domains. And if you have a longer, you can see like longer tails and shorter tails. This

00:28:19.220 would all make sense. And then once you've got your tail in place, now you start to fill that tail up

00:28:23.820 with all these oxidized phospholipids. You can start to measure. If you measure those phospholipids

00:28:29.280 normalized for APOB, you're getting a pretty good proxy also of that. And by the way,

00:28:34.440 this may actually explain, and this is one of the questions like, you know, when we

00:28:38.720 get Sam on the show, this is one of the questions I want to ask him is just taking a step back from

00:28:44.200 all of this. Sometimes you clinically know when a person has an elevated LP little a before you take

00:28:49.460 any blood out of them. These are the patients who don't seem to fit the classic picture of someone

00:28:55.740 with premature heart disease in the family. Nobody's overweight, nobody's diabetic, nobody's

00:29:00.780 smoking, or even if they are, the disease seems to come prematurely, seems to come out of nowhere.

00:29:06.880 They also tend to, you know, if you ask enough, you might even see that somebody has aortic stenosis

00:29:10.980 and you just know the answer before you get there, especially if you have their family tree and you

00:29:15.780 can trace it and you can realize that whatever's happening here is coming through dominantly.

00:29:20.720 Is that something, it's a premature cardiovascular disease? Is that a clinical term? Is there like

00:29:25.560 a cutoff when you call it premature? Yeah, I mean, I think loosely we would say someone who's having

00:29:30.300 a major adverse cardiac event before 60 would be premature. Of course, I have a different

00:29:35.440 definition of that. I would think a major adverse cardiac event before 80 is premature. But I think

00:29:41.660 someone who's having any major adverse cardiac event may before the age of 60, I think anybody would

00:29:48.120 consider that premature. So what I've never been able to figure out is, you know, that patient of mine

00:29:54.500 that had like an LDL, an LP little a of 650, family history is not outrageous. You know, when people

00:30:01.040 get heart disease, they get it in their seventies. The patient of mine who has the 500, I've tested

00:30:07.320 this patient's family. I know where it came from. I know which parent it came from. And the burden of

00:30:13.580 disease is modest. So there is something else going on here. And it's just like the case with,

00:30:18.060 we know that LDL-P alone is not the issue. We know that it's just one factor. And similarly,

00:30:24.860 not all LP little a's must be created equal. And so the question I'd want to get into with an expert

00:30:30.940 on this is, does it have to do with the lysine binding domains, the affinity for these oxidized

00:30:35.860 moieties? Is there some feature of one person's versus another's that lends to a more aggressive

00:30:42.840 oxidation within the subanethyl space or greater retention or something like that?

00:30:47.460 I think that gets into why do we have LP little a? So what would be the evolutionary basis? What's

00:30:53.580 the function of LP little a? Like it can't just be some hell particle that's just trying to kill us.

00:30:59.140 I mean, in theory it could because it kills us through basically three mechanisms that don't

00:31:07.240 tend to kill you young. So if you were taking a purely evolutionary standpoint, I think

00:31:12.140 sometimes bad things track. But it turns out that even like APOE4, which in today's environment

00:31:19.180 doesn't seem particularly protective, APOE4 was quite protective against parasitic infections

00:31:24.720 in the CNS. And hell, up until a few years ago, that would have been a pretty good thing to have.

00:31:30.120 Of course, now that we can live long enough, that upside isn't worth the downside of an increased

00:31:35.680 risk of Alzheimer's disease. So LP little a clearly does two things that are separate. And I think we

00:31:44.220 could argue, at least theoretically, that it would have provided a benefit evolutionary. The first is,

00:31:50.520 if you go back to what we talked about, you have this great homology to plasminogen. And plasminogen

00:31:56.440 being a clotting factor means that people with elevated LP little a tend to have what's called

00:32:01.840 hypercoagulability. So they have an ability to form blood clots better than someone who doesn't.

00:32:08.680 Now, in today's environment, that's not an advantage because most of us are not in an environment where

00:32:14.440 bleeding to death is a major concern. But you can imagine 50,000 years ago, bleeding to death would

00:32:22.840 actually be a significant concern. And so I think these people would have had a trauma advantage,

00:32:28.520 you know, with respect to, and I'm sure you could probably pose many benefits during childbirth.

00:32:34.600 You know, when I think about what I saw in the OBGYN rounds, how many times was a woman bleeding so

00:32:40.540 sufficiently that she required blood clotting products? It's not unheard of. And so you think

00:32:44.780 about the benefits this could have had all the way from birth, the birth of a child, right up until,

00:32:50.920 you know, getting scratched by an animal or whatever. The second benefit is more of a speculation,

00:32:56.260 I think, but it's probably that going back to those lysine binding domains, that if you're in a

00:33:01.420 relatively low oxidative environment and your LP little a's know where to go when they're done,

00:33:07.480 which is to the liver and where not to go, which is the coronary arteries in the aortic valve,

00:33:12.860 they're actually amazing scavengers. So I'm sure somebody out there has got better data on this

00:33:20.280 or has data period, because I'm obviously speculating, but you could make the case that being

00:33:25.200 able to have more particles that can scavenge more of these oxidized phospholibids and oxidized

00:33:30.360 moieties and take them back to the liver, which is the ultimate place of clearance for the LP little

00:33:34.960 a, which is a totally safe place to take these things that would pose an advantage. And it would

00:33:40.340 be the case today that maybe we're in a higher inflammatory environment and maybe we've gone too

00:33:46.520 far. In other words, maybe we're overwhelming the system's ability to clear it. And on top of that,

00:33:51.200 we may have other risk factors, hypertension, hyperinsulinemia, other drivers of inflammation

00:33:56.740 that are now giving these LP little a's another place to go, which is, yeah, you're ultimately

00:34:01.220 going to end up at the liver, but like 6% of you are going to get stuck in the subendothelial space

00:34:06.000 and wreak havoc. And on top of that, you're doing a way worse job than the LDLP because, you know,

00:34:11.200 the LDLP when it gets there is bad enough, but the LP little a is now dragging all that oxidized

00:34:15.400 crap in there with it. So the next question is what is the problem with elevated LP little a or

00:34:22.800 what are the problems with LP little a elevated? So basically they fit into sort of three categories.

00:34:28.840 The first being enhanced atherosclerosis. The second, I don't know which by magnitude would

00:34:35.140 pose a bigger threat, but probably aortic stenosis given the severity. And then the third being enhanced

00:34:40.420 venous thrombosis. So what do those things mean? So, so basically more atherosclerosis,

00:34:47.040 more aortic stenosis, I believe about two thirds of the cases of aortic stenosis are explained by

00:34:54.860 elevated LP little a. So you have four valves in the heart and one of them is called the aortic valve.

00:35:00.220 That's the valve that separates the left ventricle from the systemic system. So the proximal aorta.

00:35:07.360 So that valve is under more pressure than the other three valves by a long shot, because it's the one

00:35:14.040 that's directly in front of the most powerful chamber of the heart. That valve has three leaflets. It's a

00:35:20.900 tri-leaflet valve. And it seems that LP little a has a particular affinity for going there and inducing

00:35:31.860 bone forming proteins to create calcifications. And when that valve loses its suppleness and it

00:35:38.940 becomes calcified, you get basically a blockage of that valve called the stenosis. And so this condition

00:35:45.440 of aortic stenosis is very problematic. One of the earlier signs in the blood that somebody has aortic

00:35:53.520 stenosis would be signs of swelling or enlargement or dilation of the heart. And there are blood markers

00:36:00.740 like brain natriuretic peptide BNP or pro NT BNP that are actually used quite frequently in ERs to

00:36:09.500 assess patients very quickly for cardiomyopathy or cardiac failure. And so that's one of those things that

00:36:14.780 we like to look at. And, you know, if I see a patient with LP little a, I'm always screening them for aortic

00:36:20.500 stenosis out of the gate. I don't care if they're 30 years old. I mean, many of our patients are in their

00:36:25.700 30s and 40s, but if they have an elevated LP little a, we're doing echo at a minimum and preferably

00:36:31.900 cardiac MRI, which is much more accurate to both look at the morphology of the aortic valve and get

00:36:39.000 a very accurate gradient of pressure. And then sometimes you'll get patients. I have a patient

00:36:44.200 who has a bicuspid aortic valve, which is going to be by itself. That's predisposed to aortic

00:36:50.420 stenosis. And he also has a very elevated LP little a, about 250 or 300. So even though he's

00:36:56.880 only in his thirties, he gets a cardiac MRI annually and he's already showing a pressure gradient. So,

00:37:03.560 you know, I've explained to him that he is going to need an intervention at some point in his life,

00:37:08.560 but the good news is we're going to do it long before he experiences any strain on his heart muscle.

00:37:14.900 And the good news again for patients today is this stuff's going to be done interventionally and not

00:37:20.120 via open heart surgery as it once was. On the atherosclerosis side, I think the Mendelian

00:37:26.160 randomizations, the GWAS and the epidemiology all tell a very similar story. I suspect that it's both

00:37:35.720 its ability, it's probably an all of the above when it comes to why, meaning it's, are these particles

00:37:42.600 more likely to enter the subendothelial space? I don't know why that would be the case. Are they more

00:37:48.500 likely to be retained? Probably because they have that whole big Kringle oxidized moiety thing there.

00:37:55.740 Are they more likely to kick off an inflammatory response? Very likely because of what they're

00:37:59.960 dragging in with them. And then on top of that, to have the pro thrombotic component, I suspect

00:38:05.760 is what's driving the increase in the risk of atherosclerosis. But in truth, we don't have

00:38:14.840 definitive proof that LP little a is a more atherogenic particle. And you and I were talking

00:38:22.580 about this the other day that there was this paper that actually was looking at patients with post-MIs

00:38:28.640 and even suggesting that, well, everybody who has an MI has a rise in LP little a. And we'll probably

00:38:34.780 get to that later why we think that might be the case. But the question posed is, well, maybe LP little

00:38:40.580 a is the result of atherosclerosis and not the cause of it. I don't agree with that because

00:38:45.360 many post-MI patients don't have an elevated LP little a. And I think a better explanation for that

00:38:51.020 is that LP little a also acts as an acute phase reactant rising with inflammatory responses. But

00:38:57.880 probably not until the antisense oligonucleotide trials complete will we actually know the answer to

00:39:04.540 this question? Because really without a clinical trial, you can't actually infer cause and effect

00:39:11.620 the way we can with other aspects of atherosclerosis like the LDL particle or inflammation where we have

00:39:19.620 elegant prospective clinical trials that create a relationship between cause and effect.

00:39:26.420 The last thing that I guess I mentioned was the thromboembolism. So I used to have a practice of

00:39:32.680 putting everybody with an elevated APO a LP little a on a baby aspirin just to combat the effect.

00:39:38.860 It turns out that that was probably an oversimplified approach and that there's only a subset of people

00:39:44.440 for whom aspirin counteracts the effect. So unfortunately, this is still one of those things

00:39:49.820 where I don't think we have a great answer. I do take DVT prophylaxis. So deep vein thromboses

00:39:55.820 prophylaxis and prevention. I do take it more seriously in the LP little a patients.

00:40:00.940 And there are certain strategies you can take around flying. There's actually a commercially

00:40:05.100 available product called flight tabs, which you can buy on Amazon. I was, remember when we did the

00:40:10.280 research on this, I was blown away that you could buy these things on Amazon because they're actually

00:40:14.600 quite potent. But I do recommend that people with elevated LP little a, if they're on really long

00:40:20.760 flights. And again, I'm not recommending that for people who are listening because I can't,

00:40:25.120 but I certainly recommend to my patients, to a subset of them that we're particularly worried about,

00:40:28.980 that we look at either pharmacologic agents or even an OTC agent like that as a way to reduce

00:40:34.880 the risk of these types of events. So do we know how much elevated LP little a is associated

00:40:39.980 with these increased risks? If we're looking at the epidemiology, what are the associated risks

00:40:44.760 with cardiovascular disease? So with aortic stenosis, the hazard ratios are anywhere from two to four,

00:40:51.800 depending on the studies. And I think they probably median ends up being, you know, roughly two and a

00:40:58.600 half. With VTE, with the venous thromboembolism, I think the hazard ratio is about 3X. And again,

00:41:04.820 it's important to put this in perspective. You know, we've talked about absolute versus relative

00:41:09.160 risk. So when you talk about a 3X risk of something that occurs like 1% of the time,

00:41:16.760 that means you're going from a 1% absolute risk to a 3% absolute risk. So in other words,

00:41:20.840 it doesn't mean like if you're listening to this and you have an elevated LP little a,

00:41:23.700 you need to call an ambulance to drive you home because you're afraid you're going to have a

00:41:26.740 pulmonary embolism. And similarly, a hazard ratio of two and a half, three, even four on aortic

00:41:32.640 stenosis. As I said, it probably explains about two thirds of the total volume of aortic stenosis,

00:41:38.280 but it doesn't mean that every patient who's got this is going to get it. I think in the case of

00:41:43.320 that one patient of mine, his bicuspid valve is just a setup to make things worse because he's,

00:41:48.680 you know, he's now got a double whammy on that. And when it comes to atherosclerosis,

00:41:54.060 basically you see odds ratios of about two to four depending on the amount. So it's, uh, it's,

00:42:05.980 it looks like a pretty good dose response where it's sort of below about 30 or 40 milligrams per

00:42:14.300 deciliter. Cause unfortunately all of these studies are done with LP little a mass and not particle

00:42:19.620 number. And I can't really convert that, but we believe that that's probably about 50. That's

00:42:25.420 probably, you're going to get comparable in the 50 to 75 nanomole per liter is this sort of safe zone

00:42:31.420 where it's relatively flat. And then it starts to uptick pretty swiftly. So by the time you're at,

00:42:37.880 call it 200 milligrams per deciliter, you're at about a 60% increase. Now, if you stop for a moment

00:42:45.840 and think about that, what should you be more afraid of a three X hazard ratio for VTE or a 1.6

00:42:54.380 hazard ratio for atherosclerosis or a 2.5 X hazard ratio on aortic stenosis? This is like the advanced

00:43:03.060 clinical epidemiology question, right? I think the answer is the 1.6 on atherosclerosis is by far the

00:43:08.100 most disconcerting because atherosclerosis is infinitely more prevalent. So a 60% increase in risk

00:43:14.340 on something that is going to kill a third of people is a big effing problem. Whereas a 3%

00:43:20.640 risk on something that's going to ding 1% of people, yeah, we'll manage it. But that's not

00:43:26.800 what we stay up late thinking about. And even for that particular individual that has that risk

00:43:31.720 profile, that if they look at their absolute risks, that probably bumps up their absolute risk

00:43:36.320 the most with cardiovascular disease. Yeah. We're screening for aortic stenosis, not because I

00:43:41.980 necessarily think it's even at the population or societal level cost effective, but at the individual

00:43:47.400 level, we're not going to let that kind of stuff slide. But if you were to think about this at the

00:43:51.860 population level, the thing we have to be most concerned of is somewhere between one in five and

00:43:56.940 one in 10 people. And in some cultures, it's even higher in Southeast Asians, it's even higher

00:44:01.100 are walking around with these little time bombs. And to the point of Anahad story, I'm still

00:44:08.160 shocked at how many doctors don't understand this. Now, look, if you're a radiologist or a

00:44:13.000 dermatologist, that's okay. I don't think you need to know this. But if you sit anywhere on the

00:44:19.820 front lines of medicine, if you're a family physician, if you're a GYN even, because for

00:44:25.160 many women, their GYNs become their PCPs, their primary care physicians. If you are anywhere in the

00:44:30.700 crosshairs of taking care of a patient where you have some input into how they lower their risk of

00:44:35.420 cardiovascular disease, and you don't understand most of what we're talking about on this podcast,

00:44:40.520 I worry that you're missing an opportunity to help patients.

00:44:44.920 Okay. So another question that came in, I think you touched upon it very quickly,

00:44:49.660 is what is the prevalence of elevated LP little a? Probably what is elevated LP little a? How is

00:44:55.540 that determined?

00:44:56.440 Well, to my knowledge, everything that's done on this that's published is based on the LP little

00:45:01.060 a mass, not the particle number. But the US levels define normal as less than 30 milligrams per

00:45:09.600 deciliter. The European atherosclerotic society defines normal as less than 50 milligrams per

00:45:16.700 deciliter. And I believe both the UK and Germany consider anything over 60 sufficient for state

00:45:27.400 covered a phoresis. A phoresis is a type of treatment where a patient has a very large IV put

00:45:36.080 in one arm, typically about a 14 gauge, and blood is taken out, run through a machine that spins at a

00:45:44.440 certain frequency to generate a separation of the plasma. And you can basically fractionate the plasma

00:45:52.280 and identify something that you want to remove. So back when I was at NIH, I used to volunteer for

00:45:57.840 a phoresis every four weeks to donate lymphocytes. And then they basically put everything back that once

00:46:04.740 they strip out the piece they want, but you can actually do a phoresis and remove the APO little a. The

00:46:10.820 problem is the frequency with which you have to do it is staggering because the half-life of these

00:46:16.360 particles is, you know, a matter of days. So these patients would undergo a phoresis potentially twice

00:46:22.580 a week. So that's obviously a very difficult way to be tethered. So we got into it a little bit there.

00:46:29.560 How is abnormal LP little a treated or dealt with? So you just got into the a phoresis. Are there other

00:46:35.780 therapies currently available? So a phoresis is, that's something that we just really never resort to

00:46:41.900 or very rarely resort to. And certainly now, now that PCSK9 inhibitors are on the market, I think that

00:46:49.800 a phoresis is becoming probably less and less utilized. Historically, the agent for treatment has been

00:46:56.800 niacin. Now niacin's got kind of a checkered history because it's known to lower APOB. So you take niacin,

00:47:04.180 your LDL goes down. And this is a super contentious topic in lipid circles. But the question is,

00:47:11.900 does niacin save lives? And depending on how you look at the trial data, the answer is maybe or no.

00:47:18.320 It's like a wonder drug. Theoretically, right? It lowers LDL. HDL goes up. LP little a might go down.

00:47:25.200 Yep. So on paper, it looks, at least up to that point, it looks great.

00:47:29.280 Right. That's exactly right. It does three things that we historically know when they happen. Good

00:47:33.600 things should happen. LDL, particle, and cholesterol, APOB, all go down. HDL cholesterol goes

00:47:41.240 up. Although I would argue that that's not a good thing. I think we have a pretty good sense of why

00:47:46.700 raising HDL cholesterol inorganically, meaning pharmacologically is not going to be good.

00:47:53.100 And it lowers LP little a by probably a third. So...

00:47:57.020 CTEP inhibitors?

00:47:57.800 Yeah, exactly.

00:47:58.600 Fifth time's the charm or...

00:48:00.020 I think we're waiting for number five.

00:48:01.540 Yeah.

00:48:01.660 But it turns out that in the trial that basically doomed niacin, the trial probably wasn't designed

00:48:08.240 that well in that they were giving niacin to patients who were already on a max dose statin

00:48:13.280 and looking for the HDL increase to see if that was adding benefit. And so you basically get

00:48:20.700 lipidologists in two camps. And actually, it's not... It's quite evenly split, at least in my narrow

00:48:27.940 sampling of smart lipidologists, where you get some who say niacin should never be used. And then

00:48:33.540 you get others who say, look, it's probably not a great drug, but if you have a patient who can't

00:48:38.180 take anything else, it's still a good drug. And I know lots of lipidologists who are still putting

00:48:43.420 LP little a patients on niacin, even though there are no data to suggest that that will save their

00:48:50.220 lives. But I got to be honest with you, I'm not convinced that that's necessarily a bad thing.

00:48:54.220 I generally don't. I now move to the third thing, which is the PCSK9 inhibitor. But I guess before I

00:49:00.320 do that, I should explain statins, because everybody is probably saying, where do statins

00:49:04.240 fit into this? And it turns out statins don't clear LP little a, which is kind of counterintuitive

00:49:10.960 if you know how statins work. So statins work via two mechanisms, what we call sort of the direct

00:49:20.020 and indirect mechanism. So the direct mechanism is that they inhibit HMG-CoA reductase, which is an

00:49:29.080 enzyme that catalyzes one of the early steps, if not the first step, I believe, of cholesterol

00:49:34.140 synthesis. So if you're making less cholesterol, you would have less cholesterol. There would be

00:49:38.880 less cholesterol to carry around. You could require fewer lipoproteins. But that's not really

00:49:44.540 the main way it works. The main way it works is that the liver, in response to the statin,

00:49:53.220 upregulates something called SREBP2. And when that thing gets upregulated, it puts more LDL

00:50:04.000 receptors on the surface of the liver. So this SREBP2, which I'll just abbreviate for short,

00:50:09.420 is called the sterile regulatory element binding protein. It basically says, hey, the liver is

00:50:14.460 getting less cholesterol and it wants more cholesterol. So I'm going to put more of these

00:50:18.440 LDL receptors on my surface to pull more in. Now, I didn't know this until recently, but one of the

00:50:24.700 other things that SREBP2 does is it actually produces more PCSK9. Now PCSK9 is a protein that

00:50:33.700 degrades LDL receptors. So it's actually a bit of a check and a balance. So you have more LDL

00:50:41.380 clearance because of more LDL receptors, but you also speed up the rate at which those LDL receptors

00:50:47.560 are degraded. So the statin is causing these two indirect effects, but the net tends to be an enhanced

00:50:56.800 clearance of the LDL particle, the ApoB particle, and therefore a lowering of the LDL cholesterol.

00:51:03.700 But it doesn't lower LP little a. And if you're listening to this and you remember what we

00:51:08.340 talked about at the outset, you're probably thinking that doesn't make sense. LP little a

00:51:12.060 is just an LDL with an ApoA on it. Why wouldn't the LDL receptor clear it? Because if the LDL receptor

00:51:20.960 clears it, it should also go down. I asked Tom Dayspring about this because a really interesting

00:51:29.640 paper came out a few weeks ago that actually tried to explain this. And like all good papers,

00:51:35.560 it ended up leaving more questions than answers. The best explanation that I understood from Tom was

00:51:41.760 that LP little a will get cleared by LDL receptors eventually, but it's just the last in line.

00:51:48.020 So after the LDL is cleared and the VLDL is cleared, yeah, then you might get to the LP little a.

00:51:56.780 But the problem is you never get there. So maybe in theory, if you increase LDL receptor expression

00:52:04.320 enough, or if you could knock out PCSK9 and offset the second piece of what the statin is doing,

00:52:11.880 the statin would work. And it turns out that that is largely what this paper showed and what we've

00:52:18.600 always known, which is when you combine a PCSK inhibitor with a statin, you actually do get a

00:52:23.180 reduction of LP little a. Whereas the statin by itself is anywhere from no reduction to, in some

00:52:28.440 studies, an actual increase in LP little a. And PCSK9 alone also lowers LP little a. So to be clear,

00:52:37.860 PCSK9 inhibitors are not FDA approved for the use of lowering LP little a. But those of us who

00:52:46.400 prescribe these drugs, both for patients with other indications and with LP little a, generally

00:52:52.560 acknowledge that we're seeing about a 30% reduction in LP little a, sometimes as high as a 50% reduction

00:52:58.300 in LP little a when patients are taking PCSK9 inhibitors with or without statins. And that also

00:53:04.380 probably speaks to the fact that we know that LP little a is cleared by different receptors.

00:53:09.220 So its primary receptor is probably LRP2, but it's also probably cleared somewhat by VLDL receptors

00:53:17.320 and even something called SRB1. Although I'm not sure of that. And frankly, I don't know that anybody

00:53:22.860 is. So that what the PCSK9 inhibitor is doing is it's inhibiting PCSK9 and therefore inhibiting the

00:53:31.640 protein that degrades not just the LDL receptor, but these other receptors that clear LP little

00:53:36.580 a. Interesting. Sure. Yeah. I was just thinking about something on statins. So oftentimes I'll read

00:53:47.080 in the papers just to back up for a second too. I often read in the papers LP little a, the words

00:53:54.280 mysterious, mysterious, unknown, like in some ways we're in our infancy and understanding this.

00:54:01.240 But I think in one of those papers, Samikas looked at the effect of statins, not only just statins in

00:54:07.740 general, but different statins, atorvastatin, pravastatin, pativastatin, livolo, rosuvastatin,

00:54:15.260 and simvastatin. I think maybe that covers all the statins. If I'm just looking at his data,

00:54:20.080 the LP little a actually, it looks like it's trending up on statins. Not only that,

00:54:26.460 the oxidized phospholipids to ApoB are also going up. Is there any explanation as to why

00:54:32.140 the thing would actually be elevated? Yeah. Cause the ApoB is probably going down.

00:54:35.720 It's probably that you're lowering the denominator. Got it. That's my guess.

00:54:40.760 What's clearly acknowledged is that when you give a patient with elevated LP little a a statin,

00:54:46.820 which we do, absolutely. It's not to lower the LP little a, it's to lower the LDL. So actually,

00:54:53.080 I'm glad you brought this up because I didn't, I sort of missed the punchline in all the detail.

00:54:57.320 At least one of the punchlines is how do we treat patients with elevated LP little a?

00:55:03.700 Well, we're probably not going to give them phoresis if they can't afford to buy a PCSK9 inhibitor,

00:55:08.860 because it's certainly not going to be approved. You have only one other choice, which is to,

00:55:14.860 actually have two other choices, but I'll get to one in a moment. It's, it's, it rarely works,

00:55:19.600 but it works occasionally. But your, your real issue is you have to give them a statin because

00:55:23.840 you now have a new LDL target. So my LDL target, when I say LDL, I'm always referring to LDLP.

00:55:29.320 My LDLP target is the 20th percentile or lower for every patient, but how much lower you go than that

00:55:36.400 is a function of other risk factors. So are we talking about secondary prevention? What's the family

00:55:42.020 history? Are they insulin resistant? You know, all these other factors, but a patient who's at,

00:55:47.180 got an elevated LP little a immediately falls into the category of all things equal. They're at the

00:55:51.860 10th percentile or lower. And so you will often need a statin to get them there. Now, not always.

00:55:56.720 I have some patients who don't need a statin to get their LP little, their LDLP down to the 10th

00:56:02.880 percentile, but they're the exception and not the rule. So that's where the, I don't want to,

00:56:06.840 I don't want people to get the impression that if you have an elevated LP little a,

00:56:09.200 you shouldn't be taking a statin. Now it's quite the opposite. You probably should be taking a

00:56:12.680 statin, but just understand that the statin is there to control APOB and not LP little a.

00:56:17.780 Okay. And I don't know if we have enough ammo to cover this, but hormone therapy, estrogen,

00:56:24.220 I think has been shown to lower LP little a. I didn't know that actually.

00:56:27.640 This is an up to date, which is a nice service that compiles a lot of this information,

00:56:33.780 almost like a review, systematic review. And they have a section on lipoprotein,

00:56:39.200 little a and cardiovascular disease and lipid lowering. And one of the things that they noted

00:56:44.720 was estrogen replacement therapy reduces LP little a levels by up to 50%. And they have a

00:56:50.480 couple of references there in effect that was somewhat mitigated by concomitant progesterone

00:56:55.580 therapy in some reports. I don't know if that's a women's health initiative. So we're probably

00:56:59.500 dealing with different variables, but not the peppy trial. However, the clinical role for hormone

00:57:05.100 replacement therapy is uncertain and it is not recommended for cardiovascular disease risk

00:57:10.880 reduction. So if that HRT topic wasn't compelling enough to go over, I think this is another reason,

00:57:15.680 another just wrinkle to throw in there.

00:57:17.240 Yeah. I'd like to understand that better. That strikes me as a bit too good to be true,

00:57:20.620 frankly, because certainly there were, I mean, if that's true, that's one, it suggests it might

00:57:24.860 only be. So I guess the question I would want to know is, does that imply that women who go 0.98

00:57:28.600 through menopause, wouldn't they see an increase in LP little a, all things equal, if they did not

00:57:33.880 receive HRT? I believe so. Yeah. I'm going to go and look at the LP little a levels of my patients

00:57:39.300 who have gone through menopause while under my care. But nothing jumps out at me. There was one

00:57:45.240 other thing we didn't talk about, which is what's on the front lines here in terms of really interesting

00:57:50.300 stuff, which is these things called ASOs, which is really the first treatment that is designed

00:57:54.920 specifically to lower LP little a. So the ASO stands for antisense oligonucleotide. So

00:58:01.400 these are molecules that disrupt protein synthesis. So I can't remember exactly where they ask. I think

00:58:10.300 they act after the messenger RNA, between messenger RNA and translational RNA, but maybe they act between

00:58:18.360 DNA and messenger RNA. I should know this. I'm sure there's like, I'm sure that's a very well-known

00:58:22.500 obvious fact that I'm just forgetting. But the point is they disrupt the synthesis of ApoA, which is occurring

00:58:28.620 in the liver. So this is a drug that goes right to the heart of LP little a. And I didn't say this

00:58:35.720 earlier, but it's worth pointing this out. When you go through my whole rigmarole on why do statins

00:58:43.240 probably not decrease LP little a, it doesn't appear that anything that's going to lower LP little a is going

00:58:50.640 to do it on the catabolism side, meaning the breakdown side. It appears to be on the synthesis

00:58:56.240 side, the making side. And so while the monoclonal antibodies like the PCS canines also increase

00:59:03.800 degradation, they reduce the synthesis. They're actually reducing the synthesis of Apo little a

00:59:11.040 and therefore you just have your garden variety LDLs. These drugs have been shown to have safety

00:59:27.840 and efficacy. So they have concluded phase one and phase two trials and they are slowly enrolling in

00:59:34.360 phase three trials. I think three years ago, I said they'd be done in five years. Three years later,

00:59:40.760 I think they'll be done in five years. Consistent. Yeah. The frequency distribution figure that we'll

00:59:48.120 include somewhere, it shows effective antisense oligonucleotide and it says around 70% to up to

00:59:54.480 99%. So it could potentially wipe out, virtually wipe out LP little a. Yeah, no, even somebody who's 1.00

01:00:00.640 got an LP little a of 200 can be normalized. I'm a little leery of wiping out something entirely.

01:00:06.200 It certainly suggests that if you have this ASO, you can test a hypothesis in terms of

01:00:13.560 LP little a lowering therapy for sure. Well, that's what I was referring to at the outset,

01:00:16.960 which was until this trial is done, I don't think we can definitively know the answer of

01:00:22.560 what is the true risk? How do you quantify the true risk of LP little a? I think we got through a

01:00:28.640 lot of the major questions. This is awesome. We like didn't have to go for four hours.

01:00:33.500 There's a bonus question. We're in the bonus round. Oh, there's some other stuff we can talk

01:00:40.460 about too as well. I think getting into the oxidized phospholipids, how that works in the

01:00:45.040 LP, PLA too, we could get into. But one of the things that I was thinking about is that

01:00:50.240 with lipoproteins, with LDL, with HDL, even triglycerides, like you have some tools in your

01:00:57.380 arsenal just in terms of, let's call them behavioral modifications or things like that. If you

01:01:01.240 challenge somebody or somebody said, I need to lower my triglycerides in like in 30 days or else,

01:01:07.220 you could probably do that through diet. Absolutely. I mean, triglycerides by far

01:01:11.460 the most sensitive thing in the blood as far as lipoprotein lipid related molecules to dietary

01:01:16.780 intervention. Yeah. In theory, it sounds like you can play around with a lot of the lipoproteins,

01:01:20.700 actually a lot of the markers, biomarkers, but it seems like with LP little a, it doesn't seem like

01:01:25.760 it can be modified all that much by lifestyle. Is that right? Or at least that's the current thinking.

01:01:29.940 No, that's absolutely correct. And probably the reason for that is, as we just learned from the

01:01:35.080 PCSK9 statin comparisons, directionally speaking, there are two things that are driving LP little a,

01:01:42.540 how much you make and how much you clear. But the game seems to be won and lost on the how much you

01:01:48.380 make front. The how much you clear seems to be a second order thing. Now, when you look at LDLP,

01:01:55.440 just to contrast it, nevermind triglycerides, when you look at LDLP, you go back to four things

01:02:02.820 that determine the number of those particles. Three of them have to do with how much you carry.

01:02:09.400 One of them has to do with how much you clear. So three about the cargo, one about the port.

01:02:15.500 How many triglycerides do you have? How much cholesterol do you synthesize? How much

01:02:19.840 esterified cholesterol or non-esterified cholesterol rather do you reabsorb in the,

01:02:26.040 after it passes through the biliary system and the enterocyte? And then what's your LDL receptor

01:02:30.340 profile look like primarily in the liver, but also in the gut. Now we just established, you can clearly

01:02:35.900 lower triglycerides through nutrition. So you got somebody walking around with a triglyceride of 200

01:02:40.320 and an LDLP of 1600 and you do nothing but lower their triglycerides to 50. While I can't predict what

01:02:48.860 their reduction is going to be, it's likely going to go down. And so that's a lifestyle intervention

01:02:53.240 and that clearly does things. And it turns out that we know that diet is also going to lower or

01:02:58.880 raise. Certainly it has an impact on LDLC that is known, but it also can have an effect on LDLP

01:03:06.340 through cholesterol synthesis and absorption. Now I think that that, I think that the relationship

01:03:12.680 there is much less clearly understood. I've speculated about what, what I see occurring.

01:03:19.660 There seems to be a subset of people who, when they consume high amounts of saturated fat,

01:03:24.380 see a really significant increase in cholesterol synthesis. I think Tom Dayspring has written a really

01:03:30.940 eloquent piece on this. So we, if we can find it, if it's publicly available, we should link to it

01:03:36.180 because I think it's a great piece on the hypothesis around why certain people in the

01:03:40.980 presence of high saturated fat just start making much more cholesterol. And then of course the

01:03:46.580 contentious topic is, does it matter? I don't think we know the answer to that question, but that's a

01:03:51.420 point. I was among them at one point. I think it's, you get an NMR and it gives you your LDLP count.

01:03:58.640 Is it in nanomoles per liter? And it actually reminded me of, uh, Fletch and Gillette collecting

01:04:05.440 rent, I believe. And, uh, he picks up one of Gillette's letters and he says, Oh, a letter from

01:04:11.000 the Oakwood potency clinic. We're sorry to inform you. We can't process sperm counts as low as yours.

01:04:16.140 So in the, in the case of this NMR, I get the test back and you, you probably know the number,

01:04:22.600 maybe it's like 2,500 or the upper cutoff is 3,500, 3,500. And it has a, one of those awesome

01:04:29.420 greater than signs. So it's greater than 3,500. It's like, we're sorry. Our machines can't process

01:04:34.240 LDL particles as high as you. And I think during the time I was doing an experiment where I was

01:04:38.860 eating a lot of my calories were coming from saturated fat. It was probably a supposedly a

01:04:43.800 well-formulated ketogenic diet, but coconut oil, maybe some of it was coconut oil, butter,

01:04:47.880 et cetera. But it was heavily loaded with saturated fat, but I would love to read that article. And

01:04:54.300 because it's one of those things that's gone around the circles, is it good? Is it bad?

01:04:57.820 But it is definitely something that's, that's seen, I think.

01:05:01.180 Yeah. I mean, given how amazing we've made progress on this and how we've barely been at it for an hour

01:05:07.200 and we're almost done. I mean, I'm happy to expand on this just based on my observations,

01:05:11.160 because I'm sure someone's going to end up asking anyway. I've probably seen this now a dozen times

01:05:15.540 where either someone comes to me already on a ketogenic diet, or we put them on a ketogenic diet

01:05:20.360 and they develop this, this change in their lipids. Now, there are some people who will argue

01:05:27.340 that it's transient and it's going to go away and you know, a year or two years or whatever,

01:05:31.960 maybe so. There are others that argue that it's irrelevant, that the increase in the cholesterol

01:05:38.820 synthesis and the LDL cholesterol and the total cholesterol is actually a good thing. And there's

01:05:43.060 some reason that they offer for that, that I don't quite buy or understand. But my view is all

01:05:48.740 things equal until I know better, I'm going to assume that high LDL is probably problematic.

01:05:55.040 And more importantly, the point is, are there ways to reverse the diet and reverse the condition and

01:06:02.100 figure out what was the component within the diet that was doing it? Was it the total fat? Was this the

01:06:06.180 subset of the fat, et cetera? And in God, all but one of those cases of maybe a dozen,

01:06:11.060 when you just replace the saturated fat with monounsaturated fat, even if they stay consuming

01:06:17.400 a very high fat diet, the problem goes away, which has not that that's proof of anything,

01:06:23.800 but that really suggests to me that in those patients, they're getting more saturated fat

01:06:30.380 than they can process. Because even, you know, I had one, the first patient that I ever went through

01:06:33.920 this with, my first thought was, dude, we got to take you off this ketogenic diet, man. Like I'm not,

01:06:39.200 we can play keto camp all day long, but I'm not that comfortable with these numbers.

01:06:43.960 And he was like, but, you know, I'm not going off a ketogenic diet. Like, you know,

01:06:47.980 and he had all his reasons for why he, you know, felt better and performed better and all those

01:06:51.660 things. So I said, okay, well then we could keep you on a ketogenic diet, but we got to take,

01:06:56.180 I want to see what happens if your saturated fat goes from 75 grams a day to 25 grams a day.

01:07:01.620 And to do that, you're going to get really familiar and friendly with olives, olive oil

01:07:06.760 and macadamia nuts. And he's like, I don't care. You know, this kid, he was a young guy and he was,

01:07:11.540 uh, he'd do anything. He was kind of like a robot. And so sure enough, in like eight weeks of that

01:07:17.160 change, his LDLP went from greater than 3,500 to, you know, 1200. Same thing. I had a lot of

01:07:24.780 guacamole, macadamia nuts that replaced the saturated fat and the numbers came down and

01:07:30.020 everything else, give or take HDL triglycerides, all that stuff were sort of in the same ballpark

01:07:36.960 as before, but that LDLP came down. Yeah. And I got to tell you, I mean, I'm sure that this will

01:07:42.640 kick up a storm of people with, you know, very, very strong religious like views on,

01:07:48.620 oh, there's nothing wrong with an LDLP of 3,500. And, you know, again, I, I don't buy it because

01:07:57.080 the other thing I don't buy is a lot of those times you'll see the oxidized LDL go up as well.

01:08:01.720 And, and, and, um, how are we in the middle of Manhattan and some knucklehead drag racing on 79th?

01:08:09.700 I don't get that. It's the most gratuitous noisy city. I think we should, we should get involved.

01:08:16.780 Rev up the engines. He likes cars. We just need some jackhammers right now. So yeah,

01:08:25.460 when I see the oxidized LDL and CRP go up as well, which I often see with that, then I think,

01:08:30.520 you know, there's something else going on here. This isn't just a cholesterol synthesis problem.

01:08:35.820 It's an inflammatory problem. Something is, and look, I wasn't that guy. I mean, I probably ate

01:08:41.100 when I was in ketosis, I was probably eating 200 grams a day of saturated fat.

01:08:45.380 Maybe not quite that much, maybe 150, but I was eating a lot of saturated fat,

01:08:49.640 but I didn't have any of those response. You know, my CRP was really low. My trigs were

01:08:54.360 non-existent. My LDL particle number was probably around the 50th percentile, you know, 12 to 1300

01:09:00.480 nanomole per liter. Like I just didn't have any of those findings. And I, and again, I see a lot of

01:09:05.020 people who don't have those things. So I don't know why some people have these paradoxical reactions,

01:09:10.060 but I also don't think it's safe to ignore them just because insulin levels have gone down.

01:09:15.640 And going back to oxidized LDL, if you saw oxidized LDL going up, my newbie understanding

01:09:21.700 of this is that oxidized LDL is in a sense LP little a so that LP little a, well, that's a,

01:09:29.880 that's oxidized phospholipid. Yeah. So the LP little a picks up the oxidized phospholipids

01:09:34.720 from the lipoprotein from the LDL. Yep. And then that LP little a particle itself is now

01:09:40.400 carrying the oxidized phospholipids, but that's not a oxidized LDL. No, the ox LDL assay is different

01:09:47.180 from the ox PL assay. The ox LDL assay works independent of how many APOAs you have. Okay.

01:09:54.020 I like to see that number below 40. Again, I think the lab likes to see it below 60, but I like to see

01:10:00.960 that along with the LP PLA two, which you alluded to earlier, these are really local markers of

01:10:06.540 inflammation. And those are important because if you see a patient with an elevated C reactive protein,

01:10:11.580 should you be concerned about it? I mean, yes, probably. But the question is,

01:10:17.200 is it cardiac specific or not? You can't really tell. So that's why looking at fibrinogen and C

01:10:23.740 reactive protein and homocysteine and LP PLA two and oxidized LDL help you get a better picture of if

01:10:30.380 there's inflammation, how much of this do we think is going on locally at a vascular level,

01:10:35.240 you know, versus someplace else? This is, you know, you see this all the time in people who have

01:10:39.340 food insensitivities and things like that with respect to the fibrinogen, the CRP.

01:10:43.080 Yeah. And so we'll probably throw in those figures. He just mentioned that the

01:10:46.920 LP PLA two and LP little a, and then ox, then there's another thing called it's the oxidized

01:10:54.460 phospholipids over the APO B. And in that paper, it's a 2007 paper. And I think Samikas is the last

01:11:02.220 author on it as well. He's all over the place. They show the hazard ratios and it's a J curve

01:11:08.180 so that the very, the lowest, they call it the sextile. So they have, they partition it into six

01:11:13.560 different groups. And on the lowest, if you look at the hazard ratio, the hazard ratio is about two.

01:11:19.760 So the risk doubles if you have very low LP little a. Does he explain why he thinks that's

01:11:25.740 happening? I'm not. I wonder if it's an artifact of APO B being higher. Possibly. The denominator

01:11:31.020 going up would shrink the total number. I'll have to look at that. Are there any other LP little a

01:11:36.240 questions that came through the interwebs? Not through the interwebs. Well, then I think we can

01:11:41.220 bring to a close our inaugural chapter one, chapter one, vote on what you want to hear about.

01:11:48.600 Any final words, Bob? It's interesting. I knew about LP little a a little bit prior to Anahad's

01:11:54.800 article in January. And after doing some digging, there's, there's some other thoughts about this

01:12:00.740 stuff that I'm sure we'll get into down the road, but it's that proverb, I think Nassim Taleb

01:12:06.480 quotes. He says, he says, it's a Venetian proverb. He says, the further from the shore,

01:12:11.400 the deeper the water. And so the more you dig into this, the more you learn, the less, you know,

01:12:16.160 in a sense, it's that you've sort of exposed yourself to a lot of unknowns. So it's absolutely

01:12:22.220 fascinating. And I think it, it also gets to how most physicians don't even know about this stuff.

01:12:27.760 And you alluded to it in one of our conversations previously, that there's this lag, you know,

01:12:33.220 in terms of the medical knowledge and what's the accepted wisdom and the guidelines and things like

01:12:37.660 that. So I think LP little a is one of those cases. That's just, it's fascinating. And it's the more

01:12:44.240 you learn, the less, you know, but the more you want to learn. Yeah. And we're really, as you pointed

01:12:48.740 out earlier in our infancy of this thing, yeah, if we were just going to put numbers to it, I think

01:12:53.260 five years ago I had 50% understanding, like one unit of understanding to two units of perceived total

01:13:00.280 volume of content. Today I'm at 10% understanding, 10 units of understanding to 100 units of perceived

01:13:10.500 total content. So has my knowledge gone up in five years? Yeah, it's gone up 10 fold. The problem is

01:13:17.960 my appreciation for how much information is out there on this topic has gone up 50 fold. So my relative

01:13:22.720 insight has actually gone down five fold. What is that? It sounds like the Dunning Kruger effect a little

01:13:28.060 bit. It's that when you know, like just like the surface level, that's when you're the most

01:13:32.100 confident. You think you know everything. And then as you learn more, it's like that Dunning Kruger,

01:13:35.960 it's like a you and then your confidence and your knowledge goes down. Hey, welcome to the 24 hour

01:13:40.860 news cycle, cable TV and Twitter, man. I don't know if that's Dunning or Kruger, but it's on the left

01:13:47.680 side where it's everybody's very confident. Well, in summary, I'd say the following. If you're listening

01:13:52.040 to this as a patient, you should demand that your LP little a be known. It's non-negotiable,

01:13:56.100 especially if you have a family history of atherosclerotic disease. If you're a physician

01:14:01.420 and this is your first exposure to it, I hope that we've invited you to learn more and I hope

01:14:06.900 that we've provided you with enough information that you're sufficiently curious and we'll certainly

01:14:12.100 make a point to link to this, some of the, what we think are more relevant things. Worth noting,

01:14:17.340 I think about three days ago, an ICD-10 code was actually just issued for elevated LP little

01:14:22.880 a, that's a pretty big deal. That's like one of the signs that it's not some little nerds only

01:14:28.660 thing. Once you, once you get your ICD-9 code issued or ICD-10 rather. And if you are neither a patient

01:14:35.660 nor a physician, I don't know what you are. And therefore this podcast probably is not for you.

01:14:41.300 You can find all of this information and more at peteratiamd.com forward slash podcast. There

01:14:49.100 you'll find the show notes, readings, and links related to this episode. You can also find my blog

01:14:54.360 and the nerd safari at peteratiamd.com. What's a nerd safari you ask? Just click on the link at the

01:15:00.160 top of the site to learn more. Maybe the simplest thing to do is to sign up for my subjectively

01:15:04.380 non-lame once a week email, where I'll update you on what I've been up to, the most interesting papers

01:15:09.260 I've read and all things related to longevity, science, performance, sleep, et cetera. On social,

01:15:15.160 you can find me on Twitter, Instagram, and Facebook, all with the ID peteratiamd. But usually

01:15:20.400 Twitter is the best way to reach me to share your questions and comments. Now for the obligatory

01:15:24.720 disclaimer, this podcast is for general informational purposes only and does not constitute the practice

01:15:29.420 of medicine, nursing, or other professional healthcare services, including the giving of medical

01:15:34.640 advice. And note, no doctor patient relationship is formed. The use of this information and the

01:15:40.520 materials linked to the podcast is at the user's own risk. The content of this podcast is not intended

01:15:45.500 to be a substitute for professional medical advice, diagnoses, or treatment. Users should not

01:15:50.600 disregard or delay in obtaining medical advice for any medical condition they have and should seek

01:15:55.300 the assistance of their healthcare professionals for any such conditions. Lastly, and perhaps most

01:16:00.840 importantly, I take conflicts of interest very seriously for all of my disclosures, the companies

01:16:05.640 I invest in and or advise, please visit peteratiamd.com forward slash about.

The Peter Attia Drive - July 30, 2018

#07 - Deep Dive： Lp(a) — what every doctor, and the 10-20% of the population at risk, needs to know

Episode Stats

Summary

Transcript