Cardiovascular Clinical Conversations: Understanding the Function of HDL—Beyond Its Levels

Thanks, David, and thanks, everybody for being here. We’re excited about the HDL effects tests that we’re bringing out and the importance of HDL in managing patients atherosclerotic risk. Uh, I just wanted to give an overview of why we worked almost a decade in developing this test and why we think it’s important. They all remember the Cholesterol ester transfer protein inhibitors studies. The class of medications were designed to raise HDL as and in the illuminate trial to the left here actually increased HDL levels by 72 plus percent. Yet the increase in HDL with Torcetrapib actually led to an increase in death from all cause mortality, with a hazard ratio of about 1.58 and increased major cardiovascular events by about 25%. So clearly, just having a high HDL is insufficient. And what we recognize is it’s the function of HDL that has a role in atherosclerotic risk. The graph on the right is one you’ve probably seen from multiple trials showing that the average LDL correlates with the change in percent ATHEROMA volume. And in the illustrate trial, you’ll see that the use of placebo, which is actually atorvastatin followed that line well. But the additional torcetrapib to atorvastatin actually lowered LDL more but did not lower means after the volume suggesting that the dysfunctional HDL created by CETP inhibition is actually has an adverse effect on atherosclerosis. And again, it’s these types of data, including the data on the next slide that led us to want to develop a test around cholesterol efflux capacit of HDL. And this is work by Dan Rader and his group that has really shown over the years that enhanced Efflux capacity is actually a protective mechanism for atherosclerosis. So it’s really with the idea that less cholesterol efflux capacity is a hazard, batter efflux capacity is actually a protective mechanism. And the fact that we really have not had the ability, clinically to measure efflux capacity that led us to design and develop this test again over the last decade. On the next slide, it’s really my pleasure to introduce Dr. James Januzzi, who’s the Hutter family Professor of Medicine at Harvard and in the Division of Cardiology at Mass General in the Baim Institute for Clinical Research. Jim is an internationally recognized leader in the biomarker development space in heart failure, atherosclerosis and other cardiac disease states. He’s been a great help to Cleveland HeartLab as we’ve developed HDL effects and had the foresight to develop the Casablanca biobank, of which HDL effects was used to validate HDL effects and will be used and has been used to validate many other biomarkers going forward. It’s been a real honor and a privilege to work with Jim, and it’s my pleasure to introduce him. And Jim, thanks for being here today. Yeah, thanks very much, Dr. Penn. And I want to thank Quest Diagnostics and Cleveland Heart Lab for the opportunity not only to collaborate together on this important project, but also for the opportunity to speak today about the evolving understanding of how HDL might help supplement pretty substantially our understanding of Cardiovascular disease risk. There’s been a lot of change in the understanding about how we might think about HDL. I’ll remind you that HDL is a life of protein particle that’s involved with cholesterol transport, cholesterol transported to and from the liver by multiple carrier lipoproteins. And much of the focus of clinicians has been on LDL or APOB containing lipoproteins and HDL, which is a lipoprotein particle excuse me, that consists of multiple apolipoprotein particle particles, as we’ll talk about. And the role of HDL and how it is has come to be viewed as the, quote, good cholesterol, something that we really try to avoid using that terminology at this point has changed over time as our understanding about the biology of HDL and its role has evolved. Currently, HDL is thought to be protective against atherosclerotic risk, in part because functional HDL, and I’ll explain what I mean by that in a second, is responsible for reverse cholesterol transport. So it’s thought that HDL one function incorrectly and well is responsible for efflux of lipid of cholesterol from the periphery back to the liver where it is excreted. So this reverse cholesterol transport may be tracked as a function of what we refer to as efflux potential. And indeed, when you look at HDL cholesterol as a risk factor, this is how we sort of got to the concept of HDL being the so-called, quote, good cholesterol, end quote, which is to say that early on the belief was that higher HDL cholesterol alone would reduce the risk for cardiovascular disease. And these data demonstrate how a high HDL modifies the risk associated with a high LDL cholesterol and vice versa. So the higher the HDL in general, we see a lower risk. But but as I said, there’s more to the story. And indeed, when you look at patients that are admitted to the hospital for an acute cardiovascular event, a substantial percentage of them have HDL cholesterols that are normal, if not frankly, elevated. So how do we reconcile this finding? I mean, how do we understand HDL as a good cholesterol particle when, in fact, nearly half of the patients with acute MI, for example, have a normal to elevated HDL cholesterol? And this gets down to what Dr. Penn had already referred to, which is focusing more on function rather than form, rather than the absolute HDL concentration thinking about how HDL might work and how that function of HDL to reverse cholesterol transport or efflux might be associated with benefit. So indeed, when you look at the function of HDL, as opposed to the absolute concentration, we see a very important story, which is that HDL function, the ability to efflux cholesterol from the periphery back to the liver is actually quite substantially associated with the benefits of HDL. So the more functional HDL is associated with more cholesterol. Efflux, less inflammation and less thrombosis in a pro atherogenic state as well, in the face of treatment with cholesterol ester transfer protein inhibitor, the HDL that circulates in highest abundance is so-called dysfunctional HDL. So it’s not that the raising of HDL in the CTEP inhibitor studies itself was associated with risk. It’s that the form of HDL that was raised in this case, so-called dysfunctional HDL, which is not associated with cholesterol efflux and is in fact associated with a higher atheroma thrombotic and inflammatory risk was the particle that was increased with cetp inhibition. And so whether pharmacologically increased or just indigenously circulating dysfunctional HDL with poor efflux capacity is associated with a deleterious outcome. Now, as Dr. Penn mentioned, Dan Rader has developed a means by which efflux capacity might be assessed and I’m not going to go through the details of the methods used in order to do this cell based assay for cholesterol efflux capacity. Suffice to say that it requires many days before the results are available because it’s a cell culture based approach where macrophages are primed with cholesterol than ApoB depleted serum is then incubated together with the cells, and HDL is then added and through a method by which efflux can be assessed using scintillation counting efflux capacity may be estimated. And indeed, when you look at the results associated with cholesterol efflux capacity, you begin to realize that yes, indeed, while I’m sorry, HDL cholesterol on this force plot, you can see is on the correct side of the force plot with a hazard ratio for coronary artery disease of 0.85. It’s not statistically significant, whereas Efflux capacity is indeed significant. So how to reconcile that? Well, look, I mean, if a certain percentage of HDL has good efflux capacity, of course, a higher HDL cholesterol is going to be associated with benefit. It’s just that if you had a choice between measuring an indirect predictor of efflux capacity, specifically HDL versus estimating efflux capacity, either measuring it directly or through an indirect means, that would be expected to be a more powerful and more accurate way of predicting risk. And indeed, that is how we sort of got to the path that we got on with respect to assessing HDL functionality through the HDL effects panel. And I give a lot of credit to Dr. Penn and his team at the Cleveland Heart Lab for all of the years that they spent focusing on this concept and then approaching us with the opportunity to collaborate, to function, to evaluate and validate their findings. So let’s talk a bit about HDL effects. So Cleveland Heart Lab developed this methodology for assessing HDL function. So this is an estimate of Efflux capacity leveraging a lipidomics approach. So they looked at a broad range of HDL associated proteins. I think Dr. Penn can can correct me if I’m wrong during the question and answer session, but they looked at something like 34 proteins associated with ApoA-1 as well in HDL and from those various proteins that were identified, a core group of ApoA-1 associated lipoprotein particles or AALP were identified, and these five proteins were then quantified and reported and through the algorithm of the the so-called pCAD score, which is what the HDL effects test is assessing. One can then have an indirect assessment of cholesterol efflux capacity, and then, as I’ll show you, predict risk for obstructive coronary disease and risk for incident, cardiovascular events. So again, this is sort of the magic slide of how things happen. But the HDL effects test isolates and measures. ApoA-1 Associated lipoprotein particles from serum. The serum is incubated with ApoA-1 which attaches to the HDL particles, and the ApoA-1 has a tag that will help collecting with with collecting the HDL particles, and that HDL is then captured using affinity purification and quantified. So using LCMS. So this is a way to more rapidly assess efflux capacity without the many days and cell culture based approaches that are needed. And that’s how we got to the analysis that we performed in the Casablanca study. So Dr. Penn and his team approached us at the Massachusetts General Hospital Heart Center with the hypothesis that this HDL effects approach would be able to, through the fact that it estimates cholesterol efflux capacity, would be able to predict the presence and severity of obstructive coronary disease as well as predict incident, cardiovascular events. And the Casablanca study is very well suited for looking at both questions. So Casablanca, very briefly, is a study of over 1200 patients seen in our coronary angio Suite cardiac catheterization laboratory at the time of either coronary or peripheral angiography. It’s a mixture of patients with acute M.I, as well as other acute and chronic diagnoses. After patients were consented, we collected 15 milliliters of blood samples before and after Cath and these samples were rapidly processed and frozen. And then we assessed and follow up cardiovascular events during a mean of four years of follow up. Importantly, at the time that we enrolled patients, a coronary angiogram was performed, of course, and we recorded the presence and severity of coronary artery disease during the follow up period of time. After we enrolled patients, we adjudicated a wide range of cardiovascular events, including mortality, heart failure events and myocardial infarction. So a broad range of major adverse cardiovascular events were sought. And these data are reported in the Journal of the American College of Cardiology, as I’ll show you right now. So what we found was, first of all, after we narrowed down our study population to those without a prevalent myocardial infarction, what did people will tell you, you don’t want to be measuring lipids and lipid function in the setting of a recent within a few days acute MI because of the acute phase reaction that occurs during myocardial infarction, which may affect circulating lipid presence and function. So after excluding patients with incident MI, we had 943 participants to study. And then we looked at the the the pCAD score, the HDL effect score relative to obstructive disease and outcomes. So this is the first result. This is what’s called a corellagram, which gives us a sense of the correlation between the HDL effects panel, otherwise known as pCAD that if you want to look all the way on the far right of the of the corellagram, the second column from the right is the pCAD score. And what you see is that there are moderate correlations to the variables used for its development. So it’s not surprising that there’s a moderate correlation between pCAD and ApoA-1, HDL, HDL cholesterol, Apo C. But generally speaking, the HDL effects panel had little correlation to other lipoprotein particles or notably C-reactive protein. So it was not correlated with the presence or significance of inflammation in these subjects. And here are the beginnings of the results showing how across quartiles of score the ApoA-1 and the pCAD scores were higher among those with obstructive coronary artery disease compared to those who did not have obstructive CAD. I think more fundamentally this is a better way to look at it, which is asking the question what’s the likelihood for obstructive coronary disease on the basis of the pCAD score? And what these data show is that above the median, above the second quartile of the HDL effects score, there was a doubling a 200% increase in the likelihood for obstructive coronary artery disease. And I’ll remind you that the blood was drawn before the angiogram. So essentially what this is saying is, at the time of blood draw, we did not know that that these patients had obstructive coronary disease. And now the assay is telling us what we were soon to see during the coronary angiogram performed after the blood was drawn. So this is interesting. But of course, people will want to know how does the HDL effects panel do relative to other variables that we use to judge risk? And this is a graph showing the various C statistic. This is the area under the predictive curve, if you will, sensitivity and specificity of multiple variables, individual or collective, for predicting obstructive coronary artery disease. And what you’ll notice is that compared to other classical variables such as sex or, you know, presence of diabetes, older age or arterial hypertension, even APO B as you see all the way on the left here, the HDL effects panel had the highest C statistic for predicting the presence of obstructive coronary artery disease compared to other canonical variables that we use every day at the bedside or in the lab to predict the risk for obstructive disease. So the score, the pCAD score, otherwise known as HDL effects, is correlated with efflux capacity. And and that’s something that was shown by the Cleveland Heart Lab folks in preclinical analysis and in our data, as well as in other studies. It predicts cardiovascular disease. In the Casablanca study, the pCAD score was associated with the presence of obstructive coronary disease as well. In Casablanca, it predicted the risk for cardiovascular death during the four year follow up in another study from a registry of women actually, the pCAD score was able to identify women in the in the study who were likely to experience an incident myocardial infarction and relatively short term. So that that’s really interesting as well. We found no heterogeneity by sex in the Casablanca study. And in this analysis, the separate analysis of the the HDL effects panel had excellent predictive ability for prognosticating incident MI in women in that specific study. Okay. So, you know, I’m a clinician and I want to think like a clinician. We’ve got a tool that predicts risk. It follows what we would hope an efflux capacity assay would tell us, which is it predicts both the presence of disease but also the likelihood for progression of disease. So what are we going to do about it? And of course, there’s some challenges in this because HDL is one of those particles that we have some tools for managing, but not not the same kind of armamentarium we now have for LDL management. So what things if we’re going to go to first principles, let’s go back to therapies that have been shown to modify Efflux capacity. And of course, some of the call may recognize that there are therapies being developed now around harnessing ApoA-1 with a highly functional mutation that enhances efflux capacity. These ApoA-1 therapeutics are currently being studied in patients with acute MI so efflux capacity is improved with these ApoA-1 therapeutics. So that’s on the horizon with respect to how we might manage patients with a deleterious HDL effect score. But we have some things already available to us. Fibric acid derivatives appear to increase efflux capacity. Diet and exercise clearly in some studies has had an effect, in particular with weight loss and improvements in glycemic control. And I’m going to touch on that in a second as well. Now that we have a randomized controlled trial demonstrating the benefits of ethyl EPA, in particular, the data suggested that omega-3 fatty acids may improve efflux capacity immediately closes that loop, I think for from a therapeutic perspective. There’s a unifying thread, if I had to if I had to put my own personal opinion here, which is that after a genetic dyslipidemia, specifically the forms of dyslipidemia that are so common in the current United States population of the so-called metabolic syndrome, truncal obesity, arterial hypertension and low HDL, high triglycerides with a tendency toward disglycemia this population in particular is a really, I think, really optimal group to be thinking about this assay. And because the various therapies that go toward addressing efflux capacity have a clear niche and it’s not a small niche, a clear niche for treating these patients. So fibric acid derivatives, diet and exercise, low inflammatory diet as well as fatty acid supplementation. So, you know, when I think about, you know, in who would I use this test? You know, patients with obstructive disease, of course, regardless of their LDL and HDL, because we’re really not talking about absolute HDL levels. We’re talking about it’s functionality. Patients with pan vascular disease. You know, I think also would be an obvious patient population to assess so PAD and cerebrovascular disease and then those with at least one risk factor, notably including, among other things, obesity. I mean, I just read a study yesterday that suggested that something like 40% of the US population has a body mass index of 30 or greater at this point. That in and of itself, as a population, I would absolutely think about. Who would I not test the said. Well, I think in those patients that really have zero risk factors for cardiovascular disease, I think we have to have first principles here and sort of say, well, look, if they don’t have obvious risk factors, what’s the likelihood that we’re going to with a low pretest probability add very much with respect to an early actionable signal? And then the second point, I mean, unless and until you’ve intervened with therapy where you would expect a change in efflux capacity, I wouldn’t remeasure it. But I would point out that we don’t have data about the long term stability. And when I say stability, I don’t mean in a chemical sense. What I’m saying is we don’t have data right now about how much this does change over time. So certainly on a yearly basis, I think it would be reasonable. But in particular, what I would love to see is how the HDL effects assay is modified by therapeutic intervention. And so after therapy, I think that would be a reasonable place to think about rechecking the HDL effects. So in summary, ladies and gentlemen, it’s been a really interesting journey watching how Cleveland Heart Labs has come to this point, developing a methodology that acquires ApoA-1 associated lipoprotein particles in a scalable fashion that is a quite viable alternative to the classical HDL functionality testing. It correlates quite nicely with the efflux capacity test. And this this panel is really best thought of as a multi marker or lipidoomics approach that reports concentrations of A1, C2, 3, 4 and and groups them together in an algorithmic approach to the HDL effects pCAD score. This pCAD score derived from the methodology I just mentioned is useful not only for for prognosticating the presence of obstructive coronary disease, but also predicting the potential risk for incident myocardial infarction. So it says predicting which female patient. So it’s by no means only useful in women. But but in the study of only women, it was effective for predicting MI in these patients. And in the Casablanca study, the HDL effects panel predicted cardiovascular death, especially in those patients with obstructive coronary artery disease. So I actually have learned a lot from this journey. And just as atherogenic particle assessment has really become an essential component to LDL cholesterol assessment, it’s not just the LDL, but the components of LDL, understanding the function of HDL rather than the form the the concentration, adds clinical value to the evaluation of our patients at risk for coronary artery disease and its complications. So with that, I’m going to stop and I’ll turn it back over to Dr. Penn. Thanks so much. Once again, Mark, I think you’re on mute. Well, at least I’m consistent. Thanks, Jim. That was great and really informative and greatly appreciated. If we could go forward, please. The next slide. You know what we’ve learned and what you know, some preclinical basic science chemistry in the bench. And then as Dr. Januzzi described, the experience with CTEP inhibition has shown us is that functional HDL can become dysfunctional. and one of the measures of that is a decrease in cholesterol efflux capacity. But this dysfunctional HDL clearly becomes proinflammatory and prothrombotic, which leads to the negative consequences with respect to clinical events. Next slide, please. With that in mind, I think you know where pCAD scored through the HDL effects study really is in the middle of the inflammation, altered lipids and really fits in the setting of abnormal inflammatory testing through our approach there or directing patients to inflammation testing of HDL effects or pCAD is abnormal. We have some emerging data that demonstrates that insulin resistance and patients with insulin resistance have a decrease cholesterol efflux capacity, which isn’t a surprise given the fact that a low HDL is one of the characteristics of metabolic syndrome. And then for those interested in advanced lipid testing, whether they’re using HDL particles or HDL to be or other attempts at a surrogate for cholesterol efflux capacity, we’re now able to measure that directly. And where some of those surrogates may have fallen short in the past, an absolute measure now that’s related to efflux capacity, I think is important. And then in the setting of a standard lipid panel, you know, we get HDL in that panel, obviously. But understanding where what the efflux capacity is of the HDL in that given patient may be quite valuable in defining long term risk. So we’re again, we’ve spent a lot of time, a lot of effort, and we’re really excited where this has gotten to. And we think we’re able to give unique insight into patients metabolic risk with respect to HDL function that frankly, clinically we have not had access to before. So again, happy to answer questions. Dr. Januzzi, first, I’m happy to be there as well. And we again, really thank you for your time today and your interest. Thank you so much. Doctors Penn and Januzzi We do have a few minutes at the closing here to address some of the questions that have been submitted. Please remember, if you do have a question, use the chat feature in the WebEx to submit your question to the group. All panelists, that’s who you want to direct them to. So we’re seeing some common themes that have come across. So I’ll try to address those first. Many of you are understandably asking questions about interventions and if pCAD is modifiable specifically several around statins. Dr. Januzzi, I’ll pass this one over to you. You mentioned statins don’t change the legacy efflux assay the traditional measurement of efflux. Why do you think that is? Well, I mean, it sort of speaks to the multifaceted nature of lipid biology. You know, if you think that there’s a efferent and afferent loop, you know, statins really influence the the efferent loop, whereas, you know, efflux capacity represents the other side of the equation. So I think that that’s an overly simplistic way of explaining it, but nonetheless, you know, gives, I think, some sense of why treating one side of the equation may not necessarily affect the other side. Okay. This is another one. Another one for you, Dr. Januzzi. This is this is interesting. I haven’t thought about this. What are your thoughts in using this for some clinical data? CVD is a non obstructive individual. Do you think this has any use? Oh, absolutely. I mean, I think it’s important to emphasize that although the you know, although the data in Casablanca, what we queried was what does it predict obstructive CAD. We we chose obstructive on the basis of the sort of the canonically sort of dictated 70% or greater. There’s no reason not to expect that that intermediate scores or even high scores might be associated with variations in the degree of obstructive obstruction. What it’s basically saying is you’ve got a risk for atherogenesis. And so I think that in patients with non obstructive disease, in some ways there’s even more to be gained because these patients could have a therapeutic intervention that might be expected to retard progression of their obstruction. This next question I’d be interested to hear both both of you weigh in on people are trying to wrap their heads around the new assay and how it how it associates to what what they have been measuring. Of course, HDL cholesterol is what everyone’s mind goes to. But also some practitioners have been using particle size particle density, particle number in in terms of LDL, but HDL as well. Do we have any indication as to whether an HDL function assay like this may correlate to a certain size HDL particle or number of particle? I’m going to defer that one to Dr. Penn. I don’t know. Yeah. Yes. As I as I mentioned before, I think we’ve sought surrogates for cholesterol efflux capacity because we didn’t have a measure for it, HDL 2b particle number, things of that nature in very limited analyses, we have not seen strong correlations between those measures and pCAD score. And that’s in a poster. We we had at the HA about 350 odd patients last November. The reality is now that we have a direct measure of efflux. It’s unclear to me what the role of those surrogates are with respect to efflux capacity. Others may find other uses for them. You know, we did a fairly we did a completely unbiased approach to the Apoprotein world of HDL. The fact that we landed on C1, C2, C3 and C4, as well as ApoA-1 was interesting to us because many of the ApoC proteins in HDL have been thought to have a role in efflux capacity, and we now have a way to combine those to get a measure. You know, again, the role of surrogate markers, then for efflux capacity, I think is lessened now that we have a measure. Dr. Januzzi, in the talk you mentioned, it may be reasonable to reassess pCAD after some sort of intervention. Is there a optimal time to allow that to take to pass before we recheck? Yeah, that’s a great question. You know, we don’t have a good grasp of how rapidly efflux potential, at least using the HDL effects panel will be detected to change. Nor do we have a good read of how much change would be associated with an improvement in prognosis. But, you know, unlike, for example, statins that have a direct rapid effect on LDL cholesterol concentration, at least with with efflux capacity as a biological process, one might be expected to see it take a little bit longer to change. And so my general sense, from the knowledge of the literature would be to to reassess probably longer than just a few weeks. I would probably give it a little bit longer before remeasuring in order to really get a clear picture. This is definitely the subject of the kind of follow up studies that are needed in terms of developing some practical knowledge for clinicians. Similar to what we’d seen in questions from last evening’s webinar that we conducted. Multiple questions on the topic of Omega-3 therapeutic doses, ETA, vascepa. Could either of you sort of pontificate or hypothesize whether that may be a a potential therapeutic? And if so, the mechanism that that might modify HDL? CEC. Mark, you want to take that one or should I? Go right ahead, Jim. Yeah. So, you know, thanks. So, you know, as a clinical trial is I tend to focus on the doses studied. So I would stick with the doses in the Improve-it study. I don’t know if we’re going to necessarily get more effect from a higher dose. I certainly would not go lower. So in that respect, you know, I would just follow the approach used in Improve-it and, you know, and then follow these patients. And I’m having a lower lower bar for initiating, you know, in terms of in terms of the use of ethyl EPA. Previously, I would sort of go through the process of counseling, as I always do, with diet and exercise. But I’d become a little bit more nihilistic about the likelihood for a robust enough response to avoid prescriptions. So I’m prescribing it more frequently. Interesting. Doctor, panel, direct this one over to you. How does HDL affects pCAD compared to a predictive value provided by Corus CAD? As I understand, Corus CAD uses MRNA activity signatures in symptomatic patients to rule in with with or without obstructive CAD. How does a pCAD proteomic assessment compare to a MRNA activity signature like Corus CAD? Yeah. So to your point, Corus CAD is is designed to determine whether somebody’s a stable angina, which is a very different than having atherosclerotic disease. And as Dr. Januzzi mentioned in the paper, in the paper we show that an increase in pCAD is an increase in atheroma not necessarily just obstructive disease. So the pCAD score is reflecting atherosclerotic risk, not the presence of chest pain or stable angina. So the goals are very different. We have not compared them, but the goals of the tests and the meaning of the tests are very different. A positive pCAD test score does not mean somebody needs a catheterization looking for obstructive disease in the right patient at bay. But it really says this patient has likely significant atheroma and needs preventive therapy. Hmm. Interesting. We have time for about two more questions. I have one. One for each of you. The first one, Dr. Januzzi. Is it useful? Is this is this assay specifically useful in a low HDL? Individuals say below 40? Oh, I would actually argue it’s even more useful in a patient like this because not only are they disadvantaged by a low circulating HDL, but if that HDL is poorly functional, they’re at very high risk. So below HDL, high triglyceride, after a thrombotic, you know, metabolic syndrome phenotype is is going to be probably one very big area of focus. So I think I think it would be actually quite useful. Okay. And last but certainly not least, someone’s inquiring about key outcome studies underway. What’s next for research for HDL effects? Dr. Penn, I’ll pass that one to you. Well, it’s about what is lowering it and modifying the measure and what does it correlate with. We continue to be able to do. You know, we’re going to chat with Dr. Januzzi and his team about what we can do in follow up studies as well. We’re also looking at some biobanks to test the pCAD scoring and we’re doing some work with the HA through the one brave idea half way to look at HDL effects as well. So we’re trying to get answers to many of the excellent questions today. But in our patients currently, who we may or may not understand what their atherosclerotic risk is or what’s driving the risk, you know, this is the first time we’ll be able to measure have something related to HDL efflux. I think for now, that’s really important.