May 30, 2017
Dr. Carolyn Lam: Welcome to Circulation on the Run, your weekly podcast summary and backstage pass to the journal and its editors. I'm Dr. Carolyn Lam, associate editor from the National Heart Center, and Duke National University of Singapore. Our featured paper this week confirms the clinical utility of a polygenic risk score of common variants of cardiovascular disease. More soon after this week’s summary of articles.
The first original article describes distinct cell-specific roles for NADPH oxidase, or Nox2, in blood pressure regulation. This paper from first author, Dr. Sag, corresponding author, Dr. Shah, colleagues from King's College London British Heart Foundation Center of Excellence in the United Kingdom. The authors used novel gene modified mouse models to show that Nox2 in myeloid cells modulates basal blood pressure whereas endothelial cell Nox2 is involved in angiotensin II-dependent hypertension. The finding that Nox2 in different cell types has distinct effects on blood pressure, suggest that different diseases conditions may alter blood pressure through effects on Nox2 in different cell types. For example, it is conceivable that the effects on myeloid cells on basal blood pressure may be enhanced in inflammatory settings, whereas endothelial cell Nox2 activation may be more relevant to renin-angiotensin system-dependent hypertension. The current results are therefore relevant to the design of novel therapeutic approaches for hypertension by targeting NADPH oxidases.
The next paper provides a new, more accurate atherosclerotic cardiovascular disease risk prediction tool in familial hypercholesterolemia that may increase the efficiency of care and use of newer lipid lowering therapies. Co-corresponding authors, Dr. Mata and Pérez de Isla, from Hospital Clinicals San Carlos in Madrid, Spain, use data from SAFEHEART, a multicenter, nationwide, long-term prospective cohort study of 2,404 adult patients with molecularly-defined familial hypercholesterolemia and who have followed up for a mean of 5.5 years. They developed a robust risk prediction equation for incident atherosclerotic cardiovascular disease based on the following independent predictors; age, male gender, history of previous atherosclerotic cardiovascular disease, high blood pressure, increased body mass index, active smoking, LDL cholesterol and LPA levels. The new SAFEHEART risk equation performed better with a Harrell C index of 0.81 compared to 0.78 for the modified Framingham's risk equation and 0.8 for the ACC/AHA Pooled Cohort risk Equations. The authors therefore concluded that the risk of incident atherosclerotic cardiovascular disease may be estimated in familiar hypercholesterolemia patients, using simple clinical predictors, and that these findings may improve re-stratification and could be utilized to guide therapy in patients with familiar hypercholesterolemia.
The next study tells us that late gadolinium enhancement cardiovascular magnetic residents identifies patients with dilated cardiomyopathy but without severe left ventricular systolic dysfunction, who are still at high risk of sudden cardiac death. In this study, by first author Dr. Halliday, corresponding author Dr. Pennell, from Royal Brompton Hospital in London, United Kingdom, the authors prospectively investigated the association between mid-wall late gadolinium enhancement and the primary composite outcome of sudden cardiac death or aborted sudden cardiac death, among 399 consecutive referrals with dilated cardiomyopathy and a left ventricular ejection fraction above 40% seen at their center between 2000 and 2011. These patients were followed for a median of 4.6 years. 17.8% of patients with late gadolinium enhancement reached the pre-specified end point, compared to only 2.3% without late gadolinium enhancement.
Furthermore, following adjustment, late gadolinium enhancement predicted the composite end point, with a hazards ratio of 9.3. Thus, patients with dilated cardiomyopathy and mid-wall late gadolinium enhancement, and mild or moderate reductions of left ventricular ejection fraction should still be recognized as having a high risk of sudden cardiac death. This is important because these patients are not currently offered ICDs for the primary prevention of sudden cardiac death, based on current guidelines. Due to the low competing risk of death from non-sudden causes, it is possible that these patients will benefit from ICD implantation, but randomized trials are now required. These issues are discussed in an accompanying editorial from Dr. Markman of Johns Hopkins University, and Dr. Nazarian, Hospital of University of Pennsylvania.
The next study enhances our understanding of the role of immunity in hypertension. Now, the innate antigen-presenting cells and adaptive immune T-cells have long been implicated in the development of hypertension, however, the T-lymphocytes subsets involved in the pathophysiology of hypertension remain unclear. A small subset of innate-like T-cells expressing the gamma-delta T-cell receptor, rather than the more commonly expressed alpha-beta T-cell receptor, could play a role, and these were the focus in today's paper by first author Dr. Caillon, corresponding author Dr. Schiffrin, and colleagues from Sir Mortimer B. Davis Jewish General Hospital, McGill University, Montreal, Canada. In experimental models, the authors showed than angiotensin-2 infusion increased gamma-delta T-cell numbers and activation in the spleen of wall tite mice, as well as in increased the systolic blood pressure, and decreased mesentric artery endothelial function in wild type mice, but not in mice devoid of gamma-delta T-cells, or in mice depleted of gamma-delta T-cells by depleting antibody injections.
Furthermore, angiotensin-2 induced T-cell activation in the spleen and peri-vascular adipose tissue was blunted in null mice. In humans, there was an association between systolic blood pressure and gamma-delta T-cells. In summary, this is the first in-vivo demonstration that gamma-delta T-cells, a subpopulation of T-cells, play a fundamental role in the development of hypertension and vascular damage. These results will help design novel treatments to limit the progression of hypertension and vascular damage.
The final paper describes a novel multi-modality strategy for cardiovascular risk assessment. Dr. de Lemos and colleagues from UT Southwestern Medical Center in Dallas, Texas, hypothesized that a strategy combining promising biomarkers across multiple different testing modalities would improve global and atherosclerotic cardiovascular disease risk assessments among individuals without known cardiovascular disease. These modalities included: left ventricular hypertrophy by electrocardiogram, coronary artery calcium, N-terminal pro B-type natriuretic peptide, high sensitivity cardiac troponin-T, and high sensitivity C-reactive protein.
Using data from 6,621 individuals of the multi-ethnic study of atherosclerosis, or MESA, as well as 2,202 individuals from the Dallas heart study, the authors evaluated the association of test results with the global composite cardiovascular disease outcome, and that would include cardiovascular death, myocardial infarction, stroke, coronary or periphery revascularization, incident heart failure or atrial fibrillation, as well as atherosclerotic cardiovascular disease outcomes, which included fatal or non-fatal myocardial infarction or stroke. Over more than 10 years of follow-up, the authors found that each test result was independently associated with the global composite cardiovascular disease events in MESA. When the 5 tests were added to a base model, the C statistic improved, that was significant integrated discrimination improvement, and net reclassification improvement, and the model was well-calibrated. Using a simple integer score counting the number of abnormal tests, they showed that global cardiovascular disease risk increased with increasing score in a graded fashion. These findings were replicated in the Dallas heart study, and were similar for the atherosclerotic cardiovascular disease outcome.
This study therefore supports the potential value of a multi-modality testing strategy in selected individuals, in whom additional risk stratification is desired, beyond measurement of traditional atherosclerosis risk factors. The authors do highlight that additional studies are needed to validate the present findings, determine the optimal approach to implementation, and address direct and indirect cost implications of the additional testing.
Well, that wraps it up for your summaries. Now for our feature discussion.
Our feature paper today tells us that a polygenic risk score identifies a group of individuals with a higher burden of atherosclerosis, and greater relative benefit from statin therapy in the primary prevention setting. But perhaps even more significant, is that it addresses the fact that even relatively small effect sizes of common snips gathered together in a genetic risk score may have clinical utility in the prediction of cardiovascular disease, and to discuss this I'm so pleased to have the first author, Dr. Pradeep Natarajan from Massachusetts General Hospital, and Dr. Anand Rohatgi, associate editor from UT Southwestern. Welcome, gentlemen.
Dr. Pradeep Natarajan: Thank you very much, Carolyn.
Dr. Anand Rohatgi: Thank you, Carolyn.
Dr. Carolyn Lam: Pradeep, could you start by telling us what you did? This was a tour de force, please.
Dr. Pradeep Natarajan: Yeah, thanks so much for the invitation and the enthusiasm. So, briefly, large-scale, genome-wide association studies have discovered genetic risk variants in the population that individually associate with coronary disease risk. Many others have shown that an aggregate of these genetic risk variants predisposes to an increased risk for coronary disease by about 60%. But we sought to, with this study, understand how primary preventive statins could influence that risk, and whether these insights could be helpful in refining statin eligibility. So, among the individual variants that had been associated with coronary disease, we developed a risk score. This encapsulated 57 individual genetic variants. This risk score is independent of traditional cardiovascular risk factors, and identified individuals with a greater burden of sub-clinical atherosclerosis, defined as coronary artery calcium and carotid plaque, and two observational cohorts in individuals with a greater absolute and relative benefit from statin therapy from a subgroup analysis within the WOSCOPS clinical trial.
What we were surprised by is that the conventional wisdom, that all previously described subgroups within statin trials had the same relative benefit, and statins per unit of alveol cholesterol lowering. So, about 20 to 25% lowering of risk per 40mg per deciliter of alveol cholesterol. So we clinically identify individuals who just start out at high absolute risk, assume that the relative benefit will be the same across everyone, and optimize the number needed to treat simply by just finding individuals at high risk. But, here we didn't see the expected 20 to 25% lowering in the high genetic risk group, we saw actually a 44% relative risk reduction for the same lowering of alveol cholesterol. And we have now observed that across three different clinical trials, and these individuals are at high baseline risk, so this translates into an even more optimized number needed to treat, and really the opportunity to identify individuals earlier with an age independent biomarker.
Dr. Carolyn Lam: That's really cool, in fact, the number needed to treat in the high-risk score group was impressively low at 13.
Dr. Pradeep Natarajan: That's correct. Now, overall in the WOSCOPS trial, if you look at all individuals, it's about 38, so it is a high risk primary preventive group of men with, you know, substantial hyperlipidemia, but if you look at at least a relative difference between the two, going from 38 to 13, that's about a three-fold improvement of the number needed to treat.
Dr. Carolyn Lam: You know, what you said about it not correlating with exactly what you expected with the drop in LDL and so on, does that mean that this genetic risk score, that a lot of the snips are probably associated with LDL levels, but that a lot of them may be giving more information beyond LDL? Is that what it means?
Dr. Pradeep Natarajan: Yeah, you know, it's interesting. Most of the genetic variants that are associated with coronary disease actually do not seem to clearly influence traditional cardiovascular risk factors. The latest best estimate of that is about 39% of them associate with traditional cardiovascular risk factors, and then a subset with LDL cholesterol. So the aggregate score actually does not associate with traditional risk factors, and including with LDL cholesterol.
Dr. Carolyn Lam: Wow, and Anand, I'm sure we had so many discussions with the editors about the paper. Could you share some thoughts?
Dr. Anand Rohatgi: Yes, Carolyn. Circulation as a journal represents the best in cardiovascular science, and we're always interested in the highest-level articles related to atherosclerotic cardiovascular disease. So, when we received this manuscript from Pradeep and Sekar’s group, really leaders in the field, we were really excited, and as we went through the review process we got even more excited because it, as you said, Carolyn, it really was a tour de force, it was a high-quality article and it combined multiple things, and that's what we're really interested in seeing at Circulation, is combining several aspects, in this case genetics, sub-clinical atherosclerotic imaging, and also treatment effect.
And, you know, it's interesting because several recent manuscripts looking at genetic risk scores, they were associated with coronary disease but it wasn't clear that they were improving what we call risk prediction performance indices, at least enough to meet the bar of incorporating them into guideline-type recommendations. So I think the field wasn't sure how to move forwards with this type of information, but now I think this study really demonstrates that this type of risk score, this genetic risk score, really can inform treatment decisions in a big way. And so we were really excited to talk about that and then see it move forward.
Dr. Carolyn Lam: So a question for both of you now. Can these data be extrapolated to other cohorts of patients? I mean, WOSCOPS was predominantly white, and all were males, right? So, Pradeep, would you like to take that first?
Dr. Pradeep Natarajan: That's an excellent observation, and I think ... A clear limitation in the field, but an outstanding question that I think can be addressed going forwards. So, the main challenge is that the epidemiological cohorts that were used for genetic analysis largely have been of European ancestry, and we know that genetic background and a variety of non-genetic factors influence cardiovascular disease risk, so in genetic analysis of European individuals the influencers of coronary disease risk may not influence cardiovascular disease the same in non-European ethnicities. And, you know, we've done some work of this specifically in African-Americans, and there are some differences. You know, African-Americans are largely mixed of both African and European ancestry, some of that seems to also influence how you interpret the cardiovascular genetic risk score.
Ideally you would have a risk score that is not influenced by the genetic background, and so the next step going forward are one to look to see how well this risk score predicts in non-European ancestry, because, obviously, not as much statin clinical trial information in non-European cohorts, but I think looking at the treatment effect in non-Europeans will be important. And then, you know, the third step is we and others are participating in several now large ongoing efforts to really define what the genetic influences are in non-European ancestries, and I think that will be a very important next step that's really critical before the clinical implementation.
Dr. Carolyn Lam: Yeah, talking to you from Asia, that's music to my ears, obviously. Anand, did you have any questions for Pradeep or anything else to add about the paper?
Dr. Anand Rohatgi: Yeah, I wanted to add one or two comments. One thing that this study demonstrates is that the genetic risk scores, whether they relate to traditional risk factors or lipids, that doesn't necessarily translate to what it might mean in terms of treatment benefit, and so I think that concept is generalizable and now it needs to be tested in other ethnicities, other types of subgroups, but I think you can disentangle a relationship with risk factors and lipids to its treatment effect and this study really nicely shows that.
And I think just to take a step back, we know statins work in intermediate-risk patients, maybe even low-risk patients with the most recent studies, but at a public policy level, and just as a cognition, we really want to narrow the focus, it's something called precision medicine that the American Heart Association is promoting as a concept, and I think that this study really demonstrates that here we have now another tool that can reduce this number needed to treat, make this choice for statins more precise, maximizing the benefits and limiting cost. So, I think that concept is very generalizable, it needs to be tested now in multiple populations, like Pradeep said, and I guess one of the questions I had had for the authors is: how do we incorporate this finding that they saw with sub-clinical atherosclerosis, which we thought was very fascinating among the editors at Circulation, that now they're also linking with sub-clinical atherosclerosis, is that something that the investigators think needs to be pursued further? Would that be something that would be used clinically as well?
Dr. Pradeep Natarajan: I think there are lots of opportunities for this going forward, you know, in prior work we've done the genetic architecture for clinical coronary disease is actually very similar to sub-clinical coronary disease, and there are many influences for sub-clinical coronary disease, and clinical coronary-disease, that are both genetic and environmental, and the aggregate effect from the polygenic risk on sub-clinical atherosclerosis suggests that it's obviously not absolute and there are other factors that influence sub-clinical atherosclerosis.
Dr. Carolyn Lam: Well, listeners, you heard it right here. Thank you for joining us this week, tell all your friends about it, and don't forget to tune in again next week.