 What I'm going to focus on today, because you'll notice that we have the American College of Cardiology just after me, is really more of the complex genetics and where we are currently in the state of the field, just broadly in cardiovascular disease. And look at what we would need to do in terms of thinking about creation of a workforce that had the competencies that can address the issues of complex disease genetics more broadly. So we know there are a number of different Mendelian genetic diseases in the realm of cardiovascular diseases. We've heard hypertrophic cardiomyopathy mentioned, familial hypercholesterolemia is a second, long QT syndrome a third. So we have a number of those Mendelian conditions. So I'm just going to focus now on where we stand on the state of the art with respect to just common disease genetics and what we in the Functional Genomics Council are doing in this area. So we all know that it's hard to believe that genome-wide association studies just were named the Discovery of the Year by Science Magazine in 2007. Can you believe it? Just six years ago. And to date, these studies have grown and discovered hundreds and hundreds of interesting variants for common diseases. In the cardiovascular domain, one of the largest done to date is a genome-wide study of 200,000 individuals of European descent for the trait systolic blood pressure. What's interesting about these genome-wide association studies and how we really are at the, still at the precipice of understanding biology and still meaning the research to understand what these studies mean. We are long away yet from having things that we can even create appropriate use criteria for. In the world of systolic blood pressure, it's a common disease. It affects 70 million Americans' hypertension. And we really don't understand the complex genetic basis of hypertension. What's really unique about GWAS is that of the 28 loci on this slide for systolic blood pressure, 22 were in previously unrecognized areas that would contribute to blood pressure. If we look at stroke, a recent GWAS published just this month in January in Lancet Neurology reports on over 16,000 cases of ischemic stroke and 64,000 controls. And they identified several loci for ischemic stroke. What was really interesting about this recent study is that all stroke is not the same. So what they discovered when they broke down the stroke, which they could by virtue of this incredibly large sample size, is that the loci that emerge are different for large vessel disease at the top and small vessel or cardiothimbolic stroke on the bottom. So here on the top, you see that the HDAC-9 gene and the 9P21 are both important for large vessel disease stroke. And interestingly, the 9P21 is the same locus that shows up for myocardial infarction repeatedly across all GWAS studies. For cardiothimbolic stroke, the two loci that appear are both associated with atrial fibrillation, which makes sense in terms of cardiothimbolic stroke from a phenotypic domain. So what we're learning in the cardiovascular and stroke realm is that these diseases are complex, that we need really incredibly precise phenotyping to get closer to the genetic basis. And now we're moving into the realm, I won't call it whole, but of exomen genome sequencing. I don't know why that would confuse scientists, but maybe I'm just missing something. But the state of the science so far is that we really are just beginning to discover new genes using the technology of exomer genome sequencing in both cholesterol and blood pressure control mechanisms, and that we're really now just at the precipice again of taking these findings to their functional characteristics to understand what they mean physiologically. So our real challenge now is moving from all of these discoveries into what it means in terms of the phenotyping. We have an example here of how that has been done in cardiovascular disease with a sort one locus. So what are the clinical applications right now in terms of complex cardiovascular and stroke disease? We're really, despite this wealth of knowledge from a research domain, cardiovascular and stroke care has not yet been significantly impacted. Again, this is related just to complex diseases. And we expect that translation will ensue and accelerate over the next five to ten years. Where we think that will happen is in the development of risk prediction models using both common DNA variants and these rare exonic or whole genome variants. So in terms of what we know by common DNA variants, in most cases these have been discovered by GWAS. For those of us in the room that have dealt with common diseases, we know that each of these variants has very small effects and that not one alone is going to really help us use genetics for predictive purposes. But we can use them in aggregate to combine them into risk scores. And these genetic risk scores have indeed been tested in the cardiovascular domain. In aggregate, they show modest prediction for cardiovascular disease. For example, for incident coronary heart disease, the high genetic risk score in the paper that was on a slide a few back where we've looked at 95 different variants, it still only confers about a 70% increased risk. So really it's a limited clinical utility at the moment except if you want to better risk stratify people in order to understand those who may be at intermediate risk. In terms of rare variants, these have the potential to have very large effects to increase or decrease risk multiple fold, not just in the percentage range. But however, each of these rare variants, as we know, occur in only one or a few people at most. So it's making these creations of these common risk scores challenging because incorporation of those rare variants that occur in just a few people with these common variants, it makes it computationally challenging. So ultimately we're left with this needle in the haystack problem with the rare variants where our exome and whole genome sequence have identified hundreds of rare variants in an individual, but most of those rare variants are relevant to disease. And how do you sort through which of those are clinically important? And right now in the field of cardiovascular and stroke disease, this is where we are most challenged. And what we know of those that have been done, and Ray Hershberger who sits in the room wrote one of the first papers that really implicated going from exome sequencing in a family down to the disease itself and the mechanism. The genes that we're discovering for these complex cardiovascular disease also tend to fall in those genes that have already been implicated. So what does it mean if we do have a mutation? So here again I'm restricting my comments just to the cardiovascular and stroke domain. Even if we have a known gene, we have two problems. The first problem is that that particular mutation may or may not affect the gene function and those things are difficult to predict at the outset. And second, even if the mutation changes that function, the penetrance may be affected by genetic background and even by the environmental constraints in which the individual lives. And even within a family, some people with a mutation, the most well studied in our field is hypertrophic cardiomyopathy, may or may not develop a disease or set incomplete penetrance. So what are the dangers that we perceive in our area in the American Heart Association about genetic information? One of the real challenges, and we presented on this at our scientific sessions two years ago, is the concept of having direct to consumer marketing and how that is truly and fundamentally impacting clinical practice. So the direct to consumer testing informs patients of genetic risk scores for now, for common variants, but it can lead individuals, perhaps they have a high family risk, they get their family risk score, it can potentially lead them with this false sense of reassurance or possibly false worry. And so it may or may not be helpful in them to risk stratify for their own sake about whether or not they should pursue regular prescriptive and preventive medicine. And we think the problem will increase as more genomic technologies are applied and directly accessed by consumers. And we feel that the providers still are really not aware of the limitations that are imposed by these direct to consumer products and they don't know how to appropriately counsel patients and our provider's lack of knowledge will serve our patients very poorly and we need education. And we've tried to crack this nut of education in many different ways at our annual meetings in November. We've tried to have courses called Genetics for Dummies. Well that really offended a lot of people. We thought it would entice people who felt that it was too beyond them to go and listen to a talk and so we created the Genetics for Dummies course. That didn't go over. So we've tried different ways to engage different clinicians, scientists to come and get information. But one of the speakers earlier said only 1% of the audience wants to be there if you're lucky. So we find that these sessions, even though very well put together and by the leaders in the field of cardiovascular science are not extremely well taken up on at our meeting. By far one in the cardiovascular area, one of the biggest beacons of hope has been in the area of pharmacogenomics. Basically thinking that we can use genetic screening to get the right dose of medication to the right person and at the right medication itself. And so we would be integrating this into clinical practices, Bruce Korf pointed out at the point of care. But it's not yet adopted in the cardiovascular stroke care, even though there are nine different agents which have some kind of FDA indication for testing. There are several emerging applications for the drugs that are listed here. And by far one of the most recent sort of wins, if you can call it a win, is with the story of Clopidogrel and SIPT-2C19. And the reason that this is a potentially important area is that for every patient that presents with an acute coronary syndrome, being an infarct or just an acute coronary syndrome, they're usually given a drug, an anti-platelet agent as a discharge medication. So it's a large number of people that are affected by this prescription every year. So the routine practice is to give Clopidogrel, which is a type of thionopyridine. And that reduces the risk of further cardiovascular events as well as the risk of having a stent thrombosis. So why do you think this is an important clinical question now for anyone in the room except Alan, who knows the question very well? Do you know why this is such a hot issue now in cardiology? Well, most things are driven by payment, right? Well, the issue is that Clopidogrel is now coming off patent, which makes it hugely more affordable. So it would make sense to use this drug if it's coming off patent, because it's affecting so many people. The challenge is that there is the Cip2C19 enzyme that encodes the major active metabolism of Clopidogrel. And so these reduced function alleles result in less Clopidogrel efficacy, which led to the FDA black box warning. And if you look at the clinical trial evidence, you see that carriers of that particular reduced function allele have a higher rate of cardiovascular events when they're given Clopidogrel versus if they're a non-carrier. So an important question, so what do you do then when you have a patient that presents to the hospital with an acute myocardial infarction? You can determine their genotype. You can, if they have a normal allele, just give them Clopidogrel and send them to the cath lab. Or if they have the reduced function allele, now you have options. You can give them a second or different kind of thionoperidine, or you can just double the dose, etc. And so we really are still somewhat limited because there's little point of care access to this genotyping. It's being piloted now by a number of hospitals. It has yet been validated by a clinical study. And it's unclear at this point which patients, if any, would benefit. So we really haven't even gotten to the point. We have an FDA black box warning for this drug. We really don't have a routine clinical use yet validated in terms of a systematic test. So adoption of pharmacogenetics will be still a little ways off in the cardiovascular domain. It will be at least five years. We do expect rapid adoption by academic centers and large health care networks for things like the CIP-2C-19. But there will be this much slower adoption by individual practitioners because we do have this real knowledge gap in terms of education. So we are involved in providing education to providers. We say here that although we have a lack of provider education on genomics issues, it's not really jeopardizing clinical care now because honestly we don't have in the common disease domain a lot of products that are ready for prime time. So we're moving now aggressively to provide this literacy in two ways. The first is we're continuing our scientific statements. We've published a number of these, our first one. I wrote Genetics and Genomics at Prevention. I think that came out in 2006. And those were just state of the science at the current time. We're updating that now. It should come out this year. And we have a working group that's creating a statement on the use of genomics and genetics in cardiovascular and stroke patient care, which is really provided exclusively for the provider to give them education. And this is the outline of that statement. And then finally, we're also developing a massively open online course monitoring the latest trends in education. We're formulating that course that will cover all the topics that are outlined on the prior slide in that use of genetics and genomics in patient care statement. And that course will be freely available to the biomedical community. And we're pitching it at a broad range from the undergraduate level. We really think that genetics and genomics is going to have to be adapted by many members of the health care team, not just the physician. So we're encouraging our nursing council and pharmacists to also become involved in this kind of health genomics literacy. That massively online course will be supplemented with modules for specific cardiovascular disorders that I've already mentioned before. And the initial versions will be done in English, which will be followed by translations into other languages. Our scientific publications like circulation are provided in a number of different languages. So we have the skill set to do that. So with that, I will conclude my presentation. And Terri, I did it in 20 minutes. All right. Sure. Questions? It's always good to be in the post-brandial phase when you're giving your presentation. Terri has a question. But I should defer to David since I talked a lot. I just wanted to make, I mean, I really enjoyed that presentation. That was really terrific. And one of the things that everyone should just make sure they're focused on is that about, well, when I'm doing a genome case, I'll often find maybe a quarter of the time I'm looking up some variant that's quote, known to cause a disease, and the literature is wrong. And so that's like a whole other level of problem. And particularly in hypertrophic cardiomyopathy, that's particularly evil in terms of the quality of the literature. I hope there are no hypertrophic cardiomyology. Researchers here will yell at me, but there is quite a. I don't mean to pick on hypertrophic cardiomyopathy, but there's a huge problem with the literature that supports all of the evidence that we're trying to build on here. OK, so let's see. I think we have Maron, but maybe I'll get my comment in real quickly. So Donna, thank you. And thank you for putting up with us all eating while you talk. That was nice of you to eat your lunch early. And I was curious about the comment you made that will be five years likely before pharmacogenomics. It'll be some years before these can be implemented, because these projects are being evaluated and validated. And so my question is, what is the basis and the criteria for your evaluation and the validation that you are expecting? Emy, where did I come up with the five years? No, really, what will it take for the AHA to say, yes, this is validated? So Bill has already mentioned that we, the American College of Cardiology and the American Heart Association, write joint guidelines. At this point, the warfarin trial is still ongoing. Allen's involved in the SIP-2C-19 trial with the clopidogrel. So we have trials that are emerging, but really they're first in practice kinds of trials. So without an evidence base, we would not write a guideline, but hopefully we can develop appropriate use criteria for these genetic tests. But we do believe we need the RCT evidence to get us to at least a point in making those statements. Yeah, and I think that's a recurring theme in the cardiology field that we don't see in other fields about the RCTs, because there are other types of evidence. And so I may ask, I know Jeff Ginsburg has spoken on this. So Maureen, if you're willing to wait while he addresses this specific point, Jeff. Yes, I just wanted to ask about the RCT level, because as we've commented on already in this meeting, that not all guidelines are at the RCT level of evidence. So is there some middle ground for a consensus statement maybe jointly between AHA and ACC to begin to create that path for getting some niche uses or some subsets of the populations, perhaps, in which a consensus group might believe that it's really ready for clinical use? Well, I'll make a statement. Then I'll have Bill fill in. We do have other methods, which are called the appropriate use criteria, which we use in that gray zone, where we don't have RCT evidence. But would we launch something while these two trials are still ongoing? We would have to take that to our guidelines committee and have them evaluate. Bill? I mean, there are other mechanisms, Jeff, beyond guidelines, which have a definite methodology that I think is very rigorous. There is consensus expert documents that actually we've done together. We've done one when the FDA warning, Black Box Warning came on Clopidigrel and SIP. And that was joined between the American Heart and the American College. So whenever there is more data to do that, I mean, the consensus expert documents are based on some data and some consensus where there's no data, but there's a clinical situation and you have to make a judgment. So I think that there is still tremendous value in that. And in response to one comment before, we hope that this would be an impetus to gather more data, more research, to have more solid evidence in that direction. And there's a whole other topic around which, I don't know if our eGap colleague wants to write about when we served on eGap, how do you know that you have clinical utility? Are they CLIA-certified labs? I mean, the whole concepts of the genetic testing, we really haven't dealt with effectively, I think, from a population scientist perspective. So let me ask Julie, were you speaking specifically to the clinical trial issue? Yeah, so, I mean, I think Terry's right in cardiology, there seems to be this particular RCT bar. And certainly for pharmacogenetics, and maybe for genetics more generally. I mean, I think there's a lot of concern about this genetic conceptualism. Somehow it's different. And so, I mean, for those of us doing pharmacogenetics research, I think we view pharmacogenetics as a tool. Just like serum creatinine is a tool to dose patients who have renal impairment to dose drugs that are cleared renaly. And so, pharmacogenetics can be a tool to help you better individualized therapy in that patient. We've never had an RCT trial that says dosing based on serum creatinine is better than not dosing based on serum creatinine. And so it's, I mean, at some level, I think we just have to change the view. Because if we have to have RCT level data for everything in genetics, then we will never ever use genetic data. And I mean, there's just really a disconnect. I mean, there's nothing absolute about genetics just the way there's nothing absolute about really any tests that's done in medicine. And so I think, and in cardiovascular disease, we seem to be the most challenged with this concept because cardiovascular disease happens to have a lot that's driven by RCT evidence. But I think we have to change the discussion, at least as it relates to pharmacogenetics, but probably genetics more generally. So responses on this point. Martin, we will get back to you. Bill. I think we all agree that the RCTs are an important part of evidence in cardiovascular medicine, but I think in the study that I think Sid Smith did a year or so, two years ago, maybe where he looked at all of the recommendations from the ACCHA guidelines. And there was a surprisingly low number that were based on RCTs, and I wanna say 14%, but I can't be sure of that. But it's a remarkably low number, quite a bit of the recommendations that come out of our guidelines are not based on clinical trials. That's very helpful to know. And I think it's something that the clinician community needs to hear because we get this fed back over and over and over again, that there's not a clinical trial on Simvastatin, so for the variant on Simvastatin, so we're not gonna use it. Other comments on this, Alan? Well, I'll just add that for the CYP2C19 clopridic well story, the reason it's so exciting is also the reason why RCT of appropriate parable will never be performed, and that is that clopridic well is coming off patents. So there's no pharmaceutical company that has enough motivation to wanna fund such a study. Yes, one last comment on this topic. So one of the problems that genetics has is that everything we're looking at is so far in the future. If you're talking about someone restenosing after getting a stent, that's in a pretty soon event, that's relatively to what we're looking at with preventing a disease way down the road. Where I think this might be somewhat solvable is compared effectiveness research, which was spoken about at the last meeting, where you're looking backwards at data, where you have the genotype and look back at people's history along good phenotype information from whatever's happened in the medical record. That may be one way to get around this, but it requires a large amount of data and someone can actually go through it. Great, good points, Erwin, yes. If I may switch to... You can switch after Mara and us. Okay, but anyone else who wants to talk about the RCT issue, Rex? Yeah, I would just make the point that, and it was made earlier, but to reinforce in this case, the other paradigm difference that I think we're gonna be facing is, we may actually already have the genetic information in hand. At least many of us think that we're likely to have it. So, as Mary said earlier, I think in that case, it's a game changer because we're not asking anybody to go out and get a new test. We're simply asking them to act on some data they already have in hand. All right, thanks. So, Mara, take it. Great, thanks. Donna, thanks for the good talk. It was interesting. I was wondering with respect to family history and if that's gonna be a component of your educational program for, and including with regards to risk assessment for heart disease and stroke, but also in perhaps recognizing Mendelian forms of cardiovascular disease. Well, we have always listed family history as the first sort of first strike for any genetic disease. So, I'm not, I don't recall now if that was on that outline, but I'll talk to Karen and make sure that we do have that back. Good point. Great, Erwin. Yeah, Donna, great talk and I wanted to compliment you on focusing the educational aspects, not only on the physician provider, but also on the non-physician provider. And I think that's something that we still do not appropriately recognize. I can give you a comment from the ground up in New York City. We've recently engaged with a number of community health centers and primary care practice networks. And, you know, studying how the workflow goes there for care for most common chronic diseases that you're talking about in cardiovascular, in metabolic endocrine, et cetera, diabetes, et cetera. You know, the physician provider is really only a small part of the overall operation. And, you know, much of the care is actually delivered through non-physician providers. And I think that deserves considerable emphasis in our educational efforts going forward. We concur. Great, good point. Jean. Donna, you made the point that there are two active trials in relevant areas and that is a barrier to issuing a guideline. And I agree with that. Do you take it further and say that prior to doing a guideline, whether or not there's a clinical trial that we really need to wait? So there are other, from the American Heart Association's perspective, there are other ways of communication that are not guidelines. So there are scientific statements, which we have been advocates for producing and the second one should be coming out this year. For the, when you have trials that are currently in progress, we have taken a more conservative approach. I'll let Bill weigh in if he wants to wait until that evidence is coming. If you know it's coming, then it would be wrong to jump ahead and write a guideline with something so close to being completed. In the sense that you might be really wrong. We might be, you know, we might be. Joan, Scott. Yeah, I would like to just riff for a moment on a couple of the last concepts that came around, which is around generating the evidence base for practice, as well as a comment about the complexity and the level of the healthcare providers who are providing that practice. And not just the need to generate evidence for practice, but to generate practice-based evidence when it comes to educational efforts. So what we've been hearing as a common theme this morning is there's a very complex system. There's lots of genomic applications more increasing with all different levels of evidence around their use. And across the wide range of medicine in very complex healthcare systems with lots of providers who have different spinach levels, if you will, around their motivation for consuming this information and actually applying it in practice. And someone said this morning that there isn't a one-size-fits-all. And I think that's absolutely correct and what we need to be thinking about is generating the evidence about how to do this education in different clinical settings. And it may be that there's different levels different levels from point of care to traditional CME or some combination therein that's gonna be different for different practice settings and different providers for different needs. Just a bill on that, you know, years ago when the transformative R01s first came out, I put in an idea, which was triaged, but many were, I won't. So it may have been a great idea, but I wanted to partner with places like 23andMe and these direct-to-consumer companies to work with them to iteratively refine algorithms for treatment for pharmacogenetics or disease prediction that could be ongoingly updated as new data came on board as a new novel way of evidence generation. Because I do think we have to step away from the traditional ways if we want to move this more quickly into practice. Okay, any last comments? Not, we should probably have, I should have asked before, Bill and Robert to head up to the podium and we'll be hearing from the flip side of the heart disease community at this point now, the American College. The right side and the left side, that's right. The oxygenated, no, we won't go. No. No. Collaborative side. There you go. We're very collaborative.