Mar 5, 2018
Dr Carolyn Lam: Welcome to Circulation on the Run. Your weekly podcast summary and backstage pass to the Journal and its editors. I'm Doctor Carolyn Lam, Associate Editor from the National Heart Center and Duke National University of Singapore. This week's journal features an international external validation study of the 2014 ESE Guidelines on Sudden Cardiac Death Prevention in Hypertrophic Cardiomyopathy. A very exciting discussion coming right up after these summaries.
The first original paper this week suggests that proteomics, a tool of precision medicine may prove useful in improving the safety and efficiency of drug development. First author, Doctor Williams, Corresponding Author, Doctor Ganz, from the Zuckerberg San Francisco General Hospital retrospectively applied large scale proteomics to blood samples from Illuminate, the trial of Torcetrapib, a cholesterol estrotransfer protein inhibitor, which raised HDL and lowered LDL cholesterol. Recall that this trial was terminated due to increases in cardiovascular events and mortality.
In the current study, the authors found that plasma concentrations of 200 proteins changed significantly with Torcetrapib. Their pathway analysis revealed unexpected and widespread changes in immune and inflammatory functions, as well as changes in aldosterone function and glycemic control. A previously validated nine protein risk score was similar in the two treatment arms at baseline, but higher in participants with subsequent events. At three months, the absolute nine protein derived risk increased in the Torcetrapib plus Atorvastatin arm compared to the Atorvastatin only arm. Thus, this protein-based risk score predicted harm from Torcetrapib within just three months. A protein-based risk assessment embedded within a large proteomic survey may prove to be useful in the evaluation of therapies to prevent harm to patients. This is discussed in an accompanying editorial entitled "Harnessing the Power of Proteomics to Assess Drug Safety and Guide Clinical Trials" by Doctor Maggie Lam and Ying Ge.
The next study suggests that personalized monitoring of heart transplant outcomes may be achieved by profiling the genetic and phenotypic markers of the CD16-dependent natural killer cell activation pathway. First and corresponding author Dr. Paul from Vascular Research center in Marseilles in France and his colleagues collected blood samples from 103 patients undergoing routine coronary angiography for cardiac allograph vasculopathy diagnosis, a median of five years since their heart transplantation. They used a non-invasive natural killer cellular-humoral activation test to evaluate the association between genetic and phenotypic markers of the CD16 dependent natural killer cell activation pathway. They showed that the Fc-gamma receptor IIIAVV polymorphic variant, which encodes the highly responsive CD16-Fc receptor, was an independent baseline predictor of cardiac allograph vasculopathy, and may be useful for stratifying patients at higher risk of rejection. The implications of these findings also include the fact that individualized natural killer cell targeted therapies may limit vascular damage in responsive patients.
The next study suggests that estimation of polygenic atrial fibrillation risk is feasible, and together with clinical risk factor burden, can explain lifetime risk of atrial fibrillation. Co-first authors Dr. Weng and Preis, corresponding author Dr. Lubitz from Massachusetts General Hospital, and colleagues estimated the lifetime risk of atrial fibrillation in individuals from the community-based Framingham Heart Study. Polygenic risk for atrial fibrillation was derived using a score of approximately 1000 atrial fibrillation-associated SNPs. Clinical risk factor burden was calculated for each individual using a validated risk score for incident atrial fibrillation comprised of height, weight, systolic and diastolic blood pressure, current smoking, anti-hypertensive medication use, diabetes, history of myocardial infarction, and history of heart failure.
They found that the lifetime risk of atrial fibrillation after age 55 years was 37 percent was substantially influenced by both polygenic and clinical risk factor burden. Among individuals free of atrial fibrillation at the age of 55 years, those in the low polygenic and clinical risk tertiles, had a lifetime risk of 22 percent, whereas those in the high risk tertiles had a risk of 48 percent. Atrial fibrillation developed at an older age among individuals with a favorable clinical risk profile regardless of genetic predisposition. Nevertheless, the lifetime risk of atrial fibrillation in individuals with high genetic predisposition was substantial, even when the clinical risk factor burden was low. Thus, individualized projections of lifetime risk of atrial fibrillation may be refined by accounting for both genetic predisposition and clinical risk factor burden.
The final study tells us that in contrast to previous perceptions, Takotsubo cardiomyopathy has long-lasting clinical consequences. First and corresponding author Dr. Skally from University of Aberdeen in the UK and their colleagues did an observational case controlled study of 37 patients with prior Takotsubo cardiomyopathy and 37 age, sex, and co-morbidity matched controls. Although Takotsubo cardiomyopathy occurred 20 months before the study, the majority of patients had persisting symptoms compatible with heart failure and cardiac limitation on exercise testing. Despite a normal left ventricular ejection fraction in serum biomarkers, patients with prior Takotsubo cardiomyopathy had impaired cardiac deformation indices on echo cardiography, increased native T1 mapping values on cardio magnetic residence imaging and impaired cardiac energetic status on p31 spectroscopy. Taken together, these findings demonstrate that after Takotsubo cardiomyopathy, patients appear to develop a persistent long-term heart failure phenotype.
Well that wraps it up for our summaries. Now for our featured discussion.
Sudden cardiac death prevention and hypertrophic cardiomyopathy. Always such an important topic. I'm so pleased to have with us the author from our featured paper this week, Dr. Perry Elliot from University College London, nd our associate editor, Dr. Mark Link from UT Southwestern who also wrote a beautiful accompanying editorial with Tera Lynn Ho. So welcome both of you. Perry, I think to set us up, I'd really love if you could tell us a little bit more about the 2014 ESE guidelines for sudden cardiac death prevention and hypertrophic cardiomyopathy. And particularly pointing out how they may differ from the 2011 ACC AHA guidelines please.
Dr Perry Elliot: So, the 2014 guideline on sudden death prevention HCM, the aim of that guideline was to try to quantify the risk of sudden cardiac death. As you pointed out, sudden death is a significant complication of hypertrophic cardiomyopathy and one which we all as clinicians spend a lot of time trying to determine. If we look back over at, I don't know, a period of twenty, thirty years the approach we've developed is based upon the recognition of a number of clinical features of the disease that we know associate with a higher risk of sudden death. So things such as, you know, unexplained syncope or severity of hypertrophy. And it was that model of sort of taking those so called major risk factors which form the basis of the 2011 US guidelines and the essential model was the more of those things you have, the greater is your risk, and I suppose the higher indication for an ICD.
One of the problems with that approach was that it's not quantitative so you know, you could say, "Okay. Well I think you're at higher risk, but I can't say how much that risk is." And another problem with that way of doing things is when you start to think about some of the individual risk factors, it doesn't make a great deal of sense clinically.
And I suppose a good example of that is wall thickness. You know we have this magical number of 30 millimeters, above which we say you're at risk, but of course are we really saying that if your wall thickness is 29 millimeters you're at low risk? We know it doesn't really work that way in biology. So when we drew up the 2014 guideline we wanted to say, "Okay let's develop a model in exactly the same way that we do with atrial fibrillation or primary prevention in coronary disease so that we can say to the patient sitting in front of us, 'Based on your clinical assessment we think you've got a one, five, ten percent risk of something bad happening to you in the next five years.'" And then we can use that information to inform our decision about ICD implantation.
The model itself is not so revolutionary. I mean, it uses a lot of the conventional risk factors such as wall thickness, such as non-sustained VT on Holter monitoring, but what it did introduce was the factor of age, because we know that the age of the patient certainly determines their risk. We brought in [inaudible 00:10:12] obstruction because we've now got reasonable evidence showing that if you've got a big gradient, that certainly modulates your risk. And also probably for the first time, I suppose, left atrial size, which was one of those missing things I think in previous assessments. You know, it's a fantastic surrogate for restrictive physiology and certainly when we added it to the model it improved the predictive power of that model.
So I suppose in summary what we've done is to produce a tool which allows you to estimate risk and then use that to help you decide on whether an individual needs a defibrillator in the clinic.
Dr Carolyn Lam: You know Perry I believe you led those guidelines and I just want to congratulate you as well as that was such a beautiful explanation of what was going on behind those. Yup, but the proof is in the pudding isn't it? But you're providing that proof in today's paper. Tell us about it. So it's an external validation, a large international multi-centers study to actually validate these 2014 guidelines.
Dr Perry Elliot: That's right. I mean, I think when we generate these kind of models it's really important to test those models in different settings. The original model was based upon a relatively small number of European centers and I think what this paper does is it brings insights into different geographies and different health care systems. So we have participating centers from North America from the Middle East from the Far East and the idea here is to get as diverse a population as we can and just see if the model performs in the same way. And you know in a study just short of 4000 people, I think that we've shown that the model does indeed seem to behave in the same way. In fact, the numbers were remarkably similar. You know the ability of this model to discriminate between high and low risk patients was almost exactly the same as in the original paper, which I think gives us a level of reassurance that this model, this tool that we've developed, can be used in different health care settings.
Dr Carolyn Lam: Mark. I really enjoyed your editorial. I love the way that you started out with a case that really shows why this is so important. And I also love that you discuss some other studies that tried to validate the 2014 ESE guidelines as well. Could you just give us some of your thoughts there.
Dr Mark Link: Yeah. I first want to congratulate Perry and his fellow authors for this paper. I think it was a very nice paper. I was a champion of this paper from the time it got sent into circulation. And, you know, the big change in the 2014 European guidelines compared to the American guidelines is really the linear risk of age, wall thickness, and I'll put tract gradient. And as Perry says, I agree, it's not a simple you have it or you don't, it's a linear risk and I applaud them for including that in their risk factor stratifier. And if you look at the current paper, I mean it was very good at picking out high risk patients. So if you have greater than a four percent, six percent risk over five years, you did. And so for picking out the high risk patients it was very good. And for picking out the medium risk patients, it didn't function as well. It was best for separating out the high risk and the low risk population.
And I will say, based on this paper, I've started using the European risk stratifier in my clinical practice. So I do want to applaud them, you know, for the risk stratifier tool and this paper. But I do want to say, and I'm sure Perry will agree, that we're not there yet. We need better tools, because not only in this data set but in other data sets, because more of these individuals reside in the low risk population, more of the sudden deaths are in that population. And we need better tools. And over time they will come. You know, they may be MRI tools. They may be scar tools. They may be other tools that we aren't even aware of that are coming on the horizon, but we do need better tools as we move forward to identify those at risk for sudden death in hypertrophic cardiomyopathy.
Dr Perry Elliot: Yeah. No. I agree. I mean I think what I would say is if you sort of take a step back and look at the overall perspective in this paper, despite the fact that, you know, we've got nearly 4000 people and they're followed in different health care settings, the overall sudden death rate in hypertrophic cardiomyopathy pretty low. You know, so that's good for patients 'cause I think it shows that at least in managed populations, the risk of sudden death which is real and we've got to assess it, but I think it's really important to get that message over to patients that for most people with HCM they're at low risk. It is of course the challenge because when you're dealing with rare events, it's really hard to predict them. And this model is far from perfect. I would argue it's probably the best we've got for the moment, but you know, it's not that bad. It's not that bad.
I mean agree with you absolutely Mark, 'cause you know, either end it performs pretty well. In the middle there it's not as predictive, although what it tends to do is overpredict, rather than underpredict. So you know, I think if you use this model in your every day practice just the greatest risk is that you'd end up putting in probably more ICDs then you really need to rather than missing a lot of patients. You know, we really want to prevent every sudden death if we possibly can, but that's always going to be really hard I think and I think the fight goes on. We got to look for new risk predictors. It may be that we can interchange some of these predictors. They might be easier to assess in some practices, but I'm not a born optimist, but I really think it's amazing just how well in such a complex heterogeneous disease that this relatively simple assessment works. You know?
Dr Mark Link: Going forward, what do you think the future of HCM [inaudible 00:15:47] stratification will include? We've got the risk stratifiers in your calculator ready. And more specifically where do you think gadolinium enhancement will play a role in the future? So MRI findings of scar or gadolinium enhancement.
Dr Perry Elliot: The base we have at the moment show that the more scar you have, the greater risk of sudden death. It sort of makes sense, doesn't it? It's part of that substrate for ventricular arrhythmia. My own reading of it just so far is that I'm not sure what it adds to the existing way of doing things. I mean I think this is true of any biomarker, you know. I've got a new biomarker, what does it tell me that I don't already know? And with scar, we know the greater amount of scar, it often tracks with wall thickness. You're likely to have a thicker heart, you're more likely to have non-sustained VT. But I'm openminded on that front. The beauty about this model, for me, is that it's a tool to into which you can plug other things and you know, if we can get big enough data sets and we can use gads and the amount of scar and put that into the model and if it improves the performance of the model that's great. Those studies are underway at the moment and I think we eagerly wait the results of those studies.
For me, one of the missing things is the genetics. This is a heterogeneous disease with quite a complex genetic architecture, and despite the fact that you know it's 20, 30 years now since we identified the first gene, we haven't really factored that in to our risk models and I think that for me is one of the big challenges and opportunities over the coming years is to put together really large international data sets so that we can answer once and for all whether your mutation determines your prognosis.
Dr Mark Link: Yeah. I agree with the genetics also I think getting more information on that. And it’s been 30 years it still is not helping us prognosticate the risk of sudden death, but it should. I mean it really should. And I do think hopefully we will find other tools also as time goes on because it really is imprecise and it's very difficult when you're sitting there in front of ... You know, I just had a 20 year old yesterday come in with his family and he's got a three centimeter septum and he's got 12 percent scar and he's saying, "Gee what would you do and what would you do if I were your son?" And it's easy when you're looking at it in the aggregate. It's much more difficult when you're sitting there one on one with a patient in front of you.
Dr Perry Elliot: Of course. Of course. And I think another factor I think which is changing the dynamic of that kind of discussion is the evolution of ICD technology. You know, I think when you're dealing with young people the fear is long-term complications with leads isn't it? And I think with the advent of the SICD I sense it's already tipping the balance into perhaps a slightly more liberal approach to ICD implantation exactly in the kind of scenario you've just described Mark, you know you've got guy who's 20. He's got a really severe hypertrophy. Well you know, if you and mess ICD you know your threshold for implantation might be a bit lower.
Dr Mark Link: Yeah and in fact, after a two hour discussion that's what we decided on is that subcu ICD was the right thing for him. And everyone's very happy with that choice.
Dr Perry Elliot: Yep. I think it also raises another thing which I often think about is that as medics we're also probably not good at considering what acceptable risk actually is. You know? We develop models in different settings and hyeprtrophic cardiomyopathy, coronary disease, heart failure, and actually if you go back and critically look at the thresholds that are used to put in defibrillators, the absolute risks vary enormously. So you know, here in [inaudible 00:19:02] we're talking about an approximate annual risk of sudden death of about one percent per annum is sufficient to put in an ICD, but in long-QT world it's quite a different threshold that's used and of course that's because there is no defined number. You know the number's we used in the ESE model of greater than six percent you should have an ICD, well yeah that's the consensus number, there's nothing magical about it. There's nothing biological about it. And I think we've probably had greater debates at what acceptable risks really are.
Dr Mark Link: And that's become a big shared decision now in the States and actually everywhere. It's become a big word because it sued to be that the physicians would decide on who gets an ICD and who doesn't. And it's no longer that way it's a discussion with the patient, with the family. How much risk are they willing to take, both with an ICD and without an ICD, because there are issues with ICD, even though I'm a big fan. There are issues and especially with transvenous ICDs, but also with subcutaneous ICDs.
Dr Perry Elliot: Absolutely. Absolutely.
Dr Mark Link: You know, it's a different world now than it was 15 20 years ago.
Dr Carolyn Lam: Perry and Mark, this has been one of the most wonderful conversations I've had on these podcasts. I just can't thank you enough. I'm sure all our listeners are thanking you too. You've been listening to Circulation on the Run. You must tune in again next week for more beautiful conversations.