Oct 16, 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 feature discussion this week centers on the temporal changes in natriuretic peptides preceding heart failure hospitalizations and patients at high risk. Data that are really novel and have implications for the way we perhaps monitor and categorize these high risk patients. Well, more soon right after these summaries.
The first original paper this week provides the first epigenome-wide association study in patients with heart failure. Now, epigenetics refers to biochemical DNA modification such as methylation of gene bodies, and post-translational modification of histones, which is increasingly recognized to play a crucial, regulatory interface between genes, environment, and the transcriptome.
The lack of availability of myocardial specimens from patients has been a major roadblock for elucidating the impact of such epigenetic changes on complex cardiovascular traits. However, in today's paper from first author, Dr. Meder, corresponding author, Dr. Katiz and colleagues from University of Heidelberg, Germany. The authors performed the first multi-omic study in myocardial tissue and blood of patients with dilated cardiomyopathy compared to controls.
They detected 59 epigenetic loci that are significantly associated with dilated cardiomyopathy, with three of them reaching epigenome-wide significance. 29 of these loci could be replicated in independent cohorts and authors further linked a subset of 517 epigenetic loci with dilated cardiomyopathy and cardiac gene expression.
Finally, they identified distinct epigenetic methylation patterns that are conserved across tissues. Thus representing novel, epigenetic biomarkers for heart failure.
The next study is the first to assess a diagnostic and prognostic value of cardiac myosin binding protein-C in patients presented with possible acute myocardial infarction or AMI. Cardiac myosin binding protein-C is a cardiac restricted protein that is more abundant than the cardiac troponins and is released more rapidly following AMI.
In today's paper, first author, Dr. Kaya, corresponding author, Dr. Marber and colleagues from the Rayne Institute In St. Thomas's hospital in London evaluated cardiac myosin binding protein-C as an adjunct or alternative to cardiac troponins in the early diagnosis of AMI in 1,954 unselected patients presenting to the emergency department with symptoms suggestive of AMI.
The final diagnosis of AMI was independently adjudicated in 340 patients. The authors found that concentrations of cardiac myosin binding protein-C at presentation were significantly higher in those with versus without an AMI. The discriminatory power for AMI quantified by the area under receiver operating curve was comparable for cardiac myosin binding protein-C to high sensitivity cardiac troponins T and I, and even superior to standard sensitivity cardiac troponin I. The use of cardiac myosin binding protein-C more accurately classified patients with a single blood test and to rule out or rule in categories in early presenters, meaning those with chest pain of less than three hours.
The improvement in rule in or rule out classification with cardiac myosin binding protein-C was larger compared with higher sensitivity cardiac troponins T and I. Finally, cardiac myosin binding protein-C was superior to high sensitivity and standard troponin I and similar to high sensitivity cardiac troponin T at predicting death at three years. Thus in summary, this paper shows that cardiac myosin binding protein-C at presentation provides discriminatory power comparable to high sensitivity troponins T and I in the diagnosis of AMI and may perform favorably in patients presenting early after symptom onset.
The next paper describes the discovery of a novel candidate cardiomyopathy or arrhythmia gene. First author, Dr. Barryfield, corresponding author Dr. McNally from Center of Genetic Medicine in Chicago and colleagues studied a family with dilated cardiomyopathy and associated conducted system disease in whom prior clinical cardiac gene panel testing was unrevealing. Whole genome sequencing however, identified a premature stop codon in the gene encoding a novel myo filament component, the myosin binding protein-H-like.
Having identified this gene, they turned to experimental approaches. The myosin binding protein-H-like gene was found to have high atrial expression with low ventricular expression. The truncated protein failed to incorporate into the myo filament. Human cell modeling demonstrated reduced expression of the mutant allele. Heterozygotes and nullumites exhibited a reduction in fractional shortening and increased diastolic ventricular chamber size, aberrant atrio-ventricular conduction and an increased rate of arrhythmia associated with the expression of the myosin binding protein-H-like in the atria, as well as in discrete puncta throughout the right ventricular wall and septum.
These findings therefore support that myosin binding protein-H-like truncations may increase the risk for human arrhythmias and cardiomyopathy.
Transplantation of cells into the infarctant heart has significant potential to improve myocardial recovery. However, low efficacy of cell engraftments still limits the therapeutic benefit. In today's paper, authors describe a method for the unbiased, in-vivo selection of cytokines that may improve Mesenchymal stromal cell engraftment into the heart. In this paper from first author, Dr. Bortolotti, corresponding author Dr. Giacca, and colleagues from University of Trieste in Italy, an arrayed library of 80 secreted factors were individually cloned into adeno-associated viral vectors.
Pools from this library were then used for the batch transduction of bone marrow derived Mesenchymal stromal cells ex-vivo, followed by intra myocardial cell administration in normal and infarctant mice. Three weeks after injection, the vector genomes were recovered from the few persisting cells, and identified by sequencing DNA barcodes that were uniquely labeled for each of the tested cytokines.
Using this novel, competitive, engraftment screening methodology, the authors identified that the most effective molecule was cardiotrophin-1 a member of the IL-6 family. Intra cardiac injection of Mesenchymal stromal cells preconditioned with cardiotrophin-1 preserved cardiac function and reduced infarct size parallel to the persistence of the transplanted cells in the healing hearts for at least two months after injection. Thus, preconditioning with cardiotrophin-1 might represent an efficient manner to improve the currently poor cell retention in patients treated with Mesenchymal stromal cell therapy.
The final paper presents results of the early myo trial, a non-inferiority trial comparing a pharmacoinvasive strategy with half-dose alteplase versus primary PCI in patients with STEMI, presenting six hours or less after symptom onset but with an unexpected PCI related delay.
First author, Dr. Poole, corresponding author, Dr. Hua and colleagues from Shanghai Jiao Tong University in China randomized a total of 344 patients from seven centers to a pharmacoinvasive arm or a primary PCI arm. They found that pharmacoinvasive strategy was non-inferior to primary PCI for the primary endpoint of complete epicardial and myocardial reperfusion after PCI defined as TIMI flow grade 3, TIMI myocardial profusion grade 3, and ST-segment resolution of more than 70%.
There was no significant differences in the frequency of the individual components of the combined endpoint. Infarct size and left ventricular ejection fraction were similar in both groups and there was no significant differences in 30-day rates of total death, re-infarction, heart failure, major bleeding events, or intracranial hemorrhage. However, minor bleeding was observed more often in the pharmacoinvasive group.
Thus the authors concluded that a pharmacoinvasive approach with reduced dose alteplase seems to offer effective and safe reperfusion in low-risk patients with STEMI with an unexpected PCI related delay. Further large, randomized control trials powered for clinical endpoints are needed.
Well, that wraps it up for your summaries. Now for our feature discussion.
The measurement of natriuretic peptides BNP, NT-proNP have certainly become the cornerstone of heart failure management. We measure these levels by guidelines in patients who are presenting with symptoms and suspected heart failure, in patients who are hospitalized. We measure them for prognostication purposes at discharge. However, what we don't really know is how the preceding changes in natriuretic peptides may precede heart failure hospitalization in patients who are at high risk of developing heart failure.
For example, patients with a recent coronary event or type-2 diabetes. And this is the very subject of our feature paper today, and I am so pleased to have the corresponding author of today's paper which is really a research letter. Dr. Brian Claggett from Brigham and Women's Hospital as well as Dr. Biykem Bozkurt who's our senior editor from Baylor College of Medicine.
Welcome both, and maybe I could start, Biykem could you let us know, what are the unanswered questions in heart failure relating to natriuretic peptides and how do you see this paper falling in, clinically?
Dr. Biykem Bozkurt: Carolyn, this is a wonderful I think prelude to perhaps preventing heart failure events. And as you are aware, we in the recent year changed our guidelines at the ACC, AHA, and the HFSA incorporating screening high risk patients for development of incident heart failure. And the study that resulted in this consideration was a STOP-HF trial which was utilizing natriuretic peptides in high risk patients to determine whether their closer follow up in a multidisciplinary fashion would result in earlier detection and prevention of heart failure, and which it did.
And this study I think is straddling the concept of high risk or stage A or B patients because they are individuals who have had heart attacks, coronary events, and they have type-2 diabetes so they are definitely high risk. And doing natriuretic peptides as an outpatient, whether that would predict the heart failure hospitalizations.
And in essence I think it's a good concept. Perhaps the challenging concepts are how often should we screen our patients, and what will be the threshold of the rise that would potentially make us act in either earlier diagnostic strategies, or management strategies. I think those are the two unanswered questions that remains.
How are we gonna screen our patients? Our high risk patients to determine when they are developing heart failure before they become symptomatic? So, what threshold are we going to use?
Dr. Carolyn Lam: That is a perfect set up. I just wanted to add as well in addition to STOP-HF there was the PONTIAC study in diabetics which is very relevant to today's paper that also sort of used NT-proBNP to risk stratify patients for prevention of heart failure. But neither of these studies talked about the temporal changes in natriuretic peptides. And I think a lot of the reason for that is, is that the methods, I mean the statistical methods to do that sort of thing are mind-blowing.
And so Brian, could you now please share with us what you did, the methodology and basically what you found before we discuss the two questions that Biykem brought up?
Dr. Brian Claggett: What was really interesting is the method that we came up with to look at these questions. It's something that we like to believe will be generalizable and can be used in other scenarios and for other biomarkers. But the idea that we have is that we are always used to thinking about the design of a clinical trial as being very regimented. So, you see a patient once at baseline, and then maybe six months later, and then maybe six months after that, and so on. And so it's hard to know what's going on, on a day to day or week to week basis.
But if you think backwards, and you think backwards from the time of any sort of event, because those events whether they're hospitalizations or MIs or death, they happen not on that same schedule. And so odds are at the end of a trial, you had a patient who came to a scheduled visit and then had an event the next day. And you probably had a patient who came in for a visit two days before an event, and another patient who came in a week before an event. So if you start thinking on that time scale, you can piece together all these different time frames when you do have data collected and try to reconstruct something that looks like an actual continuous natural history of what that biomarker would have looked like over say a two year period, if it had been measured continuously.
Dr. Carolyn Lam: So, tell us what you found. First of all, let's just make sure that everyone knows you were looking at the ELIXA cohort, right?
Dr. Brian Claggett: Yes, the data that we had available for this analysis comes from the ELIXA trials, it was 6,068 patients all with type-2 diabetes and a recent ACS event. Recent meaning within the last 180 days. And they were randomized placebo versus a diabetes drug, lixisenatide. And they were followed up for cardiac outcomes.
Beyond that, the natriuretic peptides were measure systematically at baseline, month 6, month 18, and month 24 in all patients who were participating in the trial. So this was the richest collection of a large number of patients being measured multiple times, systematically and not in just a sub-sample of the population. So, we felt like this was a great opportunity to learn something about what happens. What can you learn when you measure these natriuretic peptides over and over again.
And even more interesting than that, the fact that this wasn't a heart failure trial meant that some of the patients already had heart failure at baseline. Other patients didn't have heart failure, but as the trial went on, they developed or were hospitalized for heart failure for the first time. And so we were able to also look at differences between patients experiencing their first heart failure, versus those with more long standing disease.
Dr. Carolyn Lam: And that was very, very unique methodology that you spoke about. And I fully agree that it's going to be used more. I am staring at your beautiful figure one right now. That really, really says it all. Could you walk us through the results?
Dr. Brian Claggett: Sure, I think our key finding is that, I guess no matter when you measure patients. Patients with a higher level of NT-proBNP, or a higher level of BNP at any given time are going to be at higher risk of developing heart failure in the future.
But as we start looking at this as a temporal process, what we see is that there seem to be a noticeable acceleration in these increases, specifically in the last six months before development of heart failure. Or, before a hospitalization for heart failure. And that increase in the final six months seems to occur both in patients who had no prior history of heart failure and also in patients with a history of heart failure. So that six month window I think is something that we learned that we didn't necessarily know before.
Dr. Carolyn Lam: But, going back to Biykem's questions, do you think we have answers to how often we need to survey natriuretic peptides in these high risk patients and what threshold we need to act on?
Dr. Brian Claggett: I think both are very important. I think maybe the timing and the thresholds are somewhat separate questions. I think we're better able to answer the timing question. At the very least we can say that if dramatic changes are happening over a six month window that measuring patients only once every six months probably isn't enough. Whether that means it needs to be every three months, or two months, or one month, or something more than that, I think it's hard to know exactly what the right answer is. But I think we are confident in saying that things happen relatively quickly and we need to be measuring these things more frequently.
As far as the question of thresholds, I think that's maybe even a more difficult question. Or even the idea of a threshold means that we think that there's some magic number and I am not sure that we know for sure what's more important, the absolute number or is it the ... if someone starts relatively low and that relatively low number doubles over the course of six months. That might still be prognostically just as important as someone who's been consistently edging just below or just about that threshold level.
I'm not sure that we're confident enough to say that the changes, the speed of the changes, or the relative changes, or some absolute threshold is the most important thing to be paying attention to. But, I think where these two are related is the more ... that we can start to collect this data more frequently and be able to analyze it. I think that gives us a lot better chance of being able to successfully answer that question about thresholds.
Dr. Carolyn Lam: Indeed. Stuff for future work, huh? Biykem, what do you think?
Dr. Biykem Bozkurt: I wanted to point out two things from Brian's study which was quite interesting. One is the trajectory of the rise, or the delta changes in the natriuretic peptides was quite different in the patients with no history of heart failure compared to those with a history of heart failure. The trajectory, or the linear rise, or the delta changes were more prominent in the individuals with no history of heart failure. Probably intuitively expected so because their baseline levels are not as high as the individuals with history of heart failure.
So, it almost gives the impression that maybe in low low risk, the screening or the frequency may need to be lower, and if low, then probably the likelihood of the rise may be less. But those individuals who, as you said, are edging upward, then maybe the frequency may need to be higher and there may be perhaps a linear rise or a more prominent rise about six months before the incident event.
So, it's an interesting concept just to look at people's trajectories. But, as you said, probably individualization and monitoring or targeting may need to be individualized according to personal risk and other features. And one then wonders futuristically if this would be a concept that would be point of care testing maybe done by the patients similar to glucose monitoring. And in the event that we were to be able to carry the platform to self-test.
Dr. Brian Claggett: You're talking to a statistician, so I am always going to be in favor of collecting more data all the time. So I agree with that.
Dr. Carolyn Lam: Wow, what an insightful discussion. Thank you both for joining us on this podcast today.
Ladies and gentlemen out there, you heard it right here in Circulation on the Run. Tune in again next week.