Sep 30, 2019
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.
Dr Greg Hundley: And I'm Greg Hundley, associate editor at Circulation and director of the Pauley Heart Center at VCU Health in Richmond, Virginia. Carolyn, have you ever wondered about instead of coding a stent, coding balloons with paclitaxel? Well, the feature article day is going to look at mortality assessments of paclitaxel-coated balloons in a meta-analysis from the ILLUMENATE clinical program, the three-year outcomes. Do you have a paper you want to start us off?
Dr Carolyn Lam: I sure do. First of all, we know that diabetes impairs atherosclerosis regression following cholesterol lowering in both humans and mice. Now in this process of plaque regression, what's the role of functional high density lipoprotein or HDL, which is typically low in patients with diabetes?
Well, this first paper that I chose looks just at that and it's from Dr Fischer from New York University School of Medicine and colleagues, who aimed to test if raising functional HDL levels in diabetic mice prevents monocytosis, reduces the quantity and inflammation of plaque macrophages and enhances atherosclerosis regression following cholesterol lowering. So to do this, the authors used aortic arches containing plaques, which were developed in LDL receptor null mice, and these were transplanted into either wild type or diabetic wild type or diabetic mice transgenic for human APL lipid protein A1, which have elevated functional HDL.
Dr Greg Hundley: So Carolyn, what did they find in this interesting study?
Dr Carolyn Lam: Well, diabetic wild type mice had impaired atherosclerosis regression, which was normalized by raising HDL levels. The benefit was linked to suppressed hyperglycemia-driven myelopoiesis, monocytosis and neutrophilia. Increased HDL improved cholesterol efflux from bone marrow progenitors, suppressing their proliferation and monocyte neutrophil production capacity. ACL also suppressed the general recruitability monocytes to inflammatory sites and promoted plaque macrophage polarization to the M2 phenotype, which is an atherosclerosis resolving state. There was also a decrease in plaque neutrophil extracellular traps or nets, which are atherogenic and increased by diabetes. So raising apolipoprotein AI and functional levels of HDL promoted multiple favorable changes in the production of monocytes and neutrophils and in the inflammatory environment of atherosclerotic plaques in diabetic mice after cholesterol lowering. And this may represent a novel approach to reduce cardiovascular risk in patients with diabetes.
Dr Greg Hundley: Really interesting, Carolyn. Well, I'm going to talk to you a little bit about a large study in patients with valvular heart disease and it's a contemporary presentation and management study and it's from the Euro Observational Research Program Valvular Heart Disease II, Roman numeral two, survey. And the corresponding author is Professor Bernard Iung from Bichat Hospital. So the VHDII survey was designed by the Euro Observational Research Program of the European Society of Cardiology to analyze actual management of valvular heart disease and compare practice with guidelines.
Now in short, patients with severe and native valvular heart disease or previous valvular intervention were enrolled prospectively across 28 countries over a three-month period in 2017. Indications for intervention were considered concordant if the intervention was performed or scheduled in symptomatic patients corresponding to class one recommendation specified in the 2012 ESC and in the 2014 American Heart Association American College of Cardiology valvular heart disease guidelines.
Dr Carolyn Lam: Wow. So what did they find, Greg?
Dr Greg Hundley: Okay, so there's 7,247 patients. 4,483 were hospitalized, and 2,764 were outpatients, and they were included across 222 centers. The median age was 71 years and 1,917 patients were over the age of 80, and 3,400 were women. Now, aortic stenosis was present in 2,000 plus patients, aortic regurgitation in 279, mitral stenosis and 234, mitral regurgitation in 1,114. And multiple left-sided valvular heart disease was present in 1,297, right-sided valvular heart disease in 143, and 2,028 patients had prior vascular intervention.
So the decision for intervention was concordant with class one recommendations in symptomatic patients with severe single left-sided valvular heart disease in 79.4% of those with AS, 77% with aortic regurgitation, 68.5% for mitral stenosis, and 71% for primary MR. Valvular interventions were performed in 2,150 patients during the survey. Of them, 47.8% of the patients with single left-sided native valvular heart disease were in New York Heart Association class three or four, and transcatheter procedures were performed in 38.7% of the patients with AS and 16.7% of those with MR.
Dr Carolyn Lam: Wow, Greg. So what are the take home messages? That was a lot of numbers.
Dr Greg Hundley: Yep. Lots of data there. And so couple things. First, recommendations for interventions in symptomatic patients with severe valve disease are better applied today in this paper than in the previous European survey conducted in 2001, particularly for those individuals with aortic valve disease. Second, multi-modality imaging is now more frequently used, but stress testing remains underused in asymptomatic patients. And finally, transcatheter therapies are now widely used in patients with stenotic valve disease, and we would expect that, particularly for the use in the elderly.
Dr Carolyn Lam: Great, Greg. So what are the clinical implications?
Dr Greg Hundley: Okay, so Carolyn, first, late referral for intervention shows the need for increasing awareness of valvular heart disease by general practitioners and cardiologists. Second, the high burden of elderly patients highlights the need for multidisciplinary heart team approaches to assess the risk benefit ratios of the different modalities of valvular interventions. And finally, number three, echocardiographic quantification of regurgitation should be more accurate and pay more attention to quantitative measurements. Those are the main take homes from this large registry analysis.
Dr Carolyn Lam: Nice. Thanks, Greg. My next paper is the characterization of the first transgenic mouse model of ARVC 5. Now, that is the most aggressive form of arrhythmogenic right ventricular cardiomyopathy caused by a specific mutation in transmembrane protein 43. So this paper's from co-corresponding authors, Dr Lara-Pezzi from CNIC in Madrid and Dr Garcia-Pavia from Hospital Universitario Porto de Hero in Madrid, and with their colleagues, they generated transgenic mice over expressing transmembrane protein 43 in either it's wild type or that specific mutant form in postnatal cardiomyocytes under the control of alpha-myosin heavy chain promoter.
And they found that these transgenic mice expressing the specific mutant in transmembrane protein 43 showed fibro fatty replacement of the myocardium and died at a young age. The model confirmed that transmembrane protein 43 is mostly localized at the nuclear membrane and provides new information regarding the pathophysiological mechanisms underlying ARVC five. One of them is that the GSK3 beta signaling pathway plays an important role in this disease.
Dr Greg Hundley: So that's great, Carolyn. Sounds like we have a new model that's been created by this group and certainly this disease has spread. It's something we definitely worry about. Do you see any therapeutic implications for their work?
Dr Carolyn Lam: Great question, and indeed the authors tested two new therapeutic approaches for ARVC five. In the first they found that targeting fibrosis really had no beneficial effect. But in the second, they found that inhibition of GSK3 beta improved cardiac function and survival, thus opening the way to a new therapeutic approach focused on GSK3 beta inhibition in patients with ARVC five.
Dr Greg Hundley: Very good. So we look forward to seeing what the results of that study will be. How about now we talk about some of the other articles in this issue?
Dr Carolyn Lam: I love that. I think it's a great idea to tell everybody about this amazing issue. So we start with an article from our Global Rounds, and this time from Argentina, so a great status update and future strategies for cardiovascular disease in Argentina. We also have a perspective paper and that's on the new World Symposium on Pulmonary Hypertension guidelines, really questioning some of the cutoffs that we've taken for granted and asking, "Should 21 be the new 25?" Intrigued? Well, you really need to pick this one up and read it.
And then there's a white paper, and this is a report from the 2018 NHLBI workshop that really talks about unlocking the secrets of mitochondria in the cardiovascular system and asking if this may be a path to cure in heart failure. We also have a research letter, and I love these. They're so succinct and really contain an important message. And this one talks about the evolution of Medicare formulary coverage changes for antithrombotic therapy after the guideline update. So very topical subject.
Dr Greg Hundley: Very good, Carolyn. So I've got a couple. There's a Paths to Discovery article that John Rutherford did discussing with Paul Zimmet regarding reflections of the evolving global diabetes epidemic. Second, there is a very nice On My Mind piece from Samuel Tretheway from Birmingham, England who discusses medical misinformation, kind of like medical fake news. And he discusses how this occurs and it depends on the motivation of both authors and publishers, and he reviews responsibilities of all of us, how to avoid generating this type of material. And then finally, a really interesting Cardiology News piece by Bridget Kuehn, who discusses diet and microbes in heart failure, and with that there's a very nice piece of artistry work that would be great for your office. So that's all included in the journal.
Dr Carolyn Lam: Oh, you got us all curious. Finally, I just want to highlight, we have a section called Highlights from Major Meetings, and this time from my part of the world with Dr Aijun Sun and Dr Junbo Ge summarizing the 13th Oriental Congress of Cardiology takeaways. Cool issue, isn't it?
Dr Greg Hundley: Absolutely. So how about onto our feature discussion?
Dr Carolyn Lam: You bet, Greg.
Dr Greg Hundley: Welcome everyone to our feature discussion. And this afternoon or this morning, wherever you may be, we are going to have an opportunity to discuss the utility of paclitaxel-coated balloons in terms of management of patients with peripheral arterial disease. And our article today comes to us from Bill Gray and colleagues from Mainline Health in Philadelphia, Pennsylvania. And we have our own Josh Beckman, associate editor from Vanderbilt, who will be joining us in the discussion. Bill, welcome to Circulation. We really appreciate you sending us this article. Can you tell us a little bit about the background of why you wanted to perform your study and also, what was your study design, study population?
Dr William Gray: The study was really prompted by a prior report by Katsanos et al in JAHA about nine months ago. When we started this study, it was much more fresh. And what we did was we realized we had data from multiple studies using the Stellarex drug-coated balloon that we could use to address some of the issues raised with the Katsanos paper. Just to review that briefly, the Katsanos paper suggested that there was a significant mortality signal in patients who were randomized to drug-coated balloons using paclitaxel versus PTA or patients randomized to drug eluting stent versus PTA or other stents. That signal was seen late at two years and at five years, and so we sought a given the data, the tightly controlled and well-reported data and this experience to see if we could see a signal as well.
The study design really involved taking all the data from the randomized trials, and there were two, which comprised an aggregate of about 600 patients, unequally randomized, about 400 in the drug-coated balloon arm and about 170 or 200 patients in the PTA arm. And then we also looked at all the poolable data, which was controlled data, so we had two randomized control studies I mentioned just a minute ago, as well as three single arm studies in one registry. Now, these had quality oversight and data reporting. And then those data were adjudicated for adverse events, including death, by a blinded third party CEC, and then those data reported out by Kaplan–Meier estimates as well, and then we do a multi-variable analysis looking at predictors of death, and then I can talk about that in a moment. Importantly, the data here has followed out to three years. As I mentioned before, the original paper which incited the concern had reported unequal deaths at two and five years, so we're somewhere splitting that difference. That's the genesis of the study and the study design.
Dr Greg Hundley: So Bill, tell us now about the results.
Dr William Gray: It turns out the baseline characteristics were largely similar between these trials and the patient arms, even though they weren't strictly speaking the same trials, except that the drug-coated balloon arm was a bit younger and smoked more frequently, so they were at a little bit more risk. In the randomized control analysis, which was done first, there was no difference in all-cause mortality between the PTA patients and the patients who received paclitaxel drug-coated balloons. That was true at one year, two years and three years. When we looked at the pooled analysis, which included not only the drug-coated balloon randomized trial patients, but also all the single arm studies and registries, we also found that there was no differences between those treated with drug-coated balloons in those additional studies and the control group of 170 patients in the randomized trial arm of PTA alone.
Interestingly, when we started to look at the multi-variable analyses, we did something that we ordinarily would not do, but because of the pressing issue around paclitaxel mortality, we actually did a standard covariate analysis looking at predictors and then we forced drug and drug dose into the model to see if they would come up positive as a predictor of outcome. As you might expect, not surprisingly, we found that age, congestive heart failure, diabetes and renal insufficiency were the four major predictors of mortality in a group of patients who were largely claudicates with significant peripheral vascular disease. No surprise there. We all know the patients don't die of claudication, they die of cardiovascular disease, and this I think bears that out.
When we force drug into the model, in point of fact, not a dose nor the presence of drug had any impact on death rates in the model, so there was no predictive value there whatsoever. Those are the results. Again, they're out to three years, and I think one of the important things that we have to recognize is that the numbers are relatively small and the follow-up is relatively limited and by itself, although it doesn't show any signal, it probably doesn't stand on its own to refute a larger meta-analysis, but does I think contribute to the dataset that is becoming more evident that the individual analysis do not appear to show mortality effects.
Dr Greg Hundley: Very good. So this is Dr Josh Beckman at Vanderbilt University. Josh, could you talk to us a little bit and put this paper in perspective relative to the prior published literature in terms of how you manage patients with peripheral arterial disease?
Dr Joshua Beckman: I have to say first, I'm really glad that we're able to publish this paper from Bill Gray and his group. We are, and I'm going to put this in really muted terms, in extraordinary times. I have never seen what is going on now happen with any other technology or really even medical therapy in the 20 plus years I've been a practicing physician. I think for the audience, it's really important to understand what is going on right now because if you don't pay attention to this space, you may not realize what's really been happening. Bill did a nice job at telling you why he did the study, which was this Katsanos aggregate level meta-analysis that was published in JAHA back in December.
On the basis of this paper, there has been a rapid development of worry and concern that these devices may be associated with late mortality. This concern has spread to the Food and Drug Administration, which has now put out three letters to healthcare professionals, each of them basically suggesting that you should choose non drug-coated either balloons or stents first, and if you want to use these, you have to have an extended conversation with the patients discussing the risks. And so in response to this aggregate level meta-analysis, which had an extensive number of lost to follow-up patients and didn't account for crossovers and the usual problems with this kind of information, I have been really impressed by the community of people who are interested in this topic and work with these kinds of devices.
And by that, I mean, the response has not just been a series of editorials. The response has really been, "Let's find every single piece of data that we can find to see whether or not this signal holds up," because as evidence-based physicians, we take one piece of data and say that it is one piece of data, and then we have to put it into the context of all of the other pieces of data that were published. And so I know that Dr Gray is old enough to remember 10 years ago when these devices were being used in the coronary arteries with drug eluting stents. And as far as anybody can tell with studies that were two to three times larger or meta analyses two to three times larger than the study published in December, there was no mortality signal.
It should be made clear that in doses that dwarf the doses from these devices, when these medications are given to pregnant women who have breast cancer, not only is the mother fine but the fetus is fine. And so I think paper that we are discussing this morning in particular, but the group of investigators in the space has really stepped forward to publish as much data as possible to fill out our understanding and place the original study in the correct context. And so when you understand what's happening in the community, and there's been a significant reduction in the use of these devices on the basis of that one publication at the expense of patients for whom these devices are really much better at limb outcomes, then you can understand why we were so interested in the paper by Dr Gray.
This is another brick in creating the foundation to really have a fuller and better understanding of any possible relationship between the use of these devices and a nonspecific increase in mortality two to five years later, which as far as I can tell, I've never seen something that may end up being a poison that doesn't have a specific mechanism of causing morbidity or mortality. And so when we got this paper, I was really happy to be able to work with Bill and bring it to the level that it is now so that when it's published in October, it's going to be another really important contribution and I just want to congratulate the authors for doing that work. I will say, and I'd like to get Bill's perspective on how he thinks the information that's now being published is going to help us understand what to do with these devices.
Dr William Gray: Yeah, that's a great question, and I want to emphasize something you brought up, which I did not, which is at the aggregate level data that Katsanos used to publish his analysis was really all he had access to, which means that he had some numerical data from prior published publications but did not have patient level data. And so what Josh is referring to appropriately is the concept that each individual holder of those data, those patient level data, are now coming forward with their own analysis of those data at a patient level, which allows us to look more granularly and more clearly at the causes of death. For example, in this study, the causes of death did not cluster around cancer. They were largely cardiovascular, and they were not dis-equally distributed or unequally distributed between the two groups.
So I think that patient level data, to get back to your original question, Josh, the patient level data will be incredibly important from each of the experiences with the various drug-coated balloons and drug eluting stents on the market because it does allow us to look more closely at the mechanism of death and whether there's any putative cause that might be assigned to paclitaxel. As you mentioned, the pharmacology of this is not understandable. The only type of pharmacology that would work like this was if paclitaxel was radioactive and accumulated a hazard along the way, but we know that's not true.
I think extend your question, it's important to say that both the FDA and other independent groups like VIVA have looked closely at the meta analytic data both from a patient level and aggregate level data set, and they have seen a signal at five years. The problem with that is that data starts to winnow down very quickly at five years. There's not a lot of numbers, so that's the first problem, and the meta-analysis that have followed the publication by Katsanos. The second problem is, as Josh alluded to, there's a lot of missing data. Either patients withdrew or got lost to follow-up, and that didn't happen at an equal distribution between the control and the active arms, so there's some ascertainment bias there.
And lastly, there's a crossover, that is patients who are in the control arm crossed over near as we can tell at a rate of about one in five or one in four to an active arm in the first year alone, which means they need to be reassigned to a risk pool that includes the original assignment of paclitaxel randomization. My sense is that those data will not get any better in the near-term future because the problems I just listed are not going to go away anytime soon. And so we are left with these individual patient level data and other big data, like Medicare analyses of tens of thousands of patients or Optum insurance analyses of again, tens of thousands of patients, which actually show no difference between the treatment with paclitaxel in the real world and patients treated with non-paclitaxel devices. So while we are comfortable and happy to publish these data and we think that are meaningful in terms of contributing to the larger dataset, we recognize the flaws and the limitations in the meta-analysis, which will not be solved soon or quickly.
Dr Joshua Beckman: So, I totally agree with what you just said. I will also say that every time data like this is published, it adds to the picture to make our understanding clearer. And you are responding directly to the Food and Drug Administration, who basically said they are not settled on this question either. It is noted, they are worried about it, and what they've really asked for is for more data to be published. And so when people analyze data like these, I think it is really helpful to the rest of us to create a fuller and more granular picture of the overall state of the field.
Dr Greg Hundley: We want to thank again both Josh for his time and Bill for his time. Hope you have a great week, and both Carolyn and I look forward to sharing with you again next week. Take care everyone.
Dr Carolyn Lam: This program is copyright American Heart Association 2019.