Cardiology's Hottest Trial of 2019 and Healthcare 2020

DAPA-HF, Apple Heart Study, Healthcare 2020

; C. Michael Gibson, MD


January 15, 2020

This transcript has been edited for clarity and is a continuation of Drs Harrington and Gibson's discussion Cardiology 2019: Top Trials in Review.

Robert A. Harrington, MD: Hi. This is Bob Harrington from Stanford University, here on Medscape Cardiology and Mike Gibson and I have had conversations over the past few years about what has happened in cardiovascular medicine over the prior year, and it's been a lot of fun. This is part 2 of our 2019 reflection where we cover some more trials and then get into some broader societal issues, perhaps focusing on the 2020 healthcare debates and election.

My guest is Mike Gibson from Beth Israel Deaconess, where he's an interventional cardiologist; Harvard Medical School, where he is a professor of medicine; and the Baim Institute, where he is the director. Mike, thanks for joining me on Medscape Cardiology.

C. Michael Gibson, MD: Thanks for having me back.

DAPA-HF: Dapagliflozin in Heart Failure

Harrington: One of the most exciting, interesting, and potentially game-changing clinical trials of 2019 was DAPA-HF, presented at European Society of Cardiology (ESC) 2019, with a follow-up and big secondary analysis at American Heart Association (AHA) 2019. Were you as excited as I was?

Gibson: I was astonished. I was very excited to see the data. In the build-up to this, there were several other drugs in the class—in the EMPA-REG and CANVAS trials—that showed reductions in cardiovascular events, such as cardiovascular death, myocardial infarction, and stroke. They also showed some improvements in the secondary outcome of heart failure hospitalization. For dapagliflozin, a previous study, DECLARE TIMI-58, found a reduction not in overall major adverse cardiovascular events, but in cardiovascular death or heart failure hospitalization. There were some data to suggest this might be an effective approach, and DAPA-HF builds on those data. So, this did not come out of nowhere.

Those studies were not really powered to look at the patients who had heart failure with reduced ejection fraction (HFrEF) or patients with and without type 2 diabetes. Patients with type 1 diabetes were excluded. The DAPA-HF investigators randomly assigned a large number of people (< 4700) who had an EF < 40% and New York Heart Association (NYHA) heart failure classes II-IV, in a double-blind fashion, to receive dapagliflozin or placebo on top of standard of care.

The study lasted about 18 months, and the primary outcome of cardiovascular death or heart failure was reduced from 21.2% to 16.3%, so you would only need to treat about 20 people to prevent one of these outcomes. Cardiovascular death was reduced from 11.5% to 9.6%, so you would only need to treat about 50 people to prevent a cardiovascular death. Everything else was improved: fewer hospitalizations, fewer urgent visits for heart failure, fewer heart failure hospitalizations, and better changes on the Kansas City Cardiomyopathy Questionnaire.

The benefit was seen across all subgroups—of note, in both diabetics and nondiabetics. If you are a heart failure patient, you don't have to have diabetes to get a benefit here. Also, death from any cause was reduced by 2.3%—an amazing improvement in this modern era.

We don't know why this worked. We don't know the true mechanism of benefit. It's been hypothesized to be due to a reduction in preload, some kind of natriuresis or osmotic diuresis, or what is called a "smart diuretic." It could also be reduction in afterload. Is the drug improving energetics, or cardiac metabolism? Is it somehow improving the sodium-hydrogen exchange? Is it improving fibrosis in the heart, or necrosis? All these cytokines are being upregulated or downregulated, and the adipose tissue around the heart might change.

I think this is the hottest trial of 2019...

It's important to note that most of these people had class II heart failure, not III or IV. One other sticking point is that only about 10% of people were on Entresto (sacubitril-valsartan), but the benefit was similar in people on or off Entresto.

This is very exciting and these results were amazing. Yet it was only a single trial. What do you think, Bob?

Harrington: I'll try to stress just a couple of things. The whole series of trials grew out of US Food and Drug Administration guidance a number of years ago that said you needed to rule out cardiovascular harm with these new hypoglycemic drugs. And so, a series of clinical trials have been largely designed to be noninferior on the cardiovascular outcomes while you improve glucose control. That, by itself, has been interesting, but then these observations started to be made with the sodium-glucose cotransporter 2 (SGLT2) inhibitors, that perhaps they were improving cardiovascular outcomes. So, there is a lot of interest in pursuing that. We have a diabetes drug being used in people without diabetes, and it improved their outcome.

I love the design of DAPA-HF. It is a really interesting and creative approach to the clinical trial to include both diabetic and nondiabetic patients. The mechanistic work still has to follow; there is a lot of interesting science to unravel here. These are big effects. Aside from the angiotensin receptor-neprilysin inhibitor issue, largely these patients were well-treated. I think this is the hottest trial of 2019, with really important implications for clinical practice. What did you say at ESC, Mike?

Gibson: We're back. Cardiology is back. It was exciting in the '90s, and then we were in a lull. We didn't have as many shots on target, but, boy, what a year. We've had a lot of great strategies, pharmacotherapy, and interventions assessed. It's been a great year, but this was at the top of the heap.

Apple Heart Study: Atrial Fibrillation Detection Via Smartwatch

Harrington: I agree with you. Let's move to a study, not a clinical trial, coming out of Stanford. I'm not involved with the study, but many of my colleagues were. The Apple Heart Study looked at whether you could you get people to enroll in a virtual study (meaning completely over the phone) at scale and then ask them to allow you to collect data from the sensor on their watch to try to detect atrial fibrillation (AF).

Over 400,000 subjects enrolled over 8 months. I find that the most interesting part of the study was getting such a huge sample rapidly; getting people to agree to give you their data; and being able to potentially detect AF, albeit at a low frequency. What to do with that AF, I think has yet to be determined.

It was not a definitive study, but one that really raised the game for what technology might be able to do in clinical research. That to me was the potential game changer here—a real different way of thinking how we collect data at scale. Mike, what is your view of the Apple Heart Study?

Gibson: There are two things here, as you alluded to. There are the practical questions at hand of what does it do, and what does it detect. But the more important overarching consideration is this platform for conducting trials. I'm very excited about this. The idea that you can enroll 400,000 patients in 8 months is just mind-blowing.

It's important to remember that that was the earlier generations of the Apple watch. It's not an electrocardiogram (EKG) version, but one that detects heart rate using photoplethysmography, which uses light to look at pulses in the skin. If it was irregular on five to six measurements, you got prompted to have a telemedicine visit, and then you wore a patch. The watch had a positive predictive value of about 80%. The issue there, though, is that the patch, as the true standard, was put on a week later, so it might perform better than that.

Pretest probability is really critical here. If you are at low risk of having new-onset AF, not a lot of it's going to happen. If you are under 40 years of age, only 0.16% of the population had new AF. But if you are over 65 years of age, that number rose to 3.2%.

There were legitimate concerns, and there are legitimate concerns, about the number of false-positives with these kinds of technologies. Would this trigger unnecessary tests or excess resource utilization?

We are now building upon your team's experience and doing a randomized trial to take this a step further. We are using the newer watch, which has a single-lead EKG built into it. If you get alarmed that you have an irregular pulse, it'll tell you that you need to check your EKG. You will put your fingers on the bevel and get the EKG, it will go onto your smartphone, and then you can give that to your healthcare professional. By getting an EKG, we may actually be able to improve the diagnostic accuracy of detecting AF.

The idea that you can enroll 400,000 patients in 8 months is just mind-blowing.

Can we detect it? That will be one of the questions. But the more important issue is, who cares? Does it pass the ho-hum test? Can we improve patient outcomes? We are going to randomly assign 180,000 people in the over-65 age group to see whether detection and subsequent treatment, whatever the doctor wants to do, improves the risk for all-cause death or stroke. We are doing all this virtually. You will get randomized over the app and be followed on the app. Your outcomes will be assessed by a claims database. We are cutting out a lot of the bricks and mortar.

Harrington: Using technology to facilitate what we need to do in the research world is the exciting thing to me. We need to have large populations to do things at scale; to do them more efficiently; to do them less expensively; and to do them in a way that is engaging for the patient so they really want to participate in the research process. In the Apple Heart Study, we saw that over 24,000 of the participants were over 65 years of age, so age is not going to be a barrier to doing things virtually.

Gibson: I agree. You know the high cost of doing randomized trials. At the low end, it's $30,000 per patient. In cancer studies, that number can be $150,000-$200,000 per patient. Cardiovascular trials are probably closer to $50,000. Using $30,000 per patient, for example, the cost of this kind of study will be $300 per patient, so about 1% of the usual cost of a randomized trial. Now, we all have to agree that the use of claims databases is going to be sufficiently accurate to assess the outcomes. You may be accepting a little bit of noise in using the claims databases, but we're going to overwhelm that noise with these giga-trials. With a giga-trial of 180,000 patients, I think we will be able to more definitively answer questions rather than just say, "Maybe, maybe not."

Harrington: I like the speed with which we can do it, because if we're going to truly say that we live in a learning healthcare system, we've got to be able to rapidly deploy questions, get the system geared up to answer those questions, answer those questions quickly, and then refine our approaches or refine our questions going forward.

This is an exciting time for clinical research and clinical practice. And it's a really exciting time to have technology help facilitate that research which will inform all of what we are doing in practice. There is a lot of that to come in 2020.

Gibson: I think this is going to improve care for the following reason: We had a lull in cardiovascular development because it was costing a billion dollars per trial at $30,000 per patient. Imagine if we can do trials at 1% of the cost. Imagine how many more shots on goal we will have to try and improve care with newer drugs and devices. We will be able to bring new or more innovative things to patients more rapidly, and that is the real promise here.

Healthcare in 2020

Harrington: I agree with that. This is part of the virtuous cycle of the learning healthcare system, and I am fully on board with what your vision is here.

Let's close with a little bit of commentary on what will be going on in the United States over the next 12 months. We have an election coming in 2020. Healthcare reform has been front and center in the Democratic debates. We've also seen a series of papers that revealed the disturbing statistics that not only has death from cardiovascular disease plateaued over the past couple of years, but with regard to certain aspects of cardiovascular disease, including stroke and heart failure, mortality is actually increasing.

There is a complexity of reasons [for this] that include social determinants of health—ZIP code being a big informer of health status, as opposed to genetic code. A lot is going on in 2020.

Mike, what are you hearing? One of your senators, Elizabeth Warren, is a leading candidate for president. She has some really disruptive notions of how healthcare should be adjusted—not only to guarantee people access to care but also to fundamentally change our entire system. 2020 is going to be an interesting time.

I'm not sure our current system is sustainable at its current speed of growth.

Gibson: It is. I live in Boston now, but when I was young, I spent a lot of time in a town called Stilwell, Oklahoma. Stilwell was in the Washington Post recently, because the average life expectancy there is 56 years. It's worse than a lot of parts of the world. There is a lot of lack of access to care; it is a big Indian community; and there is a lot of diabetes, alcohol, and diabesity. But a lot of these are deaths of despair.

When we were at AHA, Rob Califf asked me how technology is going to improve outcomes. I said to him, "Rob, I don't know that we're going to technology our way out of this." This morning I wrote on Twitter, "I don't think we're going to have drugs to treat deaths due to despair." These are profound societal issues that we have got to wrestle with. Perhaps technology and other things can improve access to these more rural areas of the countries where so many people are suffering. But we face enormous challenges ahead. I'm not sure our current system is sustainable at its current speed of growth, and we've got to do something to continue to innovate, but also to provide more care to more people.

Harrington: We have a paper coming out in the first quarter of 2020 on health in rural America that will, at a macro level, start to look at some of the issues as in your home town in Oklahoma. One out of every 5 Americans lives in rural America—that is 60 million people. As a country, we just can't turn our backs on what has become a real challenge in our healthcare system. Technology is going to be part of it; public policy changes are going to be part of it. We're going to have to figure out what research questions have to be answered to help improve care in areas of the country where it is death and despair, and make things better. To do less than that would leave us as a very inequitable society, and I don't think any of us in cardiovascular medicine want to see that.

The 2020 debates and election are going to be interesting. There is a lot of excitement about technology and new drugs, etc., but some attention is going to be on more fundamental issues that contribute to healthcare. Maybe paying attention to social determinants of health will be the story of 2020—not trials or technologies.

Gibson: I write you in on my form every 4 years. When are you going to announce your candidacy?

Harrington: Dr Gibson, it is always a pleasure to talk with you. We are both people of opportunity, not people of challenges. There is a lot of opportunity in 2020. I think 2019 showed us that we've made some improvement in cardiovascular medicine, but we've got a ways to go. Don't we?

Gibson: We're back, and we're going to be even more back when we have ways to do these trials less expensively.

Harrington: My guest has been Mike Gibson from Harvard Medical School, Beth Israel Deaconess in Boston, and the Bain Institute. Mike, always fun to talk with you on Medscape Cardiology.

Gibson: Thanks, Bob.

Bob Harrington, MD, is chair of medicine at Stanford University and current president of the American Heart Association. (The opinions expressed here are his and not those of the AHA.) He cares deeply about the generation of evidence to guide clinical practice. He's also an over-the-top Boston Red Sox fan.

Follow Drs Bob Harrington and C. Michael Gibson on Twitter

Follow | Medscape Cardiology on Twitter

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: