Application of the Dapagliflozin Heart Failure Indication to Clinical Practice



  • Roundtable ID: CV38639
    Citation: Published online ahead of print.
  • Summary:

    Dr Bryan Wasson from the University of Texas Southwestern Medical Center, Dallas, TX; and Bridlewood Family Healthcare, Flower Mound, TX, moderated the topic "Application of the Dapagliflozin Heart Failure Indication to Clinical Practice" with Dr Javed Butler from the Department of Medicine, University of Mississippi, Jackson, MS, and Dr Michael Shanik from the Department of Medicine, Stony Brook University Hospital, Stony Brook, NY; and Endocrine Associates of Long Island, Smithtown, NY, participating.

    The discussion focused primarily on: 

    1. Cardiovascular disease reduction trials and heart failure reduction trials in patients with type 2 diabetes; 

    2. comparisons between the DECLARE-TIMI 58 CV outcomes trial, the EMPA-REG trial,
    and the DAPA-HF clinical trial; 

    3. the results and clinical implications of the DECLARE-TIMI 58 CV outcomes trial; 

    4. the Food and Drug Administration–approved indication for dapagliflozin for heart failure hospitalization reduction; and 

    5. the application of the new dapagliflozin heart failure with diabetes data to clinical practice.

    [Published online ahead of print 2020 August] (Med Roundtable Cardiovasc Ed. 2020 August) ©2020 FoxP2Media, LLC

    This roundtable was supported by AstraZeneca. The discussants (authors) developed the discussion and reviewed the transcript for important intellectual content, and approved the final version for publication. The authors maintained control of the discussion and the resulting content of this article.


Please register or login to see this content.

Please register or login to see this content.
Please register or login to see this content.
Please subscribe to use our print features or to download PDF files.

DR WASSON: Let’s begin with introductions. Dr Butler, would you mind introducing yourself to the group?

DR BUTLER: Hi, I’m Javed Butler. I am a professor and chairman in the Department of Medicine at the University of Mississippi and a cardiologist by training.

DR WASSON: Very good. Dr Shanik, can you introduce yourself as well?

DR SHANIK: My name is Michael Shanik. I am managing partner at
Endocrine Associates of Long Island
in Smithtown, New York. I am also Associate Professor of Medicine at Stony Brook University Hospital. I’m an endocrinologist.

DR WASSON: My name is Dr Bryan Wasson. I’m an internal medicine physician and a Clinical Assistant Professor of Family and Community Medicine at the University of Texas Southwestern Medical School. I’m also an adjunct clinical instructor at the University of Texas at Arlington College of Nursing. In addition, I hold an honorary faculty position at UT Dallas. I practice internal medicine in the Dallas–Fort Worth area. It is my pleasure to welcome each of you to this roundtable discussion. If you don’t mind, I’m going to do a quick overview of the sodium-glucose cotransporter-2 (SGLT-2) inhibitor class for just a moment. I’m going to introduce a brief historical perspective of recently approved classes of diabetic medications. Following the introduction we will launch into an overview of the DECLARE-TIMI 58 trial.1 

SGLT-2 plays a key role in the tubular reabsorption of glucose. The kidney filters glucose molecules that are subsequently reabsorbed in the proximal tubules into the bloodstream. However, when the amount of glucose in the renal tubules surpasses roughly 160 mg/dL to 180 mg/dL, the kidneys are unable to reabsorb the glucose load. Therefore, glucose begins to be released in the urine. We also know that there is an overexpression of SGLT-2 in diabetes, which increases glucose reabsorption and thus hyperglycemia. By blocking SGLT-2 and the reabsorption of glucose from the proximal renal tubules we can favorably affect glycemia.

Recall in 2008, the US Food and Drug Administration released industry guidelines for the evaluation of cardiovascular (CV) safety of new anti
diabetic therapies for type 2 diabetes. As a result, we now have CV outcome trials that have spanned through all of these new types of medications, including dipeptidyl peptidase-4 (DPP-4)
inhibitors, glucagon-like peptide-1 receptor agonists, and SGLT-2 inhibitors. In 2015, EMPA-REG2 with empagliflozin showed lower composite CV end points versus placebo in patients with type 2 diabetes who had established CV diseases.

This sets the stage then for other trials to be set forth insofar as looking at heart failure (HF) benefits in the face of type 2 diabetes, which leads us then to the DECLARE trial with dapagliflozin. So, gentlemen, with your knowledge base of that trial, can you help summarize the overview of this particular trial and the significance of the trial in this setting?

DR SHANIK: As you pointed out, there is a big need for potential CV benefit with different treatments. The important part to note about the DECLARE trial is the large study population, which included over 17,000 patients, many of whom did not have atherosclerotic CV disease. The median duration was a little over 4 years and one of the dual primary outcomes was the composite of major adverse CV events, ie, CV death, myocardial infarction, or ischemic stroke. It also looked at another dual primary end point, which was hospitalization for HF and CV death. What was found in this trial was that in this large population, where they randomized patients to receive either dapagliflozin 10 mg or placebo, there was a significant reduction in the composite of hospitalization for HF or CV death. What’s notable about the trial is that it was a broad population, not necessarily all having CV disease. Many of those patients had no known CV disease and still had the benefit when it came to hospitalization for HF.

DR BUTLER: I completely agree with the characterization of the trial. Just one additional piece is that, of all the SGLT-2 inhibitor trials, this trial had the largest proportion of what we would consider a primary prevention cohort, ie, those patients who had risk factors for atherosclerotic CV disease but did not have manifest CV disease, per se, as opposed to EMPA-REG, in which all patients for secondary prevention had atherosclerotic disease, and CANVAS,3 which had about two thirds with atherosclerotic CV disease. Here, the majority of the patients were actually at high risk but did not have atherosclerotic CV disease, so this trial extends the data to the primary prevention population as well.

DR WASSON: The DAPA-HF trial tested the hypothesis that dapagliflozin would reduce the primary composite outcome of worsening HF or CV death in patients with HF and reduced ejection fraction, with or without type 2 diabetes. What drove the CV death reduction in EMPA-REG? 

DR BUTLER: That’s a little bit of a tricky thing, and we’ll have to put the results in the historical context as these trials initially came about because there was concern for a potential increased risk for ischemic events. Therefore, end points were not based on any pathophysiologic rationale but because of the regulatory guidance, and so major adverse CV event was the primary end point. The original trials had only major adverse CV event as their primary end point and in EMPA-REG including a CV mortality benefit. This trial had about 10% of the patients who had baseline HF and 90% who did not. What stimulated interest was that we came to find out that there was a substantial reduction in HF hospitalizations both in patients with and without a history of HF at baseline, which suggested the potential for both HF treatment and prevention per se. By the time we got to the DECLARE trial, because we had the data from EMPA-REG, the primary end point was expanded to dual primary end points including major adverse cardiovascular events and HF hospitalization and CV death, and the result was positive for CV death and HF hospitalization, mainly driven by the HF hospitalization benefit. 

DR WASSON: In this particular trial, again we’re referring to the DECLARE trial, patients with diabetes who were enrolled had glycated hemoglobin (HbA1c) values between 6.5% and 12%. As you know, working with diabetes, many diabetic medications have a greater effect on individuals with higher HbA1c. Does this also apply to those who have congestive HF (CHF), ie, do those patients who have the higher HbA1c tend to have a better outcome on this medication who also have concomitant CHF? Any thoughts on that?

DR SHANIK: It’s a great question. We do know that there are some historical studies looking at worsening glycemic control and increasing risk for HF. The majority of the patients in the DECLARE trial did not have a known history of HF. It was not assessed in the trial but I do think it’s an intriguing question to ask because it’s likely that poor glycemic control, especially long-term glycemic control, increases the risk of microvascular and macrovascular disease and the potential CV implications. One might presume, although we don’t know for sure at this point, that there might be a bigger benefit in patients who have even higher HbA1c and worse glycemic control, but across the board, at least in the trial, it was pretty consistent in different subgroups that there was a benefit for HF reduction.

DR WASSON: The DAPA-HF trial4 was published in the New England Journal of Medicine on November 21, 2019. To refresh, this trial studied dapagliflozin in patients with HF and reduced ejection fraction (EF). It involved 4744 patients who were enrolled with New York Heart Association (NYHA) class II, III, or IV and an EF of ≤40% randomized to 10 mg of dapagliflozin versus placebo. A total of 42% of the patients in the trial had a history of type 2 diabetes mellitus and an additional 3% (82 patients in the dapagliflozin group and 74 in the placebo group) had previously undiagnosed diabetes, which was defined as an HbA1c level of ≥6.5% (≥48 mmol/mol), as measured in a central laboratory at both screening and randomization.5 Which patients in this scenario with the classification of HF do you surmise would have the greater benefit: the one with the NYHA class II, III, or IV HF, or do we have those data available for us, and what are the implications for that then?

DR BUTLER: I look at it a little bit differently. The NYHA class II, III, IV gives us a little bit of a sense about the severity of the disease and how sick somebody is so it’s not exactly a disease classifier as much as it is a disease severity classifier. So, I would probably divide the patients into those patients at risk for HF but who don’t have HF per se, and those patients who do have HF but have HF with reduced EF, and then those patients with HF who have HF with preserved EF. So, the two nuances that you are suggesting here—I think we have plenty of data with a large number of patients. The risk for HF hospitalization, which is sort of a proxy for HF prevention, in those patients who don’t have HF at baseline––that is actually pretty well established. So, what about patients who already have HF? In other words, not prevention but can you treat HF, per se? 

There is a lot of interest in looking at both patients who have HF with reduced EF and those who have HF with preserved EF. Now, before we go into the reduced versus the preserved EF, the other interesting thing that you mentioned was the correlation with HbA1c. It turns out that the HbA1c showed no correlation with the benefits. If you had a high or a low HbA1c level you still benefited within the diabetes range, right? But, if you look at the pharmacodynamic properties of these drugs, and the different things that they do in the body, one can actually hypothesize that it has nothing to do with diabetes; that even patients who don’t have diabetes can potentially benefit. So, when we started looking into the issue of HF, the use of these drugs was expanded, and these drugs were tested in both patients with and without diabetes. The DAPA-HF study was in HF with reduced EF regardless of diabetes, and that trial was really positive in patients both with and without diabetes. So, whether you have class II, III, or IV symptoms, it doesn’t really matter the severity––if you have reduced EF there was a mortality and hospitalization benefit with the use of dapagliflozin regardless of whether you have diabetes or you don’t have diabetes. And then the last group includes those patients with preserved EF… that trial in HF with preserved EF is ongoing.

DR SHANIK: I just want to make one other point that I think is interesting. When you compare the two trials, DECLARE (all diabetes) and DAPA-HF (diabetes and nondiabetes), most of the patients in the DECLARE  trial did not have established HF and they got a benefit for hospitalization for HF reduction. In the DAPA-HF trial, where there was an established population of patients with known HF with reduced EF, they were already on HF standard of care. They were getting all the benefits of every drug that they should have gotten and on top of that, with the established HF and standard of care, they still had an additional benefit. So, I think the two studies really illustrate that for HF reduction in both diabetics and nondiabetics, you’re going to get a benefit when you add dapagliflozin, regardless of their diabetes state.

DR WASSON: That’s an excellent point. What do you hypothesize is the mechanism behind the improvement in HF? Allow me to digress a moment to highlight a few points. One facet I would like to introduce is that in the fasting state, the main source for cardiac ATP, in fact, up to 70% of total production, comes from fatty acid oxidative phosphorylation, 20% from glucose, and the remaining from amino acids and lactic and a couple  of others. In the fed state, the presence of high carbohydrate level, the principal substrate is glucose. With that understanding, there are a couple of different metabolic disturbances that account for the failing heart especially in diabetic cardiomyopathy, with a greater reliability on fatty acid oxidation but lower glucose consumption. There are a couple of mechanisms involved with this process. The altered cardiac metabolism may lead to inflammation and myocardial stiffness resulting in extracellular matrix stiffness. There are changes in mechanisms of inflammation oxidative stress that may in fact be mediated by an SGLT-2 inhibitor medication. What are your thoughts, gentlemen, on the underlying cause outside that of osmotic diuresis and blood pressure control that we think is manifesting here in the positive outcomes for the DECLARE trial?

DR BUTLER: I think you’ve characterized the metabolic profile of these drugs pretty well, but I would say that this is incredibly complex. We tend to sometimes focus on trying to find a unifying theory that explains all things but the pharmacodynamic profiles of these drugs are so incredibly vast and there are so many different ways by which they can affect the heart that the overwhelming likelihood is that it will never be one thing or another and it’s in concert that all of these things are causing the benefits. 

So, clearly, the effect of osmotic diuresis and natriuresis and volume loss in the kidneys is one thing, but also preservation of the renal function, actually in the long run in these high-risk patients who are at risk for or have chronic kidney disease by itself, can delay the development of HF. Decreased adiposity, as you mentioned, leads to reduction in oxidative stress and reduction in inflammation. There are data with improvement in aortic stiffness. There are data with endothelial function improvement. Sodium hydrogen exchange pump inhibition may favorably affect the heart. Effects on epicardial and perinephric fat may reduce progression of heart and kidney dysfunction, then the metabolic profile that you mentioned. So, I think it’s a concert effect of all of these pharmacodynamic properties rather than one over another.

DR WASSON: We know diabetic patients have a higher risk of coronary artery disease. What is the percentage of those people who have diabetes and have CHF as a presenting symptom who are independent of coronary artery disease? How do we identify these patients as being a target for this class of medication?

DR SHANIK: That’s a good question. Many of the patients we see are certainly at risk for developing HF and we’re not aware of it. I think the classic teaching was that diabetes is a CV risk equivalent but we didn’t necessarily think of it as an HF risk. We thought of coronary disease. I think now we’re very well aware that diabetic cardiomyopathy is a clear picture that seems to be somewhat unique and different than other forms of HF in patients who don’t have diabetes.

Ventricular mass changes, wall thickness, myocardial damage—all of those effects that we think are partly the glycemic effects from high blood sugars—but also all the inflammatory markers and other things that go along with obesity, diabetes, and metabolic syndrome. As far as how we address and then determine which patients are appropriate, if we follow patients with diabetes long enough, they’re going to have a higher risk of HF. What is also not as well-known but is certainly clear is that HF is probably more common depending on how you define it––more common than other classic CV morbidities. So, I think we are not always seeing it, but certainly on echocardiograms and sometimes subtle symptoms can define which patients are appropriate. In a broad population such as that studied in the DECLARE trial, we can see that many of these patients would benefit.