Ileana L. Piña, MD, MPH; Javed Butler, MD, MPH, MBA


September 18, 2017

Ileana L. Piña, MD, MPH: Hello. I'm Ileana Piña, and I am here at European Society of Cardiology (ESC) Congress 2017 with some 30,000 attendees in the magnificent city of Barcelona, Spain. Today I have my good friend Javed Butler, from Stony Brook University in New York, to talk to us a little bit about a substudy that he is going to be presenting.

There is tremendous interest right now in diabetes. It's like diabetes has been reborn because, for the first time, we have drugs that actually reduce cardiovascular mortality. I've been telling patients for years that they need to keep their sugars down and that they need to take their medicines. But I've never been able to say to them that we can, perhaps, reduce their risk for cardiovascular mortality. Welcome, Javed.

Javed Butler, MD, MPH, MBA: Thank you.

Exploring Heart Failure Risk in Type 2 Diabetes: EMPA-REG, CANVAS

Dr Piña: Tell me a little about the parent trial EMPA-REG.

Dr Butler: As you mentioned, there are all these data coming out about cardiovascular risk prevention in patients with heart failure. With the EMPA-REG trial,[1] over 7000 patients were randomized to placebo or the sodium glucose cotransporter 2 (SGLT2) inhibitor empagliflozin. They found a significant reduction of about 15% in the three-point major adverse cardiac event (MACE) outcome of myocardial infarction, stroke, and cardiovascular death.

The CANVAS trial[2] showed almost the same 15% relative risk reduction in three-point MACE with canagliflozin, another SGLT2 inhibitor. We obviously know that patients with diabetes are at a particularly high risk of developing heart failure; and what is interesting is that while there was a statistically significant 13% or 15% reduction in three-point MACE, there was also a 35% reduction in new-onset heart failure. That reduction was seen in both trials.

Dr Piña: That's impressive. We seldom hear about drugs that actually stop the onset of heart failure. We are so driven to take care of patients once it develops. Tell me about the substudy that you are presenting.

EMPA-REG Insights

Dr Butler: You do these large studies with 6000, 7000 patients and see the benefit; but if you do a cost-effective population-based intervention, you cannot give everything to everybody, whether or not you maintain the cost-effectiveness. So our question was: Does baseline risk matter? There are people all across the spectrum of risk. Some are at low risk, and some are at high risk. We wanted to know whether the risk reduction would be seen across the rest. About 10% of patients in the EMPA-REG trial already had heart failure [at baseline], so we excluded them. We divided the remaining 90% of patients, about 6300, who did not have baseline heart failure into risk categories.[3,4]

We all think about coronary risk prevention a lot; and about a decade ago, we were involved in developing a risk prediction model. The Health ABC score[5,6] is a Framingham risk score equivalent of heart failure, if you may. There are nine risk factors including all of the usual things like creatinine, blood pressure, age, smoking, and so on. Gender is not in that risk model. We put all of those things together and divided patients into three groups: Low-to-average risk was <10% risk for heart failure in 5 years, high risk was 10%-20%, and >20% was the very-high-risk group. The risk score distribution was the same in the placebo group as the treatment group, not that there was any reason not to expect that. About two thirds (67%) of the patients were in the low-to-average risk group. A quarter of the patients were in the high-risk group, and about 5%-6% were in the really-high-risk group. Length of follow-up of the entire trial was about 3.2 years. We normalized the length of follow-up for 100 patient-years.

As you would expect, the absolute risk changed across the spectrum. Risk was around 1.68 per 100 person-year follow-up for low-to-average risk patients, about 5 for intermediate risk, and over 7 for high risk. The benefit was seen across all of the risk groups. There was about a 29% relative risk reduction in the low-to-average risk group and about a 45%-50% risk reduction in high risk and over.

Dr Piña: That's very impressive because you go from 29% to 50%.

Dr Butler: Yes. That was the interesting thing there.

Study Takeaways

Dr Piña: What would you tell clinicians? Should they assess the risk in the specific patient that they have in front of them?

Dr Butler: Remember that the EMPA-REG trial, by definition, had patients with known cardiovascular disease (CVD). These were not only diabetic patients but diabetic patients with CVD, peripheral vascular disease, stroke—some vascular atherosclerotic disease. The top-level message has actually nothing to do with our substudy. The top message we have learned both with CANVAS and EMPA-REG is that these drugs reduce cardiovascular outcomes; they lead to MACE risk reduction. The CANVAS trial was just presented, so it has not gone through the regulatory pathway. Remember, based on EMPA-REG, empagliflozin has a cardiovascular mortality reduction indication now.

Dr Piña: They were the first ones to get the indication.

Dr Butler: Absolutely. Being a heart failure doctor, I get motivated from heart failure research, and this trial shows us that across the risk spectrum of people who met the entry criteria for EMPA-REG, a population that already has some risk, you are really seeing a risk reduction in heart failure risk, and we should really think about these drugs when we are treating our patients with diabetes.

Dr Piña: It would be nice to come up with some algorithm score for the clinician. Here you have a patient who already has CVD or vascular disease. What about all of the peripheral artery disease patients we see that we are not certain what to put them on? Many of those are diabetic.

Dr Butler: While we would like to put the burden on our primary care physicians' shoulders, I think we have our own homework to do. The heart failure community really does not think about prevention as much.

Dr Piña: No, we don't.

Dr Butler: We get motivated about LVADs [left ventricular assist devices] and transplants, and appropriately so.

Dr Piña: By the time they send patients to us, they are so sick that we've got to address the disease right in front of us, and we are not thinking about prevalence. Maybe general cardiology could be looking at this.

Why Does Empagliflozin Work So Well?

Dr Piña: What about this drug? What is it doing?

Dr Butler: I thought I would get by this interview without being asked the more difficult questions.

Dr Piña: No, no, I'm going to ask a tough question here.

Dr Butler: I do not know for sure, but this is my hypothesis. I think there are dual level effects. If you look at the curves, they start separating really early, then they stabilize, and then they start diverging again. I think there are dual mechanisms of action, and one is early. The early effect may be very simple. There is a natriuretic effect and a glycosuric effect, so you get rid of volume and salt. You also have some reduction in B-type natriuretic peptide (BNP) and reduction in cardiac size. Those are all acute effects. The intermediate effect is that you get some lowering of the blood pressure. There are data coming out that show improvement in arterial stiffness, but those data mostly are in type 1 diabetics.[7]

Dr Piña: What about body mass index (BMI)? Was there a reduction in BMI?

Dr Butler: That's the point. Initially there is weight loss, which may just be fluid, but then you actually lose fat weight. As you know, as you lose fat weight, the fat stores are the place where all of your oxidative stress and inflammation keep coming up. Then, in the long run, you are talking about metabolic differences. We actually just published a paper in Journal of the American College of Cardiology[8] looking at stabilization of high-sensitivity troponin and N-terminal pro-BNP on canagliflozin, a cousin drug. There were a few patients in the placebo group in whom these levels started going up, so that may be another mechanism.

Dr Piña: An inflammatory response may be happening. Do you think it's myocardial damage with the troponin?

Dr Butler: It's stability; it's not that it's decreasing. In the placebo group, it starts inching up; and in the canagliflozin group, it is stabilized.

Dr Piña: How often do we see high troponin levels in our patients? All the time.

Dr Butler: Absolutely. The other thing, which is of particular interest to me, is that the beta-hydroxybutyrate levels go up, so fatty acid levels go up, and with ischemic heart disease you have a propensity toward more efficient metabolism. I think there are a lot of mechanisms.

Dr Piña: That is so interesting. We need more biologic plausibility work on these. So often we find the end results, but we don't know what it's really doing. You've got the manuscript done on this one already?

Dr Butler: Manuscript is not only done, it's accepted and was published simultaneously in European Heart Journal.

Dr Piña: Yes, perfect. Thank you, Javed.

Dr Butler: Great. Thank you very much.

Dr Piña: We're going to have more conversations. This is a fascinating and important area. Diabetes has come back to the forefront. Take a look at these new drugs and see if your patients are candidates for them. We think about diabetes as a vascular disease to start with. Here are some answers. Thank you for joining me today from the ESC. Good day.


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.