Eldrin F. Lewis, MD, MPH, on Heart Failure's Place in Diabetes Drug Trials, and the Promise of SGLT2s in Prevention

Evidence-Based Diabetes Management, December 2017, Volume 23, Issue SP14

The payment reform movement has changed the thinking about what can be done for patients with diabetes and heart failure, for the good of patients.

Ten years ago, a stunning article in the New England Journal of Medicine linked a blockbuster diabetes drug—rosiglitazone—with a higher risk of heart attacks.1 The FDA soon required expansive cardiovascular (CV) outcomes trials for new glucose-lowering therapies, to make sure they did not raise the risk of heart attacks, strokes, or early death.2

But what about heart failure? Even though 25% of patients who develop heart failure have diabetes, and these patients tend to be far sicker and costlier to health systems,3 the FDA’s 2008 guidance focused on atherosclerotic CV disease, which occurs when plaque accumulates and hardens the arteries. Heart failure, in which the heart fails to adequately pump blood throughout the body, was not a primary endpoint in the wave of trials that followed.

To Eldrin F. Lewis, MD, MPH, a CV disease and transplant specialist at Brigham and Women’s Hospital and an associate professor at Harvard Medical School, this was a missed opportunity. In an interview with Evidence-Based Diabetes Management™ (EBDM™), Lewis said heart failure specialists tried to sound the alarm.

Since the FDA didn’t request it, 100% of the clinical trials did not include heart failure as part of the primary composite endpoint,” he said. “And for years, we in the heart failure community have been saying, ‘This is a problem; this is a problem.’”

Science has brought others around, however. The first surprise came in 2013, when a safety trial found an unexpected increase in hospitalization for heart failure for the dipeptidyl peptidase-4 (DPP-4) inhibitor saxagliptin.4 The big shock came in September 2015, when the EMPA-REG OUTCOME trial reported that the sodium-glucose cotransporter-2 (SGLT2) inhibitor empaglifozin, sold as Jardiance, caused a 38% reduction in CV death.5 Meanwhile, a secondary endpoint showed a 32% reduction in hospitalization for heart failure among a study population with longstanding diabetes and established CV disease.

Since then, 2 more type 2 diabetes drugs—the SGLT2 inhibitors canaglifozin,6 sold as Invokana, and the glucagonlike peptide-1 (GLP-1) receptor agonist liraglutide,7 sold as Victoza—have shown cardioprotective efects; results from CANVAS for canaglifozin showed a 33% risk reduction for hospitalization for heart failure.6

As the results generate waves at the FDA and among payers, heart failure is having its moment.

Diabetes specialists appeared at this year’s meeting of the American College of Cardiology (ACC) to tout upcoming trials, called EMPEROR HF, that will study empaglifozin in heart failure patients. And Lewis was part of a panel of CV specialists at the American Diabetes Association (ADA) who called for more attention to heart failure, including clinical trials with heart failure as a primary endpoint.

A Prevention Signal?

Separate EMPEROR trials, for EMPaglifozin outcome tRial in Patients With ChrOnic heaRt Failure, will study the SGLT2 inhibitor empaglifozin in patients with preserved and reduced ejection fraction (EF).8,9 Sponsored by Boehringer Ingelheim and Eli Lilly, the trials are currently recruiting patients and will not be completed before 2020. But a separate trial based on claims data, called CVD-REAL, drew notice at this year’s ACC and ADA meetings when results showed the drug class might be able to prevent heart failure in a broad population.10,11

Lewis, whose research includes work on preventing disease progression and on patient quality of life, is excited about the possibility that at long last, strategies will be found to prevent heart failure, especially in high-risk patients with diabetes. Attitudes are shifting not only because of trial data but also because payment reform has compelled health systems to think differently about patients with diabetes and heart failure.

“The management of diabetes is a team effort. We have to work with nutritionists; we have to work with psychiatrists, if patients have depression; with endocrinologists; with cardiologists; and with pharmacists to come up with strategies to optimize management of both,” he said. “These patients are often depressed because they get frustrated that they can’t do a lot of things they want to do. And they want information to try to prevent future adverse events.” What follows are excerpts from the EBDM™ interview with Lewis:

EBDM™:Why has it taken so long for heart failure among patients with diabetes to receive the attention it’s receiving now?

LEWIS: The main reason that it’s taken a while is that the understanding of diabetes and how it relates to cardiovascular disease was that diabetes was a coronary heart disease risk equivalent. That’s been considered standard for a long time. As a consequence, I think there’s been a lot of emphasis on reducing coronary heart disease—heart attacks, unstable angina, and stroke—kind of an atherosclerotic cardiac event.

Because of that understanding—and we understand the impact of diabetes on the pathophysiology of atherosclerosis—translational research [has focused on] linking the presence of diabetes, or poor glucose control, with atherosclerotic events and models. The focus for patient education, clinical trials, and public health prevention has been on reducing atherosclerotic-related events, or atherosclerotic cardiovascular disease.

What we’ve noticed in the heart failure community all along is that in almost every predictor that you evaluate for the development of heart failure, diabetes is one of the most important risk fac-tors. In fact, if you take away hypertension, diabetes would probably be the number 2 risk factor for heart failure development.

The reason I think we are seeing attention to heart failure now is 2-fold:

• First, we have 2 trials looking at SGLT2 inhibitors [empaglifozin and canaglifozin] that show a reduction in heart failure events in patients with diabetes. So people are saying, “Oh, that’s interesting. We need to pay more attention to heart failure.”

• Conversely, because there is a single DPP-4 inhibitor agent [saxagliptin] that led to an increased signal of risk for heart failure, there’s now concern for the safety of therapies for heart failure in addition to overall atherosclerotic cardiovascular disease.

EBDM™: The EMPEROR HF clinical trials are just getting under way. What interests you about these trials?

LEWIS: The EMPEROR trials are exciting for 2 reasons. First, they are looking at patients with heart failure—both a reduced and a preserved EF population. I don’t think you can take data from a high-risk cardiovascular disease trial, such as EMPA-REG and CANVAS, and then extrapolate that to a heart failure population because the heart failure patients have higher risk, and they have other factors that can influence hospitalizations for heart failure. So I think it’s interesting and important to look in both the reduced and preserved EF population to examine the use of empaglifozin in this population, to see whether the findings of EMPA-REG and CANVAS can be reproduced in a heart failure population.

The second interesting thing for me is that these trials will be looking at patients without type 2 diabetes. I’m really intrigued to look at the use of this agent in non-diabetes patients. One of the thoughts that I have is: You have a drug that happens to lower glucose, but it has non-glucose-lowering properties that may lead to an improvement of cardiovascular risk. And if that’s the case, then we can still use empaglifozin and other SGLT2 inhibitors to treat type 2 diabetes, but we can consider using them in non-diabetes patients as well. The question I would have is, whether the mechanism of action for reduction of cardiovascular risk is glucose related or not, I think it doesn’t matter—at the end of the day, you want to see how people do. You want to reduce the morbidity and mortality associated with the complications of diabetes.

EBDM ™: The CVD-REAL trial, while not a randomized clinical trial, suggested there may be potential for SGLT2 inhibitors to be used for prevention. Depending on what we see in the EMPEROR trials, do you see potential for more widespread use of SGLT2 inhibitors—as well as earlier use in the disease life cycle? Is that a possibility?

LEWIS: I think that’s a possibility. I give a little pause in terms of applying it generally, but for the moderate to high-risk population with cardiovascular disease, it looks like there is a benefit. I think the CVD-REAL trial supports the fact that, one, there is early use of SGLT2 inhibitors, and two, it looks like from a large set of payer data that the use of SGLT2 inhibitors—regardless of why—has translated into some improved outcomes in the real world. I’m encouraged by that, and I actually see that we’ll be using more SGLT2 inhibitors fairly rapidly.

EBDM ™: Has the payment reform movement—and particularly the focus on hospital readmissions— caused health systems to pay more attention to heart failure, given the high rate of readmissions and the costs associated with this condition?

LEWIS: I think so. In general, heart failure patients are complicated. The mean age of patients is in the early 70s, and so there may not be as much interest in general in the typical heart failure patient, unfortunately. The older heart failure patients come with a lot of comorbid illnesses—chronic kidney disease, COPD [chronic obstructive pulmonary disease], hypertension, end-stage renal disease—multisystem organ failure. They have frailty, sarcopenia, etc. When these patients come in, they are complicated, and sometimes we cannot dramatically improve them. They are expensive for the hospital system because of the comorbid conditions. As a consequence, there hasn’t been as much focus from mainstream cardiology on heart failure management. We traditionally had readmission rates within 30 days of higher than 30%, with the thinking, “These are complicated patients, and you can’t reduce it.”

When third-party payers started saying there will be a penalty associated with 30-day all-cause readmission, all of a sudden there’s an interest in the hospital systems’ coming up with strategies to reduce readmissions.

With ACOs [accountable care organizations], where everyone must share in coming up with efficiency measures, that alone has incentivized hospital systems to come up with strategies to prevent hospitalizations for heart failure. And the negative incentives from third-party payers to prevent readmissions will also offer a reason to provide necessary resources to prevent readmissions.

hat we’ve noticed since a lot of these changes have occurred is that you have seen a reduction in readmissions—down to the 24% range instead of well over 30%. And we’re starting to better understand why these patients are hospitalized in the first place.

EBDM ™: Preventing disease progression and hospital admissions is the goal. But we hear that when physicians try to start patients on newer therapies, such as SGLT2 inhibitors, earlier in the life cycle of the disease, payers often say no. How can we address these challenges going forward?

LEWIS: There will be challenges—payer restrictions are problematic. The tricky part is, it’s hard to know whether we’re receiving them early enough because we don’t have enough trials to inform us about the timing and the impact of the tiered approach. Metformin still becomes the foundation of diabetes management. Most primary care physicians and endocrinologists would start metformin first. There hasn’t been a head-to-head trial that asks, “In newly diagnosed diabetes, should we start metformin or an SGLT2 inhibitor?” Once we have that information, it will be easier to go away from the current guidelines. The guidelines say, “Start with metformin and then expand.”

In terms of restrictions—and I don’t have data, this is strictly my opinion—if you have a prior authorization that is required, a busy clinician will often find that hurdle somewhat challenging. You have to slow down, complete the form, and sometimes have to do a peer-to-peer review, which takes additional time. If you’re a busy clinician, it becomes challenging in between seeing your 45 to 50 patients. Prior authorization can be enough of a hurdle to reduce utilization.

Unfortunately, we don’t have primary data that evaluate primary care clinicians on why they haven’t started a new therapy. Say you have patient A, who clearly qualifies for initiating an SGLT2 inhibitor; their A1C [glycated hemoglobin] is uncontrolled. And the clinician chooses not to [prescribe the drug]. The question is to go back to the provider and ask, “Why didn’t you do it?” That kind of information would be very helpful. Sometimes it’s lack of knowledge; sometimes it’s the hurdle.

In terms of the knowledge, the other tricky part is, who manages the diabetes? Diabetes and heart failure can both be managed by a specialist; you can have an endocrinologist managing diabetes, a cardiologist or heart specialist managing the heart failure. But in the real world, about 80% of heart failure patients are managed by their primary care doctor. For diabetes, I would assume it would be a relatively high number as well. If you’re a primary care doctor who has 15 minutes to deal with all the problems that a patient has, in addition to diabetes and heart failure, it becomes challenging to not only have the knowledge gap reduced, so that they understand: This is a drug that can be used to treat diabetes, and here’s a novel drug to treat heart failure. But they have to understand when you would use an SGLT2 inhibitor, versus a DPP-4 inhibitor, versus a GLP-1 receptor agonist, versus using insulin initially. So, these algorithms can be complex, and for heart failure, it’s even more complex.

I think the future of diabetes and heart failure management rests with electronic health records [EHRs] with logic built in—to help trigger the primary care doctor to identify patients who might benefit from some of these more novel therapies. Once a new guideline comes out, you build that into the EHR decision-making process.

EBDM™: Are we doing enough to prevent heart failure in patients with diabetes? Do we need to do more?

LEWIS: Given that diabetes is the number 2 risk factor for heart failure, we have to. Especially in patients who have what I call the trifecta—hyper tension, diabetes, and preexisting atherosclerotic cardiovascular disease—those are very high-risk populations. Lipid lowering is important, as well, in these patients, but we need some type of precision medicine approach to managing the prevention of heart failure. If you look at the natural history of patients after they develop heart failure, you’re looking at a median survival of 5 years; at best, the median survival is 8 years, if we include the asymptomatic patients.

We should absolutely come up with strategies to prevent heart failure—and that’s what excites me about the SGLT2 inhibitors. I’m waiting to see the additional trials come out to add to our understanding, but the fact that we have 2 trials that have reduced not only mortality but also reduced heart failure is very important. REFERENCES

1. Nissen SE, Wolski K. Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes [published correction appears in N Engl J Med. 2007;357(1):100]. N Engl J Med. 2007;356(24):2457-2471. doi: 10.1056/NEJMoa072761.

2. FDA guidance for industry. Diabetes mellitus: Developing drugs and therapeutic biologicals for treatment and prevention. US Food and Drug Administration. fda.gov/downloads/Drugs/.../Guidances/ucm071624.pdf. February 2008, Draft. Accessed October 15, 2017.

3. Caffrey M, Szwed A. Heart failure needs more attention in diabetes drug trials, expert says. The American Journal of Managed Care® website. ajmc.com/conferences/ada-2017/Heart-Failure-Needs-More-Attention-in-Diabetes-Drug-Trials-Expert-Says. Published June 11, 2017. Accessed October 15, 2017.

4. Scirica BM, Bhatt DL, Braunwald E, et al, for the SAVOR-TIMI 53 Investigators. Saxagliptin and cardiovascular outcomes in patients with type 2 diabetes mellitus. N Engl J Med. 2013;369(14):1317-1326. doi: 10.1056/NEJMoa1307684.

5. Zinman B, Wanner C, Lachin JM, et al, for the EMPA-REG OUTCOME Investigators. Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes. N Engl J Med. 2015;373(22):2117-2128. doi: 10.1056/NEJMoa1504720.

6. Neal B, Perkovic V, Mahaffey KW, et al, for the CANVAS Program Collaorative Group. Canagliflozin and cardiovascular and renal events in type 2 diabetes. N Engl J Med. 2017; 377(7):644-657. doi: 10.1056/NEJMoa1611925.

7. Marso SP, Daniels GH, Brown-Frandsen K, et al; LEADER Steering Committee; LEADER Trial Investigators. Liraglutide and cardiovascular outcomes in type 2 diabetes. N Engl J Med. 2016;375(4):311-322. doi: 10.1056/


8. EMPagliflozin outcomE tRial in Patients With chrOnic HeaRt Failure With Reduced Ejection Fraction) EMPEROR-Reduced. www.clinicaltrials.gov/ct2/show/NCT03057977. Updated October 4, 2017. Accessed October 15, 2017.

9. EMPagliflozin outcomE tRial in Patients With chrOnic HeaRt Failure With Preserved Ejection Fraction (EMPEROR-Preserved). www.clinicaltrials.gov/ct2/show/NCT03057951. Updated October 4, 2017. Accessed October 15, 2017.

10. Caffrey M. Can SGLT2 inhibitors prevent heart failure in a broad population? The American Journal of Managed Care® website. www.ajmc.com/conferences/acc-2017/can-sglt2-inhibitors-prevent-heart-failure-in-a-broad-population-results-from-a-real-world-study. Published March 19, 2017. Accessed October 15, 2017.

11. Kosiborod M, Cavender MA, Fu AZ, et al; CVD-REAL Investigators and Study Group. Lower risk of heart failure and death in patients initiated on sodium-glucose cotransporter-2 inhibitors versus other glucose-lowering drugs: the CVD-REAL Study. Circulation. 2017;136(3):249-259. doi: 10.1161/CIRCULATIONAHA.117.029190.