A New Drug for Heart Failure? DEFINE-HF Bolsters Dapagliflozin Cardiovascular Cred

September 17, 2019

PHILADELPHIA — Once again, the antidiabetic drug dapagliflozin (Farxiga, AstraZeneca) showed that the term "antidiabetic drug" doesn't really capture all that it might offer.

In the randomized DEFINE-HF trial of patients with heart failure with reduced ejection fraction (HFrEF), either with or without type 2 diabetes, those who took dapagliflozin along with their standard heart failure meds rapidly showed fewer symptoms and improved quality of life.

On the other hand, levels of the biomarker N-terminal pro b-type natriuretic peptide (NT-proBNP) didn't fall significantly in those taking the drug compared with its placebo. That suggests decongestion is not how dapagliflozin helps in heart failure and drives home that the actual mechanism is still a bit of a mystery.

Similar patients given the drug in a study reported only a few weeks ago, DAPA-HF, benefited with reduced risk of cardiovascular death or heart failure hospitalization over about 18 months.

Although DEFINE-HF didn't look at such clinical outcomes, its findings contribute to the total profile of potential dapagliflozin benefits that patients with HFrEF care about.

"Dapagliflozin produced what I think are clinically meaningful improvements in reducing symptom burden, improving functional status, and quality of life in just 12 weeks of treatment," Mikhail Kosiborod, MD, told theheart.org | Medscape Cardiology.

And it did it in patients already on excellent guideline-directed medical therapy for HFrEF, said Kosiborod, of Saint Luke's Mid America Heart Institute, Kansas City, Missouri. Virtually everyone in the trial was on beta-blockers and renin-angiotensin-system (RAS) inhibitors of various kinds.

The patients on dapagliflozin, a sodium-glucose cotransporter 2 (SGLT-2) inhibitor, benefited regardless of which RAS inhibitor they were taking and, importantly, whether they had diabetes — also mirroring the results of DAPA-HF.

"Collectively, the results support the use of dapagliflozin as a new treatment option in patients with HFrEF regardless of diabetes status," Kosiborod said when presenting DEFINE-HF here at the Heart Failure Society of America (HFSA) 23rd Annual Scientific Meeting. He is also senior author on the trial's report, which was published online September 16 in Circulation, at about the same time as the presentation.

The symptomatic and quality-of-life improvements on dapagliflozin required a number-needed-to-treat (NNT) of only 10. That's about half the NNT to reduce the risk of the clinical endpoints seen in DAPA-HF, observed Kosiborod, who was also slated to present DEFINE-HF at the September 16-20 European Association for the Study of Diabetes (EASD) 2019 Annual Meeting in Barcelona, Spain.

Not Just "An Expensive Diuretic"

In diabetes, SGLT-2 inhibitors work in part by promoting elimination of glucose in urine, so it has been frequently questioned whether dapagliflozin in HFrEF is simply "an expensive diuretic," pointed out John Teerlink, MD, as a panelist following Kosiborod's presentation.

If true, it would possibly show up as reduction in NT-proBNP, an indirect biomarker for fluid status — which didn't happen in DEFINE-HF, said Teerlink, of San Francisco VA Medical Center and University of California, San Francisco.

"I think it's very clear, looking at various data points now, that these [agents] are not just simple diuretics. They certainly have a lot more effects other than diuresis, and in fact I'm not sure how much of the effect that we see is diuresis at all," Kosiborod replied.

Also evident in DEFINE-HF is that dapagliflozin does not entirely share mechanisms with the RAS inhibitors in heart failure, he said in an interview.

That's because its patients benefited from addition of dapagliflozin regardless of which RAS inhibitor they were also taking, or whether they were taking a RAS inhibitor at all, he said. "The mechanisms appear to be nonoverlapping."

As to which mechanism, "We're still kind of scratching our heads," Eric J. Velazquez, MD, Yale School of Medicine, New Haven, Connecticut, told theheart.org | Medscape Cardiology.

"This data suggests that the impact, on at least patient-perceived quality of life, is independent of a major effect across the population in NT-proBNP," he noted.

"That actually is probably, at the end of the day, the most important contribution of DEFINE-HF, which is to highlight that this effect that we're seeing in the clinical outcome trials is not just because of better decongestion," Velazquez said. "It's something else."

When to "Jump" in HFrEF

Despite the appeal of SGLT-2 inhibitors, at least dapagliflozin, as a new class of heart failure medication, no one is currently recommending that they be used quite that way.

In patients with diabetes and "progressive heart failure symptoms," Velazquez said, "my suspicion is that people are going to jump very quickly to adding dapagliflozin in their population as an additional therapy."

"Frankly, if they have diabetes and heart failure, using one of these medications is a no-brainer," Kosiborod agreed when interviewed. Their trials show the drugs can lead to significant reductions in morbidity, mortality, and symptom burden, he said.

"These drugs we know also protect the kidneys in patients with diabetes. And they of course improve glycemic status. So you're killing not just one or two, but three birds with one stone in many of these patients."

But in those with heart failure without diabetes, he said, "there is not an indication yet. My prediction would be that it's likely going to be incorporated in guidelines, hopefully in the relatively near future. And I certainly hope there will be an indication in the reasonably near future, but that will be up to the regulators."

Indeed, AstraZeneca announced yesterday that the US Food and Drug Administration (FDA) has given dapagliflozin "fast track" status for a proposed indication to reduce cardiovascular death or worsening of heart failure in adults with either reduced- or preserved-ejection-fraction heart failure.

Two Primary Endpoints

DEFINE-HF randomly assigned 263 patients with optimally managed HFrEF to also receive dapagliflozin 10 mg daily or placebo for 12 weeks. They were required to be in NYHA class 2-3 functional class with elevated natriuretic peptides and good renal function and without a heart failure hospitalization or coronary revascularization in the past month. Patients with type 1 diabetes were excluded.

Nearly all were on beta-blockers and RAS inhibitors, including ACE inhibitors or angiotensin receptor blockers (ARBs) in 59% and sacubitril/valsartan (Entresto, Novartis) in 33%; 60% were on aldosterone inhibitors. About 35% were on biventricular pacing for heart failure.

The two groups didn't differ significantly in the average of adjusted 6-week and 12-week NT-proBNP levels, the first of two primary endpoints. The odds ratio (OR) for the effect of dapagliflozin on the endpoint was 0.95 (95% confidence interval [CI], 0.84 - 1.08; P = .43).

For the second primary endpoint, a composite of elevation in heart failure–specific health status by at least 5 points in the Kansas City Cardiomyopathy Questionnaire (KCCQ) summary score or at least a 20% decrease in NT-proBNP levels, the OR effect for dapagliflozin was 1.8 (95% CI, 1.03 - 3.06).

Both the KCCQ-score and NT-proBNP portions of the endpoint contributed to the "nominally significant" P value of .039, Kosiborod reported.

The results were consistent in all prespecified subgroup analyses, including by diabetes status.

Table 1. Odds Ratio (OR) for Outcomes, Dapagliflozin vs Placebo, by Relevant Subgroups in DEFINE-HF

Subgroup Parameters

OR (95% CI)

Interaction P value

History of Diabetes



1.4 (0.7 - 2.9)



2.6 (0.9 - 7.4)


RAS Inhibitor Used


 ACE inhibitor or ARB

2.0 (0.9 - 4.5)



1.9 (0.6 - 5.8)



1.5 (0.3 - 7.6)


*For the second primary endpoint reflecting clinically meaningful improvement in heart failure disease-specific health status or NT-proBNP levels.

The treatment groups didn't differ significantly in weight change nor in 6-minute walk distance at either 6 or 12 weeks.

Serious adverse events in the study were few, Kosiborod said, and they were similarly infrequent in the two groups. "Drug adverse events" occurred in 2.3% in both groups. The most common were "volume depletion events" in 9.2% on dapagliflozin and 5.3% on placebo.

The benefits of dapagliflozin "very clearly" represent a SGLT-2 class effect in the arena of heart failure prevention for people with diabetes. "We saw it consistently in all the trials, with different agents and different patient populations," Kosiborod said.

"I think it's more likely than not to be a class effect in heart failure treatment as well. Of course what we have the data for right now is dapagliflozin. It will be important to see data with other agents."

DEFINE-HF was investigator-initiated and funded by AstraZeneca. Kosiborod disclosed research support from AstraZeneca and Boehringer Ingelheim; honoraria from AstraZeneca, Boehringer Ingelheim, and Novo Nordisk; and consulting fees from Amarin, Amgen, Applied Therapeutics, AstraZeneca, Bayer, Boehringer Ingelheim, Eisai, Glaxo Smith Kline, Glytec, Intarcia, Janssen, Merck, Novartis, Novo Nordisk, and Sanofi Aventis. Disclosures for the other study authors are in the full report. Velazquez reported research grants, honoraria for speaking, and consulting or serving on an advisory board for Novartis. Teerlink declared research grants from Abbott, Amgen, AstraZeneca, Bayer, Boerhringer Ingelheim, Bristol-Myers Squibb, Cytokinetics, Medtronic, Novartis, and Windtree; and consulting or serving on an advisory board for Amgen, AstraZeneca, Bayer, Boerhringer Ingelheim, Bristol-Myers Squibb, Cytokinetics, Novartis, and Windtree.

Heart Failure Society of America 23rd Annual Scientific Meeting: Late Breaking Clinical Trials. Presented September 16, 2019.

Circulation. Published September 16, 2019. Abstract

AstraZeneca. Published September 16, 2019. Press Release

Follow Steve Stiles on Twitter: @SteveStiles2. For more from theheart.org | Medscape Cardiology, follow us on Twitter and Facebook.


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.