DECLARE-TIMI 58: Dapagliflozin Cuts Heart Failure in Diabetes

Sue Hughes  

November 10, 2018

CHICAGO —  Dapagliflozin (Farxiga/Forxiga, AstraZeneca) showed a nonsignificant trend toward a reduced rate of major adverse cardiac events (MACE) but significantly reduced hospitalization for heart failure in the DECLARE-TIMI 58 trial in patients with type 2 diabetes.

The trial was presented here at the American Heart Association (AHA) Scientific Sessions 2018 by lead author Stephen Wiviott, MD, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts.  It was also simultaneously published online in the New England Journal of Medicine.

"What we're seeing in this trial is a similar theme to other major trials of  sodium glucose cotransporter-2 (SGLT2) inhibitors — a significant reduction in heart failure hospitalization and renal events," Wiviott commented to | Medscape Cardiology.

"DECLARE-TIMI 58, however, differs from the other [cardiovascular outcomes] trials in that it enrolled a much broader and healthier population including 10,000 patients without pre-existing cardiovascular disease but with multiple risk factors, as well as 7000 patients with pre-existing cardiovascular disease."   

"We found that the benefit of dapagliflozin on heart failure was similar in those patients with and without pre-existing cardiovascular disease, whereas the effect on MACE differed between these populations, with no effect in the primary prevention group and a trend towards a reduction in those with secondary prevention.

"All three SGLT2-inhbitor trials have found a large effect on the heart failure endpoint, and our trial adds to the known literature in that regard. But it also extends this heart failure benefit to the primary prevention diabetic population," Wiviott said.

He added: "If we look across all the trials, empagliflozin showed the most benefit on MACE, but this was still less than the heart failure benefit.  I think after the DECLARE-TIMI 58 trial we can definitely now say the biggest benefit of the SGLT2 inhibitors is on the prevention of heart failure, and the reduction in major cardiovascular events is limited to patients with existing underlying cardiovascular disease."

"The DECLARE TIMI-58 trial also gives very reassuring data on safety with no sign of any increases in stroke, amputations, or bladder cancer," he added.

The large cardiovascular (CV) outcome trials of newer type 2 diabetes drugs are being conducted to demonstrate safety following a mandate by the US Food and Drug Administration (FDA) in 2008 after concerns about CV harm with older  type 2 diabetes medications.

But thus far, none of the eight completed CV outcomes trials have identified excess CV risk from the drugs in question, and three have actually shown benefit.

These included two studies of oral SGLT2 inhibitors: the EMPA-REG OUTCOMES trial with empagliflozin (Jardiance, Boehringer Ingelheim/Lilly) and CANVAS with canagliflozin (Invokana, Janssen). In both studies, all patients had type 2 diabetes and existing CVD or were at high risk for CVD.

Likewise, in the third study, LEADER, with the injectable once-daily glucagon-like protein-1 (GLP-1) agonist liraglutide (Victoza, Novo Nordisk), all of the patients with type 2 diabetes had established CV disease (CVD) or chronic renal failure, or were aged 60 years and older with CVD risk factors.

DECLARE now adds to this list of studies showing cardiovascular benefit with new diabetes drugs, although its benefits are restricted to the heart failure endpoint and it didn't show the same reductions in MACE as did the other SGLT2-inhibitor trials, or LEADER. But this trial enrolled a lower-risk population of patients with type 2 diabetes than in the previous cardiovascular outcomes studies.

No Increase in Amputations With Dapagliflozin in DECLARE

For the DECLARE-TIMI 58 study, 17,160 patients with type 2 diabetes who had atherosclerotic CVD or multiple risk factors for CVD were randomly assigned to dapagliflozin 10 mg daily or placebo on top of standard therapy.

The primary safety outcome was a composite of MACE events, defined as CV death, myocardial infarction (MI), or ischemic stroke.  The two co-primary efficacy endpoints were MACE and a composite of cardiovascular death or hospitalization for heart failure. 

After a mean follow-up of 4.2 years, the primary safety outcome met the criteria for noninferiority.  

In terms of the two efficacy outcomes, MACE were numerically reduced in the dapagliflozin group, but this finding was not significant. The CV death/heart failure hospitalization endpoint was significantly reduced. This was driven by a lower rate of hospitalization for heart failure.

A key secondary outcome was a renal composite (≥40% decrease in estimated glomerular filtration rate to <60 mL/min per 1.73 m2 of body surface area, new end-stage renal disease, or death from renal or CV causes). This was also significantly reduced with dapagliflozin.

Table 1. DECLARE-TIMI 58: Main Results 

Variable Dapagliflozin (%) Placebo (%) Hazard Ratio (95% Confidence Interval)
CV death/MI/stroke 8.8 9.4 0.93 (0.84 - 1.03)
CV death/heart failure hospitalization 4.9 5.8 0.83 (0.73 - 0.95)
Heart failure hospitalization 2.5 3.3 0.73 (0.61 - 0.88)
CV death 2.9 2.9 0.98 (0.82 - 1.17)
Renal composite 4.3 5.6 0.76 (0.67 - 0.87)


After the groups were separated into patients with and those without established cardiovascular disease, MACE were nonsignificantly reduced with dapagliflozin in those with established disease, but there was no effect in those without established CVD.

Table 2. Results in Those With and Without CVD (HR for Dapagliflozin) 

Variable Hazard Ratio (95% Confidence Interval)
CV death/MI/stroke 0.90 (0.79 - 1.02) 1.01 (0.86 - 1.20)
CV death/heart failure hospitalization 0.83 (0.71 - 0.98) 0.84 (0.67 - 1.04)


In terms of adverse events, diabetic ketoacidosis was more common with dapagliflozin (0.3% vs 0.1%), as were genital infectors leading to discontinuations or considered to be serious (0.9% vs 0.1%). Wiviott noted that these are both known side effects of SGLT2 inhibitors.

He commented: "Our results are also reassuring in that we didn't see any suggestion of an increase in  amputations or strokes with dapagliflozin and this is the largest study of these agents with the longest follow-up."  

"In the [EMPA-REG OUTCOMES] empagliflozin trial, stroke went in the wrong direction and in the CANVAS trial with canagliflozin there was an increased incidence of amputations in the treated group. Because of these observations in previous trials, we evaluated these outcomes very carefully and found no evidence at all of any increase with dapagliflozin."

"In early studies of dapagliflozin, there was a small increase in bladder cancer with the drug so the FDA mandated that we had to do careful surveillance for this in the DECLARE trial and it turned out that the rate of bladder cancer was actually lower in the dapagliflozin arm.  So this is reassuring again and shows that observations in studies with small numbers are often due to chance," he added.

CV Outcomes Trials in Diabetes: Sea Change in Therapy

Wiviott noted that these newer type 2 diabetes drugs have been slow to penetrate the market. "At present cardiologists are not often prescribing these drugs, but now we have multiple studies showing cardiovascular benefits, I think their use in the cardiology community will grow in both primary and secondary prevention patients with diabetes.     

"These trials were initially conducted to show cardiovascular safety, but they have actually shown cardiovascular benefit, which was not expected, and so these drugs are now turning into cardiovascular agents that also reduce blood sugar rather than just diabetic drugs.

"That is a sea change, and studies are now ongoing with SGLT2 inhibitors as heart failure and renal prevention treatments in patients without diabetes."

"Research is also ongoing on the mechanism of action behind their beneficial effects, which is probably not just due to blood sugar lowering," he added.  "They affect the sodium/glucose transporter in the kidney so the patient excretes sodium and glucose in the urine, but they may also have direct cardiac effects," he suggested.

Asked how the different agents within the class compared, he said, "I would feel confident using any of these drugs. Instead of competing over which SGLT2 inhibitor to use, I would recommend that, when treating diabetic patients, any of the drugs in this class which have proven cardiovascular and renal benefit would be preferable to the older diabetes drugs, which have not shown such benefits."

Reducing Macrovascular and Microvascular Events: A Paradigm Shift

"I also think we are entering a paradigm change in the treatment of diabetes. Everyone has so far been fixed on lowering blood sugar to reduce microvascular complications and there has been nothing to differentiate between the various classes of [newer] diabetes drugs, but now we are starting to focus on reducing macrovascular complications (ie, cardiovascular outcomes) as well."  

Designated discussant of the study, Javed Butler, MD, University of Mississippi Medical Center, Jackson, said DECLARE-TIMI 58 was a well-conducted trial and included the highest proportion of diabetic patients without established atherosclerotic CVD of all the SGLT2 inhibitor CV outcomes trials.

"This trial shows again the benefits of SGLT2 inhibitors in diabetic patients in terms of reducing heart failure risk and renal problem," he said.

"We are also seeing from all the trials together that diabetic patients with underlying cardiovascular disease" who are taking these agents get a benefit on MACE, but that "this effect does not extend to patients without underlying cardiovascular disease."

Butler stressed that heart failure is a very important endpoint for diabetic trials.

Heart failure "is equally, or possibly even more, common than major adverse cardiovascular events in diabetes patients, and heart failure usually has worse outcomes. We also know that we can reduce cardiovascular outcomes in diabetes patients by working on lifestyle — quitting smoking, reducing weight, and blood pressure, et cetera — but this doesn't seem to have the same effect on heart failure risk."

"These trials have now conclusively demonstrated that diabetes patients with underlying cardiovascular disease, or with multiple cardiovascular risk factors, should be taking these drugs to lower their risk of heart failure."

To | Medscape Cardiology, Butler added: "Choosing between the individual SGLT 2 inhibitor drugs is difficult. The cardiovascular mortality benefit with empagliflozin was very striking — that's hard to ignore. The renal and heart failure benefits seem to be overlapping with all the drugs. There was a small increase in amputations with canagliflozin. This may been just chance and has not been seen with other SGLT2 inhibitors." 

"Then we also have the GLP-1 agonist drugs, which have shown a clear benefit on major adverse cardiovascular events but seem to be neutral on heart failure risk. I think we can make a case for use of both of these classes of agents in some cases."

A More Tempered View…

Others, however, are taking a more tempered view. One of these is David Nathan, MD, director of the diabetes center at Massachusetts General Hospital, Boston, Massachusetts. He commented to | Medscape Cardiology: "These SGLT2 inhibitors reduce the risk of heart failure hospitalization in diabetics with, or at increased risk of, heart disease, but the absolute risk reduction is quite modest — about 1%. Empagliflozin has shown better effects on major adverse cardiovascular events in patients with established heart disease."

He also points out that the adverse effects and costs of the SGLT2 inhibitors all need to be taken into account when their use is being considered.

"These drugs increase the excretion of glucose in the urine, which leads to urinary tract infections. And we could ask whether they are actually just very expensive diuretics — would we get the same effect with a low dose of furosemide or thiazide diuretic with fewer side effects and much lower cost?"

Nathan also pointed out that the glycemic effects of dapagliflozin were modest with a 0.4% reduction in hemoglobin A1c (HbA1c) (lowering HbA1c from 8.3% to 7.9%) in this trial. "This is not sufficient to satisfy the usual minimal FDA requirements for approval of a new diabetes medication."

"Whether these medications should more appropriately be viewed as treatments for heart failure in patients with diabetes, rather than glucose-lowering medications per se, is a not-so-subtle distinction that needs consideration," he concluded.

DECLARE-TIMI 58 was funded by Astra Zeneca and Bristol-Myers Squibb. Wiviott reports grants and personal fees from Astra Zeneca and Bristol-Myers Squibb.  Butler is a consultant to Astra Zeneca.

American Heart Association (AHA) Scientific Sessions 2018.  Abstract no. 19485. Presented November 10, 2018.

N Engl J Med. Published online November 10, 2018. Full text

For more from | 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.