After EMPA-REG: Docs Dissect Diabetes Study, Discuss Implications

September 24, 2015

STOCKHOLM — Experts in diabetes and cardiology are both excited and admittedly somewhat perplexed by the results of the EMPA-REG OUTCOME study reported at the European Association for the Study of Diabetes (EASD) 2015 Meeting last week and published simultaneously in the New England Journal of Medicine.

Hailed as historic, the findings were presented to a packed hall in Stockholm, where audible gasps and spontaneous applause erupted each time Silvio Inzucchi, MD, of Yale Diabetes Center, New Haven, Connecticut, showed a slide demonstrating how the survival curves for the high-risk type 2 diabetes patients taking empagliflozin (Jardiance, Lilly/Boehringer Ingelheim) diverged from those on placebo over the course of the 3-year trial, with the effect starting very early, at 3 months.

In the study, 39 patients needed to be treated with empagliflozin for 3 years to prevent one cardiovascular death.

This is the first time a glucose-lowering drug has shown superiority in a cardiovascular-outcomes trial — similar trials with diabetes drugs from other classes have demonstrated only neutrality (ie, cardiovascular safety), including most recently the TECOS trial with the DPP-4 inhibitor sitagliptin (Januvia, Merck) and the ELIXA study with the injectable glucagonlike receptor (GLP-1) agonist lixisenatide (Lyxumia, Sanofi).

Dr Inzucchi and the other investigators urged prudence, however, noting that they had only first seen the results of EMPA-REG "about 4 weeks ago" and that extensive work is still needed to try to unravel the mechanisms by which this sodium glucose cotransporter-2 (SGLT-2) inhibitor is exerting its seemingly beneficial effects.

And, he stressed, the trial population studied in EMPA-REG — type 2 patients with existing cardiovascular disease — is not representative of all patients with the condition. "We have to be cautious. We can't cut and paste these expectations to the diabetes population as a whole."

Asked to comment by Medscape Medical News, Anne Peters, MD, of University of Southern California, Los Angeles, said: "This study suggests major nonglycemic benefit from a diabetes drug, which is something we've always sought but had trouble finding. The reduction in overall death rates is very impressive."

"However, this study raises questions, as well, and we still need to individualize therapy," she said, noting for example that "3 years' worth of follow-up still doesn't give us long-term data in terms of the effects of SGLT-2 inhibitors on bone."

These findings are unique in terms of diabetes clinical trials.

There has been a recent warning from the US Food and Drug Administration (FDA) regarding increased fracture risk with another SGLT2 inhibitor, canagliflozin (Invokana, Janssen), although no increase in fracture was seen with empagliflozin vs placebo in EMPA-REG out to the 3 years of follow-up. There have also been warnings about euglycemic diabetic ketoacidosis with SGLT2 inhibitors, but again no increase in this serious adverse event was seen with empagliflozin in this trial.

Also, EMPA-REG is a secondary-prevention trial, Dr Peters observed, adding — "it is not known if there are similar benefits in primary prevention."

Nevertheless, "these findings are unique in terms of diabetes clinical trials and may serve to guide use of empagliflozin in patients who have clinical characteristics similar to those studied in this trial," she commented.

More Questions Than Answers: Is It a Class Effect?

In EMPA-REG OUTCOME, just over 7000 patients received state-of-the-art standard of care in the form of blood-pressure lowering and dyslipidemia medication, as well as optimal glucose-lowering therapy with agents other than SGLT2 inhibitors, including insulin if required. They were then randomized to one of two doses of empagliflozin (10 mg or 25 mg) or placebo.

Those taking empagliflozin (the findings for the two doses were pooled) were 38% less likely to die from cardiovascular disease and 32% less likely to die from any cause over the 3 years of the study.

But because empagliflozin did not seem to have any effect on heart attacks or strokes, the investigators are a little mystified as to how it can be superior to placebo in terms of mortality.

Many theories were proposed, with most coming down on the side of an osmotic-diuretic effect of the drug, as there was also a significantly positive effect of the agent on heart-failure hospitalizations, which were reduced by 35% among those taking the active drug compared with placebo.

A Lilly/Boehringer Ingelheim–sponsored symposium that immediately followed the conference presentation was packed with meeting attendees, too, and featured four eminent physicians in the field of cardiology and metabolic medicine.

Yet even they could not provide answers to the numerous questions submitted by the audience — including whether this is a class effect of all SGLT2 inhibitors, what dose of empagliflozin should be prescribed, and whether diabetes guidelines should be changed as a result of the study.

On the issue of a class effect, all the experts agreed that the other large cardiovascular-outcomes trials with other SGLT2 inhibitors should not be stopped early because of the EMPA-REG findings.

"It probably is a class effect," commented Naveed Sattar, MD, of University of Glasgow, Scotland, "but until we complete the other trials we won't know the answer."

These trials include the Canagliflozin Cardiovascular Assessment Study (CANVAS), which is randomizing just over 4000 type 2 diabetes patients to canagliflozin 100 mg or 300 mg daily or to placebo and is now slated for completion in 2017, and the DECLARE-TIMI 58 study with dapagliflozin (Farxiga/Forxiga, AstraZeneca).

A disabling stroke is perhaps a worse outcome than dying. I hope they will not stop the ongoing [other SGLT2 inhibitor] trials.

The latter is set to enroll more than 17,000 type 2 diabetes patients who will be randomized to either dapagliflozin 10 mg daily or placebo and followed for 4 to 5 years and is expected to report in 2019.

Cardiologist Sanjay Kaul, MD, of Cedars Sinai Medical Center, Los Angeles, CA, said: "I hope they don't take any decision in haste. Diabetes drugs have a very checkered history when it comes to cardiovascular outcomes, and there is also a concern for a stroke signal in EMPA-REG [there was a nonsignificant increase in fatal/nonfatal stroke observed with empagliflozin vs placebo (hazard ratio [HR], 1.18; P =.26)]."

And one could argue, said Dr Kaul, "that a disabling stroke is perhaps a worse outcome than dying. I hope they will not stop the ongoing trials — it's imperative that they allow them to continue."

How Is Empagliflozin Working? "We Simply Don't Know"

Musing on what could be behind the startling effect on mortality observed with empagliflozin in EMPA-REG, Dr Sattar said, "The fact that nonfatal disease (MI and stroke) didn't change means this drug is changing something else. It could be a diuretic effect on the background of some small effects on blood pressure and weight loss. Or is it cardiac remodeling? I don't think any of us has the answer."

Dr Kaul said he doesn't believe the effect of empagliflozin was mediated via blood-pressure lowering, "because if so, why is stroke going in the wrong direction?

"And we know this is not a glycemic effect, and it's not weight loss — when was the last time we saw such a profound and early benefit [due to slight weight loss] with such a treatment?"

Others commented on the discordance between the major adverse cardiovascular events (MACE) components of the primary outcome. Indeed, Dr Kaul wondered "whether MI, stroke, and CV death are a good composite end point."

During his presentation of the results, Dr Inzucchi said he thought the beneficial effect of empagliflozin was mediated "through osmotic diuresis; I think we are treating patients at a subclinical level."

But the real answer to how this is happening, said Dr Kaul, is that "we simply don't know. But the mortality benefit we are seeing is virtually double that of a recently approved heart-failure drug….It's hard to put our finger on it."

It may also be that there is some interaction with older age and lots of ventricular dysfunction, for example — but although it would seem appealing to perform subgroup analysis looking for explanations, this is "tempting but treacherous," Dr Kaul advised.

Everyone present stressed how important it is that analyses continue to be able to much better understand which diabetes patients will benefit.

Cardiologist Lars Ryden of the Karolinska Institute, Stockholm, Sweden, said: "We need to do very specific trials on this drug to be able to explain the effects. There will be a number of such trials, and the beauty of EMPA-REG is that it opens up the area."

We need to do very specific trials on this drug to be able to explain the effects.

Dr Inzucchi agreed, noting this "opens up a whole new field of investigation in diabetes."

Dr Sattar concurred: "It will be very important to know where we will use this drug. Are there other groups we should be targeting? Those patients at high risk of heart failure, for example?"

Indeed, Dr Ryden said the effects of empagliflozin were so impressive that he believes it must now be tested in heart-failure patients who don't even have diabetes.

What Dose of Empagliflozin to Start With?

Another frequently asked question was, what dose of empagliflozin should doctors start patients on?

Official discussant of the EMPA-REG results, Hertzel Gerstein, MD, from McMaster University and Hamilton Health Sciences, Ontario, said there was "no evidence of a dose-response relationship" with empagliflozin in the trial.

Dr Kaul said as there was no appreciable difference between the cardiovascular outcomes between the two different doses tested, "I would start every patient on 10 mg and stop there."

Dr Sattar agreed. "It's true that you get more bang for your buck at 10 mg."

Lawrence Leiter, director of the lipid clinic at St Michael's Hospital, Toronto, Canada, remarked, "The cardiovascular benefit was no greater at 25 mg, but the higher dose gave a little bit more HbA1c lowering, a little bit more BP lowering, and a little bit more weight loss."

And Dr Ryden added, "In my practice, I would start patients at 10 mg, but if I were to do a study, I would go with 25 mg."

In my practice, I would start patients at 10 mg.

What's also not yet clear is what impact empagliflozin is having on the kidneys. Although SGLT2 inhibitors are contraindicated in those with mild renal impairment — they are generally not recommended in those with estimated glomerular filtration rate (eGFR) of < 45 mL/min/1.73 m2 — patients were only excluded from EMPA-REG if their eGFR was less than 30 mL/min/1.73 m2.

Dr Gerstein noted "there were a lot of people with renal dysfunction in EMPA-REG….By every measure, this was a high-risk group of patients."

EMPA-REG investigator David Fitchett, MD, of the University of Toronto, Ontario, reported on the safety outcomes of the trial and told the meeting that "serum creatinine and eGFR were not changed by empagliflozin, and electrolytes didn't alter either."

Dr Sattar said the full analysis on renal outcomes from EMPA-REG has not yet been performed, adding, "It will be nice to see the microvascular outcomes."

Dr Leiter observed that, in people with impaired renal function, it appears that the glucose-lowering effect of SGLT2 inhibitors "falls off [in time]. But the BP lowering effect doesn't."

Recent research has even suggested that SGLT2 inhibitors may protect the kidneys, and this is being investigated with canagliflozin in the Evaluation of the Effects of Canagliflozin on Renal and Cardiovascular Outcomes in Participants With Diabetic Nephropathy (CREDENCE) trial, which aims to enroll 3700 participants, and the Study of the Effects of Canagliflozin on Renal Endpoints in Adult Participants With Type 2 Diabetes Mellitus (CANVAS-R), which is planning almost 6000 participants.

And Should Guidelines Alter?

The question on everyone's lips was whether diabetes guidelines should be changed in light of the EMPA-REG findings, and if so, how quickly.

Dr Gerstein ventured that empagliflozin "may become first-line therapy for middle-aged people with type 2 diabetes at [high] risk of cardiovascular events" and/or combination treatment with empagliflozin may move up the pyramid of recommendation.

A combination product of empagliflozin with metformin, which is generally regarded as first-line therapy for type 2 diabetes around the world, is already available, with the brand name Synjardy (Boehringer Ingelheim/Eli Lilly).

Lead investigator of EMPA-REG, Bernard Zinman, MD, director, Diabetes Centre, Mount Sinai Hospital, Toronto, Ontario, told meeting attendees: "Guidelines are evidence-based. My colleagues and I are really excited to see how the evidence we've presented today will affect diabetes guidelines" worldwide.

And finally, Dr Leiter noted that the EMPA-REG results at least somewhat vindicate the US FDA and its 2008 decision to mandate such cardiovascular-outcomes trials with diabetes drugs, a ruling that has been criticized in many quarters.

"Despite the fact that everyone dumped on the FDA, in a few years we will have the answer," he concluded.

The EMPA-REG study was supported by Boehringer Ingelheim and Eli Lilly. Drs Sattar, Leiter, Kaul, and Ryden were speaking during a Boehringer Ingelheim /Lilly–sponsored symposium. Dr Inzucchi has served as a consultant for Merck, Janssen, Transtech Pharma, and Regeneron/Sanofi; lectured for AstraZeneca; received research support from the National Institute of Diabetes and Digestive and Kidney Diseases, the National Institute of Neurological Disorders and Stroke, and Takeda; has served on clinical-trial steering committees for Boehringer Ingelheim, Lexicon, and Eisai; and served on data safety and monitoring boards for Novo Nordisk and Intarcia. Dr Zinman has received grant support to his institution and consulting fees from Boehringer Ingelheim, Merck, and Novo Nordisk and personal fees from Sanofi, Eli Lilly, Takeda, AstraZeneca, and Janssen outside the submitted work. Dr Gerstein has previously disclosed being on an executive steering committee and receiving travel grants and compensation for working in the committee from F Hoffmann–La Roche.


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.