EXSCEL: 'Disappointment' Spurs Discussion on GLP-1 Agonists in Diabetes

September 14, 2017

LISBON, PORTUGAL — Results of the large cardiovascular safety outcomes trial for AstraZeneca's once-weekly version of the glucagonlike peptide-1 (GLP-1) receptor agonist drug for type 2 diabetes, exenatide (Bydureon), showing that the agent met the goal of cardiovascular safety but failed to show any significant cardiovascular benefit, have been reported here at the European Association for the Study of Diabetes (EASD) 2017 Annual Meeting.

The study, top-line results for which were released earlier this year, was also simultaneously published in the New England Journal of Medicine.

The Exenatide Study of Cardiovascular Event Lowering (EXSCEL) involved 14,000 people with type 2 diabetes at a wide range of cardiovascular risk (73.1% had previous CVD), from 35 countries and randomized them to once-weekly sustained-release exenatide (2 mg subcutaneously) or placebo on top of usual care.

Putting the Findings Into Perspective

Lead author of the study, Rury R Holman, director of the University of Oxford Diabetes Trials Unit, United Kingdom, attempted to put the findings into perspective for attendees and to discuss the different agents within the GLP-1 agonist class that have so far reported cardiovascular-outcomes data.

These include the Evaluation of Lixisenatide in Acute Coronary Syndrome (ELIXA) trial, with lixisenatide (Lyxumia, Sanofi), the LEADER trial with liraglutide (Victoza, Novo Nordisk) and SUSTAIN-6 with the investigational agent, semaglutide (Novo Nordisk).

"The big question is, are the differences we see across [the cardiovascular-outcomes] trials real, or are they [the GLP-1 agonists] more similar than dissimilar?" he said.

There are certainly variations in the ways the trials were designed — for example, EXSCEL, he noted, is the "largest" CV-outcomes trial to date of the GLP-1 agonists and "the entry criteria were wider." It is 2.5 times larger than SUSTAIN-6.

And the patient populations differ. In ELIXA, all patients had recently suffered an acute coronary syndrome, so none of the patients were classified as primary prevention.

In addition, median follow-up differed between the trials — in EXSCEL, he admitted, one of the limitations is the shorter exposure to study drug, a mean of only 2.4 years compared with 3.5 years in LEADER with liraglutide.

This was driven by a shorter follow-up time and a higher rate of study drug discontinuation — there were particular issues with the complexity of the first-generation device used to deliver exenatide.

So "these are not insignificant differences in trial design and conduct," Dr Holman observed.

Nevertheless, Dr Holman believes, "the class as a whole is consistent in cardiovascular outcomes. The EXSCEL CV and mortality findings were consistent with those seen with other GLP-1 agonists in the ELIXA, LEADER, and SUSTAIN-6 trials, further supporting the use of these agents for the treatment of type 2 diabetes."

But the independent invited commentator, Francesco Giorgino of the endocrinology, andrology, and metabolic diseases department of the Universita Degli Studi di Bari, in Aldo Moro, Italy, had a slightly different take on things.

EXSCEL "adds to the knowledge of GLP-1 agonist therapy of type 2 diabetes, and we are left with some good news, but also some disappointment," he said.

The good news is that there is no cardiovascular risk with exenatide, no risk of hospitalization for heart failure, nor of pancreatic cancer or pancreatitis. There was a reduction in all-cause mortality and a "potential" cardiovascular benefit in a subgroup of patients — those aged 65 years and older.

Nonetheless, there was not a statistically significant reduction in the primary outcome, despite the fact that EXSCEL "resembles the picture of LEADER" more than any of the other GLP-1 agonist trials, he added.

But in EXSCEL, a reduction in all-cause mortality was not associated with a reduction in cardiovascular mortality, as was the case with liraglutide in LEADER. The latter led to the FDA recently approving the additional indication of reducing the risk for myocardial infarction (MI), stroke, and cardiovascular death in adults with type 2 diabetes who have established cardiovascular disease.

"The most important question — putting all these results together — is do these agents differ in their signaling capacity and biological effects?" mused Dr Giorgino.

"This is not an easy question, and the answer is still open," he said. "Although exenatide and liraglutide have similar properties, there is potential for some differences between the agents," he stressed.

EXSCEL, he concluded, "leads to more research needed to clarify these issues."

In general discussion after the results were reported, doctors said choice of GLP-1 agent would, in everyday life, depend on many practical characteristics, such as cost, type of device used to deliver the agent, and frequency of drug administration. Liraglutide, for example, must be administered once a day in contrast to the once-weekly schedule for exenatide.

Main Results of EXSCEL

EXSCEL was conducted in 14,752 patients worldwide, 25% of whom were recruited in North America, 18% from Latin America, 46% from Europe, and 10% from Asia Pacific. There were 685 study sites in 35 countries.

Baseline characteristics were well-balanced between the two study arms. The median age of participants was 62, and 38% were female. Most were obese, with a mean body mass index of 32 kg/m2, and they had had type 2 diabetes for a median of 12 years.

The aim of the trial was glycemic equipoise to examine the cardiovascular effects of exenatide, stressed M Angelyn Bethel, MD, of Oxford University, United Kingdom, who presented details of the participant characteristics.

Doctors could pretty much prescribe any other glucose-lowering agent to anyone in the trial other than another GLP-1 agonist, and there was no "run-in" period in EXSCEL, in contrast to most of the other GLP-1 agonist CV-outcomes trials.

Around 75% of all patients were using metformin, 35% were taking sulfonylureas, 45% were also using insulin, 12% DPP-4 inhibitors, and a "small" number were using sodium glucose cotransporter-2 (SGLT2) inhibitors.

Of the patients, 73% had prior cardiovascular disease, defined as a prior MI, stroke, coronary revascularization, significant coronary or carotid artery stenosis, or peripheral arterial disease (PAD). Congestive heart failure was not, however, considered as prior CVD, and 16% of patients in both the exenatide and placebo arms fell into this category.

The primary composite end point — of first occurrence of death from CV causes, nonfatal MI, or nonfatal stroke — occurred in 11.4% of patients taking exenatide over a median of 3.2 years compared with 12.2% of those taking placebo (hazard ratio 0.91 with intention-to-treat analysis; = .06 for superiority).

Presenting the cardiovascular-outcomes data, Adrian F Hernandez of Duke University, Durham, North Carolina, said EXSCEL "did not meet its primary efficacy end point," and the findings for secondary outcomes — rates of death from CV causes, fatal or nonfatal MI, fatal or nonfatal stroke, hospitalization for heart failure, and hospitalization for acute coronary syndrome — didn't differ significantly between the groups, either, and "were consistent with the primary outcome."

But exenatide did meet its primary safety hypothesis, and the incidence of acute pancreatitis, pancreatic cancer, medullary thyroid carcinoma, or any other serious adverse event did not differ between the exenatide and placebo groups.

And there was a hint of a benefit in terms of cardiovascular death in those aged > 65 years, said Dr Hernandez.

Reiterating their findings in their paper, the EXSCEL authors say: "The lack of CV efficacy in the current trial may be related to multiple factors. The median follow-up time was shorter [in EXSCEL] than that in the LEADER trial (3.2 years vs 3.8 years) as was the duration of exposure to the trial regimen (2.4 years vs 3.5 years); in addition, the baseline HbA1c in our trial was lower than in the LEADER trial (8.0% vs 8.7%), and the rate of discontinuation of the trial regimen was higher."

EXSCEL Dissatisfying as CV Benefit Is the New Norm in Type 2 Diabetes

Trials such as EXSCEL were simply designed to demonstrate cardiovascular safety of type 2 diabetes drugs in the wake of the rosiglitazone (Avandia, GlaxoSmithKline) scandal.

But over the past couple of years, doctors were surprised to find that some of these trials also demonstrated cardiovascular benefit, so this has now become the expectation as more and more of these studies are reported.

As well as the outcomes with other GLP-1 agonists, results of the landmark EMPA-REG trial — with a different class of type 2 diabetes drug, the SGLT2 inhibitors — was the first to show a reduction in cardiovascular death with a type 2 diabetes drug, empagliflozin (Jardiance, Boehringer Ingelheim/Lilly), and last December, the FDA allowed a label to be added to that product indicating that it can improve survival.

And more recently Canagliflozin Cardiovascular Assessment Study (CANVAS) showed a CV benefit for another SGLT2 inhibitor, canagliflozin (Invokana, Johnson & Johnson), in more than 10,000 patients with type 2 diabetes and a history of or at high risk for cardiovascular events — although this came at the cost of a greater risk of amputations with canagliflozin.

The full slide set for EXSCEL is available to download at www.exscel-study.org.

AstraZeneca sponsored EXSCEL. Dr Holman reports grants from AstraZeneca, during the conduct of the study; grants and personal fees from Bayer, personal fees from Novartis, grants and personal fees from Boehringer Ingelheim, personal fees from Amgen, other from Elcelyx, other from GlaxoSmithKline, other from Jannsen, personal fees from Servier, other from Takeda, grants and personal fees from Merck, outside the submitted work. Disclosures for the coauthors are listed on the journal website.

N Engl J Med. Published online September 14, 2017. Article

Follow Lisa Nainggolan on Twitter: @lisanainggolan1. For more diabetes and endocrinology news, follow us on Twitter and on 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.