LISBON -- Extended-release exenatide (Bydureon) was safe for the heart but didn't help prevent cardiovascular events in type 2 diabetes among a mixed primary and secondary prevention population, the EXSCEL trial showed.
For the primary composite endpoint of death from cardiovascular causes, nonfatal myocardial infarction, or nonfatal stroke, the once-weekly GLP-1 diabetes drug fell just short of showing a cardiovascular benefit, Rury Holman, FRCP, of the Diabetes Trials Unit at the University of Oxford in England, and colleagues reported here at the European Association for the Study of Diabetes meeting and in a paper published simultaneously in the New England Journal of Medicine.
Rates for the composite after median follow-up of 3.2 years were 11.4% with the drug versus 12.2% with placebo (HR 0.91, 95% CI 0.83-1.00), each used along with usual care. That translated to noninferiority for safety (P<0.001 for noninferiority) but missed superiority for efficacy (P=0.06 for superiority), as had been announced in topline resultsreleased in May. Per-protocol analyses were even farther from showing superior efficacy.
All-cause mortality nominally supported exenatide (6.9% versus 7.9% with placebo; HR 0.86, 95% CI 0.77-0.97). However, "this difference was not considered to be statistically significant on the basis of the hierarchical testing plan," the researchers cautioned in the NEJM paper.
Other endpoints likewise came out similar between treatment arms, including fatal or nonfatal myocardial infarction and stroke and hospitalization for heart failure and for acute coronary syndrome. Overall serious adverse events weren't elevated with exenatide, nor were acute pancreatitis, pancreatic cancer, or medullary thyroid carcinoma. But as expected, heart rate was elevated with the drug by an overall least squares mean difference of 2.51 bpm.
As to why this trial showed neutrality when two other GLP-1 drugs have shown superiority in their large cardiovascular outcomes trials, the researchers pointed out that "the direction and magnitude of the cardiovascular outcomes observed were not inconsistent with those seen in the LEADER and SUSTAIN-6 trials," which tested liraglutide and semaglutide, respectively.
Holman emphasized the similarities over the differences. "It's just that the EXSCEL trial had shorter duration of drug exposure [3.2 versus 3.8 years] than the LEADER trial, which is the other large study, and maybe if the trials had had similar exposure we may have had almost the same result."
SUSTAIN-6 had the shortest drug exposure time and was smallest of the three trials, but still showed cardiovascular prevention. Holman argued that EXSCEL fit closest to LEADER in size, population, and other characteristics.
His group acknowledged, though, that the "four GLP-1 receptor agonists evaluated to date may not all be bioequivalent," noting neutrality in the ELIXA trial with lixisenatide (Adlyxin).
Study discussant Francesco Giorgino, MD, PhD, of the University of Bari Aldo Moro in Bari, Italy, said there is reason to think that this might be the case with "pleiotropic" effects on the cardiovascular system. For example, amino acid residues critical for binding and signaling GLP-1 show the potential for differences among the GLP-1 agnoist drugs.
Whatever the reason, the take away message should be that there is "solid evidence the GLP-1 class is safe and efficacious," Holman said.
"Of course, companies and people with opinions will be trying to sway the formulary and things. That will always happen. If a trial is put in a meta-analysis and there's no difference between them, it should be a class effect," he told MedPage Today. "It won't go on the label. But there's a difference between a label and the way the physicians use the drug ... If it's not on the label and people feel that the class is similar and maybe there's a price difference or a convenience difference, then the physicians may chose to do differently. That's a physician choice. It's more complicated than just being on the label."
The pragmatic, multinational trial included 14,752 type 2 diabetes patients (of whom 73.1% had a history of cardiovascular disease) treated in the usual care setting. They were randomized to the addition of 2-mg extended-release exenatide in a once-weekly injection or matched placebo to usual care management of diabetes and cardiovascular risk factors (only GLP-1 receptor agonists were excluded) with a median 3.2 years of follow-up.
A key limitation of the trial was the high premature, permanent discontinuation rate -- 43% with exenatide and 45% in the placebo group -- driven primarily by patient decision (10% and 13% didn't want to give injections). "We speculate that probable factors for discontinuation were the complexity of the first-generation injection device that was used and the fact that our trial had no run-in period," the researchers noted.
That early syringe had to be mixed before injection and there was some pain associated with injection, Holman told MedPage Today. However, Holman pointed out that proportion of time on drug was still 75%, and mean drug exposure time was about 2.4 years.
www red DiabetologNytt