HARMONY: Albiglutide Cuts CV Events but Not Deaths in Diabetes

Liam Davenport

October 03, 2018

BERLIN —The glucagonlike peptide-1 (GLP-1) receptor agonist albiglutide (Tanzeum, GlaxoSmithKline) significantly reduces major adverse cardiovascular events (MACE) in type 2 diabetes patients with established cardiovascular disease compared with placebo without any additional safety concerns, according to the results of the latest post-approval cardiovascular outcomes study of a diabetes drug.

The HARMONY Outcomes study of almost 9500 patients treated with the drug or placebo for more than one and a half years against a background of standard diabetes and cardiovascular disease medication showed that the benefit was largely driven by a significant reduction in myocardial infarctions (MIs).

The results were presented during a well-attended session dedicated to the trial here at the European Association for the Study of Diabetes (EASD) 2018 Annual Meeting and were published simultaneously in the Lancet. The study was led by Adrian F. Hernandez, MD, Duke Clinical Research Institute, Duke University School of Medicine, Durham, North Carolina.

To spontaneous applause at EASD, John J. V. McMurray, MD, British Heart Foundation Glasgow Cardiovascular Research Center, University of Glasgow, United Kingdom, showed a slide indicating that albiglutide reduced the rate of MACE events by a significant 22% over placebo.

"These results...add to the evidence that certain GLP-1 receptor agonists reduce cardiovascular events in patients with type 2 diabetes," commented session chair Stefano Del Prato, MD, Department of Clinical and Experimental Medicine, University of Pisa, Italy, in a press release.

GlaxoSmithKline ceased all manufacturing and sales of albiglutide in 2017 for commercial reasons, but has now said it will seek a new home for the drug.

Lawrence A. Leiter, MD, from the Li Ka Shing Knowledge Institute, St. Michael's Hospital, University of Toronto, Ontario, Canada, told delegates the cardiovascular results of HARMONY are comparable to those seen with other GLP-1-based agonists, including liraglutide (Victoza, Novo Nordisk), in the LEADER trial and semaglutide (Ozempic, Novo Nordisk) in SUSTAIN-6, although there was no reduction in CV death with albiglutide as was seen in LEADER.

In contrast, the ELIXA and EXSCEL trials with lixisenatide (Lyxumia, Sanofi) and exenatide (Bydureon, AstraZeneca) did not show a significant reduction in the MACE event rate, which he suggested may be because these compounds are extendin-4 based. However, Leiter said that "one has to be cautious comparing across trials because, of course, these were not head-to-head" studies.

HARMONY "adds further support that evidence-based GLP-1 receptor agonists should be part of a comprehensive strategy to reduce the risk of cardiovascular events in type 2 diabetes, as recommended in recent cardiology and diabetes guidelines," Leiter asserted.

Summarizing, he said: "We found compelling evidence of MACE benefit, highly statistically significantly — importantly, despite only modest metabolic improvements. So we think that our findings highlight the fact that the cardiovascular benefit of this class of medication likely are above and beyond any metabolic effects of these drugs."

Nevertheless, he noted that there are some potential limitations of the study.

"The duration of follow-up was just 1.6 years," he explained, "which may have reduced our ability to investigate the effect on long-term microvascular endpoints, as well as on mortality, and the lack of measurement of lipids and urinary albumin excretion limit our ability to explore mechanistic explanations of the results."

GSK Commits to Finding a New Home for Albiglutide

GlaxoSmithKline ceased all manufacturing and sales of albiglutide in 2017 but made a commitment to continue with HARMONY to its conclusion. Following the results, GSK announced in a press release "that it will continue to seek a way to make albiglutide available to patients through another company," Leiter noted.

In an invited commentary, David R. Matthews, MD, MD, PhD, professor of diabetic medicine at the University of Oxford, said that the trial was "extremely well conducted" and "beautifully positive."

He added: "Your applause demonstrates that it was a successful trial; in other words, there was a 'win' over the major adverse cardiovascular events."

He continued by saying that trials such as these "are 'gamed,' and these investigators have gamed rather nicely."

HARMONY recruited patients who were "extremely near the edge...you're desperate for people to have events, you've got 1.6 years, and you want people to fall off the cliff. They mustn't be too near because then the therapy makes no difference...but not too far because otherwise the trial goes on and on and on, and the investigators die before the patients. So you have to bear this in mind...when...making generalizations."

Matthews also pointed out that in the current trial, patients did not achieve anywhere near the weight loss seen in previous trials with GLP-1 agonists, such as semaglutide, which he suggested may be related to differing effects of different agents in the class related to the blood-brain barrier.

This, he argued, means that albiglutide may be helpful in patients for whom weight loss is not desirable, such as young patients with type 1 diabetes.

Matthews added that GlaxoSmithKline's decision to continue to support the trial after their decision to stop manufacturing the drug "is highly ethical behavior, and GSK need to be congratulated on taking this stance."

No Reduction in CV Death With Albiglutide

In HARMONY, 9463 type 2 diabetes patients from 610 sites across 28 countries were randomly assigned to receive either a once-weekly subcutaneous injection of 30 to 50 mg of albiglutide (n = 4731) or a similar placebo injection (n = 4732), on top of standard, adjustable diabetes and cardiovascular care that followed local guidelines and practices. All diabetes drugs were allowed, including DPP-4 inhibitors and SGLT-2 inhibitors, with the exception of GLP-1 agonists.

The mean age of the patients was 64 years, about 70% were male, and a similar proportion were white. The mean duration of diabetes was 14 years, the average HbA1c level was approximately 8.7%, and the mean body mass index was 32 kg/m2.

Approximately 70% of the patients had coronary artery disease; 47% had already had an MI; and approximately 44% had undergone a percutaneous coronary intervention. Stroke was reported in around 17% of patients, and approximately 25% had peripheral artery disease. Furthermore, 20% had heart failure, and 86% had hypertension.

After a median follow-up of 1.6 years, complete data were available for 4619 patients who had received albiglutide and 4577 patients who had received placebo.

McMurray presented the cardiovascular outcomes, starting with the primary outcome of time to MACE, defined as cardiovascular death, MI, or stoke.

"After about 8 months of follow-up there was a separation in these event curves, and that separation grew wider as the trial progressed," he said. He noted that albiglutide was associated with a significant reduction in events, at 338, compared to 418 with placebo.

This translated into a hazard ratio of 0.78 (noninferiority, P < .0001; superiority, P = .0006). The event rate per 100 person-years was 4.57, vs 5.87 for placebo.

Further analysis showed that the improvement in the MACE event rate was driven by a reduction in the rate of fatal and nonfatal MI, at a hazard ratio of 0.75 (95% confidence interval, 0.61 - 0.90), rather than any reduction in stroke or cardiovascular death.

Overall, the number of patients who would need to be treated with albiglutide to prevent one CV event over a median of 1.6 years was 50.

In terms of the secondary endpoints, there was a significant reduction in the MACE or urgent revascularization event rate with albiglutide vs placebo, at a hazard ratio of 0.78 (P < .001), but the difference for cardiovascular death or heart failure hospitalization and for all-cause death was not significant.

In their Lancet article, the HARMONY investigators acknowledge: "We did not observe a significant reduction in death from cardiovascular causes, which was noted [with liraglutide] in the LEADER trial."

However, "The median duration of follow-up in our trial was considerably shorter than the 3.8 years in the LEADER trial, and it might be that an effect on death from cardiovascular causes requires additional time to accrue."

Other Outcomes: Weight Loss Less Than With Other GLP-1 Agonists

Presenting the metabolic outcomes, Del Pato said that, overall, the composite endpoint of an HbA1c level at the final visit was ≤7%, there were no severe hypoglycemic episodes, and weight gain <5% was significantly more common with albiglutide than placebo, at 26.0% vs 15.1% (P < .001).

As already noted, weight loss appears to be less with albiglutide than with some of the other GLP-1 agonists — at 8 months, albiglutide was associated with a significant but relatively small weight loss vs placebo of -0.66 kg (P < .001); the gap narrowed after 16 months and was nonsignificant at the final assessment.

There was no overall improvement in the estimated glomerular filtration rate with albiglutide vs placebo.

However, the study drug was associated with a significant reduction in the time to first clinically important microvascular event, at a hazard ratio vs placebo of 0.66 (P = .055), which was driven by a substantial reduction in diabetic retinopathy events.

"In summary, as far as the diabetes-related outcomes are concerned," Del Prato concluded, "we can say that albiglutide treatment added on to current antihyperglycemic therapy was associated with better glycemic control, with modest weight loss, with reduced need for initiation of insulin, with a lower risk of severe hypoglycemia and no excess of clinically important microvascular events, namely renal and eye events."

Christopher B. Granger, MD, from Duke Clinical Research Institute, presented the safety outcomes, focusing on noncardiovascular events and on nonserious adverse events only if they were of special interest.

There was no overall difference in the rate of serious adverse events between the patients who received albiglutide and those who received placebo, at 19.8% vs 21.7%.

Study drug injection site reactions were the only adverse events of special interest that were significantly more common with the study drug. Rates of acute pancreatitis, pancreatic cancer, and medullary thyroid carcinoma were similar.

Trial GLP-1 Agonists in Nondiabetic Patients

In a commentary accompanying the Lancet article, Marion Mafham, MD, and David Preiss, PhD, of the Medical Research Council Population Health Research Unit, Clinical Trial Service Unit and Epidemiological Studies Unit, Nuffield Department of Population Health, University of Oxford, note that the "increasing weight of evidence" in support of GLP-1 agonists in patients with diabetes and cardiovascular disease should be reflected in international guidelines.

"Further studies investigating their effects in people with diabetes but without known vascular disease are now a priority to better inform the clinical community in the selection of glucose-lowering drugs in such patients," the authors state.

Mafham and Preiss underline, however, that these studies will need to be highly streamlined and innovative to deliver results in a cost-effective manner and be long enough to reveal emerging benefits and hazards, while ensuring good adherence.

They also point out that the lack of relationship between glycemic control and cardiovascular outcomes warrants the testing of GLP-1 agonists in nondiabetic patients with cardiovascular disease. One such trial of semaglutide in obese or overweight individuals without diabetes is already underway.

"Notably, substantial weight loss has been achieved in obese nondiabetic individuals who are treated with high doses of some GLP-1 receptor agonists, supporting the potential usefulness of these drugs in populations without diabetes," they write.

They conclude that, "given the clear cardiovascular benefit observed with albiglutide in the Harmony Outcomes trial, GlaxoSmithKline should reconsider making it available to patients."

The research was supported by GlaxoSmithKline Research and Development Limited. Consistent with an announcement on July 26, 2017, GlaxoSmithKline has ceased further research and development, manufacturing, and sales activity for albiglutide. The authors' relevant financial relationships are listed in the Lancet article. Dr Preiss is an investigator on a trial of a glucose-lowering drug that is funded by a grant to the University of Oxford by Boehringer Ingelheim, but he receives no personal payment. Dr Mafham has disclosed no relevant financial relationships.

European Association for the Study of Diabetes (EASD) 2018 Annual Meeting. HARMONY Outcomes study, presented October 2, 2018.

Lancet. Published online October 2, 2018. Abstract, Commentary

For more diabetes and endocrinology news, please 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.