Albiglutide cut cardiovascular events in Harmony Outcomes, but the drug was already off the market

Once-weekly albiglutide lowered the risk of cardiovascular death, heart attack or stroke by about 22% in adults with type 2 diabetes who already had heart disease, according to the Harmony Outcomes trial published in the Lancet in October 2018 ^[1]. The finding arrived too late to help the drug commercially: albiglutide was no longer being sold, and its European marketing authorization was formally withdrawn on October 29, 2018 ^[2].

The trial was a double-blind, randomized, placebo-controlled study run at 610 sites in 28 countries. It enrolled 9,463 patients aged 40 and older who had type 2 diabetes and established cardiovascular disease. Participants were assigned in a 1:1 ratio to receive either a weekly under-the-skin injection of albiglutide, at 30 to 50 mg based on blood sugar response and tolerability, or a matching placebo injection, in addition to their standard care ^[1].

Screening ran from July 1, 2015, to November 24, 2016. Researchers screened 10,793 people, and 4,731 were assigned to albiglutide and 4,732 to placebo. Patients were followed for a median of 1.6 years. The last patient visit was on March 12, 2018 ^[1].

The primary outcome was the first occurrence of cardiovascular death, heart attack or stroke. It occurred in 338 patients (7%) taking albiglutide, an incidence rate of 4.6 events per 100 person-years, and in 428 patients (9%) taking placebo, or 5.9 events per 100 person-years. That worked out to a hazard ratio of 0.78, with a 95% confidence interval of 0.68 to 0.90. Albiglutide was first shown to be non-inferior to placebo (p<0.0001) and then superior (p=0.0006) ^[1].

Serious side effects did not differ significantly between the groups. Acute pancreatitis was reported in ten patients taking albiglutide and seven taking placebo; pancreatic cancer in six and five patients, respectively; and medullary thyroid carcinoma in zero patients in either group. Investigators who did not know which treatment patients had received judged three deaths (less than 1%) in the placebo group and two deaths (less than 1%) in the albiglutide group to be related to treatment ^[1]. The trial was funded by GlaxoSmithKline ^[1].

The authors concluded that albiglutide was superior to placebo for major adverse cardiovascular events, and wrote that evidence-based GLP-1 receptor agonists should be considered as part of a broader strategy to lower cardiovascular risk in people with type 2 diabetes ^[1]. They also noted that drugs in this class differ in chemical structure, how long they last in the body, and their effects on clinical outcomes ^[1].

Why it matters for patients

Albiglutide itself is not an option. GlaxoSmithKline Trading Services Limited asked the European Commission to withdraw the product, sold in Europe as Eperzan, and told the commission it was discontinuing it permanently for commercial reasons ^[2]. The drug had held European authorization since March 21, 2014 ^[2].

The result still matters for how the wider GLP-1 class is understood. The people in this trial all had type 2 diabetes plus established cardiovascular disease, so the findings apply to that group and not necessarily to people without heart disease or to people using these medicines only for weight loss. The trial did not test albiglutide against another GLP-1 drug, so it does not show that albiglutide was better or worse than any competitor.

It also does not mean every GLP-1 medicine carries the same heart benefit. The trial authors explicitly wrote that drugs in this class differ in structure, duration of action and clinical outcomes ^[1]. Whether the 22% reduction reflects a shared class effect or something specific to albiglutide has not been established by this trial. The semaglutide medicines Ozempic, Wegovy and Rybelsus, and the tirzepatide medicines Mounjaro and Zepbound, are different molecules and were not part of Harmony Outcomes.

Because albiglutide is off the market, no patient can be switched to it based on these results. Whether any regulatory body used the Harmony Outcomes data to change labeling for other GLP-1 drugs has not been reported.