Daily liraglutide lowered A1C more than weekly exenatide in head-to-head trial

A 26-week head-to-head trial found that a once-daily injection of liraglutide 1.8 mg lowered HbA1c by 1.48 percentage points in adults with type 2 diabetes, while a once-weekly injection of exenatide 2 mg lowered it by 1.28 points. The difference failed the study's preset test for non-inferiority, and the results appeared in The Lancet in January 2013 ^[1].

The study, called DURATION-6, ran at 105 sites in 19 countries between Jan. 11, 2010, and Jan. 17, 2011. It enrolled adults 18 and older whose type 2 diabetes was being treated with lifestyle changes and oral glucose-lowering drugs. Participants were assigned at random, in a 1-to-1 ratio, to liraglutide once daily or exenatide once weekly. Neither participants nor investigators were masked to which treatment was given. Of 912 randomized patients, 911 were included in the intention-to-treat analysis: 450 on liraglutide and 461 on exenatide ^[1].

The main endpoint was change in HbA1c, a measure of average blood sugar over roughly the previous two to three months, from the start of the trial to week 26. The estimated average change was a drop of 1.48 percentage points with liraglutide (standard error 0.05; n=386) and 1.28 points with exenatide (0.05; 390). The treatment difference was 0.21 points, with a 95 percent confidence interval of 0.08 to 0.33. The trial had defined non-inferiority in advance as an upper limit of that interval below 0.25 percent, a bar the result did not meet ^[1].

Side effects split the other way. Nausea was the most common adverse event, reported by 93 people (21 percent) taking liraglutide and 43 (9 percent) taking exenatide. Diarrhea occurred in 59 (13 percent) versus 28 (6 percent), and vomiting in 48 (11 percent) versus 17 (4 percent). These events were less frequent with exenatide and became less common over time in both groups. Twenty-four patients on liraglutide (5 percent) and 12 on exenatide (3 percent) stopped participating because of adverse events ^[1].

The trial authors wrote that both drugs improved blood sugar control, with larger reductions seen with liraglutide, and that differences in injection frequency and tolerability "could inform therapeutic decisions for treatment of patients with type 2 diabetes." The study was funded by Eli Lilly and Company and Amylin Pharmaceuticals LLC and registered on ClinicalTrials.gov as NCT01029886. It published alongside a Lancet commentary titled "Comparing diabetes drugs, helping clinical decisions?" ^[1].

Why it matters for patients

For an adult with type 2 diabetes weighing a GLP-1 medicine, this trial illustrates the kind of tradeoff that head-to-head comparisons are meant to expose. The blood-sugar difference favored liraglutide, but it was small, 0.21 percentage points. The gap in nausea, diarrhea and vomiting favored exenatide, and exenatide required one injection a week instead of one a day. Roughly one in five people on liraglutide reported nausea, about twice the rate seen with exenatide ^[1].

The results do not establish which drug any individual should use. The trial authors describe the findings only as information that could inform treatment decisions, and they note the two drugs differ in both injection frequency and tolerability ^[1]. The study also does not describe how either drug compares with semaglutide or tirzepatide, which were not part of this trial.

What the trial did not answer

The published results report A1C, injection frequency and the listed side effects, and they note in the background that GLP-1 receptor agonists have been shown to improve blood sugar control and reduce body weight in people with type 2 diabetes ^[1]. Weight change during DURATION-6 is not reported. Nor do the results include longer-term outcomes such as heart attacks, strokes or deaths, and whether the findings apply equally across age groups, races and ethnicities has not been reported.