STOCKHOLM – Patients with poorly controlled type 2 diabetes who take the combination of liraglutide and metformin are about four times more likely to hit their glycemic target than those who take lixisenatide and metformin.
After 26 weeks of treatment, hemoglobin A1cdropped a mean of 1.8% among those on the liraglutide regimen, compared with 1.2% among those taking lixisenatide. The change brought 74% of the liraglutide group into a treatment target of HbA1c less than 7%; 45% of the lixisenatide group reached that goal (odds ratio, 4.16).
Dr. Michael A. Nauck
Those taking liraglutide also were almost four times as likely to reach a target of 6.5% or less (55% vs. 26%; OR, 3.6), Dr. Michael A. Nauck said at the annual meeting of the European Association for the Study of Diabetes.
Both of the drugs are glucagon-like peptide-1 receptor agonists, and both are formulated for once-daily injection, but liraglutide’s 13-hour half-life gives it a strong clinical edge.
“Lixisenatide has a 3-hour half-life, so it’s only going to be around for 6 to a maximum of 10 hours after injection,” said Dr. Nauck, head of the Diabeteszentrum in Bad Lauterberg, Germany. “For the most part of the night there’s no exposure to lixisenatide at all, while liraglutide is present 24 hours.”
The study randomized 404 patients with poorly controlled type 2 diabetes to metformin plus either liraglutide 1.8 mg/day or lixisenatide 20 mcg/day, he reported. Patients were a mean of 56 years old, with a mean body mass index of 35 kg/m2. Their mean HbA1c was 8.4%.
At the end of the 26-week study, fasting plasma glucose measurements mirrored the HbA1c improvements. Both drugs significantly reduced glucose within the first 6 weeks, but liraglutide did so to a significantly greater extent (–2.85 mmol/L vs. –1.7 mmol/L). Each reduction held steady through the end of the trial.
Both drugs effected a similar weight loss by the study’s end – about 4 kg. Both also reduced systolic blood pressure similarly, with a mean drop of about 5 mm Hg for liraglutide and 4 mm Hg for lixisenatide. Liraglutide increased the mean pulse rate from 75 beats per minute (bpm) at baseline to 80 bpm by week 6. The mean decreased to 78 by the end of the study. Pulse declined by 1 bpm from baseline among patients taking lixisenatide.
Liraglutide seemed to exert a beneficial effect on beta cells, as measured by the homeostatic model assessment (HOMA-B). At baseline, mean HOMA-B was 60%. By 26 weeks, it had increased to near 100% among those taking the drug; among those taking lixisenatide, HOMA-B had increased to about 77% – a significant difference.
Liraglutide was associated with a significant increase in lipase, Dr. Nauck said. Baseline lipase was 40 U/L. By 12 weeks, both drugs had increased lipase significantly – lixisenatide to about 50 U/L and liraglutide to about 60 U/L. These levels were constant through the study’s end. There were, however, no cases of pancreatitis or pancreatic cancer in either treatment arm, he noted.
Liraglutide use was associated with significantly more adverse events than lixisenatide (72% vs. 64%), but few of those were serious (6% vs. 3.5%). There were no cases of severe hypoglycemia and no deaths. Less than 5% of patients experienced a treatment-emergent or serious adverse event. The most common of these were nausea (22% in each group), diarrhea (12% liraglutide vs. 10% lixisenatide), vomiting (7% vs. 9%), increased lipase (8% vs. 2.5%), nasopharyngitis (6% vs. 10%), decreased appetite (6% vs. 2.5%), and dyspepsia (5.4% vs. 3%).
Although lixisenatide was outperformed in this analysis, it is still a valuable alternative for some patients, Dr. Bo Ahren said in an interview.
“Its primary action is to reduce postprandial glucose,” said Dr. Ahren, who is dean of the Faculty of Medicine at Lund (Sweden) University. “Therefore, there may be a use for this agent in patients who have controlled fasting glucose, for example, with insulin, but still have high postprandial glucose.”
Although the adverse event data were encouraging, more study is necessary, he added. The current data aren’t enough to hone any patient selection criteria.
“The lipase increases need to be examined in more detail, although it’s true there were no cases of pancreatitis. Also, the heart rate was measured only in the clinic, not as a 24-hour measurement. Therefore it’s too early to base any patient selection on this,” Dr. Ahren said.
Both Dr. Nauck and Dr. Ahren disclosed numerous financial relationships with pharmaceutical companies.