Liraglutide Maintains Efficacy for Type 1 Diabetes, Despite Antibodies

The efficacy of liraglutide, a synthetic protein with 97% homology to human glucagon-like polypeptide-1 (GLP-1) recently approved by the US Food and Drug Administration, is not impacted in subjects with type 1 diabetes who develop anti-liraglutide antibodies, according to a pooled analysis of four phase III trials.

Overall, 102 (8.6%) subjects out of a total of 1,185 generated antibodies to liraglutide. “However, this did not change their response and it did not have any adverse effects,” Alan Garber, MD, professor of medicine, biochemistry, and cell biology at Baylor College of Medicine, Houston, Texas, said in an interview at the American Association of Clinical Endocrinologists 19th Annual Meeting and Clinical Congress.

In addition to assessing the prevalence of anti-liraglutide antibodies by radioimmunoassay, the researchers also tested all samples that were antibody positive for cross-over reactivity to native GLP-1 and in-vitro neutralizing effects on liraglutide. They also determined whether antibody generation affected efficacy based on hemoglobin A1C levels.

All subjects had end-of-treatment samples taken off drug (liraglutide 0.6 mg, 1.2 mg and 1.8 mg) for at least five days, to ensure that serum liraglutide levels did not interfere with the antibody assay.

Anti-liraglutide antibodies were detected in 8.3% of subjects treated with 1.8 mg of liraglutide, in 8.7% of subjects treated with 1.2 mg, and in 9.2% who were treated with 0.6 mg of liraglutide.

Twelve subjects who tested positive for liraglutide antibodies (11.8%) had neutralizing antibodies, and 56 (55%) subjects had antibodies that cross-reacted with native GLP-1.

The researchers reported that antibody titres were in the range of 1.6 to 10.7% bound/total, with a mean of 3.28% B/T.

Among subjects who received liraglutide 1.8 mg, those with anti-liraglutide antibodies had a mean reduction in HbA1c of -1.1%, whereas those without anti-liraglutide antibodies had a mean HbA1c reduction of -1.2%. Similar results were seen for subjects who received 1.2 mg of liraglutide. Antibody- positive subjects had a mean HbA1c reduction of -1.3%, and antibody-negative subjects had a mean HbA1c reduction of -1.2%. There was also no difference in HbA1c reduction in the liraglutide 0.6 arm in antibody-positive or antibody-negative subjects.

Nine subjects who were antibody-positive had injection site reactions, but none withdrew from the trials because of these adverse events, Garber said.

“A small number of people would develop some antibodies even if the peptide was totally identical,” he commented. “But at least, the presence of the antibodies did not change the response, and did not have severe adverse effects. The reason that this is important is because there is another GLP-1 protein that is 53% identical to human insulin and it seems to generate more antibodies, but the significance of this higher antibody level is currently not clear. What we can say now with liraglutide is that its immunogenic potential is low, and that the prevalence of low titres does not affect efficacy.”