Frequency and Magnitude of Antibody Formation Are Lower with Liraglutide Than Exenatide: LEAD-6 Results
Abstract Number: 676-P
Authors: JOHN BUSE, EDUARD MONTANYA, GIORGIO SESTI, JASON BRETT, HELENE SOLBERG, JULIO ROSENSTOCK, MICHAEL NAUCK, JENS HOLST
Institutions: Chapel Hill, NC, Barcelona, Spain, Catanzaro, Italy, Princeton, NJ, Copenhagen, Denmark, Dallas, TX, Bad Lauterberg im Harz, Germany
Results: Antibody formation to GLP-1 receptor agonists could theoretically compromise efficacy via neutralization of the drug or even native GLP-1. Liraglutide (LIRA) and exenatide (EXE) antibodies were measured during the LEAD-6 trial, a 26-wk comparison of QD LIRA 1.8mg and BID EXE 10µg both + met &/or SU with a 52-wk extension where EXE-treated patients (pts) switched to LIRA.
LIRA/EXE antibodies and cross reactivity to native GLP-1 were detected using a radioimmunoprecipitation assay and neutralizing effect was tested using an in vitro cell-based assay at wks 0, 12, 26, 40, 41, 78 & 79. In contrast to EXE, the longer half-life and higher plasma concentration of LIRA decreased sensitivity of the assay to detect low LIRA antibody levels in pts on drug. To eliminate assay interference by plasma LIRA, antibody values from ≥5 days after last dose are presented.
After 78 wks on LIRA, 4/154 (2.6%) pts had low-titer LIRA antibodies (range: 1.9-5.3% Bound/Total [B/T]); all 4 samples cross-reacted with GLP-1 and 1 had in vitro neutralizing effect. A1C changes were 0.4, 0.8, -1.4 & -1.9% over 78 wks.
After 26 wks on EXE, 113/185 (61.1%) pts had antibodies to EXE (range: 2.4-60.2%B/T); 5 samples cross-reacted with GLP-1 and 12 had in vitro neutralizing effect. EXE pts with high antibody titers (>20%B/T) had smaller A1C reductions (-0.5%, N=47) than those with low titers (-1.0%, N=132). Patients with the highest EXE antibody titers (>50%B/T, N=6) only had a -0.1% A1C reduction.
After switching from EXE to LIRA at wk 26, 90/181 (49.7%) and 25/143 (17.5%) pts had EXE antibodies at wk 40 and 78, respectively. EXE antibody positive pts had slightly greater A1C improvements at wk 40 and 78 than those who were antibody negative, indicating no adverse effect of EXE antibodies on LIRA efficacy. Only 4/134 (3.0%) pts who switched from EXE to LIRA for 1 yr had LIRA antibodies.
In conclusion, LIRA resulted in a lower incidence and level of antibody formation than EXE even when switching from EXE to LIRA. High-titer EXE antibodies affected glycemic response to EXE, but EXE antibodies that persisted up to 1 yr after switching to LIRA did not decrease efficacy of LIRA treatment.