Search in the Abstract Database

* = Presenting author

DOP028 Antibodies towards vedolizumab appear from week 2 onwards and disappear upon treatment

Bian S.*1, Tang H.T.1, Peeters M.1, Compernolle G.1, Dreesen E.1, Van Assche G.2, Ferrante M.2, Vermeire S.2, Gils A.1

1KU Leuven, Department of Pharmaceutical and Pharmacological Sciences, Leuven, Belgium 2University Hospitals Leuven, Department of Gastroenterology and Hepatology, Leuven, Belgium

Background

Vedolizumab (VDZ), a monoclonal antibody (MA) that specifically binds to α4β7 integrin, has demonstrated efficacy in patients with moderate-to-severe Crohn's disease (CD) and ulcerative colitis (UC). VDZ trough concentrations (TC) are typically >30 μg/ml during induction and >10 μg/ml during maintenance therapy, making it challenging for immunogenicity assays to detect anti-VDZ antibodies (AVA). The GEMINI trials reported AVA at ≥2 consecutive time points in 0.4% and 1% of patients with CD and UC, respectively.

Methods

We developed a drug-resistant AVA assay in which AVA are complexed with an excess of VDZ followed by VDZ/AVA complex precipitation, acidification of the precipitate, coating and detection of released AVA using biotinylated VDZ. MA-VDZ19C11, a MA towards VDZ, was validated as calibrator. Drug resistance was examined by determination of the recovery of MA-VDZ19C11 in the presence of 3 different concentrations of VDZ. The cut-off was determined using 20 VDZ naïve patients. Cross reactivity with serum from 2 anti-infliximab antibody and 2 anti-adalimumab antibody positive patients, as well as with serum containing high concentration of rheumatoid factor (150 U/mL) was determined. The assay was subsequently applied to serum samples from 75 VDZ-treated patients (46 CD, 29 UC) taken at trough during induction (w6) and maintenance (w22). VDZ TC, AVA and CRP were determined in all available sera of patients positive for AVA at w6/22.

Results

MA-VDZ19C11 yielded dose–response curve ranging from 25–1600 ng/mL in 1/125 diluted serum allowing detection of AVA concentrations up to 200 μg/mL equivalents. Spiking 1, 10 and 100 μg/mL of VDZ to 40 μg/mL MA-VDZ19C11 yielded recoveries of 89±7% (mean ± SD, n=3), 90±11% and 82±5%, respectively, confirming the complete drug resistance of the AVA assay. Cut-off for quantification was determined to be 1.1 μg/mL MA-VDZ19C11 equivalents and none of the sera tested revealed cross-reactivity. Among the 75 VDZ-treated patients, 1 patient (1.3%) had AVA antibodies at w6 determined by a classical drug-sensitive bridging assay whereas 4 patients (5.3%) were AVA positive on ≥2 time points using the drug-resistant AVA assay. AVA antibodies appeared from w2 onwards but disappeared over time (Figure 1). None of the 4 ADA-positive patients required VDZ intensification.

Conclusion

Using a drug-resistant AVA assay, AVA are detected in 5.3% of patients. Antibodies appear from w2 onwards and disappear upon time indicating their transient character.