P484. Functional cellular based assay reveals neutralising anti-drug antibodies in IBD patients treated with maintenance adalimumab
N. Vande Casteele1, M. Peeters1, M. Ferrante2, G. Compernolle1, G. Van Assche2, S. Vermeire2, A. Gils1, 1KU Leuven, Department of Pharmaceutical and Pharmacological Sciences, Therapeutic and Diagnostic Antibodies, Leuven, Belgium, 2KU Leuven, Department of Clinical and Experimental Medicine, Translational Research Center for GastroIntestinal Disorders (TARGID), Leuven, Belgium
Anti-adalimumab antibodies (AAA) have been described in 2.8–9.2% of adalimumab treated inflammatory bowel disease (IBD) patients. It is not known whether these AAA neutralise the binding of the drug to tumour necrosis factor alpha (TNF). Our aim was to investigate the prevalence and clinical impact of neutralising AAA in a large cohort of IBD patients.
Patients with undetectable adalimumab and/or detectable AAA at any given time point were selected from a large cohort of 397 IBD patients (318 CD, 72 UC and 7 IBDU) receiving maintenance adalimumab therapy. Serial adalimumab and AAA levels (measured with a previously described ELISA and bridging ELISA respectively) were available from routine follow-up at our IBD unit. Twenty patients (17 CD and 3 UC) were identified of which 89 serum samples were retrospectively tested in a functional TNF-responsive cell based assay (CBA) to determine neutralising capacity of the AAA. MA-ADM6A10, a monoclonal antibody with a high affinity and specificity for adalimumab was used as standard in CBA, hence AAA values were expressed towards the response of this MA. Neutralising AAA levels <0.132 µg/ml were regarded as undetectable.
Testing with CBA revealed neutralising AAA in 48/89 samples (53.9%) thereby classifying all 20 patients as positive for neutralising AAA on at least one time point during follow up. CBA detected AAA in 15 samples that were negative in bridging ELISA, thereby reclassifying 3 patients as AAA positive. Median (inter quartile range) levels of AAA in ELISA were 0.20 µg/ml equivalents (0.20–0.39) and 0.21 µg/ml (0.13–1.4) in CBA. Of the patients developing neutralising AAA 15/20 patients (75%) discontinued adalimumab treatment or were still treated with adalimumab but were not in remission at the end of follow-up.
Functional testing with a CBA revealed neutralising AAA in 5% of IBD patients treated with maintenance adalimumab. CBA was more sensitive than ELISA but both assays cannot detect AAA in presence of drug and therefore the amount of AAA positive patients is underestimated. However, this study demonstrates that the development of detectable neutralising AAA - although not frequent - is detrimental for the patient's clinical outcome.