Background Biological therapy has shown its efficacy in different chronic inflammatory arthropathies during the last decade. However, in practice, clinical efficacy is reduced in some patients, suggesting that drug-induced immunogenicity may be a possible mechanism

Objectives To analyze the development of tumor necrosis factor (TNF) antagonist antibodies (AB) and determine serum drug levels in patients in whom loss of efficacy or adverse treatment effects occur over time.

Methods Descriptive, retrospective study of patients attended by a rheumatology department of a tertiary hospital between February and December 2012 who were on active treatment with TNF antagonists and in whom loss of efficacy, inefficacy or adverse treatment effects occurred. Patients were included according to clinical judgement and not according to an established protocol. We measured antidrug AB and drug serum levels using an ELISA immunoassay (Promonitor®). Only patients on infliximab (IFX), etanercept (ETN) or adalimumab (ADA) treatment were analyzed. All blood samples were obtained in the 24 h before the next scheduled dose of treatment. Only patients with positive AB were finally analysed. The following variables were collected: demographic variables; diagnosis and disease duration; previous and current treatment; reason for the analysis; antidrug levels and serum drug concentrations; and clinical decision.

Results Seventy patients were included; 67% female; mean age 51±14 years. Main diagnoses were: rheumatoid arthritis (RA) (52.8%); ankylosing spondylitis (AS) (18.6%); psoriatic arthritis (PsA) (10%); and miscellaneous (18.6%) which included, among others, Behçet disease (BD), undifferentiated spondyloarthropathy and juvenile idiopathic arthritis (JIA). The reason for the analysis was: loss of efficacy in 78.6% of patients, adverse events (11.4%), partial response (7.2%), and inefficacy (2.8%). Twelve patients (17.1%) developed antidrug AB: 7 anti-ADA and 5 anti-IFX, representing 24% and 31%, respectively, of patients on these treatments included in the study. Antidrug AB were found directly in 9/12 patients and by a dissociation method in the remaining 3 patients. In 10 patients, serum drug levels were undetectable, and in 2 cases (1 ADA and 1 IFX), both found by the dissociation method, suboptimal (suboptimal level ADA: 0.04-0.80 ng/ml and IFX: 0.05-1.50 ng/ml). Diagnoses were: 6 RA; 3 AS; 1 PsA; 1 JIA and 1 BD. Seven out of 12 patients received concomitant synthetic DMARDs (4 leflunomide and 3 methotrexate), and the remaining five received biological therapy in monotherapy. Five out of 12 have been treated previously with another biological agent (2 anti-TNF and 3 non anti-TNF), which had been discontinued due to inefficacy or adverse events. After the development of antidrug AB, 10 patients discontinued current biological therapy: 7 patients were switched to another anti-TNF agent, with a good response, and 2 to a non anti-TNF drug.

Conclusions In our study, around 20% of patients with a loss of efficacy or adverse events to a TNF antagonist developed antidrug AB, which were only observed in patients receiving monoclonal antibodies.

Disclosure of Interest None Declared