Background Switching from the first Tumor Necrosis Factor α inhibitor (TNFi) to a second in rheumatoid arthritis is an effective strategy to decrease disease activity. Especially, when antidrug antibodies (ADA) against the first TNFi were detected [1,2]. In addition, patients with detectable ADA against the first TNFi had an increased risk to develop ADA against a second TNFi. In ankylosing spondylitis (AS) and psoriatic arthritis (PsA), switching from the first TNFi to the second is effective, although switchers have a poorer clinical outcome compared with patients who continued their first TNFi [3,4].
Objectives To investigate clinical response in AS and PsA patients who switched to a second TNFi versus biological naïve starters. Secondly, to study the association between reason for switching and efficacy of the second TNFi. Finally, investigate predictive value of drug level and ADA of the first TNFi and efficacy to the second.
Methods In this retrospective study 2 years data were collected from observational cohorts of patients with AS and PsA, in whom TNFi was started. Disease activity and response in AS was assessed by the use of AS Disease Activity Score (ASDAS) and Patient Global Disease Activity (PGDA). For PsA, the Disease activity Scale using 28 joint count (DAS28) and PGDA were used for this purpose. Switchers response to 2nd TNFi use was compared with response during 1st TNFi use of the control group (further mentioned as: switcher vs control). Drug- and ADA levels were measured by an Enzyme-linked immunosorbent assay and an antibody binding test, respectively.
Results A total of consecutive 486 patients were included of which 65 (AS, n=35; PsA n=30) were switchers. Generalized estimating equation (GEE) demonstrated a significant difference between ASDAS response, switchers vs control; Odds ratio (OR): 0.32 (95% CI 0.14-0.75) p=0.008; and a significant difference between PGDA, switchers vs control, regression coefficient (RC): 1.09 (95% CI 0.30-1.89) p=0.007. In PsA, a significant difference between DAS28 response, switchers vs control; OR 0.28 (95% CI 0.12-0.67) p=0.004; and a significant difference between PGDA of switchers vs control; RC: 15.0 (95% CI 8.8-21.2) p<0.001. No significant difference could be demonstrated in reason for switch, adverse events or failure. A trend, for a better and faster clinical response, was found in favor of switchers with adalimumab ADA, both AS and PsA, compared with switchers with detectable adalimumab drug levels and no ADA.
Conclusions Switchers had, during their second TNFi use, a poorer clinical response compared with biological naïve patients in both AS and PsA. There was a tendency towards patients with adalimumab ADA before switch to have a quicker and better response on disease activity compared to the patients with detectable adalimumab drug levels. However, patient numbers were small.
Ann Rheum Dis 2011 Feb;70(2):284-8.
Curr Opin Rheumatol 2012 May;24(3):306-11.
Arthritis Rheum 2013 May;65(5):1213-23.
Ann Rheum Dis 2013 Jul;72(7):1149-55.
Disclosure of Interest J. Ruwaard: None declared, E. Kneepkens Speakers bureau: Pfizer, A. Marsman: None declared, E. Vogelzang: None declared, I. van der Horst-Bruinsma: None declared, T. Rispens Speakers bureau: Pfizer, Abbvie, M. Nurmohamed Grant/research support from: Pfizer, Abbvie, Roche and BMS, Speakers bureau: Janssen, Abbvie and BMS, G. Wolbink Grant/research support from: Pfizer, Speakers bureau: Pfizer, Amgen, Abbvie, UCB and BMS