Background: After adalimumab treatment failure, TNFi and non-TNFi bDMARDs are equally viable as subsequent treatment in RA. However, preliminary data suggest that anti-drug antibodies (ADA) and adalimumab serum levels (ADL) predict response to a subsequent TNFi [1].

Objectives: To assess the association of presence of ADA and/or low ADL with response to a subsequent TNFi bDMARD or non-TNFi bDMARD.

Methods: A retrospective cohort study to assess the predictive value of ADA and ADL for response to a subsequent TNFi or non-TNFi bDMARD in RA patients.

All RA patients who received adalimumab (standard dose, ≥ 3 months) and subsequently switched to another TNFi or a non-TNFi (rituximab, tocilizumab, abatacept) in the Sint Maartenskliniek or Radboud University Medical Centre between January 2012 and January 2018 were considered for inclusion in the current study. Further inclusion criteria were the availability of (random timed) serum samples between ≥8 weeks after start, and ≤2 weeks (for ADL) or ≤12 weeks (for ADA) after discontinuation of adalimumab, and clinical outcome measurements (DAS28-CRP/BSE) between 3-6 months after treatment switch. Serum samples were derived from a period of biobanking at every visit of RA patients and an observational cohort study including consecutive bDMARD starters.

The primary outcome of this study was the association between ADL or ADA and EULAR good response (DAS28-CRP/ESR based) to the subsequent bDMARD. When DAS28-based response was unreliable due to glucocorticoid use, or low baseline DAS28 (if switching due to adverse effects), judgement of the rheumatologist was used.

A drug-tolerant competitive enzyme-linked immunosorbent assay (Sanquin, the Netherlands) was used to quantify ADA, and thereafter, ADL was determined via an ELISA. Reference values were ≥5 µg/ml for ADL and <12 AU/ml for ADA [2,3]. Treatment was blinded for ADL and ADA levels.

Prediction of response were assessed using the area under the receiver-operator characteristic (AUROC) and sensitivity/specificity. Sub-analyses were performed for primary and secondary non-responders. Correlations between ADL and ADA presence and clinical variables were also cross-sectionally explored.

Results: 137 patients were included, 47 of whom switched to a second TNFi and 90 to a non-TNFi. Sensitivity and specificity of the proposed ADA and ADL reference values were low ( table 1 ). The AUROC did not differ appreciably or significantly from 0.5. Results were similar for both primary and secondary non-responders to adalimumab.

Higher ADL (Spearman’s ρ = -0.68, p = 0.00) but not ADA (ρ = 0.23, p = 0.28) presence was associated with a lower DAS28 at the time of switching to a subsequent bDMARD, but not with follow-up DAS28 after starting the subsequent bDMARD (ρ = -0.29, p = 0.17, and ρ = 0.10, p = 0.65, respectively). In addition, higher ADL were associated with lower baseline CRP (ρ = – 0.67, p = 0.00) and ESR (ρ = – 0.546, p = 0.006) and higher ADA correlated with higher baseline ESR (ρ = 0.49, p = 0.01).

Conclusion: No predictive value for response to a second TNFi or non-TNFi was found for either ADA or random timed ADL. Limitations of this study are the retrospective design and random timed serum sampling. An ongoing randomized blinded test-treatment trial will provide more definitive answers [4].

REFERENCES:

[1]Van Herwaarden et al. Expert Opin Drug Metab Toxicol. 2017;13:843–57.

[2]Pouw et al. Ann Rheum Dis. 2015 Mar;74(3):513-8.

[3]Bartelds et al. Ann Rheum Dis. 2007 Jul;66(7):921-6.

[4]ADDORA-SWITCH study, www.trialregister.nl, no NL 8210

Disclosure of Interests: Nathan den Broeder: None declared, Evy Ulijn: None declared, Maike Wientjes: None declared, Noortje van Herwaarden: None declared, Inger Meek: None declared, Lieke Tweehuysen: None declared, Aatke van der Maas: None declared, Bart van den Bemt Grant/research support from: UCB, Pfizer and Abbvie, Consultant of: Delivered consultancy work for UCB, Novartis and Pfizer, Speakers bureau: Pfizer, AbbVie, UCB, Biogen and Sandoz., Alfons den Broeder: None declared