Background At least one third of patients with long-standing RA starting treatment with a biologic (b)DMARD will have to change the substance within three years due to inefficacy or adverse events. The question arises, which treatment responses we can expect after one or more failures of biologic therapies and whether there are differences in the effectiveness between individual substances used as second- or third-line bDMARDs.
Objectives We aimed to investigate the effectiveness of a first, second or at least third bDMARD therapy in RA patients observed in the German biologics register RABBIT.
Methods Patients who started a bDMARD therapy and were enrolled in RABBIT between 2007 and 2012 were eligible for the present analysis. They were stratified in bDMARD treatment states according to the nth bDMARD they started at enrolment (1st bDMARD: n=1,989, 2nd bDMARD: n=1,017, 3rd bDMARD: n=1,113). The 12-month course of disease activity captured by the composite disease activity score in 28 joints (DAS28CrP) was examined by linear mixed effects models. Adjustment was made for the baseline parameters sex, rheumatoid factor, DAS28CrP, the number of prior non-biologic (nb)DMARDs and the number of comorbidities, as well as for the observation time.
Results Patients who received a second or at least third bDMARD were more often female and rheumatoid factor positive, had a longer disease duration, more co-morbid conditions, lower functional capacity and more pain. In the stratum of the first bDMARD, 81% of the patients received TNF inhibitors, 8% rituximab and 9% tocilizumab. The most frequent second bDMARD was rituximab (39%), followed by a second TNF inhibitor (34%) and tocilizumab (18%). Abatacept (17%), rituximab (57%) and tocilizumab (17%) were mostly used substances in the 3rd bDMARD treatment status.
Patients of all bDMARD treatment states, including patients enrolled with at least a third bDMARD, showed significant improvement of disease activity after 12 months of treatment.
After adjustment for baseline differences, patients treated with the first, second, or at least third bDMARD improved considerably from the adjusted mean baseline DAS28CrP of 5.2 to a mean DAS28CrP (95% CI) at month 12 of 3.4 (3.3–3.6), 3.6 (3.5–3.7), and 3.8 (3.6–3.9), respectively. The difference between the 1st and 3rd bDMARD treatment status could mainly be explained by differences in clinical baseline values and treatment history of the patients. Within the first two strata no significant differences in the outcome between the individual bDMARDs were observed. Particularly in the 3rd bDMARD treatment status, we found an advantage for tocilizumab, most likely due to the direct suppression of CrP (and ESR) responses via IL-6 blockade.
Conclusions Patients with previous failures of bDMARDs had poorer baseline clinical values and lower responses than those on a first bDMARD. When using real-life data to compare different substances, it has to be taken into account that there are strong preferences which of the substances are used first. Therefore, stratification or adjustment for respective bDMARD treatment states may be appropriate for head-to-head comparisons of bDMARDs in the unselected spectrum of patients with long-standing RA.
Disclosure of Interest K. Gerhold Grant/research support: RABBIT is supported by a joint unconditional grant from AbbVie, Bristol-Myers Squibb, MSD Sharp & Dohme, Pfizer, Roche and UCB, A. Richter: None declared, A. Strangfeld: None declared, P. Herzer: None declared, M. Bohl-Bühler: None declared, S. Berger: None declared, J. Listing: None declared, A. Zink: None declared