In rheumatoid arthritis (RA) treatment with TNF inhibitors changes the relationship between disease activity and progression of radiological joint damage (the so-called ‘disconnect’ that allows arrest of damage progression even at moderate disease activity levels). In early RA intensive combination therapy of methotrexate and sulfasalazine with step-down prednisolone (COBRA schedule) has been shown to be equivalent to high-dose methotrexate and infliximab in suppression of damage progression (BeSt trial).1 We investigated whether COBRA treatment also results in a ‘disconnect’ between disease activity and damage.
In a meta-analysis we combined the data of the COBRA trial2 (COBRA v sulfasalazine monotherapy) with that of two arms of the BeSt trial (COBRA v sequential monotherapy). For each trial, 1-year progression of damage (Sharp van der Heijde score) was the dependent variable in a linear regression formula with the variables disease activity, treatment strategy (COBRA or control) and their interaction as independent factors. As estimates of disease activity we used the 1-year time-averaged Disease Activity Score (DAS44) and C-reactive protein (CRP) in separate analyses. The focus was on the interaction between disease activity and treatment strategy as proof of the existence of a ‘disconnect’. Therefore we pooled the interaction terms of the two analyses, weighting by their inverse variance according to standard technique. Because of the directionality of the hypothesis and the relative insensitivity of regression for the detection of interaction, we used one-sided significance tests.
For these analyses between 60 and 100% of original trial patients had complete data. In both trials analysed separately time-averaged DAS44 and CRP, respectively, were the only (strongly) significant independent factors related to damage progression. In the models that included disease activity, neither treatment group nor the interaction between treatment and disease activity entered as significant factors. However, pooling resulted in a (weakly) significant test for interaction:
Time-averaged-DAS44 x treatment interaction: one-sided p = 0,027.
Time averaged CRP x treatment interaction: one-sided p = 0,044.
This analysis has limitations inherent in metaanalysis, and a generic lack of power to detect interactions, especially in the context of low damage progression overall in a skewed distribution. Nevertheless, it gives some support to the suggestion that changes in the relationship between disease activity and damage progression are not limited to anti-TNF treatment, but a property of early, rapid and deep suppression of joint inflammation.
1. Goekoop-Ruiterman YP, de Vries-Bouwstra JK, Allaart CF, et al. Comparison of treatment strategies in early rheumatoid arthritis: a randomized trial. Ann Intern Med 2007;146:406-15.
2. Boers M, Verhoeven AC, Markusse HM, et al. Randomised comparison of combined step-down prednisolone, methotrexate and sulphasalazine with sulphasalazine alone in early rheumatoid arthritis. Lancet 1997;350:309-18.
Disclosure: M. Boers, None; L. H. D. van Tuyl, None; M. van den Broek, None; P. J. Kostense, None; C. F. Allaart, None.