Maxime Verhoeven

31 Systematic review of remission-induction strategies in early RA pooled RR of achieving remission for strategies within the window of opportunity was 1.43 [95%CI 1.15 - 1.77] versus 1.44 [95%CI 1.12 - 1.86] for studies outside the window of opportunity. Five studies used a single csDMARD-initiating strategy with GC bridging (i.e., 2 studies within and 3 studies outside), Supplementary Figure S2. Longer term effectiveness of remission-induction strategies started in early RA We found six studies evaluating the effect of a remission-induction strategy versus a single csDMARD-initiating strategy on the long term (4 to 10 years). In 4 studies, DAS remission was more often achieved in the initial remission-induction strategy compared to the single csDMARD-initiating strategy over time. 33–35,37 In the remission-induction strategy arm, Boolean remission, as well as SDAI remission, was less often achieved in 1 of 2 studies with no difference in the other, compared to the single csDMARD-initiating strategy arm. 34,36 No difference was found for CDAI remission, which was reported in only one study. 34 One study reported data about SR, which was achieved in almost all patients over time, without differences between the different strategy arms. 38 However, using (s)DFR as outcome, differences were shown in favor of the remission-induction strategies. 37,38 No differences were found for radiographic progression over time between the different strategies. 33–35,37 Details of these studies can be found in Table 1. Risk of bias assessment The risk of bias of the included studies was overall low. In general, 26/29 studies were RCTs, the remaining 3 were cohort studies. An overview of the risk of bias assessment is shown in Supplementary Table 1. The studies evaluating long-term effects of a remission-induction strategy compared to a single csDMARD-initiating strategy, after the initial RCT, 33–38 treatment was according to the treating physician and standard care, without detailed information on the initial trial and attrition, prohibiting to fully assess all items of the risk of bias assessment. Further, moderate/high risk of bias was present in the seven studies evaluating short-term effects. 23,25–27,30–32 DISCUSSION The current meta-analysis shows that a remission-induction strategy is more effective compared to a single csDMARD-initiating strategy, possible specifically for a bDMARD based remission-induction strategy. However, this superior effect over single csDMARD- initiating strategy is limited and can no longer be detected statistically significantly when patients are treated initially also with GCs, short-term as ‘bridging therapy’. Longer term follow-up studies showed conflicting results, but more a favorable outcome with regard to (s)DFR for the remission-induction strategy may be present. 2