Systematic review, and meta-analysis of steroid-sparing effect, of biologic agents in randomized, placebo-controlled phase 3 trials for systemic lupus erythematosus

Abstract Objectives To systematically review, and conduct a meta-analysis of steroid-sparing effect in, phase 3 randomized, placebo-controlled trials of biologic therapies for systemic lupus erythematosus (SLE). Methods Studies were identified by searching Medline (via Pubmed), EMBASE, CINAHL and SCOPUS databases, the Cochrane library, and clinicaltrials.gov. Adult human studies published in English in the last ten years (until 18/04/2017) were included. A random-effects meta-analysis comparing a common corticosteroid-reduction endpoint in the trials of rituximab, belimumab, tabalumab and epratuzumab in SLE, was conducted. Results Twenty-eight studies were identified; nine were conducted in SLE, five in lupus nephritis and the remaining 14 were post hoc analyses of phase 3 trials in SLE. All therapies trialed targeted B-cells (rituximab (anti-CD20 monoclonal antibody (mAb)), belimumab (anti-BAFF mAb), tabalumab (anti-BAFF mAb), epratuzumab (anti-CD22 mAb), atacicept (TACI-Ig), ocrelizumab (anti-CD20 mAb)), except for abetimus sodium and abatacept (CTLA4-Ig). Only the three trials of belimumab met their primary endpoints, although benefit in secondary endpoints and reduction in serological activity was often seen in the other studies. Meta-analysis showed that most therapies (belimumab, tabalumab, and epratuzumab) had a steroid-sparing effect, compared to placebo (pooled RR 1.36 (1.19, 1.56), I2 = 0, p Conclusions With the exception of belimumab, none of the phase 3 trials of biologic therapy in SLE have met their primary endpoint. However, the significant steroid-sparing effect of these agents suggests that future trials may need to include steroid dose in a composite primary endpoint.