AB0526 Sustained efficacy responses and a consistent safety profile with rituximab repeat treatment over 5 years in patients with rheumatoid arthritis and an inadequate response to tumour necrosis factor inhibitors in the reflex study

Background In the REFLEX study conducted in anti-TNF inadequate responder (TNF-IR) patients (pts) with RA, a single course of rituximab (RTX) in combination with methotrexate (MTX) significantly improved disease activity at 24 wks vs placebo (PBO) + MTX. Pts were eligible for continued RTX treatment in an open-label extension (OLE). Efficacy and safety outcomes from REFLEX and its OLE over 5 yrs are presented. Methods This was an observational, post-hoc analysis of REFLEX from baseline to 5 yrs, open label from the second study treatment. Pts originally randomized to PBO were rescued with RTX as appropriate and included in the OLE. Pts with a response to initial RTX treatment were eligible for further RTX treatment courses. RTX retreatment was administered as needed based on SJC and TJC ≥8 and at the discretion of the physician (≥24 wks following first RTX course and ≥16 wks following additional courses). PBO pts were re-baselined prior to their first RTX treatment and for this analysis PBO data were pooled with RTX pt data from time of first RTX treatment. Efficacy outcomes 24 wks after each RTX course were calculated relative to first RTX pre-treatment baseline. No imputations were made for missing data. Safety data included rates of AEs, serious AEs (SAEs), infections and serious infections (SIEs). Results Overall, 480 pts received at least one course of RTX. Subsequent RTX courses were given to 317 (≥2 courses), 259 (≥3 courses), 195 (≥4 courses), and 122 (≥5 courses) pts. Most withdrawals occurred after course 1, mainly for non-safety reasons. ACR responses were improved after the 1st course of RTX and were maintained over 5 courses (table) with a similar trend observed for EULAR responses. Table 1. ACR response rates (% pts) at 24 wks after each course Clinical measureCourse 1Course 2Course 3Course 4Course 5 (n=400)(n=279)(n=225)(n=161)(n=91) ACR2062.072.872.465.870.3 ACR5030.841.247.644.741.8 ACR7013.019.426.224.222.0 The proportion of pts achieving a minimal clinically important difference in HAQ-DI was maintained over 5 courses (66.0–71.1%). Over the 5-yrs, rates of AEs, SAEs and infections did not increase and generally remained stable in the RTX treated population (1768 pt-yrs), with overall rates per 100 pt-yrs (95% CI) of 344.87 (336.32, 353.64)for AEs, 22.34 (20.24, 24.65) for SAEs, 97.50 (93.01, 102.21)for all infections and 5.60 (4.60, 6.82)for SIEs. The most frequent SIE was pneumonia, affecting 2% of RTX pts. Conclusions In this post-hoc