Background The order of use of biologic agents after failing a TNF inhibitor remains a question for debate. Phase III trial data in TNF-IR patients show comparable efficacy results across biologic agents, and limited head-to-head studies have been published. Prospective registries offer a unique opportunity to observe the effectiveness of these agents in a real-world clinical setting where all single diagnostic patient treated in the center are included. We report here a six-year follow-up.
Objectives Our aim is to assess if patients with rheumatoid arthritis (RA) treated with rituximab (RIT) after failing a first or a second anti-TNF agent (TNF-IR) have different six-year retention rates than patients similarly prescribed anti-TNF agents (pooled adalimumab, etanercept or infliximab) and compare the treatment strategies of using RIT as second or third biologic treatment.
Methods Data from TNF-IR RA patients prescribed adalimumab (ADA), etanercept (ETA), infliximab (INF) or rituximab (RIT) as second or third biologic agents on or after January 1st, 2007 was extracted and subjects taking either ADA, ETA or INF were pooled to form the anti-TNF cohort. Baseline demographics included age, disease duration, HAQ-DI, fatigue and pain visual analog scale evaluations (VAS), TJC, SJC, DAS-28(ESR) and CDAI. Six-year drug retention rates were estimated and compared using Kaplan-Meier survival estimates. Statistical analysis was performed using SAS version 9.4. RHUMADATA® is a software used in daily clinical practice at the IRM and CORQ. All patients with RA are included.
Results The data from 231 RA patients was extracted, 155 and 76 having respectively failed a first and a second anti-TNF agent. No clinically significant differences in baseline variables were observed between treatment groups in second and third intention. The 6-year retention rates of second-line RIT and anti-TNF users were 80.1% and 19.1% respectively (overall survival log-rank p<0.0001). After two failures, subsequent use of RIT and anti-TNF agents respectively demonstrated 6-year retention rates of 53.6% and 37.2% (overall survival log-rank p=0.0473). Second versus third-line use was numerically (80.1% vs. 53.6%) and statistically better (overall survival log-rank p=0.0029). Of the censored patients treated with a second TNFi, 41%, were in remission, 27% had low disease activity (LDA), and 23% had moderate disease activity (MDA) according to DAS28(CRP) criteria. In patients treated in second intention with RIT, 23% were in remission, and 27% and 42% were respectively in LDA and MDA. Among censored patients treated with a third TNFi, 33%, 11% and 44% were in remission, had LDA and MDA respectively compared to 39%, 6% and 39% in RIT exposed patients.
Conclusions As second and third-line agent used in TNF-IR patients, RIT demonstrates a better 6-year retention rate than anti-TNF agents. Second-line use showed a statistically better retention rate than third-line use. This suggests that using RIT as a second agent after failing a first anti-TNF agent is a better strategy than waiting to use it after two successive anti-TNF failures.
Disclosure of Interest None declared