FRI0175 Indirect Treatment Comparisons and Network Meta-Analysis among Biological Agents for Rheumatoid Arthritis: A Systematic Review of Published Literature

Abstract

Background Rheumatoid arthritis (RA) is associated with significant morbidity, functional impairment, loss of quality of life, premature mortality, and substantial economic burden.1 Biological disease-modifying antirheumatic drugs (bDMARD) are key in the RA treatment pathway. Indirect treatment comparisons (ITC) and network meta-analysis (NMA) allow inferences in the absence of head-to-head comparisons.2 Objectives To perform a systematic review of publications of ITC and NMA among bDMARD in RA. Methods A search in PubMED/Medline and Google Scholar was conducted. Inclusion criteria included full-text access and publication in English or Spanish during a 2009-2013 timeframe. Studies that pooled TNF-alpha inhibitors and those focused on economic evaluations were excluded. Competing alternatives (either as monotherapy or in combination) were infliximab, etanercept (ETN), adalimumab, certolizumab pegol (CZP), golimumab, abatacept, tocilizumab (TOC), rituximab, and anakinra. Relevant characteristics regarding target population and study design (eg. prior drugs failed and if bDMARD were given as monotherapy or in combination besides its duration) of the selected clinical trials were registered. Trial quality assessment (eg. Jadad scale) and type of statistical approach (frequentist or Bayesian) used when performing ITC or NMA were also recorded. American College of Rheumatology (ACR) 20, 50, and 70 response criteria and withdrawal rates analysed throughout ITC or NMA comprised the outcomes of interest for the present review. Results Sixty-five references were retrieved, of which 10 achieved inclusion criteria for ITC and NMA.3-12 The number of clinical trials analyzed in each publication varied from 105 to 44.10 Among the 3 studies that conducted frequentist analyses, 3-5 CZP ranked highest for efficacy in 4 out of 7 efficacy ranking reports, followed by ETN (2) and TOC (1). CZP received the highest efficacy rank in 10 out of 13 ranking reports among the 7 studies with Bayesian approaches, 6-12 while ETN ranked highest in the other 3. ETN was associated with the lowest rates of withdrawal due to adverse events (AEs),3,10 whereas CZP showed the lowest rates of withdrawal due to lack of efficacy.10 Overall, CZP had the lowest rates of all-causes of withdrawal.10 Conclusions Current review indicated that published ITC and NMA show that CZP has the highest treatment effect compared to other bDMARD used in RA. In terms of withdrawal rates, ETN had the lowest rates due to AEs, and CZP was associated with the lowest all-cause withdrawal rates.