OA14b Etanercept originator versus its biosimilar (SB4) for the treatment of rheumatoid arthritis. Are they truly similar?

Abstract

Background/Aims Etanercept biosimilar SB4 (ETN-B) showed comparable efficacy to its biologic originator (ETN-O) among biologic-naive patients with rheumatoid arthritis (RA) in randomised controlled trials. Subsequently, a nationwide guideline in the United Kingdom (UK) obligated the prescription of biosimilars in 2016, resulting in significant cost-savings to the NHS. However, real-world evidence comparing ETN-O and ETN-B is limited. This analysis aims to compare the effectiveness of ETN-O versus ETN-B amongst biologic-naive patients with RA treated in routine clinical practice in the UK. Methods The British Society for Rheumatology Biologics Register for RA (BSRBR-RA) is a prospective observational cohort study. This analysis included patients starting their first biologic (ETN-O or ETN-B), recruited since 2010, with at least one follow-up. Baseline data are collected at start of therapy, including patient demographics and disease activity. Follow-up data are collected six monthly for the first three years and annually thereafter, including changes in disease activity and anti-rheumatic therapy. Key outcomes assessed were change in Disease Activity Score for 28-joints (DAS28), DAS28 remission, EULAR treatment response, change in Health Assessment Questionnaire (HAQ), and minimal clinically important difference (MCID) in HAQ. Drug retention was explored using Kaplan-Meier and Cox-regression analyses, along with reasons for treatment withdrawal. Multiple imputation was used to account for missing data and confounding by indication was accounted for using propensity-decile adjustment. Results This analysis included 1927 patients with RA starting their first biologic therapy: 996 (52%) ETN-O and 931 (48%) ETN-B. At six-months, change in DAS28 and HAQ were similar between therapies (Table). There were no differences in key imputed propensity adjusted outcomes (odds ratio [95% confidence interval]): DAS28 remission: 0.90 (0.7-1.1, p=0.3); MCID in HAQ: 0.86 (0.7-1.1, p=0.3) and EULAR treatment response: 0.96 (0.8-1.2, p=0.8). One-year retention rates were 69% (95% CI: 66-72) for ETN-O and 70% (95% CI: 67-73) for ETN-B and the main reasons for treatment withdrawal were adverse events and ineffectiveness for both drugs. Conclusion This analysis is the first to our knowledge to show similar outcomes amongst biologic-naive patients with RA treated with either ETN-O or ETN-B using real-world data, which brings further evidence for similarity between the drugs. (Table Presented).