Objectives: This exploratory analysis investigated the potential use of the multibiomarker disease activity (MBDA) score to support biosimilarity assessments using data from two randomised controlled trials (RCTs) of biosimilar infliximab (IFX-qbtx) and biosimilar adalimumab (ADL-afzb) versus EU-sourced infliximab (Remicade; IFX-EU) and adalimumab (Humira; ADL-EU) reference products, respectively, both conducted in adult patients with active rheumatoid arthritis.

Methods: In one study, patients (N=650) were randomised 1:1 to IFX-qbtx or IFX-EU (3 mg/kg intravenous at weeks 0, 2 and 6, then every 8 weeks). In the other, patients (N=597) were randomised 1:1 to ADL-afzb or ADL-EU (40 mg subcutaneous every other week). All treatments were given with MTX. Mean values of MBDA scores were calculated at baseline (BL), based on the concentrations of 12 serum proteins using the Vectra disease activity algorithm, and at timepoints throughout treatment period 1 (TP1) of the IFX (weeks 6, 14, 30) and ADL (weeks 6, 12, 26) studies. Data were summarised using descriptive statistics for the intent-to-treat population, without imputation for missing data.

Results: At BL, mean (±SD) MBDA scores were 61.3 (±12.5) and 58.8 (±13.2) for IFX-qbtx (n=236) and IFX-EU (n=248), respectively, and 57.2 (±14.44) and 58.3 (±15.34) for ADL-afzb (n=292) and ADL-EU (n=293), respectively. Mean MBDA scores were highly comparable between IFX-qbtx and IFX-EU and between ADL-afzb and ADL-EU at all measured timepoints during TP1 in each study.

Conclusions: These RCTs are the first to incorporate MBDA score as an exploratory assessment of biosimilarity. MBDA scores may provide objective, quantitative evidence of biosimilarity using an assessment of disease activity that is independent of the potential subjectivity inherent in joint counts, or in patient or physician global assessments.

Trial registration numbers: NCT02222493 and NCT02480153.