Comparative Clinical Efficacy and Safety of the Proposed Biosimilar ABP 710 With Infliximab Reference Product in Patients With Rheumatoid Arthritis

Mark C. Genovese; Juan Sanchez-Burson; MyungShin Oh; Eva Balazs; Jeffrey Neal; Andrea Everding; Tomas Hala; Rafal Wojciechowski; Gary Fanjiang; Stanley Cohen


Arthritis Res Ther. 2020;22(60) 

In This Article


Study Design

This was a randomized, double-blind, active-controlled study in adult patients with moderate to severe RA who have an inadequate response to methotrexate (MTX). Approval for this study was granted by the Human Research Ethics Committee and was conducted accordingly. The study protocol was approved by an independent ethics committee or institutional review board at each site prior to study initiation. A total of 558 patients (279 per treatment group) were randomized. Screening occurred ≤ 4 weeks before dosing. Eligible patients were randomized in a ratio of 1:1 to receive 3 mg/kg intravenous (IV) infusion of either ABP 710 or infliximab RP on day 1 (week 0), at weeks 2 and 6, and every 8 weeks thereafter until week 22. At week 22, patients initially randomized to infliximab RP were re-randomized in a 1:1 ratio to either continue receiving infliximab RP every 8 weeks (referred to as RP/RP treatment group) or to transition to receive ABP 710 every 8 weeks (referred to as RP/ABP 710 treatment group) through week 46, while patients initially randomized to ABP 710 continued receiving the same treatment every 8 weeks through week 46 (referred to as ABP 710/ABP 710 treatment group).

Study Population

Eligible patients included infliximab RP- or infliximab biosimilar-naive adult men and women aged 18–80 years with a diagnosis of RA (duration of at least 3 months). Patients were to have active RA, defined as ≥ 6 swollen joints and ≥ 6 tender joints (based on 66/68 joint count excluding distal interphalangeal joints) at screening and baseline and at least one of the following at screening: erythrocyte sedimentation rate ≥ 28 mm/h or serum C-reactive protein (CRP) > 1.0 mg/dl. Patients were to have a positive rheumatoid factor and/or anti-cyclic citrullinated peptide at screening and were to have taken MTX for ≥12 consecutive weeks and be on a stable oral or subcutaneous dose of 7.5 to 25 mg/week MTX for ≥ 8 weeks prior to dosing of investigational product (IP). Other permitted concomitant treatments included continuation of stable dose of oral corticosteroids at a dose of ≤ 10 mg prednisone (or equivalent) per day. Nonbiologic disease-modifying antirheumatic drugs (other than MTX) any biologic treatment for RA as well as other specified treatments that could impact RA were prohibited. Premedications were selected according to local practices and/or the approved product labeling for infliximab and were to be administered approximately 30 min prior to the start of the IV infusion.

Efficacy Endpoints

The primary efficacy endpoint was the response difference (RD) of achieving a 20% improvement from baseline in the American College of Rheumatology core set of measurements (ACR20) at week 22.[8,9] Secondary efficacy endpoints included the difference in means of Disease Activity Score 28-joint count-CRP (DAS28-CRP) and the RD for ACR20, ACR50, and ACR70 (20%, 50%, 70% improvement in ACR core set of measurements) at various time points throughout the study.


Key safety endpoints included adverse events (AEs), serious adverse events (SAEs), and incidence of antidrug antibodies (ADAs). AEs of interest were determined by customized queries or the Standard Medical Dictionary for Regulatory Activities (MedDRA) queries.

ADAs were assessed at baseline and weeks 2, 6, 14, 22, 30, 34, 38, and 50/end of study. Binding and neutralizing ADAs were detected with a two-tiered approach that included a screening assay and a confirmatory assay. Validated immunoassays were used to detect antibodies capable of binding ABP 710 or infliximab RP. Samples that tested positive for binding ADAs were subsequently tested in the corresponding target binding assay to determine neutralizing activity against ABP 710 or RP.

Statistical Analyses

A sample size of approximately 550 patients was chosen to achieve > 90% power to demonstrate equivalence between the ABP 710 and infliximab RP groups for the primary efficacy endpoint, RD of ACR20 at week 22, with a two-sided significance level of 0.05 and equivalence margin of (− 15%, 15%). This calculation was based on the assumption of an expected ACR20 response of 52% at week 22 for each group.

All efficacy endpoints were analyzed using the intent-to-treat (ITT) analysis set, which included all randomized patients, based on patients' randomized treatment, regardless of the actual treatment received. The per-protocol (PP) analysis set included all patients randomized in the study who completed the treatment period and did not experience a protocol deviation that affected their evaluation for the primary objective of the study and was based on actual treatment received. The analysis of safety endpoints included all randomized patients who received any amount of IP and was based on actual treatment received (safety analysis set).

Clinical equivalence for the primary endpoint was evaluated by comparing the two-sided 90% confidence interval (CI) for RD of ACR20 at week 22 between ABP 710 and infliximab RP with an equivalence margin of (− 15%, 15%). RD was estimated by the Mantel-Haenszel estimate and the CIs for RD of ACR20 were estimated by the stratified Newcombe confidence limits adjusting for two stratification factors (geographic region and prior biologic use for RA) on the ITT analysis set with non-responder imputation (NRI).

An additional sensitivity analysis was performed ad hoc to adjust for the impact of a random imbalance in baseline demographic and disease characteristics between the 2 groups.[10–12] The baseline covariates identified to be predictive of ACR20 by a stratified conditional logistic regression (with forward selection method and p value ≤ 0.25 to enter) were identified for each analysis performed in the ITT set (with NRI, last observation carried forward, and as observed) and the PP set.

For the secondary endpoints of RD of ACR20 (at scheduled visits other than week 22), RD of ACR50, and RD of ACR70, analyses were performed on the ITT analysis set with NRI using the same statistical model as used for the primary analysis of the primary endpoint. For DAS28-CRP change from baseline, the difference between means and the corresponding CIs are estimated using an ANCOVA model adjusted for the baseline DAS28-CRP and the two stratification factors on ITT analysis set with observed data.