Background Disease activity guided tapering of tumor necrosis factor inhibitors (TNFi) has been shown to be feasible, safe and cost-effective in rheumatoid arthritis (RA) patients with stable low disease activity. However, data on biologics with another mode of action, like abatacept (ABA) and tocilizumab (TCZ), is scarce. Moreover, most studies are clinical trials, leaving uncertainty on the feasibility of tapering in daily clinical practice.
Objectives To assess the difference in mean change (Δ) between baseline and 12 months follow-up in DAS28 and HAQ-DI between eligible RA patients that attempted to taper ABA or TCZ compared to eligible patients that did not attempt to taper.
Methods In this retrospective, explorative cohort study, data on clinical outcome measures were collected from all patients eligible to taper ABA or TCZ at the outpatient clinic of a specialized rheumatology hospital. Patients were eligible when fulfilling the 1987 and/or 2010 ACR RA criteria and/or clinical diagnosis by the treating rheumatologist, and when reaching low disease activity (DAS28 <3.2) after ≥6 months of treatment with ABA or TCZ. Patients that attempted to taper were included in the tapering group and patients that did not attempt to taper were included in the control group. Tapering was done according to the local protocol, in which the dose was stepwise reduced or the injection interval was prolonged until discontinuation. Successful tapering was defined as using ABA or TCZ on lower than baseline dose and having low disease activity at 12 months. Outcome measures were mean ΔDAS28 and mean ΔHAQ-DI (positive figures mean increase) and differences between groups were tested using an (unpaired) t-test.
Results From January 2007 until June 2015, 320 patients were treated with ABA and/or TCZ of whom 58 patients could be included (figure 1). Tapering was successful in 37% of ABA patients and 35% of TCZ patients. Mean ΔDAS28 at month 12 was 0.92 (95% CI -0.46 to 2.31) in the ABA tapering group versus 1.32 (95% CI 0.00 to 2.65) in the control group (p=0.05)and 0.31 (95% CI -0.24 to 0.85) in the TCZ tapering group versus 0.28 (95% CI -0.59 to 1.16) in the control group (non-significant (NS)). Mean ΔHAQ-DI at month 12 was 0.16 (95% CI -0.48 to 0.89) in the ABA tapering group versus -0.41 (95%CI -0.51 to 0.69) in the control group (NS) and 0.02 (95% CI= -0.40 to 0.44) in the TCZ tapering group versus 0.24 (95% CI= -0.31 to 0.79) in the control group (NS).
Conclusions Tapering of ABA or TCZ appears to be feasible in RA patients in daily clinical practice, with no significant difference in change of disease activity or functioning between patients that tapered and patients that did not taper, although confidence intervals are relatively large in this small study. However, the percentage of successful tapering is lower compared to studies on tapering of TNFi. Also, numbers of patient that start tapering are suboptimal, possibly explained by the perception of a more refractory patient population in which treatment is continued longer, since ABA and TCZ are reserved for patients that fail on TNFi in our clinic.
Disclosure of Interest None declared