Background The efficacy of etanercept (ETN) plus methotrexate (MTX) in rheumatoid arthritis (RA) is well-established; long-term trials have shown that continued treatment maintains clinical response and prevents radiographic damage.
Objectives To study whether ETN must be continued to maintain clinical and radiographic state, or if MTX alone, or MTX combined with a lower dose of ETN, might be equally effective.
Methods RA patients on ETN 50 mg/wk plus MTX (stable dose 7.5–25 mg/wk) and documented low disease activity or remission (LDA/REM; DAS28≤3.2) for ≥11 mths were included. Additional key inclusion criteria were: age ≥18 y, stable MTX, no prior biologics except anti-TNF, and no prior attempt to discontinue ETN due to stable disease. Patients were followed for 2 mths without changes in therapy to ensure stable LDA/REM, and stratified based on LDA/REM status and randomly assigned 1:1:1 to ETN 50 mg/wk (ETN50), ETN 25 mg/wk (ETN25), or placebo (PBO) while continuing MTX. Failure was defined as DAS28≥3.2 and an increase in DAS28≥0.6 or disease progression as determined by investigator or patient. The primary outcome was the % of non-failures at 48 wks for ETN50 vs. PBO (analyzed using Generalized Estimating Equations). Secondary outcomes included comparisons of non-failure and DAS28, time to failure, and predictors for failure (analyzed using Logistic modeling).
Results 106 patients were screened, 91 enrolled and 73 randomized. Baseline demographics were well-balanced among groups: 70% were female, average age (SD) was 57 (11) y, disease duration was 13.6 (8.8) y, and 81% had DAS28≤2.6. After 48 wks, the proportion of non-failures was 52% for ETN50 and 13% for PBO [OR 7.2 (1.7-29.8), p=0.007]. For ETN25, the % of non-failures was 44% [OR 4.2 (1.0-17.0); p=0.044 vs PBO (not sign. vs ETN50)]. Median time to failure was 48, 36, and 6 wks from randomization for ETN50, ETN25, and PBO, respectively. A greater likelihood of treatment failure was associated with the following baseline predictors (p<0.10): higher erosion scores, higher VAS pain score, and shorter duration of ETN treatment prior to screening. Taking potential predictors together, the final model for likelihood of treatment failure was associated with the following baseline predictors (p<0.05): higher erosion score and higher pain score. Adverse events were similar between groups.
Conclusions For RA patients with stable LDA/REM on ETN 50 mg/wk + MTX, continued treatment with ETN at 50 mg/wk or 25 mg/wk provides a significantly higher likelihood of maintaining stable disease state over 48 wks than MTX monotherapy. This trial suggests that an “induction-maintenance” strategy may be feasible in individual patients even in established RA, especially in patients without erosions with less pain, and with longer duration of ETN treatment. However, further and longer-term studies are needed to determine if this entails an increased risk for radiographic progression.
Acknowledgements This study was sponsored by Pfizer Inc. Medical writing support was provided by Kim Brown of UBC Scientific Solutions and was funded by Pfizer Inc.
Disclosure of Interest R. Vollenhoven Grant/research support from: Abbott, BMS, GSK, MSD, Pfizer, Roche, UCB, Consultant for: Abbott, BMS, GSK, MSD, Pfizer, Roche, UCB, K. Franck-Larsson Employee of: Pfizer Sweden (former), M. Leirisalo-Repo Consultant for: Pfizer, MSD, Roche, BMS, T. Uhlig Grant/research support from: Pfizer, Consultant for: Pfizer, M. Jansson Employee of: Pfizer Sweden, E. Larsson Employee of: Pfizer Sweden, K. Hutchinson Employee of: Quanticate, M. Østergaard Grant/research support from: Abbott, Pfizer, Centocor, Consultant for: Abbott, Pfizer, Merck, Roche, UCB, Speakers bureau: Abbott, Pfizer, Merck, BMS, UCB, Mundipharma