Background Once remission is achieved for patients with rheumatoid arthritis (RA), treatment down-titration should be attempted, for safety issues or economic reasons. One of the proposed strategies, recently tested in the STRASS trial , is to progressively space injections of TNF-blockers.
Objectives To assess the Incremental Cost-Effectiveness Ratio (ICER) of a strategy of progressive spacing of TNF-blocker injections (S-arm) over another maintaining them at full-dose (M-arm) in RA patients in stable remission.
Methods The study was a French multi-centre 18-month equivalence randomized open-label controlled trial. It included patients receiving etanercept (ETA) or adalimumab (ADA) at stable dose for ≥1 year; in remission on 28-joint Disease Activity Score (DAS28) for ≥6 months; and with stable joint damage. In the S-arm, the interval between 2 subcutaneous injections was increased every 3 months by 50% in 4 steps, to a complete stop at step 4. If remission was not maintained, spacing was suspended or reversed to the previous interval. Costs engaged within the study-period, measured in euros, were assessed using a health insurance payer perspective encompassing medical costs (drugs, consultations and medical tests, use of emergency room and hospitalizations) and costs relative to sick leave. Utilities values used to compute Quality Adjusted Life Years (QALYs) were derived from the EQ-5D, using values validated in the French population, assessed at baseline and every 6 months. The ICER was estimated. A probabilistic sensitivity analysis was performed by computing 5000 ICERs (bootstrap). The probability of cost-effectiveness (p of CE) of the maintenance strategy was computed for different Willingness to Pay (WTP) thresholds.
Results Analyses were performed on 44 patients in the S-arm and 54 in the M-arm with complete data. In the S-arm, TNF-blockers were stopped for 34.1% of the patients, tapered for 43.2%, maintained at full-dose for 18.2%. After 18 months of follow-up, 62.3% and 45.9% of the patients of the S and M-arm respectively, found their symptomatic state acceptable (p=0.08). Patients in the S-arm gained 1.106 QALYs while it was 1.264 in the M-arm (mean differences in QALYs of -0.158). After 18 months, total mean cost was 12452 euros in the S-arm and 20892 euros in the M-arm (mean cost difference of -8440 euros). The cost of TNF-blockers represented 81.0% and 92.8% of the total cost in the S and M-arms respectively. The estimated ICER was 53417 euros per QALY. The p of CE of the maintenance strategy was 0.02, 0.06, 0.22 and 0.41 for WTP thresholds fixed at 25000, 30000, 40000 and 50000 euros respectively.
Conclusions Although 62.3% of the patients found their symptomatic state acceptable, the spacing strategy was found to be less effective based on QALY measures. But, it was also found to be less costly. Thus, the maintenance strategy was assessed as the strategy with the lowest probability of being cost-effective for WTP thresholds ranging from 25000 to 50000 euros.
Fautrel et al, EULAR 2013 and ACR 2013
Disclosure of Interest None declared