Background ADA has recently been approved by the European Medicines Agency for the treatment of adults with severe axial spondyloarthritis (axSpA) without radiographic evidence of AS (nr-axSpA), but with objective signs of inflammation by elevated c-reactive protein (CRP) and/or positive magnetic resonance imaging (MRI) who have had an inadequate response to, or are intolerant to non-steroidal anti-inflammatory drugs (NSAIDs).

Objectives To evaluate, from the Spanish National Health Service perspective, the cost effectiveness of ADA (40 mg every other week) vs. conventional care (CC) in the licensed population when used according to existing international treatment guidelines.

Methods A cohort Markov model was developed to compare over 40 years the cost and quality-adjusted life years (QALYs) of CC vs. ADA in the licensed population. A central model feature is that patients (pts) who do not respond or do not maintain response stop therapy. The model was primarily based on the ABILITY-1^[1] trial data (0-12 weeks blinded, randomized phase and 12-68 weeks open-label extension phase) and supplemented with literature. Using ABILITY-1 trial data, pts were categorized into ASAS40 responders (i.e., ≥40 percent improvement from baseline using the Assessment of SpondyloArthritis International Society criteria) and non-responders at week 12. Over time, responders were further categorized among those who stayed on therapy and those who discontinued, based on modelled discontinuation curves. The BASDAI and BASFI scores of pts were modeled over time conditional on responder (yes/no) and discontinuer (yes/no) status. In the base case, ADA non-responders and discontinuers returned to baseline BASDAI and BASFI scores before experiencing a slow increase in BASFI score (0.068 points/year). Utilities were linked to BASDAI/BASFI scores based on the observed ABILITY-1 data. Costs of drug, screening, initiation and monitoring (e.g., tuberculosis test), and adverse events were included. AxSpA-related costs were linked to BASDAI scores on the basis of an analysis of the Outcome Assessment in AS International Study (OASIS) registry^[2]. Discounting was applied at 3.0%/year for costs and benefits. Uncertainty was addressed via scenario, univariate and probabilistic sensitivity (PSA) analyses.

Results ADA pts gained an estimated 0.54 QALYs (10.34 vs. 9.80) at an additional net cost of €12,196 (€146,463 vs. €134,267) relative to CC. The use of ADA resulted in higher drug (€16,865), initiation (€439) and monitoring costs (€972) and lower axSpA-related costs (−€6,080, €128,187 vs. €134,267). Therefore, ADA resulted in an incremental cost-effectiveness ratio (ICER) of €22,787/QALY. The ICER was sensitive to the relationship between BASDAI/BASFI and axSpA-related costs and utilities in univariate analyses. In scenario analyses, the ICER ranged from €17,878/QALY to €53,924/QALY, depending on assumptions. In PSA, the ICER was ≤€31,450 in 95% of runs.

Conclusions Based on the present analysis, ADA appears cost effective when used according to the international treatment guidelines in nr-axSpA pts with elevated CRP and/or positive MRI and who have failed NSAIDs.

References

1. Ann Rheum Dis Online First: 7 July 2012.

2. Arthritis Rheum 1998;41:1119-25.

Disclosure of Interest X. Ji Employee of: Pharmerit, which received payment from AbbVie for this research analysis, A. Joshi Shareholder of: AbbVie, Employee of: AbbVie, S. Vidal Pérez-Campoamor Shareholder of: AbbVie, Employee of: AbbVie, R. Sanchez Hernandez Shareholder of: AbbVie, Employee of: AbbVie, M. Botteman Employee of: Pharmerit, which received payment from AbbVie for this research analysis, Y. Bao Shareholder of: AbbVie, Employee of: AbbVie

DOI 10.1136/annrheumdis-2014-eular.2403