PMS28 COST-UTILITY ANALYSIS OF TREATMENT OPTIONS AFTER INITIAL TUMOR NECROSIS FACTOR INHIBITOR THERAPY FAILURE IN PATIENTS WITH RHEUMATOID ARTHRITIS

Abstract

To use real-world data to analyze the cost-utility of sequences of therapeutic drugs used by rheumatoid arthritis patients whose initial tumor necrosis factor inhibitor therapy failed. We developed a microsimulation Markov model to analyze 10,000 rheumatoid arthritis patients (80% female, mean age 52yrs) beginning their second biological treatment with adalimumab (cycling) or abatacept (swapping) and followed them for 10 years or until death. Patients not responding switched to the next drug in a sequence of three (adalimumab-abatacept-tocilizumab versus abatacept-tocilizumab-rituximab) and then shifted to palliative treatment. Demographics, the most common drug choice per cycling or swapping cohorts, subsequent treatment sequences, direct medical costs, and transition probabilities were derived from a cohort of rheumatoid arthritis patients in the 2008-2016 Truven Health MarketScan Research database. Baseline Health Assessment Questionnaire Disability Index (HAQ-DI) were derived from an age-adjusted comorbidity index and converted to utilities using the formula from the Birmingham Rheumatoid Arthritis Model. Scenario analysis was used to vary the time horizon, utility formula and baseline HAQ-DI. We also performed a probabilistic sensitivity analysis. Swapping to a sequence beginning with abatacept showed an incremental discounted cost of $11,357 over ten years and achieved a discounted cost per quality-adjusted life year (QALY) benefit of 0.16 compared with cycling to adalimumab. The incremental cost-effectiveness ratio of $68,950/QALY was within current willingness-to-pay thresholds. Scenario analysis produced an incremental cost-effectiveness ratio range of $44,573/QALY to $148,558/QALY: abatacept was more cost-effective over longer periods of time and when baseline HAQ was lower. Probabilistic sensitivity analysis showed that swapping to abatacept after tumor necrosis factor inhibitor therapy failure had an 80.6% likelihood of being cost-effective at $100,000/QALY. Abatacept is a cost-effective strategy for rheumatoid arthritis patients whose initial tumor necrosis factor inhibitor therapy fails. Being able to stratify patients by reason for switch would improve the validity of our conclusions.