SAT0645 Diagnostic accuracy of serum autoantibodies multi-testing in rheumatoid arthritis and economic consequences across europe

Background Rheumatoid arthritis (RA) diagnosis requires interpretation of a combination of clinical, laboratory and imaging investigations. In RA, the consequences of incorrect serology test results are particularly important: patients with False Positive results (FPs) are initially managed as RA patients, bringing about extra costs until a correct diagnosis is made. Objectives The first aim of the study was to evaluate the diagnostic performance of RF-IgA, RF-IgM, and CCP, used alone or in multi-testing parallel or sequential combinations. The secondary goal focused on the economic consequences of serology FPs in selected European countries. Methods 190 established RA patients and 197 controls (either affected by other conditions or healthy donors) were used to assess the diagnostic performance of mono- and multi-serology testing in PC; both testing in Primary Care (PC) and in Secondary Care (SC) settings were considered. For the secondary objective, a 12 month Markov model simulated, from the National Health Services perspective, 1.000 RA-suspected individuals tested in PC and SC with mono- or multi-testing. Costs come from the published literature. The European countries included in the analysis were the United Kingdom, Austria, Belgium, Denmark, France, Germany, Italy, the Netherlands, Portugal, Spain and Sweden. Uncertainty was addressed with sensitivity analysis. Results The mono-testing diagnostic performance was: - RF-IgA: sensitivity [95% CI]=40.5% [33.5%, 47.9%], specificity=92.4% [87.8%, 95.7%]; - RF-IgM: sensitivity=59.0% [51.6%, 66.0%], specificity=89.3% [84.2%, 93.3%]; - CCP: sensitivity=59.5% [52.1%, 66.5%], specificity=96.5% [92.8%, 98.6%]. In multi-testing, defining a “positive result” as “positivity to at least one test” increased sensitivity; “positivity to all the tests” increased specificity. In a PC scenario: - parallel testing: - o using CCP and RF-IgM increased specificity to 99.5% [97.2%, 100.0%]; - o the three tests used simultaneously maximised specificity (100.0% [98.1%, 100.0%]), but reduced sensitivity to 35.8% [29.0%, 43.1%]; - sequential testing: testing positive to both RF-IgA and RF-IgM followed by CCP testing led to 90.7% [81.7%, 96.2%] sensitivity and to 100.0% [54.1%, 100.0%] specificity. With respect to mono-testing, multi-testing options reduced the number of FPs. Therefore, in each of the countries considered, multi-testing allowed for important cost savings due to reduced clinical procedures and resource utilisation of FPs. Conclusions Multi-serology testing improve the diagnostic accuracy of the individual RF-IgA, RF-IgM and CCP tests. Optimal multi-testing combinations minimise the number of FPs, thus reducing avoidable costs to the National Health Services. Consequently, multi-testing for RA demonstrates superior value from patient and payer perspective. References [1] Hallert E, et al. Rheumatology2006;45(3):325–331. [2] Lajas C, et al. Arthritis Rheum2003;49(1):64–70. [3] Lundkvist J, et al. Eur J Health Econ 2008;(Suppl 2):S49–60. [4] Verstappen SM, et al. Ann Rheum Dis2004;63(7):817–24. Disclosure of Interest B. Mascialino Employee of: Thermo Fisher Scientific, S. Swiniarski Employee of: Thermo Fisher Scientific, I. Gehring Employee of: Thermo Fisher Scientific, M. Poorafshar Employee of: Thermo Fisher Scientific