FRI0008 ADDRESSING THE DIAGNOSTIC GAP IN RHEUMATOID ARTHRITIS BY COMPLEMENTING THE SEROLOGY WITH GENETIC INFORMATION

Abstract

Background: Rheumatoid arthritis (RA) is caused by an interaction of inherited and environmental factors. The genetic component was revealed in different twin studies and the shared epitope was identified as main contributor. During several genome-wide association studies (GWAS) different genes have been highlighted as relevant for developing RA but so far, none of these have been used for a diagnostic approach to address the diagnostic gap for RA. Objectives: The aim of this study was to identify genes within pathways relevant for developing RA and to combine these genetic risk factors with serologic data to improve the diagnosis of RA, especially in regards to CCP negative RA patients. Methods: The cohort consists of 804 RA patients, 159 Disease controls and 495 Healthy individuals. Serology data were obtained using the EliATM CCP IgG, EliA RF IgM and EliA RF IgA (Phadia AB, Uppsala, Sweden) tests. The genetic measurements were performed using the Next-Generation-Sequencing technology AmpliSeqTM on the Ion GeneStudioTM instruments (Thermo Fisher Scientific, Carlsbad, USA). The combined dataset was analysed by machine learning focusing on multivariate supervised models and discrete models. Results: The Disease Research Area database (Thermo Fisher Scientific, Carlsbad, USA) uses a proprietary algorithm to create connections between genes and a specific disease highlighting relevant genes and pathways using the available datasets from NCBI, DisGeNet, ClinVar and further databases. Based on this information we were able to identify genes and variants within relevant pathways for RA and designed a sequencing panel specific for diagnosis of RA. The results from the targeted sequencing approach and the serologic data of the patients were analysed by an algorithm resulting in an improved diagnosis. Each variant within the algorithm has a specific weighting score resulting in a combination of protective and destructive variants, which are used together with the serology to calculate a RA risk score. In this study the combination of genetic and serology leads to an improved sensitivity of over 10% compared to CCP alone, with a comparable specificity. This increase in sensitivity results in the identification of >20% CCP-negative RA patients. Additionally, a genetic pattern was identified distinguishing between CCP-negative and CCP-positive RA patients. Conclusion: Usage of genetic variants in combination with serologic data improves the diagnosis of RA within the measured cohort of Caucasian RA patients. Furthermore, a genetic pattern specific for CCP negative RA patients was accentuated and could be used to improve the diagnosis of CCP negative RA patients in the future. The combination of serology and genetic information for the diagnosis of RA patients could lead to an earlier diagnosis, especially for CCP negative RA patients.