BASELINE IgG-Fc N-GLYCOSYLATION PROFILE IS ASSOCIATED WITH LONG-TERM OUTCOME IN A COHORT OF EARLY INFLAMMATORY ARTHRITIS PATIENTS

Background: Rheumatoid arthritis (RA) is a systemic autoimmune disease which causes chronic joint inflammation and functional limitation . The diagnosis of rheumatoid arthritis (RA) is mainly based on clinical data and RA specific autoantibodies, while the prediction of long-term prognosis from disease outset is not clinically reliable . The importance of immunoglobulin G (IgG) and its Fc N -glycosylation in inflammation of RA has been described, with changes in the glycosylation profiles observed years before the diagnosis of RA . Objectives: We herein sought to assess the value of total serum IgG Fc N -glycosylation as a diagnostic and prognostic biomarker in patients with early inflammatory arthritis (EIA). Specifically, we aim to assess whether IgG N -glycoform levels may predict the diagnosis of RA or undifferentiated arthritis (UA) and the long-term disease‘s outcome in patients with EIA arthritis patients naïve to treatment. Methods: The “Early Arthritis Clinic” of the University Hospital of Heraklion is a prospective cohort of patients with inflammatory arthritis. For the present study, we selected a group of patients naïve to any immunosuppressive treatments with available serum at baseline evaluation (n=118). At baseline, demographics, RA clinical characteristics (DAS28, HAQ-DI) and laboratory tests [autoantibodies (RF and/or ACPA)], were also recorded. The patients were prospectively followed for two years, with clinical, laboratory and disease-related treatments documented. A diagnosis of RA or UA was based on established classification criteria . In order to assess long-term prognosis we formulated a combined “index” of favourable outcome if the patients fulfilled all the following at 24 months of follow-up: remission or low disease activity (based on DAS28 < 3.2) and normal functionality (based on HAQ ≤ 0.25) while on treatment with csDMARDs and never use bDMARDs. We applied a state-of-the-art liquid chromatography - mass spectrometry (LC-MS) based workflow for analysis of subclass-specific IgG Fc N -glycosylation at the baseline . Results: We studied 118 EIA patients [age (mean, SD) (53, 15.6) years, females (80.5%), symptoms duration (53.8, 8.7) years, ACPA positive (16%), DAS28 (4.8, 0.14)]. During the 2 years of follow-up, 60% of the patients were diagnosed with RA and 40% with UA. Although patients with UA had higher relative abundances of galactosylated and sialylated N -glycoforms (H4N4F1, H5N4F1 and H5N4F1S1) in all IgG subclasses at baseline compared to RA patients, differences were not statistically important. Interestingly, we observed a significant association between high levels of IgG2/3 galactosylation for H5N4F1 [effect 0.63, adjusted p=0.036)] and H3N4F1 [effect -0.55122, adjusted p=0.0496) and favorable outcome after two years of treatment. Conclusion: In our cohort of EIA we found IgG2/3 Fc N -glycoforms to be associated with a favorable prognosis after 2 years of follow-up. Should the present data be confirmed in a larger cohort could be of clinical value. Since currently available prognostic tools have significant limitations, further research should aim to the development of a predictive tool of high specificity and sensitivity based on the combination of clinical, serological data and novel biomarkers. REFERENCES: [1]Smolen JS, et al. Lancet (2016) 388(10055):2023-38. doi: 10.1016/S0140-6736(16)30173-8 [2]Firestein GS, McInnes IB. Immunity (2017) 46(2):183-96. doi: 10.1016/j.immuni.2017.02.006. [3]Scott DL, et al .J. Lancet (2010) 376(9746):1094-108. doi: Doi 10.1016/S0140-6736(10)60826-4. [4]Weyand CM, Goronzy JJ. Nat Immunol (2021) 22(1):10-8. doi: 10.1038/s41590-020-00816-x. [5]Ligier S, et al. Br J Rheumatol (1998) 37(12):1307-14. doi: 10.1093/rheumatology/37.12.1307. [6]Aletaha D, et al. Ann Rheum Dis (2010) 69(9):1580-8. doi: 10.1136/ard.2010.138461. [7]De Leoz MLA, et al. Molecular & Cellular Proteomics (2020) 19(1):11-30. doi: 10.1074/mcp.RA119.001677. Acknowledgements: This research was funded by the GlySign and SYSCID – European Union’s Horizon 2020 research and innovation programs under the Marie Skłodowska-Curie, grant numbers 722095 and 733100, respectively. Disclosure of Interests: Thomas Sénard: None declared, Irini Flouri: None declared, Frano Vučković Employee of: Genos Ltd, Garyfalia Papadaki: None declared, Panagiota Goutakoli: None declared, Aggelos Banos: None declared, Maja Pučić-Baković Employee of: Genos Ltd, Marija Pezer Employee of: Genos Ltd, George Bertsias: None declared, Gordan Lauc Shareholder of: Genos Ltd, a private research organization that specializes in high-throughput glycomic analyses and has several patents in this field., Prodromos Sidiropoulos: None declared Citation: , volume 81, supplement 1, year 2022, page 1176Session: Rheumatoid arthritis - aetiology, pathogenesis and animal models (Publication Only)