POS0140 PREDICTING OUTCOMES IN SYSTEMIC SCLEROSIS: STRATIFICATION BY AUTO-ANTIBODIES OUTPERFORMS CUTANEOUS SUBSETTING IN THE EUSTAR COHORT

BackgroundRisk-stratification is key in a heterogeneous disease like systemic sclerosis (SSc). Until now, SSc patients are stratified according to the extent of skin involvement into limited cutaneous, diffuse cutaneous and sine scleroderma subtypes. However, this classification remains inaccurate to capture disease heterogeneity. Autoantibodies are found in more than 90% of the patients and can be detected before onset of the disease. Among them, three predominant and specific antibodies are used: anti-centromere, anti-Scl70 and RNA polymerase III antibodies.ObjectivesTo compare the performances of stratification into LeRoy’s cutaneous subtypes versus autoantibody status in SSc versus combination of cutaneous subtypes and autoantibodies status.MethodsPatients from the EUSTAR database were classified either as (i) limited cutaneous, diffuse cutaneous or sine scleroderma (based on the recording made by the treating physician) or (ii) according to autoantibodies with the following subclassifications: (1) no specific autoantibodies, (2) isolated ANA, (3) anti-centromere antibodies, (4) anti-Scl70 antibodies and (5) anti-RNA polymerase III antibodies or (iii) according to combination of cutaneous subset and auto-antibodies. The respective performance of each model to predict overall survival (OS), progression-free survival (PFS), disease progression and different organ involvements was assessed and the three models were compared by the area under the receiver operating characteristic curve (AUC 95%CI) and the net reclassification improvement (NRI). Missing data were imputed through multiple imputation using chain equations.ResultsIn all, 10’711 patients were included: 84.6% females, mean age: 54.4±13.8 years, mean disease duration: 7.9±8.2 years. In the prospective analysis (n= 6’467 to 7’829 according to the outcome), after a mean follow-up of 56 months and a mean of three visits per patient, we did not identify any difference in AUC between the cutaneous-based model and the antibody-based model for prediction of OS and disease progression. However, the NRI showed a significant improvement in prediction of OS (0.57 [0.46-0.71] vs. 0.29 [0.19-0.39]) and disease progression (0.36 [0.29-0.46] vs. 0.21 [0.14-0.28]) at 4 years using the antibody-based model. Regarding prediction of each organ involvement in longitudinal analyses, the antibody-based model showed better performance than the cutaneous-one for renal crisis (AUC: 0.719 [0.696-0.742] vs. 0.664 [0.643-0.685]), with the highest association observed with anti-RNA polymerase III (OR: 7.47 [1.63-34.24], p= 0.010). Similarly, the antibody-based model was better than the cutaneous model in predicting lung fibrosis (AUC 0.719 [0.715-724] vs. 0.653 [0.647-0.659]) and restrictive lung fibrosis (AUC 0.759 [0.749-0.766] vs. 0.711 [0.701-0.721]) which were both associated with anti-Scl70 antibodies (OR: 9.29 [8.17-10.55] and 7.92 [5.37-11.69], respectively, pConclusionAuto-antibody status outperforms the common cutaneous subsetting to risk-stratify SSc patients in the EUSTAR cohort. This easily performed subclassification using autoantibodies specific status can be used by the clinicians to risk-stratify their patients and to adapt disease monitoring in routine practice.Disclosure of InterestsMuriel Elhai Speakers bureau: BMS outside of the submitted work, Marouane Boubaya: None declared, Nanthara Sritharan: None declared, Alexandra Balbir-Gurman: None declared, Elise Siegert: None declared, Eric Hachulla: None declared, Jeska de Vries-Bouwstra: None declared, Gabriela Riemekasten: None declared, Jörg H.W. Distler: None declared, Douglas Veale: None declared, Edoardo Rosato: None declared, Francesco Del Galdo: None declared, Fabian A Mendoza: None declared, Daniel Furst Consultant of: Abbvie, Novartis, Pfizer, R-Pharm, Grant/research support from: Emerald, Kadmon, PICORI, Pfizer,Prometheus, Talaris, Mitsubishi, Carlos De la Puente Bujidos: None declared, Anna-Maria Hoffmann-Vold Speakers bureau: Actelion, Boehringer Ingelheim, Jansen, Lilly, Medscape, Merck Sharp & Dohme, Roche, Consultant of: Actelion, ARXX, Bayer, Boehringer Ingelheim, Jansen, Lilly, Medscape, Merck Sharp & Dohme, Roche, Grant/research support from: Boehringer Ingelheim, Armando Gabrielli: None declared, Oliver Distler Speakers bureau: Bayer, Boehringer Ingelheim, Janssen, Medscape, Consultant of: Abbvie, Acceleron, Alcimed, Amgen, AnaMar, Arxx, AstraZeneca, Baecon, Blade, Bayer, Boehringer Ingelheim, Corbus, CSL Behring, 4P Science, Galapagos, Glenmark, Horizon, Inventiva, Kymera, Lupin, Miltenyi Biotec, Mitsubishi Tanabe, MSD, Novartis, Prometheus, Roivant, Sanofi and Topadur, Grant/research support from: Kymera, Mitsubishi Tanabe, Boehringer Ingelheim, Coralie Bloch-Queyrat: None declared, Yannick Allanore Consultant of: Actelion, Bayer, BMS, Boehringer-Ingelheim, Inventiva, Roche, Sanofi-Aventis, Grant/research support from: Actelion, Bayer, BMS, Boehringer-Ingelheim, Inventiva, Roche, Sanofi-Aventis