Op0163 Identification of Anca-Associated Vasculitis Relapse Risk Factors for Building a Personalized Maintenance Therapy Scheme Model

Background: ANCA associated vasculitides (AAV) are a heterogeneous group of rare diseases with unknown etiology and the clinical spectrum ranging from life-threatening systemic disease, through single organ involvement to minor isolated skin changes. Individual disease course prediction, prognosis, and maintenance treatment regimen selection create difficulties due to the heterogeneity of the AAV. The ability to predict the risk of relapse in AAV course is crucial for decision about maintenance therapy duration. It also forms an important unmet need in actual guidelines 1. Objectives: To identify risk factors of AAV relapse and build a model for personalized prediction of AAV exacerbation. Methods: We conducted a national multicenter study of adult patients diagnosed with AAV (648–GPA, 170–MPA) 2. Their clinical and laboratory data were collected in the POLVAS registry by 12 referral centers. Cox proportional hazards analyses were applied to calculate hazard ratios for the first relapse as the main endpoint. First, one-dimensional models were used to identify potentially relevant variables. Then, using stepwise regression with different order of inclusion and exclusion of variables, a multidimensional model was obtained. Results: Data analysis from 818 patients identified seven significant risk factors of AAV relapse: gender, skin, ENT or eye involvement, maximal ever creatinine 10 ng/ml at baseline (Table 1). In the next step AAV patients were divided into 5 groups with different risk of relapse (HP) over time (Figure 1) using following algorithm: hi, i ∈ 1,. . . 7 are hazard ratios for risk factors obtained in analysis. HP value determines to which group the patient belongs: : Group 1: HP≤1. 5, Group 2: 1. 55. The Relapse free survival rate for all five groups is shown in Figure 1. These results suggest the in the groups 4 and 5 over 80 % of cases will have AAV relapse, whereas in the groups 1 and 2 less than 60%. If we look at the time course of relapse occurrence we can observe the fast decrease of relapse free survival in the first 36-48 months, later on the dynamic of relapse rate in all groups is similar. Comparison of different groups suggests that in group 1 the first 24 months are crucial with ca 50% of cases having no relapse during next 240 months. In groups 2 and 3 the highest number of relapses occur in the first 36 months. In groups 4 and 5 patients have the highest risk of relapse and what is more almost all of them will experience the AAV relapse in long term perspective. Conclusion: POLVAS registry data analysis identified AAV relapse risk factors for and allowed to build a model able to define AAV patients' subsets characterized by different probability of disease relapse which may help to guide personalized decisions about the duration and type of maintenance therapy. REFERENCES: 1 Hellmich B et al. Ann Rheum Dis. 2023 Mar 16: ard-2022-223764. doi: 10. 1136/ard-2022-223764. 2 Wójcik K et al. Clin Rheumatol. 2019 Sep;38 (9): 2553-2563. Acknowledgements: This work is supported by the Jagiellonian University Medical College under Grant No. N41/DBS/001188. Disclosure of Interests: None declared.