#4183 A PHENOTYPIC DRIVEN RITUXIMAB MAINTENANCE APPROACH IS EFFECTIVE AND SAFE IN PATIENTS WITH ANCA ASSOCIATED VASCULITIS

Abstract

Abstract Background and Aims Rituximab (RTX) has emerged as the first-line treatment to maintain remission in ANCA-Associated Vasculitis (AAV). However, the ideal strategy to RTX re-dosing is still unclear [1]. Method In this monocentric and retrospective study we evaluated the efficacy and safety of two RTX-based maintenance regimens in patients with AAV. Fixed-dose (FD) RTX dosing, consisting of at least 3 pre-emptive RTX administrations every 6 months, was preferred in patients judged to be at high relapse risk. Instead, a biomarker-guided on-demand (OD) RTX dosing strategy was preferred for patients deemed at lower risk of relapse; this consisted in RTX re-treatment only in case of B-cell repopulation or rise in ANCA titer. The single patient relapse risk was assessed according to the disease phenotype (granulomatous vs vasculitic), the ANCA specificity and the disease status (onset vs relapse). Relapses were defined as increase of disease activity requiring escalation of immunosuppression; major relapses were defined in case of life/organ-threatening manifestations. Results The study included 100 patients followed for a median of 24 months (IQR 20-30), 51% treated with an OD and 49% with a FD strategy. The main baseline characteristics of our cohort are shown in the Table; patients treated with a FD strategy were more often GPA, PR3-ANCA positive, with relapsing disease and ENT involvement and less frequent renal involvement. The FD group received a median of 4 (IQR 4-4) administrations of RTX for a median cumulative dose of 2 g (IQR 2-2). Only 6 (11.8%) patients in the OD group received RTX as maintenance therapy with a median of 1 (IQR 1-3) administration of RTX for a median cumulative dose of 0.75 g (IQR 0.5-1.75). Thirteen relapses occurred, 6 in the FD group and 7 in the OD one. Remission rates were comparable in the two cohorts, with respectively 87.6% and 85.9% of the patients in remission at 24 months (Figure 1 - panel A). No significant differences in remission rates across the 2 groups were observed after considering only major flares or after stratification according to clinical diagnosis (MPA/GPA) and ANCA specificity (Figure 1 - panel B to F). Ten and 20 severe infections (p = 0.383) and 3 and 1 cancer (p = 0.581) occurred respectively in the FD and OD group. Conclusion In our case series, a phenotypic-driven approach to RTX maintenance showed comparable efficacy and safety between fixed and on-demand dosing, despite different baseline characteristics between the two groups. These results suggest that personalizing RTX maintenance may be a feasible and safe strategy.