LAUSR

Background/Aims  Time-to-relapse after rituximab for ANCA vasculitis is variable and optimal retreatment strategy is unclear. We previously showed that repopulation of naïve B-cells at 6 months predicts sustained response. Our objective was to evaluate clinical and B-cell predictors for relapse in multivariable analysis to develop a retreatment algorithm. Methods  An observational study was conducted in 60 rituximab-treated AAV patients over 10-years. Complete response was defined as a Birmingham Vasculitis Activity Score (BVAS) v3.0=0. Repeat rituximab cycles were given on clinical relapse. Naïve, memory B-cells and plasmablasts were measured using highly sensitive flow cytometry. Clinical and B cell predictors were analysed in multivariable analysis (MVA) to develop a retreatment algorithm. Results  Patients were: 33/60 (55%) male, mean(SD) age 51(19), 39/60 (65%) received concomitant immunosuppressants and 54/60 (90%) had prior therapy with cyclophosphamide. Median times to retreatment for cycles 1-5 were 87, 71, 65, 59 and 86 weeks. Over 417 patient-years follow up, 137 relapses occurred in 50 patients; 16 (in 14 patients) were major (renal=7, neurological=4, ENT=3 and lungs=2). The major-relapse rate was 3.8/100 patient-years. Results of MVA are shown in the table. With naïve B-cell repopulation at 6 months, relapse rates at 12 and 18 months were 5% and 16%, versus 45% and 59% without naïve repopulation (p < 0.001). Relapse rates at 18 and 24 months were 17% and 54% with naïve repopulation at 12 months versus and 29% and 71% without (p = 0.045). AUROC for time-to-relapse was greater if retreatment was guided by naïve B-cell repopulation than if ANCA and/or CD19+ return was used at 6 months, 0.82 and 0.52 respectively. Conclusion  These results suggest the following: (i) all patients should receive an oral immunosuppressant; (ii) patients with incomplete remission should be retreated at 6 months; (iii) patients with complete remission but absent naïve B cells at 6 months should be retreated at 6 months; (iv) patients with complete remission and naïve B-cell return may not be appropriate for fixed retreatment. For this last group monitoring of clinical symptoms and B-cells may be effective and future work will address this further. This algorithm could avoid hypogammaglobulinaemia due to unnecessary retreatment. P207 Table 1:Factors associated with time-to-relapse to first cycle rituximabRisk FactorsUnivariable analysisHR (95% CI); p-values(with multiple imputation)Multivariable analysis (MVA)HR (95% CI); p-values(with multiple imputation)Age, per 10 years (BL)1.01 (0.86 – 1.17); p = 0.954Not included in MVAFemale (BL)1.15 (0.65 – 2.02); p = 0.629Not included in MVADisease duration, years (BL)1.06 (0.98 – 1.15); p = 0.160Included in MVA but removed from final model as p < 0.20Concomitant immunosuppressant (BL)0.69 (0.39 – 1.22); p = 0.2050.48 (0.24 – 0.94); p = 0.034Positive ANCA immunofluorescence (BL)0.89 (0.46 – 1.71); p = 0.725Not included in MVAPositive anti-PR3/anti-MPO (BL)0.57 (0.31 – 1.06); p = 0.077Included in MVA but removed from final model as p < 0.20CRP, mg/L (BL)1.00 (0.99 – 1.01); p = 0.456Not included in MVABVAS 3.0 per point score (BL)0.99 (0.94 – 1.05); p = 0.763Included in MVA but removed from final model as p < 0.20VDI per point score (BL)1.14 (0.87 – 1.50); p = 0.353Not included in MVANaïve B-cells, x 10 9/L* (BL)1.00 (1.00 – 1.01); p = 0.797Not included in MVAMemory B-cells, x 10 9/L* (BL)1.01 (1.00 – 1.02); p = 0.0401.01 (1.00 – 1.02); p = 0.045Plasmablasts, x 10 9/L* (BL)1.04 (0.94 – 1.16); p = 0.459Not included in MVAComplete depletion at 6 Weeks post-RTX0.90 (0.50 – 1.61); p = 0.721Not included in MVAComplete Response (26Wk)0.34 (0.19 – 0.61); p < 0.0010.24 (0.12 – 0.50); p < 0.001Positive ANCA immunofluorescence (26Wk)0.99 (0.56 – 1.75); p = 0.962Not included in MVAPositive anti-PR3/anti-MPO (26Wk)0.79 (0.44 – 1.42); p = 0.426Not included in MVACRP, mg/L (26Wk)0.99 (0.97 – 1.02); p = 0.618Not included in MVANaïve B-cell repopulation (26Wk)0.38 (0.19 – 0.76); p = 0.0060.43 (0.22 – 0.84); p = 0.013Memory B-cell repopulation (26Wk)0.45 (0.20 – 0.99); p = 0.046Included in MVA but removed from final model as p < 0.20Plasmablast cell repopulation (26Wk)1.14 (0.61 – 2.13); p = 0.675Not included in MVABL = baseline, 26Wk = 26 weeks. Disclosure  J.E. Arnold: None. E.M.J. Vital: Honoraria; Roche, GSK, AstraZeneca. S. Dass: Honoraria; Roche, GSK. A. Aslam: None. A.C. Rawstron: None. S. Savic: Honoraria; Novartis, Swedish Orphan Biovitrum, Sire. Grants/research support; Novartis, Swedish Orphan Biovitrum, Octapharma, CSL Behring. P. Emery: Consultancies; BMS, Abbott, Pfizer, MSD, Novartis, Roche, UCB. Grants/research support; Abbott, BMS, Pfizer, MSD, Roche. M. Md Yusof: None.