Safety of rituximab for remission maintenance in relapsing anca–associated vasculitis: repeat cycles on clinical relapse are associated with low rates of hypogammaglobulinaemia.

Background: Rituximab is effective in ANCA-associated vasculitis (AAV). Since relapse often occurs after an initial response, either fixed-interval rituximab re-treatment or re-treatment-at-relapse is used to maintain long term response. Attrition of immunoglobulin levels and risk of infection are a potential concern with repeat cycles of B cell depletion with up to 67% of patients with low IgG after 36 months in one series[1]. We previously reported good long term efficacy using re-treatment-at-relapse[2]. Objectives: To evaluate immunoglobulin levels, serious adverse events and serious infections during rituximab treatment of refractory AAV based on the re-treatment-at-relapse strategy. Methods: We conducted an observational study of patients with AAV treated with at least 2 cycles of rituximab in a single centre between January 2006 and December 2012 – a total of 142 patient-years follow up. Each cycle consisted of 2x1000mg infusions repeated on clinical relapse. IgG levels were measured at baseline and 6 months after each cycle (normal range 6.0-16.0). Severe adverse events were defined as those resulting in hospitalisation which lasted more than 24 hours, flares requiring intravenous therapy, malignancies, life-threatening situations or death. Results: 30 patients (18 male and 12 female) with a median age of 58.9 years old (range 21-83) were identified. 22 patients were PR3 positive, 7 were MPO positive and 1 was ANCA negative but with a positive histology for granulomatosis with polyangiitis (GPA). Numbers of repeat cycles are shown in Table 1. Median time to re-treatment for cycles 1-4 were 82, 71, 58 and 57 weeks respectively. There were 26 serious adverse events: 1 death due to bronchopneumonia and 26 hospital admissions (median duration 6 days) were recorded in 14 patients. Of these, 9 were due to flares requiring intravenous methylprednisolone and 8 were serious infection (5.6/100 patient-years), mostly lower respiratory tract infection. 6 patients with serious infection had lower IgM at baseline and a greater reduction in IgG, although was not statistically significant. IgG data are shown in Table 1. 4 patients had low IgG before rituximab and only 1 developed low IgG during therapy. 1 patient with low IgG at baseline had two serious infections and required intravenous immunoglobulin. No other patient had to discontinue rituximab therapy due to low IgG. Conclusions: Re-treatment at relapse did not result in progressive reduction in IgG in patients with normal immunoglobulins at baseline. Serious infection rates were low. This is of particular importance since there are limited other effective agents for this relapsing disease.