Purpose: To evaluate the long-term outcomes of rituximab (RTX) treatment in patients with ocular granulomatosis with polyangiitis (GPA) with localized or generalized disease.
Design: Retrospective cohort.
Participants: Thirty-seven patients with ocular GPA receiving RTX in a multidisciplinary vasculitis clinic between 2004 and 2013.
Methods: A total of 100 patients who received a course of RTX were identified, and notes were reviewed. Baseline demographic details, clinical characteristics (including organ involvement), drugs used, and outcome measures were recorded.
Main outcome measures: The percentage in remission (inactive disease with prednisolone ≤7.5 mg with or without maintenance treatment) at 6 months, time to remission, percentage relapsing, side effects, B-cell count, antineutrophil cytoplasm antibody titers, induction, and maintenance regimens.
Results: The median follow-up time after the first RTX course was 36.5 months. Twenty patients had scleritis, and 17 patients had orbital disease; 86% achieved remission at 6 months. The percentage in remission versus partial remission was not statistically significant between patients with scleritis and patients with orbital disease (85% vs. 15% with scleritis and 82% vs. 18% with orbital disease; P = 1.00). The percentage relapsing was not statistically significant (P = 0.33) between scleritis (60%) and orbital disease (41%). Localized disease (ocular ± ear-nose-throat/lung) was observed in 57%, and generalized disease (ocular plus other organs) was observed in 43%, the former having a median duration of disease of 40 months. There was no statistically significant difference (P = 0.37) in the percentage in remission between localized and generalized ocular disease. Relapses occurred in 51%, with localized disease being a significant risk factor for relapse. Fifty percent of patients with generalized disease versus none with localized disease received cyclophosphamide (CYP) as part of the induction regimen. Patients who received CYP during induction had significantly (P = 0.027) lower ratios of baseline 12-month proteinase 3 titers than patients who did not have CYP. Infections were observed in 16% of patients, with 8% requiring hospital admission.
Conclusions: Our long-term data suggest that RTX is effective for inducing disease remission in localized and generalized ocular GPA. Localized disease is a significant risk factor for relapse, which may be related to less use of CYP in the induction regimen.
Copyright © 2015 American Academy of Ophthalmology. Published by Elsevier Inc. All rights reserved.