Importance: Results of randomized clinical trials have demonstrated rituximab's noninferiority to cyclophosphamide as induction therapy for antineutrophil cytoplasm antibody (ANCA)-associated vasculitides (AAV), with neither treatment having a specific advantage for granulomatosis with polyangiitis (GPA). However, post hoc analysis results have suggested that rituximab might be more effective than cyclophosphamide in inducing remission in patients with proteinase 3-positive AAV.
Objective: To compare the effectiveness of rituximab and cyclophosphamide in inducing GPA remission in a large population of unselected patients.
Design, setting, and participants: This comparative effectiveness study used multicenter target trial emulation observational data from 32 French hospitals in the French Vasculitis Study Group Registry. Groups were determined according to treatments received, without any intervention from the investigators. Inverse probability of treatment weighting was used to correct for baseline imbalance between groups. Participants included patients with newly diagnosed or relapsing GPA who satisfied American College of Rheumatology classification criteria and/or Chapel Hill Consensus Conference nomenclature. Data were analyzed from October 1, 2021, to May 31, 2022.
Exposures: At least 1 infusion of rituximab or cyclophosphamide for induction therapy between April 1, 2008, and April 1, 2018.
Main outcomes and measures: The primary outcome was remission rate at month 6 (±2 months), with remission defined as a Birmingham Vasculitis Activity Score (BVAS) of 0 and prednisone dose of 10 mg/d or less. The BVAS is a validated tool for small-vessel vasculitis and used to assess the level of disease activity, with a numerical weight attached to each involved organ system. The BVAS has a range of 0 to 63 points; a score of 0 indicates no disease activity. Subgroup analyses included the primary outcome for patients with a new diagnosis, for most recently treated patients, and for patients with myeloperoxidase-ANCA positivity.
Results: Among 194 patients with GPA included in the analysis (mean [SD] age, 54 [15] years; 110 men [56.7%]), 165 (85.1%) had a new diagnosis, and 147 of 182 with data available (80.8%) had proteinase 3-ANCA positivity. Sixty-one patients received rituximab and 133 received cyclophosphamide for induction therapy. In the weighted analysis, the primary outcome was reached for 73.1% of patients receiving rituximab vs 40.1% receiving cyclophosphamide (relative risk [RR], 1.82 [95% CI, 1.22-2.73]; risk difference, 33.0% [95% CI, 12.2%-53.8%]; E value for RR, 3.05). Similar results were observed in the subgroup of patients with newly diagnosed GPA and those with a more recent treatment. In the subset of 27 patients with myeloperoxidase-ANCA-positive GPA, 8 of 10 rituximab recipients and 8 of 17 cyclophosphamide recipients met the primary end point (unweighted RR, 1.73 [95% CI, 0.96-3.11]).
Conclusions and relevance: In this comparativeness effectiveness study using clinical data, rituximab induction therapy for GPA was more frequently associated with remission than cyclophosphamide. These results inform clinical decision-making concerning the choice of remission induction therapy for this subset of patients with AAV.