Background/Purpose: Randomized–controlled trials (RCTs) showed rituximab’s (RTX) noninferiority to cyclophosphamide for induction therapy of severe ANCA-associated vasculitides and significantly lower relapse rate than azathioprine maintenance. However, those RCTs enrolled granulomatosis with polyangiitis (GPA) and microscopic polyangiitis patients, who might need to be studied separately, and selected patients with low preexisting comorbidities that preclude generalization. This open-label, real-life, single-center cohort study was undertaken to assess efficacy of RTX induction-and-maintenance therapy for GPA patients.

Methods: All consecutive adults with GPA who received ≥1 RTX infusions for induction (April 2005–December 2016) at Cochin National Referral Center for Vasculitis were included. Patients had to satisfy the ACR classification criteria or revised Chapel Hill Nomenclature for GPA. Those enrolled in prospective trials were ineligible. After remission, preemptive low-dose RTX maintenance was given. Kaplan–Meier estimations of relapse-free survival were tested for significance with log-rank tests. Cox regression with time-dependent covariates was used to identify predictors of relapses, serious adverse events (SAEs) or serious infections.

Results: During the study, 114 patients, with relapsing (65%) or refractory/grumbling (22%) active GPA, received RTX induction therapy; 100 were given ≥1 RTX maintenance infusions and 90 received 500 mg every 6 months. The median [IQR] prednisone dose at first RTX infusion was 30 [20–50] mg/day. After the first RTX infusion, median follow-up was 3.6 [1.6–5.8] years; respective 1- and 2-year remission rates were 83.7% and 86.4%; respective 1-, 2-, and 3-year relapse-free survival rates were 93% (95% CI 88–98), 85% (78–92) and 82% (74–90); and the 2-year RTX-retention rate was 74% (SE 5%). Respective SAE or serious infection rates were 8.1 and 4.9 per 100-patient-years. Two patients died. Univariate analyses identified pure granulomatous disease (P=0.017), refractory/grumbling vs. new-onset/relapsing GPA (P<0.001), pachymeningitis (P=0.048) or estimated glomerular filtration rate >60 ml/min (P=0.005) as factors associated with less likely remission. Multivariate analyses retained subglottic stenosis (HR 4.88, 95% CI 1.79–13.25; P=0.002), ENT involvement (HR 2.91, 1.37–6.20; P=0.01), skin involvement (HR 5.20, 2.18–12.44; P=0.0003) and refractory/grumbling vs. new-onset/relapsing GPA (HR 4.73, 1.04–21.43; P=0.05) as independent predictors of relapse. Only the prednisone dose over time was associated with SAEs or serious infections (HR 1.03, 1.01–1.05 and HR 1.06, 1.02–1.09, respectively; both P<0.005).

Conclusion: Cohort studies and RCTs provide complementary information. The 2-year 74% RTX-retention rate with relatively low toxicity in this real-life setting has important implications. The glucocorticoid dose was the major factor associated with SAEs, even in this cohort whose median dose at inclusion was already lower than that used in the pivotal RCTs.

« Back to 2017 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/rituximab-for-induction-and-maintenance-therapy-of-granulomatosis-with-polyangiitis-a-single-center-cohort-study-on-114-patients/