Glucocorticoids vs glucocorticoids plus cyclophosphamide in eosinophilic granulomatosis with polyangiitis without poor-prognosis factors: a target trial emulation study

Abstract

Objectives: Current recommendations suggest treating eosinophilic granulomatosis with polyangiitis (EGPA) without severe manifestations with glucocorticoids (GCs) and EGPA with severe manifestations with GCs plus cyclophosphamide (CYC) regardless of poor-prognostic factors. However, GCs plus CYC and GCs alone have never been compared in EGPA without poor-prognosis factors assessed by the 1996 Five Factor Score (FFS). We aimed to compare the efficacy of GCs plus CYC vs GCs alone for the treatment of EGPA without poor-prognosis, including among patients with severe manifestations.

Methods: We emulated a target trial using observational data from a European multicentre retrospective study. We included patients with (i) newly diagnosed EGPA, (ii) a FFS = 0 at diagnosis and (iii) treated with GCs or GCs plus CYC. Primary outcome was overall relapse at 12 months. Inverse probability of treatment weighting-based analysis was used to adjust for potential confounding factors. In a subgroup analysis, we focused on patients with severe manifestations not included in the FFS.

Results: A total of 250 patients were included: 177 treated with GCs alone and 73 with GCs plus CYC. After adjustment, no reduction in the risk of overall relapse was observed between the two treatment groups. Similar results were observed in the subgroup of patients with severe manifestations.

Conclusion: This study shows that the adjunction of CYC to GCs does not reduce the risk of relapse in patients with EGPA and no poor-prognosis factors. It supports current guidelines in patients without severe manifestations but challenges the need of CYC adjunction in patients with severe manifestations but no poor-prognosis factors.

Keywords: ANCA-associated vasculitides; cyclophosphamide; eosinophilic granulomatosis with polyangiitis; poor-prognosis factor.