Long-term efficacy of mepolizumab in patients with eosinophilic granulomatosis with polyangiitis: a propensity score matching analysis in the multicenter REVEAL cohort study

Abstract

Background: Mepolizumab (MPZ) has demonstrated efficacy in clinical trials for eosinophilic granulomatosis with polyangiitis (EGPA); however, few studies compare the disease course between patients treated with MPZ (MPZ group) and those who were not treated with MPZ (non-MPZ group) in real-world settings.

Objectives: This study aimed to compare the disease course and outcomes between the two groups and assess the long-term efficacy of MPZ in a multicenter cohort in Japan. Methods: We enrolled 113 EGPA patients registered in the cohort until June 2023. Data on clinical characteristics, disease activity, organ damage, treatments, and outcomes were retrospectively collected. To minimize potential confounding factors, we conducted propensity score matching (PSM).

Results: After PSM, 37 pairs of matched patients were identified. Clinical characteristics, including age at disease onset, sex, disease duration at last observation, antineutrophil cytoplasmic antibody positivity at disease onset, Birmingham Vasculitis Activity Score (BVAS) at disease onset, and Five-factor score at disease onset, were comparable between the groups. The median BVAS at the last observation was 0 in both groups; however, more cases in the non-MPZ group exhibited elevated BVAS, resulting in a significantly higher BVAS in the non-MPZ group at the last observation (median; MPZ group: 0, non-MPZ group: 0, p=0.028). The MPZ group had significantly lower glucocorticoid (GC) doses at the last observation (median; MPZ group: 4 mg/day, non-MPZ group: 5 mg/day, p=0.011), with a higher proportion achieving a GC dose ≤ 4 mg/day at the last observation (MPZ group: 51.4%, non-MPZ group: 24.2%, p=0.027). Three models of multivariable logistic regression analyses were performed to identify factors associated with GC doses ≤ 4 mg/day at the last observation. In all models, achieving a GC dose ≤ 4 mg/day was positively associated with MPZ administration and inversely associated with asthma at disease onset. Finally, we evaluated the survival rates between the groups, and the 5-year survival rates were significantly higher in the MPZ group compared to the non-MPZ group (MPZ group: 100%, non-MPZ group: 81.3%, p=0.012).

Conclusion: Mepolizumab not only contributes to disease activity control but also reduces the GC dose, which may lead to improved survival in EGPA patients.

Keywords: eosinophilic granulomatosis with polyangiitis; glucocorticoid-sparing; mepolizumab; multicenter cohort; survival rate.

Figure 1

Changes in BVAS and VDI scores between the two groups from six months after onset to the last observation. (A) BVAS at 6 months after onset. (B) BVAS at 12 months after onset. (C) BVAS at the last observation. (D) VDI scores 12 months after onset. (E) VDI scores 24 months after onset. (F) VDI scores at the last observation. Mann-Whitney U test was used to compare differences between the two groups. *P<0.05. BVAS, Birmingham vasculitis activity score; MPZ, mepolizumab; VDI, vasculitis damage index.

Figure 2

Glucocorticoid-sparing effect of MPZ. (A) Serial changes in glucocorticoid dose in the two groups at 6, 12, and 24 months after onset and at the last observation. (B) Percentage achieving glucocorticoid-free status at the last observation. (C) Percentage achieving glucocorticoid dose ≤ 4 mg/day at the last observation. Fisher’s exact test or the Mann-Whitney U test was used to compare differences between the two groups. *P<0.05. GC, glucocorticoid; MPZ, mepolizumab.

Figure 3

The relapse-free survival rates and five-year survival rates in the two groups. (A) The relapse-free survival rate for all relapses. (B) The relapse-free survival rate for systemic vasculitis. (C) The relapse-free survival rate for asthma and ENT manifestations. (D) The five-year survival rate. (E) The five-year survival rate stratified by the median age of 60 years. (F) The five-year survival rate stratified by the median year of onset in 2015. Relapse-free survival rates and survival rates were calculated by the Kaplan–Meier method and compared using the log-rank test. *P<0.05. ENT, ear, nose, and throat; MPZ, mepolizumab.