Benralizumab reduces sputum ANCA in patients with eosinophilic granulomatosis with polyangiitis

Extract

Eosinophilic granulomatosis with polyangiitis (EGPA) is a rare vasculitis characterised by eosinophilic inflammation and autoimmune features, with anti-neutrophil cytoplasmic antibodies (ANCA) against myeloperoxidase (MPO) (pANCA) detectable in serum in up to 40% of patients. Seronegative EGPA can be associated with severe cardiac and pulmonary complications, including uncontrolled asthma [1]. We previously observed ANCA in sputum (spANCA) in 74% of EGPA patients, regardless of serum pANCA status, all of whom had severe asthma requiring high-dose oral corticosteroids [2].

Shareable abstract

In EGPA, benralizumab reduced airway autoimmune responses (spANCA and IL-17A) and eosinophilic inflammation, while mepolizumab was linked to persistent or increased autoimmune activity https://bit.ly/4hJJIYv

To the Editor:

Eosinophilic granulomatosis with polyangiitis (EGPA) is a rare vasculitis characterised by eosinophilic inflammation and autoimmune features, with anti-neutrophil cytoplasmic antibodies (ANCA) against myeloperoxidase (MPO) (pANCA) detectable in serum in up to 40% of patients. Seronegative EGPA can be associated with severe cardiac and pulmonary complications, including uncontrolled asthma [1]. We previously observed ANCA in sputum (spANCA) in 74% of EGPA patients, regardless of serum pANCA status, all of whom had severe asthma requiring high-dose oral corticosteroids [2]. These findings suggest that a localised airway autoimmune response may drive disease severity. Mepolizumab and benralizumab – monoclonal antibodies targeting interleukin (IL)-5 and the IL-5 receptor α-subunit (IL-5Rα), respectively – are approved therapies that induce remission and reduce corticosteroid use in EGPA. Benralizumab has demonstrated noninferiority to mepolizumab in relapsing or refractory cases [3]. Our primary objective was to investigate whether benralizumab and mepolizumab exert differential effects on airway autoimmune responses in EGPA. Airway autoimmunity is increasingly recognised as a key determinant of pulmonary outcomes in EGPA and its modulation may influence treatment efficacy [2, 4]. To address this, we analysed sputum, nasal lavage and blood samples from 16 patients enrolled in the randomised, double-blind MANDARA trial (www.clinicaltrials.gov identifier number NCT04157348).

This mechanistic substudy was conducted under the MANDARA clinical trial with approval from the Hamilton Integrated Research Ethics Board (#8195). A total of 24 international sites initially agreed to participate in the substudy; however, only three sites were ultimately able to recruit patients due to COVID-19-related restrictions. The final cohort (n=16) consisted of patients from these sites with no perceived selection bias. All participants provided written informed consent. Detailed inclusion and exclusion criteria were previously described [3]. Patients in the substudy were randomised to receive benralizumab (n=7; mean age 52±11 years, 57% female; median (range) inhaled corticosteroid (ICS) dose 1000 (500–2000) μg·day⁻¹; oral corticosteroid (OCS) dose 7.5 (7.5–12.5) mg·day⁻¹) or mepolizumab (n=9; mean age 58±16 years, 78% female; ICS dose 625 (500–2000) μg/day; OCS dose 7.5 (7.5–35) mg·day⁻¹). No significant differences in demographic or baseline corticosteroid use were observed between the treatment arms (p>0.05, data not shown). Sputum, nasal lavage and blood samples were collected at baseline and at week 52 (or week 48 for nasal lavage) to assess eosinophil counts, eosinophil activity markers (eosinophil peroxidase (EPX) and eosinophil-derived neurotoxin (EDN)) and inflammatory mediators. spANCA and citrullinated histone H3 (citH3) levels were measured to evaluate airway autoimmunity. Natural killer (NK) cells, group 2 innate lymphoid cells (ILC2) and B-cells were characterised to assess immune mechanisms driving therapeutic responses in EGPA [5–7].

Of the 16 patients receiving either mepolizumab or benralizumab, all but one patient (on mepolizumab) achieved remission (Birmingham Vasculitis Activity Score 0, OCS ≤4.0 mg·day⁻¹). Sputum eosinophils showed a decreasing trend in both treatment groups, although this was not statistically significant (p>0.05) (figure 1a). Notably, using a previously established cutoff of ≥2.3% sputum eosinophils [8], more patients treated with benralizumab had normalisation of sputum eosinophils compared to those treated with mepolizumab (p=0.0015) (figure 1b). Blood eosinophil and EDN levels showed a significant reduction in patients treated with mepolizumab (p<0.005) (figure 1a). In patients receiving benralizumab, blood eosinophil and EDN levels exhibited a decreasing trend (p>0.05) (figure 1a). This holds true for all but one patient, who exhibited an increase in sputum and blood eosinophils post-treatment (0% to 16.7% in sputum) despite achieving remission. This patient had low baseline NK-cell levels and reduced killer-cell immunoglobulin-like receptor (KIR)-expressing NK-cells (CD158a and CD158e1); while it is biologically plausible that low NK-cell levels could impair eosinophil depletion via antibody-dependent cell-mediated cytotoxicity, this is a single observation that requires further investigation before any conclusions can be drawn. No significant difference in total NK-cells or their subsets were observed in either treatment arm, at baseline and post-treatment (p>0.05, data not shown). IL-5 levels were comparable in sputum and nasal lavage across both treatment groups but were elevated in the blood of patients on mepolizumab (p=0.0039) (figure 1a). MPO levels remained consistent across blood, sputum and nasal lavage, with no significant differences observed between treatment arms (figure 1a).

FIGURE 1.

Changes in the inflammatory and autoimmune mediators following mepolizumab and benralizumab treatment. a) Inflammatory profile of serum, nasal lavage and sputum of eosinophile granulomatosis with polyangiitis (EGPA) patients with mepolizumab and benralizumab treatments. Data shown for two visits (baseline (week 0) and post-treatment (week 48 for nasal lavage, week 52 for blood and sputum)). Based on Shapiro–Wilk normality testing, data are presented as mean±sd or median (range) as appropriate. In addition, change is presented as mean±sd or median (range), and a paired t-test/Wilcoxon matched pairs signed rank test was performed, respectively. b) Proportion of patients with sputum eosinophilia (≥2.3%) following treatment. Three patients in the mepolizumab group and one patient in the benralizumab group had sputum eosinophilia (≥2.3%) post-treatment. c) Changes (Δ) in sputum anti-neutrophil cytoplasmic antibody (ANCA) between week 0 and week 52 of mepolizumab and benralizumab treatment. d) Correlation of Δ sputum eosinophil peroxidase (EPX) and Δ sputum ANCA. e) Correlation of sputum citrullinated histone H3 (citH3) and proportion double negative (DN) (CD27⁻ IgD⁻) B-cells in the total CD45⁺ lymphocyte population in patients on both treatment arms at baseline and post-treatment. MPO: myeloperoxidase; EDN: eosinophil-derived neurotoxin; IL: interleukin; AU: arbitrary unit. *: p<0.05; **: p<0.01.

The levels of type 2-associated cytokines in sputum, including IL-5, IL-13 and IL-4, were comparable at baseline and post-treatment in both treatment arms (p>0.05) (figure 1a). A distinct pattern was observed for IL-17A, which increased post-treatment in the mepolizumab arm but not in the benralizumab arm (p=0.040) (figure 1a). No significant differences were observed in the levels of other sputum inflammatory markers in either treatment arm (figure 1a). Reduction in blood and sputum IL-5Rα⁺ ILC2 was previously shown in severe asthma patients following benralizumab treatment [5]. In this study, neither treatment significantly affected total ILC2 or IL-5Rα⁺ ILC2 numbers, in blood or sputum, likely due to the limited number of matched pre- and post-treatment samples. A trend towards reduction of blood IL-13⁺ ILC2s was observed in patients receiving benralizumab; although this difference did not reach statistical significance (baseline 179 (0–288), post-treatment 0 (0–0) cells per mL; p=0.13). However, a significant proportion of patients treated with benralizumab exhibited a complete depletion of circulating IL-5⁺ ILC2 and IL-13⁺ ILC2 (p<0.001).

We have previously demonstrated that spANCA is associated with pulmonary complications in EGPA and can induce neutrophil extracellular trap (NETosis) and eosinophil extracellular trap (EETosis) formation, releasing pro-inflammatory mediators and self-antigens that exacerbate airway autoimmunity [2]. A significant difference in spANCA change from baseline to week 52 was observed between treatment arms (p=0.04) (figure 1c), with 62% (six out of nine) of patients on mepolizumab and 29% (two out of seven) on benralizumab remaining spANCA⁺. The increase in spANCA with mepolizumab may reflect sustained local autoimmune activity, potentially driven by immune complex formation [4]. This is supported by elevated sputum IL-17A, a cytokine that promotes Bcell activation and autoantibody production via neutrophil recruitment [9] (figure 1a). Changes in spANCA levels correlated with changes in sputum EPX, highlighting the role of eosinophilic activity in sustaining airway autoimmunity (p=0.01, ρ=0.66) (figure 1d). Eosinophil degranulation releases EPX and other nuclear/extranuclear antigens that persist in the airway tissue even in the absence of intact eosinophils [10]. These antigens activate self-reactive lymphocytes, driving autoantibody production that promotes tissue damage and inflammation, contributing to symptom worsening [10, 11]. The relationship between eosinophilic activity and autoimmunity underscores the need to address both inflammation and the underlying autoimmune drivers in EGPA management.

Given the role of B-cells in autoimmune responses and airway autoimmunity [6], we explored the effect of treatment on circulating B-cell subsets. No significant differences were observed in the proportions of naïve B-cells, CD27⁺ memory B-cells, transitional B-cells or CD27⁻ IgD⁻ double-negative (DN) B-cells in either treatment arm (data not shown). Additionally, B-cells expressing regulatory-associated markers, such as CD5 and CD25, were comparable before and after treatment in both groups (p>0.05, data not shown). A significant, positive correlation was identified between circulating DN B-cells and sputum citH3 (p=0.04, ρ=0.47) (figure 1e), while a negative correlation was observed between CD5⁺ regulatory B-cells and sputum citH3 (p=0.049, ρ=0.46; data not shown). We have previously shown DN B-cells to be both increased in the sputum of patients with severe eosinophilic asthma, and associated with eosinophilic inflammation and autoantibody production [6]. Given citH3 is a marker of both EETosis and NETosis, and is implicated in airway autoimmunity [4], these findings highlight a potential link between specific B-cell subsets and autoimmune activity in the airways. Further investigation into these associations may provide insights into the mechanisms of airway autoimmunity and the role of B-cell subsets in treatment response.

In conclusion, benralizumab reduced both airway autoimmune responses (spANCA and IL-17A) and eosinophilic inflammation in patients with EGPA, whereas mepolizumab was associated with persistent or increased autoimmune activity. Despite the limited sample size, these findings underscore the distinct immunological effects of the two therapies and suggest that benralizumab may offer added benefit for patients with prominent airway autoimmunity.

Data availability

Data underlying the findings described in this manuscript may be obtained in accordance with AstraZeneca's data sharing policy described at https://astrazenecagrouptrials.pharmacm.com/ST/Submission/Disclosure. Data for studies directly listed on Vivli can be requested through Vivli at www.vivli.org. Data for studies not listed on Vivli could be requested through Vivli at https://vivli.org/members/enquiries-about-studies-not-listed-on-the-vivli-platform/. AstraZeneca Vivli member page is also available outlining further details: https://vivli.org/ourmember/astrazeneca/.