Anti‐βc mAb CSL311 inhibits human nasal polyp pathophysiology in a humanized mouse xenograft model

Abstract

To the Editor, Chronic rhinosinusitis with nasal polyps (CRSwNP) is an in‐ flammatory disease of the nose and paranasal sinuses character‐ ized by eosinophilia, elevated levels of local IgE and type 2 (T2) inflammatory cytokines1 and mucus production. Interleukin (IL)‐5, together with the other T2 beta common (βc) cytokines IL‐3 and granulocyte‐macrophage colony‐stimulating factor (GM‐CSF), sup‐ ports the survival and enhances the differentiation and activation of myeloid cells, which play a key role in the pathogenesis of this disease.2,3 We have developed a fully human therapeutic monoclonal an‐ tibody (mAb), CSL311, with specificity for the common cytokine binding site of the human βc receptor. 4 The ability of CSL311 to block IL‐3, IL‐5 and GM‐CSF signals simultaneously may offer dis‐ tinct advantages in the treatment of inflammatory diseases com‐ pared with mAbs‐targeting single cytokines. Preclinical testing of CSL311 is limited in animal models due to the species specificity of CSL311. To overcome this constraint, we modified the human nasal polyp (NP) xenograft model described previously5 and utilized Rag2−/−Il2rg−/−hIL‐3/GM‐CSF knock‐in mice, which express human IL‐3 and GM‐CSF to support longer‐term survival, growth and dif‐ ferentiation of human myeloid cells to evaluate, for the first time, the in vivo efficacy of CSL311, on NP progression in a preclinical proof‐of‐concept study. NPs from 12 patients, who were withdrawn from any medication 1 month prior to polypectomy, were used in this study. The most fre‐ quent comorbidities in these patients were allergic rhinitis (50%) and asthma (41.7%) (Table S1); consistent with atopy, histological analysis revealed that the NPs were highly eosinophilic (11/12), demonstrat‐ ing characteristic T2 inflammation (data not shown). High levels of βc cytokines were detected in the NP compared with normal sinus tis‐ sues (Figure S1), suggesting a role of βc cytokines in contributing to the pathogenesis of CRSwNP. NP from each patient was engrafted into 4‐10 mice with all treatment groups included in each indepen‐ dent experiment (Table S2). CSL311 significantly suppressed NP progression in recipient mice after 5 weeks compared with isotype control mAb (Figure 1A, Figure S3). We found that treatment with CSL311 for 1 week resulted in an immediate reduction in NP vol‐ ume measured externally, which was not observed with positive control prednisolone treatment. After 5‐week treatment, CSL311 significantly suppressed NP volume in recipient mice when com‐ pared with isotype control mAb. To understand how CSL311 treatment inhibited NP growth, we examined the immune cell profiles in the xenografted NPs col‐ lected at the end of the 5‐week engraftment. The percentage of eosinophils, neutrophils and plasma B cells (Figure 1B, Figure S2, Table S3), as well as the number of mast cells (Figure 1C), were significantly reduced in the presence of CSL311 when compared with isotype control mAb treatment. These cells are known to express βc receptor and when activated by βc cytokines produce chemical mediators that directly promote tissue remodelling and oedema. Next, we assessed the effect of CSL311 on mucous gland hyperplasia, a general feature of NPs.6 Histological examination re‐ vealed a significant decrease in mucous gland number and mucus production in NPs treated with CSL311 that was not observed in those treated with isotype control mAb (Figure 1D, Figure S3). The stroma of NPs contains increased levels of fibroblasts.7 Fibroblast‐ specific protein 1(FSP‐1) is expressed in fibroblasts in different organs that undergo tissue remodelling and is commonly used as a marker to identify fibroblasts.8 Using immunofluorescence stain‐ ing, we found that CSL311 significantly decreased the percentage of FSP‐1‐positive fibroblasts compared with isotype control mAb (Figure 1E, Figure S3). We also performed transcriptome analysis by RNA sequencing to understand the impact of CSL311 on the global gene expres‐ sion profile in the xenografted NPs. Treatment with CSL311 re‐ sulted in differential expression of 29 genes with a fold change >2 (|log2FC| > 1) and false discovery rate <0.05. The top 25 differen‐ tially expressed (DE) genes (Figure 2A) were used to calculate a gene score based on the average counts per gene, which confirmed that CSL311 significantly reduces expression of this gene set compared with isotype control mAb; no significant effect of prednisolone was observed on expression of these genes (Figure 2B). The average log fold change (log2FC) in gene expression, induced by CSL311 vs iso‐ type control mAb and prednisolone vs saline, is shown in Figure 2C. These DE genes encode the following: (a) surface markers associ‐ ated with macrophage/dendritic cells. for example CD14, CD206 and CD209, and granulocytic cell subsets, for example CLEC4A, CD32 and FCERG1, (b) chemokines that recruit T2 cells (basophils, eosin‐ ophils and TH2 cells), for example CCL7, CCL13, CCL18 and CCL23 and (c) inflammatory response proteins including complement com‐ ponents C1QA, C1QB and C1QC. Ingenuity Pathway Analysis (IPA) upstream regulator tool, which predicts the most likely regulators of the DE genes, identified GM‐CSF (Figure 2D) and IL‐3 (Table S4) Kwok Ho Yip and Nicholas J. Wilson should be considered joint first authors.