Allergic Bronchopulmonary Aspergillosis/Mycosis: An Allergic Disease or an Eosinophilic Disease?

Authors: Koichiro Asano, Tsuyoshi Oguma

Abstract

Allergic bronchopulmonary aspergillosis/mycosis (ABPA/ABPM) is characterized by increased serum levels of total and fungi-specific immunoglobulin E (IgE) and eosinophilic mucus plugs in the airways. Its classification as either an allergic or eosinophilic disease remains controversial. In the present review, we explored this topic based on three clinical studies that analyzed the clinical characteristics of ABPA/ABPM using a cluster analysis, factor analysis, and comparison between ABPM caused by Schizophyllum commune and ABPA. We also compared therapeutic responses to biologics targeting either IgE (omalizumab) or eosinophils (mepolizumab/benralizumab) to elucidate the role of these components in the pathogenesis of ABPA/ABPM. Based on these analyses, eosinophilic mucus plug formation in the airways is considered a cardinal feature of the development of ABPA/ABPM, whereas IgE responses to fungi are important factors that modulate disease manifestation.

Introduction

Allergic airway diseases, such as atopic asthma and allergic rhinitis, develop mostly during childhood or adolescence due to immunoglobulin E (IgE)-mediated hypersensitivity reactions to allergens. IgE-mediated airway diseases often precede or co-develop with other allergic diseases in different organs, such as food allergies and eczema. The consecutive development of childhood-onset allergic diseases is termed an allergic (or atopic) march (1-4). In contrast, non-atopic asthma mostly develops in middle-aged or older patients, often with chronic rhinosinusitis (CRS) with nasal polyps or eosinophilic CRS, and is characterized by local and systemic eosinophilia (3,5,6). Patients with “adult-onset eosinophilic airway diseases” sometimes develop other eosinophilic diseases, such as nonsteroidal anti-inflammatory drug-exacerbated airway disease (aspirin-induced asthma) and eosinophilic granulomatosis with polyangiitis (3). Therefore, there are substantial differences in the genetic, immunological, and pathological backgrounds of childhood-onset IgE-mediated allergic and adult-onset eosinophilic airway diseases.

Allergic bronchopulmonary aspergillosis/mycosis (ABPA/ABPM) is another adult-onset eosinophilic airway disease, despite having an allergic component characterized by sensitization to fungal allergens (7-9). Hinson et al. (10) reported the first three cases of “allergic type of bronchopulmonary aspergillosis” in 1952 and mentioned that the condition resembled “pulmonary eosinophilia” proposed by Crofton et al. (11) or “pulmonary infiltration with eosinophilia” proposed by Reeder et al. (12). From a pathological perspective, ABPA/ABPM has also been considered an eosinophilic disease, as originally proposed by Hinson. Bosken et al. defined the pathological features of ABPA/ABPM as eosinophilic mucus plugs or bronchocentric granulomatosis with advanced eosinophilic infiltration along with fungal hyphae in these lesions without tissue invasion (13).

However, since IgE was discovered in 1966, researchers and clinicians have focused on IgE-mediated hypersensitivity to fungi and increased levels of total IgE in the serum as key components for the diagnosis of ABPA/ABPM, as in most of the current diagnostic criteria (7,14,15). According to the guidelines of the International Society for Human and Animal Mycology ABPA Working Group (ISHAM-AWG), the presence of fungus-specific IgE and total IgE levels in serum ≥500 IU/mL are considered the major criteria for the diagnosis of ABPA/ABPM (9). In contrast, the diagnosis of allergic fungal rhinosinusitis (AFRS), a similar disease in the upper airway, is based on the presence of “eosinophilic mucin” and “fungi on staining” as well as the presence of type I hypersensitivity to fungi (16). Notably, the steering group of the European Position Paper on Rhinosinusitis and Nasal Polyps 2020 discussed whether or not the term AFRS should be replaced by “eosinophilic fungal rhinosinusitis” (17).

Therefore, the classification of ABPA/ABPM as either an allergic or eosinophilic disease remains controversial. This review explored this ongoing debate through the analysis of clinical characteristics and therapeutic responses to biologics targeting either IgE or eosinophils in patients with ABPA/ABPM.

IgE and Eosinophils in Clinical Presentation of ABPA/ABPM

A cluster analysis of ABPA in a nationwide survey

The first nationwide survey of ABPA/ABPM in Japan was conducted in 2013, and data were collected from 499 physician-diagnosed ABPA/ABPM cases from 132 clinical centers (18). The obtained data were incomplete for a retrospective diagnosis using the criteria proposed by the ISHAM-AWG, which were not yet available at the time of the survey. Therefore, cases satisfying two minimum criteria of positivity for Aspergillus fumigatus-specific IgE antibody in the serum or an immediate skin reaction to A. fumigatus antigen and the presence of bronchiectasis or mucoid impaction in the central bronchi on thoracic computed tomography (CT) were used to define possible ABPA-central bronchiectasis (CB). As a result, 349 possible ABPA-CB cases were identified and analyzed.

A two-step cluster analysis was performed to examine the clinical presentation and possible phenotypes of ABPA-CB. Nine parameters were the age at the ABPA onset, sex, presence of asthma, serum total IgE levels, positivity for precipitins/IgG specific for A. fumigatus, presence of characteristic CT findings [bronchiectasis, mucoid impaction, lung consolidation/ground-glass opacity (GGO)], and recurrences/flares. Continuous variables were standardized using z-scores and categorical variables were standardized as 0 or 1. First, a dendrogram was generated to estimate the optimal number of clusters using Ward's minimum-variance hierarchical clustering method. Following the estimation of the appropriate number of clusters, a non-hierarchical analysis using the k-means clustering algorithm was performed using the same nine parameters.

Three clusters of ABPA-CB have been identified in Japan. Male-dominant cluster #1 with middle-age-onset ABPA was found to represent 28% (n=97) of the total population, whereas female-dominant clusters #2 and #3 with late-onset disease represented 34% (n=117) and 39% (n=135), respectively (Table 1). In addition to the differences in the age at the onset and sex, there were large differences in the serum levels of total IgE among the clusters. The median total IgE level was ＞3,000 IU/mL in clusters #1 and #3, whereas the median IgE level was 524 IU/mL in cluster #2. In contrast, the peripheral blood eosinophil counts were equivalent among the three clusters. These results suggested that eosinophilia and mucus plug formation can be induced independently of IgE-mediated hypersensitivity responses to fungi.

A factor analysis of ABPA in a prospective cohort study

The Japan Allergic Bronchopulmonary Mycosis Research Program prospectively registered patients ≥16 years old with physician-diagnosed ABPA/ABPM, fungus-sensitized asthma, and other related diseases from 14 medical institutes in Japan between 2013 and 2023. Data from 106 well-characterized patients with ABPA diagnosed using Asano's criteria (19) were extracted from the database and examined using a factor analysis (20).

Our analysis identified three clinical components of ABPA: allergic, eosinophilic, and fungal components (Table 2). The allergic component was defined mainly by serum levels of total and allergen (A. fumigatus and house dust mite)-specific IgE, followed by lung consolidation/GGO and younger age at ABPA onset. The eosinophilic components included peripheral blood eosinophil counts, mucus plugs, and HAM. The fungal component included a positive culture of Aspergillus spp. in sputum or bronchial lavage fluid, presence of A. fumigatus-specific precipitin/IgG antibody, an older age at the onset of ABPA, and a lack of comorbid asthma. These results support the hypothesis that IgE-mediated hypersensitivity and eosinophilic mucus plug formation are independent responses to airway fungi.

We compared the factor scores for each component of the three types of ABPA: ABPA sans asthma, ABPA with atopic asthma, and ABPA with non-atopic asthma. ABPA sans asthma refers to a disease without a history of asthma or other predisposing conditions before or at the onset of ABPA. Atopic asthma is defined as physician-diagnosed asthma accompanied by sensitization to house dust mites, cat dander, and/or cockroaches. The proportions of patients with ABPA sans asthma, ABPA with atopic asthma, and ABPA with non-atopic asthma were 24%, 54%, and 23%, respectively. Factor scores for the allergic component were highest in ABPA patients with atopic asthma, and those for the fungal component were highest in ABPA patients sans asthma (Fig. 1). In contrast, the factor scores for the eosinophilic component were not significantly different, regardless of asthma comorbidity or atopic predisposition.

We next examined which component was associated with a good ABPA prognosis, defined as disease remission achieved within six months after the diagnosis and no recurrence of ABPA within 12 months after remission. Patients who achieved and maintained disease remission showed significantly lower scores for the allergic component; however, there were no marked differences in the scores for the eosinophilic or fungal components. These results suggest that eosinophilia and mucus plug formation are cardinal features of ABPA, and that the allergic component of ABPA is an important modifier of disease presentation.

ABPM caused by Schizophillum commune (ABPM-Sc) and ABPA

S. commune is a basidiomycetous fungus commonly found in rotten wood that can cause ABPM because of small size of its conidia and thermophilicity. Since the first case of ABPM-Sc was reported by Kamei et al. in 1994 (21), almost all ABPM-Sc cases have been reported in Japan (22). Asano et al. (19) developed diagnostic criteria with excellent sensitivity and specificity that can be applied to cases of not only ABPA but also ABPM. Toyotome et al. (23) identified Sch c 1, a glucoamylase, as a major allergen of S. commune and developed an enzyme-linked immunosorbent assay to measure Sch c 1-specific IgG and IgE. With these advances, we analyzed the clinical characteristics of 33 cases of ABPM-Sc and compared them with those of ABPA (24).

ABPM-Sc develops in patients with milder asthma and a less impaired pulmonary function than those with ABPA. Although there was no marked difference in peripheral blood eosinophil counts and total serum IgE levels between the ABPM-Sc and ABPA groups, we found that the ABPM-Sc group could be divided into two subgroups with significantly different clinical characteristics according to co-sensitization to A. fumigatus. The median serum IgE level in patients with ABPM-Sc who were not sensitized to A. fumigatus (n=10) was 562 IU/mL, which was significantly lower than that in patients with ABPM-Sc who were sensitized to A. fumigatus (2,366 IU/mL, n=19) or ABPA (1,954 IU/mL, n=46) (Table 3). In contrast, there were no marked differences in peripheral blood eosinophil counts or the prevalence of mucus plugs/high attenuation mucus. These results provide further evidence that eosinophilic mucus plug formation in the airways is essential for the development of ABPA/ABPM and is independent of IgE responses to fungi.

Treatment of ABPA/ABPM with Biologics Targeting IgE or Eosinophils

Standard treatments for ABPA/ABPM include oral corticosteroids and/or triazole antifungals administered for three to five months (25). The initial response to treatment was generally good, and the response rate after 6 weeks of treatment was 88-100% (26-28). However, approximately half of the patients who received standard treatment presented with exacerbation of ABPA/ABPM during tapering or immediately after treatment discontinuation. To avoid adverse effects and the emergence of azole-resistant fungi caused by long-term treatment with oral corticosteroids and/or antifungals, biologics targeting either IgE or eosinophils can be used to treat difficult-to-treat ABPA/ABPM (25). Although there are few randomized controlled trials (RCTs) for this class of drugs in cases of ABPA/ABPM, a substantial number of case reports and case series studies have described the efficacy of omalizumab, an anti-IgE antibody; mepolizumab, a neutralizing antibody against the eosinophil growth factor interleukin (IL)-5; and benralizumab, an anti-IL-5 receptor alpha chain antibody.

Omalizumab

Eight case series studies (29-36) including ≥10 cases of ABPA and 1 small RCT (37) have reported the efficacy and limitations of omalizumab as a treatment for ABPA complicated by severe asthma. The outcomes evaluated in these studies included respiratory symptoms, exacerbation rate of asthma, doses of oral corticosteroids and antifungals, and the pulmonary function. A few studies have also evaluated radiological improvement and the exacerbation rate of ABPA.

In almost all studies, treatment with omalizumab suppressed the exacerbation rate of asthma and reduced the necessary dose of oral corticosteroids (Table 4). Four of the five studies that examined improvements in respiratory symptoms were positive. However, the effects of omalizumab on the pulmonary function are inconsistent. One study demonstrated some radiological improvement, in which omalizumab decreased endobronchial mucus plugs in three cases (35).

In a systematic review that quantitatively analyzed 14 reports, including 186 patients treated with omalizumab for asthma- or cystic fibrosis-associated ABPA, the exacerbation rate decreased from 1.75-4.25/year to 0.20-1.00/year (38). The dose of oral corticosteroids was reduced in 65% of patients, and 53% of patients were able to completely discontinue the drug. However, improvements in the imaging findings have rarely been reported. Considering these results, omalizumab is expected to have the potential to improve the control of severe asthma associated with ABPA and to reduce or discontinue systemic corticosteroids in corticosteroid-dependent patients. However, data on mucus plug removal, an important outcome of acute ABPA, are insufficient.

Several issues remain unresolved in the treatment of ABPA with omalizumab, the first of which is the dosage of omalizumab administered to patients with ABPA and very high serum IgE levels. The dose of omalizumab must be adjusted according to serum IgE levels and body weight to completely neutralize free IgE. However, a substantial proportion of patients with acute ABPA require omalizumab at doses that exceed the upper limit of the current dosing tables. Data from the basophil activation test suggested that omalizumab at the maximal dose (300-375 mg/2 weeks) could reduce the amount of IgE and IgE receptors on the cellular surface, even in patients with total IgE levels ＞1,500 IU/mL (37), suggesting that the antibody may be clinically effective even at suboptimal levels. The second unresolved question is whether or not the dose of omalizumab can be reduced after the disease stabilizes and IgE production decreases. In circumstances of low serum IgE levels, it may be rational to reduce the dose of omalizumab or extend its dosing interval; however, evidence to support this strategy is scarce.

Mepolizumab/Benralizumab

Evidence supporting the clinical usefulness of mepolizumab/benralizumab in targeting eosinophils via neutralization of IL-5 or IL-5 receptors is limited compared to that of omalizumab. Four case series examining more than nine cases treated with these biological agents have been reported (39-42). Similar to omalizumab, mepolizumab/benralizumab reduced the rate of asthma exacerbation and decreased the required dose of oral corticosteroids; however, their impact on the pulmonary function improvement was inconsistent (Table 4).

A unique feature of this class of biologics is their effect on mucus plugs in the bronchi. We recruited 29 patients with ABPA treated with mepolizumab/benralizumab (median age, 63 years old) from 15 institutes in Japan, who participated in a survey on ABPA/ABPM (42). Mucus plugs in the bronchi were present in 22 patients (76%): 17 were treated with mepolizumab and 5 with benralizumab. Radiographic improvement was observed in 13 patients (76%) treated with mepolizumab and in all 5 patients (100%) treated with benralizumab. Six patients with residual mucus plugs despite mepolizumab treatment were switched to benralizumab treatment; the mucus plugs disappeared in 4 cases (67%) after switching from mepolizumab to benralizumab. These results suggest that benralizumab, which eliminates peripheral blood eosinophils with antibody-dependent cellular cytotoxicity, is more effective than mepolizumab in removing mucus plugs from the bronchi.

Those who showed radiographic responses to mepolizumab treatment exhibited almost complete depletion of peripheral blood eosinophils (median eosinophil reduction rate of 98%), whereas patients who failed to respond to treatment demonstrated incomplete depletion of peripheral blood eosinophils (median eosinophil reduction rate of 82%, p ＜0.001) (42). In contrast, the administration of mepolizumab/benralizumab does not necessarily decrease serum IgE levels, which are essential biomarkers of ABPA disease activity (42). Therefore, active turnover of mucus plugs through the continuous recruitment of eosinophils from the systemic circulation is necessary to maintain the pathology independent of IgE.

Dupilumab/Tezepelumab

Data on the effects of dupilumab, a neutralizing antibody against the IL-4 receptor alpha chain, and tezepelumab, an antibody that neutralizes thymic stromal lymphopoietin, on ABPA/ABPM are limited to case reports. However, it should be noted that some patients refractory to omalizumab or mepolizumab/benralizumab treatment have been successfully treated with dupilumab or tezepelumab (43-49). Furthermore, radiological improvement, including the elimination of mucus plugs, has also been reported after treatment with these biologics (44-46,48-54).

A pharmaceutical company conducted a clinical trial (the LIBERTY ABPA AIRED study) to evaluate the efficacy and safety of dupilumab in patients with ABPA and severe asthma. The LIBERTY ABPA AIRED study was completed in February 2024, and the results are awaited.

IgE and Eosinophils in the Pathophysiology of ABPA/ABPM

As discussed in the previous sections, eosinophils represent the major effector cells in the clinical characteristics of ABPA/ABPM, especially in mucus plug formation. In the last decade, the role of eosinophils in the formation of highly tenacious mucus plugs, which are characteristic of adult-onset eosinophilic airway diseases such as ABPA/ABPM and eosinophilic CRS, has been elucidated (55).

Eosinophils in contact with A. fumigatus through adhesion molecule CD11b release filamentous chromatin as extracellular traps (56). In the process of extracellular trap release, the protein arginine deiminase 4 citrullinates histones and changes the structure of chromatin (57), the cytoplasmic protein galectin-10 crystallizes to form Charcot-Leyden crystals (58), and eosinophils undergo active cell death (ETosis) (59). Although neutrophils and eosinophils produce extracellular traps, the biophysical properties of extracellular traps released from these cells are quite different. Miyabe et al. (60) stimulated peripheral blood neutrophils and eosinophils in vitro and induced aggregation of extracellular traps and cell debris using shear flow. Eosinophil-derived aggregates resemble the mucus plugs in the bronchi of ABPA/ABPM in terms of color (brown) and viscosity (comparable to peanut butter) and demonstrate higher hydrophobicity, dry weight, and CT density than neutrophil-derived aggregates. There are several other mechanisms by which eosinophils induce the formation of tenacious mucus plugs. Charcot-Leyden crystals promote mucus plug formation (61), eosinophil peroxidase facilitates the polymerization of Muc5AC by crosslinking the thiol groups of cysteine residues (62), and tissue factors expressed on the cellular surface of cytokine-stimulated eosinophils activate the coagulation cascade to fibrin nets (63).

However, whether IgE itself is essential for the pathogenesis of ABPA/ABPM or whether it is merely a biomarker of activated IL-4/IL-13 signals, which are indispensable for mucus production in bronchial epithelial cells, is not yet clear.

Conclusions

Some researchers have questioned whether ABPA/ABPM is an independent disease entity different from fungus-sensitized asthma and have proposed the term “allergic fungal airway disease,” which covers both ABPA/ABPM and fungus-sensitized asthma (64,65). However, the presence of mucus plugs in the central bronchi is a unique feature of ABPA/ABPM that has a significant impact on prognosis. Analyses of the clinical characteristics and treatment responses to biologics, together with recent discoveries in eosinophil biology, have identified that eosinophils play a major role in mucus plug formation in patients with ABPA/ABPM. Therefore, it is rational to target eosinophils during treatment (Fig. 2). However, the allergic component of the disease is also an important modifier of the disease prognosis. In patients who exhibit a high allergic predisposition, it is necessary to target both the eosinophilic and allergic pathways for successful disease management.

Author’s disclosure of potential Conflicts of Interest (COI).

Koichiro Asano: Lecture fee, AstraZeneca, Nippon Boehringer Ingelheim and Sanofi.

Financial Support

This study was partially supported by a research grant on allergic diseases and immunology from the Japan Agency for Medical Research and Development (Grant Number JP22ek0410098).