Clinical relevance of Aspergillus fumigatus sensitization in cystic fibrosis

The clinical relevance of sensitization to Aspergillus (A) fumigatus in cystic fibrosis (CF) is unclear. Some researchers propose that specific A fumigatus IgE is an innocent bystander, whereas others describe it as the major cause of TH‐2‐driven asthma‐like disease.


| INTRODUC TI ON
Aspergillus (A) fumigatus is a ubiquitous fungal organism commonly found in house dust, water damaged walls or ceilings, and decomposing organic material. Its conidia are approximately 2-3.5 µm in diameter, which allows deposition in terminal airways and alveoli. 1 Several hundred A fumigatus conidia are inhaled by the human lung every day. 1 The host interaction with A fumigatus in cystic fibrosis (CF) lung disease is diverse in several aspects. 1-3 (a) The fungus chronically colonizes the CF airways, rarely triggering systemic, or invasive infections. (b) While 60% of adult patients are sensitized to A fumigatus with the presence of specific IgE and IgG, a subset of these adults mounts a robust allergic response with a high total IgE and substantially increased eosinophils. 4 (c) Only a small subset of those patients develops significant pulmonary symptoms, the full picture of bronchial pulmonary aspergillosis (ABPA). [3][4][5] The underlying mechanisms seem to be complex, and they are modulated by a variety of factors, including the cystic fibrosis transmembrane conductance regulator (CFTR), non-CFTR genetic immune host susceptibility and gene modifiers, patient's age and sex, atopy, microbial interactions, medication. [1][2][3] Early diagnosis of these different disease entities is critical in order to evaluate individual risk profiles and future treatment options, for example itraconazole, corticosteroids, and/or Vitamin D3. To date, the role and clinical relevance of specific IgE against A fumigatus are unclear. Some authors suggest that IgE is an innocent bystander, whereas others showed that neutralizing IgE with the monoclonal antibody omalizumab is effective in ABPA. 2,6,7 It can be hypothesized that the production of IgE and associated Th2-driven inflammation (IL-5, IL-4, IL-13) induce the transition from sensitization to Aspergillus fumigatus to an asthma-like condition leading to endstage ABPA. 1,2,6 To test this hypothesis, patients with sensitization to A fumigatus were analysed by lung function and airway inflammation before and after bronchial allergen provocation (BAP) with lyophilized A fumigatus antigen. BAP is the gold standard to demonstrate the bronchial relevance of an allergen. Earlier studies from our working group showed that BAP is a safe and highly reproducible method in children and adults suffering from respiratory diseases. [3][4][5][6] Although great effort has been put into standardizing the procedure, BAP is still rarely used by clinicians due to possible side-effects and the fear of severe asthmatic reactions. [8][9][10][11][12][13] For safety reasons, only patients with mild CF (FEV1 > 75%) and proven sensitization to A fumigatus underwent BAP.

| Patients
Patients were recruited from the Division of Pediatric Pulmonology, Allergy and Cystic fibrosis, Goethe University, Frankfurt, Germany. The population of this study consisted of 35 clinically stable patients with CF (6 were P aeruginosa-infected), aged 4-41 years (average 14 years), and 20 non-smoking healthy control subjects. The CF patients were allocated to one of two groups: group 1 (n = 18): negative for sIgE against A fumigatus; and group 2 (n = 17): sIgE positive ≥ 0.7 KU/L, ≥ class 2.
Before inclusion in the study, a detailed verbal and written explanation took place with all patients. The course of the study, the goals, and the risks were discussed in detail with patients and/or their legal guardians. Prior to the start of the study, the subjects and/or their legal guardians signed the consent form. BAP is time-consuming for the patients-at least 9 hours-and we therefore deemed it inconvenient and not ethical to perform BAP in sIgE-negative patients when we designed the study protocol.
At visit 3, 24 hours after BAP, a physical examination, measurement of exhaled NO, lung function testing, methacholine testing, and sputum induction was performed.

| Measurement of exhaled NO
Exhaled NO (eNO) was measured using the NIOX1 (Aerocrine, Solna, Sweden) according to American Thoracic Society guidelines. 14

| Pulmonary function test
Pulmonary function tests were performed according to the recommendations of the American Thoracic Society and the European Respiratory Society. 15 The following measurements were obtained: FVC, FEV1, FEV1/VC, maximum FEV1 fall in early asthmatic reaction (EAR) and late asthmatic reaction (LAR), and PD20 FEV1-A fumigatus.

| Methacholine test
The methacholine test was performed using the Aerosol Provocation System (APS) (VIASYS Healthcare GmbH, Höchberg, Germany), as described previously in detail by our group. 16 During tidal breathing, the system determined the exact administered dose of methacholine automatically. Methacholine with a concentration of 16 mg/ mL was inhaled in five steps: 0.01, 0.1, 0.4, 0.8, and 1.6 mg. The individual cumulative provocation dose (PD) causing a 20% drop in FEV1 (PD20FEV1) was calculated by logarithmic interpolation using an integrated programme.

| Bronchial allergen provocation
All patients were instructed not to take any medication for at least 24 hours before each BAP test. The BAP was performed in the morning, between 8:00 am and 12:30 pm using the flow-  (SBU/mL). All extracts from Allergopharma GmbH&Co.KG were standardized by in-house ELISA and by prick testing. The A fumigatus allergen was a non-modified native extract, completely free of endotoxins, and licensed in Germany for BAP, but no information on major allergen contents is available.
After the baseline FEV1 was determined, the patient inhaled a 0.9% saline solution without allergen. Two minutes later, the FEV1 was measured. In case of a decrease of more than 10%, the BAP test was postponed. Otherwise, the BAP test was conducted by stepwise inhalation of increasing amounts of the standardized allergen solution. Ten minutes after each step, spirometry was performed.
The first step covered an inhalation of 5 SBU/mL. Afterwards, the dose was doubled until a decrease of FEV1 ≥ 20% was reached or a cumulative dose of 635 SBU/mL was administered. [8][9][10] At the end of each test, every patient inhaled 2 puffs of salbutamol (200 µg) to improve the FEV1 to at least 80% of the baseline value. To detect late asthmatic response (LAR), FEV1 was measured every hour with the asthma monitor AM1® (VIASYS Healthcare GmbH) for up to 12 hours after BAP. Symptoms were recorded by the patients or their parents. The LAR was defined as a maximum FEV1 fall ≥ 15%. Patients were instructed to use salbutamol as rescue medication. 24 hours after BAP sputum was collected, patients first performed three baseline lung function tests according to ERS guidelines.

| Sputum collection and sputum cells
Afterwards, they inhaled 400 µg salbutamol, and 20 minutes after administration, three more lung function tests were performed.
Consecutively, nebulized hypertonic saline was administered at concentrations of 3%, 4%, and 5% every 7 minutes as described previously. [17][18][19] After each inhalation of the saline concentra-  Table S2. The amount of IL-5, IL-8, IL-13, INF-γ, T-bet, GATA-3, and FoxP3 mRNA expression was normalized with endogenous control GAPDH (∆Ct values), and the relative quantification and calculation of the range of confidence was performed using the comparative threshold cycle (2 −∆∆Ct ) method (relative gene expression) as previously described. 24 All amplifications were carried out at least in duplicate.

| IgE Measurements
Total IgE and sIgE to A fumigatus were routinely determined according to the manufacturer´s instructions in our laboratory by chemiluminescence immunoassay (IMMULITE, Siemens Healthcare, Erlangen, Germany). 25

| Statistical analysis
Data were analysed using the statistical program GraphPad Prism (version 5) and Microsoft Excel. The Mann-Whitney test was used for unpaired comparisons between patients (sIgE-negative vs. sIgE-positive patients) and controls. Paired samples (data before and after bronchial provocation) were analysed using the Wilcoxon rank-sum test. For non-normally distributed samples, the corresponding non-parametric tests were used. Spearman rank correlation was performed to test the relationship between the cumulative allergen dose and the specific IgE levels. Statistically significant differences were defined as P values *P < .05, **P < .01, and ***P < .001.

| Characteristics of the study population
The population of this study consisted of 35 mild CF patients with and without sIgE to A fumigatus and 20 non-smoking healthy control subjects (see Table 1). It is important to note that sputum induction was not successful in all CF patients. Induced sputum was obtained from 13 to 18 (72.2%) patients without sensitization and 10 to 17 patients (58.8%) in the group with sensitization. The clinical characteristics of patients with and without successful sputum were comparable, as shown in Table S1. A significant decrease in FVC and FEV1 became noticeable compared with healthy control subjects (Table 1).

| Total and specific IgE levels
As shown in Table 1, there were significant differences for total IgE and sIgE between non-sensitized and sensitized patients. Note: For all parameters, mean ± SD are shown; significant differences were found between all patients and controls for FVC and FEV1 (*P < .05) and for total IgE and sIgE between the sIgE-negative and sIgE-positive patient group (**P < .01).

| Bronchial Allergen Provocation leads to asthma-like symptoms
At visit 2, 13 from 17 (76.5%) of the sIgE-positive patients underwent BAP with A fumigatus since their lung function was above FEV1 > 75% according to the ATS guidelines. 26 Three patients showed an early asthmatic reaction (EAR), three patients had an EAR and a LAR, and two exhibited LAR only. Five patients showed no response. However, one patient who was considered "negative" due to the pre-defined cut-off level of 20% had a drop in the FEV1 of 19%. Interestingly, there was a significant correlation between sIgE quantification to cumulative BAP dose sensitivity (r = −.579; P < .05) in the Spearman correlation test, indicating that patients with high levels of sIgE are more likely to have a positive BAP than those with lower sIgE levels, as shown in Figure 1.
All patients with an EAR showed mild symptoms like cough and wheezing and were advised to use salbutamol. In addition, four of the five patients with an LAR used salbutamol for symptom relief.
However, two of these patients exhibited a severe LAR with clinical symptoms such as shortness of breath, chest tightness, and cough.
These two patients had a drop in FEV1 of 48% and 54%, respectively. Both were treated with salbutamol and oral steroids (prednisolone 50 mg). Lung function recovered completely within 24 hours, but patients showed a slight irritant cough for 3-4 days.

| Exhaled NO in sensitized and non-sensitized CF Patients
Baseline Exhaled NO (eNO) levels between sIgE-negative and sIgEpositive patients were invariant (

| Analysis of the inflammatory cell distribution in sputum
We analysed the inflammatory cell distribution in patient groups and healthy controls (

| Expression of mRNAs
The mRNA expression of the pro-allergic TH-2 key cytokines IL-5 and IL-13, the TH-1 key cytokine IFN-γ, the pro-inflammatory chemotactic factor IL-8, and the TH-1 and TH-2 master transcription F I G U R E 1 Correlation of sIgE A fumigatus to cumulative allergen dose. The Y-axis represents the cumulative allergen dose (SBU/ml), and the x-axis represents sIgE to Aspergillus (K/U/L). There was a significant correlation between sIgE and allergen doses of the BAP (r = −0.579, P < .05) Statistics: Spearman test Note: At least 400 sputum cells were counted for each specimen. Macrophages, neutrophils, eosinophils, and lymphocytes were expressed as percentages of the total cell count. Mean ± SD are shown. A significant increase of neutrophils in CF patients and controls was found (*P < .05). 24 h after BAP, a significant increase of eosinophils ( §,* P < .05) was found.

TA B L E 2 Sputum cell counts in induced sputum
factors T-bet and GATA-3, as well as the master transcription factor of regulatory T cells FoxP3, were differentially regulated in sputum derived from CF patients compared with healthy controls ( Table 3).
As described before, the expression of the pro-inflammatory cy- fold expression), GATA-3 (4.9-fold expression), and FoxP3 (2.9-fold expression) were significantly elevated compared with healthy controls. However, no differences were found for the TH-2 related cytokines IL-5 (1.9-fold expression) and IL-13 (1.7-fold expression; see Table 3). At baseline, no significant differences could be seen within the groups of sIgE-positive and sIgE-negative patients. In contrast to baseline findings, expression of TH-2 related cytokines IL-5 and IL-13, the TH-2 master transcription factor GATA-3, and the homeostatic transcription factor FoxP3 were distinctly elevated, indicating an asthma-like inflammation in the airway 24 hours after BAP with A fumigatus (Figure 2).

| D ISCUSS I ON
The lung disease cystic fibrosis is characterized by perpetuating inflammation, recurrent infection, and mucus hypersecretion. 17,18,27 Among the major cell types within CF airways, after neutrophils both B and T lymphocytes are found in large numbers beneath the surface epithelium. 28,29 In infants with CF, exaggerated production of pro-inflammatory cytokines was observed in lower airways. 17  FoxP3. The significant increase of IL-5, IL-13, transcription factors, and eosinophils as well as the inverse decrease of IFN-γ and T-bet described in this study were also observed in previous publications of segmental allergen challenges in asthmatics and CF mouse models. 36  However, this study has some limitations. The number of CF patients who underwent BAP was very small. There were several reasons for this. According to ATS guidelines, only patients with a lung function > 75% could be challenged, because the risk of severe side-effects has to be considered. In addition, even in our CF community, there was great fear of severe side-effects. This was the major reason why some patients did not give their consent for the study and why our sample of patients with BAP was relatively small.

TA B L E 3 mRNA expression changes in sputum cells
The study could also be criticized due to the fact that patients without sIgE were not challenged as a control group to rule out nonspecific toxicity or even contaminants in the A. fumigatus extract.
BAP is a long-lasting procedure and patients have to be monitored carefully for 9 hours; therefore, we found it intrusive and not ethical to subject sIgE-negative patients to BAP. In addition, the A fumigatus used in the study was endotoxin free, and no pro-inflammatory signals, such as an increase of IL-8 or T-bet, were detected in sputum after BAP.
It would have been nice to be able to measure Aspergillus IgG or its relevant recombinant epitopes. However, this was not done, since that was assay is not available at our university, and sending it to an external laboratory was cost-prohibitive without a sponsor or other source of funding.
In conclusion, BAP with A fumigatus caused both a significant decrease in FEV1 in the majority of sIgE-positive patients as well as asthma-like symptoms with increased salbutamol use. A marked TH-2 mediated inflammation involving eosinophils, IL-5, IL-13, FoxP3, and eNO was demonstrated. Current clinical practice should be aware of the possible clinical relevance of A fumigatus-induced asthma-like disease.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data from this study are available on request from the