Cystic fibrosis (CF) is the most common lethal genetic disease in Caucasians and affects approximately between 70,000 and 100,000 people worldwide. The leading cause of death is the progressive lung disease, driven by a complex and diverse inflammatory immune syndrome induced by recurrent and chronic bacterial and fungal infections of the lung (Stephenson et al., 2015; Bell et al., 2020).

A. fumigatus (Af), the most frequently detected mold fungus in CF airways, is the fungal pathogen with the most impact on morbidity and mortality in CF (Pihet et al., 2009; Amin et al., 2010; Middleton et al., 2013; Schwarz et al., 2018a). The defective chloride channel in CF together with impaired mucociliary clearance and structural changes resulting from infectious complications (e.g. cavitary lung disease, bronchiectasis) result in a local mucosal immunodeficiency (Cohen and Prince, 2012), which allows airway pathogens like Af conidia to colonize and germinate in the airways. Allergic bronchopulmonary aspergillosis (ABPA), bronchitis, aspergilloma and invasive infection (pneumonia) are the known disease entities caused by Af in CF (Moss, 2010; Baxter et al., 2013a; Middleton et al., 2013; Hong et al., 2018).

ABPA is an immune mediated lung disease with a prevalence of up to 15% in CF. It manifests with a poorly controlled obstructive disease and recurrent pulmonary infiltrates with or without bronchiectasis (Stevens, 2003). According to the 2003 CFF (Cystic Fibrosis Foundation) consensus criteria, classic diagnosis of ABPA comprises clinical deterioration not attributed to other causes, serum IgE >1000 IU/ml, a positive skin prick test or positive specific IgE, antibodies against Af, precipitating or IgG antibodies against Af, and radiological changes that are not treatable by antibiotics. For the minimal ABPA criteria, serum IgE >500 IU/ml and fewer of the criteria mentioned above are sufficient to suspect ABPA (Stevens, 2003). Invasive pulmonary Af infections primarily occur in severely immunocompromised patients, but they also can occur in CF if additional risk factors are present (e.g. severe lung disease, diabetes, steroid therapy) (Kousha et al., 2011; Hong et al., 2018). The prevalence is extremely low, thus precise data are not available. In CF, invasive pulmonary Af infections often show a subacute, slowly progressing clinical course. and is induced by local invasion of Af (Schwarz et al., 2018b). Radiographic imaging and clinical courses are often non-specific (Felton and Simmonds, 2014; Kosmidis and Denning, 2015). This bears the risk of delayed diagnosis in CF. Af bronchitis is another Af entity with an estimated prevalence of 9% in CF. Increasing evidence exists that Af in CF can cause airway symptoms without hypersensitivity and without tissue invasion of fungal hyphae (Felton and Simmonds, 2014; Brandt et al., 2018). Shoseyov et al. were the first to report on Af bronchitis in six patients (Shoseyov et al., 2006). The group suggested that bronchitis should be considered, and antifungal therapy initiated when deteriorating respiratory function is not responding to antibacterial therapy, Af is growing in sputum cultures and ABPA was excluded. Chrdle et al. proposed a definition using clinical criteria: symptomatic chronic lower airway disease, detection of Af in sputum or bronchial alveolar lavage (BAL) by culture or PCR, and detection of Aspergillus-specific IgG antibodies (Chrdle et al., 2012). In summary, repeatedly detected Af in sputum samples and persistent respiratory symptoms without signs of Af hypersensitivity and without radiographic evidence of infection are characteristics of Af bronchitis in patients with CF (Felton and Simmonds, 2014; Kosmidis and Denning, 2015). Due to the polymicrobial colonization, overlapping clinical, radiographic, and laboratory characteristics, it is still difficult to clearly distinguish Af conditions from other causes of bronchopulmonary exacerbation in CF (Stevens, 2003). This bears the risk of Af diseases being missed and can lead to situations in which a potentially harmful differential therapeutic strategy must be attempted in order to achieve successful treatment. Therefore, the search for reliable, widely available biomarkers for Af diseases is still an important task today (el-Dahr et al., 1994; Kurup, 2005; Latzin et al., 2008; Hafen GM et al., 2009; Delhaes et al., 2010; Scheffold et al., 2018; Keown et al., 2019). In this light, it seems useful to evaluate single biomarkers in terms of their significance for Af diseases. Af IgG antibodies are used for diagnosis of chronic pulmonary aspergillosis (CPA) and aspergillus diseases in patients with chronic lung disorders and are referred to as diagnostic criteria in many guidelines (Stevens, 2003; Delhaes et al., 2010; Agarwal et al., 2017). They are found in many patients with ABPA, bronchitis and invasive aspergillosis, and several assays are available. In the clinical routine, in recent years, technologies like ImmunoCAP have continuously replaced older methods like immunoprecipitation (Van Hoeyveld et al., 2006; Baxter et al., 2013b). The ImmunoCAP technique provides quantitative measures and reproducible results and therefore a higher sensitivity and specificity in detecting aspergillus conditions (Baxter et al., 2013b). However, there is still a lack of standardization in terms of cut-off values, as several working groups examining Af IgG in Af related conditions obtained discrepant results, which is in part due to different study designs and different laboratory methods used, but this may also reflect the different host-pathogen-interaction of heterogeneous study populations with different underlying diseases (Agarwal et al., 2017; Sehgal et al., 2018; Alghamdi et al., 2019; Sehgal et al., 2019). We therefore evaluated specific Af IgG in ABPA, Af bronchitis and invasive aspergillosis in CF.

This prospective study was conducted between January 2016 and December 2019 at Christiane Herzog Cystic Fibrosis Center, Charité—Universitätsmedizin Berlin, Germany. Ethical aspects were considered and approval for the study was gained by the Ethics Committee of Charité—Universitätsmedizin Berlin, Germany (Number EA2/121/16).

We collected sera from 109 patients with CF, ages 6–69 years, independent of their history of Af related conditions. To obtain both a control group and a disease group, Af-specific IgG was determined either as part of a diagnostic workup when Af conditions were suspected or without such suspicion in pulmonary exacerbated and in clinically stable patients during routine checkup. Af-specific IgG (ImmunoCAP Gm3) was determined in our routine laboratory. Depending on the diagnosis criteria for Af related conditions in CF, the patients were categorized into four groups: control, Af bronchitis, ABPA, and Af pneumonia. As a few patients were assessed several times during the course of the study, some follow-up measurements for Gm3 are available and listed below.

In this study, we evaluated the utility of Af specific Gm3 for the diagnosis of Af associated diseases in CF. Mean Gm3 levels revealed a significant difference between control group vs. ABPA and vs. pneumonia group, but not vs. bronchitis group. Af specific IgG may therefore be a useful biomarker for acute ABPA and Af pneumonia in cystic fibrosis, but not for Af bronchitis in cystic fibrosis.

However, we also found a wide distribution of Gm3 in the control, bronchitis and ABPA group. Thus, for the diagnosis of ABPA and Af pneumonia in CF, Gm3 should only be interpreted alongside other biomarkers.

Interestingly, a significant proportion of the Gm3 of our control group exceeded the manufacturer`s cut-off-level of 39 mg/L. Due to the high exposure of the CF lung to respiratory pathogens, the immune response in CF—here in the form of Gm3—may fundamentally differ from that of healthy individuals. Specific cut-offs may therefore be required in CF to describe pathogen-mediated conditions. Positive and even high titers of Gm3 do not necessarily represent Af-associated disease. On the other hand, the pathogenic impact and the need for treatment of chronic colonization with Af in CF is still under discussion. In our cohort, 33% of the control group patients had Af positive sputum (see Table 1). There could be cases of bronchitis in the control group that did not meet the definition of bronchitis. Another reason for high Gm3 levels in the control group may be cross reactivity to other fungi colonizing the respiratory tract in CF, as m3 is the crude Af antigen and shows cross reactivities to other opportunistic fungi (Fukutomi et al., 2016). In our cohort 85.3% were co-colonized with Candida spp. and 11% with other fungi (see Table 1). If this problem could be overcome by using recombinant antigens, is unclear, because so far, no investigation has been performed under this aspect. The studies conducted with recombinant antigens to date show vastly different results (Nikolaizik et al., 1995; Kurup, 2005; Fricker-Hidalgo et al., 2010; Alghamdi et al., 2019). Finally, in some patients in the cohort of acute Af diseases, Gm3 exceeds the measurement range of 200 mg/L. Quantification above 200 mg/L could therefore improve sensitivity and specificity.

The non-significant difference in median Gm3 levels between control and bronchitis group could be explained by a predominantly endobronchial host-pathogen interaction in Af bronchitis, which may less likely lead to IgG production than in the more invasive forms of aspergillosis. This may also explain why a significant proportion of the bronchitis group had Gm3 levels below the manufacturer`s cut-off. On the other hand, also a small but still significant proportion of the ABPA group had Gm3 levels below the manufacturer`s cut-off, which may be due to a predominant Type 1 hypersensitivity reaction in these cases.

Our results of a Gm3 cut-off-level of 78.05 mg/L with a sensitivity of 75% and a specificity of 74.6% in Af pneumonia and ABPA in CF differ substantially from those found in other populations. Agarwal et al. examined Af specific Gm3 in asthmatic patients with ABPA and Af sensitization without ABPA and found a sensitivity of 88% and specificity of 100% when using a cut-off level of 26.9 mg/L ImmunoCAP Gm3 (Agarwal et al., 2017). In chronic pulmonary aspergillosis in non-CF-patients, Sehgal et al. found a sensitivity and specificity of 95.6% and 100% using a Gm3 cut-off of 27.3 mg/L (Sehgal et al., 2018). For simple aspergilloma in pulmonary tuberculosis patients, the same working group found a sensitivity of 63.5% and a specificity of 98.3% using a cut-off of 27.3 mg/L. However, the Gm3 cut-off-level is still under discussion, as a recent multicentric study found a median Gm3 of 21.3 mg/L and a 90% quantile for Gm3 of 78 mg/L in a population of 121 healthy donors (Raulf et al., 2019). The working group suggested that IgG values higher than at least 90% of the healthy donor group indicate elevated levels, which is more consistent with our findings of a cut-off-level of 78.05 mg/L.

Our follow-up data, where some patients showed a rise in Gm3 at a second point of time after diagnosis are consistent with the data of Agarwal et al. in asthmatic patients with ABPA, who observed an increase of Gm3 in 23.1% of the patients and a decrease in 23.1% (Agarwal et al., 2017). However, immunoglobulins are often maintained long-time following antigen contact, and, so far, no longitudinal data exist for titer progression of Gm3. Considering the small number of cases in our cohort, no statement can be made if and after what time Gm3 is suitable as a follow-up marker. Due to the low prevalence of Af diseases in CF, the suitability of Gm3 as a screening and outcome parameter should be investigated in a multicenter study.

Af-specific IgG may be a useful biomarker for acute ABPA and Af pneumonia, but not for Af bronchitis in CF. However, due to the large interindividual variability of Gm3, it should only be interpreted alongside other biomarkers. Therefore, due to its broad availability, it could be suitable as a biomarker for ABPA and Af pneumonia in CF, if the results can be supported by a larger multicenter cohort. Standardized, longitudinal studies and a better knowledge of underlying immunological mechanisms may improve its usability.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Ethikkommission Charité Universitätsmedizin Berlin. Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin.

PE, CG, and CS contributed to the conception and design of the study. PE and CG organized the database. CG performed the statistical analysis. PE, CG, and CS wrote the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by the German Federal Ministry of Education and Science (BMBF)—Project InfectControl 2020 (“ART4Fun”; 03ZZ0813E).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.