Cystic fibrosis (CF) lung disease involves impaired mucociliary clearance leading to pulmonary inflammation, infection, and a more rapidly declining lung function. The CF airway milieu contains higher levels of pro-inflammatory cytokines in comparison to healthy controls (Sagel, 2009). Whilst the association between pathogenic bacteria and airway inflammation, infection and disease progression is well established (Hauser et al., 2011), less is known about the role of fungi in the CF airways and their impact on airway inflammation and lung disease progression. Allergic bronchopulmonary aspergillosis (ABPA) is a well-established complication in CF associated with sensitization to Aspergillus fumigatus (Thia and Balfour Lynn, 2009). In the absence of ABPA, it is not clear whether the detection of fungi in the CF airway in asymptomatic patients reflects simple colonization (innocent bystander effect) or infection contributing to increased airway inflammation (Liu et al., 2013; Chotirmall and McElvaney, 2014).

A. fumigatus is the commonest filamentous fungus isolated from CF airways (Pihet et al., 2009; Sudfeld et al., 2009). Improved mycological methods enable the detection of other fungi, including Aspergillus terreus, Scedosporium apiospermum complex, Exophiala dermatitidis, and Penicillium species (Borman et al., 2010).

The clinical significance of A. fumigatus colonization in CF is uncertain. Whilst Saunders et al. (2016) report a steeper decline in Forced Expiratory Volume in 1 second (FEV1) in their fungal culture positive pediatric CF cohort, studies by Baxter et al. (2013b) and De Vrankrijker et al. (2011) found no association between A. fumigatus airway colonization and lung function. Other studies have reported greater risk of pulmonary exacerbations requiring hospitalization (Amin, 2010) and higher rates of bronchiectasis (McMahon et al., 2012).

The lack of standardization in processing respiratory samples for fungal culture, differing fungal recovery rates between standard culture methods and the insensitivity of certain fungal culture methods (Chen et al., 2017) is likely to contribute to these contradictory findings. The use of modified culture methods using higher amounts of sputum inoculum have shown increased sensitivity in the detection of fungi (Pashley et al., 2012; Fraczek et al., 2014).

Exploring the relationship between FC positivity and inflammatory airway biomarkers may help in clarifying the clinical significance of fungi. Attempts at profiling the inflammatory cytokine profile in conjunction with clinical and microbiological profiles in CF is limited to a few well-established biomarkers (Frayman et al., 2017) such as neutrophil elastase (NE) and interleukin-8 (IL-8) found in higher concentrations in the CF airway (Frayman et al., 2017). There is a need to identify biomarkers that are specific for pulmonary infections including fungal infection. There is emerging evidence for the novel biomarkers galactomannan and tumor necrosis factor receptor 2 (TNF-R2) for monitoring non-invasive fungal lung disease in CF and asthma (Baxter et al., 2013a; Ghebre et al., 2017)

Neutrophil-driven airway inflammatory responses play a central role in CF. Activation of the innate immune system following pathogenic exposure results in phagocytosis, release of NE and cytokines (Hartl et al., 2012; Lammertyn et al., 2017). IL-1β and TNF-α facilitate recruitment of immune cells to the site of inflammation. Recruitment cytokines such as IL-8 have high neutrophil attracting capacity (Strieter, 2002), further exacerbating the inflammatory process and resulting in large amounts of neutrophilic intracellular protease and oxidant being released. Activation cytokines (interferon-γ, IL-2, IL-12, and TNF-α) augments the T-helper cell response.

Inhalation of fungal spores may result in germination and hyphal growth within the bronchioles (Chotirmall and McElvaney, 2014). Fungi may subsequently colonize, infect, or induce a spectrum of hypersensitivity reactions within the airways. Galactomannan, a fungal cell wall component, is released during hyphal tissue invasion by Aspergillus species and can be detected in body fluids (Mennink-Kersten et al., 2004). Current literature focusses on the role of galactomannan in invasive pulmonary aspergillosis, but only a few studies to date have aimed to evaluate its role in non-invasive fungal lung disease (Nguyen et al., 2007; Baxter et al., 2013a).

Differentiating between filamentous fungi as innocent bystanders and fungal infection, using a combination of clinical parameters and airway biomarkers, may inform antimicrobial strategies. This is particularly important given emerging evidence that CF transmembrane conductance regulator (CFTR) modulators may not alter susceptibility to A. fumigatus (Fritsch et al., 2019). Consequently the role of fungi in CF is likely to present an ongoing clinical challenge, whilst at the same time limiting our therapeutic options due to complex drug interactions (Tracy and Moss, 2018).

We report a prospective observational cohort study. Our principal aim was to describe the clinical characteristics and airway inflammatory biomarker profile (NE, IL-8, TNF-R2 and galactomannan) in filamentous fungi culture-positive CF patients. We hypothesized that fungal culture positivity in CF is associated with worse lung function and increased airway inflammation.

Our study is the first to describe CF airway biomarkers in relation to fungal culture positivity. We report the relationship between airway fungal culture positivity, airway biomarkers and clinical, radiological and microbiological characteristics in children and adults with CF.

The clinical relevance of fungal airway colonization and its impact on lung disease progression in CF is unclear. In keeping with previous reports, we found that A. fumigatus was the most frequently isolated fungal species (Pihet et al., 2009). Colonization with A. fumigatus has been associated with more advanced CF lung disease (Ramsey et al., 2014), lower lung function (Amin, 2010; Saunders et al., 2016) and greater risk of pulmonary exacerbations requiring hospitalization (Amin, 2010). In our study, the fungal culture positive group, with the exception of the over 30-year old sub-group, had a lower mean FEV1 across all younger age groups. Our findings of a higher FEV1 within the FC positive subgroup for the over 30-year age group, may be a reflection of participants surviving with milder CF lung disease.

LCI derived from multiple breath washout is a more sensitive marker of lung disease progression compared to spirometry, allowing non-invasive monitoring of small airways disease. A study by Walicka-Serzysko et al. demonstrated a higher LCI in CF patients with A. fumigatus (Walicka-Serzysko et al., 2019). We also found higher LCI’s in fungal culture positive patient but this did not reach statistical significance.

The primary focus of previous studies have included exploring the association between A. fumigatus and clinical outcomes, there is a paucity of information exploring fungal airway colonization and its impact on markers of airway inflammation. It is well established that the CF airway milieu contains higher levels of pro-inflammatory cytokines in comparison to healthy controls (Sagel, 2009). Evidence for pulmonary inflammation in surveillance BAL such as raised NE and IL-8 in the CF airways is present, even in the absence of clinically apparent lung disease (Armstrong et al., 2005; Frayman et al., 2017; Khan et al., 2019). Sly et al. have reported a link between high NE levels within the CF airways and the subsequent development of bronchiectasis (Sly et al., 2009). Whether fungal colonization exacerbates underlying airway inflammation, resulting in higher levels of airway biomarkers and therefore further driving lung disease progression requires further study. Exploring this relationship is complicated by the difficulty in separating the effect of bacteria and fungi present in respiratory cultures.

In our study, only one-third of patients were fungal culture positive without the presence of concomitant pathogenic bacteria, making the direct attribution of inflammation to a positive fungal culture problematic. We found no significant difference between NE and IL-8 between the fungal culture positive and negative groups. Two patients with very high levels of NE were fungal culture positive but these low numbers preclude any firm conclusion. Studies have shown improvements in NE and IL-8 levels following treatment of an exacerbation (Ordonez et al., 2003; Colombo et al., 2005). Evidence suggests NE activity is higher during exacerbation and is responsive to antibiotics (Chalmers et al., 2017). Whilst we acknowledge our small number of paired patient samples due to difficulty in obtaining sufficient sputum volumes, we report higher levels of NE and IL-8 during pulmonary exacerbation compared to the stable state.

Galactomannan may be a useful screening test for invasive pulmonary aspergillosis (Kimura et al., 2009; Talento et al., 2017). The diagnostic usefulness of galactomannan assays in patients with non-invasive fungal lung disease in CF, such as ABPA and chronic pulmonary aspergillosis remains uncertain (Nguyen et al., 2007; Kitasato et al., 2009; Fayemiwo et al., 2017). We found that only four patients (13%) within our cohort had a positive result for galactomannan. There were no significant difference in the absolute values of galactomannan between the fungal culture positive and negative groups. Contrary to our study findings, galactomannan has been reported as a sensitive method of detecting Aspergillus in sputum in comparison to fungal cultures (Baxter et al., 2013a), even in the absence of invasive disease. Baxter et al. used sputum samples homogenized with Sputasol (commercial product containing DTT) and sonication. In comparison we used homogenized sputum supernatants without sonication, which may have decreased the overall yield.

There is emerging evidence implicating TNF-R2 as a possible airway biomarker in fungal associated asthma. Ghebre et al. noted significantly higher TNF-R2 levels in adults with asthma who were fungal culture positive (Ghebre et al., 2017). Our study is the first to analyze TNF-R2 as a potential biomarker of inflammation in adults and children with cystic fibrosis. Overall, we detected TNF-R2 in only 19% of our patients. Within the limitations of our modest number of patients with detectable TNF-R2 levels, we found no association with adverse clinical outcomes. Whilst Ghebre et al. concluded that TNF-R2 may diminish the inflammatory response to A. fumigatus, the exact mechanism for this and its clinical implications in CF is unknown. Interestingly, Ghebre et al. concluded that TNF-R2 appears to be a strong discriminator of exacerbation in children and adults with asthma (Ghebre et al., 2019). Further studies in CF are required to determine if a similar relationship exists between fungal culture positivity, exacerbation states and higher TNF-R2 levels.

We found that the fungal culture positive subgroup had greater previous and/or current positive sputum isolates for non-tuberculous mycobacteria. This is in keeping with previous literature where it has been proposed that Mycobacterium superinfection can occur alongside aspergillosis (Cui et al., 2013). The proposed theories include a synergistic relationship which enhances their resistance to environmental change such as use of broad-spectrum antibiotics (Kerr, 1994; Cui et al., 2013).

Within our CF cohort, we report higher rates of fungal culture positivity with indoor mold exposure. Evidence suggests a link between indoor fungal contamination and ABPA. Reports linking airway fungal colonization and mold exposure in CF are scarce (Rocchi et al., 2015). Fairs et al. reported environmental mold exposure may predispose asthmatics to airway colonization (Fairs et al., 2013) and whether this also occurs in CF patients is unclear and requires further study.

Although not statistically significant we observed a higher rate of bronchiectasis in the fungal culture positive cohort with 76% having radiological evidence of this on CT chest, in comparison to 64% of the culture negative cohort. Similar associations have been reported between A. fumigatus and higher radiological scores with more significant bronchiectasis (McMahon et al., 2012).

Whilst over 35% of our fungal culture positive cohort received three-monthly intravenous antibiotics, only 16% of fungal culture negative group were exposed to regular antibiotics. Several studies have shown an association between the use of broad-spectrum antimicrobials and A. fumigatus airway colonization (Burns et al., 1998; Hodson et al., 2002). The use of regular intravenous antibiotics are typically reserved for patients with more severe lung disease and/or bacterial colonization. The concept of regular exposure to broad spectrum antimicrobials reducing competition for surfaces and limited nutrients, therefore promoting fungal adhesion, growth and colonization has therefore been proposed (Oever and Netea, 2014). However, a study by Baxter et al. reported a significant decline in the bioburden of Aspergillus when they compared pre- and post-antibiotic sputum samples from 26 adult CF patients (Baxter et al., 2013b). Therefore, the use of antimicrobials for well-established CF pathogens and its impact on A. fumigatus remains poorly understood.

We report a significantly higher rate of inhaled corticosteroid use within the FC positive cohort. This is a well-established finding in existing literature whereby the use of inhaled and/or systemic corticosteroids are known to promote A. fumigatus growth (Noni et al., 2015).

Our study limitations include a modest sample size, in particular for the paired sample analysis. The presence of significant numbers of non-sputum producing children in our study cohort means that we were unable to obtain sufficient volumes to complete ELISA biomarker and sputum cell differential analysis for all patients. We included a combination of sputum and BAL samples to enhance our sample size and to allow inclusion of non-sputum producing children within our study cohort. In order to minimize bias resulting from salivary contamination of respiratory samples, we included BAL samples when clinically indicated and where we were unable to obtain spontaneous or induced sputum samples.

Secondly, published studies report a wide range of prevalence in the detection of filamentous fungi (Nagano et al., 2010; Pashley et al., 2012) due to sampling techniques and culture conditions, it is likely that the true prevalence of fungi in CF is underestimated. To overcome this we used a modified fungal culture approach, which has been shown to be a more sensitive for the detection of filamentous fungi (Pashley et al., 2012). Whilst it is possible that our approach is detecting low levels of fungi that are not truly colonizing the lower airways, the association between higher culture rates and worse lung function in both this and the adult asthma studies (Fairs et al., 2010; Agbetile et al., 2012) that have utilized this more sensitive culture method would suggest the detection may be clinically relevant. An alternative approach to culture may be to utilize a molecular technique such as quantitative polymerase chain reaction (qPCR), which is commercially available for A. fumigatus and a limited number of other Aspergillus species (Guegan et al., 2018). Both modified fungal culture and Aspergillus quantitative PCR have been shown to be more sensitive than the NHS routine approach (Public Health England, 2019) for identifying A. fumigatus from respiratory secretions (Fraczek et al., 2014). It should be noted, however, that the current commercial qPCR assays, whilst based on quantitative technology, are used for determining the presence or absence of a fungus only, and will amplify DNA from dead as well as live fungal material.

Our study reports on the short-term clinical outcomes associated with FC positivity in CF given the cross-sectional methodology used. Whilst the long-term outcomes of ABPA and serological sensitization are well-described with evidence of lung function decline (Kraemer et al., 2006), increased need for intravenous antibiotics (Peetermans et al., 2014) and bronchiectasis (Agarwal et al., 2010), literature on the outcomes of Aspergillus colonization and non-invasive infection is scarce. It is clear that further prospective longitudinal studies are required to describe the clinical outcomes of A. fumigatus colonization and non-invasive infection.

Finally, statistical analysis for LCI and bronchiectasis were excluded from the binomial regression model due to the risk of bias from missing data. A limited number of patients had LCI results available as this is not routine practice in our adult CF cohort and is carried out as part of the annual comprehensive review process in children. However, not all children are capable of completing this procedure. We also note a limited number of patients underwent a CT chest as this was only carried out as part of routine care when indicated. In addition, our results obtained for NTM and mold exposure have high confidence intervals despite clinical significance, reflecting the highly variant estimates derived for these parameters.

In conclusion, we found no difference between the airway biomarker profile between sputum fungal culture positive and negative patients with CF. The role of galactomannan and TNFR2 as more fungal specific airway biomarkers in CF remains uncertain. Fungal culture positivity is associated with a lower FEV1 in patients less than 30 years old, a lower LCI, higher rates of NTM positivity, bronchiectasis and intravenous antibiotic exposure. Larger trials using a broader range of inflammatory markers in conjunction with fungal culture and quantitative PCR measures are needed.

The original contributions presented in the study are included in the article/supplementary materials. Further inquiries can be directed to the corresponding authors.

The studies involving human participants were reviewed and approved by Leicester longitudinal study of respiratory infections and microbiomics in CF by the East Midlands Research Ethics Committee, reference number 12/WM/0285. Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin.

DP, CP, and EG co-designed the study. DP was responsible for analyzing the data and writing the manuscript. CP supervised the laboratory analyses (modified culture and ELISA) and EG the clinical aspects of the study. DP was responsible for patient recruitment, the design and custodianship of the clinical database, and processing clinical samples. DP and KD undertook the ELISA analysis. All authors contributed to the article and approved the submitted version.

Funding from Gilead Sciences Grant Program (grant no. 02261) and University Hospitals of Leicester charitable funds was awarded for this study alongside our separate study termed “Changes in the Cystic Fibrosis Airway Mycobiome Over Time and its Link with Bacterial Species Rarefication”.

DP reports grants from Gilead Sciences Grant Program, outside the submitted work. This grant was awarded for a separate ongoing study termed: Changes in the Cystic Fibrosis Airway Mycobiome Over Time and Its Link with Bacterial Species Rarefication. CP has received research support and honorariums for advisory board from Pulmocide who are developing a new antifungal agent. EG reports from Consultancy work for Boehringer Ingelheim November 2016, from Consultancy work for Anaxsys July 2018, grants from Circassia research grant, grants from Gilead sciences research grant, grants from Chiesi Limited, from Medimmune research collaboration, and outside the submitted work: 1. Consultancy work for Boehringer Ingelheim November 2016. Money, paid to the institution (The University of Leicester); 2. Consultancy work for Anaxsys July 2018, money paid to the institution (The University of Leicester); 3. Circassia research grant, investigator-led research grant; 4. Gilead sciences research grant, investigator-led research grant; and 5. Medimmune research collaboration, research collaboration.

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