Abstract
Allergic bronchopulmonary aspergillosis (ABPA) is a significant complication in people with cystic fibrosis (pwCF), driven by hypersensitivity to Aspergillus fumigatus. This study aimed to identify factors associated with the development of ABPA in pwCF, using data from the French CF Registry (FCFR). We conducted a multicenter case-control study utilizing anonymized data from the FCFR, spanning the period from 2016 to 2021. A total of 312 ABPA cases were matched to 936 controls. Various clinical factors, including CFTR variants, nutritional status, glucose disorders, respiratory function, chronic bacterial colonization, and treatments such as antibiotics, corticosteroids, and antifungals, were analyzed. Multivariate analyses and logistic regression models were used to identify associations with ABPA. PwCF who received more frequent intravenous antibiotics (OR = 2.47, P = .013), long-term inhaled corticosteroids (OR = 1.82, P < .001), or antifungals (OR = 5.83, P < .0001) exhibited a higher likelihood of developing ABPA. Additionally, glucose disorders were significantly associated with ABPA (OR = 1.41, P = .03). In contrast, a higher body mass index (BMI >25 kg/m²) appeared to be a protective factor (OR = 0.47, P = .03). No significant associations were observed with lung function, CFTR variants, or chronic Pseudomonas aeruginosa colonization. These findings suggest that certain clinical factors and treatments, particularly glucose disorders, frequent antibiotic use, and corticosteroid therapy, are associated with the development of ABPA in pwCF. Notably, a higher BMI may have a protective effect. Further research is needed to explore the underlying mechanisms of these associations and optimize treatment strategies for ABPA in CF, especially as CF therapies continue to evolve.
Keywords: cystic fibrosis, allergic bronchopulmonary aspergillosis, national registry, case-control study
Introduction
Cystic Fibrosis (CF) is an autosomal recessive disease affecting over 160 000 individuals worldwide, with approximately 7700 cases reported in France alone.1,2 It is caused by variants in the CFTR gene, with over 2200 distinct variants identified to date.3 These variants result in abnormal production and/or expression of the CFTR protein, an ion transporter situated on the surface of epithelial cells. The absence of functional CFTR protein leads to dysfunction in multiple organs, most notably the respiratory tract, digestive system, and pancreas.4
In the lungs, the absence of functional CFTR protein leads to airway dehydration, thick mucus production, and impaired ciliary clearance. Consequently, pathogens are more likely to colonize the airways, triggering bronchial inflammation and increasing susceptibility to respiratory infections.5,6 This cascade of events ultimately results in the destruction of airway walls and decline in respiratory function, often leading to terminal respiratory insufficiency.
A particularly concerning pathogen for people with CF (pwCF) is Aspergillus fumigatus (A. fumigatus), the most commonly isolated filamentous fungus in this population, with a prevalence ranging from 10% to 50%.2,7 Moreover, concurrent airway colonization by A. fumigatus and Pseudomonas aeruginosa (P. aeruginosa) is associated with a greater decline in respiratory function and worse overall prognosis.8Aspergillus fumigatus, which is ubiquitous in soil, plants, and organic matter, poses a significant risk due to its conidia's ability to reach the smallest bronchioles.9 A notable complication of A. fumigatus colonization is allergic bronchopulmonary aspergillosis (ABPA), a severe respiratory complication observed primarily in individuals with asthma or CF, with a prevalence ranging from 6% to 25% among the latter.2,7,10,11 ABPA involves a hypersensitivity reaction mediated by a Th2 immune response, characterized by elevated serum total IgE, specific A. fumigatus-IgE levels, and blood eosinophilia, contributing to lung inflammation.12 This immune response results in mucus plugs, airway obstruction, and lung infiltrates visible on imaging. Consequently, compared to those without ABPA, pwCF who develop ABPA are believed to experience a faster decline in lung function, more frequent hospitalizations, and a significant impact on quality of life.13,14
Given the challenges in diagnosing and managing ABPA, it is crucial to identify populations at higher risk of developing this condition to facilitate early detection and prevention. Therefore, this study aimed to investigate factors associated with ABPA in pwCF. Leveraging extensive data from the French CF Registry (FCFR), we conducted a comparative analysis between pwCF diagnosed with ABPA, referred to as ‘ABPA cases’, and those without ABPA, referred to as ‘controls’.
Materials and methods
Patients
This multicenter observational case-control study was conducted using anonymized data from the FCFR. Established in 1992, the FCFR aggregates clinical information from 47 French CF centers, capturing over 95% of the French CF population with minimal loss to follow-up (<3%). The registry annually collects comprehensive data including anthropometric measurements, medical follow-up, prescribed treatments, and respiratory and microbiological data, with the latest published FCFR documenting records of 7 743 pwCF.2
For this study, ABPA cases and controls were selected from FCFR data spanning from 2016 to 2021(inclusive). During this period, French CF centers defined ABPA based on the criteria established by the CF Foundation Consensus Conference in 2003.15 These criteria include: 1) a (sub-)acute clinical deterioration not attributable to another etiology; 2) total serum IgE > 1000 IU/ml in steroid-free patients; 3) positive immediate cutaneous hypersensitivity to Aspergillus antigens or elevated Aspergillus fumigatus-specific IgE levels; 4) precipitating or IgG antibodies against A. fumigatus in serum; and 5) new radiological abnormalities that have not resolved with physiotherapy and antibiotic treatment. ABPA diagnosis is then recorded in the registry as a binary outcome (yes/no). The annual occurrence of ABPA in the FCFR ranged from 9.9% in 2016 to 6.3% in 2021 (Supplementary Table 1).
ABPA cases were defined as those with a diagnosis of ABPA at any point between birth and the most recent age recorded in the registry after 2016. If a patient was diagnosed with ABPA in multiple years, the first diagnosis during the 2016–2021 period was considered for analysis. Subsequent diagnoses were not included to prevent duplication. Controls were selected from pwCF without an ABPA diagnosis during the same period. To minimize bias, cases and controls were matched for age, sex, exocrine pancreatic function status, and year of follow-up, using a propensity score matching with a Greedy algorithm, resulting in a three-to-one ratio of controls to cases.
The study focused on ABPA occurrence, with data from cases analyzed from the year preceding the initial ABPA diagnosis. Data from transplanted patients were included, provided they were collected before the first transplantation. Several factors were examined for their association with ABPA, including CFTR variant, nutritional status (as indicated by body mass index, BMI), glucose disorder (e.g., glucose intolerance or CF-related diabetes), severe CF-liver disease (cirrhosis), respiratory function (percentage of predicted forced expiratory volume in 1 s, ppFEV1), asthma diagnosis, chronic bacterial airway colonization (defined as more than 50% of sputum samples being positive in the last 12 months, with at least four samples during this period), viral respiratory infection, advanced pulmonary disease therapies (e.g., long-term oxygen therapy and noninvasive ventilation, NIV), number of annual intravenous antibiotic courses, and long-term use (e.g., continuous use for 6 months or more in the year preceding ABPA diagnosis) of specific treatments including inhaled antibiotics, mucolytics (e.g., rhDNase or saline), inhaled or oral corticosteroids, azithromycin, antifungal agents, and CFTR modulators (ivacaftor, lumacaftor/ivacaftor, tezacaftor/ivacaftor, and elexacaftor/tezacaftor/ivacaftor).
Statistical analyses
The analyses were carried out on all the available data present in the FCFR, except for BMI and ppFEV1, which were imputed using values from previous years (up to 2 years prior) when missing. This imputation approach helped preserve statistical power and reduce bias. Cases and controls were compared in a descriptive way using Wilcoxon or Student t test for continuous variables, and χ2 or Fisher's exact tests for categorical variables. Subsequently, cases and controls were compared with logistic regression models. First, the univariate analysis was conducted on individual parameters, with selection based on a significance threshold of 25% for the overall effect. Then, the multivariate analysis was performed incorporating the previously selected significant variables and any identified interactions. To ensure robustness, various models were tested to determine the optimal balance among variables with differing codings (e.g., chronic colonization, glucose disorder vs. dichotomous variables such as glucose intolerance and diabetes; respiratory therapy vs. dichotomous variables like oxygen and NIV). Model selection was guided by Akaike's information criterion (AIC), which balances model complexity and goodness of fit.16 AIC is calculated as: AIC = 2k – 2ln(L) where k is the number of estimated parameters and L is the maximum likelihood of the model. Lower AIC values indicate better-fitting models with fewer parameters, reducing the risk of overfitting. The final model retained variables based on their clinical relevance and statistical significance, ensuring the most meaningful interpretation of factors associated with ABPA.16
Ethics
This observational study was conducted using anonymized data from the patient cohort registered within the FCFR after approval by the French Institute for Health Data (approval n° 217, on 12/01/2016) and the French Data Protection Authority (approval n° DE-2018-001, on 03/12/2018). Written informed consent was not required for participation in this study, in accordance with national legislation and institutional requirements.
Results
After selection and matching, the final study population comprised 1248 pwCF, including 312 ABPA cases and 936 controls (without ABPA) (details by year of follow-up in Supplementary Table 2). In both groups, males were predominant (56.1% vs. 43.9% female), with a median age of 17 years.
Descriptive analysis
In the descriptive analysis (summarized in Table 1), ABPA cases exhibited a higher prevalence of glucose disorder (P = .013), asthma (P = .0003), chronic airway colonization, particularly by Pseudomonas aeruginosa (P = .005), and viral infections (P = .04) compared to controls. ABPA cases also received more antibiotic courses (intravenous and inhaled, P < .0001 and P = .0002, respectively), steroids (inhaled and oral, P < .0001 and P = .002, respectively), azithromycin (P = .007), and antifungal (P < .0001) treatments. Lung function, as measured by median ppFEV1, was lower in ABPA cases (P = .002), and they were more often prescribed advanced pulmonary disease treatments (oxygen or NIV, P = .001). Additionally, ABPA cases were less likely to have a high BMI (P = .025). No significant differences were observed regarding the type of CFTR variant or the different CFTR modulators, although few pwCF were on modulators like elexacaftor/tezacaftor/ivacaftor during the study period (Supplementary Table 2), as its approval in France occurred after 2020 (Supplementary Tables 3 and 4).
Table 1.
Descriptive comparison of the ABPA cases and matched controls in people with cystic fibrosis from the French CF Registry.
| Characteristics | ABPA cases n = 312 | Controls n = 936 | P-value |
|---|---|---|---|
| Males, n (%) | 175 (56.1%) | 525 (56.1%) | 1.0000b |
| Insufficient exocrine pancreatic function, n (%) | 296 (94.9%) | 888 (94.9%) | 1.0000b |
| Age* (years), median [IQR] | 17.0 [11–25] | 17.0 [11–25] | .9499a |
| CFTR genotypes, n (%) | |||
| F508del homozygotes | 131 (42) | 421 (45) | .3569b |
| F508del heterozygotes | 129 (41.3) | 369 (39.4) | .5480b |
| Others | 52 (16.7) | 146 (15.6) | .6546b |
| Nutritional status, BMI | |||
| Median [IQR] | 18.8 [16–21] | 19.1 [16.2–21.5] | .1002a |
| Low (<18.5), n (%) | 147 (47.1) | 416 (44.4) | .4702b |
| Normal (18.5–24.9), n (%) | 154 (49.4) | 451 (48.2) | .7191b |
| High (>25), n (%) | 11 (3.5) | 66 (7.1) | .0250 b |
| Glucose disorder | |||
| None, n (%) | 129 (41.3%) | 463 (49.5%) | .0129 b |
| Glucose intolerance, n (%) | 129 (41.3%) | 343 (36.6%) | .1381b |
| CF-related diabetes, n (%) | 54 (17.3%) | 130 (13.9%) | .1402b |
| Severe CF-liver disease (cirrhosis), n (%) | 45 (14.4%) | 125 (13.4%) | .6338b |
| Respiratory status, ppFEV1 | |||
| Median [IQR] | 83.7 [65–95.3] | 87.1 [69.9–100.4] | .0025 a |
| <40%, n (%) | 19 (6.1%) | 44 (4.7%) | .7575b |
| 40%–70%, n (%) | 74 (23.7%) | 176 (18.8%) | .0603b |
| ≥70%, n (%) | 201 (64.4%) | 655 (70%) | .0671b |
| Diagnosis of asthma, n (%) | 157 (50.3%) | 363 (38.8%) | .0003 b |
| Chronic airway colonization, n (%) | 181 (58%) | 473 (50.5%) | .0220 b |
| Pseudomonas aeruginosa, n (%) | 57 (18.3%) | 108 (11.5%) | .0050 b |
| MRSA, n (%) | 8 (2.6%) | 12 (1.3%) | .0529b |
| MSSA, n (%) | 97 (31.1%) | 308 (32.9%) | .9185b |
| Burkholderia species, n (%) | 5 (1.6%) | 10 (1.1%) | .5473c |
| Achromobacter species, n (%) | 7 (2.2%) | 21 (2.2%) | .9136b |
| Stenotrophomonas maltophilia, n (%) | 7 (2.2%) | 14 (1.5%) | .3738b |
| Viral respiratory infection, n (%) | 12 (3.8%) | 17 (1.8%) | .0393 b |
| Therapy for advanced pulmonary disease | |||
| None, n (%) | 292 (93.6%) | 912 (97.4%) | .0014 b |
| Oxygenotherapy, n (%) | 10 (3.2%) | 12 (1.3%) | .0254 b |
| Noninvasive ventilation, n (%) | 8 (2.6%) | 7 (0.7%) | .0167 c |
| Number of annual intravenous antibiotic courses | |||
| 0, n (%) | 193 (61.9%) | 725 (77.5%) | <.0001 b |
| 1–3, n (%) | 99 (31.7%) | 189 (20.2) | <.0001 b |
| >3, n (%) | 20 (6.4%) | 22 (2.4%) | .0006 b |
| Long-term inhaled antibiotics, n (%) | 101 (32.4%) | 204 (21.8%) | .0002 b |
| Long-term inhaled mucolytics, n (%) | 167 (53.5%) | 447 (47.8%) | .0775b |
| Long-term inhaled corticosteroids, n (%) | 142 (45.5%) | 297 (31.7%) | <.0001 b |
| Long-term oral corticosteroids, n (%) | 19 (6.1%) | 23 (2.5%) | .0021 b |
| Long-term azithromycin, n (%) | 128 (41%) | 305 (32.6%) | .0067 b |
| Antifungal treatment, n (%) | 24 (7.7%) | 8 (0.9%) | <.0001 b |
| CFTR modulator therapy | |||
| None, n (%) | 244 (78.2%) | 711 (76%) | .4181b |
| Elexacaftor/tezacaftor/ivacaftor, n (%) | 1 (0.3) | 7 (0.7) | .6875c |
| Lumacaftor/ivacaftor, n (%) | 59 (18.9) | 182 (19.4) | .8360b |
| Ivacaftor, n (%) | 6 (1.9) | 27 (2.9) | .3593b |
| Tezacaftor/ivacaftor, n (%) | 2 (0.6) | 9 (1) | .7407c |
Note: Significant P-values are in bold. IQR: interquartile range, CFTR: cystic fibrosis transmembrane conductance regulator, ppFEV1: predicted forced expiratory volume in 1 s, BMI: body mass index, MRSA: Methicillin resistant Staphylococcus aureus MSSA: Methicillin susceptible Staphylococcus aureus.
Age at time of matching.
Wilcoxon signed-rank test.
Pearson's χ2 test,
Fisher's exact test.
Univariate analysis
In the univariate analysis (Table 2), ABPA was significantly associated with glucose disorders (P = .005), asthma (P = .0004), and chronic airway colonization by P. aeruginosa (P = .003). There was also an association between ABPA and treatments, including intravenous and inhaled antibiotics (both P < .0001), inhaled and oral corticosteroids (P < .0001 and P = .003, respectively), azithromycin (P = .005), and antifungals (P < .0001).
Table 2.
Univariate associations between explicative factors and ABPA in people with cystic fibrosis from the French CF Registry.
| Odds ratio | 95% confidence interval | P-value | |
|---|---|---|---|
| CFTR genotypes | |||
| F508del/F508del vs. other/other | 0.87 | [0.60; 1.27] | .4712 |
| F508del/other vs. other/other | 0.98 | [0.68; 1.43] | .9316 |
| Nutritional status, BMI | |||
| BMI (continuous) | 0.93 | [0.88; 0.98] | .0070 |
| Low (<18.5) vs. normal (18.5–24.9) | 1.15 | [0.79; 1.67] | .4532 |
| High (>25) vs. normal (18.5–24.9) | 0.47 | [0.24; 0.93] | .0297 |
| Glucose disorder | |||
| Glucose disorder vs. none | 1.52 | [1.13; 2.05] | .0055 |
| Glucose intolerance vs. none | 1.48 | [1.08; 2.02] | .0147 |
| CF-related diabetes vs. none | 1.68 | [1.11; 2.53] | .0144 |
| Severe CF-liver disease (cirrhosis) | 1.10 | [0.75; 1.61] | .6224 |
| Respiratory status | |||
| ppFEV1 (continuous) | 0.99 | [0.98; 1.00] | .0012 |
| ppFEV1 < 40% vs. ≥ 70% | 1.28 | [0.75; 2.17] | .3675 |
| ppFEV1 40%–70% vs. ≥ 70% | 1.45 | [1.03; 2.04] | .0324 |
| Asthma | 1.60 | [1.24; 2.08] | .0004 |
| Chronic airway colonization | |||
| Pseudomonas aeruginosa colonization vs. none | 1.64 | [1.18; 2.27] | .0030 |
| MRSA colonization vs. none | 2.02 | [0.98; 4.14] | .0559 |
| MSSA colonization vs. none | 1.02 | [0.75; 1.37] | .9090 |
| Burkholderia species colonization vs. none | 1.53 | [0.51; 4.62] | .4490 |
| Achromobacter species colonization vs. none | 0.95 | [0.40; 2.25] | .9136 |
| Stenotrophomonas maltophilia colonization vs. none | 1.50 | [0.61; 3.72] | .3811 |
| Therapy for advanced pulmonary disease | |||
| Any (oxygen or NIV) vs. none | 2.77 | [1.47; 5.24] | .0017 |
| Oxygenotherapy vs. none | 2.27 | [1.04; 4.97] | .0396 |
| NIV vs. none | 2.60 | [1.10; 6.14] | .0300 |
| Number of annual intravenous antibiotic courses | |||
| 1–3 vs. 0 | 2.13 | [1.57; 2.90] | <.0001 |
| >3 vs. 0 | 3.74 | [1.99; 7.04] | <.0001 |
| Long-term inhaled antibiotics | 1.91 | [1.39; 2.62] | <.0001 |
| Long-term inhaled mucolytics | 1.46 | [1.05; 2.04] | .0257 |
| Long-term inhaled corticosteroids | 2.03 | [1.52; 2.73] | <.0001 |
| Long-term oral corticosteroids | 2.48 | [1.35; 4.55] | .0034 |
| Long-term azithromycin | 1.48 | [1.13; 1.96] | .0050 |
| Antifungal treatment | 9.00 | [4.04; 20.03] | <.0001 |
| CFTR modulator therapy | 0.87 | [0.63; 1.20] | .3951 |
Note: Significant P-values are in bold.
CFTR: cystic fibrosis transmembrane conductance regulator, ppFEV1: predicted forced expiratory volume in 1 s, BMI: body mass index, MRSA: Methicillin resistant Staphylococcus aureus MSSA: Methicillin susceptible Staphylococcus aureus, NIV: noninvasive ventilation.
Lung function (ppFEV1) as a continuous variable was significantly associated with ABPA, showing a slight protective effect (OR = 0.99, P = .0012), though the effect size was small. When categorized into three groups, ppFEV1 between 40% and 70% was associated with ABPA (P = .032), while ppFEV1 < 40% showed no significant association (P = .367). Similarly, BMI as a continuous variable was associated with a lower risk of ABPA (OR = 0.93, P = .007), but with a small effect size. Categorization into three groups confirmed this trend, with individuals having a higher BMI (>25) exhibiting a reduced risk of ABPA (P = .03) compared to those with a lower BMI.
Multivariate analysis
The multivariate analysis (Table 3 and Fig. 1) confirmed the following factors associated with ABPA: glucose disorders (OR = 1.41, P = .03), higher numbers of annual intravenous antibiotic courses (OR = 1.69, P < .01 for 1–3 courses; OR = 2.47, P = .013 for > 3 courses), long-term antifungal treatment (OR = 5.83, P < .0001), and long-term inhaled corticosteroids (OR = 1.82, P < .001). The protective effect of higher BMI against ABPA remained significant after adjustment (OR = 0.94, P = .041). There were no associations with advanced pulmonary disease treatments, ppFEV1 (continuous or categorical), long-term azithromycin use, or chronic P. aeruginosa colonization.
Table 3.
Multivariate associations between explicative factors and ABPA in people with cystic fibrosis from the French CF Registry.
| Odds ratio | 95% confidence interval | P-value | |
|---|---|---|---|
| Nutritional status, BMI | |||
| BMI (continuous) | 0.94 | [0.89; 1.00] | .0412 |
| Low (<18.5) vs. normal (18.5–24.9) | 1.05 | [0.70; 1.58] | .8121 |
| High (>25) vs. normal (18.5–24.9) | 0.47 | [0.24; 0.95] | .0365 |
| Glucose disorder vs. none | 1.41 | [1.02; 1.93] | .0353 |
| Respiratory status | |||
| ppFEV1 (continuous) | 1.00 | [1.00; 1.01] | .7745 |
| ppFEV1 < 40% vs. ≥ 70% | 0.75 | [0.40; 1.40] | .3619 |
| ppFEV1 40%–70% vs. ≥ 70% | 0.90 | [0.61; 1.34] | .5994 |
| Pseudomonas aeruginosa colonization vs. none | 1.31 | [0.90; 1.89] | .1575 |
| Therapy for advanced pulmonary disease (oxygen or NIV) vs. none | 1.75 | [0.83; 3.67] | .1384 |
| Number of annual intravenous antibiotic courses | |||
| 1–3 vs. 0 | 1.69 | [1.19; 2.39] | .0033 |
| >3 vs. 0 | 2.47 | [1.21; 5.05] | .0132 |
| Long-term inhaled corticosteroids | 1.82 | [1.33; 2.48] | .0002 |
| Long-term azithromycin | 1.24 | [0.92; 1.67] | .1644 |
| Antifungal treatment | 5.83 | [2.55; 13.35] | <.0001 |
Note: Significant P-values are in bold.
BMI: body mass index, ppFEV1: predicted forced expiratory volume in 1 s, NIV: noninvasive ventilation.
Figure 1.
Multivariate analysis of factors associated with ABPA in people with cystic fibrosis from the French Cystic Fibrosis Registry. Significant associations include annual intravenous antibiotic courses (OR = 1.69, P < .01 for 1–3 courses; OR = 2.47, P = .013 for > 3 courses), glucose disorders (OR = 1.41, P = .03), long-term inhaled corticosteroids (OR = 1.82, P < .001), and long-term antifungals (OR = 5.83, P < .0001). Higher BMI was protective (OR = 0.47, P = .037). No associations were found with chronic Pseudomonas aeruginosa colonization, lung function (percentage of predicted forced expiratory volume in 1 s, ppFEV1), therapy for advanced pulmonary disease (oxygen or noninvasive ventilation), or long-term azithromycin use.
Discussion
This case-control study, using data from the FCFR, identified several factors associated with ABPA in pwCF. These include glucose disorders, increased frequency of intravenous antibiotic courses prior to ABPA diagnosis, long-term inhaled corticosteroids, and long-term antifungal therapy. Conversely, a higher BMI (>25 kg/m²) emerged as a potential protective factor against ABPA, while no associations were found with worse lung function or chronicPseudomonas aeruginosa colonization.
The link between ABPA and glucose disorders in pwCF has been noted in few studies. Previous research has shown that pwCF with CF-related diabetes are at higher risk of persistent Aspergillus fumigatus respiratory isolation, worsening lung function, and increased mortality.17,18 Hyperglycemia and diabetes can promote A. fumigatus colonization, potentially leading to ABPA. PwCF with glucose disorders tends to exhibit increased bacterial colonization, likely due to elevated glucose levels in alveolar surface fluid.18 Furthermore, hyperglycemia can induce pulmonary dysbiosis, which correlates with disease severity in CF, though its association with ABPA remains understudied.19 In non-CF diabetic patients, hyperglycemia has been linked to decreased neutrophil function, which may impair immune response against A. fumigatus.20 These factors—microbial colonization, respiratory microbiome disruption, and immune dysregulation—could contribute to the increased susceptibility of pwCF with glucose disorders to ABPA.21
Our study also identified several treatments frequently prescribed to pwCF as being associated with ABPA. PwCF frequently experience recurrent infections requiring antibiotic treatment to prevent lung function decline. Our findings suggest that antibiotic treatments are associated with the development of ABPA, with a stronger association observed as the number of intravenous courses increases. While this correlation has also been reported by others,22 it remains difficult to discern whether it reflects the worsening condition of patients prior to ABPA onset or the potential for repeated antibiotic therapies to select for microbial populations, such as Aspergillus species, that favor colonization. Similar associations have been noted between long-term use of inhaled antibiotics and persistent respiratory colonization by A. fumigatus.17,23
Inhaled corticosteroids were also associated with ABPA, aligning with previous findings that long-term corticosteroid therapy may predispose pwCF to A. fumigatus colonization and ABPA development.17,24 This highlights the importance of monitoring patients on long-term corticosteroid therapy for ABPA signs. Interestingly, antifungal treatment was associated with ABPA, likely due to clinicians initiating antifungal therapy after detecting fungal colonization. This may reflect a response to A. fumigatus detection rather than a direct causal relationship. Future studies should explore antifungal resistance and the implications of prolonged antifungal use in ABPA.
The protective effect of a higher BMI against ABPA, observed in our analysis, is a novel finding. Previous studies suggest that overweight or obese individuals may have a Th1-skewed immune response, reducing Th2-mediated eosinophilic inflammation, which is central to ABPA development.25 With the increase in BMI in pwCF, partly due to CFTR modulator therapy, this observation warrants further investigation.26 Contrary to previous studies, we did not observe strong associations between ABPA and malnutrition or the use of azithromycin or mucolytics. This could be due to differences in study populations or evolving treatment practices.17,22,27
Our study did not identify a significant association between lung function (ppFEV1) and ABPA development, which contrasts with previous findings suggesting a link between ABPA and poorer lung function.14,22,28,29 However, the relationship between lung function and ABPA remains complex and somewhat controversial. While ABPA can contribute to lung function decline due to airway inflammation and bronchiectasis,12,15 it is unclear whether lower lung function itself predisposes pwCF to ABPA or if the decline is a consequence of ABPA rather than a risk factor. Notably, our univariate analysis did reveal an association between ppFEV1 and ABPA, suggesting a potential link between lung function impairment and ABPA risk. However, this association did not persist in the multivariate analysis, indicating that other confounding factors may explain the observed relationship. The absence of an independent association after adjustment suggests that lower lung function alone may not be a primary risk factor for ABPA in pwCF but rather an indicator of overall disease severity. Further, longitudinal studies are needed to clarify the temporal relationship between lung function decline and ABPA development, particularly to determine whether lung function impairment precedes ABPA onset or is primarily a consequence of disease progression.
This study has several limitations. First, the use of registry data, while comprehensive, may be subject to reporting biases or missing information, particularly in cases where data collection is incomplete or inconsistent across different centers. Second, as the FCFR does not collect data on Aspergillus airway cultures or fungal sensitization, we were unable to assess the potential role of fungal colonization in ABPA development. This also prevented us from determining whether antifungal treatments in the year preceding ABPA diagnosis reflected persistent Aspergillus colonization or were prescribed for other indications. Additionally, the registry does not provide information on antifungal treatment indications, duration, or effectiveness, limiting a more detailed analysis of their impact. Consequently, some controls may have received antifungal treatment for reasons unrelated to ABPA, such as persistent Aspergillus colonization, but this could not be verified. Third, while all French CF centers adhere to standardized ABPA diagnostic criteria, the anonymized nature of the registry prevented us from assessing potential center effects, which could introduce variability in diagnosis or treatment approaches. Fourth, although chronic Aspergillus airway colonization is not included in the ABPA diagnostic criteria, its role as a risk factor could not be evaluated due to the lack of systematic data collection on fungal colonization in the registry. Additionally, while propensity score matching was employed to reduce confounding, unmeasured variables not captured in the registry, such as environmental factors or detailed treatment adherence, may still influence the results. Moreover, given the number of statistical comparisons performed, the possibility of Type I error (false positives) cannot be excluded. Although adjustments were made to control for confounding, multiple testing increases the likelihood of finding associations by chance, and caution is warranted when interpreting findings. Lastly, the small number of patients on newer CFTR modulators, particularly elexacaftor/tezacaftor/ivacaftor, limits the generalizability of findings related to these therapies. Despite these limitations, this study provides valuable insights into factors associated with ABPA in pwCF. The large sample size and matching methodology strengthen our findings, while the identification of novel factors such as glucose disorders and the potential protective effect of a higher BMI underscores the complexity of ABPA pathogenesis in CF.
In conclusion, this study provides important insights into the factors associated with ABPA in pwCF, drawing from data in the FCFR. Our findings emphasize the need for careful assessment of certain treatments, such as antibiotics and inhaled corticosteroids, due to their association with ABPA progression. Additionally, pwCF with glucose disorders appear to be particularly vulnerable to ABPA, suggesting the importance of more frequent screening in this subgroup. The potential protective effect of a higher BMI also warrants further investigation, especially with the growing use of CFTR modulators, which may promote weight gain in CF patients. As more data become available over time, particularly regarding the effects of CFTR modulators, future studies should continue to explore the evolving relationship between ABPA and CF, with the goal of optimizing patient care and improving outcomes for individuals affected by both conditions.
Supplementary Material
Acknowledgments
This study was supported by the association Vaincre La Mucoviscidose, which provided data from its Registry, and participated in the interpretation of data and writing of the report.
Contributor Information
Marine Tarizzo, Sorbonne Université, Inserm U938, Centre de Recherche Saint-Antoine (CRSA), 75012 Paris, France; Sorbonne Université, Assistance Publique - Hôpitaux de Paris (AP-HP), Hôpital Trousseau, Service de Pneumologie Pédiatrique, GRC SoLID, 75012 Paris, France.
Lydie Lemonnier, Association Vaincre la Mucoviscidose, Registre français de la mucoviscidose, 75013 Paris, France.
Soline Leblanc, IT&M STATS, Division Interne, 92100 Boulogne-Billancourt, France.
Jeanne Bigot, Sorbonne Université, Inserm U938, Centre de Recherche Saint-Antoine (CRSA), 75012 Paris, France; Sorbonne Université, Assistance Publique - Hôpitaux de Paris (AP-HP), Hôpital Saint-Antoine, Laboratoire de Parasitologie-Mycologie, GRC SoLID, 75012 Paris, France.
Guillaume Thouvenin, Sorbonne Université, Inserm U938, Centre de Recherche Saint-Antoine (CRSA), 75012 Paris, France; Sorbonne Université, Assistance Publique - Hôpitaux de Paris (AP-HP), Hôpital Trousseau, Service de Pneumologie Pédiatrique, GRC SoLID, 75012 Paris, France.
Loïc Guillot, Sorbonne Université, Inserm U938, Centre de Recherche Saint-Antoine (CRSA), 75012 Paris, France.
Harriet Corvol, Sorbonne Université, Inserm U938, Centre de Recherche Saint-Antoine (CRSA), 75012 Paris, France; Sorbonne Université, Assistance Publique - Hôpitaux de Paris (AP-HP), Hôpital Trousseau, Service de Pneumologie Pédiatrique, GRC SoLID, 75012 Paris, France.
Author contributions
Marine Tarizzo (Conceptualization, Data curation, Investigation, Methodology, Writing – original draft, Writing – review & editing), Lydie Lemonnier (Data curation, Formal analysis, Investigation, Methodology, Writing – original draft), Soline Leblanc (Data curation, Formal analysis, Investigation, Methodology, Validation, Writing – review & editing), Jeanne Bigot (Investigation, Methodology, Writing – original draft, Writing – review & editing), Guillaume Thouvenin (Investigation, Methodology, Writing – original draft, Writing – review & editing), Loïc Guillot (Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing), and Harriet Corvol (Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Validation, Writing – original draft, Writing – review & editing).
Funding
This work was supported by the association Vaincre la Mucoviscidose (RC20220502999).
Declaration of interest
The authors have no relevant financial or non-financial interests to disclose.
References
- 1. Guo J, Garratt A, Hill A. Worldwide rates of diagnosis and effective treatment for cystic fibrosis. Eur Cystic Fibrosis Soc. 2022; 21(3): 456–462. 10.1016/j.jcf.2022.01.009 [DOI] [PubMed] [Google Scholar]
- 2. Vaincre-la-Mucoviscidose . Registre Français de la mucoviscidose - Bilan des données 2022. 2023. https://www.vaincrelamuco.org/sites/default/files/registre_francais_mucoviscidose_bilan2022.pdf [Google Scholar]
- 3. Grasemann H, Ratjen F. Cystic fibrosis. N Engl J Med. 2023; 389(18): 1693–1707. 10.1056/NEJMra2216474 [DOI] [PubMed] [Google Scholar]
- 4. Corvol H, Thompson KE, Tabary O, le Rouzic P, Guillot L. Translating the genetics of cystic fibrosis to personalized medicine. J Lab Clin Medi. 2016; 168: 40–49. 10.1016/j.trsl.2015.04.008 [DOI] [PubMed] [Google Scholar]
- 5. Mésinèle J, Ruffin M, Guillot L, Boëlle PY, Corvol H. Airway infections as a risk factor for Pseudomonas aeruginosa acquisition and chronic colonisation in children with cystic fibrosis. Eur Cystic Fibrosis Soc. 2023; 22(5): 901–908. 10.1016/j.jcf.2023.06.007 [DOI] [PubMed] [Google Scholar]
- 6. Mésinèle J, Ruffin M, Kemgang A, Guillot L, Boëlle PY, Corvol H. Risk factors for Pseudomonas aeruginosa airway infection and lung function decline in children with cystic fibrosis. Eur Cystic Fibrosis Soc. 2022; 21(1): 45–51. 10.1016/j.jcf.2021.09.017 [DOI] [PubMed] [Google Scholar]
- 7. Cystic Fibrosis Foundation Patient Registry . 2022 Annual Data Report. 2023. https://www.cff.org/media/31216/download
- 8. Reece E, Segurado R, Jackson A, McClean S, Renwick J, Greally P. Co-colonisation with Aspergillus fumigatus and Pseudomonas aeruginosa is associated with poorer health in cystic fibrosis patients: An Irish registry analysis. BMC Pulmonary Medi. 2017; 17(1): 70. 10.1186/s12890-017-0416-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Latgé JP, Chamilos G. Aspergillus fumigatus and aspergillosis in 2019. Clin Microbiol Rev. 2019; 33(1): e00140–18. 10.1128/CMR.00140-18 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Roboubi A, Audousset C, Fréalle É, et al. Allergic bronchopulmonary aspergillosis: A multidisciplinary review. J Mycol Med. 2023; 33(3): 101392. 10.1016/j.mycmed.2023.101392 [DOI] [PubMed] [Google Scholar]
- 11. Simmonds NJ, Southern KW, De Wachter E, et al. ECFS standards of care on CFTR-related disorders: Identification and care of the disorders. Eur Cystic Fibrosis Soc. 2024; 23, 590–602. 10.1016/j.jcf.2024.03.008 [DOI] [PubMed] [Google Scholar]
- 12. Agarwal R, Sehgal IS, Muthu V, et al. Revised ISHAM-ABPA working group clinical practice guidelines for diagnosing, classifying and treating allergic bronchopulmonary aspergillosis/mycoses. Eur Respir J. 2024; 63(4): 2400061. 10.1183/13993003.00061-2024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Janahi IA, Rehman A, AR A-N. Allergic bronchopulmonary aspergillosis in patients with cystic fibrosis. Annals Thoracic Med. 2017; 12(2): 74–82. 10.4103/atm.ATM_231_16 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Kaditis AG, Miligkos M, Bossi A, et al. Effect of allergic bronchopulmonary aspergillosis on FEV(1) in children and adolescents with cystic fibrosis: A European Cystic Fibrosis Society Patient Registry analysis. Arch Dis Child. 2017; 102(8): 742–747. 10.1136/archdischild-2016-311132 [DOI] [PubMed] [Google Scholar]
- 15. Stevens DA, Moss RB, Kurup VP, et al. Allergic bronchopulmonary aspergillosis in cystic. fibrosis–state of the art: Cystic Fibrosis Foundation Consensus Conference. Clin Infect Dis. 2003; 37:Suppl 3: S225–S264. 10.1086/376525 [DOI] [PubMed] [Google Scholar]
- 16. Burnham KP, Anderson DR. Model Selection and Inference: A Practical Information-Theoretic Approach, 2nd edin. New York: Springer-Verlag, 2002. [Google Scholar]
- 17. Hong G, Psoter KJ, Jennings MT, et al. Risk factors for persistent Aspergillus respiratory isolation in cystic fibrosis. Eur Cystic Fibrosis Soc. 2018; 17(5): 624–630. 10.1016/j.jcf.2018.01.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Prentice BJ, Jaffe A, Hameed S, Verge CF, Waters S, Widger J. Cystic fibrosis-related diabetes and lung disease: An update. Eur Respiratory Soc. 2021; 30(159): 200293. 10.1183/16000617.0293-2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Françoise A, Héry-Arnaud G. The microbiome in cystic fibrosis pulmonary disease. Genes. 2020; 11(5): 536. 10.3390/genes11050536 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Joshi MB, Lad A, Bharath Prasad AS, Balakrishnan A, Ramachandra L, Satyamoorthy K. High glucose modulates IL-6 mediated immune homeostasis through impeding neutrophil extracellular trap formation. FEBS Lett. 2013; 587(14): 2241–2246. 10.1016/j.febslet.2013.05.053 [DOI] [PubMed] [Google Scholar]
- 21. Schei K, Simpson MR, Øien T, Salamati S, Rudi K, Ødegård RA. Allergy-related diseases and early gut fungal and bacterial microbiota abundances in children. Clin Translational Allergy. 2021; 11(5): e12041. 10.1002/clt2.12041 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. De Baets F, De Keyzer L, Van Daele S, et al. Risk factors and impact of allergic bronchopulmonary aspergillosis in Pseudomonas aeruginosa-negative CF patients. Pediatr Allergy Immunol. 2018; 29(7): 726–731. 10.1111/pai.12953 [DOI] [PubMed] [Google Scholar]
- 23. Düesberg U, Wosniok J, Naehrlich L, Eschenhagen P, Schwarz C. Risk factors for respiratory Aspergillus fumigatus in German cystic fibrosis patients and impact on lung function. Sci Rep. 2020; 10(1): 18999. 10.1038/s41598-020-75886-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Noni M, Katelari A, Dimopoulos G, et al. Inhaled corticosteroids and Aspergillus fumigatus isolation in cystic fibrosis. Med Mycol. 2014; 52(7): 715–722. 10.1093/mmy/myu038 [DOI] [PubMed] [Google Scholar]
- 25. Kuruvilla ME, Lee FE, Lee GB. Understanding asthma phenotypes, endotypes, and mechanisms of disease. Clin Rev Allergy Immunol. 2019; 56(2): 219–233. 10.1007/s12016-018-8712-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Litvin M, Yoon JC, Leey Casella J, Blackman SM, Brennan AL. Energy balance and obesity in individuals with cystic fibrosis. Eur Cystic Fibrosis Soc. 2019; 18(2): S38–S47. 10.1016/j.jcf.2019.08.015 [DOI] [PubMed] [Google Scholar]
- 27. Jubin V, Ranque S, Stremler L, Bel N, Sarles J, Dubus JC. Risk factors for Aspergillus colonization and allergic bronchopulmonary aspergillosis in children with cystic fibrosis. Pediatr Pulmonol. 2010; 45(8): 764–771. 10.1002/ppul.21240 [DOI] [PubMed] [Google Scholar]
- 28. Mastella G, Rainisio M, Harms HK, et al. Allergic bronchopulmonary aspergillosis in cystic fibrosis. A European epidemiological study. Epidemiologic registry of cystic fibrosis. Eur Respir J. 2000; 16(3): 464–471. 10.1034/j.1399-3003.2000.016003464.x [DOI] [PubMed] [Google Scholar]
- 29. Chesshyre E, Warren FC, Shore AC, Davies JC, Armstrong-James D, Warris A. Long-term outcomes of allergic bronchopulmonary aspergillosis and Aspergillus colonization in children and adolescents with cystic fibrosis. J Fungi (Basel, Switzerland). 2024; 10(9): 599. 10.3390/jof10090599 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.