Allergic bronchopulmonary aspergillosis (ABPA) in asthma

Allergic bronchopulmonary aspergillosis (ABPA) is a pulmonary condition described in patients with asthma and cystic fibrosis (CF) caused by hypersensitivity to Aspergillus. It was first described in 1952 from the London Chest Hospital (Hinson et al, 1952). Although inhaled fungal conidia are normally removed from the airways of healthy people by defence mechanisms, defective clearance in patients with asthma and CF allows germination of conidia to hyphae, which then induce the production of pro-inflammatory cytokines that are responsible for the development of symptoms. ABPA complicating asthma (non-CF) is discussed here.

ABPA complicates about 1-4.1% of patients with asthma in secondary care (Benatar et al, 1980; Varshokar 2002; Donnelly et al, 1991; Eaton et al, 2000; Al-Mobeireek et al, 2001; Ma et al, 2011). It manifests as poorly controlled asthma, and exacerbations consist of fever, malaise, expectoration of mucus plugs and haemoptysis. Imaging may show pulmonary infiltrates and bronchiectasis may develop with time. Many patients present with the complication of bronchiectasis well established. The diagnostic approach consists of a screening test for Aspergillus hypersensitivity (either skin test or Aspergillus-specific IgE), followed by testing for total IgE levels, which have to be elevated beyond a certain value (e.g. >1000 IU/mL), although there is no widespread consensus regarding the cut-off value. (Agarwal et al, 2014) 

Recognition of ABPA is important as it may, if uncontrolled, eventually lead to bronchiectasis and permanent lung damage. Therefore, the goal of treatment is to control asthma symptoms, reduce the frequency of exacerbations, and prevent development of bronchiectasis and CPA.

An acute exacerbation of ABPA is suspected on the basis of poorly controlled asthma, typical symptoms (mucous plugs, fever, malaise, and haemoptysis), presence of new pulmonary infiltrates and sometimes high-attenuation mucus impaction on imaging, and elevated total IgE and Aspergillus IgE titres. The presence of high-attenuation mucoid impaction is characteristic of ABPA and, although previously thought to be rare, 18% of patients had high-attenuation mucus on high resolution CT (Agarwal et al, 2007). It is not clear why some patients develop mucus impaction, but there may be a genetic element predisposing to more severe inflammation. It was associated with higher levels of serological markers like total IgE and was a predictor of ABPA relapse in a recent study (Agarwal et al, 2007).

 Acute exacerbations are treated with systemic corticosteroid therapy. There are no established guidelines on the recommended regimen, and a dose of 0.5mg/kg or higher for a few weeks or months is usually used. A regimen recommended by Greenberger (2002) consists of prednisolone 0.5mg/kg/day for 1-2 weeks, then on alternate days for 6-8 weeks, then taper by 5-10 mg every 2 weeks until stopping. Higher steroid doses (prednisolone 0.75mg/kg) tapered more gradually over 6-12 months were associated with longer periods of remission but there are no direct comparisons between regimens (Agarwal et al, 2006). Shorter periods of high dose oral corticosteroids (ie 40mg prednisolone daily for 3 weeks tapered to zero over 2-4 weeks may also be effective. A randomised trial comparing low and high-dose steroids has been completed but results are not available yet (; NCT00974766).  Response to therapy can be gauged both clinically and serologically; serum IgE levels should fall and radiological infiltrates should clear.  Patients who fail to improve with corticosteroids may benefit from a therapeutic bronchoscopy to remove obstructing plugs. Older data suggested direct instillation of N-acetylcysteine may be helpful in relieving obstructed airways, but there are few recent data. Nebulised hypertonic (6-7%) saline twice daily may also be useful in clearing sputum plugs.

Patients who respond to steroids are classified as being in remission. Patients with frequent exacerbations or with recurrent symptoms when steroids are tapered are classified as steroid-dependent and may benefit from treatment with antifungal agents which may have a steroid-sparing effect.

The first report on antifungal treatment for ABPA was a case report of nystatin inhalations published in 1967 (Stark et al, 1967), followed by a report on inhaled natamycin (Henderson et al, 1968). In a small randomised trial, oral ketoconazole led to improvement in symptoms and serological markers; however a larger trial was abandoned because of significant toxicity (Shale et al, 1987), and was ineffective in other series (Fournier et al, 1984). Ketoconazole has no anti-Aspergillus activity. In the first controlled trial of antifungal therapy, inhaled natamycin twice daily was also ineffective (Currie et al, 1990).

Subsequently, the release of itraconazole yielded an alternative therapeutic avenue for ABPA. Itraconazole 200-400 mg/d showed benefit in several open studies. In one open study, all 3 patients experienced improvement in pulmonary function (Denning et al, 1991). In another series, as part of an international open-labelled compassionate use program several patients responded but the stage of disease and measurement of response was not well characterised (de Beule et al, 1988). Another study compared a 2-year period before introduction of itraconazole with that on 200mg daily of itraconazole in 14 patients with ABPA and asthma (Salez et al, 1999) and reported improvement in blood eosinophilia, total IgE, Aspergillus precipitating antibody and FEV1. Ιtraconazole had a steroid-sparing effect. Similar results were reported by Germaud et al (1995).

In two placebo-controlled trials in asthmatic ABPA patients, there was a clear benefit of itraconazole (Stevens et al, 2000 and Wark et al, 2003). Stevens et al compared itraconazole 200mg twice daily to placebo for 16 weeks and then all patients received 16 weeks of 200mg daily. Response was defined by 3 components - at least a 50% reduction in steroid dose and at least a 25% reduction in total IgE and either an increase by at least 25% in exercise tolerance or pulmonary function tests or an absence of pulmonary infiltrates. In the first phase of the study, 13 of 28 (46%) of patients receiving itraconazole responded compared with 5 or 27 (19%) receiving placebo (p=0.04). Patients older than 50 were more likely to respond (p=0.045), as were those without bronchiectasis as determined on chest X-rays. No relapses occurred on the lower dose in the second phase of the study in those who responded, and an additional 12 (36%) patients responded, 8 of whom had received placebo previously. Only a few isolates of Aspergillus were collected. Those that were resistant (MIC 6.25g/ml) or tolerant (MIC <6.25g/ml and MFC 6.25g/ml) to itraconazole were associated with a lack of response to 200mg daily of itraconazole. Wark et al used itraconazole 400mg/day and noted a significant effect on number of exacerbations requiring steroids (p=0.03) and a significant decrease in IgE (p<0.01) but no change in FEV1.

For those with ABPA who are corticosteroid-dependent or severely disabled by the disease, a trial of itraconazole 400 mg/d is warranted. If there is no improvement over 16-32 weeks, measurement of serum concentrations may be useful as a guide to appropriate dosing or susceptibility testing of the infecting isolate of Aspergillus. For those who respond, at least 200mg daily should be continued until no further improvement is seen (usually 6-9 months). However, resistance has been noted to develop in ABPA patients, especially when itraconazole levels have been subtherapeutic (Howard et al, 2010). Therefore monitoring for evolving resistance is indicated in patients on itraconazole. Other azoles like voriconazole and posaconazole are also effective and can be used in cases of failure or intolerance of itraconazole (Chishimba et al, 2012; Mulliez 2010). Finally, a thorough search for drug interactions is essential for all azoles. For example, itraconazole has significant drug interactions with omeprazole, simvastatin and steroids, including inhaled steroids which are used in most patients with concomitant asthma.

There is limited experience on the use of other therapies. Nebulized amphotericin B (e.g. Fungizone 10mg twice daily) has been used in cases non-responsive or intolerant to itraconazole (Chishimba et al, 2014; Godet et al 2012). In the Chishimba et al study, only 14.3% of 21 patients had a benefit, measured with quality of life questionnaires and FEV1. However, 1 in 3 patients failed initial dose because of bronchospasm. Therefore, nebulised amphotericin B can only be used when other alternatives are not available. Inhaled corticosteroids do not appear to have a therapeutic effect in ABPA other than control of asthma (Agarwal et al, 2011). The monoclonal IgE antibody omalizumab, used for treatment of asthma, was shown to have a steroid-sparing effect in 4 patients with asthma and ABPA (Collins et al, 2012).

Bronchiectasis is the most important complication of ABPA. It may be present initially when ABPA is diagnosed, or develop subsequently, especially when the disease exacerbations are not well controlled. The typical presentation is of central bronchiectasis, involving the inner two thirds of the lung parenchyma. Once bronchiectasis develops, it should be managed according to guidelines (e.g Pasteur et al, 2010) in order to prevent exacerbations and further decline in lung function.

Finally, it may be helpful to enquire about possible sources of exposure to Aspergillus and try to eliminate these (e.g. mouldy or damp areas in the home, gardening, and occupational exposure). 



ABPA complicates up to 18% of patients with CF based on prevalence studies from various parts of the world (Geller 1999, Mastella 2000, Carneiro 2008, Sharma 2014). Aspergillus colonisation is very frequent, up to 58% in CF patients, but not all will develop ABPA. ABPA presents at a younger age in CF patients compared to patients with asthma.  

The criteria used for ABPA diagnosis in asthma patients are not very satisfactory in CF because of the overlapping symptoms and findings caused by other processes as well as the deterioration caused by the underlying disease. In addition, precipitating antibodies to Aspergillus are common in adult CF patients and most of these do not have ABPA. Baxter et al proposed a classification system for the diagnosis of ABPA in CF patients (Baxter et al, 2013).

Systemic corticosteroids are recommended for exacerbations of ABPA in CF in the same doses used for ABPA in asthma as there are no data pertaining specifically to CF patients. Less evidence exists regarding the use of antifungals in CF compared to asthma and there are no randomised trials. In the first report of itraconazole use in CF, Denning et al reported improvement in FEV1 in all 3 patients (Denning et al, 1991). One study of itraconazole in 16 patients with ABPA and CF demonstrated fewer exacerbations and a reduction in oral corticosteroid dose (Nepomuceno et al, 1999). In two studies of 13 and 21 patients, an improvement in FEV1 and in IgE levels was seen with itraconazole (Skov et al, 2002) and voriconazole (Hilliard et al, 2005). In two other series, antifungals had a steroid-sparing effect (Proesmans et al, 2010; Glackin et al, 2009). No benefit was seen in another study (Casaulta et al, 2005).     

There has been more experience on the use of nebulized amphotericin B in CF patients with ABPA compared to asthmatics, but still only a few case series exist. Some series report a steroid-sparing effect (Proesmans et al 2010; Tiddens et al 2003). Nebulised amphotericin has also been beneficial in case reports (Suzuki et al 2002; Hayes et al 2010) and in another small series (Laoudi et al 2008).  

The IgE monoclonal antibody omalizumab has been used in patients with ABPA and CF. In the first report of 3 children, it led to clinical improvement and discontinuation of steroids (Zirbes and Milla, 2008). Case reports or small case series were subsequently published (Kanu and Patel, 2008; Lebecque et al 2009; Elmallah et al, 2012; Wong et al, 2013; Zicari et al, 2014). In a series of 6 patients who received omalizumab because of steroid toxicity or treatment failure, all patients had an improvement in FEV1 (from an average of 57.5% to 62.3% after 3-4 months of omalizumab), and steroids could be tapered or discontinued.