Abstract
Severe asthma with fungal sensitisation and allergic bronchopulmonary aspergillosis encompass two closely related subgroups of patients with severe allergic asthma. Pulmonary disease is due to pronounced host inflammatory responses to noninvasive subclinical endobronchial infection with filamentous fungi, usually Aspergillus fumigatus. These patients usually do not achieve satisfactory disease control with conventional treatment of severe asthma, i.e. high-dose inhaled corticosteroids and long-acting bronchodilators. Although prolonged systemic corticosteroids are effective, they carry a substantial toxicity profile. Supplementary or alternative therapies have primarily focused on use of antifungal agents including oral triazoles and inhaled amphotericin B. Immunomodulation with omalizumab, a humanised anti-IgE monoclonal antibody, or "pulse" monthly high-dose intravenous corticosteroid, has also been employed. This article considers the experience with these approaches, with emphasis on recent clinical trials.
Abstract
Treatment of fungal asthma includes glucocorticoids, azoles, amphotericin and anti-IgE. Trial validation is needed. http://ow.ly/uavHn
Introduction
It is thought that up to 10% of people with asthma have poorly controlled disease with major life impact despite guideline-based combination high-dose inhaled corticosteroid/long-acting bronchodilator therapy, i.e. severe asthma. One-third to one-half of these severe asthmatics has atopic sensitisation to filamentous fungi, most prominently to Aspergillus fumigatus [1]. Evidence has mounted that fungal sensitisation is associated with a more severe asthma phenotype [2–8]. Thus, an important identifiable subgroup of asthma, termed severe asthma with fungal sensitisation (SAFS), has emerged [4, 9]. Identification of SAFS as a recognisable asthma phenotype appears to carry important therapeutic implications.
It is also becoming clear that many asthmatics with an even more severe form of fungal inflammatory lung disease, usually due to A. fumigatus and known as allergic bronchopulmonary aspergillosis (ABPA), are often not properly diagnosed and have significant unmet diagnostic and therapeutic needs [10–13]. ABPA occurs almost exclusively in people with asthma or cystic fibrosis (CF). It results from atopic sensitisation to hyphal antigens of filamentous fungi (A. fumigatus in >90% of cases), which provokes a florid innate and adaptive immunoinflammatory response clinically characterised by: wheezy dyspnoea; malaise; and productive cough; very high IgE levels, elevated IgE and IgG antibodies to A. fumigatus; pronounced granulocytic (eosinophilic>neutrophilic) endo- and peribronchial pulmonary inflammation, pulmonary infiltrates with mucoid impaction of bronchi; proximal bronchiectasis; and, if left untreated, pulmonary fibrosis with the progressive loss of lung function (table 1) [11, 13, 15]. The pathophysiology of ABPA results from florid T-helper cell (Th)2 innate and adaptive immune responses in susceptible hosts who are unable to efficiently clear the respiratory epithelium of inhaled fungal spores (fig. 1) [15–18].
Pathophysiology of allergic bronchopulmonary aspergillosis. Pathogen-associated molecular pattern (PAMP) structures of Aspergillus fumigatus germinating spores and hyphae activate dendritic and respiratory epithelial cells via innate recognition receptors, such as Toll-like receptor (TLR)2/4 to produce T-helper cell (Th)2 immune-deviating cytokines; chemokines and costimulatory molecules that include thymic stromal lymphopoetin (TSLP), interleukin (IL)-25, IL-33, OX40 ligand (OX40L) and CCL17 (thymus-activated and -regulated chemokine), which in turn orchestrate differentiation, chemotaxis and activation of CD4+ Th2 cells. CCL17 also attracts regulatory T-cells (Treg) capable of suppressing protective Th1 responses and suppressing macrophage activation, thereby impairing fungal killing. Th2 cells produce a suite of cytokines including IL-4 and IL-5 that attract and activate eosinophils and drive differentiation of B-cells to IgE-secreting plasmacytes. IgE antibodies affix to tissue mast cells and circulating basophils to trigger immediate hypersensitivity reactions upon re-exposure to A. fumigatus allergens. The resulting granulocytic luminal mucus plugging and bronchocentric mucosal inflammation lead to bronchiectasis that may progress to fibrosis if untreated. The actions of current treatments are shown by red arrows; note the pleiotropic potential of glucocorticosteroids to act on multiple cell types and levels of the disease process. AM: alveolar macrophage; DC: dendritic cell; TNF: tumour necrosis factor. Reproduced and modified with permission from [16]; see also [17] and [18] for detailed discussion.
This article will consider SAFS and ABPA as closely related (and probably overlapping) nosological categories of severe asthma caused by noninvasive fungal airway infection, with emphasis upon recent therapeutic approaches and trials in these patients. With the World Health Organization estimated worldwide asthma prevalence of 300 000 000 cases, the possibility of up to 30 000 000 of these falling into the SAFS category is a very sobering prospect. In addition, deductive epidemiological modelling based upon literature reports suggests that ABPA causes an estimated worldwide illness burden of nearly 5 000 000 adult cases [11, 19].
Oral glucocorticosteroids have been employed as the fundamental therapy of ABPA for several decades, based on what appears to be clear efficacy in widespread empirical experience, despite a lack of randomised placebo-controlled trials [20]. Recently, a randomised open-label controlled trial of two oral prednisolone dose regimens has been completed but the results of this study have not yet been reported (ClinicalTrials.gov identifier NCT00974766). As conventional high-dose inhaled corticosteroid therapy is insufficient to control SAFS and ABPA, and chronic recurrent oral steroid therapy carries a troublesome toxicity profile, other therapies have been explored to address the unmet treatment needs for both of these related diseases. These are reviewed herein.
Antifungal therapy in ABPA and SAFS
The most prominent of alternatives to long-term oral steroids is the use of antifungal agents as an add-on or even as primary treatment of ABPA and SAFS.
Antifungal treatment of ABPA, and more recently SAFS, rests upon an assumption that allergic inflammatory responses arise in part from noninvasive airway fungal infection. Evidence for this assumption arises from several lines of investigation. First, serological studies in patients with ABPA and SAFS demonstrate IgE responses to fungal products derived from in vivo germination of inhaled conidia into hyphae, as these antibodies exist as a response to fungal exoproducts, hyphal cell wall components expressed during growth phase and cytoplasmic antigens [21, 22]. Secondly, there is mounting evidence that fungal hyphal products, including β-glucan and proteases, activate epithelial cells to secrete pro-inflammatory and Th2-polarising cytokines and chemokines directly via Toll-like receptor-induced signalling, via other innate sensors or via protease-activated receptors [23–26]. Thirdly, fungal asthma has been linked to host innate immune responses to chitin, a major fungal cell wall component, as chitinase promoter polymorphisms and associated alterations in chitinase activity and elevated chitinase-like protein YKL-40 levels have been found in SAFS in adults and children [27–33]. Finally, despite variably reported rates of fungal recovery from conventional respiratory cultures taken from patients with SAFS and ABPA, the use of PCR- or deep sequencing-based nonculture methods reveals that the vast majority of these patients have fungal DNA in these ex vivo samples. In addition, these patients usually have substantial levels of the hyphal cell wall component, galactomannan, in their sputum samples, often at levels consistent with invasive disease when measured in the blood [34, 35]. Direct ex vivo microscopy of such cases clearly demonstrates active hyphal growth in sputum plugs (fig. 2).
Endobronchial fungal infection in a young adult with cystic fibrosis and chronic Aspergillus fumigatus in sputum cultures. A spontaneously expectorated sputum plug was stored overnight at 4°C. Without processing, the plug was flattened between a coverslip and slide and viewed with an upright Nikon Eclipse E600FN Series microscope (Nikon, Tokyo, Japan) equipped for differential interference contrast microscopy. Digital imaging was via a Retiga-1300, cooled, 12-Bit, colour-Bayer Mosaic CCD camera with RGB Liquid Crystal Color Filter Module (QImaging, Surrey, Canada). Images were made with a ×40 (aperture 0.8, 2-mm working distance) water immersion lens (drop of water added on top of the cover slip, not touching the plug) producing an optical slice of ∼1 μm. a) Hyphal mat in sputum plug, possibly organising into biofilm (more parallel packed hyphal organisation). Older hyphae suggested by thicker walls and slight pigmentation. Granulocytes and exfoliated ciliated columnar epithelial cells can be seen in lower left quadrant and inset. b) Active endobronchial growth of younger thin-walled hyphae shown by tip extension and septation. Photograph courtesy of J.J. Wine (Cystic Fibrosis Research Laboratory, Stanford University, Stanford, CA, USA). a) Scale bars=20 μm; b) scale bar=50 μm.
To date, antifungal therapy for SAFS and ABPA has been directed against the main fungal pathogen, A. fumigatus (table 2). Initial trials of the imidazole ketoconazole and the polyene natamycin (antifungal agents lacking high activity against Aspergillus spp.) in patients with ABPA were disappointing [36, 37]. However, case reports and uncontrolled series over several decades have reported treatment success upon addition of the Aspergillus-active oral triazole agent itraconazole to steroids for ABPA in asthma patients [38–41]. Pacheco et al. [39] showed that itraconazole reduced specific IgG titres in a patient. Germaud and Tuchais [40] showed effectiveness in 11 out of 12 treated patients, including prevention of exacerbations in six patients weaned off oral steroids. Using a before–after methodology, Salez et al. [41] showed reduction in exacerbations and oral steroid doses when patients were started on itraconazole.
The effectiveness of itraconazole in ABPA was demonstrated in two randomised, double-blind, placebo-controlled trials in patients with asthma (both trials excluded patients with CF) [42, 43]. A multicentre trial in 55 patients in the United States by Stevens et al. [42] found more responders in those randomised to receive itraconazole (n=28) for 16 weeks as compared with placebo (n=27). The primary efficacy end-point was a composite measure consisting of at least a 50% reduction in oral steroid dose, at least a 25% reduction in total IgE and at least a 25% improvement in exercise tolerance or resolution of pulmonary infiltrates. Using these criteria, 46% of patients receiving itraconazole responded compared with 19% of those receiving placebo (p=0.04). Additionally, a third of the nonresponders in the placebo-controlled portion of the trial then responded during a subsequent 16-week open-label extension. No relapses occurred in patients receiving itraconazole during the study. In a second trial, conducted at a single centre in the UK, Wark et al. [43] extended these observations in stable asthma patients with ABPA. This trial randomised patients to itraconazole (n=15) or placebo (n=14) for 16 weeks. The study population here was different from that of the study by Stevens et al. [42] in that only one-third of these patients were receiving oral steroids during the trial; all were on inhaled steroids on an average dose of 2000 μg daily, and half were also receiving a daily leukotriene receptor antagonist. In the study by Wark et al. [43], the main outcome indicators evaluated were immunological biomarkers. Patients receiving itraconazole showed normalisation of sputum eosinophilia and eosinophil cationic protein level, and a decrease in serum total IgE level and Aspergillus-specific IgG level. With regard to clinical outcomes, fewer itraconazole-treated patients had exacerbations than patients receiving placebo. These results suggest an anti-inflammatory benefit of itraconazole in ABPA in asthma patients, which may be due to a reduction in fungal burden or perhaps other nonantimicrobial mechanisms. Overall, pooled data from the placebo-controlled trials indicate that itraconazole is effective in ∼60% of asthma-ABPA patients (number needed to treat=3.58; personal communication: D. Denning, National Aspegillosis Centre, University of Manchester, Manchester, UK). The use of azoles for ABPA in asthma patients was reviewed and recommended by the Cochrane collaboration [58, 59].
ABPA occurs in ∼8% of CF patients (meta-analysis 95% confidence interval 6–10%) [60]. Extended courses of oral corticosteroids are considered first-line treatment for ABPA in asthma, and this is also the case in patients with CF [61–65]. As in asthma, itraconazole add-on therapy has been reported to be clinically beneficial in several uncontrolled studies of ABPA in CF patients [38, 44–47]. In these studies, reductions in oral steroid dose and stabilisation of lung function have been found. Nepomuceno et al. [45] also reported a significant decrease in exacerbations compared with a control period [46]. The Cystic Fibrosis Foundation Consensus Conference on ABPA in patients with CF recommended the use of itraconazole as an add-on therapy to oral steroids in patients with slow or poor response to oral steroids, relapse, steroid toxicity or steroid-dependence [65]. A 2000 Cochrane Collaboration review of the use of itraconazole for ABPA in patients with CF cautioned that use was “experimental” in the absence of controlled trials, but the 2012 update concluded that azole therapy is “potentially useful” in CF while in need of further trials with clear outcome measures [66, 67]. Some reports have suggested that itraconazole monotherapy may be a viable alternative to azole add-on treatment after oral glucocorticosteroids, but as yet there is no data from a controlled trial comparing azoles with steroids as monotherapy for ABPA [46, 47, 68–71]. However, recently, a randomised, open-label 3-month trial with a 3-month follow-up in 50 adolescent and adult asthma patients with ABPA comparing itraconazole with oral prednisolone monotherapy (ClinicalTrials.gov identifier NCT01321827), and a similar 50 patient trial comparing voriconazole with prednisolone (ClinicalTrials.gov identifier NCT01621321) have been initiated.
Use of itraconazole is limited by issues of poor absorption and bioavailability, pharmacogenetic variability in cytochrome P450 enzyme-mediated hepatic metabolism and toxicities [72]. Therefore, therapeutic drug monitoring has been recommended [73]. These problems are exaggerated in patients with CF as compared with asthma and, therefore, higher doses of itraconazole capsule or use of the cyclodextrin liquid formulation have been recommended for CF patients [74–76]. It may be difficult to achieve optimal efficacy with itraconazole in CF patients. This is due to its poor bioavailability, but also due to the aggravated absorption defects associated with pancreatic insufficiency, concomitant hepatobiliary disease and small bowel involvement, as well as the requirement for acidic gastric pH to ensure optimal itraconazole absorption being hindered by widespread use of gastric acid-suppressing agents. As itraconazole is highly lipophilic, a suspension in cyclodextrin is 20–50% more bioavailable than the capsule formulation. In order to ameliorate these problems, monitoring of blood levels is recommended in CF whenever therapeutic response is disappointing or there is concern about toxicity [65]. The recommended therapeutic steady-state itraconazole level, based on typical Aspergillus minimal inhibitory concentrations, and clinical studies in a variety of disease states, is 1–5 μg·mL−1, as measured by the most commonly used method, i.e. high-pressure liquid chromatography [73]. It should be noted that recommendations for target dosing for therapeutic efficacy based on monitoring drug levels in the blood have been derived from studies of invasive aspergillosis (where subtherapeutic trough levels have been associated with treatment failure) rather than direct evidence from treating ABPA [77]. Toxicities reported in ≥4% of patients (peripheral neuropathy, fluid retention, gastrointestinal intolerance, elevated hepatic transaminases, rash, headache, tremor and sleep disturbance) have been found with high steady-state triazole levels in patients with chronic pulmonary aspergillosis [72, 78]. In addition, an important drug–drug interaction exists between itraconazole and several corticosteroids, including oral or intravenous methylprednisolone and inhaled budesonide and fluticasone; the azole impairs metabolism of these exogenous glucocorticosteroids resulting in potential adrenal suppression, including overt Cushing syndrome [79–87]. It is, therefore, safer to use oral prednisone or prednisolone (neither of which has these interactions), or perhaps inhaled beclomethasone (which to date has not been shown to have an azole interaction but has also not been systematically studied in this regard), or ciclesonide (a prodrug with topical respiratory metabolism) [88], if using itraconazole or other triazoles in treating ABPA or SAFS.
Newer oral triazoles with excellent anti-Aspergillus activity (voriconazole and posaconazole) have also been reported as beneficial in the treatment of ABPA, particularly in patients with CF [48, 49, 89–91]. In one study, voriconazole was used as monotherapy in 13 CF patients with ABPA; significant and sustained improvements in clinical status, lung function and serologies occurred with prolonged treatment, although nine patients required oral steroids [48]. In another study, 10 out of 11 steroid-dependent patients were able to reduce oral steroid needs while having a marked drop in IgE levels [49]. Voriconazole has the advantage of excellent oral bioavailability. However, voriconazole has strong inhibitory effects on hepatic cytochrome P450 enzymes CYP3A4, CYP2C19 and CYP2C9; making for notoriously unpredictable steady-state levels and drug–drug interactions. This complex metabolism results in much greater inter-individual variability in steady-state voriconazole levels when compared with itraconazole (up to 100-fold range for voriconazole versus 15-fold range for itraconazole), making therapeutic drug monitoring highly advisable [73, 92, 93]. In addition, there appears to be both greater incidence and severity of toxicities with voriconazole when compared with itraconazole [91, 94]. Voriconazole levels of >6 μg·mL−1 are predictive of increased toxicity, including: hepatic; ophthalmological and photosensitive dermatological adverse reactions; and rare, but more serious, cardiac (Torsades de pointes) and neurological events. Voriconazole is also much more expensive. Most recently, posaconazole, a triazole with higher activity against Aspergillus spp. and fewer side-effects than voriconazole, has also been reported to have benefits in treatment of ABPA as well as chronic pulmonary aspergillosis, but it is even more expensive than voriconazole and therapeutic drug monitoring is also advised [50, 95, 96].
The use of azoles for treatment of ABPA has recently been extended to patients with SAFS. Severe asthma inadequately controlled despite combination inhaled corticosteroid/long-acting bronchodilator therapy may be found in up to 20% of asthmatics, and up to half of this large group of difficult patients have atopic fungal sensitivities, most commonly to Aspergillus species but often to multiple fungi. SAFS patients do not meet the necessary constellation of clinical, serological and radiological criteria for a diagnosis of ABPA, usually because total IgE levels are <1000 IU·mL−1 and/or key radiographic findings, such as mucoid impaction or bronchiectasis, are lacking. As noted previously, the link between fungal sensitisation and severe asthma has been increasingly recognised as a significant piece of the larger asthma puzzle [1–9]. Recently this association between fungal sensitisation and the severe asthma phenotype has been further supported by significant correlations between indoor Aspergillus spore air sample concentrations, recovery of fungi (most commonly but not solely A. fumigatus) from respiratory tract cultures and more severe clinical asthma [97, 98]. In one recent paediatric study, 59% of children with severe persistent asthma were found to have fungal sensitisation [99].
After anecdotal experience in adults with SAFS suggested antifungal therapy with itraconazole may be beneficial with reduced hospital admissions and steroid courses, Denning et al. [52] conducted a randomised, double-blind, placebo-controlled trial of itraconazole in 58 patients with SAFS, >40% of whom had been hospitalised within the previous year. The treatment effect on the primary end-point and a validated asthma quality-of-life score was significant. In addition, the rhinitis score, morning peak-flow rates and serum IgE levels were also significantly improved. However, it is important to note that side-effects leading to discontinuation occurred in five out of 29 treated patients and drug–drug interactions resulting in suppression of cortisol levels in half the treated subjects were reported. Similarly to ABPA, ∼60% of SAFS patients were responders to itraconazole (number needed to treat=3.22). In a real-life effectiveness study, outcomes in 22 SAFS patients, as well as 11 asthmatic ABPA patients, treated with open-label itraconazole for at least 6 months were followed [52]. Lung function was improved, while dosage and courses of oral steroids were decreased, and 40% of patients were weaned off oral steroids after 6 months of therapy. Serological measures (total and Aspergillus-specific IgE) and eosinophils were decreased in patients treated for 6–12 months. Recently, similar success in treating SAFS in children with itraconazole has also been reported [30, 53].
While oral azoles, thus far, appear to be an effective component in successful management of ABPA and SAFS, several important caveats exist. These include inter-individual variability in absorption and metabolism, toxicity, drug–drug interactions and cost. It has not yet been clearly demonstrated that the beneficial effects of azoles in ABPA and SAFS are due to their antifungal activity as opposed to alterations in concomitantly administered glucocorticosteroid metabolism or independent anti-inflammatory azole effects [43, 51]. Most troubling, however, is the emerging evidence that increased azole usage for various medical conditions and (at least in some geographical regions) agricultural applications is leading to a higher prevalence of azole resistance in clinical A. fumigatus isolates, most commonly due to point mutations in the cyp51A gene [100–108]. The Aspergillus cyp51A gene encodes cytochrome P450 sterol 14α-demethylase and is the target for azole drugs. Between 5% and 20% of CF patients exposed to recent itraconazole courses were found to be either colonised or infected with azole-resistant A. fumigatus in recent studies, while, in another study, 4% of A. fumigatus respiratory isolates from a variety of different patient groups (including CF, chronic obstructive pulmonary disease, intensive care unit cases and ABPA) were itraconazole-resistant [104, 105, 108]. In some instances, azole cross-resistance has also been documented. Resistance to both itraconazole and voriconazole has been found in patients with ABPA [102, 109].
Alternatives to azoles
In part due to the potential problems of metabolism, tolerance and resistance associated with azole therapy of ABPA and SAFS, further alternative approaches have been investigated utilising both anti-infective and anti-inflammatory target modalities.
Amphotericin B
Anti-infective alternatives to azole therapy for SAFS and ABPA have, thus far, been limited to the use of inhaled formulations of amphotericin B, as topical delivery avoids most systemic toxicity issues. Amphotericin deoxycholate has been used by inhalation to treat pulmonary fungal infection for over half a century, primarily in the settings of cancer treatment or lung transplantation [110]. Unfortunately the literature on inhaled amphotericin is muddied by a plethora of unstandardised and often poorly validated delivery systems and dose regimes; by the availability of multiple formulations (approved for i.v. use) including water-soluble lyophilised amphotericin deoxycholate and three commercially available lipid preparations; and, finally, by a substantial diversity of the diseases for which these agents have been nebulised. A variety of nebulisation devices can deliver amphotericin B particles with good tolerability to the lower respiratory tract in doses capable of exceeding typical Aspergillus minimal inhibitory concentrations in the epithelial lining fluid [111]. Systemic levels are low, reducing the risk of renal and other toxicities. Thus, inhaled amphotericin is a plausible therapy for the chronic or recurrent noninvasive Aspergillus respiratory infection seen in SAFS and ABPA.
However, clinical results for ABPA, while positive, are available for scrutiny in only a few case reports and one small open-labelled series in CF patients [54–57, 112]. Two recent reports utilised amphotericin deoxycholate or liposomal amphotericin in aerosol doses of 10 mg twice daily or 20 mg thrice weekly, respectively, with success [56, 57]. There have been no published reports of inhaled amphotericin use in SAFS. Based on the available literature, it is unclear what the optimal amphotericin formulation, dose, schedule and delivery system should be. Care should be taken to initiate inhaled amphotericin B therapy under observation as cough and bronchospasm may occur, especially with low baseline lung function [113]. An interesting future prospect is the development of a dry powder inhalational formulation and rapid portable delivery system for amphotericin B [114].
Alternative anti-inflammatory approaches to use of oral corticosteroids in ABPA have included the use of high-dose inhaled glucocorticosteroids, i.v. monthly “pulse” high-dose glucocorticosteroids and immunomodulation of the allergic response with omalizumab (humanised monoclonal anti-IgE). Inhaled steroids are already the basic treatment of all severe asthma phenotypes, and none of the other modalities have been reported as yet in SAFS. Leukotriene antagonists have not been evaluated in the treatment of SAFS or ABPA, but would not be expected to provide much benefit given their recommended use for milder asthma phenotypes [115].
Alternative corticosteroid regimes
Inhaled corticosteroids, while useful for concomitant asthma management in patients with ABPA, do not control the pathophysiology or clinical manifestations of ABPA [116–120]. In contrast, “pulse” steroid therapy (10–20 mg·kg−1·day−1 i.v. methylprednisolone infused on three consecutive days every 3–4 weeks) was generally safe and effective in two open-labelled series of 13 steroid-dependent ABPA CF patients selected for this treatment because they were either not well controlled or had severe corticosteroid side-effects on conventional oral prednisone treatment [121, 122]. In most cases, pulse i.v. steroid therapy was well tolerated, with disease control allowing discontinuation of pulse therapy after 6–12 months. However, long-term follow-up data is not available and this published experience is uncontrolled and sparse.
Anti-IgE
Omalizumab, a monoclonal antibody to IgE that prevents allergen-induced IgE-mediated signalling of the classic allergic inflammatory cascade, is licensed in many countries for use in patients with severe allergic asthma [123]. It is increasingly utilised in the treatment of ABPA. While SAFS patients have undoubtedly been included in the many large clinical trials and effectiveness studies of omalizumab that have focused on the approved indication of patients with severe allergic asthma, SAFS is not identifiable as a distinct subgroup for analysis in these studies as selection criteria in most trials, and in subgroup analyses when reported, focused on allergic sensitisation to perennial indoor allergens, i.e. dust mite, cockroach and cat or dog dander [124, 125]. The rationale and pharmacodynamic issues involved in the use of omalizumab for ABPA are reviewed elsewhere [126]. Omalizumab-treated ABPA patients reported in the literature have generally responded well, with reduced exacerbation rates, decreased oral steroid exposure and decreased steroid toxicity being the three major observed benefits of therapy [127–142]. For example, two open-labelled series from Spain and France (34 subjects when pooled, including two with CF-ABPA) showed significant reductions in exacerbations and oral steroid doses [137, 138]. However, a recent multicentre open-labelled retrospective series from France found variable results over an average 21-month observation period in 32 CF-ABPA patients on omalizumab, with a reduction in steroid need but no change in lung function or use of i.v. antibiotics [143].
As yet, no placebo-controlled trials of omalizumab in ABPA or SAFS have been completed, leading to a call by the Cochrane Collaboration for completion of a randomised controlled trial [144]. A multicentre, randomised, double-blind, placebo-controlled, 6-month trial with a 6-month open-labelled extension in CF-ABPA patients aged ≥12 years concomitantly treated with prednisone (on a prescribed tapering regime) and itraconazole 400 mg twice daily was initiated in Europe in 2008. It had rescue oral steroid use as the primary outcome (ClinicalTrials.gov identifier NCT00787917). The study was terminated by the sponsor, Novartis Pharmaceuticals, in 2011 after enrolment of only 14 subjects (mean±sd age 23±7 years). Both enrolment and dropouts apparently were impacted by an arduous study design that included daily subcutaneous injections of omalizumab at doses of up to 600 mg or placebo. Of nine subjects randomised to omalizumab, only four completed the 6-month placebo-controlled trial; discontinuations were attributed to an adverse event in one, lack of efficacy in one and “administrative problems” in three, and two out of five subjects randomised to placebo also dropped out due to “administrative problems.” Of the seven subjects going on to the 6-month open-labelled extension, only three completed it, with dropouts attributed to unsatisfactory therapeutic effect (n=1) and “administrative problems” (n=3). Crucially, this failed trial utilised omalizumab in a much more intensive and intrusive regime than the way it is marketed and clinically used for allergic asthma (i.e. maximum dose 375 mg every 2 weeks), leaving open the question of whether a similar “real world” design (as in the published off-labelled case reports and series) might not be a more feasible and potentially successful way to examine efficacy in a controlled trial.
The package insert dosing table for omalizumab treatment of asthma, which caps recommended maximal dosing at 375 mg every 2 weeks, encompasses a baseline (free) serum total IgE range of 30–700 IU·mL−1 and a bodyweight range of 20–150 kg. The dosing table is based on clinical trial doses targeted to reduce free IgE levels in blood to <25 IU·mL−1 in ≥95% of recipients meeting the baseline IgE level range requirements [145]. However, the dosing table recommendations generally correspond well to a published formula of 0.016 mg·kg−1·IgE−1 (in IU·mL−1) monthly that is based on calculations of dose required to bind >90% free IgE in vitro [146, 147]. As patients with ABPA by definition have baseline IgE levels exceeding the dosing table upper limit (>1000 versus 700 IU·mL−1, respectively), and many SAFS patients may also have IgE levels above the dosing table range, the apparent clinical efficacy in the literature, generally using doses at or only modestly greater than the dosing table, suggests that dosing for ABPA and SAFS is not substantially greater than that currently recommended for patients with lower baseline IgE levels and may suffice for clinical benefit. Recently, two CF-ABPA patients with baseline IgE levels of 1039 and 1782 IU·mL−1 were treated on the basis of the formula, resulting in omalizumab regimes of 450 mg monthly in the first patient and 450 mg every 2 weeks in the second. They had reductions in free IgE of 88% and 96% after 6 and 3 months treatment, respectively, with corresponding marked clinical improvement [148]. Altogether the clinical experience suggests that omalizumab treatment is rational in patients with SAFS or ABPA, despite IgE >700 IU, especially if the formula, rather than dosing table, is utilised to calculate an optimal dose and interval adjusted for tolerability. In a recent alternative strategy of interest, a recent case report of omalizumab therapy in a patient with SAFS suggested that using the immunological effect of azole therapy to reduce the baseline IgE value may help establish an omalizumab regime within the dosing table [43, 149].
An important difficulty in properly evaluating any emerging therapy for ABPA is the potential differential response to therapeutic interventions in patients with different degrees of structural lung damage. Only one published trial, stratified or otherwise, distinguished ABPA patients without bronchiectasis (“ABPA-serologic”) from those with bronchiectasis [120, 150]; those with hyperattenuating mucoid impaction have not been compared with those without impaction, despite apparent differences in severity of immunopathology and prognosis [13, 151].
In conclusion, active noninvasive endobronchial fungal infection is likely to play an important role in a significant segment of asthma and CF patients with more severe pulmonary pathology and illness. Therapies aimed at lowering the fungal burden and at down-regulating host allergic immune response show indications of efficacy, which are supported by the improved understanding of SAFS and ABPA pathophysiology, distinct nosological entities along a spectrum of asthma that is induced by fungal infection and Th2-biased immune response. Their role in the overall management of these patients remains to be determined, hopefully by controlled trials, where such evidence is lacking, and by comparative effectiveness studies comparing conventional with alternative treatments and alternative treatments with each other.
Acknowledgments
The author thanks J. Wine (Cystic Fibrosis Research Laboratory, Stanford University, Stanford, CA, USA) for kindly providing the photomicrograph shown in the figure 2.
Footnotes
Support statement: Work described herein was funded by a research grant from Genentech, Inc. (San Francisco, CA, USA) (grant number Genentech C4-150174), the manufacturer of omalizumab (Xolair).
Conflict of interest: Disclosures can be found alongside the online version of this article at www.erj.ersjournals.com
- Received August 10, 2013.
- Accepted November 20, 2013.
- ©ERS 2014
References
