Is low dose inhaled corticosteroid therapy as effective for inflammation and remodeling in asthma? A randomized, parallel group study

Corticosteroids are the most effective medication for the prevention and treatment of asthmatic inflammation. Many dose-ranging studies have been conducted to examine the dose-response relationship of ICSs in the treatment of asthma in adults and children [1‐17]. The adverse effects of inhaled corticosteroids (ICSs) have been clearly shown to be dose-related in both adults and children in numerous studies [18‐25]. Although it has been suggested that most of the clinical benefits of ICSs occur at low doses [6, 26], the results of dose-ranging studies are inconsistent. Some data show no dose-response for FEV1 or asthma symptoms [8‐11, 13, 27‐29], for airway hyperresponsiveness [7, 14, 15] or exhaled nitric oxide [15], while other data show a significant dose-response for clinical outcomes [2‐5, 7, 9, 12, 16, 17, 30, 31].

Evidence for equivalent effects on underlying airway inflammation and remodeling of low dose and high dose ICS therapy is lacking. In vitro, the degree of suppression of some inflammatory cytokines and the degree of upregulation of anti-inflammatory cytokines is dependent on the dose of corticosteroid [32, 33]. However, there are no in vivo randomized studies comparing the effects of low and high dose ICS on production of clinically relevant cytokines.

Alveolar macrophages (AMs) normally represent the majority of cells isolated by bronchoalveolar lavage (BAL) and express allergen-specific low affinity immunoglobulin E (IgE) receptors on their surface [34] which, when activated, release pro-inflammatory mediators and anti-inflammatory regulators including chemokines (IL-8, MCP-1, MIP-1α), anti-inflammatory cytokines (IL-1ra and IL-10) and pro-inflammatory cytokines (TNF-α, GM-CSF, IL-6 and IL-1β) [35]. Endobronchial allergen challenge significantly increases expression of activation markers on AMs and correlates with AHR to methacholine [36]. The pro-inflammatory cytokine granulocyte macrophage-colony stimulating factor (GM-CSF), secreted by AMs, promotes myeloid cell differentiation, stimulates antigen presenting cells, prolongs eosinophil survival [37] and is elevated in BAL and bronchial biopsies in asthma and after allergen challenge [38‐41]. Neutralisation by GM-CSF antibodies in mice abolishes ovalbumin-induced AHR and mucus cell hyperplasia [42]. In vitro incubation of peripheral blood mononuclear cell cultures with beclomethasone dipropionate (BDP) or fluticasone propionate (FP) dose-dependently inhibits allergen-induced GM-CSF in atopic asthmatics [32]. Tumour necrosis factor-alpha (TNF-α) is another pro-inflammatory mediator secreted by AMs which is increased in symptomatic asthmatics and enhanced by allergen challenge. In vitro incubation with TNF-α increases human airway smooth muscle responsiveness [43]. Inhaled recombinant TNF-α increases AHR to methacholine [44]. In vitro incubation of alveolar macrophages with fluticasone or budesonide results in dose-dependent inhibition of lipopolysaccharide (LPS)-induced TNF-α protein secretion [33].

AMs secrete the anti-inflammatory cytokine interleukin-10 (IL-10) which inhibits production of many pro-inflammatory cytokines [45], suppresses allergic inflammation by inhibiting T helper 2 cell cytokines [46], shortens eosinophil survival and attenuates induction of IgE synthesis by IL-4 [47]. Studies of the effects of corticosteroid treatment on IL-10 are mixed. IL-10 expression is greater during budesonide treatment compared with placebo [48] and serum IL-10 increases after triamcinolone [49] while inhibition of IL-10 by corticosteroid treatment can also occur [50].

Another anti-inflammatory cytokine secreted by AMs is interleukin-1 receptor antagonist (IL-1ra) which is an endogenous means of protection against inflammatory responses in diseases such as asthma and rheumatoid arthritis by inhibiting the pro-inflammatory mediator interleukin-1beta (IL-1β). More macrophages express IL-1β in asthma [51] and IL-1ra serum levels increase during asthma exacerbations [52]. In animal studies, aerosol IL-1ra prior to histamine challenge prevents bronchial hyperresponsiveness [53]. Inhaled BDP in asthmatic subjects inhibits epithelial expression of IL-1β without inhibition of IL-1ra, favouring an anti-inflammatory shift in the ratio of IL-1β/IL-1ra [54].

There is also some evidence for dose related differences in the effects of corticosteroids on bronchial mucosal basement membrane thickness (BMT) [55, 56] but whether low dose ICS therapy, while sufficient to control asthma symptoms, is also adequate to improve airway remodeling is unknown.

The null hypothesis of this study is that low dose ICS therapy is as efficacious as high dose therapy, for asthma control, airway inflammation and remodeling in adults. Our aim was to examine differences in response to high and low dose ICS, in asthma control, airway inflammation, AHR and airway remodeling in mild to moderate adult asthmatics who were previously corticosteroid free for over two months.

In this study we found that seven weeks of low dose ICS treatment in adults with mild to moderate asthma, who were corticosteroid free for over 2 months, is as effective in controlling airway inflammation as high dose therapy, measured by constitutive mRNA expression of GM-CSF, TNF-α and constitutive or LPS induced IL-1ra mRNA and protein in BAL. Low and high doses were also equally effective in controlling asthma in terms of symptoms, spirometry, eNO and AHR after 6 weeks, which is consistent with previous data [6, 26]. We found no significant effects on airway remodeling by either dose after 7 weeks, as measured by change in basement membrane thickness.

There have been previous ICS dose-ranging trials which have not found differential clinical effects between dose arms. Meta-analyses have revealed plateaus in the ICS dose-response curve beginning at 100-200 μg of FP per day with a peak benefit at 500 μg/day [11] or most of the clinical efficacy attained with an FP dose of 200 μg/day [6] or 300-600 μg/day of budesonide (BUD) [26]. However, control of underlying inflammation and remodeling was not assessed and individual patients with more severe asthma may require higher doses to achieve the corresponding maximal effects. Previously published evidence suggests that AHR, which is loosely related to inflammation, may be more likely to show ICS dose-dependence than FEV1, PEF or symptoms [9, 12, 31, 61] and a dose-dependent rate of reduction of eNO and asthma symptoms has been observed [7, 16, 17]. In a meta-analysis of sixteen trials using multiple drug formulations (FP, triamcinolone, BUD and mometasone), a statistically significant inhaled corticosteroid dose-response relationship was found for morning PEF and a dose-response relation for FEV1 was found for BUD [2]. Asthma symptom score was also differentially affected by the dose of FP or triamcinolone, in contrast to the present study finding of no difference in change in Juniper score between the two doses. This absence of difference in the present study may be due to a milder degree of asthma and smaller sample size.

Despite the small sample size in a study by Chetta et al. [56] (8 in each group), there were significant differences in a number of in vitro inflammatory indices between low and high dose inhaled steroid treatment. High dose treatment resulted in a significant decrease in basement membrane thickness, number of vessels, vascular area and mean vessel size whereas no change was observed with low dose. Like the current study findings, there were no significant differences in the improvements in AHR and asthma symptom scores between the groups. The lack of difference between high and low dose ICS effect on inflammatory markers in the present study, compared with the findings of Chetta et al. [56] could be due to milder asthma in our subjects since none had previously taken long term ICS regularly. This is supported by the correlation observed between baseline severity and magnitude of improvement in outcomes.

Modulation of alveolar macrophage inflammatory responses is an important mechanism underlying ICS efficacy in asthma [34‐36, 62‐65]. There are no published randomized studies examining the dose-response relationship of the effect of ICSs on pro- and anti-inflammatory cytokine production in asthma. In the present study inhaled FP significantly reduced mRNA expression of the pro-inflammatory cytokines GM-CSF and TNF-α. The change in GM-CSF protein levels after FP treatment was smaller than the estimated effect size of 1500 pg/mL. This estimate was derived from a study which had assessed the effect of inhaled budesonide (800 μg twice daily) in comparison to placebo [48], given the lack of available data quantifying the dose-dependent effects of inhaled corticosteroids on airway cytokines. A smaller effect size would be expected between two active treatment arms of different doses than between active treatment and placebo but, while this difference may be considered clinically significant, a greater number of participants in each group would have been required to show statistical significance. The effect size from two previous studies which measured alveolar macrophage TNF-α protein reduction post inhaled flunisolide [60] (in the order of 50 pg/mL) differed greatly compared with post inhaled beclomethasone [66] (in the order of 140,000 pg/mL). Therefore, although TNF-α is an important biological endpoint in this study, the effect of inhaled steroid treatment on TNF-α may depend on factors that are, as yet, undefined making the lack of significant effect of FP on TNF- α protein in the present study difficult to interpret. Inhaled or systemic corticosteroids suppress inflammatory cytokines in vitro including GM-CSF or TNF-α [40, 48, 66‐73]. Previous in vitro studies have shown that inhibition of pro-inflammatory cytokines by corticosteroids is dose-dependent [32, 33] and it was postulated that the in vivo effects of treatment on the same inflammatory markers would also be dose-dependent. The present study, however, found no significant difference between low and high dose inhaled FP groups in the change in quantity of constitutive and LPS induced cytokine mRNA and protein from alveolar macrophages.

The anti-inflammatory cytokine IL-1ra mRNA expression and IL-1ra protein was also significantly suppressed by inhaled FP in the present study. Previous studies examining the effects of ICSs on IL-1ra show conflicting results ranging from no effect [54, 74] to elevation of IL-1ra levels post treatment [52, 74]. In the current study, the finding of a reduction in IL-1ra mRNA and protein after treatment with FP was unexpected considering this cytokine has anti-inflammatory properties. Epithelial expression of IL-1β is significantly inhibited by inhaled BDP in asthmatic subjects without any significant concurrent inhibition of IL-1ra thus favouring an anti-inflammatory shift in the ratio of IL-1β/IL-1ra [54]. It is possible that the pro-inflammatory cytokine IL-1β, which was not measured in this study, may have been attenuated by FP to a greater extent than was IL-1ra, thus resulting in a lower, more anti-inflammatory ratio of IL-1β to IL-1ra levels.

The present study found no significant changes in the anti-inflammatory cytokine IL-10 gene expression or protein secretion resulting from treatment with FP. The small differences in IL-10 protein levels detected after FP treatment were substantially smaller than the 260 pg/mL reported previously. The levels of IL-10 were close to the lower limit of detection in many subjects making resolution of changes impossible. Previous studies have revealed mixed effects of corticosteroids on IL-10, including no effect of in vitro dexamethasone [75], elevation after high dose inhaled budesonide [48] or inhaled triamcinolone [49] and inhibition in nonasthmatics [50].

Basement membrane thickness in asthma is reduced by corticosteroid treatment in some studies [73, 76‐80]. Although dose related differences in the effects of corticosteroids on bronchial mucosal BMT are apparent in some studies [55, 56], the current study did not demonstrate a significant reduction in BMT after FP treatment for 7 weeks. This is consistent with the findings of Boulet et al. for high dose FP [81]. Although two previous studies have shown a significant effect of short term treatment (six weeks) with medium or high dose inhaled corticosteroid on BMT [56, 76], longer treatment duration (three to twelve months) is more likely to alter BMT [73, 77, 78, 82‐84].

An important feature of current asthma management guidelines is the use of a maintenance dose of ICS which is the minimum effective dose to achieve asthma symptom control to limit the potential for adverse effects. Current therapy guidelines also advocate alternative approaches to achieve asthma control without increasing the ICS dose. The addition of a long-acting beta agonist, leukotriene antagonist and/or low dose theophylline represent examples of this approach. At least on a short term basis, our results suggest that in stable mild or moderate adult asthmatic patients, who are corticosteroid free for over 2 months, an FP equivalent dose of 200 μg/day may be just as effective as 1000 μg/day in controlling cytokine driven inflammation. Combining the different anti-asthmatic mechanisms of other therapies with the attenuation of cytokine driven inflammation by ICS might be a superior and safer approach to achieving asthma control.

It is possible that differences in inflammation may occur with longer term treatment and it would be important to conduct longitudinal studies with larger sample size to re-evaluate the dose-response relationship of a wide repertoire of inflammatory mediators, along with the probably smaller influence of dose on remodeling and fibrotic markers. Additionally, prevention of asthma exacerbations and decline in lung function would be important endpoints to assess in a long term study. It would also be useful to assess any differential dose effects of combination inhaled corticosteroid/long acting beta-agonist maintenance treatment, to firmly establish whether such low dose therapy optimally suppresses underlying inflammation and prevents long term airway remodeling and not just short term symptoms.