Do inhaled corticosteroids reduce serum IgE levels? The answer is maybe but how relevant is the question?

Two years ago saw the thirtieth anniversary of the discovery of IgE [ 1, 2] and its identification as the reaginic antibody first described by Prausnitz and Kustner [ 3]. It has since become increasingly apparent, through epidemiological and immunological studies, that IgE plays a central role in the pathogenesis of atopic disease. The finding therefore that inhaled beclamethasone dipropionate (BDP) reduces serum IgE levels in atopic asthma is interesting, and its relevance deserves critical examination. A number of epidemiological studies have demonstrated a link between circulating levels of IgE and atopic disease. Burrows et al. [ 4] measured circulating levels of IgE in a population of 2657 randomly selected subjects. The prevalence of self-reported asthma, as determined by questionnaire answers, showed a highly significant association with serum IgE levels with asthma prevalence rates increasing linearly with log IgE values. In a later study Sears et al. [ 5] showed an association between bronchial hyperresponsiveness to methacholine and serum IgE levels in a population of 1037 children. This association remained significant even when children with diagnosed asthma were excluded, and was also significant in the smaller subgroup of totally asymptomatic children. In this study only one child with a serum IgE level of less than 32 IU/mL had a PC20 of less than 8 mg/mL whereas 28% of children with IgE levels greater than 1000 IU/mL had a PC20 of less than 8 mg/mL. IgE plays a central role in the immediate hypersensitivity-type responses to allergen, such as the early phase of airway obstruction that follows experimental allergen challenge. Allergen-specific IgE binds to the α chain of the high affinity receptor (FcɛR1) which is expressed by mast cells and basophils, as well as other cell types [ 6]. In asthma, cross-linking of cell-bound IgE by inhaled multivalent allergens leads to aggregation of receptors and release of preformed and newly generated vasoactive and bronchoconstrictive mediators such as histamine, neutral proteases, prostaglandins and cysteinyl leucotrienes. It is now recognized that IgE-mediated mast cell activity may also play an important role in the late phase reaction. Immunoreactivity for the cytokines, interleukin (IL)-4, IL-5, IL-6 and TNF-α, has been demonstrated in mast cells from nasal [ 7] and bronchial biopsies [ 8] from both normal and asthmatic/rhinitic subjects, and the presence of IL-3 and GM-CSF has also been reported [ 9]. Purified mast cells from the normal human lung have been shown to release IL-4 in response to IgE-dependent cross-linking of FcɛR1, and TNF-α has been shown to be released in acute rhinitis [ 10]. Less is known about the effect of IgE cross-linking on the other mast cell cytokines, although it is possible that they too are released following mast cell stimulation. All these cytokines have pro-inflammatory actions, causing an influx into the airways of eosinophils, mast cells, and activated T cells, and leading to their increased maturation, survival and activation. Such an action would attribute an important role to IgE-mediated activation of the mast cell in the inflammatory process underlying the late phase response, and hence the bronchial hyperreactivity that characterizes asthma. There is now evidence that IgE may also act independently of the mast cell and basophil to increase T-cell responsiveness. Antigen–IgE complexes bound to the low affinity FcɛR2 receptor, and possibly the FcɛR1 receptor recently identified on antigen-presenting cells [ 11, 12], have been shown to amplify the T-cell response to allergen by facilitating antigen presentation to antigen-specific T cells [ 13]. In an elegant series of experiments, Coyle et al. [ 14] showed that treatment of mice with a monoclonal anti-IgE antibody prior to allergen challenge inhibited both recruitment of eosinophils into the lung and production of the TH2-type cytokines IL-4 and IL-5. This was independent of the effect of monoclonal antibodies on circulating IgE and was not mast cell dependent, since the effect was also seen in mast cell-deficient mice. Administration of a monoclonal antibody directed against the low affinity receptor for IgE had a similar inhibitory effect on eosinophil infiltration, an effect that was not potentiated by further administration of the monoclonal anti-IgE. This would suggest that IgE, acting through the low affinity receptor, plays an important role in the induction of a TH2-type cytokine response and forms part of a positive feedback loop, leading to further IgE production as well as eosinophil recruitment. It would seem, therefore, that IgE contributes to the early and late airway response both through its action on the mast cell-bound high affinity IgE receptor and by augmenting allergen presentation to T cells via the low affinity IgE receptor. The finding that inhaled steroids can act to reduce circulating IgE levels is thus an important one. However, is this finding real, and could this effect contribute to the therapeutic effect of inhaled steroids? In this issue of the Journal, Ohrui et al. [ 15] report that 3 months treatment with 800 μg per day of inhaled BDP led to improvement of symptoms and a significant reduction of circulating total and allergen-specific IgE levels in a group of atopic asthmatics. This was a well-designed double-blind study. It included a placebo group to allow for seasonal variations in IgE levels, and the study was performed outside the allergen season. The study group was small, but this is more likely to cause type II rather than type I errors. Although not stated in the paper, it would appear that there was no statistically significant difference between the baseline characteristics of the two groups and, encouragingly, the decrease in serum IgE was consistent for all members of the steroid-treated group. Furthermore, previous studies have demonstrated a similar effect. Naclerio et al. [ 16] showed that intranasal steroids inhibited the seasonal increase in serum ragweed-specific IgE in ragweed-sensitive allergic rhinitics, and Settipane, Pudupakkam and McGowan [ 17] showed that following an initial increase, systemic administration of corticosteroids lead to a significant reduction in circulating IgE levels. The authors cite other studies where steroids have had no demonstrable effect, or have caused increases in circulating IgE levels. Kerstjens et al. [ 18, 19] showed no effect of 2.5 years of inhaled corticosteroid treatment on serum IgE levels but their study included nonatopic patients and patients with chronic obstructive pulmonary disease. Zieg et al. [ 20] demonstrated increases in systemic IgE production following corticosteroid therapy. Ohrui et al. cited differences in route of administration as a possible cause of these discrepancies. Equally plausible would be differences in timing of the samples. Zieg et al. only reassessed IgE levels at 1 week, whereas in the study of Settipane et al. [ 17] systemic corticosteroids caused an initial increase in serum IgE levels lasting for 2 weeks before they began to fall. Although this transient rise in IgE is difficult to explain, it is consistent with the present study in which IgE levels were only measured after 3 months. An issue not discussed by the authors is the discrepancy between this in vivo study and previous in vitro work that has suggested that corticosteroids may act to increase IgE production. Jabara et al. [ 21] showed that a combination of hydrocortisone and rIL-4 could induce IgE synthesis in highly purified cultures of normal peripheral blood B cells. Other workers have shown that hydrocortisone can potentiate IL-4-induced IgE synthesis by peripheral blood mononuclear cells [ 22, 23]. We have demonstrated an analogous situation with salbutamol where findings in vitro — which suggested that salbutamol potentiated IL-4-induced production of IgE [ 24, 25] — could not be confirmed in vivo [ 26]. This discrepancy probably reflects the much more complex nature of the in vivo situation. In vitro models of IgE synthesis employ peripheral blood mononuclear cells stimulated by IL-4, whereas the in vivo situation involves allergen penetration of the airway mucosa, processing and presentation of allergen by antigen presenting cells and the involvement of other cytokines that potentiate IgE production, such as IL-6 and IL-13. It is therefore unsurprising that previous in vitro observations were not, in the current study reproduced in vivo. Could a reduction in circulating IgE be part of the mechanism by which steroids exert their clinical effects? Certainly the authors show that clinical improvement, as measured by asthma symptom scores, accompanied the fall in IgE levels and, indeed, demonstrated a significant correlation between these two outcome measures. Sensibly though, the authors do not jump to this conclusion. Treatments aimed at reducing circulating IgE using specific monoclonal antibodies have been tested in both atopic asthmatics and rhinitic subjects and have been shown to lead to reduction in symptoms and bronchial hyperreactivity [ 272829]. However, these treatments led to the elimination of measurable IgE both locally and systemically. In contrast, in the current study, IgE levels fell from a geometric mean of 2.94 log IU/mL at the start of treatment to 2.64 log IU/mL at completion. Although this is a statistically significant fall and represents a halving of serum total IgE (from 871 IU/mL to 436 IU/mL), the levels of IgE at the end of the study remained high. It is unlikely that such a fall would have any effect on IgE tightly bound to the high affinity receptor on mast cells and basophils, since even elimination of circulating IgE has not been shown to reduce basophil histamine releasablility [ 27]. It is conceivable that there may be some effect on IgE bound to the low affinity receptor, although it is difficult to speculate on the clinical relevance of this. Certainly in the population studies described previously, levels of IgE of around 400 IU/mL were compatible with symptomatic asthma, and indeed around 40% of adult asthmatics were found to have levels below 100 IU/mL [ 4]. It may be that local airway expression of IgE is being affected. In the present study inhaled corticosteroids were administered directly into the airways, yet systemic levels of IgE were being measured. Some authors have suggested that local production of IgE may be important [ 30] and a more dramatic effect on local levels of IgE may be being missed. Indeed studies using sodium cromoglycate have shown that it is possible to decrease IgE levels in bronchoalveolar lavage (BAL) without effecting serum levels [ 31]. If a larger study is being contemplated then measurement of IgE in BAL would be a useful addition. An alternative and more likely explanation is that the fall in serum IgE simply reflects other actions of corticosteroids. Serum IgE levels are known to fall when exposure to allergen is reduced [ 32]. Thus, any treatment that reduces airway inflammation and so reduces airway permeability to allergen is likely to cause a reduction in serum IgE levels. Significant improvements in airway morphology are known to follow even short courses of BDP treatment [ 33]. Corticosteroids are also known to have wide-ranging effects on cell cytokine production, through an action on glucocorticoid receptors which act to regulate gene transcription [ 34], perhaps by binding to transcription factors such as NF-κB [ 35]. Administration of corticosteroids has been shown to inhibit the transcription of a large number of cytokines such as interleukin (IL)-1, IL-3, IL-4, IL-5, IL-6, IL-8, TNF-α and GM-CSF [ 36]. Reduction in the levels of these cytokines will attenuate the recruitment and activation of inflammatory cells within the airways but will also have important effects on synthesis of IgE, mainly due to a reduction in levels of IL-4 and perhaps IL-6 and TNF-α. It is likely, therefore, that reductions in serum IgE will parallel the other anti-inflammatory effects of cortico-steroids. If the improvement in symptoms is due to these effects they may still show a positive correlation with the effect on circulating IgE levels. Finally, it is possible that the effect on IgE levels is due to changes in the use of other drugs. Unfortunately, it is not clear from the paper whether volunteers in the BDP group continued to take β2-agonists on a regular or a required basis. If the latter was the case then the improved symptom control resulting from BDP use could lead to reduced levels of β2-agonist use. It has been suggested by some that administration of β2-agonists may upregulate IgE production and so a reduction in their use could lead to a fall in serum levels [ 24, 37]. This remains a possible explanation although, as described above, we have failed to demonstrate any such effect with systemic administration of salbutamol [ 26]. The authors have demonstrated in the clinical situation that 3-months' treatment with inhaled corticosteroids leads to a reduction in serum IgE. A larger study, with more controlled use of β2-agonists and measurement BAL IgE levels needs to be performed to confirm and extend this finding. It is encouraging that in vitro findings, showing that steroids potentiated IgE synthesis, have not been confirmed. It is, however, unlikely that this is an effect that contributes to the clinical efficacy of this commonly prescribed group of compounds.