New Developments |
 |
INTRODUCTION |
|---|
TOP
INTRODUCTION
REFERENCES
|
|---|
Inhaled corticosteroids have revolutionized the treatment of asthma and have now become the mainstay of therapy for patients with chronic disease (1). In a supplement to this journal published 5 yr ago the currently available data on efficacy and safety of inhaled corticosteroids was reviewed (2). Since then there have been many important developments in our understanding about how inhaled corticosteroids work in asthma, and much more data on the efficacy and safety of inhaled corticosteroids have been published. In addition, there are now many more studies comparing different inhaled corticosteroids, so that inhaled corticosteroids are now among the most carefully studied drugs in clinical use. The new information about inhaled corticosteroids has reflected an enormous increase in their prescription for use in asthma, including earlier introduction in adults and children. In order to discuss and evaluate this new information, we organized an international meeting of leading investigators in this field. Their contributions to the meeting and the ensuing discussions have formed the basis for this review.
This supplement concentrates on developments occurring since the previous review in 1993. Where necessary, we have included some older studies in order to provide a balanced and comprehensive picture.
| |
MOLECULAR MECHANISMS |
|---|
There has recently been an enormous increase in our understanding of the molecular mechanisms whereby glucocorticoids suppress inflammation in asthma. This has shed new light on the molecular mechanisms of asthma and may point the way to the development of more specific therapies in the future (3, 4).
Glucocorticoid Receptors
Corticosteroids exert their effects by binding to glucocorticoid receptors (GRs), which are localized to the cytoplasm of target cells. The affinity of cortisol binding to GR is approximately 30 nM, which falls within the normal range for plasma concentrations of free hormone. There is a single class of GR that binds corticosteroids, with no evidence for subtypes of differing affinity in different tissues. Glucocorticoid receptors are widely distributed within human lung; immunocytochemical localization studies and in situ hybridization indicate the greatest levels of expression in airway epithelial cells and bronchial vascular endothelial cells (5).
Recently a splice variant of GR, termed GR-
, has been
identified that does not bind corticosteroids, but binds to
DNA, and may therefore interfere with the action of steroids
(6). The structure of GR has been elucidated using site-directed
mutagenesis, which has revealed distinct domains (7, 8). The
glucocorticoid binding domain is at the C-terminal end of the
molecule, and in the middle of the molecule are two finger-like projections that interact with DNA. Each of these "zinc
fingers" is formed by a zinc molecule bound to four cysteine
residues (Figure 1). An N-terminal domain (
1) is involved in
transcriptional trans-activation of genes once binding to DNA
has occurred, and this region may also be involved in binding
to other transcription factors (9). This is the least conserved
part of the molecule. Deletion analysis has demonstrated a 41 amino acid core at the C-terminal end of the 1 domain that is
critical for trans-activation. In human GR there is another
trans-activating domain (2) adjacent to the steroid-binding
domain, and this region is also important for the nuclear translocation of the receptor. Glucocorticoid receptor is phosphorylated (predominantly on serine residues at the N-terminal), but the role of phosphorylation in steroid actions is not yet certain (10).
|
The inactivated GR is bound to a protein complex (~ 300 kilodaltons [kD]) that includes two molecules of 90 kD heat shock protein (hsp 90) and a 59 kD immunophilin protein and various other inhibitory proteins. The hsp 90 molecules act as a "molecular chaperone," preventing the unoccupied GR from localizing to the nuclear compartment. Once the glucocorticoid binds to GR, hsp 90 dissociates, exposing two nuclear localization signals and allowing the nuclear localization of the activated GR-steroid complex and its binding to DNA (Figure 2).
|
Effects on Gene Transcription
Corticosteroids produce their effect on responsive cells by activating GR to directly or indirectly regulate the transcription of certain target genes (11). The number of genes per cell directly regulated by steroids is estimated to be between 10 and 100 in any particular cell, but many genes are indirectly regulated through interaction with other transcription factors, as discussed below. Upon activation, GR forms a dimer that binds to DNA at consensus sites termed glucocorticoid response elements (GREs) in the 5'-upstream promoter region of steroid-responsive genes. This interaction changes the rate of transcription, resulting in either induction or repression of the gene. The consensus sequence for GRE binding is the palindromic 15-base pair sequence GGTACAnnnTGTTCT (where n is any nucleotide), although for repression of transcription the putative negative GRE (nGRE) has a more variable sequence (ATYACnnTnTGATCn). Crystallographic studies indicate that the zinc finger binding to DNA occurs within the major groove of DNA, with each finger interacting with one-half of the palindrome. In contrast to these simple GREs, there are composite GREs that do not share these GRE sequences but depend on the presence of other transcription factors binding to DNA (14). Interaction with other transcription factors may also be important in determining differential steroid responsiveness in various cell types. Other transcription factors binding in the vicinity of GRE may have a powerful influence on steroid inducibility, and the relative abundance of different transcription factors may contribute to the steroid responsiveness of a particular cell type. Glucocorticoid receptors may also inhibit protein synthesis by reducing the stability of mRNA via enhanced transcription of specific ribonucleases that break down mRNA containing constitutive AU-rich sequences in the untranslated 3'-region, thus shortening the turnover time of mRNA. There is increasing recognition that corticosteroids may also affect the translation of proteins.
Interaction with Transcription Factors
Glucocorticoid receptors may interact directly with other
transcription factors, which bind to each other via so-called
leucine zipper interactions (15, 16). This could be an important determinant of steroid responsiveness and is a key mechanism whereby corticosteroids exert their anti-inflammatory
actions (3). This interaction was first demonstrated for the collagenase gene, which is induced by the transcription factor activator protein-1 (AP-1), which is a heterodimer of Fos and
Jun oncoproteins. AP-1 binds to a specific DNA binding site
(TRE or TPA response element, TGACTCA). Steroids are
potent inhibitors of collagenase gene transcription induced by
tumor necrosis factor-
(TNF-) and phorbol esters, which activate AP-1. AP-1 forms a protein-protein complex with activated glucocorticoid receptors, and this prevents glucocorticoid receptors from interacting with DNA and thereby reduces
steroid responsiveness (17).
In human lung, TNF- and phorbol esters increase AP-1
binding to DNA; this is inhibited by corticosteroids (20).
Glucocorticoid receptors may interact with other transcription
factors that are activated by inflammatory signals, including
nuclear factor-kappa B (NF-
B) in a similar manner (20, 21,
23) (Figure 3). There is increasing evidence that NF-B
may play a pivotal role in the orchestration of chronic inflammatory diseases, including asthma (26, 27). Many of the stimuli that lead to an increase in airway inflammation activate
NF-B, and this transcription factor induces many of the inflammatory genes that are abnormally expressed in asthma,
including genes for proinflammatory cytokines that amplify
inflammation, chemokines involved in recruitment of eosinophils, inflammatory enzymes that synthesize mediators, adhesion molecules involved in the recruitment of eosinophils, and
inflammatory receptors (Figure 4). By inhibiting NF-B, corticosteroids would therefore inhibit many aspects of the inflammatory process in asthma.
|
|
There is also evidence that 2-agonists, via cyclic AMP formation and activation of protein kinase A, result in the activation of the transcription factor CREB that binds to a cyclic
AMP responsive element (CRE) on genes. A direct interaction between CREB and GR has been demonstrated (28). -Agonists increase CRE binding in human lung and epithelial
cells in vitro and at the same time reduce GRE binding, suggesting that there may be a protein-protein interaction between CREB and GR within the nucleus (29, 30). These interactions between activated GR and transcription factors occur
within the nucleus, but recent observations suggest that these
protein-protein interactions may also occur in the cytoplasm (31).
Evidence suggests that several transcription factors, including GRs, interact with large co-activator molecules, such as CREB-binding protein (CBP) and the related p300, which bind to the basal transcription factor apparatus (32). Since binding sites on this molecule may be limited, this may result in competition, and several transcription factors, including CREB itself and AP-1, may compete with GR for binding, thus yielding an indirect rather than a direct protein-protein interaction.
Target Genes in Control of Asthmatic Inflammation
Corticosteroids may control airway inflammation in asthma by inhibiting many aspects of the inflammatory process through increasing the transcription of anti-inflammatory genes and decreasing the transcription of inflammatory genes (3, 4, 33) (Table 1).
|
Anti-inflammatory proteins. Corticosteroids may suppress inflammation by increasing the synthesis of anti-inflammatory proteins. Steroids increase the synthesis of lipocortin-1, a 37-kD protein that has an inhibitory effect on phospholipase A2 (PLA2), and therefore may inhibit the production of lipid mediators. Steroids induce the formation of lipocortin-1 in several cells and recombinant lipocortin-1 has acute anti- inflammatory properties (34). However, corticosteroids do not induce lipocortin-1 expression in all cells, and this may be only one of many genes regulated by corticosteroids. Inhaled corticosteroid treatment does not appear to increase the release of lipocortin-1 into bronchoalveolar lavage (BAL) fluid in asthmatic patients (35). Corticosteroids increase the synthesis of secretory leukocyte protease inhibitor (SLPI) in human airway epithelial cells by increasing gene transcription (36). Secretory leukocyte protease inhibitor is the predominant antiprotease in conducting airways and may be important in reducing airway inflammation by counteracting inflammatory enzymes such as tryptase.
Nuclear factor-kappa B may play a pivotal role in asthmatic inflammation and is normally controlled by an inhibitory protein, IB, which binds to NF-B within the cytoplasm.
Corticosteroids have been found to increase the expression of
one form of IB, IB- in certain cell types (37, 38), although
this may not apply to all cell types (39, 40). Increased transcription of IB may thus lead to inhibition of NF-B and
control of inflammation. Gene transfer of IB- inhibits the
expression of adhesion molecules regulated by NF-B in endothelial cells (41).
Interleukin (IL)-10 is an anti-inflammatory cytokine secreted predominantly by macrophages in the lung, which inhibits the transcription of many pro-inflammatory cytokines
and chemokines (42) and appears to be mediated via an inhibitory effect on NF-B (43). Interleukin-10 secretion by alveolar macrophages may be impaired in asthmatic patients, resulting in increased macrophage cytokine secretion (44, 45).
Glucocorticoid treatment in asthmatic patients increases IL-10 secretion by these cells, although this appears to be an indirect effect, since treatment of alveolar macrophages in vitro
with corticosteroids tends to decrease IL-10 secretion (45).
Interleukin-1 receptor antagonist (IL-1ra) is another anti-inflammatory cytokine that specifically inhibits the actions of IL-1; it is synthesized by several cells, including airway epithelial cells. Corticosteroids increase the expression of IL-1ra in these cells in vitro (46) and in vivo (47). Another mechanism whereby corticosteroids may block IL-1 effects is through increased synthesis of a second type of IL-1 receptor (IL-1rII) that binds IL-1 without signaling and therefore acts as a decoy receptor (48).
The enzyme neutral endopeptidase (NEP) degrades bronchoconstrictor and inflammatory peptides, such as bradykinin and tachykinins. There is evidence that expression of this enzyme is increased in cultured human epithelial cells in vitro (49) and that patients treated with inhaled corticosteroids have a higher level of NEP expression in airway epithelial cells (50).
2-Adrenoceptors. Steroids increase the expression of 2-adrenoceptors by increasing the rate of transcription; the human 2-receptor gene has three potential GREs (51). Steroids
double the rate of 2-receptor gene transcription in human
lung in vitro, resulting in increased expression of 2-receptors
(52). Using autoradiographic mapping and in situ hybridization in animals to localize the increase in 2-receptor expression, there appears to be an increase in all cell types, including
airway epithelial cells and airway smooth muscle after chronic
glucocorticoid treatment (53). This may be relevant in asthma
because it may prevent downregulation in response to prolonged treatment with 2-agonists. In rats, corticosteroids prevent the downregulation and reduced transcription of 2-receptors in response to chronic -agonist exposure (53). However,
inhaled corticosteroids do not appear to prevent the tolerance
that develops to the protective effects of inhaled 2-agonists in
asthmatic patients (54), although intravenous steroids provide
rapid reversal of the bronchodilator subsensitivity (55).
Cytokines. Although it is not yet possible to be certain of
the most critical aspects of steroid action in asthma, it is likely that their inhibitory effects on cytokine synthesis are of particular relevance (56). Steroids inhibit the transcription of
several cytokines that are relevant in asthma, including IL-1, TNF-, granulocyte/macrophage colony-stimulating factor
(GMCSF), IL-2, IL-3, IL-4, IL-5, IL-6, IL-11, and the chemokines IL-8, RANTES, MCP-1, MCP-3, MIP-1, and eotaxin.
These effects were at one time thought to be mediated directly
via interaction of glucocorticoid receptors with a nGRE in the
upstream promoter sequence of the cytokine gene, resulting in
reduced gene transcription. Surprisingly, there is no apparent
nGRE consensus sequence in the upstream promoter region
of these cytokines, suggesting that corticosteroids inhibit transcription indirectly. Thus, the 5'-promoter sequence of the human IL-2 gene has no GRE consensus sequences, yet corticosteroids are potent inhibitors of IL-2 gene transcription in
T lymphocytes. Transcription of the IL-2 gene is predominantly regulated by a cell-specific transcription factor, nuclear
factor of activated T cells (NF-AT), which is activated in the
cytoplasm on T-cell receptor stimulation via calcineurin. A
nuclear factor is also necessary for increased activation, and
this factor appears to be AP-1, which binds directly to NF-AT
to form a transcriptional complex (57). Corticosteroids therefore inhibit IL-2 gene transcription indirectly by binding to
AP-1, thus preventing increased transcription due to NF-AT
(58). Other examples of cytokine genes negatively regulated
by corticosteroids that do not have a GRE in their promoter region include IL-8, which is regulated predominantly via
NF-B (59), and RANTES, which is regulated by NF-B and
AP-1 (60). There may be marked differences in the response
of different cells and different cytokines to the inhibitory action of corticosteroids, and this may be dependent on the relative abundance of transcription factors. Thus, in alveolar macrophages and peripheral blood monocytes, GMCSF secretion
is more potently inhibited by corticosteroids than IL-1 or IL-6
secretion (61).
Some cytokines may have anti-inflammatory effects in
asthma, however, and there is evidence that corticosteroids
may increase the expression of these cytokines, such as IL-10
and IL-1ra, as discussed above. IL-12 may play a key role in
regulating the balance between T helper 1 (Th1) and Th2
cells, increasing the proliferation of Th1 cells and the secretion
of IFN-
(62). Inhaled corticosteroids are reported to increase
the expression of IL-12 in airways (63).
Inflammatory enzymes. Nitric oxide synthase (NOS) may
be induced by pro-inflammatory cytokines, resulting in increased nitric oxide (NO) production. NO may increase airway blood flow and plasma exudation and may amplify the
proliferation of Th2 lymphocytes, which orchestrate eosinophilic inflammation in the airways (64) and act as a chemotactic agent for eosinophils (67). Inducible NOS (iNOS) is
potently inhibited by corticosteroids. In cultured human pulmonary epithelial cells, proinflammatory cytokines result in
increased expression of iNOS and increased NO formation
(68, 69). This is due to increased transcription of the iNOS
gene and is inhibited by corticosteroids. There is no nGRE in
the promoter sequence of the iNOS gene, but NF-B appears
to be the most important transcription factor in regulating
iNOS gene transcription (70). Since TNF-, IL-1, and oxidants activate NF-B in airway epithelial cells, this accounts
for their activation of iNOS expression (71). Corticosteroids
may therefore prevent induction of iNOS by inactivating NF-B, thereby inhibiting transcription.
Corticosteroids inhibit the synthesis of several inflammatory
mediators implicated in asthma through an inhibitory effect on enzyme induction. Corticosteroids inhibit the induction of the gene coding for inducible cyclo-oxygenase (COX-2) in monocytes and epithelial cells, and this also appears to be via NF-B
activation (72). Corticosteroids also inhibit the gene transcription of a form of PLA2 induced by cytokines (75). Whether
steroids also modulate expression of 5'-lipoxygenase has not
yet been established, but studies of cysteinyl-leukotriene formation in asthmatic patients in vivo indicate that doses of oral
or inhaled corticosteroids that are effective clinically do not
significantly reduce the excretion of leukotriene E4 (LTE4),
the major stable metabolite of leukotriene D4 (LTD4) (76, 77).
Steroids also inhibit the synthesis of endothelin-1 (ET-1) in lung and airway epithelial cells, and this effect may also be via inhibition of transcription factors that regulate its expression (78). Patients on inhaled corticosteroids have lower levels of ET-1 in BAL fluid than asthmatic patients treated only with bronchodilators (79).
Inflammatory receptors. Corticosteroids decrease the transcription of gene coding for certain receptors. Thus, the NK1-receptor that mediates the inflammatory effects of substance P in the airways may show increased gene expression in asthma (80). This may be inhibited by steroids through an interaction with AP-1 since the NK1 receptor gene promoter region has no GRE but has an AP-1 response element (81).
Corticosteroids are potent inducers of IL-1rII, resulting in release of a soluble form of the receptor, thus reducing the functional activity of IL-1, as discussed above (82).
Cell survival. Steroids markedly reduce the survival of certain inflammatory cells, such as eosinophils. Eosinophil survival is dependent on the presence of certain cytokines, such as IL-5 and GMCSF. Exposure to steroids blocks the effects of these cytokines and leads to programmed cell death or apoptosis (83). Steroids may increase the transcription of specific endonucleases, which may be relevant to the action of steroids on eosinophil and mast cell survival in the airways of asthmatic patients.
Adhesion molecules. Adhesion molecules play a key role in
the trafficking of inflammatory cells to sites of inflammation. The expression of many adhesion molecules on endothelial
cells is induced by cytokines. Steroids may lead indirectly to a
reduced expression via their inhibitory effects on cytokines,
such as IL-1 and TNF-. Steroids may also have a direct inhibitory effect on the expression of adhesion molecules, such
as intercellular adhesion molecule-1 (ICAM-1) and E-selectin
at the level of gene transcription (84). Intercellular adhesion
molecule-1 expression in bronchial epithelial cell lines and
monocytes is inhibited by corticosteroids (85).
| |
EFFECTS ON CELL FUNCTION |
|---|
Steroids may have direct inhibitory actions on several inflammatory cells implicated in pulmonary and airway diseases (Figure 5).
|
Macrophages
Steroids inhibit the release of inflammatory mediators and cytokines from alveolar macrophages in vitro (61, 86), although their effect after inhalation in vivo is modest (87). Steroids may be more effective in inhibiting cytokine release from alveolar macrophages than in inhibition of lipid mediators and
reactive oxygen species in vitro (88, 89). Inhaled corticosteroids reduce the secretion of chemokines and proinflammatory
cytokines from alveolar macrophages from patients with
asthma, whereas the secretion of IL-10 is increased (45). Oral
prednisone inhibits the increased gene expression of IL-1 in
alveolar macrophages obtained by BAL from patients with
asthma (90).
Eosinophils
Steroids have a direct inhibitory effect on mediator release from eosinophils, although they are only weakly effective in inhibiting secretion of reactive oxygen species and eosinophil basic proteins (91, 92). Steroids inhibit the permissive action of cytokines such as GMCSF and IL-5 on eosinophil survival (93, 94), and this contributes to the reduction in airway eosinophils seen with steroid therapy. One of the best-described actions of steroids in asthma is a reduction in circulating eosinophils, which may reflect an action on eosinophil production in the bone marrow (95). In patients with asthma there is an increase in the proportion of low-density eosinophils in the circulation that may reflect an effect of cytokines (96). Inhaled corticosteroids inhibit the increase in circulating eosinophil count at night in patients with nocturnal asthma and also reduce plasma concentrations of eosinophil cationic protein (97). After inhaled corticosteroids (budesonide 800 µg b.i.d.), there is a marked reduction in the number of low-density eosinophils, presumably reflecting inhibition of cytokine production in the airways (98).
T Lymphocytes
An important target cell in asthma may be the T lymphocyte,
since steroids are very effective in inhibition of activation of
these cells and in blocking the release of cytokines, which are
likely to play an important role in the recruitment and survival of inflammatory cells involved in asthmatic inflammation. In
an experimental model of asthma that involves sensitization
and repeated exposure to allergen in an IgE-producing strain
of rat, there is an influx of eosinophils and lymphocytes into
the lung with a concomitant increased airway responsiveness
to inhaled methacholine (99). Pretreatment with steroids completely inhibits the increased eosinophil and lymphocyte numbers and the increase in airway responsiveness, whereas pretreatment with cyclosporin A, which similarly inhibits cellular
influx, fails to block the increased airway responsiveness (100).
This suggests that the effect of steroids on airway hyperresponsiveness may be through other cells in addition to T lymphocytes. In T-cell clones derived from BAL, corticosteroids
inhibited the secretion of IL-4 and IL-5 to a greater extent
than IFN-, which may indicate that the balance is tipped in
favor of Th1 cells (101).
Mast Cells
While steroids do not appear to have a direct inhibitory effect
on mediator release from lung mast cells (102, 103), chronic steroid treatment is associated with a marked reduction in mucosal mast cell number (104, 105). This may be linked to a reduction in IL-3 and stem cell factor (SCF) production, which is
necessary for mast cell expression in tissues. Mast cells also secrete various cytokines (TNF-, IL-4, IL-5, IL-6, IL-8), but
whether this is inhibited by steroids is not yet clear.
Dendritic Cells
Dendritic cells in the epithelium of the respiratory tract appear to play a critical role in antigen presentation in the lung because they have the capacity to take up allergen, process it into peptides, and present it via major histocompatibility complex molecules on the cell surface for presentation to uncommitted T lymphocytes (106). In experimental animals the number of dendritic cells is markedly reduced by systemic and inhaled corticosteroids, thus dampening the immune response in the airways (107). Topical steroids markedly reduce the numbers of dendritic cells in the nasal mucosa (108), and it is likely that a similar effect would be seen in airways.
Neutrophils
Neutrophils, which are not prominent in the biopsies of patients with asthma, are not very sensitive to the effects of steroids. Indeed, systemic steroids increase peripheral neutrophil counts, which may reflect an increased survival time due to an inhibitory action of neutrophil apoptosis (in complete contrast to the increased apoptosis seen in eosinophils) (109, 110).
Endothelial Cells
Glucocorticoid receptor gene expression in the airways is most prominent in endothelial cells of the bronchial circulation and airway epithelial cells. Steroids do not appear to directly inhibit the expression of adhesion molecules, although they may inhibit cell adhesion indirectly by suppression of cytokines involved in the regulation of adhesion molecule expression (111). Steroids may have an inhibitory action on airway microvascular leak induced by inflammatory mediators (112, 113). This appears to be a direct effect on postcapillary venular epithelial cells. The mechanism for this antipermeability effect has not been fully elucidated, but there is evidence that synthesis of a 100-kD protein distinct from lipocortin-1, termed vasocortin, may be involved (114). Although there have been no direct measurements of the effects of steroids on airway microvascular leakage in asthmatic airways, regular treatment with inhaled corticosteroids decreases the elevated plasma proteins found in BAL fluid of patients with stable asthma (115).
Epithelial Cells
Epithelial cells may be an important source of inflammatory mediators in asthmatic airways and may drive and amplify the inflammatory response in the airways (116, 117). Airway epithelium may be one of the most important targets for inhaled corticosteroids in asthma (33, 118). Steroids inhibit the increased transcription of the IL-8 gene induced by TNF in cultured human airway epithelial cells in vitro (119, 120) and the transcription of the RANTES gene in an epithelial cell line (121, 122). Eotaxin is a highly selective and potent eosinophil chemoattractant that is expressed in airway epithelial cells (123) and may be an important target for inhaled corticosteroids. Inhaled corticosteroids inhibit the increased expression of GMCSF, MIP-1, and RANTES in the epithelium of patients with asthma (124).
There is increased expression of iNOS in the airway epithelium of patients with asthma (127), and this may account for the increase in NO in the exhaled air of patients with asthma compared with normal subjects (128, 129). Patients with asthma who use inhaled corticosteroids regularly, however, do not show such an increase in exhaled NO (128), suggesting that corticosteroids have suppressed epithelial iNOS expression. Furthermore, double-blind randomized studies show that oral and inhaled corticosteroids normalize the elevated exhaled NO in patients with asthma (130, 131).
Corticosteroids decrease the transcription of other inflammatory proteins in airway epithelial cells, including COX-2, cPLA2, and endothelin-1 (72, 78). Airway epithelial cells may be the key cellular target of inhaled corticosteroids; by inhibiting the transcription of several inflammatory genes, inhaled corticosteroids may reduce inflammation in the airway wall (Figure 6).
|
Mucus Secretion
Steroids inhibit mucus secretion in airways, perhaps through direct action on submucosal gland cells (132). The inhibitory effect of steroids may involve lipocortin-1 synthesis (133). Recent studies suggest that steroids may also inhibit the expression of mucin genes, such as MUC2 and MUC5A (134). In addition, there are indirect inhibitory effects due to the reduction in inflammatory mediators that stimulate increased mucus secretion.
| |
EFFECTS ON ASTHMATIC INFLAMMATION |
|---|
Bronchial Biopsies and Lavage
Corticosteroids are remarkably effective in controlling the inflammation in asthmatic airways, and it is likely that they have multiple cellular effects. Several biopsy studies in patients with asthma have now confirmed that inhaled corticosteroids reduce the number and activation of inflammatory cells in the airway (104, 105, 125, 135, 136). Similar results have been reported from BAL of patients with asthma after inhaled budesonide, with a reduction in both eosinophil number and eosinophil cationic protein concentrations, a marker of eosinophil degranulation (137, 138). These effects may be due to inhibition of cytokine synthesis in inflammatory and structural cells. There is also a reduction in activated CD4+ T cells (CD4+/ CD25+) in BAL fluid after inhaled corticosteroids (139). The disrupted epithelium is restored and the ciliated cell/goblet cell ratio is normalized after 3 mo of therapy with inhaled corticosteroids (104). There is also some evidence that a reduction in the thickness of the basement membrane may occur (125), although in patients with asthma taking inhaled corticosteroids for over 10 yr, the characteristic thickening of the basement membrane was still present (140).
Induced Sputum
The anti-inflammatory effect of inhaled corticosteroids has been confirmed in less invasive studies using induced sputum. In untreated subjects with asthma there is an increase in eosinophils and eosinophil cationic protein (ECP); these inflammatory markers are reduced by oral and inhaled corticosteroid therapy (141, 142), the use of which is also associated with evidence of apoptosis of eosinophils (143). Conversely, reductions in the dose of steroid results in an increase in sputum eosinophils and an increase in asthma symptoms (144).
Exhaled NO
Exhaled NO correlates with sputum eosinophil counts in patients with asthma (145) and may also be a marker of airway inflammation (146). Patients with asthma who are treated with inhaled corticosteroid have significantly lower levels of exhaled NO than subjects with untreated asthma (128). Controlled studies in patients with asthma demonstrate that oral and inhaled corticosteroids reduce the elevated exhaled NO in a dose-related manner (147). Furthermore, reduction in the dose of inhaled corticosteroids in patients with well-controlled asthma results in an increase in exhaled NO which precedes the increase in symptoms and the deterioration in lung function (150). Thus, measurement of exhaled NO may be useful in early detection of loss of anti-inflammatory control in the clinical setting.
Airway Hyperresponsiveness
By reducing airway inflammation, inhaled corticosteroids consistently reduce airway hyperresponsiveness (AHR) in adults and children with asthma (151). Chronic treatment with inhaled corticosteroids reduces responsiveness to histamine, cholinergic agonists, allergen (early and late responses), exercise, fog, cold air, bradykinin, adenosine, and irritants such as sulfur dioxide and metabisulfite. The reduction in AHR takes place over several weeks and may not reach maximum until after several months of therapy. The magnitude of reduction varies among patients but is in the order of one to two doubling dilutions for most challenges; it often fails to return to the normal range. This may reflect suppression of the inflammation but persistence of structural changes that cannot be reversed by steroid therapy. Inhaled corticosteroids not only make the airways less sensitive to spasmogens, but they also limit the maximal airway narrowing in response to spasmogens (152).
| |
CLINICAL EFFICACY OF INHALED CORTICOSTEROIDS |
|---|
Inhaled corticosteroids are very effective in controlling symptoms in asthmatic patients of all ages and disease severity (1, 153). The clinical efficacy of inhaled corticosteroids in the treatment of asthma has been further established through the research produced during the last few years. New information pertains mainly to:
Dose-Response Relationships
The clinical effect of inhaled corticosteroids is best evaluated in dose-response trials. A number of clinical studies have assessed the effects of different doses of specific corticosteroids. These studies have provided clinically important information that can also be of use when interpreting results from trials which have compared two or more different corticosteroids or inhalers. Therefore, a brief review of the published dose- response trials of the various inhaled corticosteroids will be presented.
Beclomethasone Dipropionate
An early 5-mo clinical study of inhaled beclomethasone dipropionate (BDP) looked at 15 patients with asthma who were dependent on inhaled corticosteroids. Despite increasing the dose of inhaled BDP delivered via a pressurized metered-dose inhaler (pMDI) from 400 µg/d to 1,600 µg/d, no additional benefit in FEV1 was derived from treatment (154). These results contrast with those of a larger 56-wk study, which found that a reduction of oral steroid dose was easier to achieve when an initial dose of 800 µg/d BDP pMDI was used rather than an initial dose of 400 µg/d (155).
Another open, graded-dose trial of BDP included adult patients with oral steroid-dependent asthma who increased their
dose of BDP stepwise every 2 wk from 200 µg up to 1,600 µg/
day (156). A dose-response effect was seen in terms of symptoms, 2-agonist inhaler use, and frequency of exacerbations
during the 26-wk study. However, this study was not randomized, so it is possible that the progressive improvements
noted may have reflected the effect of duration of inhaled corticosteroid treatment rather than dose. No significant difference between the effects of adjacent doses was seen. At a later
stage in the study, the patients were able to progressively reduce their dose of oral steroid, revealing an apparent dose-
response for the oral steroid-sparing effect of the inhaled
BDP (157), but again with little difference between individual
dose steps.
A more recent randomized, double-blind comparison of the oral steroid-sparing effect of high and low doses of BDP (1,500 µg/d and 300 µg/d, respectively, delivered via pMDI with plastic spacer) in a group of patients with severe asthma showed no dose-response effect (158). Both groups of patients achieved a mean reduction of 5 mg/d in oral prednisolone dose over the 6-mo study period. Finally, control of asthma was better in patients receiving high doses of BDP (1,000- 2,000 µg per day) than in patients who received 400 µg/d (159).
Budesonide
An acute dose-response relationship in airway function was seen when single doses of budesonide (BUD) (100 µg, 400 µg, and 1,600 µg) were given by pMDI to 12 patients with chronic stable asthma (160). Another study of 34 patients showed a linear relationship between dose and peak expiratory flow (PEF) when BUD was given by pMDI in the dose range of 400 µg/d to 1,600 µg/d for 2 wk (161). This study was not randomized, however, so the results could reflect the effect of total treatment time with corticosteroid rather than dose. As in most other studies, there was no significant difference between the effects of individual adjacent doses. A 15-wk study of 45 patients with steroid-dependent asthma (162) and a 4-wk crossover study of 24 patients with steroid-dependent asthma (163) both found that 1,600 µg/d BUD given via pMDI produced greater improvement in lung function than 400 µg/d BUD. In a double-blind, crossover study of 18 patients with moderate chronic asthma, dose-dependent increases in PEF were seen when BUD was given at 100 µg, 400 µg, and 1,600 µg/d by pMDI (each dose given for 2 wk) (164). No significant differences were seen between individual dose steps.
Recent double-blind, placebo-controlled, dose-finding studies of 473 adults (165) and 404 children (166) both found a dose-response relationship for lung function with BUD delivered through a multiple-dose dry powder inhaler (Turbuhaler) (Figure 7). The dose-response effect was statistically significant but relatively small, and the difference between placebo and low-dose BUD (200 µg/d) was greater than the difference between low-dose and high-dose BUD (1,600 µg/d). The differences between individual dose steps were not significant in these studies; in particular, there was no difference between the effects of 800 µg/d and 1,600 µg/d.
|
A double-blind, crossover study of 19 schoolchildren with moderate to severe asthma compared the effects of 100 µg, 200 µg, and 400 µg/d BUD given by pMDI with large-volume plastic spacer (167). Morning PEF values and FEV1 for all three doses were significantly better than for placebo. There was no dose-response effect in morning PEF values measured at home, but a dose-response effect with significant differences between the effects of individual doses was seen in the FEV1 values measured at the clinic. The fall in FEV1 or FEF25-75% after exercise proved to be a sensitive marker of dose-response with significant differences detected between adjacent doses.
The results of this particular study illustrate an important
phenomenon that often goes unrecognized: the dose-response
curve differs for the various measurable effects of inhaled corticosteroids. This is reflected by another recent study which
reported that low doses of inhaled corticosteroid produced a
marked effect on symptoms and lung function, whereas somewhat higher doses were required to normalize NO concentration in exhaled air (168). In an early study, Toogood and colleagues (156) also found that the shape of the dose-response
curve depended on the parameter measured, with symptom
control and 2-agonist use revealing steeper curves than those
of PEF and lung functions. The dose of inhaled corticosteroids
necessary to prevent asthma exacerbations may differ from
that needed to control chronic asthma.
Fluticasone Propionate
A dose-dependent increase in PEF was seen in 672 adult patients with asthma treated for 4 wk with 100-800 µg/d fluticasone propionate (FP) delivered by pMDI. Mean morning PEF was 364 liters/min at a dose of 100 µg/d and 378 liters/min at 800 µg/d (169). No significant difference was seen between any of the individual doses used in the study.
Similar results were found in three randomized, double-blind, parallel group studies, each including over 300 adults with mild to moderate asthma (170). Significant differences were seen between the effects of placebo and those of 50 µg/d, 200 µg/d, and 1,000 µg/d FP pMDI (170), between placebo and 100 µg/d, 200 µg/d, and 500 µg/d (Diskhaler) (171), and between placebo and 200 µg/d, 500 µg/d, and 1,000 µg/d (pMDI). Once again, however, no significant differences were seen between the effects of the different doses of FP (although there was a nonsignificant trend toward greater efficacy of the higher doses in two of the three studies).
Another study compared the effect of two high doses of FP pMDI (1,500 µg/d and 2,000 µg/d) with that of placebo in a group of 96 adult oral steroid-dependent patients (173). Both FP doses were significantly better than placebo in eliminating oral steroid dependence and in leading to improvements in FEV1. A significant difference between the two doses was seen in effect on FEV1 but not on oral steroid dose.
The effect of 100 µg/d and 200 µg/d FP Diskhaler was
compared with that of placebo among 169 children with
asthma (174). Both doses of FP led to significant improvements in PEF and asthma symptom scores over 6 and 12 wk,
but no significant difference was seen between the effects of
the two doses of FP in any other measured parameter. Similar
findings were reported in a study comparing daily doses of
100, 200, and 500 µg FP Diskhaler in 331 adults with moderate asthma (FEV1 > 50% and < 80% predicted normal) in a
randomized, placebo-controlled trial. All doses significantly improved asthma symptom scores and lung function and decreased the need for rescue 2-agonists. However, no differences were detected among the three FP treatments in any of
the other outcome measures (171).
Summary of Dose-Response Studies
All dose-response studies reported so far share some common
features. They all show marked and statistically significant differences between the effects of all doses of inhaled corticosteroid and placebo, and most of them also show significant
dose-response relationships between dose and effect. However, virtually all have failed to show statistically significant
differences between the clinical effects of adjacent doses on
the dose-response curve. Normally, a 4-fold or greater difference in dose has been required to detect a statistically significant
(but often small) difference in effect on commonly measured
outcomes such as symptoms, PEF, use of rescue 2-agonist,
and lung functions; even such large differences in dose are not
always associated with significant differences in response. These
findings suggest that pulmonary function tests or symptoms may
have a rather low sensitivity in the assessment of the effects of
inhaled corticosteroids. An awareness of this is important for
the interpretation of clinical comparisons between different
inhaled corticosteroids or inhalers.
More studies are needed to assess whether other outcome measures such as AHR or a steroid-sparing effect may be more sensitive than traditional outcome measures such as symptoms or lung function tests. One study measuring changes in response to provocation by methacholine showed a dose-response effect of two doses of BUD (200 µg/d and 800 µg/d) (175). The fall in FEV1 after exercise testing may also act as a surrogate measure for underlying airway inflammation; a dose-response relationship for inhaled BUD therapy has been demonstrated using this technique (167). Unlike most dose-response studies of inhaled corticosteroids, this study showed significant differences in response between each dose step when this outcome parameter was measured. As for other outcome parameters, the lowest dose was very effective, producing about 50% of the maximum achievable response.
Clinical Comparisons of Different Inhaled Corticosteroids
Many factors other than dose and drug may influence the clinical and systemic effects of inhaled corticosteroids. These factors should always be considered and presented in the resulting publication to allow accurate interpretation and comparison of results. Some of the more important factors are briefly discussed.
Each of these factors may influence the measured clinical and systemic effects of an inhaled corticosteroid to at least the same extent as doubling or halving the specified dose and thus are of critical importance. The value of comparative studies of inhaled corticosteroids that ignore these factors is doubtful. Furthermore, the difficulties encountered when attempting to establish whether significant clinical differences exist between individual doses on the dose-response curve for a specific inhaled corticosteroid are likely to be multiplied when comparisons between different inhaled corticosteroids are performed. These difficulties may be particularly prominent when standard outcome parameters, such as symptoms, FEV1, or morning PEF, are used.
Comparisons of the effects of different inhaled corticosteroids, or of the same inhaled corticosteroid administered with different dosing frequencies, or via different inhaler devices are subject to many inaccuracies if careful consideration is not given to the study design and subsequently to the resulting publication. The optimal design of a comparative trial of corticosteroids has not yet been identified, but true differences between corticosteroids are most likely to be detected in well-designed trials that control for the factors mentioned above. While our present knowledge does not allow us to accurately specify the ideal study design, examples of inappropriate and unsuitable study designs abound.
One reason for the apparently flat dose-response curve for inhaled corticosteroids is probably that most studies have included a very heterogeneous study population with marked differences in individual dose-response curves. Alternatively, the studies may have measured insensitive outcome parameters. This problem may be reduced by studying different doses in more homogeneous populations using crossover designs (167) or by using designs that make multiple individual dose adjustments to assess the minimal effective dose. Further studies are needed to confirm this.
Many trial designs have been used to compare different inhaled corticosteroids, and again there is no consensus view on the best design. The various trial designs published to date are listed below in descending order of what is currently believed to be their likely value.
Studies based on other designs may also contribute to the assessment of the effects of different forms of inhaled corticosteroid therapy but are of more limited value in the quantitative comparison of the effects of different corticosteroids. Such studies take two main forms:
The foregoing classification of studies will be used in the next sections which discuss the key comparative studies reported to date.
BUD and BDP
Dose-response comparison. In a double-blind, four-period,
crossover study, BDP pMDI 200 µg and 500 µg twice daily
was compared with BUD pMDI 200 µg and 400 µg twice
daily. The effects of both doses of both drugs on PEF, symptom scores, and 2-agonist use were comparable and significant
when compared with the effect of placebo. No dose-response
effect was seen for either drug (179). Although the study was
well designed, no firm conclusion about equivalence of dose
can be drawn from it because of the absence of a dose-response
with either drug.
Dose down-titration comparisons. Two studies have indicated a greater efficacy for BUD Turbuhaler than for BDP pMDI (182, 183). These studies were open in design and included extensive dose adjustments of the two drugs. After 3 mo of treatment, the two groups of patients in the first study had equivalent control of their asthma at a mean dose of BUD of 900 µg/d and a mean dose of BDP of 1,350 µg/d (182). The second study (183) suggested that BUD Turbuhaler 600 µg/d is equivalent in efficacy to BDP 1,000 µg/d.
Equal dose comparisons. In a double-blind, crossover-designed, oral steroid-sparing study of 40 adults, BDP pMDI (100 µg q.i.d.) had a greater effect than an equal dose of BUD pMDI (184).
In a double-blind, crossover study of 28 patients receiving high doses of either BDP (750 µg twice daily) or BUD (800 µg twice daily), no differences in asthma control were observed (185). Another double-blind, crossover trial involving 21 children showed that equal doses of BDP and BUD (100 µg twice daily) had similar positive effects on lung function (186). The power of these two studies to detect differences between the two drugs was very low, however, because of the study designs and the small groups of patients.
Finally, results from an open crossover study suggested a greater efficacy for 400 µg/d BUD Turbuhaler than for 400 µg/d BDP Rotahaler (187). More pronounced increases in FEV1, FVC, and FEF50 were seen in the BUD group than in the BDP group, and only the BUD group showed a significant improvement in AHR after provocation with histamine.
Two-to-one comparison. A comparison between BDP pMDI (1,500 µg/d) and BUD Turbuhaler (800 µg/d) found no significant difference in any outcome parameters between the two treatments (188). In contrast, an open multicenter study found that BUD Turbuhaler 400 µg/d was significantly more effective than BDP pMDI 800 µg/d in 227 patients with moderate asthma (189).
No firm conclusions can be drawn from these studies. BUD Turbuhaler may be more potent than BDP pMDI and Rotahaler, but further studies are needed to confirm this.
BDP and FP
Dose-response versus one-dose comparison. The most comprehensive comparison between BDP and FP to date compared multiple doses of FP (100-800 µg/d by pMDI) with a single dose of BDP (400 µg/d by pMDI) in 672 patients (169). A flat dose-response curve of lung function was seen, with no statistically significant differences in clinical efficacy between any dose of FP compared with the single dose of BDP, nor between the various doses of FP. The authors' conclusion was that 400 µg BDP pMDI was equivalent to 200 g FP pMDI.
Equal dose comparison. One 3-mo study comparing nearly equal doses of BDP (1,600 µg/d) and FP (2,000 µg/d), both delivered by pMDI, showed that the two were similar in improving the efficacy of asthma control when prescribed to 134 patients who had previously received a lower dose of inhaled corticosteroid (190). Another double-blind, parallel group study in 274 adults found that treatment with 1,500 µg/d FP pMDI resulted in significantly higher morning and evening PEF values and fewer exacerbations than the same dose of BDP pMDI (191).
Two-to-one comparisons. Three studies have compared BDP with half the dose of FP; no difference in clinical effect was found between the two treatments in any of them. One compared BDP (2,000 µg/d) and FP (1,000 µg/d) in 154 adults with severe asthma (192). Another compared BDP (400 µg/d) and FP (200 µg/d), both given by pMDI with plastic spacer to 398 children with asthma (193). The third compared BDP (400 µg/d) and FP (200 µg/d), both given by pMDI to 261 adult patients with mild to moderate asthma (194).
As seen with the comparisons between BDP and BUD, it is not possible to draw any firm conclusion about the comparative potency of BDP and FP on the basis of these studies. Fluticasone propionate may be clinically more potent than BDP. Further studies are needed, however, to accurately assess their potency ratio.
BUD and FP
Dose down-titration comparison. A recent double-blind study compared the efficacy of BUD Turbuhaler and FP Diskhaler in 217 children with moderate asthma (180). On entry, all the children were receiving BUD by pMDI with large-volume plastic spacer; the dose was gradually reduced to define the minimal effective dose. After this, the children entered a run-in period, followed by randomization to half this dose of BUD Turbuhaler or FP Diskhaler for 5 wk. If there was no deterioration over that period, the dose was further reduced by 50% at 5-wk intervals until deterioration in asthma control was seen, as defined by criteria based on diary card variables and exercise testing. Compliance with therapy was monitored, and inhalation technique was known to be optimal for both devices. The mean minimal effective doses for the two treatments were nearly identical, and neither dose was significantly different from the other in clinical effect.
Dose-halving comparison. A randomized open study in 171 adults compared BUD Turbuhaler and FP Diskhaler, both in doses of 200 µg/d and 400 µg/d, in patients who had previously been treated for mild to moderate asthma with conventional inhaled corticosteroids (mainly BDP) at twice the dose of BUD or FP used in the study (181). No difference was seen between the effects of equal doses of the two drugs in this study, and reducing the dose of BDP by half proved possible since patients were switched to either BUD or FP without loss of asthma control. The study results were not analyzed to detect differences between the effects of the two doses of each drug.
Dose-response versus one-dose comparison. A double-blind, multicenter study in 671 patients with severe asthma compared FP pMDI (1,000 µg/d and 2,000 µg/d) with BUD pMDI (1,600 µg/d) (195). Both doses of FP led to significantly greater increases in lung function than the single dose of BUD. The increases in mean morning PEF from baseline for FP 2,000 µg/d was 24 L/min, for FP 1,000 µg/d was 21 L/min, and for BUD 1,600 µg/d was 13 L/min. No dose-response effect was seen between the two doses of FP.
Equal dose comparison. An 8-wk randomized, parallel, double-blind, double-dummy study in 229 children with already well-controlled asthma compared the effects of BUD Turbuhaler (400 µg/d) and FP Diskhaler (400 µg/d) (196). Both forms of therapy led to small improvements in mean morning and evening PEF; during the middle part of the study the increase for the FP treatment group was significantly greater than for the BUD group. However, the difference was transient and had disappeared by 8 wk. There were no differences between the two drugs in the other measured efficacy markers, including day and nighttime asthma scores and rescue bronchodilator use. In contrast, a multicenter, randomized, open, parallel group study involving 230 adult patients with mild to moderate asthma comparing BUD Turbuhaler (400 µg/d once daily; 200 µg twice daily) with that of FP Diskhaler (200 µg twice daily) found no difference in clinical effect between any of the treatments (197).
Two-to-one comparisons. A number of open, randomized, parallel group studies utilizing unequal doses of the two drugs have been performed in adult patients with asthma. One compared FP (200 µg/d) with BUD (400 µg/d), both via pMDI, in 122 patients for 8 wk (198). Another compared FP pMDI (500 µg/d) with BUD Turbuhaler (1,200 µg) in 456 patients for 8 wk (199). Five more compared the dry powder preparations of BUD and FP. One of these compared FP Diskhaler (400 µg/d) with BUD Turbuhaler (800 µg/d) in 243 patients during 8 wk (200). Another compared FP Diskhaler (200 µg/d) with BUD Turbuhaler (400 µg/d) in 164 patients, also over 8 wk (201). A third (double-blind) study compared FP Diskhaler (800 µg/d) with BUD Turbuhaler (1,600 µg/d) in 486 patients during 12 wk (202). A fourth open study compared FP Diskus/Accuhaler (500 µg/d) with BUD Turbuhaler (1,200 µg/d) in 259 patients for 4 wk (203). A similar open study compared FP Diskus (200 µg/d) with BUD Turbuhaler (400 µg/d) in 321 children during 4 wk (204). The authors of these studies all concluded approximate therapeutic equivalence between the compared doses of the two drugs. A recent meta-analysis of seven of the studies with this design was said to show that the FP:BUD potency ratio is at least 2:1 (205).
As seen with the comparisons of other drugs, it is impossible to draw unequivocal conclusions about the comparative efficacy of FP and BUD based on the studies published to date. The studies suggest that FP pMDI is more potent than BUD pMDI, whereas FP pMDI and FP Diskhaler appear to be approximately equipotent with BUD Turbuhaler (by microgram nominal dose).
FP and Triamcinolone Acetonide
Treatment with 500 µg/d FP Diskhaler was found to be significantly more effective than 800 µg/d triamcinolone acetonide (TAA) with regard to PEF measurements, albuterol use, and exacerbation rate in 304 patients with moderate asthma in a 6-mo study (206). Both drugs were more effective than placebo. In light of the problems discussed earlier associated with identifying clinical differences between inhaled corticosteroids, these findings suggest quite marked differences in clinical potency (at least 4-fold) between the two drugs.
FP, TAA, and Flunisolide
A recent study compared dose-response relations between BDP, flunisolide (FLU), and TAA by evaluating their blocking effect on acute antigen-induced bronchoconstriction (207). Results reflected a dose-related increase in effect. It is possible, however, that prolonged treatment contributed to the increased effect (rather than dose alone) since the order of doses was not randomized. No significant differences were detected between adjacent doses on the dose-response curve for any of the outcome parameters measured. On the basis of the trial design, the validity of the conclusion of this study, that BDP, FLU, and TAA have similar potencies, must be questioned.
Conclusions
Although a substantial number of comparative studies have been performed, it is difficult to draw firm conclusions about the comparative efficacy of different inhaled corticosteroids. This may be partially explained by differences between the designs of studies, the flat dose-response relationship for inhaled corticosteroids, the differences between inhalers, and the lack of control over important confounding factors in many studies. Additional well-designed comparisons, perhaps examining different outcome parameters, are required before any definite conclusions can be made.
Clinical Efficacy in Adults
Inhaled corticosteroids are also very effective in the treatment of asthma in adults (208). Their efficacy has been shown by a reduction of symptoms and exacerbations, improvement of lung function, and a decreased need for bronchodilator rescue therapy (209, 210). Reduction of airway inflammation, manifested both by airway histology findings and improved AHR, has also been documented (104, 135, 139, 211).
Systemic corticosteroids are effective in the management of severe asthma. However, their long-term use is associated with significant and undesirable side effects. Early studies with inhaled corticosteroids demonstrated their effectiveness in reducing or eliminating the need for systemic steroids while maintaining symptom control and lung function. Despite use of high-dose inhaled corticosteroids, a number of patients are unable to dispense with oral corticosteroid therapy. Noonan and collaborators (173) identified 96 patients in a multicenter trial who required oral prednisone despite significant doses of inhaled corticosteroids. These patients with severe, persistent asthma were randomized to treatment with either 1,500 µg/d or 3,000 µg/d FP via pMDI or placebo. Only 3% of the placebo-treated patients were able to eliminate their oral prednisone. In contrast, 69%-88% of the FP-treated groups were able to safely and effectively stop prednisone, and moreover, these patients demonstrated improved FEV1 values. Those treated with inhaled FP also showed greater symptomatic control of asthma. These results suggest that inhaled corticosteroids may have greater effectiveness in the control of asthma than oral prednisone when chronic disease is evaluated, although the mechanisms for this difference in effect is not apparent.
There has been some discussion as to whether single or divided doses of inhaled corticosteroids are most effective.
Weiner and colleagues (212) compared single versus twice-daily dosing and Malo and associates (213) compared four
times versus twice-daily dosing in patients requiring less than
1,200 µg BUD daily. Although different in design and frequency of daily dosing, both trials were able to demonstrate
that more frequent dosing yielded more effective asthma control, as manifested by lower 2-agonist use and decreased peak
flow variability, symptoms, and rate of exacerbations. The differences in results between the two dosing approaches of each
study were not striking, but they do indicate that overall effectiveness is likely greater with more frequent dosing. The mechanisms behind this effect are not established. Issues such as compliance, convenience, and acceptability, factors influenced in part by dosing frequency, need to be considered
when selecting the dosing regimen for individual patients.
Clinical Efficacy in Children
Controlled trials have established that inhaled corticosteroids are effective in all children regardless of asthma severity. Continuous treatment controls both day and nighttime symptoms, reduces the frequency of acute exacerbations and the number of hospital admissions, and improves lung function and AHR (214) both in patients treated at hospital clinics and in patients seen in general practice.
Normally, quite marked and rapid clinical improvements and changes in lung function are seen at very low daily doses (around 100 µg), even in children with moderate and severe asthma (167, 214, 215) (Figure 8). These improvements in lung function and symptoms precede and reach a plateau before a reduction in AHR (229). Further improvement in these parameters with increasing doses is rather small; often it may take an additional 4-fold increase in dose to produce supplementary significant effect on symptoms or peak flow measurements. A daily dose of 400 µg BUD pMDI with plastic spacer produced about 80% of the maximum achievable protection against exercise-induced asthma. This indicates that the vast majority of schoolchildren can achieve optimal symptom control on quite low daily doses of inhaled corticosteroids, around 100-200 µg (167, 214). Somewhat higher doses and/or longer treatment are required to control AHR as assessed by protection against exercise-induced asthma or histamine challenge (167, 217, 229).
|
Clinical Effect in Preschool Children and Infants
Though the number of controlled studies in preschool children and infants is not as high as in schoolchildren, an increasing number of placebo-controlled clinical trials in this age
group demonstrate marked clinical effects of inhaled corticosteroids, including improvement in lung function, reduction of
AHR, oral steroid requirement, and need for rescue bronchodilators (233). Results from these studies indicate that
benefits similar to those achieved in schoolchildren may also
be achieved in preschool children. Virus-induced wheeze,
which is common in these age groups, is also modified (239),
and two controlled trials have found that nebulized BDP
reduced the frequency (but not the severity) of respiratory
symptoms and improved lung function in infants with wheezing postbronchiolitis (240, 241). Finally, three uncontrolled studies and one double-blind, placebo-controlled trial have reported clinical improvement among infants with severe
asthma who had not responded to other treatment (238, 242-
244). Two dose-finding studies have been performed with nebulized BUD (245, 246). A mean minimal effective daily dose
of around 1,000 µg was found in one study (245); however,
marked individual variations were seen and the conclusion of
both studies was that the dose of nebulized BUD must be individualized. A problem with young children and infants
in addition to difficulties with effective drug delivery to the intrapulmonary airwaysis that in the daily clinical situation it is
impossible to distinguish between children with recurrent wheeze who have asthma and children with no asthma suffering from recurrent virus-induced wheeze. The optimal use of
inhaled corticosteroids may differ between these two groups.
Just as with older children, treatment with inhaled corticosteroids has also been shown to be cost-effective in children aged 1-3 yr (247). No formal comparisons with other treatments have been performed in young children. However, the children recruited for the controlled trials had all tried unsuccessfully to obtain asthma control with other anti-asthma medications, including theophylline, inhaled and oral bronchodilators, cromolyn sodium, and alternate-day oral steroids. Still, inhaled corticosteroids improved their condition, indicating that this treatment is also more effective than any other anti-asthma treatment in young children. Further studies are needed to confirm this.
Comparisons with Other Drugs
In comparing inhaled corticosteroids with other therapies for asthma, efficacy outcome measures represent a fundamental problem. Since inhaled corticosteroids differ from other agents in their mechanism of action, results of trials comparing steroids with other agents may be heavily influenced by the outcome parameter chosen for measurement. Though inhaled corticosteroids affect more outcome measures than any other class of anti-asthma drugs, the risk of reaching a misleading conclusion is great when the main emphasis is placed on a single outcome parameter, such as PEF. When an inhaled corticosteroid is compared with a bronchodilator, for example, and no difference in PEF is found between the two treatments, it is incorrect to conclude that the two treatments are equally effective. The results might likely have been quite different if another outcome measure, such as exacerbations or control of AHR, had been evaluated. Finally, recognizing that asthma is heterogeneous (i.e., patients with long history of symptoms may respond less well to inhaled corticosteroids than those with a shorter history), comparisons may also depend on the subgroup of patients with asthma chosen for study.
Adults
Bronchodilators. Several studies have compared the effects of inhaled corticosteroids with bronchodilators. One series of studies, conducted by the Dutch Chronic Nonspecific Lung Disease Study Group, compared the addition of inhaled corticosteroid, placebo, and the anticholinergic bronchodilator, ipratropium bromide, in a mixed group of patients with chronic obstructive pulmonary disease (COPD) and asthma (248). Addition of inhaled corticosteroid, but not placebo or ipratropium bromide, was associated with improvements in airflow as reflected by FEV1 and improvement in AHR as assessed by peak flow variation. Interestingly, these benefits were achieved with increased cost, even allowing for improved health status (251). Assessing cost/benefit, therefore, very much depends on assessing the value of reduced symptoms.
In another study, treatment with BUD was associated with a significant improvement in AHR, morning and evening PEF and peak flow variations, asthma symptom scores, and rescue medication usage compared with terbutaline (209). The improvement in AHR observed with BUD was persistent throughout the 2 yr of study. Bambuterol and BUD have also been compared in a study of 4 wks' duration (252). Both improved airflow, AHR, and nocturnal symptoms, although inhaled BUD was more effective than oral bambuterol.
Cockcroft and colleagues (253) compared the effectiveness
of 2-agonists and inhaled BUD on AHR with chronic dosing.
The study confirmed previous observations that chronic usage
of 2-agonists can induce tolerance to the protective effect
against methacholine challenge and can increase reactivity to
inhaled allergen. In contrast, BUD was associated with improved AHR. When the two therapies were combined, the inhaled corticosteroid was not able to block the increase in
AHR induced by the 2-agonist.
These studies clearly show that inhaled corticosteroids are
more beneficial in reducing nocturnal symptoms and reactivity of the airways than long-acting 2-agonists. Short-acting 2-agonists may have adverse effects on airway response to 2-agonist and on reactivity to antigen, effects that corticosteroids may not be able to block.
As discussed above, inhaled corticosteroids reduce airway
inflammation as measured by biopsies, induced sputum, and
exhaled NO. In contrast, salmeterol was found to have little
effect on eosinophils or mast cell tryptase in BAL fluid (254).
Similarly, terbutaline has also been shown to be ineffective in
improving airway inflammation (104, 255). Another study
showed corticosteroids were superior to terbutaline in improving airway epithelial cell metaplasia (104). It appears,
then, that inhaled corticosteroids are superior to 2-agonists in
modulating airway inflammation.
Cromones. Data are also available comparing inhaled corticosteroids with cromolyn sodium. In an open, parallel group study, FP was superior to cromolyn sodium in improving morning and evening peak flows and symptoms (216). Treatment with nedocromil sodium did not have any significant effect on inflammatory markers in bronchial biopsies of patients with asthma (256).
In summary, corticosteroids are superior to 2-agonists,
anticholinergics, and cromones, when compared head-to-head and examined for effects on AHR, FEV1, morning and
evening peak flow variation, rate of exacerbations, and airway
inflammation. However, additive effects may be obtained in
some patients when other therapy is combined with inhaled
corticosteroids.
Children
Generally, the beneficial effects of inhaled corticosteroids in children are more pronounced than for any other anti-asthma drug (216, 221, 224, 257). In two studies, children with mild asthma seen in general practice achieved markedly better symptom control and significantly higher morning and evening PEF rates and lung function during treatment with 50 µg FP twice daily as compared with children treated with cromolyn sodium 20 µg four times daily (216, 258) (Figure 9). In addition, FP treatment was associated with markedly fewer exacerbations. In three studies, 200 and 400 µg/d BDP was significantly more effective than continuous theophylline treatment in optimal doses and inhaled bronchodilators (224, 226, 227). Verbene and coworkers (257) compared 200 µg BDP twice daily with 50 µg salmeterol twice daily for 1 yr in children with mild and moderate asthma. Marked differences were seen in favor of BDP in reduction of AHR, number of acute exacerbations, and improvements in lung function, both before and after inhalation of a bronchodilator. A trend toward a decrease in lung function and a worsening in AHR over the year was seen in the patients treated with salmeterol, while significant improvements were seen in these parameters among the children receiving BDP.
|
p < 0.0001 (BUD was better than nedocromil sodium and cromolyn sodium in controlling asthma and reducing exacerbations in two studies (221, 259). In another long-term study, BUD was better than combinations of all other anti-asthma drugs in controlling symptoms and improving lung function, and in reducing peak flow variability, hospitalizations, and the use of other anti-asthma drugs (228). Furthermore, BUD significantly increased the growth rate of lung function to normal levels, while an annual decline was seen in percent predicted lung functions of the children not receiving