- Email: [email protected]
Leukotrienes in respiratory disease
PAEDIATRIC RESPIRATORY REVIEWS (2001) 2, 238–244 doi:10.1053/prrv.2001.0146, available online at http://www.idealibrary.com on
SERIES: BASIC PHARMACOLOGY
Leukotrienes in respiratory disease R. M. McMillan AstraZeneca, Mereside, Alderley Park, Macclesfield, SK10 4TG, UK KEYWORDS 5-lipoxygenase inhibitors, leukotriene B4 , cysteinyl leukotriene antagonists.
Summary Arachidonic acid metabolism via 5-lipoxygenase gives rise to a group of biologically active lipids known as leukotrienes: leukotriene B4 , which is a potent activator of leukocyte chemotaxis, and cysteinyl leukotrienes (leukotriene C4 , D4 and E4 ) which account for the spasmogenic activity previously described as slow-reacting substance of anaphylaxis. The biological actions of leukotrienes and the observations that leukotrienes are synthesised in the lung following antigen provocation and are elevated in asthma, stimulated considerable activity in the pharmaceutical industry to find drugs that modulate the synthesis or actions of leukotrienes. Three cysteinyl leukotriene antagonists (zafirlukast [Accolate], montelukast [Singulair] and pranlukast) and one 5-lipoxygenase inhibitor (zileuton) have received regulatory approval for the treatment of asthma. The clinical data obtained from using these drugs are generally consistent and complimentary. As a class the leukotriene modulators produce a rapid improvement in lung function after the first oral dose. Lung function improvements are maintained on chronic administration and are associated with reductions in a variety of asthma symptom scores. All of the available data are consistent with the hypothesis that all the leukotriene modulators exert their clinical benefit primarily through interference with cysteinyl leukotrienes. There are no compelling clinical data for an additional contribution by leukotriene B4 in human asthma. In other respiratory conditions such as COPD, which are characterised by pronounced neutrophil infiltration, it may be that the chemotactic properties of leukotriene B4 are more important and therefore evaluation of 5-lipoxygenase inhibitors in this condition is warranted. The introduction of the leukotriene modulators into clinical practice is the culmination of over 60 years of research since the initial discovery of the slow-reacting substances. The leukotriene modulators, and in particular the cysteinyl leukotriene antagonists, provide respiratory physicians with an oral therapeutic option and have set an efficacy standard which new oral anti-inflammatory approaches will have to beat. ° C 2001 Harcourt Publishers Ltd
INTRODUCTION Arachidonic acid metabolism via 5-lipoxygenase gives rise to a family of biologically active lipids known as leukotrienes (Fig. 1): leukotriene B4 , which is a potent activator of leukocyte chemotaxis, and the so-called cysteinyl leukotrienes, leukotriene C4 , leukotriene D4 and leukotriene E4 . The latter group accounts for the spasmogenic activities previously ascribed to slow-reacting substance(s) of anaphylaxis (SRS-A).
1526–0550/01/030238 + 07 $35.00/0
The biological activities of leukotriene B4 and the cysteinyl leukotrienes generated considerable interest in the pharmaceutical industry. Some companies already had been seeking antagonists of SRS-A. Now that it had been characterised structurally and shown to be related to another class of biologically active lipids, there was an explosion of interest. By the early 1980s, most major phamaceutical companies had initiated programmes to search for drugs that would modulate leukotrienes either by blocking their actions (cysteinyl leukotriene ° C 2001 Harcourt Publishers Ltd
LEUKOTRIENES IN RESPIRATORY DISEASE
Figure 1 Arachidonic acid metabolism via 5-lipoxyenase.
antagonists) or their biosynthesis (5-lipoxygenase inhibitors) Examples of both classes have been developed clinically for the treatment of asthma. In this review, the actions of the leukotrienes will be summarised briefly, then the status of both classes of leukotriene modulator will be described. Finally the clinical data supporting the use of leukotriene modulators in asthma will be summarised.
LEUKOTRIENES AND SRS-A This term ‘slow-reacting substance’ was first coined in 1938 to describe a substance which produced a pronounced contraction of smooth muscle which is slower in onset and longer in duration than that produced by histamine.1 Two years later a similar biological activity was shown to be released during anaphylactic shock and was termed ‘slow-reacting substance of anaphylaxis’ (SRS-A). Despite research over the next 40 years, the identity of SRS-A remained elusive. However, by the late 1970s a variety of biochemical and physical chemical properties had been discovered. For example, it was known that SRS-A contained a conjugated triene, that sulphur was present in the molecule, that SRSs could be formed in a variety of cells including leukocytes and that arachidonic acid was a precursor of SRS-A formation. Independent research at that time in the Karolinska Institute had focussed on novel pathways of arachidonic acid metabolism in leukocytes. Known pathways of arachidonate metabolism (i.e., cyclo-oxygenase, 12-lipoxygenase) were not prominent in leukocytes and Samuelsson and colleagues discovered that a novel enzymatic pathway, 5-lipoxygenase, predominated in leukocytes (for review see2 ). They discovered that 5-lipoxygenase catalysed the formation of two labile intermediates, 5-hydroperoxy eicosa-tetraenoic acid (5HPETE) and leukotriene A4 . Leukotriene A4 is an unstable epoxide which is subject to non-enzymatic hydrolysis to form a series of di-hydroxy acids. In addition enzymatic hydrolysis produces a 5,12di-hydroxy acid, leukotriene B4 . The term leukotriene was coined to indicate that these products are formed in leukocytes and that they contain conjugated trienes.
239
The similarities in some of the properties between SRS-A and the products described by Samuelsson were striking. Subsequently, Samuelsson’s colleagues and others carried out large scale purification and demonstrated that SRS-A was also formed from leukotriene A4 . In fact, SRS-A was not a single substance, but a group of lipids which contained a hydroxyl substituent at C5 and a sulphur-containing peptide at C6. They represented sequential metabolites containing either glutathione (leukotriene C4 ), cysteinyl glycine (leukotriene D4 ) or cysteine (leukotriene E4 ). The term cysteinyl leukotrienes was introduced to describe this group of metabolites. Biological studies with naturally produced leukotrienes, and later synthetically produced material, confirmed that the cysteinyl leukotrienes accounted for all the biological activity previously known as SRS-A, including spasmogenic activity in airway and intestinal smooth muscle as well as vascular inflammatory actions. The wide availability of synthetic materials facilitated evaluation of their potencies and also allowed clinical studies to be performed. The cysteinyl leukotrienes were demonstrated to be approximately 100 times more potent than histamine or methacholine in man when administered by aerosol and asthmatic patients were shown to exhibit enhanced sensitivity to the bronchoconstrictor effects of cysteinyl leukotrienes. Pro-inflammatory effects were also observed including, in the case of leukotriene E4 , induction of eosinophil recruitment into the lung when it was administered by inhalation to asthmatic volunteers. Identification of the receptors responsible for mediating the actions of cysteinyl leukotrienes eluded investigators for 20 years after their initial discovery. The term CYSLT1 was introduced in 1995, based on pharmacological characterisation of smooth muscle tissues, but it was not until 1999 that molecular cloning led to the identication of a 336 amino acid, G-protein-coupled membrane protein which exhibited the pharmacological properties of a CYS-LT1 receptor.3 In parallel studies, leukotriene B4 was recognised as a potent mediator of inflammation via its chemotactic activity. These effects were highly specific; thus isomers of leukotriene B4 with different stereochemistry or double bound geometry were much less potent or, in some cases, devoid of chemotactic activity. Leukotriene B4 induced chemotaxis of polymorphonuclear leukocytes (both neutrophils and eosinophils) in vitro. This also occurred in vivo and in some species leukotriene B4 acted synergistically with vasodilators to induce plasma exudation. Leukotriene B4 was shown to be a potent activator of human leukocytes and to induce inflammatory reactions in human volunteers. Like the cysteinyl leukotriene receptor, the binding site for leukotreine B4 (B-LT) was characterised pharmacologically prior to its molecular characterisation as a 352 amino acid, membrane-bound G-protein-coupled receptor.
240
Synthesis of leukotrienes has been observed in a range of animal models of allergy and inflammation which is not surprising since this was the way in which the activity of SRS-A was first discovered. It is also well established that leukotriene synthesis is a feature of human respiratory diseases. Human pulmonary tissue, lung mast cells and mononuclear leukocytes have the capacity to synthesise leukotrienes and cysteinyl leukotrienes are released during asthmatic attacks and are increased in bronchoalveolar lavage fluid during both early and late phase reactions following antigen provocation of allergic patients. There are several potential approaches to modulation of leukotrienes including 5-lipoxygenase inhibition, CYSLT1 antagonism and B-LT antagonism (Fig. 1). All these approaches, as well as other sites in the 5-lipoxygenase pathway, have been pursued with varying degrees of success. The following sections will describe only the approaches to CYS-LT1 antagonism and 5-lipoxygenase inhibition, since these have resulted in successful clinical development. To date, B-LT antagonists and other approaches have not proceeded beyond early clinical evaluation.
CYSTEINYL LEUKOTRIENE ANTAGONISTS Prior to the elucidation of the leukotriene structures, a number of companies had been engaged in the search for SRS antagonists and a prototypic compound FPL55712 (Fisons, now AstraZeneca) had been reported. The structural characterisation of SRS-A stimulated an enormous increase in pharmaceutical activity and by the early 1980s most pharmaceutical companies had initiated programmes to search for antagonists of cysteinyl leukotrienes. At that time there was no information on the leukotriene receptor genes and the synthetic challenges in producing leukotrienes meant that drug discovery programmes employed natural SRS-A preparations and conventional bio-assays. It was not until the mid-1980s that synthetic preparations of cysteinyl leukotrienes became generally available and the cloned human receptors did not arrive until after the launch of the first three cysteinyl leukotriene antagonists! The initiation of programmes to discover cysteinyl leukotriene antagonists also preceded the revolution in high throughput screening. As a result, companies relied on rational design which was based either on the structure of the Fisons’ compound FPL55712 or on the structures of the leukotrienes themselves. FPL55712 suffered from several major disadvantages: relatively weak potency, modest selectivity and poor oral bio-availability. Oral activity was an essential criterion for most cysteinyl leukotriene programmes. The most common asthma therapies (beta agonists and steriods) were and still are administered by inhalation. The main reason for this is concern about systemic side effects. Cysteinyl
R. M. McMILLAN
leukotriene antagonists were not expected to match the efficacy of inhaled steroids or beta agonists but their side effect liabilities were expected to be much lower. Therefore oral administration was regarded as feasible and was thought to be a significant advantage in terms of patient acceptability and compliance. In the last 15 years a large number of cysteinyl leukotriene antagonists have been evaluated in humans. The earliest compounds tested usually either lacked oral activity, and therefore had to be administered by inhalation, and/or they exhibited relatively short halflives which limited their clinical utility. However, during the 1990s three orally active and selective cysteinyl leukotriene antagonists were successfully developed and launched. These were ICI204219 (zafirlukast [Accolate] AstraZeneca), MK0476 (montelukast [Singulair] Merck) and ONO1078 (pranlukast, ONO). These structurally diverse antagonists (Fig. 2) produce similar pre-clinical pharmacological profiles which will be illustrated briefly here using the data for zafirlukast.5 Potent competitive antagonism of the actions of cysteinyl leukotrienes at cys-LT1 receptors was observed with Ki values ranging from 0.2 nM to 1 nM depending on the cysteinyl leukotriene used as agonist and the species studied. In contrast, zafirlukast is virtually inactive at cys-LT2 receptors as indicated by the Ki
Figure 2 Stuctures of Cysteinyl leukotriene antagonists.
LEUKOTRIENES IN RESPIRATORY DISEASE value of >3 µM for antagonism of binding of leukotriene C4 to human lung membranes. Selectivity is confirmed by the fact that zafirlukast at 10−5 M has virtually no activity against a wide range of other receptors including adrenergic, histaminergic, serotinergic, muscarinic and prostanoid. In vivo studies of CYS-LT1 antagonists were most commonly performed in guinea pigs.5 In that species zafirlukast demonstrated dose-dependent inhibition of dyspnoea induced by aerosolised LTD4 with an oral EC50 value of approximately 0.5 µmoles per kg. At this oral dose the effects of zafirlukast were maintained for up to 16 hours. Inhibition of allergic bronchospasm was demonstrated in several species. Thus zafirlukast inhibited the increase in pulmonary resistance and decrease in dynamic compliance induced by albumin in sensitised guinea pigs. These effects were dose- and time-dependent. In allergic sheep, zafirlukast inhibited both the early and the late phase of bronchospasm. A number of other in vivo responses were also blocked by zafirlukast. Most notable, in terms of respiratory diseases, was the antagonism of eosinophilia in guinea pig lungs induced by aerosolised leukotriene D4 . These data and similar results for pranlukast and montelukast supported the nomination of these CYS-LT1 antagonists for clinical evaluation in asthma.
5-LIPOXYGENASE INHIBITORS Lipoxygenases are enzymes that catalyse hydroperoxidation of fatty acids and other molecules. They are distinguished by the position from the carboxyl terminal at which the hydroperoxidation occurs. The most common forms of the enzymes are 5-, 12- and 15lipoxygenases. Lipoxygenases are widely distributed in both the plant and animal kingdom and they have been extensively studied since their first discovery in the 1930s. Indeed, soyabean lipoxygenase was one of the first enzymes obtained in pure crystalline form. The discovery of 5-lipoxygenase, and the subsequent demonstration of biologically active cysteinyl leukotrienes and leukotriene B4 , stimulated considerable activity in the pharmaceutical industry. Theoretically, 5-lipoxygenase inhibitors could have advantages over antagonists of either of the leukotriene classes since the actions of all the biologically active metabolites should be blocked. Moreover, since considerable mechanistic information was available from the plant lipoxygenases, it was envisaged that 5-lipoxygenase inhibitors would be more amenable to rational design. The amino acid sequence of 5-lipoxygenase was not derived until the enzyme was first cloned in 1988.6 Until then, purification from a variety of human and rat sources was employed. These studies demonstrated that the activity of the enzyme, in contrast to other lipoxygenases, was dependent on calcium and ATP. 5-lipoxygenase was
241
shown to be a 78 kDa protein containing 674 amino acids. The rat and human enzymes had 93% identity and they showed significant sequence homology with previously reported lipoxygenase enzymes. In mammalian cells, 5-lipoxygenase is normally present in the cytosol but on activation the enzyme undergoes calcium dependent translocation to the nuclear membrane. This translocation is required for cellular leukotriene biosynthesis. At the nuclear membrane 5-lipoxygenase becomes closely associated with a membrane protein known as 5-lipoxygenase activating protein (FLAP) which appears to be required for presentation of acachidonic acid to 5-lipoxygenase. Inhibitors of leukotriene synthesis has been an extremely active area and in terms of patent publication it has been even more productive than cysteinyl leukotriene antagonism. Several compounds have been evaluated clinically but to date only one (zileuton) has obtained successful regulatory approval.7 Most early reports of 5lipoxygenase inhibitors were redox-based inhibitors. Such compounds generally inhibited not only 5-lipoxygenase but also other lipoxygenases and a range of other oxidative enzymes, including cyclo-oxygenase. As in the case of CYS-LT1 antagonists, oral activity was considered essential for 5-lipoxygenase inhibitors. However, compounds of the redox class generally lacked suitable oral bioavailability and their poor safety profile precluded their clinical evaluation. Since the redox reaction catalysed by 5-lipoxygenase was iron-catalysed, a number of approaches using potential iron ligands were adopted. One such approach led to the development of zileuton (Abbot). Iron ligand inhibitors exhibited only modest selectively for 5-lipoxygenase compared to cyclo-oxygenase (10–30-fold). This was considered to be a potential disadvantage because of the well-known intolerance of some asthmatic patients to aspirin and related cyclo-oxygenase inhibitors. Two classes of leukotriene synthesis inhibitors have demonstrated markedly increased levels of selectively with essentially no inhibition of cyclo-oxygenase (for review see8 ). The first of these are the so-called nonredox 5-lipoxygenase inhibitors, exemplified by ZD2138 (AstraZeneca). The second group are FLAP antagonists, e.g., MK886 (Merck), which do not directly inhibit 5lipoxygenase but block leukotriene synthesis by binding to 5-lipoxygenase activating protein and, therefore, preventing activation of 5-lipoxygenase. Although nonredox inhibitors and FLAP antagonists have been evaluated in both volunteers and patients, no members of these classes have yet progressed beyond phase II studies. The pre-clinical pharmacology of leukotriene synthesis inhibitors has many similarities to that obtained with cysteinyl leukotriene antagonists. 5-lipoxygenase inhibitors and FLAP antagonists do not block the actions of administered leukotrienes but they are effective in models of allergic bronchospasm in both guinea pig and sheep. In
242
addition, such compounds exhibit activity in inflammatory models which are not sensitive to cysteinyl leukotriene antagonists and these effects have been ascribed to inhibition of synthesis of leukotriene B4 . Such actions include inhibition of neutrophil infiltration and vascular oedema in mouse or rabbit skin.
EVALUATION OF LEUKOTRIENE MODULATORS IN ASTHMA A substantial amount of data is now available on the effects of leukotriene modulators in asthma. Figure 3 illustrates the approach which has been taken to evaluation of either CYS-LT1 antagonists or 5-lipoxygenase inhibitors. After initial evaluation of safety and tolerability, the pharmacological activity of the molecules was demonstrated. In the case of CYS-LT1 antagonists, this involved evaluating the inhibitory effects of the antagonists on the bronchoconstrictor activity of inhaled leukotrienes. For 5-lipoxygenase inhibitors, activity was initially confirmed by measurement of ex vivo synthesis of leukotrienes. In later studies this was replaced by measurement of urinary leukotriene E4 levels which provided a better estimate of lung leukotriene synthesis than did the peripheral blood systems normally used for the ex vivo studies. The pharmacological efficacy models allowed selection of doses for future clinical evaluation. The first studies in asthma patients generally involved a range of challenge models
Figure 3 Clinical evaluation of leukotriene modulators.
R. M. McMILLAN
which allowed the study of the effects of compounds on the bronchoconstrictor effects of stimuli such as allergen, exercise and cold air. This approach was used to study a variety of early CYS-LT1 antagonists in the late 1980s including LY171883, L648051 and MK571. Modest beneficial effects were seen with these prototypic compounds but the levels of efficacy were insufficient to justify progression into prolonged evaluation in clinical asthma. However the studies did confirm the relevance of the challenge models for evaluation of leukotriene modulators and did increase the confidence that cysteinyl leukotrienes were important mediators of bronchospasm in humans. Subsequently, CYS-LT1 antagonists and 5-lipoxygenase inhibitors with improved potency and pharmacokinetics became available and a number of these compounds moved on to full-blown asthma trials. In this section the discussion will be limited to those compounds which have completed clinical development and have received regulatory approval, namely the CYS-LT1 antagonists zafirlukast, montelukast and pranlukast and the 5-lipoxygenase inhibitor zileuton. All four agents were effective against bronchoconstriction induced by allergen, cold air and exercise. In accord with earlier studies with weaker antagonists and inhibitors, the agents were effective against both the early and late phase of bronchoconstriction and also reduced hyperreactivity. Since 5-lipoxygenase inhibitors will prevent the
LEUKOTRIENES IN RESPIRATORY DISEASE
formation of both classes of leukotrienes, it has been speculated that such agents might provide greater efficacy than the CYS-LT1 antagonists. The demonstration that zileuton-surpressed allergen-induced eosinophilia in the lung was one of the first demonstrations of an anti-inflammatory effect of 5-lipoxygenase inhibitors.9 At the time it was speculated that this effect might arise from blocking synthesis of the chemotactic leukotriene B4 . However, later studies demonstrated that CYS-LT1 antagonists could also reduce allergen-induced cell infiltration. Thus zafirlukast has been shown to reduce accumulation of eosinophils, lymphocytes and basophils in bronchoalveolar lavage fluid after allergen challenge.10 The likely explanation of this effect of CYS-LT1 antagonism is related to the fact that leukotriene E4 stimulates eosinophil influx when administered by inhalation to asthmatics. The beneficial effects of the range of leukotriene modulators in challenge models provided confidence for taking CYS-LT1 antagonists and 5-lipoxygenase inhibitors forward to evaluation in asthma. A striking feature of the effects of leukotriene modulators in asthmatics is an acute improvement in baseline lung function which is observed rapidly after the first dose is administered. The magnitude of the effects vary with the treatment regimen and with the extent of lung function impairment in the patients being studies. However, improvements of 10–15% in FEV1 are commonly observed in studies with zafirlukast and montelukast. This improvement in lung function is additive with inhaled beta angonists and occurs in the presence or absence of inhaled corticosteroids. Similarly, the efficacy of zileuton in asthma involves both an acute and a chronic improvement in lung function. The acute improvement in FEV1 is similar in magnitude to that observed with the cysteinyl leukotriene antagonists and is observed within hours of the first oral dose.12 Since leukotriene modulators do not improve lung function in non-asthmatics, the acute effects cannot be due to a direct bronchodilator action in the smooth muscle. Rather it appears that synthesis of leukotrienes in the airways of asthmatics contributes to the bronchoconstrictor tone in the patients. In view of the speed of onset of the effects of leukotriene modulators, it would appear that at least part of this is related to direct effect of the cysteinyl leukotrienes on airway smooth muscle. However additional actions of cysteinyl leukotrienes and possibly leukotriene B4 on cellular components cannot be excluded. The improvement in FEV1 is maintained in longer term trials on CYS-LT1 antagonists or zileuton (3– 6 months’ duration). Improvement in a variety of asthma symptoms including night time awakening, morning asthma symptoms and total asthma symptoms have been demonstrated in three month studies of zafirlukast or montelukast.13 As with lung function, the improvements in symptoms varies depending on the nature of the
243
symptoms and the severity of the asthma being treated but they are generally in the range of 25–50%. A 4 week study of pranlukast in the United States has shown similar improvements in symptoms. Comparisons with the majority of the studies with pranlukast is more difficult since they have been carried out in Japan using a different protocol. A 3 month study of zileuton in steroid-na¨ıve patients with moderate asthma in the US confirmed the acute improvement in FEV1 and demonstrated an even more pronounced improvement in lung function following chronic treatment.12 This was associated with improvement in nocturnal asthma and beta agonist use and the magnitude of effects of zileuton were similar to those observed with CYS-LT1 antagonists. Since corticosteroids, both inhaled and oral, have become the mainstay of anti-inflammatory therapy in asthma there has been considerable interest in the interactions of leukotriene modulators and steroids. Several studies have attempted to investigate whether there is a ‘steroid sparing’ effect of these agents. Different approaches have been used. One protocol compared the effect of leukotriene modulators combined with a lowdose inhaled corticosteroid with patients who received a doubling of the steroid dose. In a 13 week study using this paradigm, zafirlukast-treated patients obtained a comparable improvement in asthma symptoms and morning peak expiratory flow to those who received the doubled steroid dose.12 An alternative approach is to investigate whether drug treatment can reduce the required steroid dose. Using this protocol, montelukast was shown to produce a small but significant reduction in inhaled steroid dose compared to patients receiving placebo rather than the leukotriene modulator.13 As described above, a large number of studies now support the additive effects of leukotriene modulators with inhaled steroids. Thus, the improvements of FEV1 produced by either cysteinyl leukotriene antagonists or zileuton are observed in patients maintained on inhaled corticosteroids and, in some cases, improvements in lung function even occur in asthmatics receiving oral prednisolone. A reduction in asthma exacerbations, defined as asthma attacks requiring treatment with oral steroids, has been shown in a 13-week study of montelukast.14 Meta-analysis of five clinical trials with zafirlukast has confirmed this. Despite the fact that the definition of exacerbations varied in the individual trials, a reduction of approximately 50% in exacerbations was reported in the zafirlukast-treated patients in this meta-analysis.15
STATUS OF LEUKOTRIENE MODULATORS During the last 15 years the role of leukotrienes in asthma has moved from concept to proof, with the successful development of three cysteinyl leukotriene antagonists and
244
a 5-lipoxygenase inhibitor. The clinical data obtained with these drugs and earlier studies with other leukotriene modulators resulted in the introduction of the first new class of asthma therapy for over 20 years. The clinical data obtained with the different drugs are generally consistent and complimentary. As a class the leukotriene modulators produce a rapid improvement in lung function after the first dose. Lung function improvements are maintained on chronic administration and are associated with reductions in a variety of asthma symptom scores. Data obtained with zileuton are generally indistinguishable from results obtained with the cysteinyl leukotriene antagonists. As a result, the latter have become the preferred leukotriene modulators. In the absence of any additional clinical benefit, the disadvantages of zileuton in terms of size and frequency of dose are too great. The absence of any compelling evidence of additional benefit of zileuton also calls into question the contribution played by leukotriene B4 in human asthma. All of the available data are consistent with the hypothesis that all the leukotriene modulators exert their clinical benefit solely through interference with cysteinyl leukotrienes. It may be that differential advantage of 5-lipoxygenase inhibitors will be observed in other respiratory diseases. In chronic obstructive pulmonary disease (COPD) there have been anecdotal reports of beneficial effects of cysteinyl leukotriene antagonists and recently the first controlled trial in this disease demonstrated that zafirlukast had a modest beneficial effect on lung function in COPD patients. Since COPD, in contrast to asthma, is characterised by pronounced neutrophil infiltration, it may be that the chemotactic properties of leukotriene B4 will be more important and, therefore, evaluation of 5-lipoxygenase inhibitors in this condition is warranted. The success of the leukotriene modulators in asthma is in contrast to disappointing results in clinical trials with other single mediator approaches such as PAF antagonists, which had produced at least as impressive a preclinical profile as the leukotriene modulators. The introduction of the leukotriene modulators into clinical practice is the culmination of over 60 years of research since the initial discovery of the SRSs. These agents, and in particular the cysteinyl leukotriene antagonists, not only provide respiratory physicians and paediatricians with an oral therapeutic option but have set an efficacy standard which new oral anti-inflammatory approaches will have to beat.
R. M. McMILLAN
REFERENCES 1. Feldberg W, Kelloway CH. Liberation of histamine and formation of lysolecithin-like substance by cobra venom. J Physiol 1938; 94: 187–226. 2. Samuelsson B. The discovery of leukotrienes. Am J Respir Crit Care Med 2000; 161 (suppl): 41–45. 3. Lynch KR, O’Neill GP, Liu Q, Im D-S, Sawyer N, Metters KM, Coulombe N, Abramovitz M, Gigueros DK, Zeng Z, Connolly BM, Bai C, Austin CP, Chateauneuf A, Stocco R, Greig GM, Kargman S, Hooks SB, Hosfield E, Williams DL Jr, Ford-Hutchinson AW, Caskey CT, Evans JF. Characterisation of the human cysteinyl leukotriene CysLT1 receptor. Nature 1999; 399: 789–793. 4. Adkins JC, Brogden RN: Zafirlukast. A review of its pharmacology and therapeutic potential in the management of asthma. Drugs 1998; 55: 1: 121–144 . 5. Carter GW, Young PR, Albert DH, Bouska J, Dyer R, Bell RL, Summers JB, Brooks DW. 5-Lipoxygenase inhibitory activity of zileuton. J Pharmacol Exp Ther 1991; 256: 929–937. 6. McMillan RM, Walker ERH. Designing therapeutically effective 5-lipoxygenase inhibitors. Trends Pharm Sci 1992; 13: 8: 323–330. 7. Kane GC, Pollice M, Kim C-J, Cohn J, Dworski RT, Murray JJ, Sheller JR, Fish JE, Peters SP. A controlled trial of the effect of the 5-lipoxygenase inhibitor, zileuton, on lung inflammation produced by segmental antigen challenge in human beings. J Allergy Clin Immunol 1996; 97: 646–654. 8. 9. Hui KP, Barnes NC. Lung function improvement in asthma with a cysteinyl leukotriene receptor antagonist. Lancet 337: 1062–1063. 10. Israel E, Rubin P, Kemp JP, Grossman J, Pierson W, Siegel SC, Tinkelman D, Murray JJ, Busse WW, Segal AT, Fish J, Kaiser HB, Ledford D, Wenzel SE, Rosenthal R, Cohn J, Lanni C, Pearlman H, Karahalios P, Drazen JM. The effect of inhibition of 5-lipoxygenase by zileuton in mild-to-moderate asthma. Ann Intern Med 1993; 119: 1059–1066. 11. Fish JE, Kemp RF, Lockey M, Glass M, Hanby LA, Bonuccelli CM. Zafirlukast improves symptomatic mild to moderate asthma: A 13 week multicenter study. Clin Ther 1997; 19: 675–690. 12. Nayak AS, Anderson P, Kharous BL, Williams K, Simonson S. Equivalence of adding zafirlukast versus double-dose inhaled corticosteroids in asthmatic patients symptomatic on low dose inhaled corticosteroids (abstract). J Allergy Clin Immunol 1998; 101(Pt2): S333. 13. Lofd¨ahl C-G, Reiss TF, Leff JA, Israel E, Noonan MJ, Finn AF, Seidersberg BC, Capizzi T, Kunda S, Godard P. Randomised, placebo controlled trial of effect of a leukotriene receptor antagonist, Montelukast on tapering inhaled corticosteroids in asthmatic patients. BMJ 1999; 319: 87–90. 14. Reiss TF, Chervinsky P, Dockhorn RJ, Shingo S, Seidenberg B, Edwards TB. Montelukast, a one-daily leukotriene receptor antagonist in the treatment of chronic asthma: a multicenter randomised, double-blind trial. Arch Intern Med 1998; 35: 57–62. 15. Barnes NC, Black B, Syrett N et al. Reduction of exacerbations of asthma in multinational trials with zafirlukast (Accolate). Allergy 51 1996; (Suppl 30): 84.
SERIES: BASIC PHARMACOLOGY
Leukotrienes in respiratory disease R. M. McMillan AstraZeneca, Mereside, Alderley Park, Macclesfield, SK10 4TG, UK KEYWORDS 5-lipoxygenase inhibitors, leukotriene B4 , cysteinyl leukotriene antagonists.
Summary Arachidonic acid metabolism via 5-lipoxygenase gives rise to a group of biologically active lipids known as leukotrienes: leukotriene B4 , which is a potent activator of leukocyte chemotaxis, and cysteinyl leukotrienes (leukotriene C4 , D4 and E4 ) which account for the spasmogenic activity previously described as slow-reacting substance of anaphylaxis. The biological actions of leukotrienes and the observations that leukotrienes are synthesised in the lung following antigen provocation and are elevated in asthma, stimulated considerable activity in the pharmaceutical industry to find drugs that modulate the synthesis or actions of leukotrienes. Three cysteinyl leukotriene antagonists (zafirlukast [Accolate], montelukast [Singulair] and pranlukast) and one 5-lipoxygenase inhibitor (zileuton) have received regulatory approval for the treatment of asthma. The clinical data obtained from using these drugs are generally consistent and complimentary. As a class the leukotriene modulators produce a rapid improvement in lung function after the first oral dose. Lung function improvements are maintained on chronic administration and are associated with reductions in a variety of asthma symptom scores. All of the available data are consistent with the hypothesis that all the leukotriene modulators exert their clinical benefit primarily through interference with cysteinyl leukotrienes. There are no compelling clinical data for an additional contribution by leukotriene B4 in human asthma. In other respiratory conditions such as COPD, which are characterised by pronounced neutrophil infiltration, it may be that the chemotactic properties of leukotriene B4 are more important and therefore evaluation of 5-lipoxygenase inhibitors in this condition is warranted. The introduction of the leukotriene modulators into clinical practice is the culmination of over 60 years of research since the initial discovery of the slow-reacting substances. The leukotriene modulators, and in particular the cysteinyl leukotriene antagonists, provide respiratory physicians with an oral therapeutic option and have set an efficacy standard which new oral anti-inflammatory approaches will have to beat. ° C 2001 Harcourt Publishers Ltd
INTRODUCTION Arachidonic acid metabolism via 5-lipoxygenase gives rise to a family of biologically active lipids known as leukotrienes (Fig. 1): leukotriene B4 , which is a potent activator of leukocyte chemotaxis, and the so-called cysteinyl leukotrienes, leukotriene C4 , leukotriene D4 and leukotriene E4 . The latter group accounts for the spasmogenic activities previously ascribed to slow-reacting substance(s) of anaphylaxis (SRS-A).
1526–0550/01/030238 + 07 $35.00/0
The biological activities of leukotriene B4 and the cysteinyl leukotrienes generated considerable interest in the pharmaceutical industry. Some companies already had been seeking antagonists of SRS-A. Now that it had been characterised structurally and shown to be related to another class of biologically active lipids, there was an explosion of interest. By the early 1980s, most major phamaceutical companies had initiated programmes to search for drugs that would modulate leukotrienes either by blocking their actions (cysteinyl leukotriene ° C 2001 Harcourt Publishers Ltd
LEUKOTRIENES IN RESPIRATORY DISEASE
Figure 1 Arachidonic acid metabolism via 5-lipoxyenase.
antagonists) or their biosynthesis (5-lipoxygenase inhibitors) Examples of both classes have been developed clinically for the treatment of asthma. In this review, the actions of the leukotrienes will be summarised briefly, then the status of both classes of leukotriene modulator will be described. Finally the clinical data supporting the use of leukotriene modulators in asthma will be summarised.
LEUKOTRIENES AND SRS-A This term ‘slow-reacting substance’ was first coined in 1938 to describe a substance which produced a pronounced contraction of smooth muscle which is slower in onset and longer in duration than that produced by histamine.1 Two years later a similar biological activity was shown to be released during anaphylactic shock and was termed ‘slow-reacting substance of anaphylaxis’ (SRS-A). Despite research over the next 40 years, the identity of SRS-A remained elusive. However, by the late 1970s a variety of biochemical and physical chemical properties had been discovered. For example, it was known that SRS-A contained a conjugated triene, that sulphur was present in the molecule, that SRSs could be formed in a variety of cells including leukocytes and that arachidonic acid was a precursor of SRS-A formation. Independent research at that time in the Karolinska Institute had focussed on novel pathways of arachidonic acid metabolism in leukocytes. Known pathways of arachidonate metabolism (i.e., cyclo-oxygenase, 12-lipoxygenase) were not prominent in leukocytes and Samuelsson and colleagues discovered that a novel enzymatic pathway, 5-lipoxygenase, predominated in leukocytes (for review see2 ). They discovered that 5-lipoxygenase catalysed the formation of two labile intermediates, 5-hydroperoxy eicosa-tetraenoic acid (5HPETE) and leukotriene A4 . Leukotriene A4 is an unstable epoxide which is subject to non-enzymatic hydrolysis to form a series of di-hydroxy acids. In addition enzymatic hydrolysis produces a 5,12di-hydroxy acid, leukotriene B4 . The term leukotriene was coined to indicate that these products are formed in leukocytes and that they contain conjugated trienes.
239
The similarities in some of the properties between SRS-A and the products described by Samuelsson were striking. Subsequently, Samuelsson’s colleagues and others carried out large scale purification and demonstrated that SRS-A was also formed from leukotriene A4 . In fact, SRS-A was not a single substance, but a group of lipids which contained a hydroxyl substituent at C5 and a sulphur-containing peptide at C6. They represented sequential metabolites containing either glutathione (leukotriene C4 ), cysteinyl glycine (leukotriene D4 ) or cysteine (leukotriene E4 ). The term cysteinyl leukotrienes was introduced to describe this group of metabolites. Biological studies with naturally produced leukotrienes, and later synthetically produced material, confirmed that the cysteinyl leukotrienes accounted for all the biological activity previously known as SRS-A, including spasmogenic activity in airway and intestinal smooth muscle as well as vascular inflammatory actions. The wide availability of synthetic materials facilitated evaluation of their potencies and also allowed clinical studies to be performed. The cysteinyl leukotrienes were demonstrated to be approximately 100 times more potent than histamine or methacholine in man when administered by aerosol and asthmatic patients were shown to exhibit enhanced sensitivity to the bronchoconstrictor effects of cysteinyl leukotrienes. Pro-inflammatory effects were also observed including, in the case of leukotriene E4 , induction of eosinophil recruitment into the lung when it was administered by inhalation to asthmatic volunteers. Identification of the receptors responsible for mediating the actions of cysteinyl leukotrienes eluded investigators for 20 years after their initial discovery. The term CYSLT1 was introduced in 1995, based on pharmacological characterisation of smooth muscle tissues, but it was not until 1999 that molecular cloning led to the identication of a 336 amino acid, G-protein-coupled membrane protein which exhibited the pharmacological properties of a CYS-LT1 receptor.3 In parallel studies, leukotriene B4 was recognised as a potent mediator of inflammation via its chemotactic activity. These effects were highly specific; thus isomers of leukotriene B4 with different stereochemistry or double bound geometry were much less potent or, in some cases, devoid of chemotactic activity. Leukotriene B4 induced chemotaxis of polymorphonuclear leukocytes (both neutrophils and eosinophils) in vitro. This also occurred in vivo and in some species leukotriene B4 acted synergistically with vasodilators to induce plasma exudation. Leukotriene B4 was shown to be a potent activator of human leukocytes and to induce inflammatory reactions in human volunteers. Like the cysteinyl leukotriene receptor, the binding site for leukotreine B4 (B-LT) was characterised pharmacologically prior to its molecular characterisation as a 352 amino acid, membrane-bound G-protein-coupled receptor.
240
Synthesis of leukotrienes has been observed in a range of animal models of allergy and inflammation which is not surprising since this was the way in which the activity of SRS-A was first discovered. It is also well established that leukotriene synthesis is a feature of human respiratory diseases. Human pulmonary tissue, lung mast cells and mononuclear leukocytes have the capacity to synthesise leukotrienes and cysteinyl leukotrienes are released during asthmatic attacks and are increased in bronchoalveolar lavage fluid during both early and late phase reactions following antigen provocation of allergic patients. There are several potential approaches to modulation of leukotrienes including 5-lipoxygenase inhibition, CYSLT1 antagonism and B-LT antagonism (Fig. 1). All these approaches, as well as other sites in the 5-lipoxygenase pathway, have been pursued with varying degrees of success. The following sections will describe only the approaches to CYS-LT1 antagonism and 5-lipoxygenase inhibition, since these have resulted in successful clinical development. To date, B-LT antagonists and other approaches have not proceeded beyond early clinical evaluation.
CYSTEINYL LEUKOTRIENE ANTAGONISTS Prior to the elucidation of the leukotriene structures, a number of companies had been engaged in the search for SRS antagonists and a prototypic compound FPL55712 (Fisons, now AstraZeneca) had been reported. The structural characterisation of SRS-A stimulated an enormous increase in pharmaceutical activity and by the early 1980s most pharmaceutical companies had initiated programmes to search for antagonists of cysteinyl leukotrienes. At that time there was no information on the leukotriene receptor genes and the synthetic challenges in producing leukotrienes meant that drug discovery programmes employed natural SRS-A preparations and conventional bio-assays. It was not until the mid-1980s that synthetic preparations of cysteinyl leukotrienes became generally available and the cloned human receptors did not arrive until after the launch of the first three cysteinyl leukotriene antagonists! The initiation of programmes to discover cysteinyl leukotriene antagonists also preceded the revolution in high throughput screening. As a result, companies relied on rational design which was based either on the structure of the Fisons’ compound FPL55712 or on the structures of the leukotrienes themselves. FPL55712 suffered from several major disadvantages: relatively weak potency, modest selectivity and poor oral bio-availability. Oral activity was an essential criterion for most cysteinyl leukotriene programmes. The most common asthma therapies (beta agonists and steriods) were and still are administered by inhalation. The main reason for this is concern about systemic side effects. Cysteinyl
R. M. McMILLAN
leukotriene antagonists were not expected to match the efficacy of inhaled steroids or beta agonists but their side effect liabilities were expected to be much lower. Therefore oral administration was regarded as feasible and was thought to be a significant advantage in terms of patient acceptability and compliance. In the last 15 years a large number of cysteinyl leukotriene antagonists have been evaluated in humans. The earliest compounds tested usually either lacked oral activity, and therefore had to be administered by inhalation, and/or they exhibited relatively short halflives which limited their clinical utility. However, during the 1990s three orally active and selective cysteinyl leukotriene antagonists were successfully developed and launched. These were ICI204219 (zafirlukast [Accolate] AstraZeneca), MK0476 (montelukast [Singulair] Merck) and ONO1078 (pranlukast, ONO). These structurally diverse antagonists (Fig. 2) produce similar pre-clinical pharmacological profiles which will be illustrated briefly here using the data for zafirlukast.5 Potent competitive antagonism of the actions of cysteinyl leukotrienes at cys-LT1 receptors was observed with Ki values ranging from 0.2 nM to 1 nM depending on the cysteinyl leukotriene used as agonist and the species studied. In contrast, zafirlukast is virtually inactive at cys-LT2 receptors as indicated by the Ki
Figure 2 Stuctures of Cysteinyl leukotriene antagonists.
LEUKOTRIENES IN RESPIRATORY DISEASE value of >3 µM for antagonism of binding of leukotriene C4 to human lung membranes. Selectivity is confirmed by the fact that zafirlukast at 10−5 M has virtually no activity against a wide range of other receptors including adrenergic, histaminergic, serotinergic, muscarinic and prostanoid. In vivo studies of CYS-LT1 antagonists were most commonly performed in guinea pigs.5 In that species zafirlukast demonstrated dose-dependent inhibition of dyspnoea induced by aerosolised LTD4 with an oral EC50 value of approximately 0.5 µmoles per kg. At this oral dose the effects of zafirlukast were maintained for up to 16 hours. Inhibition of allergic bronchospasm was demonstrated in several species. Thus zafirlukast inhibited the increase in pulmonary resistance and decrease in dynamic compliance induced by albumin in sensitised guinea pigs. These effects were dose- and time-dependent. In allergic sheep, zafirlukast inhibited both the early and the late phase of bronchospasm. A number of other in vivo responses were also blocked by zafirlukast. Most notable, in terms of respiratory diseases, was the antagonism of eosinophilia in guinea pig lungs induced by aerosolised leukotriene D4 . These data and similar results for pranlukast and montelukast supported the nomination of these CYS-LT1 antagonists for clinical evaluation in asthma.
5-LIPOXYGENASE INHIBITORS Lipoxygenases are enzymes that catalyse hydroperoxidation of fatty acids and other molecules. They are distinguished by the position from the carboxyl terminal at which the hydroperoxidation occurs. The most common forms of the enzymes are 5-, 12- and 15lipoxygenases. Lipoxygenases are widely distributed in both the plant and animal kingdom and they have been extensively studied since their first discovery in the 1930s. Indeed, soyabean lipoxygenase was one of the first enzymes obtained in pure crystalline form. The discovery of 5-lipoxygenase, and the subsequent demonstration of biologically active cysteinyl leukotrienes and leukotriene B4 , stimulated considerable activity in the pharmaceutical industry. Theoretically, 5-lipoxygenase inhibitors could have advantages over antagonists of either of the leukotriene classes since the actions of all the biologically active metabolites should be blocked. Moreover, since considerable mechanistic information was available from the plant lipoxygenases, it was envisaged that 5-lipoxygenase inhibitors would be more amenable to rational design. The amino acid sequence of 5-lipoxygenase was not derived until the enzyme was first cloned in 1988.6 Until then, purification from a variety of human and rat sources was employed. These studies demonstrated that the activity of the enzyme, in contrast to other lipoxygenases, was dependent on calcium and ATP. 5-lipoxygenase was
241
shown to be a 78 kDa protein containing 674 amino acids. The rat and human enzymes had 93% identity and they showed significant sequence homology with previously reported lipoxygenase enzymes. In mammalian cells, 5-lipoxygenase is normally present in the cytosol but on activation the enzyme undergoes calcium dependent translocation to the nuclear membrane. This translocation is required for cellular leukotriene biosynthesis. At the nuclear membrane 5-lipoxygenase becomes closely associated with a membrane protein known as 5-lipoxygenase activating protein (FLAP) which appears to be required for presentation of acachidonic acid to 5-lipoxygenase. Inhibitors of leukotriene synthesis has been an extremely active area and in terms of patent publication it has been even more productive than cysteinyl leukotriene antagonism. Several compounds have been evaluated clinically but to date only one (zileuton) has obtained successful regulatory approval.7 Most early reports of 5lipoxygenase inhibitors were redox-based inhibitors. Such compounds generally inhibited not only 5-lipoxygenase but also other lipoxygenases and a range of other oxidative enzymes, including cyclo-oxygenase. As in the case of CYS-LT1 antagonists, oral activity was considered essential for 5-lipoxygenase inhibitors. However, compounds of the redox class generally lacked suitable oral bioavailability and their poor safety profile precluded their clinical evaluation. Since the redox reaction catalysed by 5-lipoxygenase was iron-catalysed, a number of approaches using potential iron ligands were adopted. One such approach led to the development of zileuton (Abbot). Iron ligand inhibitors exhibited only modest selectively for 5-lipoxygenase compared to cyclo-oxygenase (10–30-fold). This was considered to be a potential disadvantage because of the well-known intolerance of some asthmatic patients to aspirin and related cyclo-oxygenase inhibitors. Two classes of leukotriene synthesis inhibitors have demonstrated markedly increased levels of selectively with essentially no inhibition of cyclo-oxygenase (for review see8 ). The first of these are the so-called nonredox 5-lipoxygenase inhibitors, exemplified by ZD2138 (AstraZeneca). The second group are FLAP antagonists, e.g., MK886 (Merck), which do not directly inhibit 5lipoxygenase but block leukotriene synthesis by binding to 5-lipoxygenase activating protein and, therefore, preventing activation of 5-lipoxygenase. Although nonredox inhibitors and FLAP antagonists have been evaluated in both volunteers and patients, no members of these classes have yet progressed beyond phase II studies. The pre-clinical pharmacology of leukotriene synthesis inhibitors has many similarities to that obtained with cysteinyl leukotriene antagonists. 5-lipoxygenase inhibitors and FLAP antagonists do not block the actions of administered leukotrienes but they are effective in models of allergic bronchospasm in both guinea pig and sheep. In
242
addition, such compounds exhibit activity in inflammatory models which are not sensitive to cysteinyl leukotriene antagonists and these effects have been ascribed to inhibition of synthesis of leukotriene B4 . Such actions include inhibition of neutrophil infiltration and vascular oedema in mouse or rabbit skin.
EVALUATION OF LEUKOTRIENE MODULATORS IN ASTHMA A substantial amount of data is now available on the effects of leukotriene modulators in asthma. Figure 3 illustrates the approach which has been taken to evaluation of either CYS-LT1 antagonists or 5-lipoxygenase inhibitors. After initial evaluation of safety and tolerability, the pharmacological activity of the molecules was demonstrated. In the case of CYS-LT1 antagonists, this involved evaluating the inhibitory effects of the antagonists on the bronchoconstrictor activity of inhaled leukotrienes. For 5-lipoxygenase inhibitors, activity was initially confirmed by measurement of ex vivo synthesis of leukotrienes. In later studies this was replaced by measurement of urinary leukotriene E4 levels which provided a better estimate of lung leukotriene synthesis than did the peripheral blood systems normally used for the ex vivo studies. The pharmacological efficacy models allowed selection of doses for future clinical evaluation. The first studies in asthma patients generally involved a range of challenge models
Figure 3 Clinical evaluation of leukotriene modulators.
R. M. McMILLAN
which allowed the study of the effects of compounds on the bronchoconstrictor effects of stimuli such as allergen, exercise and cold air. This approach was used to study a variety of early CYS-LT1 antagonists in the late 1980s including LY171883, L648051 and MK571. Modest beneficial effects were seen with these prototypic compounds but the levels of efficacy were insufficient to justify progression into prolonged evaluation in clinical asthma. However the studies did confirm the relevance of the challenge models for evaluation of leukotriene modulators and did increase the confidence that cysteinyl leukotrienes were important mediators of bronchospasm in humans. Subsequently, CYS-LT1 antagonists and 5-lipoxygenase inhibitors with improved potency and pharmacokinetics became available and a number of these compounds moved on to full-blown asthma trials. In this section the discussion will be limited to those compounds which have completed clinical development and have received regulatory approval, namely the CYS-LT1 antagonists zafirlukast, montelukast and pranlukast and the 5-lipoxygenase inhibitor zileuton. All four agents were effective against bronchoconstriction induced by allergen, cold air and exercise. In accord with earlier studies with weaker antagonists and inhibitors, the agents were effective against both the early and late phase of bronchoconstriction and also reduced hyperreactivity. Since 5-lipoxygenase inhibitors will prevent the
LEUKOTRIENES IN RESPIRATORY DISEASE
formation of both classes of leukotrienes, it has been speculated that such agents might provide greater efficacy than the CYS-LT1 antagonists. The demonstration that zileuton-surpressed allergen-induced eosinophilia in the lung was one of the first demonstrations of an anti-inflammatory effect of 5-lipoxygenase inhibitors.9 At the time it was speculated that this effect might arise from blocking synthesis of the chemotactic leukotriene B4 . However, later studies demonstrated that CYS-LT1 antagonists could also reduce allergen-induced cell infiltration. Thus zafirlukast has been shown to reduce accumulation of eosinophils, lymphocytes and basophils in bronchoalveolar lavage fluid after allergen challenge.10 The likely explanation of this effect of CYS-LT1 antagonism is related to the fact that leukotriene E4 stimulates eosinophil influx when administered by inhalation to asthmatics. The beneficial effects of the range of leukotriene modulators in challenge models provided confidence for taking CYS-LT1 antagonists and 5-lipoxygenase inhibitors forward to evaluation in asthma. A striking feature of the effects of leukotriene modulators in asthmatics is an acute improvement in baseline lung function which is observed rapidly after the first dose is administered. The magnitude of the effects vary with the treatment regimen and with the extent of lung function impairment in the patients being studies. However, improvements of 10–15% in FEV1 are commonly observed in studies with zafirlukast and montelukast. This improvement in lung function is additive with inhaled beta angonists and occurs in the presence or absence of inhaled corticosteroids. Similarly, the efficacy of zileuton in asthma involves both an acute and a chronic improvement in lung function. The acute improvement in FEV1 is similar in magnitude to that observed with the cysteinyl leukotriene antagonists and is observed within hours of the first oral dose.12 Since leukotriene modulators do not improve lung function in non-asthmatics, the acute effects cannot be due to a direct bronchodilator action in the smooth muscle. Rather it appears that synthesis of leukotrienes in the airways of asthmatics contributes to the bronchoconstrictor tone in the patients. In view of the speed of onset of the effects of leukotriene modulators, it would appear that at least part of this is related to direct effect of the cysteinyl leukotrienes on airway smooth muscle. However additional actions of cysteinyl leukotrienes and possibly leukotriene B4 on cellular components cannot be excluded. The improvement in FEV1 is maintained in longer term trials on CYS-LT1 antagonists or zileuton (3– 6 months’ duration). Improvement in a variety of asthma symptoms including night time awakening, morning asthma symptoms and total asthma symptoms have been demonstrated in three month studies of zafirlukast or montelukast.13 As with lung function, the improvements in symptoms varies depending on the nature of the
243
symptoms and the severity of the asthma being treated but they are generally in the range of 25–50%. A 4 week study of pranlukast in the United States has shown similar improvements in symptoms. Comparisons with the majority of the studies with pranlukast is more difficult since they have been carried out in Japan using a different protocol. A 3 month study of zileuton in steroid-na¨ıve patients with moderate asthma in the US confirmed the acute improvement in FEV1 and demonstrated an even more pronounced improvement in lung function following chronic treatment.12 This was associated with improvement in nocturnal asthma and beta agonist use and the magnitude of effects of zileuton were similar to those observed with CYS-LT1 antagonists. Since corticosteroids, both inhaled and oral, have become the mainstay of anti-inflammatory therapy in asthma there has been considerable interest in the interactions of leukotriene modulators and steroids. Several studies have attempted to investigate whether there is a ‘steroid sparing’ effect of these agents. Different approaches have been used. One protocol compared the effect of leukotriene modulators combined with a lowdose inhaled corticosteroid with patients who received a doubling of the steroid dose. In a 13 week study using this paradigm, zafirlukast-treated patients obtained a comparable improvement in asthma symptoms and morning peak expiratory flow to those who received the doubled steroid dose.12 An alternative approach is to investigate whether drug treatment can reduce the required steroid dose. Using this protocol, montelukast was shown to produce a small but significant reduction in inhaled steroid dose compared to patients receiving placebo rather than the leukotriene modulator.13 As described above, a large number of studies now support the additive effects of leukotriene modulators with inhaled steroids. Thus, the improvements of FEV1 produced by either cysteinyl leukotriene antagonists or zileuton are observed in patients maintained on inhaled corticosteroids and, in some cases, improvements in lung function even occur in asthmatics receiving oral prednisolone. A reduction in asthma exacerbations, defined as asthma attacks requiring treatment with oral steroids, has been shown in a 13-week study of montelukast.14 Meta-analysis of five clinical trials with zafirlukast has confirmed this. Despite the fact that the definition of exacerbations varied in the individual trials, a reduction of approximately 50% in exacerbations was reported in the zafirlukast-treated patients in this meta-analysis.15
STATUS OF LEUKOTRIENE MODULATORS During the last 15 years the role of leukotrienes in asthma has moved from concept to proof, with the successful development of three cysteinyl leukotriene antagonists and
244
a 5-lipoxygenase inhibitor. The clinical data obtained with these drugs and earlier studies with other leukotriene modulators resulted in the introduction of the first new class of asthma therapy for over 20 years. The clinical data obtained with the different drugs are generally consistent and complimentary. As a class the leukotriene modulators produce a rapid improvement in lung function after the first dose. Lung function improvements are maintained on chronic administration and are associated with reductions in a variety of asthma symptom scores. Data obtained with zileuton are generally indistinguishable from results obtained with the cysteinyl leukotriene antagonists. As a result, the latter have become the preferred leukotriene modulators. In the absence of any additional clinical benefit, the disadvantages of zileuton in terms of size and frequency of dose are too great. The absence of any compelling evidence of additional benefit of zileuton also calls into question the contribution played by leukotriene B4 in human asthma. All of the available data are consistent with the hypothesis that all the leukotriene modulators exert their clinical benefit solely through interference with cysteinyl leukotrienes. It may be that differential advantage of 5-lipoxygenase inhibitors will be observed in other respiratory diseases. In chronic obstructive pulmonary disease (COPD) there have been anecdotal reports of beneficial effects of cysteinyl leukotriene antagonists and recently the first controlled trial in this disease demonstrated that zafirlukast had a modest beneficial effect on lung function in COPD patients. Since COPD, in contrast to asthma, is characterised by pronounced neutrophil infiltration, it may be that the chemotactic properties of leukotriene B4 will be more important and, therefore, evaluation of 5-lipoxygenase inhibitors in this condition is warranted. The success of the leukotriene modulators in asthma is in contrast to disappointing results in clinical trials with other single mediator approaches such as PAF antagonists, which had produced at least as impressive a preclinical profile as the leukotriene modulators. The introduction of the leukotriene modulators into clinical practice is the culmination of over 60 years of research since the initial discovery of the SRSs. These agents, and in particular the cysteinyl leukotriene antagonists, not only provide respiratory physicians and paediatricians with an oral therapeutic option but have set an efficacy standard which new oral anti-inflammatory approaches will have to beat.
R. M. McMILLAN
REFERENCES 1. Feldberg W, Kelloway CH. Liberation of histamine and formation of lysolecithin-like substance by cobra venom. J Physiol 1938; 94: 187–226. 2. Samuelsson B. The discovery of leukotrienes. Am J Respir Crit Care Med 2000; 161 (suppl): 41–45. 3. Lynch KR, O’Neill GP, Liu Q, Im D-S, Sawyer N, Metters KM, Coulombe N, Abramovitz M, Gigueros DK, Zeng Z, Connolly BM, Bai C, Austin CP, Chateauneuf A, Stocco R, Greig GM, Kargman S, Hooks SB, Hosfield E, Williams DL Jr, Ford-Hutchinson AW, Caskey CT, Evans JF. Characterisation of the human cysteinyl leukotriene CysLT1 receptor. Nature 1999; 399: 789–793. 4. Adkins JC, Brogden RN: Zafirlukast. A review of its pharmacology and therapeutic potential in the management of asthma. Drugs 1998; 55: 1: 121–144 . 5. Carter GW, Young PR, Albert DH, Bouska J, Dyer R, Bell RL, Summers JB, Brooks DW. 5-Lipoxygenase inhibitory activity of zileuton. J Pharmacol Exp Ther 1991; 256: 929–937. 6. McMillan RM, Walker ERH. Designing therapeutically effective 5-lipoxygenase inhibitors. Trends Pharm Sci 1992; 13: 8: 323–330. 7. Kane GC, Pollice M, Kim C-J, Cohn J, Dworski RT, Murray JJ, Sheller JR, Fish JE, Peters SP. A controlled trial of the effect of the 5-lipoxygenase inhibitor, zileuton, on lung inflammation produced by segmental antigen challenge in human beings. J Allergy Clin Immunol 1996; 97: 646–654. 8. 9. Hui KP, Barnes NC. Lung function improvement in asthma with a cysteinyl leukotriene receptor antagonist. Lancet 337: 1062–1063. 10. Israel E, Rubin P, Kemp JP, Grossman J, Pierson W, Siegel SC, Tinkelman D, Murray JJ, Busse WW, Segal AT, Fish J, Kaiser HB, Ledford D, Wenzel SE, Rosenthal R, Cohn J, Lanni C, Pearlman H, Karahalios P, Drazen JM. The effect of inhibition of 5-lipoxygenase by zileuton in mild-to-moderate asthma. Ann Intern Med 1993; 119: 1059–1066. 11. Fish JE, Kemp RF, Lockey M, Glass M, Hanby LA, Bonuccelli CM. Zafirlukast improves symptomatic mild to moderate asthma: A 13 week multicenter study. Clin Ther 1997; 19: 675–690. 12. Nayak AS, Anderson P, Kharous BL, Williams K, Simonson S. Equivalence of adding zafirlukast versus double-dose inhaled corticosteroids in asthmatic patients symptomatic on low dose inhaled corticosteroids (abstract). J Allergy Clin Immunol 1998; 101(Pt2): S333. 13. Lofd¨ahl C-G, Reiss TF, Leff JA, Israel E, Noonan MJ, Finn AF, Seidersberg BC, Capizzi T, Kunda S, Godard P. Randomised, placebo controlled trial of effect of a leukotriene receptor antagonist, Montelukast on tapering inhaled corticosteroids in asthmatic patients. BMJ 1999; 319: 87–90. 14. Reiss TF, Chervinsky P, Dockhorn RJ, Shingo S, Seidenberg B, Edwards TB. Montelukast, a one-daily leukotriene receptor antagonist in the treatment of chronic asthma: a multicenter randomised, double-blind trial. Arch Intern Med 1998; 35: 57–62. 15. Barnes NC, Black B, Syrett N et al. Reduction of exacerbations of asthma in multinational trials with zafirlukast (Accolate). Allergy 51 1996; (Suppl 30): 84.