Antileukotriene agents in the management of asthma

INTRODUCTION — Leukotrienes are powerful lipid mediators of inflammation generated principally by myeloid cells and mucosal chemosensory cells. Their name derives from their cells of origin (leukocytes) and a positionally conserved triad of double bonds (triene). The potency of the leukotrienes and their production during inflammatory responses, including the airway inflammation of asthma, made them a promising molecular target for therapeutic development.

This topic reviews the biology of the leukotrienes, their role in asthma and inflammation, and the use of 5-lipoxygenase (5-LO) inhibitors and cysteinyl leukotriene (CysLT) receptor antagonists as treatments for asthma, rhinitis, and other allergic diseases. An overview of asthma management and the integrated use of asthma therapies are discussed separately. (See "An overview of asthma management in children and adults".)

LEUKOTRIENE BIOLOGY

Synthesis — Arachidonic acid is a 20-carbon polyunsaturated fatty acid liberated by activated cells and is the substrate from which both leukotrienes and prostaglandins are generated. In myeloid cells, a substantial fraction of released arachidonate is presented by the membrane-bound 5-lipoxygenase-activating protein (FLAP) [1] to 5-lipoxygenase (5-LO), an enzyme that translocates to the nuclear membrane to catalyze the formation of leukotriene A₄ (LTA₄) (figure 1) [2]. LTA₄ is an unstable epoxide, which depending upon the cell type of origin, is either hydrolyzed by a specific LTA₄ hydrolase (LTA₄H) [3] to form LTB₄, a potent chemoattractant for neutrophils and some T cell populations [4], or is conjugated to reduced glutathione by membrane-bound leukotriene C₄ synthase (LTC₄S) [5] to form LTC₄. LTC₄ is the parent molecule of the cysteinyl leukotrienes (CysLTs). CysLTs are a potent class of bronchoconstrictors and activators of innate immune effector cells [6] and have a validated role in asthma.

●LTB₄ – The roles of LTB₄ and its receptor in asthma are not clear. LTB₄ is the dominant 5-LO pathway product of neutrophils [7] and is also generated by macrophages [8] and mast cells [9]. One study suggested a weak capacity of eosinophils from patients with severe asthma to generate LTB₄ as well [10]. Once converted by LTA₄H, LTB₄ is released to the extracellular space, where it exerts its actions at two different G protein coupled receptors (GPCRs), termed BLT₁ and BLT₂, respectively [11,12]. LTB₄ is a potent chemoattractant for neutrophils in both humans and in mice [13]. It is also chemotactic for populations of CD4+ [4,14] and CD8+ effector memory cells [15], including a population of interleukin (IL)-13-expressing CD8+ cells that are present in the airways of patients with severe asthma [16]. Studies using gene-deleted mice and selective receptor antagonists suggest that BLT₁ is the dominant receptor for LTB₄ in mediating chemotaxis of leukocytes and suggest that this receptor may be involved in facilitating eosinophilic airway pathology [15]. Although LTB₄ can be detected in the bronchoalveolar lavage fluid from severely asthmatic subjects [17], no studies have yet addressed the role of LTB₄ or BLT₁ receptors in asthma due to a lack of approved receptor antagonists.

●CysLTs – Mast cells, eosinophils, and basophils, the principal effector cells of type 2 (eosinophil-rich) immunopathology, all express LTC₄S, and they preferentially or exclusively generate LTC₄ rather than LTB₄ when activated [18-20]. Macrophages and myeloid dendritic cells also express LTC₄S [21,22], as do platelets [23]. Platelets lack FLAP and 5-LO but can convert LTA₄ from granulocytes to LTC₄, especially during inflammatory responses when they often form P-selectin-dependent complexes with granulocytes in the peripheral circulation [24]. In asthma, the frequencies of these platelet-leukocyte complexes are higher than in nonasthmatic controls [25] and are markedly increased in the blood and sinonasal tissues of patients with aspirin-exacerbated respiratory disease (AERD; a clinical triad of asthma, nasal polyps, and pathognomonic respiratory reactions to the ingestion of aspirin and other nonselective inhibitors of cyclooxygenase, which is associated with especially high levels of CysLT production) [23]. (See "Aspirin-exacerbated respiratory disease".)

Although most nonhematopoietic cells cannot generate leukotrienes, specialized populations of rare chemosensory cells in the intestine (tuft cells) and in the airway (brush cells) express FLAP, 5-LO, and LTC₄S and generate substantial quantities of LTC₄ on a per cell basis when activated [26,27]. The diverse cell sources of LTC₄ imply complex functions that vary depending on context. Once released, LTC₄ is converted by a gamma glutamyl leukotrienase to LTD₄ by removal of the glutamic acid residue from the glutathione adduct [28]. LTD₄ is extremely short lived due to its rapid conversion by extracellular dipeptidases to the stable metabolite, LTE₄, by removal of glycine [29]. LTE₄ is excreted in the urine without further modification [30] and is often used a marker of systemic production of CysLTs in clinical studies [31,32].

All three CysLTs are bioactive mediators, stimulating three different cognate GPCRs that differ in ligand preference and distribution of expression [33].

•The type 1 CysLT receptor (CysLT₁R) is the target of the available receptor antagonists approved for the treatment of asthma and allergic rhinitis [34,35]. It has high affinity for LTD₄ and is expressed by airway and vascular smooth muscle and several immune cells [36-38]. CysLT₁R is responsible for the bronchoconstricting effects of the CysLTs in human subjects, as well as the activation of certain relevant immune cells (mast cells, innate group 2 lymphoid cells, T helper type 2 [Th2] cells) [39-43].

•CysLT₂R, expressed by endothelial cells, brain, cardiac Purkinje cells, and leukocytes, recognizes LTC₄ and LTD₄ with equal affinity [44], while CysLT₃R (also known as GPR99) is expressed by epithelial cells of the airway and proximal convoluted tubule [45,46]. Although animal models suggest that CysLT₂R and CysLT₃R both play important roles in allergic pulmonary inflammation [47,48], knowledge of their role in human asthma is limited by the absence of selectively targeted drugs.

●Variations in leukotriene synthesis – Human subjects display a wide range of variability in leukotriene pathway activity, likely reflecting genetic, microenvironmental, and hormonal influences. Leukocytes from female subjects generate higher quantities of leukotrienes than do those from male subjects, presumably reflecting the ability of androgens to suppress the assembly of leukotriene synthetic enzymes [49].

Polymorphic variants in either the coding or the promoter regions of 5-LO, FLAP, LTA₄ hydrolase, and LTC₄S may affect enzyme expression and function [50-53]. A 5-LO promoter polymorphism predicts the therapeutic response to the 5-LO inhibitor zileuton. A common promoter polymorphism in the LTC4S gene may increase the expression of this enzyme and is associated with increased LTC₄ synthesis by eosinophils ex vivo [54]. This polymorphism was originally reported as a risk allele for AERD [53], although subsequent studies failed to replicate this association [55].

Bronchial biopsies from patients with AERD display sharply increased expression of LTC₄S protein that localizes to eosinophils [56] and correlates with the severity of reactions to aspirin challenge. Additionally, blood and sinonasal tissue granulocytes from patients with AERD show sharply increased percentages of adherent LTC₄S-expressing platelets [23]. IL-4, a key cytokine involved in type 2 immunity, potently upregulates the expression of LTC₄S by mast cells and dramatically enhances their capacity to generate LTC₄ when activated [57]. IL-5 elicits the translocation of 5-LO to the nucleus of human eosinophils [58] and mast cells [57], priming them for augmented agonist-induced LTC₄ production. Collectively, these findings suggest a complex regulation of leukotriene pathway activity in human subjects. Individual variations in CysLT production (expressed as a ratio of urinary LTE₄ to exhaled nitric oxide levels) are reported to predict response to CysLT₁R antagonists [59].

Functions of leukotrienes relevant to asthma pathogenesis

Bronchoconstriction — CysLTs are the most powerful bronchoconstrictors known to exist. LTC₄ and LTD₄ exceed the potencies of histamine and methacholine by approximately 1000-fold [60,61]. Asthmatic and nonasthmatic subjects display similar degrees of airway sensitivity to LTC₄ and LTD₄ [61,62], which elicit constriction directly through airway smooth muscle-associated CysLT₁R.

Supporting the physiologic importance of CysLTs as bronchoconstrictors in asthma, CysLT₁R inhibitors reduce bronchoconstriction, as shown in the following studies:

●In atopic asthmatic subjects, the selective CysLT₁R antagonist pranlukast (available in Japan) attenuates the decline in forced expiratory volume in one second (FEV₁) induced by inhalation challenges with specific allergens by approximately 50 percent [63], and it also blocks bronchoconstriction elicited by exercise [64], cold air [65], and inhalations of adenosine [66] or mannitol [67], all of which elicit LTC₄ synthesis in vivo.

●Patients with asthma presenting to the emergency department with exacerbations display increased levels of urinary LTE₄ (figure 2) [68]. Intravenously administered montelukast, another selective CysLT₁R antagonist, significantly increased peak expiratory flow in emergency department patients presenting with asthma attacks, although it failed to alter discharge rates [69,70].

Patients with asthma display selectively enhanced sensitivity to LTE₄-induced bronchoconstriction compared with nonasthmatic controls [71], even though LTE₄ is a less potent constrictor than its precursors (likely reflecting its weak binding affinity for CysLT₁R) [36,72]. Subjects with AERD, display even greater degrees of LTE₄ hyperresponsiveness than do aspirin tolerant asthmatic subjects [73,74] without differences in reactivity to histamine or LTC₄.

The basis for selective LTE₄ hyperresponsiveness is unknown but could reflect increased numbers of receptors or increased numbers of LTE₄-responsive cell types. The numbers of CysLT₁R-expressing cells in the airways of subjects with AERD exceed those observed in aspirin-tolerant asthmatic controls [75,76]. Notably, therapeutic desensitization to aspirin, a procedure that improves disease control in many subjects with AERD, sharply reduces CysLT₁R expression in sinonasal tissues [75] and elicits a loss of end-organ LTE₄ hyperresponsiveness [74]. This may explain why asthma frequently improves in patients with AERD after desensitization to aspirin [77], despite substantially increased levels of urinary LTE₄ compared with pretreatment levels [78]. (See "Aspirin-exacerbated respiratory disease" and "Aspirin-exacerbated respiratory disease: NSAID challenge and desensitization".)

Mucus secretion — Studies of mucus secretion provide evidence for the presence of a third CysLT receptor (CysLT₃R), in addition to CysLT₁R and CysLT₂R, and for a role of platelets in LTE₄-induced mucin elaboration. Both LTC₄ and LTD₄ elicit the release of mucus from human bronchi ex vivo [79].

●In mouse models of allergen-induced pulmonary disease, exogenous administration of LTE₄ markedly increases the numbers of mucus-producing goblet cells in the airways [80]. Notably, this response persists even in mice lacking both CysLT₁R and CysLT₂R, suggesting the involvement of an additional receptor, such as CysLT₃R [80].

●The above response to LTE₄ requires both platelets and the platelet-associated adenosine diphosphate receptor (P₂Y₁₂) [80]. Because LTE₄ neither binds P₂Y₁₂ nor activates platelets, these findings argue for an indirect, downstream role of platelets.

●Intranasal administration of LTE₄ to immunologically naïve mice causes the release of mucins from nasal goblet cells [46]. This response is abolished in mice lacking CysLT₃R, suggesting that at least some of the direct stimulatory effects of the CysLTs on mucus production may be due to this receptor [80].

Mast cell activation — Mast cell activation by LTE₄ is largely mediated through CysLT₁R, although interactions between CysLT₁R and CysLT₂R on mast cells may modulate the response to LTE₄.

●Urinary levels of a prostaglandin D₂ (PGD₂) metabolite, a marker of mast cell activation, increase sharply in response to inhalation of LTE₄ by human asthmatic subjects [39]. Both the increase in PGD₂ and the reduction in FEV₁ elicited by LTE₄ challenge are blocked by the administration of the selective CysLT₁R antagonist montelukast. Stimulation of human cord blood-derived mast cells with CysLTs ex vivo results in strong calcium flux, production of cytokines and chemokines, and production of PGD₂, responses that are blocked by CysLT₁R antagonists [81,82].

●Like many hematopoietic cells, mast cells express both CysLT₁R and CysLT₂R. CysLT₂R heterodimerizes with CysLT₁R on the surfaces of mast cells, restricting signaling and surface expression of the latter receptor [83]. Thus, CysLT-induced mast cell activation in vivo may be controlled by a balance of CysLT₁R and CysLT₂R expression. Notably, CysLT₁R expression, but not CysLT₂R expression, is upregulated in nasal and bronchial tissue mast cells in subjects with AERD [75,76], suggesting a potential mechanism involved in the enhanced reactivity of these subjects to LTE₄. Whether this reactivity depends on secondary production of bronchoconstricting agonists (such as PGD₂) by mast cells is not known.

●Proliferation of mast cells in epithelial surfaces is a feature of Th2-associated allergic disease. CysLTs contribute to modest proliferation of human and mouse mast cells ex vivo through CysLT₁R-dependent transactivation of the Kit tyrosine kinase [84].

Dendritic cell priming — Dendritic cells are antigen-presenting hematopoietic cells that are important in priming T lymphocytes for the development of Th2 memory responses. They reside in close proximity to the respiratory mucosa, capturing antigens for presentation to T cells in the regional lymph nodes [85]. Dendritic cells express all components of the CysLT synthetic pathway [21], as well as both CysLT₁R and CysLT₂R [86,87].

When stimulated with the clinically relevant allergens from the house dust mites Dermatophagoides farinae and Dermatophagoides pteronyssinus or the mold Aspergillus fumigatus, dendritic cells generate substantial quantities of CysLTs [20] and acquire the ability to prime naïve mouse T lymphocytes for subsequent generation of Th2 memory responses [87]. Notably, dendritic cells lacking either LTC₄S (which are incapable of generating LTC₄) or CysLT₁R display markedly blunted capacity to elicit Th2 priming in vivo [87].

Conversely, dendritic cells lacking CysLT₂R show enhanced Th2 priming, possibly due to the loss of CysLT₂R interaction with CysLT₁R, which can downregulate surface expression and signaling through CysLT₁R, as is the case for mast cells [86] (see 'Mast cell activation' above). Although these findings suggest that agents that block CysLT production or CysLT₁R signaling could hold promise in the prevention of early life sensitization to common antigens that are associated with asthma, no human studies have addressed this possibility. One small, controlled trial demonstrated reductions in total serum immunoglobulin E (IgE) in asthmatic children treated with the CysLT₁R antagonist montelukast [88].

Lymphocyte activation — Human Th2 cells express CysLT₁R to a greater extent than do Th1 cells [40] and respond to LTE₄ by generating IL-5 and IL-13. This response is substantially potentiated by simultaneous stimulation with PGD₂ [40], suggesting that lipids generated by activated mast cells are sufficient to elicit polyclonal activation of Th2 cells.

CysLTs may potentiate type 2 immune responses to helminths and other stimuli that breach epithelial barriers and activate group 2 innate lymphoid cells (ILC2s). ILC2s are poised effector cells that lack T cell receptors and can generate large quantities of IL-5, IL-13, and other cytokines when activated by epithelial-derived cytokines (eg, IL-25, IL-33, thymic stromal lymphopoietin) [89]. ILC2s express CysLT₁R and respond to exogenous CysLTs by generating IL-4, IL-5, and IL-13 [41,42,90]. Additionally, CysLTs markedly potentiate the production of cytokines by ILC2s stimulated with IL-25 or IL-33 [27,91]. Additionally, CysLT₁R antagonists and 5-LO inhibitors may exert part of their therapeutic effects by attenuating cytokine secretion by lymphoid effectors.

Eosinophil and basophil recruitment — Experimental inhalation of LTE₄ by subjects with mild to moderate asthma results in sustained increases in eosinophils [92,93] and basophils [93] in induced sputum and bronchial biopsies. This response is attenuated by treatment with CysLT₁R antagonists [94]. Murine studies suggest that effector cell recruitment involves input from all three CysLT receptors [47,48,93,95,96].

CLINICAL USE OF LEUKOTRIENE-MODIFYING DRUGS IN ASTHMA — Antileukotriene agents reduce asthmatic responses to bronchoprovocation challenge, as expected based on the effects of leukotrienes described above. (See 'Functions of leukotrienes relevant to asthma pathogenesis' above.).

Physiologic findings of airflow obstruction induced by exercise, allergen, aspirin, and the pollutant sulfur dioxide are all substantially attenuated by pretreatment with either cysteinyl leukotriene-1 (CysLT₁) antagonists or the 5-lipoxygenase (5-LO) inhibitor, zileuton [97-101]. In the case of allergen-induced asthmatic reactions, both early- and late-phase bronchoconstrictor responses are attenuated [102]. These agents are also useful in long-term asthma control.

Approaches to the treatment of asthma in children and adults are provided separately. (See "An overview of asthma management in children and adults" and "Asthma in children younger than 12 years: Management of persistent asthma with controller therapies".)

Available antileukotriene agents — In the United States, three antileukotriene agents (also known as leukotriene modifying agents) are available for the treatment of asthma (table 1). Availability in other countries varies, as specified below. Safety in pregnancy and lactation are discussed below. (See 'Safety in pregnancy and breastfeeding' below.)

●Montelukast and zafirlukast are cysteinyl leukotriene (CysLT₁) receptor antagonists. Pranlukast, available in Japan, is another CysLT₁ receptor antagonist. In most of Europe, only montelukast is available. Montelukast is administered once daily, and zafirlukast and pranlukast are administered twice daily; the usual doses are provided in the table (table 1). Montelukast is approved for use in children older than one year, while zafirlukast is approved for children older than five years.

●Zileuton is an inhibitor of 5-LO and thus acts “further upstream” and inhibits the formation of both the CysLTs (and, hence, attenuates stimulation of all three CysLT receptors) and other 5-LO metabolites including LTB₄ (figure 1). This agent is only available in the United States and is approved for use in children 12 years and older. Zileuton is available in two preparations: immediate release (600 mg, four times daily) and controlled release (1200 mg, twice daily) (table 1) [103]. Regular monitoring of serum alanine aminotransferase is recommended, as described below. (See 'Adverse effects' below.)

Comparisons among antileukotriene agents — Comparisons among antileukotriene agents are limited, but include the following observations:

●Montelukast versus zafirlukast – There are no efficacy data favoring either one of the CysLT₁ antagonists available in the United States (montelukast and zafirlukast) over the other (table 1). However, montelukast is commonly preferred because it is once daily and can be taken at any time in relation to meals.

●Zileuton versus montelukast – In a head-to-head comparison, 210 patients with mild to moderate asthma on no other medications were assigned to take the 5-LO inhibitor, zileuton or montelukast, for 12 weeks [104]. Zileuton demonstrated slightly greater efficacy than montelukast in improving peak flow (64.8 L/min [95% CI 54.8-74.7] versus 40.6 L/min [95% CI 31.3-49.9], respectively) and symptom scores, with comparable tolerability. This study supports the theoretical advantage of zileuton over the receptor antagonists since it inhibits production of all the CysLTs and other 5-LO metabolites including LTB₄, although the magnitude of the difference appears to be small (figure 1). (See 'Leukotriene biology' above.)

We are aware of a number of anecdotal experiences with patients who clinically benefited from zileuton after failing CysLT₁ antagonist therapy, particularly those with aspirin-exacerbated respiratory disease (AERD). Post-hoc analysis of data from clinical trials suggests that zileuton has a greater impact in patients with more severe rather than milder airflow obstruction, although conflicting data also exist [105,106]. (See 'Special populations' below.)

Use as initial controller therapy — Initial controller therapies for asthma include two groups of medications: inhaled glucocorticoids (inhaled GCs) and antileukotriene agents [107,108]. Both have beneficial effects on clinical symptoms and lead to decreases in markers of inflammation, although these properties have been best characterized for inhaled GCs. (See "Initiating asthma therapy and monitoring in adolescents and adults", section on 'Alternative approach: leukotriene receptor antagonists (LTRAs) plus SABA' and "Asthma in children younger than 12 years: Management of persistent asthma with controller therapies", section on 'Daily LTRA'.)

Predicting response — Individuals differ in their response to antileukotriene agents. Some patients respond dramatically, but others show no response at all. Symptom-based outcomes appear to be favorably affected in 60 to 80 percent of patients, whereas improvements in lung function are noted in a lower proportion of subjects (35 to 50 percent) [109,110].

Therapeutic trial — A therapeutic trial is necessary to assess the utility of an antileukotriene agent in any individual patient, since it is impossible to reliably predict responsiveness. It is our practice to employ a one to two month trial, regardless of which agent is selected and whether the leukotriene modifier is being utilized as first-line or add-on therapy. Both objective and subjective parameters must be assessed in order to comprehensively gauge benefit [111,112].

It is possible to see benefit as early as the first day of treatment with antileukotriene agents in both adults and children [34,113,114]. This likely reflects the intrinsic bronchodilator activity of antileukotrienes, and the fact that leukotriene overproduction at baseline contributes to increased bronchial tone. An anti-inflammatory effect, as judged by reductions in exhaled nitric oxide and bronchial hyperresponsiveness, has been observed within one and two weeks of initiation of therapy, respectively [115,116]. Improvements in lung function and symptoms may occur more gradually over several weeks to months [112,117].

The kinetics of improvement for different endpoints within individual patients have not been studied.

Relative efficacy

Compared with placebo — Antileukotriene agents can be effective as monotherapy in the treatment of mild-to-moderate persistent asthma. Both zileuton and CysLT₁R antagonists elicit rapid and sustained effects on measures of airflow (figure 3). Each agent has been shown in randomized trials to be superior to placebo in the following outcome measures [97,103,113,118-121]:

●Lung function (typically resulting in a 10 to 15 percent improvement in forced expiratory volume in one second [FEV₁] and improvement in measures of distal lung function, with a >10 percent improvement generally being considered to be a minimally important difference)

●Daytime and nighttime asthma symptoms and asthma-specific quality of life

●Need for rescue beta-agonist therapy

●Frequency of asthma exacerbations

Monotherapy with leukotriene receptor antagonists was evaluated in a 2015 systematic review and meta-analysis of studies involving adults and adolescents with mild asthma and reduced the risk of exacerbation (risk ratio [RR] 0.60, 95% CI 0.44-0.81) [120]. However, the review noted a high risk of bias in some studies, in addition to heterogeneity in results and an inability to assess the effect of asthma severity on the results.

Montelukast and zafirlukast are also effective in children. Specifically, children with moderate persistent asthma who were treated with montelukast experienced improved lung function at baseline, decreased asthma symptoms, increased asthma-specific quality of life, and decreased need for asthma rescue medication use, compared with placebo [34,122,123].

A randomized trial of zileuton monotherapy in adults with mild-to-moderate asthma demonstrated an increase in FEV₁ of 0.32 L (CI 0.16-0.48 L), a 13 percent increase, compared with a 0.05 L (CI -0.10 to 0.20 L) increase in patients given placebo [97]. Another trial found that after 12 weeks of therapy, FEV₁ improved by a mean of 0.39 L (20 percent) and 0.27 L (13 percent) with zileuton and placebo, respectively [103]. Symptoms and frequency of beta-agonist use also decreased with zileuton [97,103].

No evidence of tolerance to these beneficial effects has been noted during treatment for periods up to two years.

Compared with inhaled glucocorticoids — Leukotriene modifiers exert anti-inflammatory effects, such as reducing the numbers of circulating and sputum eosinophils, exhaled nitric oxide, and nonspecific bronchial hyperresponsiveness, in addition to their bronchodilatory actions [115,124,125]. In general, however, the magnitude of such anti-inflammatory effects is less than those of inhaled GCs [126].

In clinical trials comparing these two classes of controller agents, inhaled GCs are usually superior [126]. However, the nature and magnitude of the differences between these treatments are variable. In some of these trials, inhaled GCs were superior in all endpoints examined, while in others, they were superior only in some outcomes [111,127,128]. Specifically, inhaled GCs, when compared with leukotriene modifiers, tend to improve lung function measures more than inflammation or patient-centered outcomes (symptoms, quality of life, or resource utilization) [112,126,129]. A few studies have found no significant differences between the two drug classes [130-133]. Comparative studies typically utilize mean group data to compare the two active agents, a practice which obscures interpatient variation and limits extrapolation to clinical practice. (See 'Comparisons among antileukotriene agents' above.)

"Real-world" benefits — While efficacy comparisons assessing physiologic outcomes generally favor inhaled GCs over antileukotriene agents, "real-world" effectiveness also depends upon other factors, especially compliance [133].

●It is well known that patient adherence to inhaled GC is suboptimal, and most studies have demonstrated superior patient adherence to once-daily montelukast than to inhaled GC in both children and adults [131,132,134,135]. The superior adherence to montelukast may explain its comparable beneficial effects on asthma control to those of inhaled GC in some "real-world" studies [135,136]. Despite the importance of this issue in clinical practice, it is largely circumvented in clinical trials by having study coordinators provide frequent reminders to patients and by excluding those patients whose adherence (as documented by electronic monitors built into the inhaler devices) is deemed insufficient. (See "Enhancing patient adherence to asthma therapy".)

●It is also apparent that primary care clinicians tend to under-prescribe inhaled GCs.

Thus, no matter how efficacious inhaled GC may be, their utility in real-world settings is limited by suboptimal dosing, inhaler technique, and adherence. While not the preferred choice based upon efficacy in study populations, antileukotrienes may be considered a reasonable first-line controller agent for patients who either will not take or cannot tolerate inhaled GC. Validation of this approach is supported by a so-called "pragmatic" trial conducted in 306 patients managed in primary care practices, in which montelukast was demonstrated to be comparable to inhaled GC as a first-line controller therapy [133].

Use as add-on controller therapy — Antileukotriene agents may provide a modest degree of additive benefit for some patients with moderate or severe persistent asthma whose disease is inadequately controlled with inhaled GCs [107,108,137], although addition of a long-acting beta agonist to inhaled GC generally results in superior asthma control and fewer exacerbations compared with addition of an antileukotriene. (See "Treatment of severe asthma in adolescents and adults", section on 'Antileukotriene agents' and "Asthma in children younger than 12 years: Management of persistent asthma with controller therapies", section on 'Inhaled glucocorticoid plus LTRA'.)

Additive benefit to inhaled GC — Antileukotriene agents may provide modest additive benefit when used as adjunctive therapy with inhaled GC, although differences among patients in response to combined therapy have not been studied [138,139].

Several studies support the conclusion that glucocorticoids have minimal effects on leukotriene biosynthesis or responses, and persistent overproduction of leukotrienes has been repeatedly documented in patients on substantial doses of inhaled GC [140-145]. A combination of the two types of medications, therefore, may afford complementary anti-inflammatory and bronchoprotective actions.

Two clinical questions have been addressed regarding the additivity between inhaled GCs and antileukotriene agents:

●Does addition of an antileukotriene improve asthma control in a patient whose control is inadequate on inhaled GC alone? This question has been addressed in both adults and children with variable results [120,138,146-151]. A systematic review and meta-analysis (four studies, 815 participants) found that the number of exacerbations was reduced (RR 0.50, 95% CI 0.29-0.86) with addition of an antileukotriene agent [152].

When comparing the addition of an antileukotriene agent to inhaled GC with increasing the dose of the inhaled GC, a systematic review and meta-analysis (eight studies, 2008 participants) found no difference in lung function tests or in the number of participants with exacerbations requiring oral glucocorticoids [152].

Some studies note that subjective measures or indices of inflammation are more sensitive to the beneficial effects of add-on leukotriene modifier than are lung function measurements [129,142]. A possible explanation is that the intrinsic bronchodilatory actions of antileukotriene drugs may be difficult to discern when patients' lung function has already improved from inhaled GC.

●Does addition of an antileukotriene, in a patient with controlled asthma, allow control to be maintained despite reducing the dose of inhaled GC (ie, a steroid-sparing effect)? A systematic review and meta-analysis (seven studies, 1150 adults and adolescents) found that adding an antileukotriene agent did not significantly improve the likelihood of success in tapering the inhaled GC dose [152].

Compared with long-acting beta-agonists — For patients who remain symptomatic despite inhaled GC therapy, treatment guidelines recommend the addition of a second agent from a different therapeutic class [107,108,153]. Several randomized trials have compared montelukast with LABAs in this setting. Predictably, reduction in blood markers of inflammation (eg, eosinophilia) favored montelukast over salmeterol, while improvements in lung function favored salmeterol [117,154-157]. A systematic review and meta-analysis (eight studies, 5923 adults and 334 children) found that addition of a LABA tends to provide superior asthma control and a reduced risk of exacerbation compared with the addition of a leukotriene receptor antagonist (risk ratio 0.87, 95% CI 0.76-0.99) [158].

However, not all studies comparing the effects of adding antileukotriene agents or LABAs to inhaled GC therapy have concluded that LABAs are better. Comparable benefit was seen in some, particularly in the outcome of reduced rates of exacerbation, suggesting that both are useful in this regard [117,133,154]. As noted above, in a "pragmatic" trial in primary care practices, montelukast was found to be comparable to a LABA as add-on therapy for patients inadequately controlled on inhaled GC alone [133].

Special populations — Individuals differ in their responses to antileukotriene agents [111,112]. Certain clinical characteristics are associated with greater responsiveness, although substantial variability characterizes responses within each of these patient populations.

●Aspirin-exacerbated respiratory disease (AERD) – Subjects with AERD are consistently overrepresented in population studies of severe asthma [159,160] and also tend to have severe, recalcitrant nasal polyposis [161]. In addition to displaying selective hyperresponsiveness to exogenous inhaled LTE₄ [73,74], these individuals also exhibit markedly dysregulated 5-LO pathway activity and CysLT overproduction (as evidenced by urinary LTE₄ levels that exceed those observed in aspirin-tolerant asthmatic controls) [32,162]. Compared with placebo, both zileuton and CysLT₁R antagonists improve baseline symptomatic control and olfaction in patients with AERD [109,163].

Oral challenges to aspirin are commonly used in clinical practice to elicit reactions that establish the diagnosis and to induce desensitization [77]. These reactions produce changes in lung function, as well as sinonasal, ocular, and (occasionally) gastrointestinal and cutaneous symptoms, accompanied by dramatic increases in urinary LTE₄ [32,162]. Prophylaxis with zileuton markedly attenuates all features of these reactions [164-166]. Similarly, the administration of CysLT₁R antagonists prior to the challenge can attenuate the decline in FEV₁, increasing the safety margin of the challenge [167]. These findings collectively validate the pathogenetic role of the CysLTs in AERD. Prophylaxis with either zileuton or with CysLT₁R antagonists is commonly used to improve the safety of therapeutic desensitization procedures in AERD. (See "Aspirin-exacerbated respiratory disease" and "Aspirin-exacerbated respiratory disease: NSAID challenge and desensitization", section on 'Premedication for oral protocols'.)

●Exercise-induced symptoms – Antileukotriene agents are generally highly protective against exercise-induced bronchospasm (EIB). Montelukast, zafirlukast, and zileuton all ameliorate EIB to a similar degree [168]. Montelukast can be particularly useful in young children with EIB, who exert themselves unpredictably throughout the day.

Montelukast and salmeterol have been compared with respect to their capacity to inhibit EIB [169,170]. On the third day of treatment, the agents had similar effects, but after four and eight weeks the effects of montelukast were sustained while those of salmeterol had waned. The combination of montelukast plus salmeterol was shown in one study to provide greater protection against EIB than either agent alone [171].

Protection against EIB with montelukast is detectable as early as two hours after a single oral dose and persists for up to 24 hours [172]. For patients who require specific pretreatment prior to exercise, regardless of whether they are on chronic controller therapy, a leukotriene modifier dosed two hours prior to exercise is a reasonable and effective alternative to pretreatment with short-acting beta-agonists and is US Food and Drug Administration approved for this indication. (See "Exercise-induced bronchoconstriction".)

●Viral upper respiratory infection (URI)-induced symptoms – There is abundant evidence that leukotrienes are generated during various viral respiratory infections, and montelukast reduces the frequency of infection-associated asthma exacerbations in children and ameliorates the symptoms of respiratory syncytial virus-associated exacerbations [122,173]. However, a meta-analysis found that its utility in children without chronic asthma but with episodic viral wheeze is variable and not significant overall [174].

Several other patient characteristics have been proposed, although their ability to predict a response to antileukotriene agents has not been convincingly demonstrated.

●Children – On average, children may be more likely than adults to benefit from leukotriene modifier therapy [110]. In a study of mild-to-moderate persistent asthma, younger children were more likely to benefit compared with older children [175]. Whether this might reflect a greater prevalence of symptoms triggered by viral URIs in young children or the tendency of children to be physically active throughout the day remains to be determined.

●Females – Consistent with the greater magnitude of leukotriene synthesis in females than males [49,176], sexual dimorphism in treatment responses was evaluated in a large registry-based retrospective study of adult and pediatric patients administered montelukast [177]. In adults, males were more likely than females to discontinue montelukast, switch to another therapy, or require rescue therapy with oral GCs or short-acting beta-agonists. By contrast, no sex-related differences were observed in pediatric patients. Conclusions about possible variation in antileukotriene responses by sex must await prospective studies.

●Symptoms triggered by air pollutants – Leukotrienes are produced upon respiratory tract exposure to pollutants, such as sulfur dioxide and particulates. CysLT₁ receptor antagonists reduce airway inflammation and bronchoconstrictor responses to both types of pollutants in laboratory challenge settings [101,178,179]. However, the specific utility of antileukotrienes in ameliorating or preventing pollutant-induced asthma has not been explored.

●Asthma severity – Guideline-based management does not advocate using antileukotriene agents as monotherapy in moderate or severe persistent asthma, although it can be given as an adjunct therapy. (See "Treatment of severe asthma in adolescents and adults", section on 'Antileukotriene agents' and "Ongoing monitoring and titration of asthma therapies in adolescents and adults", section on 'Alternative agents for escalation from Steps 2 or 3'.)

In our experience, severity alone is not particularly helpful in predicting the likelihood of a response to a leukotriene modifier versus an inhaled GC. This was supported by a retrospective study that found that although zileuton was more often prescribed to patients with severe asthma, asthma severity did not predict responsiveness [106].

●Allergic rhinitis – As allergic rhinitis coexists with asthma in many patients and randomized trials indicate that antileukotrienes have modest efficacy in allergic rhinitis, antileukotrienes may ameliorate both upper and lower airway disease. In a retrospective, observational study of several thousand patients with both conditions, approximately 80 percent of patients reported meaningful improvement in both conditions with montelukast treatment [180]. In a large open-label study of montelukast addition to either inhaled GC or inhaled GC plus LABA in adult asthmatics with inadequately controlled mild to moderate asthma, subjects with coexisting allergic rhinitis had numerically superior asthma control scores than did those without allergic rhinitis [181].

However, it has not been our experience that patients with both conditions or patients with atopic triggers for their asthma are any more likely to benefit from leukotriene modifiers than are patients without concomitant allergic rhinitis or atopy. In addition, a boxed warning was placed on montelukast advising against its use for allergic rhinitis or mild asthma. (See 'Adverse effects' below.)

●Obesity – Obesity is associated with an increased relative risk and severity of asthma and reduced efficacy of asthma medications. While some studies suggest a benefit of leukotriene modifiers in maintaining asthma control, most studies have found that inhaled GC is superior to montelukast in all body mass index categories. For patients with obesity and suboptimally controlled asthma on an inhaled GC, addition of a long-acting bronchodilator (LABA) appears more effective than addition of an antileukotriene agent. (See "Obesity and asthma", section on 'Pharmacologic therapy'.)

●Cigarette smoking – Cigarette smoking stimulates CysLT production; urinary LTE₄ is increased in a dose-dependent fashion in smokers [182]. However, antileukotriene agents do not appear to have particular efficacy in patients with asthma who smoke. In a trial of over 1000 currently smoking patients with asthma randomized to treatment with either inhaled fluticasone, montelukast, or placebo for six months, both active treatments were superior to placebo (and no different from each other) in improving asthma control [183].

Safety in pregnancy and breastfeeding

●Data regarding use of the cysteinyl leukotriene (CysLT₁) receptor antagonists (montelukast, zafirlukast, and pranlukast) during pregnancy are reassuring, but limited. These drugs are passed to the breast milk but in very small amounts that are generally considered safe for infants [184].

●In contrast, zileuton is generally avoided during pregnancy because adverse effects were noted in animal reproduction studies. Passage to breast milk has not been studied.

Specific studies of use in antileukotriene medications in pregnancy are reviewed elsewhere. (See "Management of asthma during pregnancy", section on 'Leukotriene modifiers'.)

ADVERSE EFFECTS — The leukotriene receptor antagonists, montelukast, zafirlukast, and pranlukast, are generally well-tolerated. Zileuton is associated with more frequent adverse effects, including liver inflammation and drug interactions.

Leukotriene receptor antagonists — Adverse effects have been described in small numbers (≤2 percent) of individuals receiving zafirlukast or montelukast, including anaphylaxis, angioedema, dizziness, dyspepsia, muscle weakness, and elevated transaminases.

●Boxed warning about montelukast and neuropsychiatric adverse effects – The US Food and Drug Administration (FDA) has placed a boxed warning on montelukast, advising that all patients should be warned about potential behavior and mood-related changes; providers should consider the potential risks and benefits of montelukast, particularly when the patient has mild asthma or allergic rhinitis or a history of psychiatric illness; and patients should be monitored for neuropsychiatric symptoms when taking montelukast [185]. This warning is based on postmarketing surveillance of neuropsychiatric events (eg, agitation, depression, insomnia, suicidal thoughts and actions) among patients taking montelukast. It should be noted that depression and suicidality are more common in patients with asthma than in the general population [186].

Published data regarding potential neuropsychiatric effects are mixed [187-193]. As examples:

•A scientific review of clinical trial data requested by the FDA did not find an increase in suicidal ideation in subjects taking montelukast compared with those on placebo [187].

•A systematic review concluded that the evidence for and against an effect was of low quality and relied on pharmacovigilance more than randomized trials or high-quality observational data [191].

•Subsequent observational studies suggest a small increased risk of neuropsychiatric events associated with montelukast prescription. An 11-year study of children aged 5 to 18 years in Ontario, Canada, found that the 898 children experiencing new-onset neuropsychiatric events (eg, anxiety, sleep disturbance) were almost twice as likely as the 3497 controls to have been prescribed montelukast in the year before the event (OR 1.91) [190]. A health records database study of 72,490 adolescents and adults with asthma found that those exposed to montelukast had a small increase in insomnia (19 versus 16 cases per 1000, OR 1.13) anxiety disorders (64 versus 57 cases per 1000, OR 1.21), and new antidepressant use (59 versus 52 per 1000, OR 1.16) in the first year after montelukast treatment compared with matched montelukast-unexposed patients [192]. A Danish national registry study of adults showed similar overall results, but the effect was only evident in younger patients (new neuropsychiatric medication use after montelukast exposure compared with unexposed patients: OR 1.38 in adults aged 18 to 29 years, OR 1.17 in adults aged 30 to 44 years, and OR 0.95-1.0 in adults aged >45 years) [194].

Zileuton — Side effects associated with zileuton include headache (23 to 25 percent), dyspepsia (8 percent), myalgias (7 percent), leukopenia (1 to 3 percent), elevated transaminases (2 to 5 percent), sleep disorders (<1 percent), and behavior changes (<1 percent) [195]. Use in patients with a history of liver disease or substantial alcohol consumption should be avoided. The manufacturer recommends monitoring serum alanine aminotransferase (ALT) monthly for the first three months, every two to three months for the rest of the first year, and periodically thereafter. Zileuton should be discontinued if ALT is ≥3 times normal, which occurred in 2 percent of patients treated with zileuton for at least one year, compared with 0.2 percent of placebo recipients [196]. These elevations were usually transient, asymptomatic, and rapidly reversible, and most often appeared within four to eight weeks of starting therapy.

●Drug interactions are possible; zileuton inhibits cytochrome CYP1A2 and has a mild interaction with CYP1A2, 2C9, 3A4. For this reason, theophylline dosing may have to be adjusted downwards.

●An eosinophilic granulomatosis with polyangiitis (EGPA, Churg-Strauss)-like syndrome develops as a rare complication in steroid-dependent asthmatics treated with antileukotriene agents, typically the leukotriene receptor antagonists [197]. In most instances, this occurs following tapering of the oral glucocorticoid dose, suggesting that underlying EGPA may be unmasked by glucocorticoid withdrawal, rather than being caused by the drugs themselves. However, at least one analysis argues against dismissing a possible causal relationship [198]. The pathogenesis of EGPA is discussed separately. (See "Epidemiology, pathogenesis, and pathology of eosinophilic granulomatosis with polyangiitis (Churg-Strauss)", section on 'Leukotriene modifying agents'.)

FUTURE DIRECTIONS — Areas of ongoing investigation include the use of biomarkers and genotypic markers to predict which patients are most likely to respond to antileukotriene agents and a possible role for antileukotriene agents in acute asthma. In addition, the potential therapeutic value of antagonists to leukotriene receptors other than cysteinyl leukotriene-1 (CysLT₁), of inhibitors of biosynthetic enzymes other than 5-lipoxygenase, and of combinations of antileukotriene drugs with different mechanisms of action all require further research.

●Biomarkers – Biomarkers that reflect leukotriene production are an active area of investigation, although these markers are not currently used in clinical practice [59,109,199,200]. One obvious candidate biomarker is the degree of leukotriene overproduction itself. One study found that mean levels of ex vivo CysLT production by stimulated whole blood leukocytes were higher in patients deemed to be responders to a CysLT1 antagonist than in nonresponders, but the sensitivity and specificity of this parameter were inadequate to guide clinical decision-making [199].

Studies examining urinary leukotriene E4 (LTE₄) levels as a predictor of response to antileukotriene drugs have yielded variable results. For example, some studies in patients with moderate persistent asthma or aspirin intolerant asthma have shown no relationship [109,200]. An inverse relationship was reported between urinary LTE₄ and response to a CysLT₁R antagonist in another study [201]. In the largest series reported (albeit still only comprising 48 patients), a correlation among mild-moderate asthmatics between urinary LTE₄ and responsiveness was observed, but again the sensitivity and specificity were inadequate to be clinically useful [202]. (See "Aspirin-exacerbated respiratory disease".)

●Genotypic markers – Polymorphisms in genes encoding 5-lipoxygenase (5-LO) and LTC₄S may predict responsiveness to antileukotriene drugs. However, it is not clear whether the possible predictive value of these variant genotypes is actually explained by altered leukotriene synthetic capacity because of uncertainty regarding the relationship between the degree of leukotriene overproduction and response to leukotriene modifiers, as well as the precise impact of these variant genotypes on leukotriene synthetic capacity.

Variants of the 5-LO promoter are associated with diminished responses to a 5-LO inhibitor [203]. However, these variants occur in fewer than 10 percent of asthmatics and therefore cannot explain much of the observed nonresponsiveness to leukotriene modifiers.

A polymorphism in the LTC₄ synthase gene is more common, occurring in approximately 20 to 40 percent of asthmatics. A positive relationship between this variant allele and clinical response to antileukotriene drugs has been reported in two studies, although not in a third [54,200,204].

Minimal data exist on the possible role of variant leukotriene receptors as determinants of responsiveness to these drugs; one study found that variants in the CysLT₂R gene were associated with greater responsiveness to a CysLT₁R antagonist [205]. This puzzling observation can perhaps be explained by in vitro studies demonstrating that CysLT₂R can interact with and attenuate responses to CysLT₁R ligation [83]. (See 'Mast cell activation' above.)

●Pharmacogenomics – Variation in drug absorption may contribute to the variability of response to leukotriene modifiers. Genetic variation in the transporter gene SLCO2B1 (Solute Carrier Organic Anion Transporter family, member 2B1, MIM 604988) has been associated with variation in montelukast plasma levels and with variability of response to montelukast among patients with asthma [206]. The first study to utilize genome-wide association, rather than candidate gene analysis, to interrogate the pharmacogenomics of responses to antileukotriene drugs identified previously unreported loci in genes involved in processing transfer RNA and in glycosylation that were associated with differential responsiveness to leukotriene modifiers [207]. (See "Overview of pharmacogenomics", section on 'Drug transport'.)

●Higher dose montelukast – Anecdotally, some patients describe a benefit from montelukast only when it is used at doses higher than its standard recommended dose of 10 mg daily. Early studies with this drug used at doses up to 600 mg daily showed increases in lung function that were generally greater than those obtained with the 10 mg dose [208]. Since the cytokine-rich milieu of an inflamed asthmatic lung would be expected to increase expression of the CysLT₁R [75], it is conceivable that higher concentrations of this competitive antagonist might be required for effective receptor blockade, at least in some patients. We agree with the suggestion that montelukast dosing be reconsidered, which would require dedicated clinical trials in various subpopulations [39].

●Novel receptor antagonists – CysLT₁R is the only CysLT receptor that has successfully been targeted by US Food and Drug Administration-approved drugs in asthma. A dual CysLT₁R/CysLT₂R antagonist showed nearly identical capacity to block allergen-induced early and late phase changes in lung function in allergen-challenged atopic mildly asthmatic subjects [209]. This likely reflects the dominance of CysLT₁R in mediating CysLT-induced airway smooth muscle responses. Given that both CysLT₂R [48] and CysLT₃R [47] contribute prominently to features of eosinophilic inflammation and mucus secretion in preclinical models, it remains plausible that one or both of these receptors will prove to be a fruitful target, particularly in aspirin-exacerbated respiratory disease where both CysLT synthesis and CysLT-associated features of immunopathology are prominent [210].

●Acute treatment of exacerbations – While not available for clinical use, preliminary studies suggested that addition of an intravenous or high-dose antileukotriene agent to inhaled beta-agonists and systemic glucocorticoids was beneficial in the treatment of asthma exacerbations in the emergency department. The use of antileukotriene agents in this setting is presented separately. (See "Acute exacerbations of asthma in adults: Home and office management".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Asthma in adolescents and adults".)

SUMMARY AND RECOMMENDATIONS

●Leukotrienes are proinflammatory mediators derived from arachidonic acid that contribute to the pathogenesis of asthma. The cysteinyl leukotrienes (CysLTs), LTC₄, LTD₄, and LTE₄, mediate pro-asthmatic effects such as sustained bronchoconstriction, mucus secretion, and airway edema (figure 1). (See 'Introduction' above and 'Leukotriene biology' above.)

●Antileukotriene agents (also known as leukotriene modifying agents) inhibit the action of leukotrienes by blocking the CysLT₁ receptor (eg, montelukast, zafirlukast, pranlukast) or by interrupting production by 5-lipoxygenase (eg, zileuton) (figure 1). The usual doses of these agents are provided in the table (table 1). (See 'Available antileukotriene agents' above.)

●Some patients respond dramatically to antileukotriene agents, while others show no response at all, and clinical characteristics are poorly predictive of responsiveness in individual patients. To assess the utility of an antileukotriene agent in an individual patient, we typically perform a one to two-month trial, regardless of which agent is selected and whether the leukotriene modifier is being utilized as first-line or add-on therapy. (See 'Therapeutic trial' above.)

●Antileukotriene agents are effective as monotherapy in the treatment of some patients with mild to moderate persistent asthma, although the mean efficacy of antileukotriene agents is lower than that of inhaled glucocorticoids (inhaled GCs). Thus, inhaled GCs are preferred first-line controller agents, and antileukotriene agents are considered alternatives, particularly for patients who have difficulty with adherence or inhaler technique. (See 'Relative efficacy' above and "Initiating asthma therapy and monitoring in adolescents and adults" and "Ongoing monitoring and titration of asthma therapies in adolescents and adults".)

●In patients who respond, antileukotriene agents improve lung function and symptom-related quality of life and decrease the frequency of exacerbations and the need for symptom-relieving beta-agonist medications. Thus, they can be useful in patients who cannot tolerate, are unwilling to take, or who have not responded to inhaled GCs. In addition, compliance with oral therapies is generally higher than with inhalers. (See 'Use as initial controller therapy' above and '"Real-world" benefits' above.)

●For patients with asthma that is inadequately controlled with inhaled GCs, long-acting beta agonists (LABAs) will generally provide superior improvement in lung function compared with antileukotriene agents. However, the advantages of LABAs over antileukotriene agents in other endpoints are more modest. (See 'Compared with long-acting beta-agonists' above and 'Additive benefit to inhaled GC' above.)

●Antileukotriene agents are highly protective against exercise-induced bronchoconstriction (EIB), but short-acting beta agonists are usually preferred as they can both prevent and relieve bronchoconstriction. The role of antileukotriene agents in the management of aspirin-exacerbated respiratory disease (AERD) is discussed separately. (See 'Special populations' above and "Exercise-induced bronchoconstriction" and "Aspirin-exacerbated respiratory disease", section on 'Leukotriene-modifying agents'.)

●The leukotriene receptor antagonists, montelukast, zafirlukast, and pranlukast (not available in the United States or Canada), are generally well-tolerated. Prescribers and patients should be aware of potential behavioral and neuropsychiatric disturbances with montelukast. Zileuton is associated with more frequent adverse effects, including liver inflammation and drug interactions. (See 'Adverse effects' above.)

ACKNOWLEDGMENT — The editorial staff at UpToDate acknowledge Marc Peters-Golden, MD, who contributed to earlier versions of this topic review.