The Effect of Inhibition of 5-Lipoxygenase by Zileuton in Mild-to-Moderate Asthma

Methods

Patient Selection

Patients with mild-to-moderate asthma were recruited at 14 centers, which included university hospitals and private allergy and pulmonary practices. Patients with symptoms that corresponded with the American Thoracic Society definition of asthma [28] were screened. All patients had to have a forced expiratory volume in 1 second (FEV₁) of 40% to 75% of predicted value and a 15% or greater increase in FEV₁ 30 minutes after inhalation of two puffs of albuterol. Additionally, patients were required to be 18 to 65 years old; women of childbearing potential were excluded. Before enrollment in the study, none of the patients had used oral or inhaled steroids or cromolyn sodium for 4 weeks. Beta-blockers, calcium-channel blockers, and nonsteroidal anti-inflammatory drugs could not have been used for at least 1 week before entry into the study. All patients were required to be able to achieve adequate symptomatic asthma control without using theophylline, oral -agonists, or antihistamines; none of these medications was permitted throughout the entire study period.

Study Design and Intervention

A randomized parallel design was used in this double-blind, placebo-controlled study. Patients were chosen randomly to receive either 600 mg of zileuton (four times a day), 800 mg of zileuton (twice a day), or placebo. A 1-week, single-blind, placebo lead-in qualification period (dummy lead-in period) was followed by random allocation to one of the three treatment groups for a 4-week, double-blind phase.

During the single-blind, dummy lead-in and the double-blind study periods, all patients took capsules four times a day. Self-determined peak expiratory flow rates were recorded in the morning (before medication) and evening (2 hours after the third set of capsules) in a study diary. Albuterol inhaler use and asthma symptoms were recorded in the diary as well. Daytime asthma symptoms were self-rated on a scale of 1 to 5 (1 = no symptoms, 5 = severe symptoms; maximum weekly score of 35).

After the 1-week dummy lead-in period, patients returned to their study center. Inhaled albuterol was withheld for at least 8 hours before the study visit. Spirometry was done on patients who had no clinically significant laboratory abnormalities, who had successfully completed their diary card, who had moderately symptomatic asthma (a total score of 12 but 28 in the previous 7 days), and who had used their albuterol inhaler at least 7 times during the dummy lead-in week. If the FEV₁ was 40% to 75% of the predicted value, the patient was assigned randomly to a group according to a predetermined code. All patients took visually identical capsules four times per day that contained either 600 mg of zileuton four times daily, 800 mg of zileuton twice daily (active drug first and last dose daily), or placebo, which were supplied by Abbott Laboratories in a blind manner. The first dose of study medication was administered at the study center, and spirometry was done 30, 60, and 120 minutes later. In the 800-mg group, each day's drug card contained both placebo and active drug, and as a result, on the first day, an undetermined number of patients received placebo instead of 800 mg of zileuton as their first dose of drug. Therefore, the 800-mg group was not included in the analysis of the acute response to the first dose of drug.

During the 4-week double-blind period, patients returned to the study center at the same time of day on a weekly basis to have spirometry done and to review diary cards and medication use. During the second and third weeks of the double-blind randomization period, spirometry was also repeated 30 minutes after inhalation of two puffs of albuterol.

Urine Collection and Analysis

Urine was collected for 4 hours beginning at 8:00 a.m. before the dummy lead-in period and on day 28 of the study. Urinary LTE₄ levels were determined by reverse-phase high-performance liquid chromatography and enzyme immunoassay using minor modifications of established procedures [29]. The recovery of the internal LTE₄ standard was 74% 6%. The LTE₄ content of the urine was expressed as picograms of immunoreactive LTE₄ per milligram of creatinine.

Adverse Events

Routine complete blood counts, serum chemistries, urinalyses, and electrocardiograms were obtained throughout the study. Adverse symptoms were elicited daily through a diary question and were reviewed at the weekly visit to the study site.

Statistical Analysis

All values were expressed as means with associated 95% CIs; all outcome indicators were normally distributed. Paired t-tests were used to assess the statistical significance of any within-group changes from the baseline dummy lead-in phase. The statistical significance of differences among the placebo and active treatment groups during the 4 weeks of double-blind treatment was evaluated using a two-way analysis of variance model with effects for center, treatment, and center-treatment interaction. When statistical differences were noted among groups in the dummy lead-in, the groups were compared using an analysis of covariance adjusting for baseline differences. Available data were analyzed up to the point of withdrawal for patients who did not complete the study protocol. The Fisher test for the protected least significant difference was used to make pair-wise comparisons.

Results

Patients

A total of 188 patients entered the single-blind dummy lead-in period; 143 fulfilled the enrollment criteria and were randomly assigned to receive study drug or placebo (46 patients received 2.4 g/d, 49 patients received 1.6 g/d, and 48 patients received placebo). Two patients withdrew during the first week of the double-blind study1 for personal reasons and the other because of worsening asthma (both received 1.6 g/d of zileuton). Two patients were not included in the final analysis, because they were enrolled in a center that did not have representation in all three treatment groups (1 received 1.6 g/d of zileuton and 1 received placebo). The characteristics of the 139 evaluated patients are given in Table 1. Of the 139 patients who were still in the trial after 1 week, 12 evaluated patients left the study before completing the trial protocol (all their data were included up to the point of termination): 4 patients because of worsening asthma (1 received 2.4 g/d, 2 received 1.6 g/d, and 1 received placebo); 2 patients because of upper respiratory infections (1 received 2.4 g/d and 1 received placebo); 1 patient because of sinusitis (placebo); 1 patient because of urticaria (1.6 g/d); 3 patients because of personal reasons (1 in each group); and 1 patient because of headaches that had begun before randomization (2.4 g/d).

Acute Effects on Airway Obstruction

A single 600-mg dose of zileuton produced rapid bronchodilation (Figure 1). Compared with the mean FEV₁ measured just before study drug ingestion (0 minutes), the mean FEV₁ improved 30 minutes after a single 600-mg dose of zileuton and remained increased for the entire 2-hour observation period (P < 0.005 for all observation points). The maximum increase (14.6%) in the mean FEV₁ was 0.35 L (CI, 0.25 to 0.45 L) (P < 0.001), which occurred at 60 minutes. No improvement of the FEV₁ occurred in the placebo group (0.09 L [CI, 0.01 to 0.19 L]; P = 0.075). The improvement in the mean FEV₁ after zileuton was greater than that after placebo at 60 and 120 minutes (P < 0.001 and P = 0.01, respectively).

Effects of 4 Weeks of Zileuton Administration on Airway Obstruction

All three groups of patients had an initial improvement in mean FEV₁ after enrollment (Figure 2). By the fourth week of the study, only the zileuton groups were still statistically improved compared with the dummy lead-in period. The 0.32-L (CI, 0.16 to 0.48 L) improvement in mean FEV₁ at 4 weeks in the group receiving 2.4 g/d of zileuton was greater than that in the placebo group (P = 0.02). This represented a 13.4% improvement in FEV₁ compared with the dummy lead-in period. The group receiving 1.6 g/d of zileuton had a mean 0.24-L (CI, 0.08 to 0.40 L) (10.9%) improvement (P = 0.09 compared with placebo). A similar pattern of improvement was seen in the forced vital capacity (FVC). At the end of 4 weeks, the FVC had increased by 0.35 L (CI, 0.17 to 0.53 L) (P < 0.001) and 0.23 L (CI, 0.07 to 0.39 L) (P = 0.006) compared with the dummy lead-in period in patients taking 2.4 g/d and 1.6 g/d of zileuton, respectively. No statistical improvement was noted in the FVC in the placebo group (0.07 L; CI, 0.09 to 0.23 L) (P > 0.2). At 28 days, improvement in FVC occurred with 2.4 g/d of zileuton compared with that in the placebo group (P = 0.02).

Figure 3 shows the weekly, average morning and evening peak expiratory flow rates throughout the study. Compared with placebo, 2.4 g/d of zileuton produced a statistical improvement in the weekly, mean morning peak expiratory flow starting with the second week of drug administration through the end of the study (Figure 3, left). The maximum weekly change in the mean, morning peak expiratory flow was 39.5 L/min (CI, 24.8 to 54.2 L/min) (P < 0.001 compared with the dummy lead-in period), which represented a 10% increase in peak expiratory flow. Although the evening peak expiratory flow using 2.4 g/d of zileuton (Figure 3, right) was statistically improved compared with the dummy lead-in period throughout the entire 4 weeks, the comparison with the placebo group was not different (P = 0.064 at 3 weeks). The maximum improvement in evening peak expiratory flow was 30.5 L/min (CI, 15.8 to 45.2 L/min) in the group receiving 2.4 g/d (P = 0.004 compared with the dummy lead-in value). During the last 2 weeks of the study, there was less difference between the evening and morning peak expiratory flow compared with placebo (P = 0.015) in the group taking 2.4 g/d of zileuton.

Asthma Symptoms

Baseline symptom scores ranged from 1.98 to 2.28 among treatment groups (Table 1). After 4 weeks of treatment, the asthma symptoms reported by the patients decreased in all three treatment groups (P < 0.01), but the improvement was greater in the group receiving 2.4 g/d of zileuton (P = 0.02 compared with placebo). Overall symptom scores decreased by 37% (CI, 26.1% to 47.9%), 29% (CI, 20.3% to 37.7%), and 17% (CI, 5.7% to 28.3%) for the groups receiving zileuton, 2.4 g/d; zileuton, 1.6 g/d; and placebo, respectively.

-Agonist Use

The average daily frequency of -agonist use decreased only in the zileuton groups (Figure 4). During the last week of the trial, the frequency of -agonist use per day decreased compared with the dummy lead-in value by 0.76 (CI, 0.35 to 1.17) occasions in patients taking zileuton, 2.4 g/d (P < 0.001); 0.57 (CI, 0.19 to 0.95) occasions in patients taking zileuton, 1.6 g/d (P = 0.005); and 0.23 (CI, 0.15 to 0.61) occasions in patients taking placebo (P = 0.24). These represent a 24%, 17%, and 7% change in -agonist use, respectively (P = 0.03, 2.4 g/d of zileuton compared with placebo).

Bronchodilator Response

The improvement in the FEV₁ after administration of the -agonist albuterol was assessed at the screening visit before the dummy lead-in period and after 2 and 3 weeks of therapy. Despite the fact that the pre-bronchodilator FEV₁ had improved in the zileuton groups (see Figure 2), no reduction occurred in the acute bronchodilator response to albuterol in patients receiving zileuton compared with those given placebo. Before the dummy lead-in period, in response to inhaled albuterol, the FEV₁ increased 0.76 L (CI, 0.64 to 0.88 L), 0.71 L (CI, 0.59 to 0.83 L), and 0.78 L (CI, 0.66 to 0.90 L) in patients receiving 2.4 g/d of zileuton, 1.6 g/d of zileuton, and placebo, respectively (P > 0.2 between groups). After 3 weeks of therapy, in response to albuterol, the FEV₁ increased 0.62 L (CI, 0.50 to 0.74 L), 0.54 L (CI, 0.42 to 0.66 L), and 0.59 L (CI, 0.47 to 0.71 L) in the three groups, respectively (P > 0.2 between groups).

Urinary Leukotriene E₄

Sulphidopeptide leukotriene production, as reflected by recovery of LTE₄ in the urine, was decreased after 4 weeks of zileuton administration (Figure 5). Zileuton, 1.6 g/d, decreased the urinary LTE₄ by 26.5 pg/mg creatinine (CI, 6.6 to 46.5 pg/mg creatinine) (P = 0.009), which was equivalent to a 26% decrease from the dummy lead-in value. Zileuton, 2.4 g/d, decreased urinary LTE₄ recovery by 39.2 pg/mg creatinine (CI, 18.1 to 60.4 pg/mg creatinine) (P < 0.001), a 39% decrease. No change was noted in the urinary LTE₄ level in the group receiving placebo (P > 0.2). The reduction in the groups using 1.6 g/d and 2.4 g/d of zileuton was significant compared with the placebo group (P = 0.049 and P = 0.007, respectively).

Adverse Events

No differences were noted in the occurrence of adverse events between the patients receiving zileuton and those given placebo. Adverse effects were reported by 39.7% of patients taking 2.4 g/d of zileuton, by 44.9% taking 1.6 g/d of zileuton, and by 44.9% taking placebo. The most common adverse event reported was headache (10% in patients receiving 2.4 g/d of zileuton, 16% in those receiving 1.6 g/d, and 14% in those receiving placebo [P > 0.2]). Three patients receiving 2.4 g/d of zileuton and three receiving 1.6 g/d reported dyspepsia, although no patients receiving placebo reported dyspepsia (P = 0.18 using a two-tailed Fisher exact test).

No clinically significant changes occurred in the hematology determinations, serum chemical analyses, electrocardiograms, or urinalysis test results. Hives and increased liver function test results occurred in one patient after 24 days of zileuton, 800 mg twice a day. Both the hives and the abnormal liver chemistry test results resolved after discontinuation of zileuton therapy.

Discussion

Our data show that treatment with zileuton, a drug that blocks the action of the enzyme 5-lipoxygenase, is associated with improvements in both objective and subjective measures of disease severity in patients with mild-to-moderate asthma. Zileuton produced immediate bronchodilation after ingestion of a single dose. Of greater importance, the regular administration of 600 mg of zileuton four times a day for 4 weeks produced progressive improvement in the FEV₁, FVC, and peak expiratory flow and a decrease in asthma symptoms and need for -agonist use. The salutary effects were evident with a dose that produced greater inhibition of leukotriene production as reflected by measurement of urinary LTE₄. Because zileuton's only known pharmacologic effect is to inhibit the action of 5-lipoxygenase [27, 30], an enzyme critical to the formation of the leukotrienes, our data show that inhibition of 5-lipoxygenase can be of therapeutic benefit in persons with asthma and provide important evidence for the leukotrienes as mediators of airway narrowing in mild-to-moderate chronic asthma.

We observed acute bronchodilation after administration of the first dose of zileuton (Figure 1). The maximum bronchodilator effect, a 14.6% increase in the FEV₁, was observed within 1 hour of ingestion of the drug. Previous work has shown that zileuton has no significant effect on FEV₁ in patients with minimal baseline airway obstruction [14, 17]. However, in patients with abnormal baseline function, antagonists active at the leukotriene receptor have been shown to produce bronchodilation [19, 31]. The bronchodilation that we observed after a single dose of a 5-lipoxygenase inhibitor suggests that ongoing synthesis of leukotrienes may be necessary to mediate part of the increased bronchomotor tone characteristic of asthma in patients with impaired baseline function.

Although zileuton appeared to have a substantial bronchodilator effect during 2 hours, the bronchodilation achieved was less than half that which could be achieved by the inhaled -agonist albuterol. There are two possible explanations for zileuton's failure to reverse airway obstruction as completely as inhalation of -adrenergic aerosols. First, the urinary leukotriene data show that zileuton does not completely inhibit 5-lipoxygenase. Because the inhibition of 5-lipoxygenase is not complete, part of the residual bronchoconstrictor effect observed may result from the action of 5-lipoxygenase products that are produced despite zileuton administration. Second, it is likely that a component of reversible airway obstruction exists in asthma that results from mechanisms other than the actions of 5-lipoxygenase products on contractile elements. However, it is important that the bronchodilation produced by 5-lipoxygenase blockade, approximately 40% of the bronchodilation observed after -agonists, compares favorably with the approximately 50% response seen with oral steroids or very-high-dose inhaled steroids [32].

In addition to the immediate bronchodilator action of zileuton, a progressive improvement was noted in airflow obstruction and asthma symptoms that occurred with continued administration during the 4 weeks of the study. For the FEV₁, the importance of this effect became more apparent with resolution of the initial placebo effect frequently noted in studies of persons with asthma. That the observed change in FEV₁ was due to zileuton's action on 5-lipoxygenase is further substantiated by the observation that greater and more consistent inhibition of 5-lipoxygenase, as measured by the decrease in urinary LTE₄ to nearly normal levels [29] in patients using 2.4 g/d of zileuton, was also associated with a greater improvement in airway function. The progressive effects occurring during 4 weeks suggest that in addition to the acute effect on airway smooth muscle, inhibition of 5-lipoxygenase may affect a process other than smooth muscle constriction. For example, zileuton, by preventing the formation of the cysteinyl leukotrienes, could prevent microvascular leak and airway edema. It is also possible that preventing the formation of LTB₄, another 5-lipoxygenase product with potent chemotactic properties, could partially prevent the cellular infiltration associated with asthma [33]. These processes may take days to weeks to reverse and could account for the additional actions of zileuton during the 4 weeks of the study beyond its acute effects. Our data do not allow us to distinguish among these potential mechanisms of the prolonged effects of zileuton. However, because the long-term improvement in airflow obstruction was not associated with any change in -agonist responsiveness compared with placebo, it appears that the long-term bronchodilation associated with 5-lipoxygenase inhibition occurs through a mechanism that differs from that of the -agonists.

The magnitude of improvement in airway function after 4 weeks of administration of 2.4 g/d of zileuton (that is, a 13.4% increase in the FEV₁ and an approximately 40-L/min increase in morning peak expiratory flow) is similar to the effects of many of the therapeutic agents used in asthma treatment, such as theophylline [34] or inhaled steroids [35-37]. Of particular importance is the magnitude of the effect of 5-lipoxygenase inhibition compared with the effects of long-term administration of moderately high doses of inhaled steroids in persons with mild-to-moderate asthma. For example, Dutoit and colleagues [35] noted a 12% improvement in the FEV₁ after treatment with the equivalent of 19 puffs per day of inhaled beclomethasone dipropionate for 4 weeks, and Haahtela and colleagues [36] and Vathenen and colleagues [37] noted an improvement in the morning peak flow rate of approximately 30 to 35 L/min after 4 weeks of treatment with 1200 and 1600 g/d of inhaled budesonide, respectively. Because the doses of inhaled steroids used in these studies may result in some degree of adrenal suppression [38] and possibly increased bone turnover [39, 40], the magnitude of the effect of 5-lipoxygenase inhibition is likely to be clinically important.

The effects of zileuton compared with -agonists are also of interest in view of recent data that have questioned increased [22] or regular use [21, 24] of -agonists for the treatment of asthma and their effects on asthma symptoms and airway reactivity. At the completion of 4 weeks of treatment, there was a statistically significant 24% decrease in -agonist use and a 37% decrease in asthma symptoms. The magnitude of the decrease in -agonist use (0.76 occasions/d) compares favorably with that seen with high-dose inhaled steroids [36]. At the same time that -agonist use decreased, an improvement occurred in baseline airway function without a change in the acute bronchodilator response to -agonists compared with placebo, thus suggesting that inhibition of 5-lipoxygenase can preserve the acute response to -agonists while decreasing the need for their regular use. Additionally, high-dose zileuton narrowed the difference between morning and evening peak expiratory flow. Because the magnitude of the variation between morning and evening peak expiratory flow has been associated with the degree of airway responsiveness in asthma [41, 42], this outcome suggests the possibility that the decrease in -agonist use may be because of, or associated with, a decrease in airway responsiveness.

Zileuton has not been associated with substantial toxicity. More than 1700 people have received the drug, and adverse events have been generally mild and self-limited, with an incidence that has not been different from that in the placebo group. Isolated cases of increased serum transaminase levels have occurred, with a rapid return to normal levels without therapy.

In conclusion, this double-blind, placebo-controlled trial shows that treatment of mild-to-moderate chronic asthma with an inhibitor of 5-lipoxygenase is associated with improved airway function and decreased symptoms and a need for less bronchodilator therapy. The degree of improvement compares favorably with that achieved by clinically useful drugs such as theophylline and inhaled steroids. In addition to an acute bronchodilator effect, progressive improvement in airway function was observed during 4 weeks of treatment by a mechanism that appears to be separate from that of -agonist-induced relaxation of bronchial smooth muscle. Dosing regimens associated with greater inhibition of leukotriene production were also associated with greater clinical and physiologic improvement. These findings provide important evidence for the role of the 5-lipoxygenase products of arachidonic acid in the pathogenesis of spontaneously occurring asthma and indicate that agents that can prevent the formation of these products are of therapeutic value in asthma. It is possible that even greater improvement may be seen with more prolonged therapy or more complete enzyme inhibition.