INTRODUCTION

TOP ABSTRACT INTRODUCTION CONCLUSION DISCUSSION REFERENCES

Among the strategies for regulating leukotriene (LT) effects, three approaches have yielded drugs with established clinical efficacy: cysteinyl leukotriene type 1 (cysLT₁) receptor antagonism, 5-lipoxygenase (LO) inhibition, and disruption of the association between arachidonic acid and 5-LO by inhibiting the 5-lipoxygenase-activating protein (FLAP). Results from blinded, randomized, and placebo-controlled clinical trials with several LT modulators have demonstrated the importance of cysteinyl leukotrienes (cysLTs) in asthma. These modulators include tomelukast, zafirlukast, pranlukast, and montelukast, selective antagonists of the cysLT₁ receptor; zileuton and ZD-2138, active-site inhibitors of 5-LO; and MK886 and BAY x1005, agents that prevent the interaction of arachidonic acid with FLAP. Other potential approaches, including inhibition of cytosolic phospholipase A₂ (cPLA₂) or LTC₄ synthase, have not yet yielded published reports of clinical efficacy.

	POTENCY OF LEUKOTRIENE RECEPTOR ANTAGONISTS AND 5-LIPOXYGENASE INHIBITORS

The potency of cysLT₁ receptor antagonists can be estimated from their inhibition of LTD₄-induced bronchoconstriction in man. A first generation cysLT₁ receptor antagonist, tomelukast (LY 171,883), produced a four- to sixfold rightward shift of the LTD₄ dose-response curve; pranlukast and pobilukast (SKF 104,353), subsequently developed, produced approximately 30-fold shifts; whereas the most potent agents, zafirlukast and montelukast, produced more than 100-fold shifts. These latter compounds exhibit clinical activity at doses in the 10 to 50 mg range.

Estimation of the potency of 5-LO inhibitors and FLAP antagonists has been based on inhibition of ex vivo LTB₄ production from whole blood leukocytes. Following administration of therapeutic doses, zileuton, BAY x1005, and MK 886 each produced approximately 90% inhibition of ex vivo LTB₄ production (1, 2). Based upon the clinical activity of these agents, this degree of LT inhibition appears sufficient for meaningful efficacy.

	EFFECT OF LEUKOTRIENE RECEPTOR ANTAGONISTS AND 5-LIPOXYGENASE INHIBITORS IN ASPIRIN-INDUCED ASTHMA

In patients with aspirin-sensitive asthma, aspirin ingestion produces a dramatic fall in FEV₁; patients also develop tearing, sneezing, itchy eyes, runny nose, and, often, severe nausea and vomiting. These symptoms begin 1-2 h after aspirin ingestion and persist for 3-4 h. Baseline urinary LTE₄ excretion, which is higher in patients with aspirin sensitivity than in control patients with asthma, increases approximately fivefold following aspirin ingestion before returning to baseline as the aspirin response resolves (3). This observed increase in urinary LTE₄ excretion is consistent with the increased production of cysLTs during episodes of aspirin-induced asthma.

Eight patients with known sensitivity to aspirin that was accompanied by increases in urinary LTE₄ excretion were recruited and participated in a randomized, double-blind, placebo-controlled, crossover trial to study the effects of zileuton on aspirin-induced asthma (4). These patients received randomized, blinded treatment with zileuton (600 mg four times per day) or placebo for 6 to 8 d before an aspirin challenge. Zileuton decreased basal urinary LTE₄ excretion from 469 to 137 picograms (pg)/mg creatinine, and blunted the aspirin- induced increase in urinary LTE₄ from 3,539 to 1,120 pg/mg creatinine. Zileuton did not affect basal FEV₁ but prevented the aspirin-induced decrease in FEV₁. In this study, patients were able to subjectively discriminate the drug's efficacy because of the telltale nasal symptoms, which on aspirin challenge generally appear slightly before the bronchospastic effects. When patients received blinded placebo treatment, they developed significant nasal symptoms; however, after receiving zileuton, they had very few nasal symptoms following aspirin challenge. In addition, gastrointestinal and dermal symptoms were present in five and three patients, respectively, following aspirin challenge in the placebo period, but these symptoms did not occur following aspirin challenge in any patient who had been pretreated with zileuton.

In a similar study, another 5-LO inhibitor, ZD-2138, protected against aspirin-induced bronchoconstriction (5). However, this study differed in that patients were treated with only a single 350-mg dose of ZD-2138 prior to aspirin challenge. The aspirin-induced decrease in FEV₁ was 20.3% among patients receiving placebo, but only 4.9% following pretreatment with ZD-2138. The protective effect was accompanied by a 72% inhibition of ex vivo LTB₄ generation from whole blood leukocytes and a 74% inhibition of urinary LTE₄ excretion.

The cysLT₁ receptor antagonists also exhibit efficacy in aspirin-sensitive patients with asthma. One study is especially noteworthy because it did not evaluate the drug effect against aspirin challenge; rather, it looked at patients' resting pulmonary function (6). Patients received either 825 mg verlukast (MK-0679) or placebo; FEV₁ was then assessed. Verlukast produced an average 18% increase (range, 5-34%) in FEV₁; this bronchodilation lasted for 9 h after drug treatment. This bronchodilatory effect on resting FEV₁ also has been noted with other leukotriene modulators. It is important to note that these agents are not bronchodilators; rather, they produce their effects by inhibiting cysLT-induced airway tone that is present in many patients with aspirin-sensitive asthma.

	EFFECT OF LEUKOTRIENE RECEPTOR ANTAGONISTS AND 5-LIPOXYGENASE INHIBITORS IN CHRONIC, STABLE ASTHMA

As of September 1997, the largest "database" of LT modulator effects in chronic, stable asthma is in the form of trials published using zileuton or zafirlukast as the active agent. There are many data published in abstract form with montelukast and pranlukast; these will not be covered herein.

4- to 6-Week Trials

Following a 1-wk placebo lead-in period, zileuton was administered at doses of 600 mg four times per day or 800 mg twice per day in a randomized, double-blind, placebo-controlled 4-wk trial involving patients with mild-to-moderate asthma (7). Each patient had an FEV₁ of 40 to 75% of predicted (mean approximately 60%) and exhibited at least a 15% increase in FEV₁ after two puffs of albuterol. There are two important points to stress about this trial. First, there was an approximate 8% placebo effect on FEV₁ that continued through the first 3 wk of the trial, then decreased. At the 4-wk endpoint, patients receiving zileuton (600 mg four times per day) exhibited a 15% improvement in FEV₁, relative to baseline. Second, the measurement of peak expiratory flow rate was not plagued by a placebo effect; drug effects were therefore evident after 2 wk of treatment (Figure 1). In this trial, zileuton reduced urinary LTE₄ excretion by 39%; it also reduced asthma symptoms and the frequency of -agonist use.

View larger version (14K): [in this window] [in a new window]   View larger version (21K): [in this window] [in a new window]   Figure 1.   Effect of zileuton in chronic, stable asthma. Patients were treated with either placebo or zileuton at doses of 600 mg four times per day or 800 mg twice per day for 4 wk. The mean change from baseline FEV1 (A) and morning peak flow (B) are shown. Reprinted from Reference 7 by permission.

Zafirlukast was evaluated in a similar patient population using a similar trial design (8). Four treatment groups were evaluatedplacebo, and zafirlukast at doses of 5, 10, and 20 mg twice per day. Each treatment group had a mean FEV₁ of 60 to 62% of predicted during patient screening. In this study, a placebo effect on FEV₁ did not occur, enabling a dose- related effect to be observed. By 2 wk, there was a 10 to 15% improvement in FEV₁ compared with baseline; this effect was maintained for the 6-wk duration of the trial. Further, changes in FEV₁ correlated to zafirlukast serum concentrations (Figure 2). As pulmonary function improved, -agonist use dropped from about six puffs per day to four, a 30% decrease. Zafirlukast (20-mg twice daily dose) decreased the number of nighttime awakenings by 46%, compared with placebo treatment.

View larger version (11K): [in this window] [in a new window]   Figure 2.   Association between change in FEV1 and zafirlukast plasma concentration. Patients with asthma received placebo or zafirlukast at doses of 5, 10, or 20 mg two times per day for 6 wk. The relationship between trough zafirlukast concentrations and the change in FEV1 is shown (*p < 0.01 for the change in FEV1 between 40 mg daily dose and placebo; p < 0.05 for the linear trend with the dose of zafirlukast). Reprinted from Reference 8 by permission.

In these 4- to 6-wk trials, LT modulators produced a sustained 10 to 15% improvement in FEV₁, decreased asthma symptoms, especially those occurring at night, and decreased -agonist use. In these short-term trials, both agents were well tolerated.

13- to 26-Week Trials

Zafirlukast was evaluated in a 13-wk, randomized, double-blind trial in 762 nonsmokers who were 12 yr of age or older with mild-to-moderate asthma (9, 10). Entry criteria dictated that weekly daytime asthma scores had to be at least 8 out of a maximum score of 21, based on a daily score of 0 to 3, and that patients had FEV₁ values greater than or equal to 55% of predicted at least 6 h after -agonist use. Because this trial was designed to show an improvement in asthma symptoms, it should be noted that the difference between a daily score of 2 and a score of 1 was clinically significant.

By the end of 13 wks' treatment with zafirlukast, pulmonary function (i.e., morning peak expiratory flow rate, FEV₁, percent predicted FEV₁, and percent of baseline personal-best FEV₁) significantly improved relative to placebo (p < 0.05 for each parameter) (9). Among zafirlukast-treated patients, FEV₁ increased to greater than 80% predicted in significantly more patients than in the placebo group (p < 0.01), which was particularly meaningful because patients receiving zafirlukast used less ₂-agonist (p < 0.01). Patients receiving active treatment also showed significant improvements relative to placebo, with respect to daytime asthma symptom scores (p < 0.05) (Figure 3A), nighttime awakenings (p < 0.05) (Figure 3B), and morning awakenings with asthma (p < 0.01).

View larger version (14K): [in this window] [in a new window]   View larger version (16K): [in this window] [in a new window]   Figure 3.   Effect of zafirlukast on asthmatic symptoms. Patients with mild asthma were treated with either placebo or 20 mg twice daily zafirlukast for 13 wk. The effects of treatment on daytime (A) and nocturnal (B) symptoms are shown. From J. E. Fish, editor. 1995. Zafirlukast: introduction to a new class of asthma therapy (slide kit). Zeneca Pharmaceuticals, Wilmington, DE. Reprinted by permission.

In this group of patients with mild-to-moderate asthma, 6.5% of placebo-treated patients developed asthma exacerbations that required additional asthma medications (10). This rate of asthma exacerbations over a 13-wk period was quite similar to that found in a large NIH-sponsored trial of patients who had asthma of similar severity (11). However, only 3.1% of zafirlukast-treated patients experienced asthma exacerbations during the same 13-wk, double-blind period (10).

Health economic data collected from a subset (n = 146) of patients in this trial were also evaluated (12). As described in greater detail in the article by William Calhoun (p. S238- S246), patients who received zafirlukast experienced significantly fewer days per month with asthma symptoms or asthma episodes than did patients who received placebo. They also had fewer absences from work or school, required less rescue -agonist, and had fewer health care contacts. Taken together, these effects are highly meaningful, both for patients and physicians, because they suggest that fewer episodes of unscheduled health care result from antileukotriene therapy.

Zileuton has been evaluated in both 13- and 26-wk studies (13, 14). Following a placebo lead-in period, patients were randomized to receive either placebo or zileuton (400 or 600 mg four times per day). In this study, the baseline FEV₁ was ~ 60% of predicted. Pulmonary function in these patients was significantly impaired; there was therefore more "room" for one to see a drug effect on FEV₁ than was the case for patients in the previously described zafirlukast trial. In these trials of zileuton, FEV₁ was measured at the time of trough serum levels, i.e., before patients took their morning dose. Zileuton improved FEV₁ by 15 to 20%, which was significantly better than the 6 to 7% improvement seen in the placebo group. After over 8 wk of treatment, the magnitude of zileuton's effect on FEV₁ was slightly greater than that observed with zileuton (7) or zafirlukast (8) in the 4- to 6-wk trials in patients with similar baseline FEV₁ values.

The number of patients requiring corticosteroid rescue over the course of treatment was reduced by zileuton. Patients were stratified into groups based upon baseline FEV₁ (> 70%, 50-70%, and < 50% of predicted). Zileuton exhibited its greatest effects in the most severely affected patients, i.e., the ones who required corticosteroids most frequently (13) (Figure 4). Corticosteroid rescue was needed by 35% of the patients who received placebo but by only 8% who received the highest dose of zileuton. In patients with less severe asthma, corticosteroid use was also reduced by zileuton, but by a smaller degree.

View larger version (30K): [in this window] [in a new window]   Figure 4.   Zileuton reduces corticosteroid use in asthma patients. Patients treated with zileuton or placebo were stratified by baseline FEV1. The percentage of patients requiring corticosteroid rescue treatment is plotted. Groups were stratified by percent predicted FEV1 values at study entry (111 patients had FEV1 > 70% predicted; 187 patients had FEV1 between 50 and 70% predicted; and 103 patients had FEV1 < 50% predicted). Reprinted from Reference 13 by permission.

Airway Inflammation

Both zileuton and zafirlukast have been assessed for their effects on indices of airway inflammation. In a trial with zileuton, 12 patients with nocturnal asthma and 6 normal control subjects participated in a randomized, placebo-controlled, crossover study consisting of a 2-wk, single-blind, lead-in period followed by two 1-wk, double-blind periods separated by a 1-wk washout period (15). The subjects underwent pulmonary function testing, bronchoscopy, and urine collection at 4:00 P.M. and 4:00 A.M. The 4:00 A.M. measurement was repeated after patients had received zileuton at 600 mg four times per day for 1 wk. Whereas LT levels in bronchoalveolar lavage (BAL) fluid were similar in the asthma patients and control subjects at 4:00 P.M., they were significantly higher at the 4:00 A.M. sampling in asthma patients. Furthermore, LTB₄ levels significantly correlated with decreases in nocturnal FEV₁, and urinary LTE₄ excretion was elevated among patients with nocturnal asthma relative to control subjects. Treatment with zileuton was associated with decreased LTB₄ levels in BAL fluid and decreased urinary LTE₄; there was a trend toward improvement in the nocturnal FEV₁. These features were associated with decreased blood eosinophil counts while patients received zileuton; this observation provides a cellular correlate of the drug's anti-inflammatory activity.

As part of a subprotocol at some centers participating in the 13-wk zafirlukast trial, patients were challenged with methacholine before and after 2 and 10 wk of treatment with 20 mg zafirlukast twice daily (16). The provocative cumulative dose of methacholine that produced a 20% fall in FEV₁ (PD₂₀FEV₁) was measured. Patients treated with zafirlukast required 2.5-fold higher methacholine challenge doses to elicit the stipulated decline in pulmonary function than did placebo-treated patients. This effect was statistically significant after 2 wk of treatment and approached statistical significance after 10 wk, possibly because a smaller number of patients was tested. The change in PD₂₀FEV₁ was on the same order as has been seen with inhaled budesonide (17). Thus, zafirlukast reduced methacholine reactivity, which is often considered a measure of airway inflammation (18).

Safety Profile

Leukotriene modulators have been well tolerated in clinical trials. Of the 4,000 patients treated with zafirlukast, there have been rare cases of hepatitis (non-A, -B, or -C); it is unknown whether these were drug related. With zileuton in a safety trial of more than 3,000 patients, 4.6% of patients on drug versus 1.1% on usual care for asthma developed reversible elevations in hepatic transaminases that were greater than three times the upper limit of the reference range. Of these, 80% occurred within the first 3 wk of treatment. Use of zileuton was discontinued in patients experiencing elevations greater than five times the upper limit of the reference range in hepatic transaminases; all patients returned to normal liver function when the zileuton was discontinued.

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION CONCLUSION DISCUSSION REFERENCES

Leukotriene modulators are clearly effective in asthma. Both drug classes improve FEV₁ and morning and evening peak expiratory flow. Both decrease asthma symptoms, especially those occurring at night, which is highly important. Both decrease -agonist use. Further, studies suggest that these drugs may reduce asthmatic exacerbations and indices of airway inflammation. Precise comparison of the database between the two agents is difficult because, overall, the patients treated with zafirlukast have had asthma of diminished severity compared with those treated with zileuton. Nevertheless, both drugs have shown clinically meaningful efficacy as asthma treatments; this has occurred with zafirlukast in patients with mild asthma and with zileuton in patients with mild-to-moderate asthma.

	DISCUSSION

TOP ABSTRACT INTRODUCTION CONCLUSION DISCUSSION REFERENCES

Dahlén: The effects of nasal function in the aspirin-intolerant asthmatic is very important. There are data, particularly with zafirlukast but also with zileuton, in allergen-induced nasal obstruction where one gets a protective effect. I think that this new conceptthat you can treat both the upper and lower airways with a single, orally active drugis very important.

I think that it is also very important to stress that these drugs are not a new corticosteroid. These drugs still have important effects in asthma, including interference with bronchial hyperresponsiveness, and some effects on cells that may be important in the inflammatory process. Also, I think that it is important not to jump onto the "steroid similarity bandwagon."

Busse: What hasn't been emphasized enough is that these drugs provide a rather consistent pattern of rapid improvement in airflow obstruction, even though they are not bronchodilators. They are, no doubt, displacing an intrinsic leukotriene tone in the airway, although this response is highly variable. An important aspect that should be mentioned is the additive effect these compounds may have with -agonists.

Bernstein: In both the zafirlukast and zileuton trials, patients were taking -agonists as needed. It appeared to me that in both studies -agonist use in the placebo group went up during the course of the study.

Drazen: In the study with zafirlukast, -agonist use went up in the placebo group. In the zileuton trial, -agonist use actually went down, although you can say that there was significantly less -agonist use in the active treatment group than in the placebo group. Since the trials are double-blind, placebo-controlled, I think that the patients are using -agonists only as needed. Every time they use their inhaler, it tells you that their asthma is causing symptoms; treatment is needed less often when they are on active medication.

Bernstein: The efficacy of these agents at equivalent doses so far is the same. I think that is a very important point.

Drazen: I believe that if tomelukast had not been toxic, it would have been on the market years ago. In fact, everything that I've seen to date with these agents indicates that they all have virtually identical clinical profiles in patients with mild asthma. Therefore, the only differences boil down to cost, convenience, and safety, which are the only issues to debate among the various agents.

Peters-Golden: These drugs aren't going to work in all patients, but you're going to get an indication of whether they do work in an individual patient fairly quickly, which is really important. If you want to determine how a patient is going to respond to inhaled steroids, you need to administer them for a fairly long time, because some people are quite slow to reach a plateau of corticosteroid effects. From what I've seen, patients will reach the plateau much more quickly with leukotriene modulators.

Bernstein: Are you suggesting that you might start a patient with mild asthma on one of these drugs?

Peters-Golden: What I am saying is that there are a number of obstacles to starting a patient on inhaled steroids at the present time. One is clearly the compliance problems with MDI use. The second is that patients don't get a rapid beneficial response. Thus, there is little in the way of positive reinforcement with corticosteroid use. Education needs to occur on a continuing basis. The third obstacle to corticosteroid use is that it will often take weeks or longer for patients to obtain a beneficial response. All three of these obstacles to corticosteroid use favor use of cysLT receptor antagonists and 5-LO inhibitors.