ORIGINAL INVESTIGATION Montelukast, a Once-Daily Leukotriene Receptor Antagonist, in the Treatment of Chronic Asthma A Multicenter, Randomized, Double-blind Trial

Theodore F. Reiss, MD; Paul Chervinsky, MD; Robert J. Dockhorn, MD; Sumiko Shingo, MS; Beth Seidenberg, MD; Thomas B. Edwards, MD; for the Montelukast Clinical Research Study Group

Objectives: To determine the clinical effect of oral mon- Results: Montelukast improved airway obstruction telukast sodium, a leukotriene receptor antagonist, in asth- (forced expiratory volume in 1 second, morning and matic patients aged 15 years or more. evening peak expiratory flow rate) and patient-reported end points (daytime asthma symptoms, “as-needed” Design: Randomized, multicenter, double-blind, placebo- ␤-agonist use, nocturnal awakenings) (PϽ.001 com- controlled, parallel-group study. A 2-week, single- pared with placebo). Montelukast provided near- blind, placebo run-in period was followed by a 12- maximal effect in these end points within the first day week, double-blind treatment period (montelukast of treatment. Tolerance and rebound worsening of asthma sodium, 10 mg, or matching placebo, once daily at bed- did not occur. Montelukast improved outcome end points, time) and a 3-week, double-blind, washout period. including asthma exacerbations, asthma control days (PϽ.001 compared with placebo), and decreased periph- Setting/Patients: Fifty clinical centers randomly allo- eral blood eosinophil counts (PϽ.001 compared with pla- cated 681 patients with chronic, stable asthma to receive pla- cebo). The incidence of adverse events and discontinu- cebo or montelukast after demonstrating a forced expira- ations from therapy were similar in the montelukast and tory volume in 1 second 50% to 85% of the predicted value, placebo groups. at least a 15% improvement in forced expiratory volume in 1 second (absolute value) after inhaled ␤-agonist adminis- Conclusions: Montelukast, compared with placebo, sig- tration, a minimal predefined level of daytime asthma symp- nificantly improved asthma control during a 12-week toms, and inhaled ␤-agonist use. Twenty-three percent of treatment period. Montelukast was generally well toler- the patients used concomitant inhaled corticosteroids. ated, with an adverse event profile comparable with that of placebo. Primary End Points: Forced expiratory volume in 1 second and daytime asthma symptoms. Arch Intern Med. 1998;158:1213-1220

STHMA IS a significant matory cells, including eosinophils and mast worldwide health prob- cells, and are at least 1000 times more po- lem, accounting for $4.2 tent bronchoconstrictors than histamine or From the Departments of billion of health care costs methacholine in normal and asthmatic sub- Pulmonary/Immunology in the United States in jects.4 The leukotrienes mediate many of the (Drs Reiss and Seidenberg) 1995.1 Despite the development and in- pathophysiologicalprocessesassociatedwith and Biostatistics (Ms Shingo), A 2 Merck Research Laboratories, stitution of treatment guidelines, asthma asthma, including microvascular leakage, Rahway, NJ; NE Research remains a costly clinical problem, with a bronchoconstriction,andeosinophilrecruit- 5 Center, Inc, North Dartmouth, continuous need for new, innovative treat- ment into the airways. Agents that interrupt Mass (Dr Chervinsky); ments. Current therapies have limita- theactionoftheleukotrienes(5-lipoxygenase International Medical tions, including poor compliance (inhal- inhibitors and leukotriene receptor antago- Technology Consultants Inc, ers, dosage frequency) and side effects.3 nists) have demonstrated improvement of Prairie Village, Kan New, effective, well-tolerated oral thera- chronic asthma in clinical trials, thus pro- (Dr Dockhorn); and Allergy pies may have a substantial impact on the viding evidence for their role in asthma.6-9 and Asthma Center, Albany management of asthma. Montelukast sodium is a potent and Medical Center, Albany, NY The role of the cysteinyl leukotrienes specific leukotriene receptor antagonist10 (Dr Edwards). A list of the members of the Montelukast (leukotrienes C4,D4, and E4) in asthma has that has been shown to have substantial Clinical Research Study Group been clearly established. These leukotrienes blockade of airway leukotriene receptors 11 is given on page 1219. are produced and released from proinflam- 24 hours after oral dosing. This long du-

ARCH INTERN MED/ VOL 158, JUNE 8, 1998 1213

©1998 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ on 09/25/2021 PATIENTS AND METHODS at the prestudy visit. Patients who demonstrated compe- tence withthe use of these instruments and the ability to perform reproducible spirometry at each clinic visit were STUDY DESIGN eligible for period 2.

This multicenter, randomized, double-blind, placebo- EXCLUSION CRITERIA controlled, 3-period, parallel-group trial compared the clini- cal effect of oral montelukast sodium, 10 mg once daily at Active upper respiratory tract infection within 3 weeks, acute bedtime, and placebo. The study consisted of a 2-week, sinusdiseaserequiringantibiotictreatmentwithin1week,emer- single-blind, placebo run-in period (period 1); a 12-week, gency department treatment for asthma within 1 month, or double-blind, active treatment period (period 2); and a 3- hospitalization for asthma within 3 months before the prestudy week, double-blind, placebo washout period (period 3). visit were study exclusions. Excluded medications included Clinic visits occurred every 2 weeks during period 1 and oral, inhaled (concomitant inhaled medication use was allowed every 3 weeks thereafter. for a subset of patients), and parenteral corticosteroids within The study was conducted at 50 study centers in the 1 month; cromolyn sodium, nedocromil sodium, terfenadine, United States between October 21, 1994, and August 13, and loratadine within 2 weeks; theophylline (oral and intra- 1995; 681 patients were randomly assigned, according to venous), ␤-agonists (oral or long-acting inhaled), and anticho- a computer-generated allocation schedule, to receive ei- linergic agents within 1 week; astemizole within 3 months; and ther a film-coated tablet of montelukast sodium, 10 mg, or immunotherapy initiated within 6 months before the prestudy matching placebo. In period 3, a subset of patients was visit. According to a standardized protocol, oral corticosteroids blindly switched from montelukast to placebo according were allowed for treatment of worsening asthma during pe- to the computer-generated allocation schedule. riods 2 and 3. Patients who required rescue during period 1, Written informed consent approved by the respec- more than 2 rescues during periods 2 and 3, or change in im- tive institutional review boards was obtained from each pa- munotherapy were discontinued from the study. tient. If the patient was younger than 18 years, consent was also obtained from the patient’s parent or guardian. EVALUATIONS

INCLUSION CRITERIA The FEV1 and daytime asthma symptom score were pre- specified as primary end points. Other prespecified end Healthy, nonsmoking patients (male and female), aged 15 points were morning and evening peak expiratory flow rate years and older with at least 1 year of intermittent or per- (PEFR), daily use of inhaled short-acting as-needed ␤- sistent asthma symptoms, were enrolled. Female patients agonist, nights per week with nocturnal awakenings, asthma- had a negative serum ␤-human chorionic gonadotropin test specific quality of life, physician’s and patient’s global evalu- at the prestudy visit. All patients used short-acting in- ations, change in peripheral blood eosinophil counts, and haled ␤-agonists as needed to treat their asthma, and a per- asthma outcome end points including episodes of wors- centage of patients (not to exceed 25%) were allowed con- ening asthma (percentage of days with asthma exacerba- comitant inhaled corticosteroids at a constant dosage tions), use of rescue oral corticosteriods (percentage of pa- beginning at least 4 weeks before the prestudy visit. Pa- tients), discontinuation because of worsening asthma tients with non–life-threatening, clinically stable, concomi- (determined by whether additional asthma medications were tant diseases could be enrolled in the study. required), and asthma control days. Patients were eligible for randomization if they had, Spirometry was performed at each clinic visit between 6 on at least 2 of the 3 visits during period 1, a forced expi- and 9 AM, approximately 10 to 12 hours after the previous dose ratory flow in 1 second (FEV1) between 50% and 85% of of study medication and after ␤-agonist and short-acting an- the predicted value (after withholding ␤-agonist for at least tihistamines had been withheld for at least 6 and 48 hours, re- 6 hours) and an absolute increase in FEV1 of at least 15%, spectively.Patientsusinginhaledcorticosteroidswereinstructed 20 to 30 minutes after inhalation of ␤-agonist. In addi- to take the morning dose either an hour before or after a clinic tion, patients were required to have a minimum total 2- visit. The spirometry measurements were collected with a stan- week daytime asthma symptom score of 64 (a maximum dard spirometer (Nellcor/Puritan-Bennett PB 100/PB110, Len- score of 336 was possible) and to have used a daily aver- dena, Kan) and transmitted via modem to a central spirom- age of at least 1 puff of ␤-agonist during period 1. etry quality control center, where the data were reviewed to Patients received a peak flow meter (Mini-Wright; ensure uniform adherence to American Thoracic Society stan- Clement Clark, Columbus, Ohio) and a practice diary card dards of acceptability and reproducibility.12 Continual per-

ration of action distinguishes it from other leukotriene RESULTS receptor antagonists. Dose-finding studies in adults have identified 10 mg once daily at bedtime as the minimal PATIENTS dose to achieve the maximal response with no dose- limiting toxic effects.9 Six hundred eighty-one patients entered period 2, the The purpose of this 12-week clinical trial was to de- double-blind treatment period; 408 were allocated to mon- termine the effect of oral montelukast on asthma con- telukast and 273 to placebo treatment (Figure 1). The pa- trol end points (airway obstruction, patient-reported end tients enrolled were white (89%), African American (4.1%), points, and asthma outcomes) and to evaluate its safety and Hispanic (4.7%), with a similar distribution between and tolerability profile. treatment groups. Other baseline demographic character-

ARCH INTERN MED/ VOL 158, JUNE 8, 1998 1214

©1998 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ on 09/25/2021 formance feedback was given to clinical centers to main- and blood eosinophil counts (determined by an automated tain and enhance spirometry quality. The largest FEV1 from cell counter in a central laboratory) were performed. Fe- a set of at least 3 maneuvers was the visit value. Airway re- male patients had serum ␤-human chorionic gonadotropin versibility was evaluated at each visit during period 1 and measured at the prestudy visit and either a serum or urine at predefined visits during periods 2 and 3. pregnancy test at each visit. A complete physical examina- The daily diary card contained daytime asthma symp- tion and 12-lead electrocardiogram were performed before tom and nighttime awakening scales, previously shown to and after randomization; vital signs were recorded at each visit. have acceptable evaluative measurement properties.13 The 4 daytime asthma symptom questions addressing the se- STATISTICAL METHODS verity and bothersomeness of asthma symptoms (using a 7-point scale where 0 indicates best and 6, worst) were com- The primary analysis was an intention-to-treat approach bined into a mean daily score. Nighttime awakenings were including all randomized patients with a baseline value evaluated by the response to a single question by means of and at least 1 treatment period measurement. The data a 4-point scale (“no awakenings” to “awake all night”).13 were analyzed as averages during the treatment period, The change in nocturnal awakenings was determined for and data points were not carried forward. For end points the prespecified group of patients with 2 or more nights analyzed as change or percentage change from baseline, with awakenings per week during the run-in period. the average period 1 measurement was the baseline value. The PEFR was measured by the patient in the morning, The mean period 2 response was compared between treat- on arising, and in the evening, at bedtime, before taking the ment groups by means of an analysis of variance study medication. The largest of 3 measurements was recorded (ANOVA) model that included terms for treatment, on the diary card, and measurements performed within 4 hours inhaled corticosteroid use (stratum), and study center. of ␤-agonist use were identified. The patients also recorded The between-group differences of within-group change as-needed ␤-agonist use during the day and at night, and oral and the 95% confidence interval (CI) were computed on corticosteroid rescue, visit to a physician’s office, or hospital- the basis of the ANOVA model. Quality of life and the ization because of worsening asthma. At the completion of pe- global evaluations were analyzed by the ANOVA model. riod 2 (week 12), physicians and patients independently evalu- In addition, the 7 categories of the global evaluations were ated the change in the patient’s asthma (global evaluations) collapsed into 3 categories (better, no change, and worse) by selecting the most appropriate response by means of a 7- and analyzed with a Cochran-Mantel-Haenszel test to cor- point scale (“very much better,” “moderately better,” “a little roborate the ANOVA results. better,”“unchanged,”“alittleworse,”“moderatelyworse,”“very The presence of quantitative interactions between demo- much worse”). At the randomization visit (before patients re- graphic subgroups and changes in the study end points were ceived study medication) and at the end of period 2 (week 12), testedbytheANOVAmodel.Interactionswereconsideredclini- the patients also completed the validated Asthma Quality of cally significant if a demographic characteristic had a signifi- Life Questionnaire.14 The questionnaire contained 32 ques- cantinteractionwithatleast2of4endpoints.Correlationanaly- tions divided into 4 quality-of-life domains—activity, symp- sis among baseline and changes in end point values (FEV1 and toms,emotions,andenvironment—withresponsesona7-point daytime asthma symptoms) were also performed. scale where 0 indicates worst and 6, best. Assumptions of normality and homoscedasticity were An asthma exacerbation day was defined as a day when assessed. All statistical tests were 2 tailed, and PՅ.05 was any 1 of the following occurred: a decrease of more than 20% considered statistically significant. from baseline in morning PEFR, PEFR less than 180 L/min, The safety evaluations included all randomized pa- an increase of more than 70% from baseline in ␤-agonist use tients. The number and percentage of patients reporting (a minimum increase of 2 puffs), an increase of more than 50% clinical adverse experiences and laboratory abnormalities frombaselineinsymptomscore,“awakeallnight”withasthma, were summarized by treatment group. or worsening asthma requiring oral corticosteroid rescue, visit to a physician’s office, or hospitalization. An asthma control POWER AND SAMPLE SIZE daywasdefinedasanydaywhennoneofthefollowingoccurred: worsening asthma requiring oral corticosteroid rescue, visit The study was designed with a sample size of 300 and 200 to a physician’s office or hospitalization, nocturnal awaken- patients for montelukast and placebo groups, respec- ings, or use of more than 2 puffs of ␤-agonist. tively, to have 95% power to detect (␣=.05, 2-tailed test) a Blood samples were obtained before and 3, 6, 12, and mean difference between treatment groups of 5.4 percent- 15 weeks after randomization to period 2. Clinical labora- age points in FEV1 (percentage change from baseline) and tory tests (ie, hematology, serum chemistry, and urinalysis) 9.1% in daytime symptom score (change from baseline).

istics were comparable between the montelukast and pla- EFFICACY cebo groups (Table 1). Six hundred seven patients (89%) completed periods 2 and 3; the discontinuation rate was Montelukast significantly improved (PϽ.001 compared significantly (PϽ.05) higher in the placebo (14.3%) than with placebo) airway obstruction, as shown by an in- the montelukast (8.6%) group (Table 2). crease in FEV1 of 13.1% (placebo, 4.2%), in morning PEFR Five patients (2 and 3 in the montelukast and pla- of 24.0 L/min (placebo, 4.6 L/min), and in evening PEFR cebo groups, respectively) and 8 patients (4 in each treat- of 15.9 L/min (placebo, 4.2 L/min). The mean differ- ment group) were excluded from the intention-to-treat ence compared with placebo, based on ANOVA, was 8.9% analysis because of missing baseline or treatment period (95% CI, 6.8% to 11.0%) for FEV1, 19.4 L/min (95% CI, data for FEV1 and daytime symptom score, respectively. 14.2 to 24.5 L/min) for morning PEFR, and 11.6 L/min

ARCH INTERN MED/ VOL 158, JUNE 8, 1998 1215

©1998 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ on 09/25/2021 1515 Table 1. Demographic and Baseline Characteristics: Screened Patients Randomized Patients* 681 Patients Randomized After 2-wk Placebo Run-In Placebo Montelukast Sodium, (n = 273) 10 mg (n = 408) 273 Patients 408 Patients Received Received Placebo Montelukast Sodium Age, median (range) 31 (15-79) 31 (15-78) Duration of asthma, y 15 (0.5-62) 16 (0.5-64) Clinic Visits Every 3 wk Clinic Visits Every 3 wk 12-wk median (range) Active 35 Patients (12.8%) 31 Patients (7.5%) Sex, % of patients Treatment ∗ Discontinued in Period 2 Discontinued in Period 2∗ Male 46.9 43.4 Female 53.1 56.6 238 Patients (87.2%) 377 Patients (92.5%) Concomitant inhaled 23.4 22.3 Completed Period 2 Completed Period 2 corticosteroids, % of patients Continued Switched to History of exercise-induced 95.2 92.4 Montelukast Placebo asthma, % of patients 330 (80.8%) 47 (11.5%) History of allergic rhinitis, 91.9 88.5 % of patients 4 Patients (1.5%) Montelukast Placebo 3-wk ∗ Discontinued in Period 3 4 (0.1%) 0 (0%) FEV1, L 2.5 ± 0.7 2.5 ± 0.7 Placebo ∗ Discontinued in Period 3 FEV1, % of predicted 68.5 ± 11.1 66.7 ± 10.6 Washout PEFR, L/min Period 234 Patients (85.7%) Montelukast Placebo Completed Study 326 (79.9%) 47 (11.5%) Morning 391 ± 90 381 ± 88 Completed Study Evening 421 ± 90 413 ± 90 Daytime asthma 2.5 ± 0.8 2.5 ± 0.8 Figure 1. Study profile. Of the 1515 screened patients, 834 were not symptom score† randomized; the most common reason was failure to meet the spirometry ß-Agonist, puffs/d 5.3 ± 3.2 5.4 ± 3.2 criteria. Patients were randomized to receive either oral montelukast sodium, Nocturnal awakenings, nights/wk 4.0 ± 2.6 4.0 ± 2.4 10 mg, or matching placebo. Approximately 89% of the patients completed Quality of life score, 3.6 ± 0.9 3.6 ± 0.9 the study. Asterisk indicates that the reasons for discontinuation are listed in pooled domains Table 2. Eosinophils, ϫ109/L 0.31 ± 0.23 0.33 ± 0.26

*Data are mean ± SD unless otherwise specified. FEV1 indicates forced (95% CI, 6.9 to 16.3 L/min) for evening PEFR. The im- expiratory volume in 1 second; PEFR, peak expiratory flow rate. †The daytime asthma symptom score was the average of the responses provement observed in evening PEFR indicated that mon- to 4 questions scored on a scale of 0 to 6. telukast provided protection throughout the 24-hour dos- ing interval. Also, patient-reported end points, eg, daytime asthma symptoms and as-needed ␤-agonist, were signifi- Table 2. Reasons for Discontinuation During Periods 2 and 3 cantly (PϽ.001 compared with placebo) improved by montelukast (Figure 2). Furthermore, patients re- No. (%) ported significantly less nocturnal awakening (−1.66 and Montelukast −0.80 nights per week for montelukast and placebo, re- Placebo Sodium, 10 mg spectively); the mean difference, based on ANOVA, was (n = 273) (n = 408) −0.87 (95% CI, −1.22 to −0.53). Clinical adverse experiences 12 (4.4) 9 (2.2) The improvements observed in airway obstruction Positive pregnancy tests 1 (0.4) 1 (0.2) and patient-reported end points were maintained con- Withdrawal of consent 12 (4.4) 13 (3.2) sistently throughout the 12-week treatment period 2 (Fig- Protocol deviations 9 (3.3) 6 (1.5) ure 2). The prespecified patient subgroup that was blindly Unavailable for follow-up 5 (1.8) 6 (1.5) switched from montelukast to placebo during period 3 Total Discontinued 39 (14.3) 35 (8.6)* showed the treatment effects returned toward, but not past, the placebo group, confirming the beneficial ef- *PϽ.05 compared with placebo. fects of montelukast, and withdrawal of montelukast did not cause rebound worsening of asthma (Figure 2). perienced fewer days with asthma exacerbations (a de- Within 1 day of dosing, montelukast achieved near- crease of 31%) and more asthma control days (an increase maximal effect as shown by the response during the first of 37%) than patients receiving placebo (PϽ.001) 21 days of period 2. Figure 3 illustrates this rapid, ben- (Figure 6). Fewer patients (a decrease of 28%) treated eficial response for ␤-agonist use, daytime asthma symp- with montelukast required oral corticosteroid rescues toms, and morning PEFR. Similar improvements were (6.9% compared with 9.6% for placebo; P=.20), and fewer seen in nocturnal awakenings and evening PEFR. In ad- patients (a decrease of 59.5%) discontinued therapy be- dition, each asthma-specific quality-of-life domain had cause of worsening asthma (1.5% compared with 3.7% significantly higher scores for patients treated with mon- for placebo; P=.07). telukast (P≤.001 compared with placebo) during the 12- Montelukast significantly decreased peripheral blood week treatment period (Figure 4). Also, patients’ and eosinophil counts (PϽ.001 compared with placebo) physicians’ global evaluations demonstrated that pa- (Figure 7). tients receiving montelukast had significantly improved There was no correlation between the improve- asthma control compared with patients receiving pla- ments in FEV1 or daytime asthma symptom scores and cebo (Figure 5). Patients treated with montelukast ex- patients’ baseline values. Furthermore, there were no clini-

ARCH INTERN MED/ VOL 158, JUNE 8, 1998 1216

©1998 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ on 09/25/2021 Montelukast Sodium Group Placebo Group Montelukast Sodium Group Placebo Group

30 15 0.0

–0.1 , 1 10 –0.2 20 –0.3

5 –0.4 Morning FEV –0.5 Change From Baseline % Change From Baseline

Daytime Symptoms Score, 10 0 –0.6 063912 15 063912 15

10 40 0 0 30

–10 20 Morning PEFR, L/min Change From Baseline

–20 10 –10 -Agonist Use, 1 3751791311 15 19 21 β –30 0 Morning PEFR, L/min Change From Baseline 0.1 % Change From Baseline –40 –10 0 3 6159 12 0 3961512 Washout Washout 0.0 Weeks in Active Treatment Weeks in Active Treatment

Figure 2. The effect of montelukast sodium and placebo on the 2 primary –0.1

end points (forced expiratory volume in 1 second [FEV1] and daytime asthma symptoms), morning peak expiratory flow rate (PEFR), and as-needed –0.2 ␤-agonist use during the 12-week, active treatment period and the 3-week, placebo washout period. The FEV1 was measured at every visit; PEFR, daytime asthma symptoms, nocturnal awakenings, and as-needed ␤-agonist –0.3 were recorded daily by the patients. Montelukast, compared with placebo,

caused significant ( PϽ.001) improvements in all end points. The dashed line Change From Baseline –0.4 represents the patient subgroup switched to placebo in the placebo washout

period. The values are reported as mean±SE. Daytime Asthma Symptoms Score, –0.5

–0.6 cally significant interactions between the prespecified sub- 1 3751791311 15 19 21

groups of age, sex, race, history of allergic rhinitis, his- 20 tory of exercise-induced asthma, study center, and concomitant use of inhaled corticosteroid and these study end points. For example, patients taking concomitant in- 10 haled corticosteroids had an increase in FEV1 of 10.3% with montelukast (1.6% with placebo), and patients with- 0 out corticosteroids had an increase in FEV1 of 13.9% with montelukast (5.0% with placebo). –10

SAFETY –20

The overall frequency of clinical adverse events reported by -Agonist Use, % Change From Baseline β –30 patients was similar between the montelukast and placebo groups. Upper respiratory tract infection and headache were –40 the most frequently reported clinical adverse events, simi- 1 3751791311 15 19 21 lar in incidence between treatments (Table 3). Twelve pa- Days in Active Treatment tients (4.4%) in the placebo group and 9 (2.2%) in the mon- Figure 3. The effect of montelukast sodium and placebo during the first 21 telukast group discontinued treatment because of adverse days in the active treatment period. The values are reported as mean±SE. experiences. Six of the 12 patients in the placebo group dis- PEFR indicates peak expiratory flow rate. continued because of asthma, 2 because of bronchitis, and the other 4 because of depression, facial edema, endome- triosis, and headache. Three of the 9 montelukast-treated 2 times above the upper limit of normal were infrequent patients discontinued treatment because of asthma; the other in both the montelukast and placebo groups (Յ0.9% and 6 patients discontinued because of anxiety, depression, Յ1.5%, respectively). Also rare (Յ0.7%), elevations of al- dyspnea, gastritis, back pain, and respiratory failure. kaline phosphatase and serum bilirubin levels were simi- There was no difference in the frequency of labora- lar in incidence between treatment groups. Laboratory ab- tory adverse events between the montelukast (7.1%) and normalities either returned toward normal while study placebo (5.5%) groups. The most frequently reported event therapy was continued, or had explanations not related was increased levels of alanine aminotransferase: 2.5% with to study medications, such as weight-lifting and minor montelukast and 1.5% with placebo treatment. Serum ala- blunt trauma injuries. No laboratory adverse event caused nine and aspartate aminotransferase elevations more than discontinuation.

ARCH INTERN MED/ VOL 158, JUNE 8, 1998 1217

©1998 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ on 09/25/2021 Montelukast Sodium Group Placebo Group Montelukast Sodium Group Placebo Group 0.80 ∗ 40 ∗ 0.70 ∗

0.60 ∗ ∗ 30 0.50

0.40 20 0.30 % of Days

0.20 ∗

Score, Change From Baseline 10 0.10

0.00 Activity SymptomsEmotions Environment 0 Domains Asthma Exacerbation Days Asthma Control Days Figure 4. Asthma-specific quality of life score in the montelukast sodium Figure 6. Effect of montelukast sodium compared with placebo on asthma and placebo groups. The values are reported as mean±SE. Asterisk indicates exacerbation and asthma control days during period 2. See “Patients and PϽ.001, montelukast compared with placebo. Methods” section for definition of the end points. The values are reported as percentage of total study days (mean±SE). Asterisk indicates PϽ.001, montelukast compared with placebo.

Montelukast Sodium Group Placebo Group

80 Montelukast Sodium Group Placebo Group 70 0.00 /L 9

60 10 × 50 –0.05 40

% of Patients 30

20 –0.10 10

0

Better No Change Worse Eosinophil Change From Baseline, –0.15 80 063912 15 Weeks in Active Treatment Washout 70 Figure 7. The effect of montelukast sodium on peripheral blood eosinophils 60 during the active treatment period and washout period. The values are reported as mean±SE change from baseline. PϽ.001 compared with placebo 50 during the 12-week treatment period. 40

% of Patients 30 Table 3. Clinical Adverse Experiences Occurring 20 in 6% or More of Patients in Either Treatment Group 10 No. (%) 0 Better No Change Worse Montelukast Figure 5. Mean percentage of patients with specific responses to the patient Placebo Sodium (top) and physician (bottom) global evaluations. The 0- to 6-point scale was (n = 273) (n = 408) collapsed to 3 categories: better (0, 1, and 2), no change (3), and worse (4, 5, and 6) for montelukast sodium and placebo. Treatment with montelukast Headache 57 (20.9) 73 (17.9) sodium, compared with placebo, showed significant improvement ( PϽ.001 Worsening asthma 32 (11.7) 46 (11.3) with Cochran-Mantel-Haenszel test). Upper respiratory tract infection 96 (35.2) 129 (31.6) Pharyngitis 29 (10.6) 22 (5.4) Sinusitis 22 (8.1) 31 (7.6) COMMENT This clinical trial demonstrates that montelukast pro- vided clinical benefit during the 12-week treatment pe- For each end point, the effect of montelukast was con- riod by consistent and significant improvement of all sistent throughout the double-blind treatment period (pe- asthma control variables compared with placebo. Mon- riod 2), indicating that tolerance did not develop. Toler- telukast improved airway obstruction, patient-reported ance can be a clinical problem with some therapies, end points, and asthma outcomes (protection against including receptor antagonists.15,16 After 12 weeks of treat- worsening asthma episodes), consistent with the goals ment, removal of montelukast did not cause rebound wors- of asthma therapy as outlined in the Global Initiative for ening of asthma in any end point. Rebound worsening on Asthma.2 treatment discontinuation has been experienced with re-

ARCH INTERN MED/ VOL 158, JUNE 8, 1998 1218

©1998 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ on 09/25/2021 The Montelukast Clinical Research Study Group

Leonard C. Altman, MD, Allergy Clinic, Pacific Medical Center, Seattle, Wash; George Bensch, MD, Stockton, Calif; William E. Berger, MD, Southern California Research Center, Mission Viejo, Calif; Jonathan A. Bernstein, MD, Bernstein Allergy Group, Inc, Cincinnati, Ohio; Kathryn Blake, PharmD, The Nemours Children’s Clinic, Jacksonville, Fla; Milan L. Brandon, MD, Cali- fornia Research Foundation, San Diego, Calif; Edwin Bronsky, MD, AAAA Medical Research Group, Salt Lake City, Utah; Chris- topher Brown, MD, California Pacific Medical Center, San Francisco, Calif; William Busse, MD, University of Wisconsin, Madi- son; Paul Chervinsky, MD, New England Research Center, Inc, Allergy & Asthma Center, North Dartmouth, Mass; John J. Condemi, MD, Allergy, Asthma, Immunology of Rochester, PC, Rochester, NY; David L. Daniel, MD, Wenatchee Valley Clinic, Wenatchee, Wash; Robert J. Dockhorn, MD, International Med Tech Consultants, Inc, Prairie Village, Kan; Thomas B. Ed- wards, MD, Allergy and Asthma Center, Albany Medical Center, Albany, NY; Albert F. Finn, MD, Allergy and Asthma Center of Charleston, Pa, N Charleston, SC; Stanley J. Galant, MD, Orange, Calif; Marc F. Goldstein, MD, The Asthma Center, Phila- delphia, Pa; Jay Grossman, MD, Allergy Care Consultants, Ltd, Tucson, Ariz; William G. Harris, MD, Magan Medical Clinic, Inc, Covina, Calif; Leslie Hendeles, PharmD, University of Florida, Health Science Center, Gainesville, Fla; Mani Kavuru, MD, Cleveland Clinic Foundation, Cleveland, Ohio; James P. Kemp, MD, Allergy and Asthma Medical Group and Research Center, San Diego; Philip E. Korenblat, MD, Barnes West County Hospital, The Asthma Center, St Louis, Mo; Michael Kramer, MD, Spokane Allergy and Asthma Clinic, Spokane, Wash; Craig LaForce, MD, North Carolina Clinical Research, Raleigh, NC; Tho- mas Littlejohn, MD, Piedmont Research Assoc, Winston-Salem, NC; Richard Lockey, MD, University of South Florida, Asthma, Allergy and Immunology, Clinical Research Unit, Tampa; Zev Munk, MD, Breco Research, Houston, Tex; Anjuli Seth Nayak, MD, Asthma and Allergy Associates, SC, Normal, Ill; Harold Nelson, MD, National Jewish Center for Immunology and Respi- ratory Medicine, Denver, Colo; Michael J. Noonan, MD, Allergy Assoc, PC Research Center, Portland, Ore; Gregory Owens, MD, University of Pittsburgh Medical Center, Presbyterian University Hospital, Pittsburgh, Pa; Stephen Park, MD, Daly City, Calif; David S. Pearlman, MD, Colorado Allergy and Asthma Clinic, PC, Aurora, Colo; Andrew Pedinoff, MD, Princeton Allergy & Associates, Princeton, NJ; Bruce Prenner, MD, Allergy Associates Medical Group, Inc, San Diego; Joan Reibman, MD, Bel- levue Hospital/General Clinical Research Center, New York, NY; Alan Segal, MD, Allergy Associates Research, Dallas, Tex; James M. Seltzer, MD, Clinical Research Institute, San Diego; Frank F. Snyder, MD, Lovelace Scientific Resources, Albuquerque, NM; William Storms, MD, Asthma and Allergy Associates, PC, Colorado Springs, Colo; Mary Strek, MD, University of Chicago, Chi- cago, Ill; William Stricker, MD, Clinical Research of the Ozarks, Inc, Columbia, Mo; Richard J. Sveum, MD, Park Nicollet Clinic, Health System Minnesota, Minneapolis; David G. Tinkelman, MD, Atlanta Allergy and Immunology, Research Foundation, Riverdale, Ga; Alan A. Wanderer, MD, Clinical Research Group of Colorado, Englewood; James R. Taylor, MD, Pulmonary Consultants, Tacoma, Wash; Stephan Weisberg, MD, Allergy and Asthma Specialists, Minneapolis; Richard White, MD, Uni- versity of California–Davis Medical Center, General Internal Medicine Investigative Clinic, Sacramento, Calif; and James D. Wolfe, MD, Allergy and Asthma Associates, San Jose, Calif.

ceptor antagonists,17 possibly because of target cell recep- use of an electronic, centralized spirometry control sys- tor up-regulation.18 Since receptor up-regulation is thought tem in a therapeutic asthma clinical trial. to occur within the first week of exposure,19 it is unlikely The diary card measures (daytime asthma symp- that longer treatment durations with montelukast would tom scores, nocturnal awakening, ␤-agonist use, and result in rebound worsening of asthma. PEFR) demonstrated a near-maximal effect of mon- Studies with zafirlukast, another leukotriene recep- telukast within the first day of treatment, indicating a rapid tor antagonist, and zileuton, a 5-lipoxygenase inhibitor, therapeutic benefit. Such a rapid onset has not been seen have demonstrated that these compounds also provide ben- with other leukotriene receptor antagonists or 5- efits in chronic asthma. Zafirlukast improved airway ob- lipoxygenase inhibitors used in the treatment of struction in a 6-week study,6 and zileuton improved air- asthma.7,21,22 Other controller agents for asthma, includ- way obstruction and patient-reported end points in a 12- ing cromolyn, nedocromil, and inhaled corticosteroids, week study.20 These trials showed large variability in the also require a longer treatment duration before their ef- treatment effects across end points,6,7,20 in contrast to the fects become apparent.23,24 consistent effect shown with montelukast in this trial. Significant improvements in all quality-of-life domains In this study, all end points were measured with high (symptoms, activity, environment, and emotions) occurred precision, leading to accurate and consistent treatment with montelukast treatment. Previous work in this area25 effect estimates. Spirometry data, transmitted via mo- suggests that the magnitude of the treatment-related im- dem, were collected and assessed centrally, with timely provements observed in this study were clinically mean- feedback given to study centers. We believe that the stan- ingful. The evaluation of quality of life is important because dardized, centralized spirometry quality control insti- it determines the impact of therapy on the patient’s daily tuted in this study was the reason not only for the accu- life that is not captured by other end points.14 racy of the spirometry measurements, but indirectly for Another important objective of chronic asthma the precision of all study end points. Further evidence therapy is the protection against episodes of worsening of the benefit of centralized quality control was shown asthma.26 Such episodes have been shown to contribute in the decreased variability (the root mean square error to morbidity and consume substantial asthma-related from the ANOVA model) of the data in this large clini- health resources. We found that montelukast protected cal trial compared with that of a smaller dose-ranging significantly against asthma worsening. A 31% decrease study.9 To our knowledge this is the first report of the in asthma exacerbation days and a 37% increase in asthma

ARCH INTERN MED/ VOL 158, JUNE 8, 1998 1219

©1998 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ on 09/25/2021 control days were observed. In addition, montelukast pro- Gertrude Noonan for their excellent coordination of the study. vided protection against episodes of worsening asthma Reprints: Theodore F. Reiss, MD, Merck Research Labo- (need for oral corticosteroid rescue treatment or discon- ratories, RY 33-648, PO Box 2000, Rahway, NJ 07065. tinuation from study therapy). These results confirmed findings from a previous montelukast trial9 and were con- sistent with the improvements in the primary end points REFERENCES of this study. The effect of montelukast was generally consistent 1. Weiss KB, Gergen PJ, Hodgson TA. An economic evaluation of asthma in the across patient prerandomization characteristics, includ- United States. N Engl J Med. 1992;326:862-866. ing demographic variables and baseline values for the end 2. Global Initiative for Asthma. A six-part asthma management program. In: Global Strategy for Asthma Management and Prevention. NHLBI/WHO Workshop Re- points (FEV1 and daytime symptom score), indicating that port. Bethesda, Md: National Heart, Lung, and Blood Institute, National Insti- there was a similar clinical response to montelukast across tutes of Health; 1995:70−117. Publication 95-3659. subgroups of the asthmatic population studied. Mon- 3. Cochrane GM. Compliance in asthma. Eur Respir Rev. 1995;28:164-165. 4. Griffin M, Weiss JW, Leitch AG, et al. Effects of leukotriene D4 on the airways in telukast provided clinical benefit in patients using con- asthma. N Engl J Med. 1983;308:436-439. comitant inhaled corticosteroids, thus confirming previ- 5. Laitinen LA, Laitinen A, Haahtela T, Vilkka V, Spur BW, Lee TH. Leukotriene E4 8,27,28 and granulocytic infiltration into asthmatic airways. Lancet. 1992;341:989-990. ous clinical trials with montelukast. It has been shown 6. Spector SL, Smith LJ, Glass M, Accolate Asthma Trialist Group. Effects of 6 weeks that oral corticosteroids do not inhibit the production of of therapy with oral doses of ICI 204,219, a leukotriene D4-receptor antagonist, in leukotrienes in the airways of asthmatic patients; this pro- subjects with bronchial asthma. Am J Respir Crit Care Med. 1994;150:618-623. 7. Israel E, Cohn J, Dube´ L, Drazen JM, for the Zileuton Clinical Trial Group. Effect vides the biological basis for the additive effects of leuko- of treatment with zileuton, a 5-lipoxygenase inhibitor, in patients with asthma. triene receptor antagonists and corticosteroids.29 JAMA. 1996;275:931-936. 8. Reiss TF, Altman LC, Chervinsky P, et al. Effects of montelukast (MK-0476), a It is currently believed that asthma is a syndrome new potent cysteinyl leukotriene (LTD4) receptor antagonist, in patients with chronic of airway inflammation, characterized in part by in- asthma. J Allergy Clin Immunol. 1996;98:528-534. creased numbers of blood eosinophils, which, with other 9. Noonan MJ, Chervinsky P, Brandon M, et al, for the Montelukast Asthma Study 30 Group. Montelukast, a potent cysteinyl leukotriene antagonist, causes dose re- inflammatory cells, infiltrate the airways. Leuko- lated improvements in chronic asthma. Eur Respir J. In press. trienes have been shown to enhance proliferation of bone 10. Jones TR, Labelle M, Belley M, et al. Pharmacology of montelukast sodium marrow eosinophil and basophil precursors,31 to attract (SINGULAIRா), a potent and selective leukotriene D4-receptor antagonist. Can J 5 Physiol Pharmacol. 1995;73:191-201. eosinophils into the lung, and to cause microvascular 11. De Lepeleire I, Reiss TF, Rochette F, et al. Montelukast causes prolonged, po- leakage.5 The decrease in blood eosinophil counts over tent, leukotriene D4-receptor antagonism in the airways of patients with asthma. 9,28 Clin Pharmacol Ther. 1995;61:83-92. time, consistent with previous montelukast studies and 12. Standardization of Spirometry, 1994 update: American Thoracic Society crite- similar to that seen with inhaled corticosteroids, sug- ria. Am J Respir Crit Care Med. 1995;152:1107-1136. gests that montelukast may have important effects on mea- 13. Santanello NC, Barber BL, Reiss TF, et al. Measurement characteristics of two asthma symptom diary scales for use in clinical trials. Eur Respir J. 1997;10:646-651. sures of asthmatic inflammation. A study with a 5- 14. Juniper EF, Guyatt GH, Epstein RS, et al. Evaluation of impairment of health- lipoxygenase inhibitor has shown similar results.7 These related quality of life in asthma. Thorax. 1992;47:76-83. observations suggest that the therapeutic effect of anti- 15. Raftery EB. Cardiovascular drug withdrawal syndromes. Drugs. 1984;28:371-374. 16. Adelroth E, Inman MD, Summers E, Pace D, Modi M, O’Byrne PM. Prolonged leukotriene compounds may, in part, be caused by ef- protection against exercise-induced bronchoconstriction by the leukotriene D4- fects on inflammatory measures. receptor antagonist cinalukast. J Allergy Clin Immunol. 1997;99:210-215. 17. Rangno RE, Langlois S. Comparison of withdrawal phenomena after proprano- In this study, montelukast was generally well tol- lol, metoprolol and pindolol. Br J Clin Pharmacol. 1982;13:345S-351S. erated. Clinical adverse events occurred with similar fre- 18. Nattel S, Rangno RE, Van-Loon G. Mechanism of propranolol withdrawal phe- quencies with montelukast and placebo treatments. Ad- nomena. Circulation. 1979;59:1158-1164. 19. Wolfe BB, Harden TK, Molinoff PB. In vitro study of beta-adrenergic receptors. verse events that occurred were generally transient and Ann Rev Pharmacol Toxicol. 1977;17:575-604. self-limited, and did not require discontinuation from 20. Israel E, Rubin P, Kemp JP, et al. The effect of inhibition of 5-lipoxygenase by study therapy. Laboratory adverse experiences were in- zileuton in mild to moderate asthma. Ann Intern Med. 1993;119:1059-1066. 21. Bateman ED, Aitchison JA, Summerton L, Harris A. The early onset of action of frequent, mild, transient, and similar in frequency in the zafirlukast (ACCOLATE™) in patients with asthma. Am J Respir Crit Care Med. montelukast and placebo treatment groups. There were 1997;155:A663. 22. Sahn S, Galant S, Murray J, et al. Pranlukast (ULTAIR™) improves FEV1 in no differences in the occurrence of serum aminotrans- patients with asthma. Am J Respir Crit Care Med. 1997;155:A665. ferase elevations between the montelukast and placebo 23. Petty TL, Rollins DR, Christopher K, Good JT, Oakley R. Cromolyn sodium is groups, as have been reported with the 5-lipoxygenase effective in adult chronic asthmatics. Am Rev Respir Dis. 1989;139:694-701. 7 24. Tinkelman DG, Reed CE, Nelson HS, Offord KP. Aerosol beclomethasone dipro- inhibitor zileuton. pionate compared with theophylline as primary treatment of chronic, mild to mod- In conclusion, montelukast sodium, given orally at erately severe asthma in children. Pediatrics. 1993;92:64-77. 10 mg once daily at bedtime during a 12-week treat- 25. Juniper JF, Guyatt GH, Willan A, Griffith LE. Determining a minimal important change in a disease-specific quality of life questionnaire. J Clin Epidemiol. 1994;47:81-87. ment period, provided significant clinical benefit to pa- 26. Barnes N. Efficacy and effectiveness in treatment of asthma. Eur Resp Rev. 1995; tients with chronic asthma. It was generally well toler- 30:284-287. 27. Reiss TF, Sorkness CA, Stricker W, et al. Effects of montelukast (MK-0476), a ated, with an adverse event profile comparable with that potent cysteinyl leukotriene receptor antagonist, on bronchodilation in asthmatic of placebo. subjects treated with and without inhaled corticosteroids. Thorax. 1997;52:45-48. 28. Altman LC, Munk Z, Seltzer J, et al, for the Montelukast Asthma Study Group. A placebo controlled, dose ranging study of montelukast, a cysteinyl leukotriene Accepted for publication November 4, 1997. receptor antagonist. J Allergy Clin Immunol. In press. This study was supported by a grant from Merck Re- 29. Dworski R, Fitzgerald GA, Oates JA, et al. Effect of oral prednisone on airway inflammatory mediators in atopic asthma. Am J Respir Crit Care Med. 1994;149: search Laboratories, Rahway, NJ. 953-959. We thank Kerstin Malmstrom, PhD, for help in pre- 30. Paul WE, Metcalfe DD, Busse W, Reece ER; Goldstein RA, moderator. Asthma. paring the manuscript; Barbara Knorr, MD, for critical re- Ann Intern Med. 1994;121:698-708. 31. Wickrmasinghe RG, Khan MA, Hoffbrand AV. Do leukotrienes play a role in the view of the manuscript; Elizabeth V. Hillyer, DVM, and Judy regulation of proliferation of normal and leukemic hemapoietic cells? Prosta- Evans for editorial assistance; and Jacquelyn McBurney and glandins Leukot Essent Fatty Acids. 1993;48:123-126.

ARCH INTERN MED/ VOL 158, JUNE 8, 1998 1220

©1998 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ on 09/25/2021