This was a phase IIa, randomized, double-blind, parallel-group, placebo-controlled study conducted at 70 investigative sites in 16 countries around the world (clinicaltrials.gov study number NCT00398645; GSK study number FFA106783). The study was conducted between November 2006 and August 2007 and comprised a 2-week pre-treatment screening period (Day -14 to Day 0) for evaluation of eligibility and asthma status, an 8-week double-blind treatment phase, and telephonic follow-up contact 1 week after completing the study medication. During the double-blind treatment phase, patients were required to attend 5 on-treatment morning clinic visits (weeks 1, 2, 4, 6, and 8) and 3 on-treatment evening clinic visits (weeks 2, 4, and 8). Patients were issued with electronic daily diaries (eDiary; Asthma Monitor plus [AM 2+], Jaeger, Hoechberg, Germany), which were used to enter information including morning and evening PEF (measured using the AM 2+ device), daytime and night-time asthma symptom score, daytime and night-time use of salbutamol rescue medication. These data were then used to establish eligibility during the screening period and to establish a baseline from which to determine symptomatic worsening of asthma during the double-blind treatment period. Patients were also asked to record in their eDiary their use of non-study issued maintenance ICS during the screening period and their use of blinded study medication to assess compliance with study medication. eDiary data for PEF, rescue medication and symptom score were not data based or analyzed and the information collected was used by the study investigator for safety purposes only. All patients were assessed and treated on an out-patient basis.
The study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines and was approved by local ethics committees and institutional review boards as appropriate. All patients provided written informed consent prior to participating in the study.
Male and female patients aged ≥ 12 years with a documented history of asthma as defined by the National Institute of Health [2, 23] were eligible for study entry. Other inclusion criteria were baseline morning forced expiratory volume in one second (FEV1) 50-80% of the predicted normal value, and reversibility of baseline FEV1 (≥ 12% and ≥ 200 ml) in response to inhaled salbutamol. Study participants had to be able to replace their current short-acting beta2 agonist (SABA) therapy with salbutamol inhalation aerosol during the screening period, and to be able to withhold all inhaled short-acting beta sympathomimetic bronchodilators for 6 hours before each study visit. They also had to have been taking ICS for ≥ 3 months before screening, with a stable daily dose for 4 weeks before screening. The maximum daily ICS dose was FP 200 mcg or equivalent.
Patients were excluded if they had a history of life-threatening asthma, a respiratory infection within 4 weeks of screening, an asthma exacerbation within 4 weeks of screening, or that required oral corticosteroids within 3 months or hospitalization within 6 months of screening, clinically significant uncontrolled disease, oropharyngeal candidiasis, or a recent (1 year) or heavy (> 10 pack years) smoking history. Female patients of childbearing potential who were not using an acceptable method of contraception and patients with severe milk protein allergy or an adverse drug reaction to any beta2 agonist, sympathomimetic drug or intranasal, inhaled, or systemic corticosteroid were also excluded.
After the 2-week screening period, patients were randomized if morning pre-dose FEV1 was 50-80% of the predicted normal value and within ± 15% of the pre-bronchodilator FEV1, and if they continued to have symptoms requiring salbutamol use or a 24-hour asthma symptom score of ≥ 1 on at least 4 of the last 7 days of screening. Patients were excluded if during screening they had changes to their asthma medication, a lower or upper respiratory tract infection, asthma exacerbation, oral candidiasis, or were non-compliant with the eDiary.
Patients who successfully completed the screening period were randomized to one of six treatments (ratio, 1:1:1:1:1:1) administered via a Diskus®/Accuhaler® for 8 weeks: FF 200 mcg or 400 mcg once daily in the morning, FF 200 mcg or 400 mcg once daily in the evening, FF 200 mcg twice daily, or placebo (twice daily). Patients had stopped their usual ICS therapy one day prior to randomization. Patients who were randomized to once-daily treatment received a matching placebo Diskus®/Accuhaler®. Patients were instructed to administer one inhalation from one inhaler in the morning and one inhalation from the other inhaler in the evening, approximately 12 hours apart. Use of the salbutamol Diskus®/Accuhaler® device or nebulized salbutamol (excluding for reversibility testing during screening) was not allowed during the study. The use of a salbutamol metered-dose inhaler was, however, permitted for symptom relief.
Patients, investigators and study personnel were all blinded to study treatment. The central randomization schedule was generated by the sponsor using a validated computerized system (RandAll). Patients were randomized using Registration and Medication Ordering System (RAMOS), an automated, interactive telephone based system, which was used by the investigator or designee to register and randomize the patient and receive medication assignment information.
Patients were observed by appropriately trained site personnel during each clinic visit to ensure that they were able to administer the study drug correctly. The eDiary was used to question patients on their compliance with study medication each morning and evening; patients who were not compliant were counselled on the appropriate way to administer the study drug.
The following anti-asthma medications were not allowed ≤ 2 weeks before screening or during the study: combination therapy comprising an inhaled beta2 agonist and ICS, slow-release bronchodilators, anticholinergics, long-acting beta2 agonists (LABA), ketotifen, nedocromil sodium, sodium cromoglycate, and oral LABA. Other drugs prohibited before screening included oral SABA (within 24 hours), anti-leukotrienes or potent CYP3A4 inhibitors (within 4 weeks), and systemic, oral, parenteral, or depot corticosteroids, or anti-IgE therapy (within 3 months). Immunotherapy was permitted if initiated before screening and used at a stable dose for the treatment of allergies. Drug therapies for other medical conditions, with the exception of systemic corticosteroids, were permitted throughout the study provided the dose remained constant and their use was not expected to affect the patient's lung function or asthma status.
The primary, single, efficacy endpoint was the mean change from baseline at week 8 in the pre-dose (morning or evening [depending on regimen], pre-rescue bronchodilator) FEV1 (measured using a MasterScope CT spirometer [Viasys, Hoechberg, Germany]). This was recorded electronically at the morning clinic visits between 6 am and 11 am or at the evening clinic visits between 6 pm and 11 pm using flow-volume curves generated from calibrated spirometers and according to ATS/ERS guidelines . Patients were required to withhold their salbutamol therapy for ≥ 6 hours before each clinic visit. Treatment compliance was derived from the number of positive answers in the patient's eDiary, which were divided by the number of non-missing answers and expressed as a percentage for both morning and evening doses.
The following safety endpoints were evaluated: incidence of adverse events (AEs) and serious AEs (SAEs), vital signs, hematology, clinical chemistry, and urinalysis parameters, oropharyngeal examinations, and withdrawals due to worsening asthma. AEs/SAEs were coded using the Medical Dictionary for Regulatory Activities. Twenty-four hour urinary-free cortisol excretion was also measured at baseline (week 0) and at the end of the double-blind treatment period (week 8) to assess hypothalamic-pituitary-adrenal axis (HPA) function. A central laboratory was used for all cortisol measurements.
Assuming a common standard deviation of 450 ml, a sample size of 648 patients (108 per group) was required to provide 90% power to detect a treatment difference of 200 ml in pre-dose FEV1 between FF and placebo at the (two-sided) 5% significance level. The study was not powered to formally assess differences between once-daily treatment and twice-daily treatment or differences between morning treatment and evening treatment, therefore statistical comparisons of all FF treatment groups were only against placebo. However it was pre-specified in the study protocol that provided the FF treatment groups demonstrated a statistically significant difference relative to placebo, the relative effects of once-daily and twice-daily dosing and of morning and evening dosing would be evaluated by assessing the degree of overlap between the 95% confidence intervals relating to the treatment differences with placebo. If the point estimate of the treatment/placebo difference for any given FF regimen lay within the 95% confidence interval for another FF regimen, the treatment effect estimates would be within 0.12 L of each other.
The intent-to-treat (ITT) population, which included all randomized patients who received at least one dose of study medication, was the primary population for all efficacy and safety (excluding urinary cortisol) analyses. The per protocol (PP) population (all subjects in the ITT population who did not have any full protocol deviations) was used for confirmatory analysis of the primary endpoint.
Analysis of the primary efficacy endpoint was conducted using an analysis of covariance (ANCOVA) model with effects due to baseline pre-dose FEV1, country, sex, age, and treatment group. Any patient with a missing FEV1 measurement at week 8 was included in the analysis of the primary endpoint by imputation using the preceding non-missing FEV1 value (last observation carried forward). The analysis was performed separately for morning and evening time-points, with the placebo and twice-daily regimens used in both cases. Estimated treatment differences for pair-wise comparisons against placebo were presented together with 95% confidence intervals for the difference and p values.
The patient population for urinary cortisol (UC) analyses comprised all patients whose urine samples were not considered to have confounding factors that would affect the interpretation of the results. The 24-hour UC excretion was log-transformed and analyzed using an ANCOVA model with effects due to baseline, country, sex, age, and treatment group.