Nathan et al.

BMC Pulmonary Medicine 2012, 12:67

[Link]

RESEARCH ARTICLE Open Access

Safety and efficacy of fluticasone/formoterol

combination therapy in adolescent and adult
patients with mild-to-moderate asthma:
a randomised controlled trial
Robert A Nathan1*, Anthony D’Urzo2, Viktor Blazhko3 and Kirsten Kaiser4

Abstract
Background: This study investigated the efficacy and safety of a new asthma therapy combining fluticasone
propionate and formoterol fumarate (fluticasone/formoterol; flutiformW), administered twice daily (b.i.d.) via a single
aerosol inhaler, compared with its individual components administered separately and placebo, in patients with
mild-to-moderate asthma.
Methods: Patients aged ≥12 years were evenly randomised to 12 weeks of treatment with fluticasone/formoterol
(100/10 μg b.i.d.), fluticasone (100 μg b.i.d.), formoterol (10 μg b.i.d.), or placebo, in this double-blind, parallel group,
multicentre study. The three co-primary endpoints were: a) change in forced expiratory volume in the first second
(FEV1) from morning pre-dose at baseline to pre-dose at week 12 for the comparison with formoterol; b) change in
FEV1 from morning pre-dose at baseline to 2 hours post-dose at week 12 for the comparison with fluticasone, and
c) time to discontinuation due to lack of efficacy from baseline to week 12 for the comparison with placebo. Safety
was assessed based on adverse events, clinical laboratory tests and vital sign evaluations.
Results: Statistically significant differences were demonstrated for all the three co-primary endpoints. Fluticasone/
formoterol combination therapy showed significantly greater improvements from baseline to end of study in the
change in pre-dose FEV1 compared with formoterol (Least Squares (LS) mean treatment difference: 0.101 L; 95%
Confidence Interval (CI): 0.002, 0.199; p = 0.045) and the change in pre-dose compared with 2 hours post-dose FEV1
versus fluticasone (LS mean treatment difference: 0.200 L; 95% CI: 0.109, 0.292; p < 0.001). The time to
discontinuation due to lack of efficacy was significantly longer for patients in the combination therapy group
compared with those receiving placebo (p = 0.015). Overall, the results from multiple secondary endpoints
assessing lung function, asthma symptoms, and rescue medication use supported the superior efficacy of the
combination product compared with fluticasone, formoterol, and placebo. The fluticasone/formoterol combination
therapy had a good safety and tolerability profile over the 12 week treatment period.
Conclusions: Fluticasone/formoterol had a good safety and tolerability profile and showed statistically superior
efficacy for the three co-primary endpoints compared to fluticasone, formoterol, and placebo, in adolescents and
adults with mild-to-moderate asthma.
EudraCT number: 2007-002866-36; US NCT number: NCT00393991

* Correspondence: drrnathan@[Link]
1
Asthma and Allergy Associates PC, 2709 North Tejon Street, Colorado
Springs, CO, USA
Full list of author information is available at the end of the article

© 2012 Nathan et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative
Commons Attribution License ([Link] which permits unrestricted use, distribution, and
reproduction in any medium, provided the original work is properly cited.
Nathan et al. BMC Pulmonary Medicine 2012, 12:67 Page 2 of 15
[Link]

Background Aurora, Ontario; Research Ethics Board MUHC-MGH

Asthma is a chronic inflammatory disorder of the air- Site, Montreal, Quebec; Europe: Ethics Commission of
ways. It is associated with variable airflow obstruction State Pharmacological Center of Health Ministry of
related to airway hyperresponsiveness and bronchocon- Ukraine, Kiev, Ukraine). Written informed consent was
striction. For persistent asthma, inhaled corticosteroids obtained from all patients (or the parents or guardians of
(ICSs) are recommended as one of the most effective patients under 18 years of age) before they were enrolled
treatments for airway inflammation. Nonetheless, for a into the study.
significant number of patients symptoms persist and The efficacy and safety of fluticasone/formoterol com-
additional therapy is required [1-5]. bination therapy 100/10 μg, administered twice daily
Landmark studies in adult and adolescents have (b.i.d.) (50/5 μg, 2 inhalations b.i.d.) via a single hydro-
demonstrated that patients using ICS and long-acting fluoroalkane (HFA) pressurised metered-dose inhaler
β2-agonist (LABA) combination therapy achieved better (pMDI), was compared with the individual components
asthma control compared to more than doubling the administered separately (fluticasone, 100 μg b.i.d. pMDI
dose of ICS or administration of ICS in combination [50 μg, 2 inhalations b.i.d.]; formoterol 10 μg b.i.d. pMDI
with other therapeutic agents [2,4-12], while further [5 μg, 2 inhalations b.i.d.]), and placebo (pMDI [2 inhala-
studies specifically in children and adolescents aged up tions, b.i.d.]).
to 16 years have reported at least similar efficacy
with ICS/LABA compared with doubling the ICS dose Patients
[13,14]. In addition, research has also shown that the Patients of both sexes, aged 12 years and over, with a
interactions between ICSs and LABAs potentiate each history of asthma of at least 12 months prior to screen-
other’s respective therapeutic effects at the molecular ing, as defined by the National Asthma Education and
level [15-18]. Prevention Program [31], were considered eligible for
The ICS, fluticasone propionate (fluticasone), has a study enrolment. Eligible patients had either a documen-
well-established safety and efficacy profile, and exerts a ted history of ICS use for at least 4 weeks before screen-
potent and sustained anti-inflammatory effect [19-22]. ing, at a daily dose of not more than 500 μg fluticasone
The LABA, formoterol fumarate (formoterol), has rapid, HFA pMDI (or equivalent), or were not on ICS therapy
dose-dependent bronchodilatory effects, with an onset of for at least 12 weeks prior to the screening visit. All
action of 1 to 3 minutes [23], similar to salbutamol and patients were required to have a Forced Expiratory Vol-
faster than that of salmeterol [24-27]. Extensive research ume in the first second (FEV1) between 60% and 85%
into the safety and efficacy of these two molecules is (inclusive) of predicted normal values at both screening
widely documented in the literature [19-30], and sug- and baseline visits. Patients also needed to demonstrate
gests that the fluticasone/formoterol combination may FEV1 reversibility, i.e. reversible bronchoconstriction for
provide clinicians with a new and efficacious treatment patients who did not have a history of documented re-
for the management of persistent asthma. versibility within the 12 months prior to the screening
The study presented here evaluated the efficacy and visit. These patients underwent a reversibility testing
safety of fluticasone and formoterol combination therapy procedure, after the pulmonary function tests at screen-
(fluticasone/formoterol; flutiformW), administered via a ing, defined as a ≥14.5 % increase 15–30 minutes follow-
single aerosol inhaler, in adolescent and adult patients ing albuterol/salbutamol aerosol inhalation (200 to 400
with mild-to-moderate asthma. μg, as appropriate, i.e. two inhalations of 100 μg, sepa-
rated by a period of 1 minute. If reversibility was not
Methods met, FEV1 was re-assessed within another 30 minutes,
Study design and if still not met, two further inhalations of albuterol/
This was a 12-week, randomised, double-blind, placebo- salbutamol were administered and reversibility was
and active-controlled, parallel-group study, conducted at re-assessed). All patients had to be able to demonstrate
59 centres in North America and Europe. The study was satisfactory aerosol technique and accurate use of the
conducted in accordance with ICH GCP and as per the telephone diary system.
ethical principles of the Declaration of Helsinki. The Insti- Patients were excluded from the study if they had a
tutional Review Boards or Independent Ethics Committees history of life-threatening asthma within the previous 12
at each participating centre reviewed and approved the months or during the run-in period. Patients with a his-
protocol (United States: Schulman Associates Institutional tory of systemic corticosteroid use within the previous 3
Review Board Inc., Cincinnati, Ohio; University of Florida months, omalizumab use within the previous 6 months,
Health Science Center IRB-01, Gainesville, Florida; Baylor or leukotriene antagonist use within the week before
Research Institute IRB, Dallas, Texas; Marywood Univer- screening, were also excluded. Other exclusion criteria
sity IRB, Scranton, Pennsylvania; Canada: IRB Services, included significant, non-reversible pulmonary disease,
Nathan et al. BMC Pulmonary Medicine 2012, 12:67 Page 3 of 15
[Link]

respiratory tract infections within the 4 weeks prior to thoroughly after dosing. All other asthma medications
screening visit or during the run-in period, significant were prohibited during the study, except for albuterol/
medical illness, a smoking history of at least 10 pack- salbutamol, the use of which was permitted, as needed,
years or current smoking history within the previous in case of worsening asthma symptoms. An Interactive
year, and hypersensitivity to study medication. Patients Voice Response System was used for patient enrolment,
were also excluded if they had received β-blockers, tri- treatment allocation, and generation of patient identifi-
cyclic antidepressants, monoamine oxidase inhibitors, cation number. The use of dummy placebo inhalers
quinidine-type antiarrhythmics, or drugs known to in- ensured that blinding was maintained throughout the
hibit CYP3A4, within the week prior to screening. How- study. The investigators, study site personnel, and repre-
ever, use of a LABA prior to screening was permitted. sentatives involved in monitoring, data management,
any other aspect of the study, including sponsor
Interventions personnel, were blinded throughout the study. Treat-
The run-in period was used to confirm that all patients ment assignment was strictly confidential and accessible
were symptomatic and to ensure that the baseline only to authorised persons until the time of unblinding.
assessments were standardised across all patients after Patient adherence to assigned study medication regi-
discontinuing their respective asthma medications. For men was assessed based on the data recorded via a tele-
patients who were ICS-requiring prior to screening, the phone diary system. Each patient recorded the number
run-in period lasted 14 ± 3 days during which time they of actuations of study and rescue medication they had
received fluticasone (pMDI; 50 μg b.i.d.) as maintenance used during both the run-in and the treatment periods.
therapy. For patients with no history of ICS use, the A safety follow-up was carried out two weeks after last
run-in lasted between 14 to 28 days and they received dose of study medication by telephone.
no maintenance therapy during this time. Rescue medi-
cation was available to all patients for deteriorating
asthma symptoms. During any 7 consecutive days of the Efficacy assessments
run-in, patients were required to use at least two inhala- The efficacy of fluticasone/formoterol combination ther-
tions per day of rescue albuterol/salbutamol medication apy in comparison with fluticasone, formoterol, and pla-
for at least 3 days and to have either 3 or more days with cebo was evaluated using three co-primary endpoints:
asthma symptoms or one night with sleep disturbance the mean change in FEV1 (as measured in the clinic)
due to asthma. At the baseline visit, which was defined from pre-dose at baseline to pre-dose at week 12 (a
as week 0 and followed the run-in period, patients comparison of fluticasone/formoterol versus formoterol
returned to the study site to complete the randomisation alone) was used to assess the contribution of the antiin-
procedures (assessment of pulmonary function and gen- flammatory component from the fluticasone/formoterol
eral asthma symptom-based endpoints) and to confirm combination; the mean change in FEV1 (in clinic) from
that randomisation criteria were met. pre-dose at baseline to 2 hours post-dose at week 12 (a
At the end of the run-in period, eligible patients were comparison of fluticasone/formoterol versus fluticasone
randomised equally into one of the following four alone) was used to assess the contribution of the bron-
blinded treatment arms using minimisation with biased chodilator component from the combination product,
coin assignment [32], stratified according to prior steroid and discontinuation due to lack of efficacy was used to
use, study site, and the subgroup of patients aged 12 to evaluate the efficacy of the fluticasone/formoterol com-
18 years. Patients were provided with two inhalers: one bination compared with placebo. Lack of efficacy was
for fluticasone/formoterol, formoterol or placebo (which defined by asthma exacerbations and loss of asthma con-
were identical in appearance), and one for fluticasone or trol (see below for definitions), and these two classifica-
a visually identical fluticasone placebo. Study medication tions were combined for the analysis. In order to
was administered twice daily for 12 weeks, taking two demonstrate superior efficacy, fluticasone/formoterol
actuations from each device twice daily (8 inhalations therapy had to achieve statistical significance over the
per day): fluticasone/formoterol 100/10 μg (50/5 μg, 2 relevant comparator treatments for each of the three co-
inhalations b.i.d.) and placebo b.i.d., fluticasone 100 μg primary endpoints.
(50 μg, 2 inhalations b.i.d.) and placebo b.i.d., formoterol Secondary efficacy endpoints comprised additional pul-
10 μg (5 μg, 2 inhalations b.i.d.) and placebo b.i.d., or monary function tests including FEV1 % predicted normal,
placebo (2 inhalations, 2 devices, b.i.d.) (Figure 1). All Forced Vital Capacity (FVC), frequency of asthma exacer-
study medications were administered via a pMDI with- bations, and data gathered from patients’ telephone diaries
out the use of a spacer. Patients were required to have a including morning and evening Peak Expiratory Flow Rate
1-minute interval between inhalations, always use the (PEFR), asthma symptom scores, sleep disturbance scores,
pMDIs in the same sequence and rinse their mouth and frequency of rescue medication use.
Nathan et al. BMC Pulmonary Medicine 2012, 12:67 Page 4 of 15
[Link]

Screening Run-In Period* Treatment Period*

Patients with a
Fluticasone/formoterol 100/10 µg b.i.d.
history of ICS use 14±3 days
prior to screening (50 µg fluticasone pMDI b.i.d.) Fluticasone 100 µg b.i.d.
Patients with no 14–28 days Formoterol 10 µg b.i.d.
history of ICS use
Placebo b.i.d.
prior to screening

Stratification by steroid use prior to screening

12 weeks

Figure 1 Study design. *Albuterol/salbutamol pro re nata (as needed) as rescue medication. b.i.d. = twice daily; ICS = inhaled corticosteroid;
pMDI = pressurised metered dose inhaler.

FEV1 was measured in the clinic at baseline and at more than 25% from baseline on more than 3 of the 7
weeks 2, 4, 8, and 12 by spirometry in accordance with days before a study visit, iv) excessive use (more than 12
the American Thoracic Society/European Respiratory actuations per day) of rescue medication on more than 3
Society Task Force guidelines [33]. Predicted FEV1 of the 7 days before a study visit, or v) nocturnal awa-
values were determined using the values of Polgar and kening due to asthma, that required rescue medication,
Promadhat [34] for patients aged 12–17 years, and those on more than 2 of 7 days before a study visit.
of Crapo et al. [35] for patients aged 18 years and older.
Spirometry values were also adjusted for race. PEFR was
Safety assessments
measured twice daily, pre-dose, by means of a Micro-
Safety assessments were carried out throughout the study
Peak peak flow meter (Micromedical, Chatham
based on adverse events reported, vital signs, a 12-lead
Maritime, Kent, UK), and patients recorded the results
electrocardiogram (ECG), and clinical laboratory testing.
using the telephone diary system.
The frequency and severity (defined as either ‘mild-to-
moderate’ or ‘severe’) of asthma exacerbations were Statistical analyses and sample size calculation
recorded throughout the study. Mild-to-moderate Efficacy analyses were performed on the full analysis set
exacerbations were defined as any of the following (FAS) (all patients who received at least one dose of
occurring for at least 2 consecutive days: i) pre-dose study medication, had a baseline FEV1 measurement and
PEFR measurements more than 30% below the values at least one post-baseline pre-dose and 2-hour post-dose
measured at baseline, or ii) awakening during the night FEV1 measurement), the per-protocol (PP) population
because of asthma, or iii) the use of additional rescue (all patients in the FAS who did not have a major proto-
medication of more than three inhalations per day com- col violation, which included patients who did not take
pared with baseline. Severe exacerbations were defined study medication on at least 50% of the days that the pa-
as the deterioration in asthma that required additional tient was in the study or if the patient did not return for
therapy (for example, systemic steroids), or an emer- two study visits in a row), and the safety population (all
gency visit or hospitalisation due to asthma. randomised patients who received at least one inhalation
Asthma symptoms, scored on a six-point scale ranging of study medication).
from 0 to 5 (0 = no symptoms; 5 = asthma so severe that The change in morning pre-dose FEV1 from baseline
the patient was unable to go to work or school or to to pre-dose at weeks 2, 4, 8, and 12, and change in
carry out normal daily activities), and sleep distur- morning pre-dose FEV1 from baseline to 2-hours post-
bances, scored on a five-point scale ranging from 0 to dose at weeks 2, 4, 8, and 12 were compared between
4 (0 = slept through the night, no asthma; 4 = could the treatment groups using analysis of covariance
not sleep at all because of asthma), were recorded via (ANCOVA), with treatment group, centre, and previous
the telephone diary system. steroid use as main effects, and baseline FEV1 as a con-
Patients were withdrawn from the study because of tinuous covariate. Missing data were replaced using the
lack of treatment efficacy (asthma exacerbations and loss last observation carried forward (LOCF) approach. To
of asthma control) if any of the following five criteria analyse the time to discontinuation due to lack of effi-
were met: i) a severe asthma exacerbation requiring cacy, a stratified log-rank test was performed adjusting
emergency treatment, hospitalisation, or use of any for treatment group and previous steroid use. In this su-
asthma medication not permitted in the study protocol, periority analysis (which used a two-sided t-test), super-
ii) a decrease in pre-dose FEV1 (as measured in the iority for each the three co-primary endpoints was
clinic) of more than 20% from baseline, iii) a decrease in confirmed if the lower limit of the 95% confidence inter-
morning pre-dose PEFR (from telephone diaries) of val (CI) for the between-treatment difference did not
Nathan et al. BMC Pulmonary Medicine 2012, 12:67 Page 5 of 15
[Link]

span 0, and hence p < 0.05. There was no requirement order, using the same Hochberg methodology as
for the CI to be above a pre-defined threshold. described above: percentage of symptom-free days
For the secondary endpoints, the differences between (defined as days with an asthma symptom score of zero),
groups for the change from baseline in PEFR, asthma percentage of rescue medication-free days (days with no
symptom scores, sleep disturbance scores, and rescue use of rescue medication), percentage of asthma control
medication use were analysed using the ANCOVA days (days with asthma symptom score of zero, sleep
model described above, with the relevant baseline value disturbance score of zero, and no use of rescue medica-
as the continuous covariate. Differences in the frequency tion), the proportion of patients with treatment-
of exacerbations between groups were tested by logistic emergent asthma exacerbations, sleep disturbance scores,
regression analysis with effects for treatment group and and the percentage of awakening-free nights (nights with
previous steroid use, and differences between groups for a sleep disturbance score of zero). Pre-specified sub-
the percentage of days with an asthma exacerbation and group analyses were performed for all three co-primary
percentage of asthma control days were assessed using endpoints based on prior ICS use, using an ANCOVA,
van Elteren’s method for combining Wilcoxon rank sum with age and study site as factors, and using the LOCF
test results from independent strata, with prior steroid approach on the FAS.
use as the stratum for the analysis. Safety analyses were performed for all randomised
Provided all three co-primary endpoints were statisti- patients who received at least one inhalation of study
cally significant, the secondary endpoints were then eval- medication (the safety population).
uated using a sequential gatekeeper approach [36] for For pre-dose or 2-hours post-dose FEV1 measures, a
the three treatment comparisons, according to the fol- sample size of 92 patients per treatment group in the
lowing order: i) fluticasone/formoterol combination study would have 85% power to detect a significant dif-
therapy versus placebo, and ii) fluticasone/formoterol ference between two treatment groups using a two-sided
combination therapy versus fluticasone alone and fluti- t-test with α=0.05, assuming a difference of 0.2 L with
casone/formoterol combination therapy versus formo- respect to mean change from morning pre-dose baseline
terol alone. to either morning pre-dose FEV1 at week 12 or 2-hour
The first four secondary endpoints were analysed post-dose FEV1 at week 12, and a common standard de-
firstly for combination therapy vs. placebo, in the follow- viation (SD) of 0.45. It was therefore planned to enrol
ing order, based on the mean change from baseline to 108 patients in each group to account for an approxi-
week 12: morning PEFR, evening PEFR, use of rescue mately 15% drop out rate. Assuming that 10% of flutica-
medication, and asthma symptom scores. Provided that sone/formoterol and 30% of placebo group patients
each of these analyses returned statistically significant would discontinue due to lack of efficacy, with 92
results for combination product vs. placebo (p < 0.05), patients per treatment group there would be 90% power
the subsequent analyses (combination therapy vs. flutica- to detect this difference using a two-sided log-rank test
sone alone and vs. formoterol alone) could then be with α = 0.05.
carried out in the same order. Statistical analyses were
two-sided and significance was measured at the 0.05α Results
level. If both tests were significant at the 0.05α level, the A total of 475 patients were randomised to treatment,
next endpoint could be evaluated for confirmatory statis- including 33 adolescents (6.9%). Of the 475 patients, 333
tical significance. If one of the two tests was not significant took part at sites based in the United States, 80 were
at the 0.05α level, for example the analysis of morning based in Canada, and 62 in the Ukraine, and, overall,
PEFR for combination therapy vs. fluticasone alone, the 367 (77.3%) patients completed the study (Figure 2).
other test (morning PEFR for combination product vs. Treatment groups were well-matched with regard to
formoterol alone) could be evaluated for statistical sig- demographics and baseline characteristics, with little dif-
nificance at the 0.025α level, however all formal testing ference between groups with respect to lung function re-
of the remaining secondary endpoints was suspended versibility. Prior to screening, a total of 29.4% of patients
(i.e. for both combination product vs. fluticasone and had received ICS monotherapy, and 20.0% had received
vs. formoterol). If the analyses were not statistically sig- combined ICS and LABA combination therapy (Table 1).
nificant for the combination product versus either com- The median FEV1 % predicted value at baseline ranged
parator then, once again, all remaining confirmatory from 72.0 to 75.0 (Table 1). Mean compliance rates ran-
sequential testing was formally suspended. If the sequen- ged from 84% to 85% across treatment arms. There were
tial gatekeeper approach for the three comparative tests 459 randomised patients in the FAS (115, 117, 116, and
was statistically significant for each of the four end- 111 in the combination, fluticasone, formoterol, and pla-
points, confirmatory sequential testing of the remaining cebo groups, respectively); 408 in the PP population
secondary endpoints was carried out in the following (103 in each of the combination and fluticasone groups,
Nathan et al. BMC Pulmonary Medicine 2012, 12:67 Page 6 of 15
[Link]

Randomised
n = 475

Fluticasone/ Fluticasone Formoterol Placebo b.i.d. Overall

formoterol 100 µg b.i.d. 10 µg b.i.d. N = 118 N = 475
100/10 µg b.i.d. N = 119 N = 120
N = 118

Discontinued Discontinued Discontinued Discontinued Discontinued

n = 19 (16.1%) n = 22 (18.5%) n = 30 (25%) n = 37 (31.4%) n = 108 (22.7%)
Adverse Event Adverse Event Adverse Event Adverse Event Adverse Event
n = 1 (0.8%) n = 1 (0.8%) n = 1 (0.8%) n = 3 (2.5%) n = 6 (1.3%)
Consent Consent Consent Consent Consent
Withdrawn Withdrawn Withdrawn Withdrawn Withdrawn
n = 4 (3.4%) n = 5 (4.2%) n = 4 (3.3%) n = 4 (3.4%) n = 17 (3.6%)
Lost to follow-up Lost to follow-up Lost to follow-up Lost to follow-up Lost to follow-up
n = 2 (1.7%) n = 3 (2.5%) n = 2 (1.7%) n = 2 (1.7%) n = 9 (1.9%)
Lack of efficacy Lack of efficacy Lack of efficacy Lack of efficacy Lack of efficacy
n = 7 (5.9%) n = 9 (7.6%) n = 13 (10.8%) n = 22 (18.6%) n = 51 (10.7%)
Other Other Other Other Other
n = 5 (4.2%) n = 4 (3.4%) n = 10 (8.3%) n = 6 (5.1%) n = 25 (5.3%)

Completed Completed Completed Completed Completed

n = 99 (83.9%) n = 97 (81.5%) n = 90 (75.0%) n = 81 (68.6%) n = 367 (77.3%)

Figure 2 Patient flow diagram.

and 101 in each of the formoterol and placebo groups); (LS mean difference = 0.200 L; 95% CI: 0.109, 0.292;
all 475 patients were in the safety population. p < 0.001) (Table 2). The improvements in pre-dose FEV1
(Figure 3A) and 2-hour post-dose FEV1 (Figure 3B) with
Primary efficacy endpoints fluticasone/formoterol were demonstrated throughout the
The three co-primary endpoints demonstrated superior entire treatment period as shown by pulmonary function
efficacy of fluticasone/formoterol combination therapy tests carried out at Weeks 2, 4, 8, and 12. Secondary ana-
compared to fluticasone, formoterol, and placebo, lyses also showed that fluticasone/formoterol provided sig-
respectively (Table 2). nificantly greater improvements than fluticasone alone in
The fluticasone/formoterol combination showed clin- FEV1 from pre-dose at baseline to pre-dose at week 12, and
ically relevant improvements in FEV1 from pre-dose at numerically greater improvements in FEV1 from pre-dose
baseline to pre-dose and 2-hour post-dose at week 12 at baseline to 2 hours post-dose at week 12 compared with
(Table 2). Furthermore, the contribution of the flutica- formoterol alone (Table 2).
sone component in the combination product, as ana- Fluticasone/formoterol combination therapy was also
lysed by the mean change in FEV1 from pre-dose at shown to be superior to placebo with respect to the time
baseline to pre-dose at week 12, demonstrated statisti- to discontinuation due to lack of efficacy (due to either
cally significant improvements for patients in the com- asthma exacerbation or to loss of asthma control) (log-
bination therapy treatment arm compared with those rank p = 0.015) (Table 2). Furthermore, fewer patients
administered formoterol alone (LS mean difference = 0.101 discontinued due to lack of efficacy in the combination
L; 95% CI: 0.002, 0.199; p = 0.045). Similarly, the therapy group (6.1%) compared with those in the flutica-
contribution of the formoterol component of the sone group (7.7%), formoterol group (11.2%) or the pla-
combination product, as analysed by the mean change in cebo group (16.2%) (Table 2).
FEV1 from pre-dose at baseline to 2 hours post-dose at
week 12, demonstrated statistically significant improve- Secondary efficacy endpoints
ments for patients in the combination therapy treatment The secondary efficacy endpoints evaluated lung func-
arm compared with those administered fluticasone alone tion, disease control and asthma symptoms. Overall, all
Nathan et al. BMC Pulmonary Medicine 2012, 12:67 Page 7 of 15
[Link]

Table 1 Patient baseline demographic and asthma characteristics, Full Analysis Set
Characteristic Treatment group
Fluticasone/formoterol Fluticasone Formoterol Placebo Overall
100/10 μg b.i.d. 100 μg b.i.d. 10 μg b.i.d. b.i.d.
N = 115 N = 117 N = 116 N = 111 N = 459
Gender, n (%)
Female 72 (62.6) 71 (60.7) 63 (54.3) 70 (63.1) 276 (60.1)
Male 43 (37.4) 46 (39.3) 53 (45.7) 41 (36.9) 183 (39.9)
Ethnic origin, n (%)
White/Caucasian 89 (77.4) 88 (75.2) 83 (71.6) 79 (71.2) 339 (73.9)
Black 13 (11.3) 16 (13.7) 21 (18.1) 22 (19.8) 72 (15.7)
Asian 6 (5.2) 4 (3.4) 5 (4.3) 4 (3.6) 19 (4.1)
Hispanic 6 (5.2) 8 (6.8) 6 (5.2) 5 (4.5) 25 (5.4)
Other 1 (0.9) 1 (0.9) 1 (0.9) 1 (0.9) 4 (0.9)
Age, years
Mean (SD) 39.8 (14.54) 38.3 (14.45) 39.1 (15.26) 38.1 (13.67) 38.8 (14.47)
Age categories, n (%)
12 to 17 years 7 (6.1) 9 (7.7) 9 (7.8) 6 (5.4) 31 (6.8)
≥ 18 years 108 (93.9) 108 (92.3) 107 (92.2) 105 (94.6) 428 (93.2)
Steroid use, n (%)
Freea 59 (51.3) 60 (51.3) 58 (50.0) 55 (49.5) 232 (50.5)
b
Requiring 56 (48.7) 57 (48.7) 58 (50.0) 56 (50.5) 227 (49.5)
Prior ICS and ICS/LABA use, n (%)
ICS only 31 (27.0) 39 (33.3) 30 (25.9) 35 (31.5) 135 (29.4)
ICS and LABA 25 (21.7) 18 (15.4) 28 (24.1) 21 (18.9) 92 (20.0)
c
Duration of asthma, years
Mean (SD) 18.9 (13.40) 20.6 (13.84) 20.3 (14.48) 21.4 (12.83) 20.3 (13.64)
FEV1 % predictedd at baselinee
Mean (SD) 73.2 (7.54) 73.5 (8.14) 73.2 (7.79) 72.0 (7.97) 73.0 (7.86)
Median 72.0 75.0 73.0 72.0 73.0
FEV1 at baselinee, L
Mean (SD) 2.416 (0.5790) 2.425 (0.6625) 2.459 (0.6231) 2.352 (0.6114) 2.414 (0.6192)
Median 2.370 2.330 2.375 2.250 2.340
Reversibility at screening, %
n = 114 n = 117 n = 116 n = 111 n = 458
Mean (SD) 23.2 (10.1) 22.8 (9.0) 21.8 (8.4) 22.8 (8.3) 22.6 (9.0)
Median 19.3 19.5 18.7 20.0 19.2
N = total number of patients; n = number of patients in specified category; SD = standard deviation; b.i.d. = twice daily; ICS = inhaled corticosteroids; LABA = long
acting beta agonist; FEV1 = Forced Expiratory Volume in the first second.
a. Patient with no history of steroid use for at least 12 weeks prior to the screening visit.
b. Patient who used an inhaled steroid regimen at a dose not greater than 500 μg/day fluticasone (or equivalent steroid) for at least 4 weeks prior to screening.
c. Duration of asthma calculated as (Date of screening visit from Demographics CRF - Asthma diagnosis date)/ 365.25 and rounded to 1 decimal place.
d. Based on standardised predicted FEV1 values.
e. Baseline was the last available value prior to dosing at the baseline/week 0 visit.

of these evaluations supported the superior efficacy of The mean increase in morning and evening PEFR
fluticasone/formoterol combination therapy compared values from baseline to week 12 was statistically signifi-
with the individual components and placebo. The com- cantly greater (p < 0.01) for patients on the combination
bination product demonstrated numerically greater product compared with those administered fluticasone,
improvements for a number of the secondary endpoint formoterol or placebo (Figure 4).
evaluations versus all three of the comparators, with Disease control, as evaluated by asthma control days,
many endpoints meeting the criteria for statistical sig- rescue medication-free days, symptom-free days, and
nificance as per the sequential gatekeeping approach. awakening-free nights (Table 3), demonstrated numerically
Nathan et al. BMC Pulmonary Medicine 2012, 12:67 Page 8 of 15
[Link]

Table 2 Mean change in FEV1 (L) from pre-dose at baseline to pre-dose and 2-hour post-dose at week 12, (LOCF), time
to discontinuation due to lack of efficacy and study duration for each treatment group, Full Analysis Set
Treatment group
Fluticasone/formoterol Fluticasone Formoterol Placebo
100/10 μg b.i.d. 100 μg b.i.d. 10 μg b.i.d. b.i.d.
N = 115 N = 117 N = 116 N = 111
Baseline FEV1 (L)
Mean (SD) 2.416 (0.5790) 2.425 (0.6625) 2.459 (0.6231) 2.352 (0.6114)
Change in FEV1 from pre-dose at baseline to pre-dose at week 12
LS Mean (SE) 0.195 (0.038) 0.092 (0.037) 0.094 (0.038) 0.047 (0.037)
Difference from fluticasone/formoterol 100/10 μg b.i.d.: contribution from fluticasone component
LS Mean (SE) 0.103 (0.050) 0.101 (0.050) 0.147 (0.051)
95% CI 0.005, 0.201 0.002, 0.199 0.048, 0.247
p-value 0.040 0.045 0.004
Change in FEV1 from pre-dose at baseline to 2 hours post-dose at week 12
LS Mean (SE) 0.392 (0.035) 0.191 (0.034) 0.330 (0.035) 0.124 (0.035)
Difference from fluticasone/formoterol 100/10 μg b.i.d.: contribution from formoterol component
LS Mean (SE) 0.200 (0.047) 0.062 (0.047) 0.267 (0.047)
95% CI 0.109, 0.292 −0.030, 0.153 0.175, 0.360
p-value < 0.001 0.187 < 0.001
Discontinuation due to lack of efficacy
Number, % 7 (6.1) 9 (7.7) 13 (11.2) 18 (16.2)
Time to discontinuation, weeksa
Mean 6.9 4.7 4.6 5.5
p-value b 0.015
b.i.d. = twice daily; N = total number of patients; SD = standard deviation; LS = least squares; SE = standard error; NA = not applicable; CI = confidence interval;
LOCF = last observation carried forward; FEV1 = Forced Expiratory Volume in the first second.
a. Time to discontinuation due to lack of efficacy calculated as (Date of early discontinuation - Date of first study drug administration+1)/7 and rounded to 1
decimal place (based on all patients who discontinued).
b. p-value based on stratified log-rank test adjusting for prior steroid use for fluticasone/formoterol 100/10 μg b.i.d. versus placebo.

greater improvements for fluticasone/formoterol compared Similarly, a larger mean increase in rescue medication-
to all the comparator treatments. However, significant in- free days was observed in the combination therapy arm
ferential statistical testing was only exploratory based on than in any of the comparator groups. In the flutica-
the sequential gatekeeping approach. Patients adminis- sone/formoterol group, a greater than three-fold in-
tered fluticasone/formoterol 100/10 μg b.i.d. demon- crease was seen in the number of rescue medication-free
strated a five-fold increase in the percent of asthma days from baseline (21.8%) to week 12 (77.7%), corre-
control days from baseline (12.8%) to week 12 (69.1%), sponding to an improvement from 1.5 days/week to 5.4
corresponding to 0.9 days per week at the start of the days/week. Overall, the mean increase in percent rescue
study compared to 4.8 days by the end of treatment. The medication-free days was 55.9% in the combination ther-
mean increase in percent of asthma control days was apy group compared to 43.3%, 41.9%, and 39.4% for the
56.3% for the combination product, 44.0% for the flutica- fluticasone, formoterol, and placebo groups, respectively
sone, 41.9% for the formoterol, and 36.0% for the placebo (Table 3).
groups. The mean percentage of symptom-free days at week
Overall, a lower percentage of patients on combination 12 (77.4%, corresponding to 5.4 days/week) in the com-
therapy experienced any asthma exacerbation (20.0%) bination therapy group was 2.5-fold than seen at base-
compared to those administered the monotherapies (23.9% line (28.0%, corresponding to 2.0 days/week). The mean
on fluticasone; 28.4% on formoterol) or placebo (32.4%), increase in symptom-free days in this group was 49.4%,
although the differences did not reach statistical signifi- compared with 37.3%, 38.0% and 35.6% in the flutica-
cance. For patients in the fluticasone/formoterol group, sone, formoterol and placebo groups, respectively.
2.6% experienced a severe exacerbation, compared to 3.4% This trend was also seen in the number of awakening-
on fluticasone, 6.9% on formoterol, and 9.0% of patients on free nights for patients in the fluticasone/formoterol
placebo (p = 0.048 for placebo vs fluticasone/formoterol). group. The mean percentage of awakening-free nights
Nathan et al. BMC Pulmonary Medicine 2012, 12:67 Page 9 of 15
[Link]

a Fluticasone/formoterol 100/10 µg
0.30 Fluticasone 100 µg
Formoterol 10 µg
Mean change in FEV1(L) 0.25 Placebo

0.20

0.15

0.10 * *
*
* *
0.05 *
* *
0.00
0 2 4 6 8 10 12
Week

b Fluticasone/formoterol 100/10 µg
0.6 Fluticasone 100 µg
Formoterol 10 µg
Mean change in FEV1(L)

0.5 Placebo

0.4
*
0.3
* * * *
0.2
* * *
*
0.1 *

0
0 2 4 6 8 10 12
Week
Figure 3 A – Mean change in FEV1 (L): mean change from baseline to pre-dose at weeks 2, 4, 8, and 12, Full Analysis Set (LOCF). * P-
value ≤ 0.05 versus fluticasone/formoterol 100/10 μg b.i.d. combination therapy treatment group. Baseline means were 2.416 L, 2.425 L, 2.459 L,
and 2.352 L for the fluticasone/formoterol, fluticasone, formoterol, and placebo treatment groups, respectively, for all patients in the Full Analysis
Set. b.i.d. = twice daily; FEV1 = forced expiratory volume in the first second; LOCF = last observation carried forward. Figure 3B – Mean change
in FEV1 (L): mean change from baseline to 2 hours post-dose at weeks 2, 4, 8, and 12, Full Analysis Set (LOCF). * P-value ≤ 0.05 versus
fluticasone/formoterol 100/10 μg b.i.d. combination therapy treatment group. Baseline means were 2.416 L, 2.425 L, 2.459 L, and 2.352 L for the
fluticasone/formoterol, fluticasone, formoterol, and placebo treatment groups, respectively, for all patients in the Full Analysis Set. b.i.d. = twice
daily; FEV1 = forced expiratory volume in the first second; LOCF = last observation carried forward.

was approximately 1.5-fold greater by week 12 (87.9%) Sub-group analyses

compared to baseline (59.1%), which corresponds to 4.1 Pre-specified subgroup analyses were performed based
awakening-free nights at baseline and 6.2 awakening- on ICS use prior to study entry, although it should be
free nights at the end of study. The mean increase acknowledged that this study was not powered to assess
was 28.8% for patients in the combination therapy these endpoints statistically. For patients with no history
group compared to 25.4%, 19.6%, and 20.9% for of ICS use prior to study enrolment, the difference be-
patients in the fluticasone, formoterol, and placebo tween the mean change in pre-dose FEV1 from baseline
groups, respectively. to week 12 between the combination therapy and formo-
Asthma symptoms, as evaluated by rescue medication terol groups was not statistically significant (LS mean
use, asthma symptom scores, and sleep disturbance treatment difference: 0.094 L; 95% CI: -0.050, 0.237;
scores, recorded daily by the patients, demonstrated nu- p = 0.200). However, the mean increase in FEV1 from
merically greater improvement for those administered pre-dose at baseline to 2 hours post-dose at week 12 was
fluticasone/formoterol compared to the individual com- statistically significant in the combination therapy group
ponents and placebo. Rescue medication use was statisti- compared with fluticasone alone (LS mean treatment differ-
cally significant for the combination versus all three ence: 0.212 L; 95% CI: 0.070, 0.354; p = 0.004).
comparator groups, as evaluated using the sequential For patients with a prior history of ICS use, the com-
gatekeeping approach (Table 4). bination product was statistically significantly superior
Nathan et al. BMC Pulmonary Medicine 2012, 12:67 Page 10 of 15
[Link]

Fluticasone/formoterol 100/10 µg b.i.d. (n = 115)

Fluticasone 100 µg b.i.d. (n = 117)
50 Formoterol 10 µg b.i.d. (n = 116)
Placebo (n = 111)
45

from baseline to final week (L/min)

40

Change in PEFR 35
30
† †
25 *
†
20
15 †
†

10
5
0
Mean baseline PEFR 364.6 371.6 371.0 357.9 372.0 382.3 380.9 367.6
(L/min)
Morning Evening

Figure 4 Morning and evening PEFR (L/min): mean change from baseline to week 12, Full Analysis Set. * P-value < 0.01 versus
fluticasone/formoterol 100/10 μg b.i.d. combination therapy treatment group. † P-value < 0.001 versus fluticasone/formoterol 100/10 μg b.i.d.
combination therapy treatment group. b.i.d. = twice daily; PEFR = peak expiratory flow rate; SE = standard error. Changes from baseline are
shown as least-squares mean ± SE for the full analysis set.

to each of the comparators for the mean change from patient receiving fluticasone/formoterol therapy which
pre-dose FEV1 at baseline to both pre-dose at week was not considered by the Investigator to be treatment-
12 (LS mean treatment difference combination prod- related.
uct compared with formoterol alone: 0.139 L; 95% CI: The most common adverse event leading to premature
0.000, 0.277; p = 0.050) and 2 hours post-dose at discontinuation of treatment in any group was asthma
week 12 (LS mean treatment difference combination (fluticasone/formoterol combination therapy group,
product vs. fluticasone alone: 0.172 L; 95% CI: 0.054, 2.5%; fluticasone group, 3.4%; formoterol group, 6.7%,
0.291; p = 0.005). placebo group, 11.9%). The most frequently reported ad-
With respect to discontinuations due to lack of treat- verse events occurring in more than 2% of patients in
ment efficacy, no statistically significant treatment group any treatment group are summarised in Table 5. There
difference was identified among patients with no history were no incidences of oropharyngeal candidiasis or dys-
of prior steroid use (log-rank p = 0.795); 4 patients phonia in any of the treatment groups. In addition, there
(7.3%) from the combination therapy group and 3 (5.8%) were no clinically relevant changes or group differences
from the placebo group discontinued prematurely. For for laboratory values (including glucose and potassium),
patients with a history of ICS use, fluticasone/formoterol vital signs, or ECG parameters.
combination was demonstrated to be statistically signifi-
cantly superior to placebo (p = 0.002, log-rank test); 3 Discussion
patients (5.7%) receiving fluticasone/formoterol discon- The study presented here evaluated the efficacy and
tinued early compared to 15 patients (36.6%) adminis- safety of fluticasone/formoterol 100/10 μg b.i.d. combin-
tered placebo. ation therapy compared to the individual components
administered separately and placebo over a 12-week
Safety and tolerability treatment period. The patients who took part in the
Combination therapy with fluticasone/formoterol was well study were adolescents and adults with mild-to-
tolerated. Adverse events were reported by 38 (32.2%) moderate asthma who were either already on ICS medi-
patients in the fluticasone/formoterol group, 47 (39.5%) cation (either with or without a LABA) or who were
patients in the fluticasone group, 44 (36.7%) patients in ICS-free prior to screening.
the formoterol group, and 46 patients (39.0%) in the The three co-primary endpoints all demonstrated that
placebo group (Table 5). No deaths or asthma exacerba- the fluticasone/formoterol combination product was su-
tions requiring hospitalisation were reported. Most ad- perior in efficacy compared to each of the comparators.
verse events were mild or moderate in severity. Only The first two co-primary efficacy endpoints evaluated
one serious adverse event occurred during the study, a lung function, based on FEV1 measurements, and com-
case of right-sided renal colic in a 70-year-old male pared the combination product with fluticasone alone
Nathan et al. BMC Pulmonary Medicine 2012, 12:67 Page 11 of 15
[Link]

Table 3 Asthma control days (%), rescue medication-free days (%), symptom-free days (%), and awakening-free nights
(%): mean change from baseline to week 12, Full Analysis Set
Characteristic Treatment group
Fluticasone/formoterol Fluticasone Formoterol Placebo
100/10 μg b.i.d. 100 μg b.i.d. 10 μg b.i.d. b.i.d
N = 115 N = 117 N = 116 N = 111
Asthma control days (%) n = 109 n = 114 n = 112 n = 105
Baselinea, mean (SD) 12.8 (20.14) 14.3 (22.62) 11.5 (19.21) 10.0 (18.11)
Week 12, mean (SD) 69.1 (37.69) 58.3 (42.02) 53.4 (40.11) 46.0 (41.22)
Change to week 12
Mean (SD) 56.3 (39.11) 44.0 (39.49) 41.9 (42.13) 36.0 (39.27)
Difference from fluticasone/formoterol 100/10 μg b.i.d.b
p-value* 0.017† 0.117 0.012†
Rescue medication-free days (%) n = 112 n = 116 n = 115 n = 109
Baselinea, mean (SD) 21.8 (24.38) 21.4 (25.55) 19.5 (24.51) 17.2 (20.14)
Week 12, mean (SD) 77.7 (32.12) 64.8 (39.49) 61.4 (37.19) 56.6 (39.95)
Change to week 12
Mean (SD) 55.9 (36.43) 43.3 (37.69) 41.9 (39.49) 39.4 (38.69)
Difference from fluticasone/formoterol 100/10 μg b.i.d.b
p-value* 0.020† 0.125 0.012†
Symptom-free days (%) n=110 n=114 n=115 n=108
Baselinea, mean (SD) 28.0 (26.72) 28.6 (29.97) 22.6 (27.01) 22.7 (27.75)
Week 12, mean (SD) 77.4 (35.21) 65.9 (38.59) 60.5 (38.67) 58.3 (38.99)
Change to week 12
Mean (SD) 49.4 (38.17) 37.3 (39.79) 38.0 (42.64) 35.6 (42.18)
Difference from fluticasone/formoterol 100/10 μg b.i.d.b
p-value* 0.027† 0.195 0.151
Awakening-free nights (%) n = 112 n = 116 n = 115 n = 108
Baselinea, mean (SD) 59.1 (33.79) 62.1 (33.73) 62.9 (34.51) 56.3 (37.40)
Week 12, mean (SD) 87.9 (26.73) 87.5 (26.77) 82.6 (31.57) 77.2 (34.17)
Change to week 12
Mean (SD) 28.8 (33.91) 25.4 (35.92) 19.6 (35.68) 20.9 (41.05)
Difference from fluticasone/formoterol 100/10 μg b.i.d.b
p-value* 0.790 0.055 0.052
b.i.d. = twice daily; N = total number of patients; n = number of patients in treatment group; FAS = full analysis set; SD = standard deviation.
a. Baseline was the 7-day average calculated on the last 7 days prior to the first dose of study drug.
b. Analysis method was Cochran-Mantel-Haenszel using van Elteren’s method for combining Wilcoxon rank sum test results from independent strata, with
previous steroid use and site as the strata for the analysis.
*All p values were exploratory.
† p ≤ 0.050 versus fluticasone/formoterol 100/10 μg b.i.d. combination therapy but not statistically significant as per the sequential gatekeeping approach.

and with formoterol alone, respectively. The improve- 12 weeks of treatment in this study. This improvement
ments in FEV1 from pre-dose at baseline to pre-dose was seen throughout the study period, as shown by pul-
and 2 hours post-dose at week 12 were clinically relevant monary function tests at weeks 2, 4, 8, and 12, and were
for patients receiving fluticasone/formoterol. Moreover, supported by the secondary efficacy endpoints evaluating
improvements seen in the combination therapy group lung function, for example the morning and evening
were numerically and statistically significantly greater for PEFR measurements.
the combination product compared with the monother- The third co-primary endpoint, evaluated for the flu-
apies administered alone. Tachyphylaxis has been ticasone/formoterol versus placebo treatment arms,
reported with formoterol monotherapy [37] and should showed that the combination product was statistically
be considered when interpreting the results. Nonethe- significantly superior to placebo with respect to the
less, fluticasone/formoterol combination therapy pro- time to discontinuation due to lack of efficacy. Fewer
vided improvements from baseline in lung function over patients in the combination therapy group prematurely
Nathan et al. BMC Pulmonary Medicine 2012, 12:67 Page 12 of 15
[Link]

Table 4 Use of rescue medication (number of inhalations/day), asthma symptom scores, and sleep disturbance scores:
mean change from baseline to week 12, Full Analysis Set
Characteristic Treatment group
Fluticasone/formoterol Fluticasone Formoterol Placebo
100/10 μg b.i.d. 100 μg b.i.d. 10 μg b.i.d. b.i.d.
N = 115 N = 117 N = 116 N = 111
Rescue medication use (inhalations/day) n = 112 n = 116 n = 115 n = 109
Baselinea, mean (SD) 2.8 (2.05) 3.0 (2.24) 3.0 (2.07) 3.0 (1.68)
Change to week 12
Mean (SE)b −2.22 (0.165) −1.64 (0.160) −1.62 (0.163) −1.16 (0.162)
Difference from fluticasone/formoterol 100/10 μg b.i.d.b
LS Mean (SE) −0.58 (0.217) −0.60 (0.218) −1.06 (0.222)
95% CI −1.01, -0.15 −1.03, -0.17 −1.50, -0.63
p-value* 0.008** 0.006** <0.001**
Asthma symptom scores n = 110 n = 114 n = 115 n = 108
Baselinea, mean (SD) 1.0 (0.60) 1.0 (0.64) 1.1 (0.61) 1.1 (0.62)
Change to week 12
LS Mean (SE)b −0.72 (0.060) −0.59 (0.058) −0.54 (0.059) −0.51 (0.059)
Difference from fluticasone/formoterol 100/10 μg b.i.d.b
LS Mean (SE) −0.13 (0.079) −0.18 (0.079) −0.21 (0.081)
95% CI −0.29, 0.03 −0.33, -0.02 −0.37, -0.05
p-value* 0.100 0.027† 0.011**
Sleep disturbance scores n = 112 n = 116 n = 115 n = 108
Baselinea, mean (SD) 0.5 (0.54) 0.5 (0.45) 0.4 (0.49) 0.6 (0.57)
Change to week 12
LS Mean (SE)b −0.36 (0.033) −0.34 (0.032) −0.28 (0.033) −0.27 (0.033)
Difference from fluticasone/formoterol 100/10 μg b.i.d.b
LS Mean (SE) −0.02 (0.043) −0.09 (0.044) −0.10 (0.044)
95% CI −0.11, 0.07 −0.17, 0.00 −0.18, -0.01
p-value* 0.632 0.053 0.031†
b.i.d. = twice daily; N = number of patients in treatment group; n = number of patients with data available; CI = confidence interval; FAS = full analysis set; LS =
least squares; SD = standard deviation; SE = standard error.
a. Baseline was the 7-day average calculated in the last 7 days prior to the first dose of study drug.
b. LS mean, SE, CI and p-value are from ANCOVA with factors for treatment group, site, and prior steroid use, with baseline value as a continuous covariate.
* All p-values were exploratory.
** p ≤ 0.050 versus fluticasone/formoterol 100/10 μg b.i.d. combination therapy and statistically significant as per the sequential gatekeeping approach.
† p ≤ 0.050 versus fluticasone/formoterol 100/10 μg b.i.d. combination therapy but not statistically significant as per the sequential gatekeeping approach.

left the study because of lack of treatment efficacy were seen for patients administered the combination
compared to those in any of the other three treatment product compared to those receiving either of the indi-
groups. vidual treatments or placebo. Patients treated with the
Probably the most important and clinically relevant combination product also reported fewer asthma exacer-
endpoint for patients is disease control. This was evalu- bations throughout the study compared to each of the
ated in this study by analysing the number of asthma other treatment arms.
control days, asthma exacerbations, rescue medication- A potential criticism of this study could be the recruit-
free days, symptom-free days, and awakening-free nights. ment of patients who were not on ICS monotherapy at
Although not a validated endpoint, the definition of baseline, perhaps suggesting the potential for over-
asthma control used in this study was robust and highly treatment of patients with milder asthma. This would not
relevant (recorded as days with no asthma symptoms, be consistent with GINA guidelines, which suggest a step-
no sleep disturbance due to asthma, and no use of res- wise treatment approach whereby patients with persistent
cue medication). The use of the gatekeeping approach asthma should be initiated on a low to medium dose of
meant that these secondary endpoints could not be sub- inhaled corticosteroids prior to treatment escalation in the
jected to confirmatory statistical testing. However, the event of a suboptimal response. Nonetheless, the median
greatest improvements in these symptomatic endpoints FEV1 as a percentage of predicted value at baseline was
Nathan et al. BMC Pulmonary Medicine 2012, 12:67 Page 13 of 15
[Link]

Table 5 Overview of Adverse Events, Safety Population

Treatment group
Fluticasone/formoterol Fluticasone Formoterol Placebo
100/10 μg b.i.d. 100 μg b.i.d. 10 μg b.i.d. b.i.d.
N = 118 N = 119 N = 120 N = 118
Any AE, n (%) 38 (32.2) 47 (39.5) 44 (36.7) 46 (39.0)
Any serious AE, n (%) 1 (0.8) 0 (0.0) 0 (0.0) 0 (0.0)
Any severe AE, n (%) 6 (5.1) 6 (5.0) 11 (9.2) 16 (13.6)
Any AE leading to study discontinuationa, n (%) 4 (3.4) 5 (4.2) 9 (7.5) 17 (14.4)
Any AE with probably or possible relationship to study drug, n (%) 5 (4.2) 9 (7.6) 10 (8.3) 12 (10.2)
Any AE leading to death, n (%) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0)
Treatment-emergent AEs reported for >2% of patients in any treatment group, n (%)
Infections and infestations 20 (16.9) 27 (22.7) 14 (11.7) 15 (12.7)
Upper respiratory tract infection 7 (5.9) 6 (5.0) 3 (2.5) 4 (3.4)
Nasopharyngitis 3 (2.5) 9 (7.6) 3 (2.5) 3 (2.5)
Urinary tract infection 3 (2.5) 3 (2.5) 1 (0.8) 1 (0.8)
Respiratory, thoracic and mediastinal disorders 8 (6.8) 8 (6.7) 14 (11.7) 17 (14.4)
Asthmab 3 (2.5) 4 (3.4) 9 (7.5) 14 (11.9)
Cough 4 (3.4) 2 (1.7) 1 (0.8) 1 (0.8)
Nervous system disorders 5 (4.2) 10 (8.4) 6 (5.0) 11 (9.3)
Headache 2 (1.7) 6 (5.0) 3 (2.5) 9 (7.6)
Gastrointestinal disorders 3 (2.5) 5 (4.2) 3 (2.5) 5 (4.2)
Diarrhoea 2 (1.7) 0 (0.0) 1 (0.8) 3 (2.5)
AEs = adverse events; N = total number of patients; n = number of patients in specified category; b.i.d. = twice daily.
Adverse events were coded using MedDRA™ version 9.0. At each level of summation (preferred term, system organ class and overall), each patient was counted
only once. Percentages were based on the number of patients in the population for each treatment group.
a. Six of the 35 patients (1 fluticasone/formoterol, 1 fluticasone, 1 formoterol, and 3 placebo) had treatment-emergent adverse events reported as the primary
reason for early discontinuation from the study. The remaining 29 patients (3 fluticasone/formoterol, 4 fluticasone, 8 formoterol, and 14 placebo) had lack of
efficacy reported as the primary reason for early discontinuation from the study. A serious adverse event was one which resulted in death, was life threatening,
resulted in persistent or significant disability/incapacity, inpatient hospitalisation or prolongation of existing hospitalisation, congenital anomaly or birth defect, or
an important medical event.
b. An exacerbation of asthma was considered as an adverse event if it did not resolve with the study drug, including rescue albuterol/salbutamol, and additional
medication was required (e.g., systemic glucocorticosteroids).

relatively low (72–75% across the treatment groups), sug- study, and the adverse event profile of the combination
gesting that a notable proportion of patients had moder- therapy was similar to that of its individual components
ately severe asthma and significant pulmonary impairment. administered separately and placebo.
Furthermore, whilst the authors would not advocate a This study provides strong evidence of the benefits of a
change to established treatment paradigm, two observa- new combination of fluticasone/formoterol, administered
tions are nonetheless evident from this study. Firstly, via a single aerosol inhaler, in adolescent and adult
pre-specified subgroup analysis of the populations of patients with mild-to-moderate asthma.
patients with no history of ICS use at baseline and those
who were on an ICS at baseline demonstrate that both
subgroups showed improvements for the two co- Conclusions
primary lung function endpoints over 12 weeks of treat- The data presented here are consistent with those
ment. Secondly, the treatment benefits for ICS-requiring observed in previous studies exploring the benefits of
patients were similar to or better than those for ICS-naïve this ICS/LABA combination therapy [27,28]. The co-
patients on all three co-primary endpoints (versus the administration of formoterol and fluticasone shows
relevant comparator). These data may reflect the clinical superiority to the individual components administered
benefit that patients could receive when stepping-up treat- separately and placebo for all three co-primary end-
ment to combination therapy. points and confers significant benefits in terms of lung
Fluticasone/formoterol combination therapy demon- function, disease control, and asthma symptoms. This
strated a good safety profile and was well-tolerated dur- study therefore demonstrates that fluticasone/formoterol
ing the 12-week treatment period. There were no deaths combination therapy is an efficacious and well tolerated
or asthma exacerbations requiring hospitalisation in this treatment for asthma.
Nathan et al. BMC Pulmonary Medicine 2012, 12:67 Page 14 of 15
[Link]

Competing interests 9. Woolcock A, Lundback B, Ringdal N, Jacques LA: Comparison of addition of

RAN has received grant/research support from Abbott, Alcon, Amgen, salmeterol to inhaled steroids with doubling of the dose of inhaled steroids.
AstraZeneca, Berringer-Ingleheim, Ception, Dey, Dyax, Genentech, Am J Respir Crit Care Med 1996, 153:1481–1488.
GlaxoSmithKline, MAP, MedImmune, Novartis, Sanofi-Aventis, Sepracor, Shire, 10. O’Byrne PM, Barnes PJ, Rodriguez-Roisin R, Runnerstrom E, Sandstrom T,
and TEVA; has been a consultant/scientific advisor for Genentech, Svensson K, Tattersfield A: Low dose inhaled budesonide and formoterol in
GlaxoSmithKline, Merck, Novartis, ADVISOR and TEVA, and participated in a mild persistent asthma. Am J Respir Crit Care Med 2001, 164:1392–1397.
speaker’s bureau for AstraZeneca, Genentech, GlaxoSmithKline, MERCK, 11. Thomas M, Kay S, Pike J, Williams A, Rosenzweig JR, Hillyer EV, Price D: The
Novartis, Sanofi-Aventis, and UCB. VB had no proprietary interest in the asthma control test (ACT) as a predictor of GINA guideline-defined
tested product which was subject of the study and did not receive any asthma control: analysis of a multinational cross-sectional survey. Prim
payment from the sponsor (excluding the costs of conducting the study). Care Respir J 2009, 18(1):41–49.
ADU has received research, consulting and lecturing fees from 12. Frois C, Wu EQ, Ray S, Colice GL: Inhaled corticosteroids or long-acting
GlaxoSmithkline, Sepracor, Schering Plough, Altana, Methapharma, Roche, beta-agonists alone or in fixed-dose combinations in asthma treatment:
AstraZeneca, Nycomed, ONO pharma, Novartis, and KOS Pharmaceuticals. KK a systematic review of fluticasone/budesonide and formoterol/
was an employee of SkyePharma, which sponsored this study. salmeterol. Clin Ther 2009, 31(12):2779–2803.
13. Vaessen-Verberne AA, van den Berg NJ, van Nierop JC, Brackel HJ, Gerrits
Authors' contributions GP, Hop WC, Duiverman EJ, Group CS: Combination therapy salmeterol/
RAN was the Principal Investigator for the study and participated in study fluticasone versus doubling dose of fluticasone in children with asthma.
design and conduct, ADU and VB were Investigators contributing to the Am J Respir Crit Care Med 2010, 182:1221–1227.
conduct of the study, and KK participated in the study design and 14. Gappa M, Zachgo W, von Berg A, Kamin W, Stern-Sträter C, Steinkamp G,
coordination for the Sponsor. All of the authors revised the manuscript Group VS: Add-on salmeterol compared to double dose fluticasone in
critically for important intellectual content and read and approved the final pediatric asthma: a double-blind, randomized trial (VIAPAED). Pediatr
manuscript. Pulmonol 2009, 44:1132–1142.
15. Barnes PJ: Scientific rationale for inhaled combination therapy with
Trade marks long-acting beta2-agonists and corticosteroids. Eur Respir J 2002,
W 19(1):182–191.
FLUTIFORM is a registered trade mark of Jagotec AG and is used under licence.
16. Barnes PJ: Scientific rationale for using a single inhaler for asthma
Acknowledgements control. Eur Respir J 2007, 29(3):587–595.
The authors would like to thank all of the study Investigators and 17. Roth M, Zhong J, S’ng CT, Tamm M: Effect of fluticasone and
participants, the members of the study team at SkyePharma, and colleagues formoterol combination therapy on airway remodeling. Eur Resp J.
at Mundipharma Research Ltd. Writing assistance was provided by Dr Evelin 2011, 38(Suppl 55):307s.
Kozma, Mundipharma Research Ltd, a member of the European Medical 18. Roth M, Zhong J, S’ng CT, Tamm M: The long-acting β2-agonist
Writers’ Association (EMWA), and by Dr Mark Rolfe (Oxford PharmaGenesis formoterol re-establishes the anti-proliferative effect of glucocorticoids
Ltd). This work was prepared in accordance with the Good Publication in asthmatic airway smooth muscle cells (ASMC). Eur Resp J. 2011,
Practice (GPP2) guidelines. The study was sponsored by SkyePharma, 38(Suppl 55):612s.
Switzerland. 19. Kavuru M, Melamed J, Gross G, Laforce C, House K, Prillaman B, Baitinger L,
Woodring A, Shah T: Salmeterol and fluticasone propionate combined in
Author details a new powder inhalation device for the treatment of asthma: a
1
Asthma and Allergy Associates PC, 2709 North Tejon Street, Colorado randomized, double-blind, placebo-controlled trial. J Allergy Clin Immunol
Springs, CO, USA. 2Department of Family and Community Medicine (DFCM), 2000, 105(6 Pt 1):1108–1116.
University of Toronto, Ontario, Canada. 3Kharkiv City Clinical Hospital #13, 20. Ketchell RI, Jensen MW, Spina D, O’Connor BJ: Dose-related effects of
Kharkiv, Ukraine. 4SkyePharma, Muttenz, Switzerland. formoterol on airway responsiveness to adenosine 50-monophosphate
and histamine. Eur Respir J 2002, 19(4):611–616.
Received: 10 October 2011 Accepted: 11 October 2012 21. Luijk B, Kempsford RD, Wright AM, Zanen P, Lammers JW: Duration of
Published: 18 October 2012 effect of single-dose inhaled fluticasone propionate on AMP-induced
bronchoconstriction. Eur Respir J 2004, 23(4):559–564.
References 22. Adams NP, Bestall JC, Lasserson TJ, Jones P, Cates CJ: Fluticasone versus
1. Van Noord JA, Schreurs AJM, Mol SJM, Mulder PGH: Addition of salmeterol placebo for chronic asthma in adults and children. Cochrane Database
versus doubling the dose of fluticasone propionate in patients with mild Syst Rev 2008, (4):CD003135.
to moderate asthma. Thorax 1999, 54:207–212. 23. AstraZeneca UK Ltd: Symbicort turbohaler 100/6, inhalation powder. Summary
2. Pauwels RA, Lofdahl CL, Postma DS, Tattersfield AE, O’Byrne P, Barnes PJ, of product characteristics. 2010. [[Link] Accessed
Ullman A: Effect of inhaled formoterol and budesonide on exacerbations January 2011.
of asthma. N Eng J Med 1997, 337:1405–1411. 24. Politiek MJ, Boorsma M, Aalbers R: Comparison of formoterol, salbutamol
3. Ward C, Pais M, Bish R, Reid D, Feltis B, Johns D, Walters EH: Airway and salmeterol in methacholine-induced severe bronchoconstriction. Eur
inflammation, basement membrane thickening and bronchial Respir J 1999, 13(5):988–992.
hyperresponsiveness in asthma. Thorax 2002, 57:309–316. 25. Berger WE: The use of inhaled formoterol in the treatment of asthma.
4. GINA report, global strategy for asthma management and prevention. [http:// Ann Allergy Asthma Immunol 2006, 97(1):24–33.
[Link]/] 2009 Accessed January 2011. 26. Palmqvist M, Arvidsson P, Beckman O, Peterson S, Lotvall J: Onset of
5. Naedele-Risha R, Dorinsky P, Craig TJ: Dual components of optimal asthma bronchodilation of budesonide/formoterol vs. salmeterol/fluticasone in
therapy: scientific and clinical rationale for the use of long-acting single inhalers. Pulm Pharmacol Ther 2001, 14(1):29–34.
b-agonists with inhaled corticosteroids. J Am Osteopath Assoc 2001, 27. Lotvall J, Ankerst J: Long duration of airway but not systemic effects of
1001:526–533. inhaled formoterol in asthmatic patients. Respir Med 2008,
6. Ducharme FM, Lasserson TJ, Cates CJ: Long-acting beta2-agonists versus 102(3):449–456.
anti-leukotrienes as add-on therapy to inhaled corticosteroids for 28. Bodzenta-Lukaszyk A, Dymek A, McAulay K, Mansikka H: Fluticasone/
chronic asthma. Cochrane Database Syst Rev 2006, 4:CD003137. formoterol combination therapy is as effective as fluticasone/salmeterol
7. Ducharme FM, Ni Chroinin M, Greenstone I, Lasserson TJ: Addition of long- in the treatment of asthma, but has a more rapid onset of action: an
acting beta2-agonists to inhaled corticosteroids versus same dose open-label, randomized study. BMC Pulm Med 2011, 11:28.
inhaled corticosteroids for chronic asthma in adults and children. 29. Bodzenta-Lukaszyk A, Pulka G, Dymek A, Bumbacea D, McIver T, Schwab B,
Cochrane Database Syst Rev 2010, 5:CD005535. Mansikka H: Efficacy and safety of fluticasone and formoterol in a single
8. Greening AP, Ind PW, Northfield M, Shaw G: Added salmeterol versus pressurized metered dose inhaler. Respir Med 2011, 105(5):674–682.
higher-dose corticosteroid in asthma patients with symptoms on 30. Bousquet J, Winchester C, Papi A, Virchow JC, Haughney J, Costa D, Usmani
existing inhaled corticosteroid. Lancet 1994, 344:219–224. O, Bjermer L, Price D, Global A, Asthma European N: Inhaled corticosteroid/
Nathan et al. BMC Pulmonary Medicine 2012, 12:67 Page 15 of 15
[Link]

long-acting beta(2)-agonist combination therapy for asthma: attitudes of

specialists in Europe. Int Arch Allergy Immunol 2012, 157(3):303–310.
31. Guidelines for the Diagnosis and Management of Asthma: Expert Panel
Report 2: National Asthma Education and Prevention Program; National
Institute of Health; National Heart, Lung, and Blood Institute. 1997, publication
No.97-4051. Bethesda, USA: NHLBI; 1997.
32. Pocock SJ, Simon R: Sequential treatment assignment with balancing for
prognostic factors in the controlled clinical trial. Biometrics 1975,
31(1):103–115.
33. Miller MR, Hankinson J, Brusasco V, Burgos F, Casaburi R, Coates A, Crapo R,
Enright P, van der Grinten CP, Gustafsson P, Jensen R, Johnson DC,
MacIntyre N, McKay R, Navajas D, Pedersen OF, Pellegrino R, Viegi G,
Wanger J, Force AET: Standardisation of spirometry. Eur Respir J 2005,
26(2):319–338.
34. Polgar G, Promdhat V: Pulmonary function testing in children: techniques and
standards. Philadelphia: W. B. Sauncers Co; 1970.
35. Crapo RO, Morris AH, Gardner RM: Reference spirometric values using
techniques and equipment that meet ATS recommendations. Am Rev
Respir Dis 1981, 123(6):659–664.
36. Hochberg Y: A sharper Bonferroni procedure for multiple tests of
significance. Biometrika 1988, 75:800–802.
37. García R, Guerra P, Feo F, Galindo PA, Gómez E, Borja J, Fernandez-Pacheco
R: Tachyphylaxis following regular use of formoterol in exercise-induced
bronchospasm. J Investig Allergol Clin Immunol. 2001, 11(3):176–82.

doi:10.1186/1471-2466-12-67
Cite this article as: Nathan et al.: Safety and efficacy of fluticasone/
formoterol combination therapy in adolescent and adult patients with
mild-to-moderate asthma: a randomised controlled trial. BMC Pulmonary
Medicine 2012 12:67.

Submit your next manuscript to BioMed Central

and take full advantage of:

• Convenient online submission

• Thorough peer review
• No space constraints or color figure charges
• Immediate publication on acceptance
• Inclusion in PubMed, CAS, Scopus and Google Scholar
• Research which is freely available for redistribution

Submit your manuscript at

[Link]/submit