Respiratory Medicine (2015) 109, 170e179

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.elsevier .com/locate /rmed

Therapeutic equivalence of budesonide/formoterol delivered via breath-actuatedinhaler vs pMDI

Kevin R. Murphy a,*, Rajiv Dhand b, Frank Trudo c,Tom Uryniak c, Ajay Aggarwal c, Göran Eckerwall d

a Allergy, Asthma, and Pulmonary Research, Boys Town National Research Hospital, Department ofPediatrics, University of Nebraska Medical Center, Creighton University School of Medicine, Omaha, NE68132, United Statesb Department of Medicine, University of Tennessee Graduate School of Medicine, 1924 Alcoa Highway,U114, Knoxville, TN 37920, United Statesc AstraZeneca LP, 1800 Concord Pike, Wilmington, DE 19850, United Statesd AstraZeneca, Pepparedsleden 1, Mölndal, Sweden

Received 18 August 2014; accepted 23 December 2014Available online 3 January 2015

KEYWORDSBreath-actuated;Asthma;Budesonide;Formoterol;Pressurized metered-dose;Inhaler

* Corresponding author.E-mail address: Kevin.Murphy@boy

http://dx.doi.org/10.1016/j.rmed.200954-6111/ª 2015 The Authors. Pubcreativecommons.org/licenses/by-nc-

Summary

Rationale: To assess equivalence of twice daily (bid) budesonide/formoterol (BUD/FM)160/4.5 mg via breath-actuated metered-dose inhaler (BAI) versus pressurized metered-doseinhaler (pMDI).Methods: This 12-week, double-blind, multicenter, parallel-group study, randomized adoles-cents and adults (aged �12 years) with asthma (and �3 months daily use of inhaled corticoste-roids) to BUD/FM BAI 2 � 160/4.5 mg bid, BUD/FM pMDI 2 � 160/4.5 mg bid, or BUD pMDI2 � 160 mg bid. Inclusion required prebronchodilator forced expiratory volume in one second(FEV1) �45 to �85% predicted, and reversibility of �12% in FEV1 (ages 12 to

Budesonide/formoterol via breath-actuated inhaler 171

Conclusions: BUD/FM 2 � 160/4.5 mg bid BAI is therapeutically equivalent to BUD/FM conven-tional pMDI. The introduction of BUD/FM BAI would expand options for delivering inhaledcorticosteroid/long-acting b2-agonist combination therapy to patients with moderate-to-severe asthma.

ClinicalTrials.gov NCT01360021.ª 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Introduction

Inhaled corticosteroids (ICSs) are recommended as first-line treatment for patients with persistent asthma, andthe addition of an inhaled long-acting b2-agonist (LABA)can be considered to improve lung function and symptomsin patients whose asthma is not well controlled on ICSalone [1,2]. Combination ICS/LABAs are available asbreath-actuated dry powder inhalers (DPI) and manually-actuated pressurized metered-dose inhalers (pMDIs).However, to date, a breath-actuated option in a pMDIdevice has not been available for the delivery of a com-bination of ICS/LABA. For some patients, poor coordina-tion between actuation and inhalation with use of a pMDIcan lead to dosing errors [3]. A breath-actuated pMDIinhaler (BAI) could provide an alternative option for thosepatients unable to overcome poor hand-lung coordination[4]. This study was designed to compare the therapeuticefficacy and safety of combined budesonide/formoteroldelivered by a BAI currently in clinical development withthose of combined budesonide/formoterol delivered by apMDI in patients with moderate-to-severe, symptomaticasthma.

Methods

Patients

Inclusion and exclusion criteria were designed to selectasthmatic patients aged �12 years who required medium-to high-dose ICSs and demonstrated reversibility of airflowobstruction with an inhaled bronchodilator after a run-inperiod. Patients must have used ICSs daily for �3 monthsbefore study entry and required consistent use of stabledaily doses of medium- to high-dose ICSs in the 30 daysbefore study entry. Minimum daily doses permitted were:beclomethasone dipropionate 504 mg/d (chlorofluorocarbonpMDI) or 160 mg/d (actuation counter pMDI), budesonide400 mg/d, flunisolide 1000 mg/d, fluticasone 264 mg/d (chlorofluorocarbon pMDI) or 300 mg/d (Diskus), triam-cinolone acetonide 800 mg/d, mometasone 400 mg/d, orciclesonide 160 mg/d.

Patients had to have asthma symptom scores (nighttimeor daytime) of >0 on �3 of the last 7 days of the run-inperiod and prebronchodilator FEV1 �45% and �85% of pre-dicted normal at baseline. Asthma symptom score isdefined as: 0, no asthma symptoms; 1, you are aware ofyour asthma symptoms but you can easily tolerate thesymptoms; 2, your asthma is causing you enough discomfort

to cause problems with normal activities (or with sleep); or3, you are unable to do your normal activities (or to sleep)because of your asthma.

To ensure reversibility with b2 agonists, patients had tomeet baseline postbronchodilator criteria (4e6 actuationsof albuterol pMDI [90 mg/inhalation] or salbutamol pMDI[100 mg/inhalation] or after inhalation of 2.5 mg nebulizedalbuterol) of a change of �12% in FEV1 from baseline forpatients aged �12 and

Figure 1 Flow chart of study design. BAI, breath-actuated metered-dose inhaler; bid, twice daily; BUD, budesonide; FM, for-moterol; pMDI, pressurized metered-dose inhaler; TC, telephone conversation.

172 K.R. Murphy et al.

The reversibility requirement was designed to ensure pa-tients are capable of responding to beta agonists andtherefore have a potential to show a difference in theprimary objective.

Patients were randomized at visit 4 and attended theclinic on 3 further occasions (week 3, 7, and 12), followedby a telephone call follow-up 2 weeks later. Lung functiontesting, eDiary collection, and adverse event (AE) checkoccurred at each visit, and device functionality questioningoccurred at weeks 3, 7, and 12. AEs were the only datacollected on the telephone call. At visit 4, patients wererandomized to receive 1 of the 3 following double-blindtreatments shown in Fig. 1: BUD/FM BAI 160/4.5 mg (Fig. 2),2 inhalations bid plus placebo pMDI, 2 inhalations bid; BUD/FM pMDI 160/4.5 mg, 2 inhalations bid plus placebo BAI, 2inhalations bid; or BUD pMDI 160 mg, 2 inhalations bid plusplacebo BAI, 2 inhalations bid. At visit 2 and visit 4, patientswere instructed on proper inhaler use, and devices wereavailable at each study center for training purposes and forpatients to practice. Instruction and practice also occurredprior to dispensing blinded study drug at visit 4. Confirma-tion of proper device use was accomplished via patientresponse in their eDiary to device functionality questions.Additionally, for further confirmation at return visits, theinvestigator or staff asked the patient if device instructionswere being followed and provided additional training ifrequired.

Figure 2 Schematic representation of the breath-actuatedinhaler device. Trigger flow rate is less than 28 L/min. Dataon file [9].

Concomitant medications

Patients were allowed the use of a short-acting b2-agonist(SABA) as rescue medication throughout the study includingenrollment, run-in, and treatment periods; albuterol pMDI(90 mg/inhalation)/salbutamol pMDI (100 mg/inhalation)were provided by the study site as rescue medication.Medications not allowed during the study included paren-teral, oral, or rectal glucocorticosteroids; leukotriene

antagonists; inhaled disodium cromoglycate; inhalednedocromil sodium or 5-lipoxygenase inhibitors; methyl-xanthines; inhaled anticholinergics; any monoclonal orpolyclonal antibody therapy taken for any reason; orCYP3A4 inhibitors (eg, ketoconazole).

Use of the following medications was allowed during thestudy: mucolytics and expectorants not containing bron-chodilators; antihistamines (other than terfenadine, aste-mizole, mizolastine); allergen-specific immunotherapy ifthe patient had been on a maintenance regimen for �3months before visit 1 and remained on a maintenanceregimen during the study; topical, nasal, and/or ocularformulations of glucocorticosteroids; topical, nasal and/orocular disodium cromoglycate and/or nedocromil sodium.Other medication, which was considered necessary for thepatient’s safety and well-being, could be administered atthe discretion of the investigator(s).

Budesonide/formoterol via breath-actuated inhaler 173

Study objectives and evaluations

The primary objective of this study was to evaluate ther-apeutic equivalence of BUD/FM delivered by BAI with BUD/FM delivered by pMDI. The primary efficacy end points forassessing therapeutic equivalence were comparison of theeffects of BUD/FM BAI with BUD/FM pMDI on FEV1 at60 min postdose and FEV1 predose. To ensure that any po-tential differences in efficacy could be detected, it wasprespecified that BUD/FM pMDI was superior to BUD pMDI on60-min postdose FEV1. Subsequent assessment of thera-peutic equivalence would be established if the 95% confi-dence interval (CI) limits for the ratio of treatment effectsof BUD/FM delivered by BAI versus pMDI was containedwithin the equivalence limits of 80e125% [5].

Secondary efficacy end points collected by means of anelectronic diary (eDiary) included mean change frombaseline for the overall treatment period for morning andevening peak expiratory flow (PEF), daytime and nighttimeand total asthma symptom scores, awakening-free nights(nights without awakening due to asthma symptoms), day-time and nighttime rescue medication use, and symptom-free days.

A secondary objective was to assess patient-reportedfunctionality of the 2 devices. Patients used the eDiary tocomplete an end-of-study questionnaire through which thepatient-reported ease of use of each device was evaluated.This questionnaire contained 2 questions (‘How easy was itto use the inhaler?’ and ‘How easy is it to determine whenyou will run out of medicine [from the inhaler]?’). Eachquestion was scored on a 7-point scale, with options rangingfrom ‘Extremely easy’ to ‘Extremely difficult’: (0,extremely easy; 1, very easy; 2, somewhat easy; 3, neithereasy nor difficult; 4, somewhat difficult; 5, very difficult; 6,extremely difficult).

Safety evaluations

The safety profile of BUD/FM delivered by pMDI or BAI wasassessed by comparing the nature, intensity, and severity ofAEs occurring in each treatment group.

Figure 3 Patient disposition. BAI, breath-actuated metered-dospMDI, pressurized metered-dose inhaler.

Statistical analyses

To assess the therapeutic equivalence of newly developedBUD/FM BAI device with the marketed BUD/FM pMDI de-vice, a step-down procedure was used to address multi-plicity. First, superiority needed to be demonstrated forBUD/FM pMDI versus BUD pMDI for the difference in post-dose FEV1 with a statistical significance level of 5%. If thisrequirement was met, then the 95% CI for the ratio oftreatment effects was to be used to assess therapeuticequivalence of BUD/FM pMDI versus BUD/FM BAI. The pri-mary variable for this comparison was the mean of thepostdose FEV1 measurements obtained during treatmentperiod (visits 4e7) expressed as ratio of the predose FEV1 atrandomization (visit 4) to create the treatment:baselineratio. The logarithm of this value was used for analysis usingthe analysis of covariance (ANCOVA) model with the fixedfactors treatment and country, and the logarithm of thepredose FEV1 at randomization as a covariate.

If the superiority condition was met in comparison to BUDpMDI, a 95% CI for the ratio of treatment effects containedwithin equivalence limits of 80e125% [5] for FEV160 min postdose and FEV1 predose would establish thera-peutic equivalence of BUD/FM BAI and BUD/FM pMDI. Themodel used to estimate the ratio of treatment effects wasthe same for predose FEV1 as for postdose FEV1 describedabove (ie, a multiplicative ANCOVA with the treatment andcountry as factors and baseline predose FEV1 as a covariate).

For all variables other than these, pair-wise comparisonswere made and nominal (unadjusted for multiplicity) pvalues are reported. All hypothesis testing was conductedusing 2-sided tests. The p values were rounded to 3 decimalplaces, and all p values �0.05 after rounding were consid-ered statistically significant.

Results

Patients

Patient disposition is shown in Fig. 3, and the key de-mographics of the randomized population are shown in

e inhaler; bid, twice daily; BUD, budesonide; FM, formoterol;

Table 1 Patient demographics and baseline asthma characteristics.

Characteristic BUD/FM BAI (n Z 71) BUD/FM pMDI (n Z 71) BUD pMDI (n Z 72)

Sex, n (%)Male 34 (47.9) 24 (33.8) 37 (51.4)Female 37 (52.1) 47 (66.2) 35 (48.6)

Age, y 42.8 (16.2) 42.6 (16.9) 42.7 (14.4)Age group, n (%)12 to

Figure 4 Ratio of geometric mean of treatment to baseline: 60-min postdose FEV1 (L) over 12 weeks and treatment average.Ratio is calculated as geometric mean of treatment average divided by geometric mean at baseline. Treatment average is definedas the mean of all available variables after randomization. Baseline for postdose FEV1 is predose FEV1 at visit 4 (week 0). BAI,breath-actuated metered-dose inhaler; BUD, budesonide; FEV1, forced expiratory volume in 1 s; FM, formoterol; pMDI, pressurizedmetered-dose inhaler; Trt. Avg., treatment average.

Budesonide/formoterol via breath-actuated inhaler 175

pMDI was estimated to be 1.03, with a CI of 0.99e1.08,confirming therapeutic equivalence.

Secondary outcomes

The mean change from baseline over time in morning andevening PEF for the 3 treatment groups is shown in Fig. 6.

Table 2 Treatment group comparisons for postdose andpredose FEV1 (L).

Treatment Ratio of treatment to baseline

LS mean(CV%)a

95% CI p Value

Postdose FEV1BUD/FM BAI vs BUD pMDI 1.11 (1.99) 1.07e1.16

Figure 5 Ratio of geometric mean of treatment to baseline: predose FEV1 (L) over 12 weeks and treatment average. The ratio iscalculated as geometric mean of treatment average divided by geometric mean at baseline. Treatment average is defined as themean of all available variables after randomization. Baseline is defined as the last predose value before first dose of randomizedtherapy at visit 4 (week 0). BAI, breath-actuated metered-dose inhaler; BUD, budesonide; FEV1, forced expiratory volume in 1 s;FM, formoterol; pMDI, pressurized metered-dose inhaler; Trt. Avg., treatment average.

176 K.R. Murphy et al.

The most frequently reported AEs were viral upper res-piratory tract infections, asthma, bronchitis, bacterialupper respiratory tract infection, nasopharyngitis, oralcandidiasis, and enteritis. Viral upper respiratory tract in-fections occurred in 2 (2.8%), 5 (7.0%), and 3 (4.2%) of pa-tients in the BUD/FM BAI, BUD/FM pMDI, and BUD pMDIgroups, respectively. Asthma was listed as an AE in 1 (1.4%),2 (2.8%), and 3 (4.2%) of the patients in the BUD/FM BAI,BUD/FM pMDI, and BUD pMDI groups, respectively. Thecorresponding rates for bronchitis were 1 (1.4%), 3 (4.2%),and 0 (0.0%). Other AEs occurred in �2 patients inany group.

Eight patients discontinued treatment because of AEs, ofwhom 5 discontinued because of asthma exacerbations (1,1, and 3 patients in the BUD/FM BAI, BUD/FM pMDI, and BUD

Figure 6 Self-reported morning (A) and evening (B) PEF. Baselimetered-dose inhaler; BUD, budesonide; FEV1, forced expiratorypMDI, pressurized metered-dose inhaler.

pMDI groups, respectively). Other causes of discontinuationwere gout (BUD/FM BAI), eczema and gingival pain (BUD/FMpMDI), and bacterial upper respiratory tract infection(BUD pMDI).

Discussion

The aim of this study was to evaluate the therapeuticequivalence of a BAI and conventional pMDI devicedelivering the BUD/FM combination. For both postdoseand predose FEV1, the CIs for the ratio of the treatmenteffects for BUD/FM BAI and pMDI were within the 95% CIequivalence limits of 80e125%, demonstrating therapeu-tic equivalence. All secondary end points provide

ne is the mean of run-in period values. BAI, breath-actuatedvolume in 1 s; FM, formoterol; PEF, peak expiratory flow;

Table 3 Treatment group differences for secondary end points.

Treatment ANCOVA summary treatment comparisons

LS mean (SE) 95% CI p Valuea

Morning PEF (L/min)

BUD/FM BAI minus BUD pMDI 35.01 (6.72) 21.77 to 48.26

178 K.R. Murphy et al.

The BAI device is similar to the pMDI device in manyways including identical drug formulations, canister, valvetype, and stem orifice, as well as very similar mouth-pieces. The basic mode of operation is also the same be-tween the pMDI and BAI devices: a force is applied to thecanister base, which depresses the metering valve andreleases a dose. For a pMDI inhaler, the force is createdwhen the patient presses down on the top of the inhalerwhile inhaling. For the BAI device, a patient’s inhalationtriggers a spring, which applies the force for release of themedication (therefore eliminating the need to coordinateinhalation with manual actuation). The trigger inspiratoryflow rate is less than 28 L/min [9] which almost all pa-tients can achieve [10]. The BAI was a new device to allpatients in the study, and results from the patient func-tionality study indicated that the ease of use was similarto the pMDI device.

Patients in both the pMDI and BAI groups reported thattheir device delivered a dose on greater than 99% of oc-casions. However, there were differences in the ease ofidentifying when medication was running out. More pa-tients using the BAI device (55%) than the pMDI device (41%)found it ‘extremely easy,’ which may be because of thedifferent dose counters used on each device. The pMDIdevice has a fuel-gauge-style dose counter, with an arrowpointing to a circular gauge ranging from 120 to 0, withdemarcations every 5 actuations and numerals every 10actuations. The BAI device has a mechanical digital counterthat displays the exact number of actuations remaining,starting at 120 and counting down in increments of 1 aftereach actuation.

The safety profile of both devices was consistent withthat of previously published results [6,7]. Noonan et al. [6]compared BUD/FM pMDI with BUD pMDI, FM DPI, BUD plusFM in separate inhalers (BUD pMDI þ FM DPI), and placeboin moderate to severe asthma patients. As with the pre-sent study, all treatments were well tolerated, and mostAEs were mild to moderate [6]. In the Noonan study, theincidence of oral candidiasis in the BUD/FM pMDI groupwas higher than that in other groups [6]. However, in thepresent study, oral candidiasis was experienced by only 1patient in each patient group.

Corren et al. [7] also reported safety results from a 12-week trial of BUD/FM pMDI versus BUD pMDI and FM DPI inpatients with mild to moderate asthma. As with the presentstudy, the treatments were well tolerated, and most AEswere of mild or moderate intensity [7]. In the Corren study[7], the most common AE possibly related to BUD/FM pMDItreatment was cough (2 patients; 1.6%), with single (0.8%)additional cases of headache, pharyngolaryngeal pain,tremor, and jitteriness. In the present study, headacheoccurred in just 1 patient (1.4%) in the BUD/FM BAI groupand in none of the patients receiving BUD/FM pMDI. Theonly case of cough was reported in the BUD/FM BAI group,and 1 patient in each of the BUD/FM groups experiencedtremor.

Conclusion

BUD/FM 160/4.5 mg � 2 inhalations bid administered by aBAI device is therapeutically equivalent to BUD/FM 160/

4.5 mg � 2 inhalations bid delivered by a conventionalpMDI based on both predose and postdose FEV1.Furthermore, no difference in safety profiles was identi-fied. The introduction of BUD/FM BAI would represent anexpansion of options to help tailor effective ICS/LABAcombination therapy for patients with moderate-to-severe asthma.

Funding

This study was supported by AstraZeneca LP. F.T., A.A., andG.E. are employees of AstraZeneca LP. T.U. is a formeremployee of AstraZeneca LP.

Author declaration of financial interest

Kevin R. Murphy: Consultancy/Advisory board: Aerocrine,AstraZeneca LP, Genentech, Greer, Novartis, Meda, Merck,Mylan, and Teva. Speaker: Aerocrine, AstraZeneca LP,Genentech, Novartis, Meda, Merck, Mylan, and Teva. Hon-oraria: Aerocrine, AstraZeneca LP, Genentech, Greer,Novartis, Meda, Merck, Mylan, and Teva. Research grant:Teva.

Rajiv Dhand: Speaker’s honorarium: Glaxo Smith-Kline.Consultant: Bayer. Clinical trials: Mylan, Bayer, CardeasPharma.

Frank Trudo: Full-time employee of and shareholder inAstraZeneca LP.

Tom Uryniak: Former employee of and shareholder inAstraZeneca LP.

Ajay Aggarwal: Full-time employee of and shareholderin AstraZeneca LP.

Göran Eckerwall: Full-time employee of and share-holder in AstraZeneca LP.

Acknowledgments

The authors acknowledge Scientific Connexions (Lyndhurst,NJ, USA) for medical writing assistance funded by Astra-Zeneca LP (Wilmington, DE, USA).

References

[1] National Heart, Lung, and Blood Institute. National AsthmaEducation and Prevention Program. Expert panel report 3:guidelines for the diagnosis and management of asthma. Fullreport 2007. http://www.nhlbi.nih.gov/guidelines/asthma/asthgdln.pdf. [accessed: 06.06.13].

[2] Global Initiative for Asthma (GINA). Pocket guide for asthmamanagement and prevention. Updated 2012. http://www.ginasthma.org/local/uploads/files/GINA_Pocket2013_May15.pdf. [accessed: 12.07.13].

[3] Melani AS, Bonavia M, Cilenti V, Cinti C, Lodi M, Martucci P,et al. Inhaler mishandling remains common in real life and isassociated with reduced disease control. Respir Med 2011;105:930e8.

[4] Gross G, Cohen RM, Guy H. Efficacy response of inhaled HFA-albuterol delivered via the breath-actuated Autohaler�inhalation device is comparable to dose in patients withasthma. J Asthma 2003;40:487e95.

http://www.nhlbi.nih.gov/guidelines/asthma/asthgdln.pdfhttp://www.nhlbi.nih.gov/guidelines/asthma/asthgdln.pdfhttp://www.ginasthma.org/local/uploads/files/GINA_Pocket2013_May15.pdfhttp://www.ginasthma.org/local/uploads/files/GINA_Pocket2013_May15.pdfhttp://www.ginasthma.org/local/uploads/files/GINA_Pocket2013_May15.pdfhttp://refhub.elsevier.com/S0954-6111(14)00447-8/sref1http://refhub.elsevier.com/S0954-6111(14)00447-8/sref1http://refhub.elsevier.com/S0954-6111(14)00447-8/sref1http://refhub.elsevier.com/S0954-6111(14)00447-8/sref1http://refhub.elsevier.com/S0954-6111(14)00447-8/sref1http://refhub.elsevier.com/S0954-6111(14)00447-8/sref2http://refhub.elsevier.com/S0954-6111(14)00447-8/sref2http://refhub.elsevier.com/S0954-6111(14)00447-8/sref2http://refhub.elsevier.com/S0954-6111(14)00447-8/sref2http://refhub.elsevier.com/S0954-6111(14)00447-8/sref2

Budesonide/formoterol via breath-actuated inhaler 179

[5] Centers for Drug Evaluation and Research. Guidance for in-dustry. Statistical approaches for establishing bioequivalence.Rockville, MD: US Dept of Health and Human Services, Foodand Drug Administration; 2001.

[6] Noonan M, Rosenwasser LJ, Martin P, O’Brien CD, O’Dowd L.Efficacy and safety of budesonide and formoterol in onepressurised metered-dose inhaler in adults and adolescentswith moderate to severe asthma: a randomised clinical trial.Drugs 2006;66:2235e54.

[7] Corren J, Korenblat PE, Miller CJ, O’Brien CD,Mezzanotte WS. Twelve-week, randomized, placebo-controlled, multicenter study of the efficacy and tolera-bility of budesonide and formoterol in one metered-dose

inhaler compared with budesonide alone and formoterolalone in adolescents and adults with asthma. Clin Ther 2007;29:823e43.

[8] Spector SL, Martin UJ, Uryniak T, O’Brien CD. Budesonide/-formoterol pressurized metered-dose inhaler versus budeso-nide: a randomized controlled trial in black patients withasthma. J Asthma 2012;49:70e7.

[9] Data on file. AstraZeneca. Specification for Symbicort BA de-vice. GEL version ID: RITA.000-366-893.2.0.

[10] Melani AS, Bracci LS, Rossi M. Reduced peak inspiratory effortthrough the Diskus� and the Turbuhaler� due to mishandling iscommon in clinical practice. Clin Drug Investig 2005;25:543e9.

http://refhub.elsevier.com/S0954-6111(14)00447-8/sref3http://refhub.elsevier.com/S0954-6111(14)00447-8/sref3http://refhub.elsevier.com/S0954-6111(14)00447-8/sref3http://refhub.elsevier.com/S0954-6111(14)00447-8/sref3http://refhub.elsevier.com/S0954-6111(14)00447-8/sref4http://refhub.elsevier.com/S0954-6111(14)00447-8/sref4http://refhub.elsevier.com/S0954-6111(14)00447-8/sref4http://refhub.elsevier.com/S0954-6111(14)00447-8/sref4http://refhub.elsevier.com/S0954-6111(14)00447-8/sref4http://refhub.elsevier.com/S0954-6111(14)00447-8/sref4http://refhub.elsevier.com/S0954-6111(14)00447-8/sref5http://refhub.elsevier.com/S0954-6111(14)00447-8/sref5http://refhub.elsevier.com/S0954-6111(14)00447-8/sref5http://refhub.elsevier.com/S0954-6111(14)00447-8/sref5http://refhub.elsevier.com/S0954-6111(14)00447-8/sref5http://refhub.elsevier.com/S0954-6111(14)00447-8/sref5http://refhub.elsevier.com/S0954-6111(14)00447-8/sref5http://refhub.elsevier.com/S0954-6111(14)00447-8/sref5http://refhub.elsevier.com/S0954-6111(14)00447-8/sref6http://refhub.elsevier.com/S0954-6111(14)00447-8/sref6http://refhub.elsevier.com/S0954-6111(14)00447-8/sref6http://refhub.elsevier.com/S0954-6111(14)00447-8/sref6http://refhub.elsevier.com/S0954-6111(14)00447-8/sref6http://refhub.elsevier.com/S0954-6111(14)00447-8/sref7http://refhub.elsevier.com/S0954-6111(14)00447-8/sref7http://refhub.elsevier.com/S0954-6111(14)00447-8/sref7http://refhub.elsevier.com/S0954-6111(14)00447-8/sref7http://refhub.elsevier.com/S0954-6111(14)00447-8/sref7http://refhub.elsevier.com/S0954-6111(14)00447-8/sref7http://refhub.elsevier.com/S0954-6111(14)00447-8/sref7http://refhub.elsevier.com/S0954-6111(14)00447-8/sref7

Therapeutic equivalence of budesonide/formoterol delivered via breath-actuated inhaler vs pMDIIntroductionMethodsPatientsStudy design and treatmentConcomitant medicationsStudy objectives and evaluationsSafety evaluationsStatistical analyses

ResultsPatientsEfficacySecondary outcomesSafety

DiscussionConclusionFundingAuthor declaration of financial interestAcknowledgmentsReferences