Efficacy and safety of dry powder fluticasonepropionate in children with persistent asthmaCraig F LaForce, MD*; David S Pearlman, MD†; Michael E Ruff, MD‡; William S Silvers, MD§;Steven W Weinstein, MD�; Diane S Clements¶; Alison Brown¶; Susan Duke, MS¶;Stuart M Harding, MD¶; and Karen W House, MS¶

Background: Flovent Diskus is a powder formulation of the inhaled corticoste-roid fluticasone propionate (FP) delivered via a breath-actuated, multidose inhaler.Objective: To determine the efficacy and safety of dry powder FP administered

once or twice daily (200 �g per day) to children with persistent asthma.Methods: Twelve-week, randomized, double-blind, placebo-controlled, multi-

center trial with a 52-week, open-label extension. Children aged 4 to 11 wererequired to have pulmonary function 50% to 85% of predicted values. The popu-lation was stratified for baseline therapy (inhaled corticosteroid/cromolyn or bron-chodilators only). After a 2-week placebo run-in, 242 patients received dry powderFP 200 �g each morning, dry powder FP 100 �g BID, or placebo for 12 weeks; 192were rerandomized to the QD or BID regimen for an additional 52 weeks ofopen-label treatment. Primary endpoints were mean changes in FEV1 and morningPEF recorded at clinic visits.Results: Both dry powder FP regimens significantly improved FEV1, evening

PEF, and asthma symptoms at the double-blind phase endpoint (P� .017 comparedwith placebo). The BID regimen also significantly improved morning PEF andnighttime awakenings due to asthma (P � .005). Among patients previously treatedwith inhaled corticosteroids/cromolyn, improvements observed with the QD andBID regimens were similar. Patients switched from BID to open-label QD treatmentshowed additional improvements at week 52 generally comparable to patients whoreceived the BID regimen during both phases. Fluticasone propionate was welltolerated for up to 64 weeks with few reports of drug-related adverse events ormorning plasma cortisol abnormalities.Conclusions: Once daily dosing of dry powder FP 200 �g is an effective and

convenient alternative for children whose asthma is controlled with a more frequentdosing regimen of inhaled corticosteroids.

Ann Allergy Asthma Immunol 2000;85:407–415.

INTRODUCTIONAirway inflammation is considered tobe the primary pathogenic mechanismunderlying asthma.1,2 Airway inflam-mation has been detected in patients

with even mild or newly reported asth-ma,2,3 and there is evidence of de-creased pulmonary function by 6 yearsof age.4 Early intervention with anti-inflammatory medications such as in-haled corticosteroids may delay pro-

gression of the disease as well asmaximize lung function.5 Asthmamanagement guidelines now recom-mend the daily use of anti-inflamma-tory agents such as inhaled corticoste-roids, cromolyn sodium, or nedocromilfor all children requiring symptomatictreatment more than twice weekly (ie,persistent asthma).1; however, adher-ence to anti-inflammatory therapy inasthma is often poor.6,7 A low rate ofadherence to inhaled corticosteroid usehas been associated with more frequentasthma exacerbations in children.8 Asimplified treatment plan that includesfewer medication administrations perday may improve adherence.9Flovent Diskus (Glaxo Wellcome

Inc) is a powder formulation of thetopically active10 inhaled corticoste-roid fluticasone propionate (FP) in amultidose, breath-operated inhaler.The Diskus provides medication for upto 1 month (60 sealed blisters) alongwith a dose-counter. Large-scale clin-ical trials have established the safetyand efficacy of FP powder deliveredtwice daily via the Diskus and Dis-khaler (Glaxo Wellcome Inc) devicesin patients as young as 4 years old.11,12The present study was designed toevaluate the safety and efficacy of FPDiskus 200 �g, the maximum dailydosage recommended in pediatric pa-tients, when administered once daily(200 �g QD) or twice daily (100 �gBID) to patients 4 to 11 years old. Theinitial 12-week treatment phase wasplacebo controlled. To evaluate thelong-term safety and effectiveness ofthe drug under less controlled condi-tions, patients received open-label FPDiskus 200 �g once daily or 100 �gtwice daily for an additional 52 weeks.

* North Carolina Clinical Research, Raleigh,North Carolina.† Colorado Allergy and Asthma Centers,

P.C., Aurora, Colorado.‡ Dallas Allergy and Asthma Center, Dallas,

Texas.§ Allergy, Asthma and Immunology Clinic of

Colorado, Englewood, Colorado.� Allergy and Asthma Specialists Medical

Group, Huntington Beach, California.¶ Glaxo Wellcome Inc., Research Triangle

Park, NC. Note: Ms. Brown now is affiliatedwith CIBA Vision, Duluth, Georgia.

This study was supported by a grant fromGlaxo Wellcome Inc, Research Triangle Park,North Carolina.These results were presented in part in Feb-

ruary 1997 at the Annual Meeting of the Amer-ican Academy of Allergy, Asthma, and Immu-nology in San Francisco, California.Received for publication December 22, 1999.Accepted for publication in revised form

March 23, 2000.

VOLUME 85, NOVEMBER, 2000 407

MATERIALS AND METHODSPatient SelectionMale and premenarchal female pa-tients aged 4 to 11 years were enrolledin the study (FLTA2008) if they hadasthma13 requiring daily pharmaco-therapy for 3 months immediatelyprior to the study. Patients were re-quired to have a baseline PEF of�85% of predicted values (all pa-tients) and an FEV1 of 50% to 85% ofpredicted values (aged 6 to 11).14 Pa-tients aged 6 to 11 demonstrated a�15% reversibility to albuterol or�15% variability in FEV1 within 6months prior to the study; reversibilityin 4- to 5-year-olds was assessed byPEF. Patients also were required tohave stable asthma. Criteria for accept-able asthma stability during the last 7days of the run-in period included nomore than 3 days of �12 puffs ofalbuterol aerosol per day or �6 dosesof albuterol powder per day, no morethan 3 mornings when the PEF de-creased more than 20% below the pre-vious evening’s PEF, and no more thanthree nights with awakenings resultingfrom asthma that required treatmentwith albuterol.Patients were excluded from the

study if they had a history of life-threatening asthma or other severechronic disease; any respiratory tractinfection within 2 weeks prior to the

study; had been exposed to or hadchickenpox within 3 weeks prior to thestudy; used oral or parenteral cortico-steroids within 1 month prior to thestudy; used methotrexate or gold saltsor any prescription or over-the-countermedication which might have affectedthe course of asthma or its treatment;or had participated in any previous trialinvolving the use of the Diskus device.

Study DesignThis randomized, double-blind, paral-lel-group, placebo-controlled studywas conducted at 21 clinical centers inthe United States. Informed consentsigned by a parent or legal guardianand Institutional Review Board ap-proval were obtained before enroll-ment. After a 2-week, single-blind,placebo run-in period, patients wererandomly assigned by stratum (base-line therapy of inhaled corticosteroids/cromolyn or bronchodilators alone) toone of the following blinded treat-ments administered twice daily for 12weeks: FP Diskus 200 �g QD in themorning/placebo in the evening, FPDiskus 100 �g BID, or placebo BID.Patients completing the double-blindphase or discontinuing for worseningasthma were rerandomized to receiveopen-label 200 �g QD or 100 �g BIDfor 52 additional weeks of treatment.

During the placebo run-in, patientswere allowed to continue receivingtheophylline, inhaled cromolyn, or in-haled corticosteroids other than FP ifthe treatment was started at least 3months earlier with no change in thedose or regimen within the last month.At randomization, patients receivinginhaled corticosteroids or cromolyndiscontinued these medications. Theo-phylline could be continued during thestudy but was withheld for 24 to 36hours prior to clinic visits. During theopen-label phase, additional asthmamedications including oral corticoste-roids were allowed.Patients were required to maintain

stable asthma during the study. Pa-tients failing to meet predefined con-tinuation criteria (Table 1) at clinicvisits, and those who had a clinicalasthma exacerbation requiring hospi-talization or asthma medication ex-cluded by the study protocol, were dis-continued from the study. Patients alsocould be discontinued for worseningasthma at the discretion of the investi-gator.MeasurementsDouble-blind phase. Patients returnedto the clinic weekly for the first 4weeks and at weeks 6, 8, 10, and 12.All visits were conducted between 7and 10 AM, and baseline pulmonaryfunction tests and other assessmentswere performed prior to the morningdose of study drug administration. Allpatients withheld albuterol use for �6hours prior to clinic visits. The formu-lation of “as needed” albuterol (aerosolor powder) and the specific spacer de-vice used, if any, were consistentthroughout the study.The primary efficacy variables were

predose morning FEV1 and morningPEF obtained at each clinic visit (clinicPEF). Secondary efficacy variables in-cluded duration of study participationand the following information re-corded by patients on daily dairy cards:morning PEF (diary PEF), eveningPEF; asthma symptom scores; albu-terol use; and nighttime awakeningsrequiring albuterol use. Safety assess-ments included adverse events moni-

Table 1. Stability Criteria

Stability Criteria for Continuation in Study

At each clinic visit during thedouble-blind phase, patientswere required todemonstrate pulmonaryfunction greater than thestability limit determined atbaseline

Patients were also required to meetthe following continuation criteriabased on diary card data from the 7days preceding each clinic visit:

● FEV1 stability limit (aged 6 to11): a 20% decrease in thebest of three FEV1 values atbaseline

● PEF stability limit (allpatients): a 20% decrease inmean morning PEF obtainedduring the last 7 days of therun-in period.

● �2 days of �12 puffs of albuterolaerosol/day or �6 doses ofalbuterol powder/day

● �2 nights with awakenings resultingfrom asthma requiring treatmentwith albuterol

● �2 days with a morning or eveningPEF below the PEF stability limit

408 ANNALS OF ALLERGY, ASTHMA, & IMMUNOLOGY

tored throughout the study and routineclinical laboratory tests, unstimulatedmorning plasma cortisol concentra-tions, 12-lead ECGs, and physical ex-aminations performed at endpoint. Anabnormal plasma cortisol concentra-tion was defined post hoc as a value�4 �g/dL (�110 nmol/L),15 as chil-dren tend to have lower values thanadults.Open-label phase. Patients returned

to the clinic every 2 weeks for the first4 weeks and then every 4 weeks forpulmonary function testing and anevaluation of adverse events. Betweenvisits, patients maintained diary cardson which they recorded the following:morning peak flow measurements; thenumber of days without asthma symp-toms; and the number of nights withawakenings requiring treatment. Rou-tine clinical laboratory tests, morningplasma cortisol concentrations, ECGs,and physical examinations were as-sessed at 26 weeks and at endpoint.Statistical AnalysesThe sample size in this study (at least80 patients per treatment) was selectedin order to provide �80% power indetecting a difference in FEV1 of 0.25L and a difference in PEF of 16 L/minbetween any two treatment groupsbased on a two-sample t-test with asignificance level of .05. Mean changefrom baseline at endpoint was the pri-mary analysis for all efficacy measure-ments. Endpoint was defined as thefinal available measurement of eachphase regardless of whether the patientcompleted the study. The intent-to-treat population (all patients exposedto study drug) was used for all safetyand efficacy analyses. All statisticaltests were two-sided, and testing wasperformed on combined data from allinvestigators, controlling for investiga-tor effect. Treatment differences at orbelow the .05 level were consideredstatistically significant. Pairwise com-parisons were performed without ad-justing P values for the number ofcomparisons made, but were inter-preted only when the overall testamong treatment groups was statisti-cally significant. FEV1 and PEF values

collected during the double-blindphase were tested for treatment differ-ences using an analysis of varianceF-test. Analysis of efficacy at individ-ual clinic visits was performed usingthe last observation carried forwardmethod. Determination of the proba-bility of remaining in the study overtime (ie, duration of study participa-tion) was based on the Kaplan-Meierestimates of survival. Patient-ratedsymptom scores (0 � none, 1 � mild,2 � moderate, and 3 � severe), albu-terol use, and nighttime awakeningswere averaged over 7 days, andchanges from baseline as well as pair-wise testing, were analyzed using thevan Elteren variation of the WilcoxonRank Sum test.16 Study drug adherencewas calculated by dividing the numberof doses taken (as indicated by thedose counter) by the number of dosesscheduled.To evaluate the long-term efficacy

of FP Diskus, patients participating inthe open-label phase were divided intosix groups based upon the switch intreatment from the double-blind phase(eg, placebo to 200 �g QD, 100 �gBID to 200 �g QD). Changes in effi-cacy parameters from the beginning ofthe open-label phase to week 52 weretabulated for each of the six switchgroups. Mean change in morning pre-dose FEV1 at each visit was depictedgraphically using the last observationcarried forward method. No statisticalanalyses were performed due to thesmall sample sizes in each group.For both study phases, frequencies

of patients having any adverse eventand frequencies of patients with anyadverse event considered by the inves-tigator to be potentially drug-relatedwere tabulated by body system andby treatment received. Frequencies ofpatients with clinically significant lab-oratory tests, clinically significantECGs, or morning plasma cortisol ab-normalities also were generated. Fish-er’s exact test was used to compare thedistribution of patients within eachtreatment group having an increase, adecrease, or no change in morningplasma cortisol concentrations from

baseline to double-blind endpoint(week 12 or early termination).

RESULTSDouble-Blind PhaseTwo hundred forty-two children, in-cluding 21 4-to 5-year-olds, were en-rolled at 21 clinical sites. At baseline,more patients were receiving inhaledcorticosteroids or cromolyn (ICS stra-tum; 57%) than bronchodilator therapyonly (BDT stratum; 43%). Demo-graphic characteristics and baselinepulmonary function test results weresimilar among treatment groups (Table2). Mean study medication adherencerates at endpoint were similar amongtreatment groups (87% to 93%). Sixty-two percent, 73%, and 41% of patientsin the QD, BID, and placebo groups,respectively, completed 12 weeks ofdouble-blind therapy before participat-ing in the open-label phase. The mostcommon reason for discontinuationwas the failure to meet stringent, pre-defined continuation criteria at clinicvisits. Patients receiving either activedrug regimen were significantly morelikely to meet these criteria comparedwith placebo (P � .001; Fig 1). Atendpoint, 54% of the patients in theplacebo group were withdrawn due tofailure to meet these criteria comparedwith 30% and 20% in the QD and BIDgroups, respectively.Both active treatments significantly

improved morning predose FEV1 at allclinic visits compared with placebo(P � .028, last visit carried forwardanalysis; Fig 2). The two active groupswere similar for mean change in FEV1.There also were significant increasesin the percentage of predicted FEV1values in the two active groups com-pared with placebo (P � .005) and inthe BID group compared with the QDgroup (P� .045 at endpoint). The BIDregimen significantly improved clinicand diary morning PEF compared withplacebo (P � .005; Table 3); a similartrend was observed with the QD regi-men (P� .082). Both the QD and BIDgroups significantly improved eveningPEF (P � .017; increases of 24 and 27L/min, respectively). Patients in the

VOLUME 85, NOVEMBER, 2000 409

BDT stratum who were randomized tothe BID group showed greater im-provement in pulmonary function atendpoint than patients treated with theQD regimen or placebo. Within theICS stratum, improvements in pulmo-nary function observed in the BID andQD groups were comparable.At baseline, most patients reported

mild symptoms and few nighttimeawakenings. Both active treatmentssignificantly reduced asthma symptomscores at endpoint with decreases ofapproximately 27% in the QD groupand 41% in the BID group comparedwith a decrease of 1% in the placebogroup (P � .001). The number ofnighttime awakenings decreased inboth FP Diskus groups at endpoint, butwas significantly different from pla-cebo only in the BID group (P� .001).There was not a significant overallchange in albuterol use across treat-ment groups (P � .061). Mean dailyalbuterol use in FP Diskus-treated pa-tients declined by 22% and 33% in theQD and BID groups, respectively,compared with no change in the pla-cebo group.Both FP Diskus dosing regimens

were well tolerated during the double-blind phase with no reports of seriousdrug-related adverse events or treat-ment-related trends. Table 4 shows alladverse events reported by �10% ofpatients in any treatment group. Feverwas reported for more patients in theactive groups (12% to 13%) comparedwith placebo (1%); however, becausemore patients in the active groupscompared with placebo completed theentire 12 weeks of double-blind treat-ment, the higher incidence of fevermay be due to exposure and is notconsidered clinically significant. Therewere few reports of adverse eventsconsidered potentially drug-related(2%, 3%, and 1% in the QD, BID, andplacebo groups, respectively). Theseincluded candidiasis, throat irritation,and gastrointestinal discomfort andpain in one patient each after FP Dis-kus treatment and asthma exacerbationin one patient treated with placebo.Three patients withdrew from the dou-

ble-blind phase because of adverseevents; none of these adverse eventswas considered treatment related. Ofthese patients, one was a placebo re-cipient who had a serious adverseevent (status asthmaticus) and waswithdrawn from the study because oflack of efficacy.

Mean unstimulated morning plasmacortisol concentrations were similar inthe active and placebo groups at end-point compared with baseline (Table5). FP Diskus had little effect on thenumber of patients with a morning cor-tisol value �4 �g/dL (�110 nmol/L).At baseline, nine patients (4%) had a

Table 2. Demography and Baseline Assessments*

FP, 200 �g QD FP, 100 �g BID Placebo

Number of patients enrolled 84 80 78ICT stratum 45 (54%) 47 (59%) 47 (60%)BDT stratum 39 (46%) 33 (41%) 31 (40%)

Age (mean, SEM) 8.5 (0.2) 8.5 (0.2) 8.8 (0.2)Sex, m % 70 71 73FEV1, % predicted 70 � 0.8 70 � 1.0 73 � 0.8FEV1, L 1.40 � 0.04 1.36 � 0.04 1.51 � 0.04

ICT stratum 1.38 � 0.05 1.37 � 0.05 1.52 � 0.05BDT stratum 1.42 � 0.07 1.34 � 0.05 1.50 � 0.07

Clinic AM PEF, L/min 208 � 7 208 � 6 219 � 7ICT stratum 204 � 10 215 � 7 221 � 9BDT stratum 212 � 12 197 � 11 217 � 12

Diary AM PEF, L/min 220 � 8 222 � 7 234 � 8PM PEF, L/min 235 � 8 233 � 7 244 � 8Asthma symptom scores† 0.88 � 0.06 0.93 � 0.07 0.85 � 0.06Albuterol use, puffs/day 2.1 � 0.3 2.0 � 0.3 1.7 � 0.2Nighttime awakenings/night 0.10 � 0.02 0.10 � 0.02 0.08 � 0.02

* Unless otherwise expressed, values presented are means � SEM (standard error of themean).† Asthma symptom scores (0 � none; 1 � mild; 2 � moderate; and 3 � severe).FP � fluticasone propionate; BID � twice-daily; QD � once-daily; ICT � inhaled corticosteroidtherapy; and BDT � bronchodilator only therapy.

Figure 1. Probability of remaining in the study (Kaplan-Meier Survival Curve; P � .001 for eachactive group versus placebo). Abbreviations: FP� fluticasone propionate, BID� twice-daily, and QD�once-daily.

410 ANNALS OF ALLERGY, ASTHMA, & IMMUNOLOGY

morning plasma cortisol value lessthan 4 �g/dL; all of these patients ex-cept one had normal values at the dou-ble-blind endpoint. Seven patients(3%) had a morning plasma cortisolless than 4 �g/dL at their last treatmentvisit: two in the QD group, three in theBID group, and two in the placebogroup. Two of the patients in the BIDgroup were concurrently receiving

prednisone; a third patient in the BIDgroup had a history of significant pred-nisone use and had an abnormal valueat baseline. Both FP Diskus groupswere statistically no different from pla-cebo in the distribution of patients hav-ing an increase, a decrease, or nochange in morning plasma cortisolconcentrations from baseline to the lasttreatment visit (Table 5).

Open-Label PhaseOne hundred ninety-two patients werererandomized to receive either FP 200QD (n � 94) or FP 100 BID (n � 98).Treatment groups were comparable fordemographic and baseline characteris-tics. Mean adherence rates were com-parable for the two regimens (84% to87%). More patients treated with theQD regimen (82%) completed thestudy compared with the BID regimen(71%). Few patients were withdrawnfor worsening asthma (1% in eachgroup). Approximately 30% of pa-tients reported taking a course of sys-temic corticosteroids. Most oral corti-costeroid use was associated with theonset of an upper respiratory tract in-fection.Patients treated with active drug

during the double-blind phase contin-ued to improve pulmonary functionduring the open-label phase regardlessof regimen (Fig 3). For patients re-maining in the study at week 52, per-centage increases from the end of the12-week double-blind phase rangedfrom 10% to 24% for FEV1 and from20% to 25% for clinic morning PEF(Table 6). Patients switched from BIDto QD dosing, and patients who re-ceived the QD regimen in both phases,showed improvement generally com-parable to patients who received BIDdosing during both phases. Patientswho previously received placeboshowed large improvements in pulmo-nary function (increases in FEV1 from30% in the QD group and 40% in theBID group). The QD and BID groupsalso showed similar asthma control asevidenced by patient reports of night-time awakenings and symptoms.Both regimens were well tolerated

with few reports of drug-related ad-verse events (4% to 6% of patients ineach group). Two patients in the BIDgroup withdrew from the open-labelphase due to an adverse event consid-ered to be potentially treatment related(morning plasma cortisol concentra-tion of 5 �g/dL (140 nmol/L) in onepatient, fatigue and moodiness in an-other patient). The only potentiallydrug-related event reported for more

Table 3. Change in Efficacy Variables from Baseline to Last Visit of Double-Blind Phase(mean � SEM)*

Efficacy variableFP, 200 �g QD

n � 84FP, 100 �g BID

n � 80Placebon � 78

FEV1 AM predose, L 0.13 � 0.04* 0.23 � 0.04* �0.04 � 0.04ICT stratum† 0.10 � 0.05 0.18 � 0.05 �0.05 � 0.05BDT stratum† 0.16 � 0.06 0.31 � 0.06 �0.02 � 0.05

FEV1, change in % predicted 6.5 � 2.0* 12.2 � 2.0*‡ �1.5 � 1.8Clinic AM PEF, L/min 33 � 7 42 � 6* 19 � 7

ICT stratum† 34 � 9 36 � 8 23 � 10BDT stratum† 31 � 11 51 � 10 14 � 9

Diary AM PEF, L/min 24 � 4 34 � 4* 14 � 3Diary PM PEF, L/min 24 � 4* 27 � 3* 11 � 3Asthma symptom scores �0.24 � 0.07* �0.38 � 0.08* �0.01 � 0.07Albuterol use, puffs/day �0.5 � 0.2 �0.7 � 0.3 0.0 � 0.2Nighttime awakenings/night �0.04 � 0.03 �0.06 � 0.02* 0.04 � 0.02

Abbreviations: FP � fluticasone propionate; BID � twice-daily; QD � once-daily; SEM �standard error of the mean; ICT � inhaled corticosteroid therapy; and BDT � bronchodilatoronly therapy.* Indicates a significant (P � .05) effect versus placebo.† No statistical testing performed due to small sample sizes.‡ Indicates a significant (P � .05) effect between two active treatments.

Figure 2. Mean change from baseline in morning predose FEV1 (*P � .028 for each active groupversus placebo; †P� .035 for the BID group versus the QD group) using last observation carried forwardmodel. Abbreviations: FP � fluticasone propionate, BID � twice-daily, and QD � once-daily.

VOLUME 85, NOVEMBER, 2000 411

than one patient was candidiasis (twopatients in the BID group, four patientsin the QD group). Nine patients (5%)had a morning plasma cortisol lessthan 4 �g/dL (110 nmol/L) at their lastvisit: three in the BID group, and six inthe QD group. Four of these patientswere receiving concurrent prednisone.No clinically relevant changes in meanmorning plasma cortisol concentra-tions were seen at the end of 52 weekscompared with pretreatment concen-trations (Table 7).

DISCUSSIONIn this study of children with mild-to-moderate persistent asthma, we haveshown that both once- and twice-dailydosing regimens of the inhaled corti-costeroid FP Diskus (200 �g per day)significantly improved FEV1, eveningPEF, and asthma symptoms comparedwith placebo. Both regimens con-trolled asthma significantly more thanplacebo as assessed by the frequencyof patients meeting stringent continua-tion criteria at each clinic visit duringthe 12-week double-blind phase. Al-though there was not a significantoverall difference among groups in thechange in albuterol use, both activegroups decreased mean use by 20% to30% (approximately one-half puff perday) compared with no change for theplacebo group. Patients receiving ac-tive treatment in the double-blindphase continued to improve in theopen-label phase, regardless of regi-men. Both regimens were well toler-ated for up to 64 weeks with few re-ports of drug-related adverse eventsand few abnormal morning plasmacortisol concentrations. During thedouble-blind phase, FP Diskus groupswere statistically no different from pla-cebo in the distribution of patients hav-ing an increase, a decrease, or nochange in morning plasma cortisolconcentrations. The incidence ofplasma cortisol abnormalities reportedwith long-term administration of FPDiskus was similar to the incidence atstudy enrollment.During the double-blind phase, only

the BID regimen significantly im-proved morning PEF and the fre-

quency of nighttime awakenings. Oneexplanation for why the QD regimen,administered in the morning, signifi-cantly improved evening PEF but notthe following morning’s PEF involvesthe timing of dose administration.There may be some acute effects ofinhaled corticosteroids that diminish at24 hours compared with 12 hours afterdose administration. Another explana-tion involves sample size. The QDgroup did not achieve a 16 L/min dif-ference from placebo in morning PEF,which was defined as a clinically sig-nificant difference in power calcula-tions for this study. However, therewas a statistical trend for greater im-provement in the QD group comparedwith placebo. Statistical significancewas not reached due to insufficientsample size to detect the smaller dif-

ference than expected between the QDand placebo groups (14 L/min).The BID regimen provided consis-

tently greater improvement than theQD regimen during the 12-week dou-ble-blind phase in the overall popula-tion, although the difference achievedstatistical significance only for percentpredicted FEV1. Among patients re-ceiving inhaled corticosteroids or cro-molyn at baseline, however, increasesin pulmonary function at endpointwere comparable between the QD andBID regimens. Data from the 52-weekopen-label phase suggest that thegreater improvements initially seenwith the BID regimen compared withthe QD regimen appeared to diminishover time. Patients switched from BIDdosing in the first 12 weeks to QDdosing in the open-label phase, and

Table 4. Adverse Events in �10% of Patients in any Group (double-blind phase)

Adverse EventFP, 200 �g QD

n � 84FP, 100 �g BID

n � 80Placebon � 78

Number of patients with one ormore adverse events (%)

64 (76%) 61 (76%) 57 (73%)

Ear, nose, and throatUpper respiratory tractinfection

17 (20%) 12 (15%) 12 (15%)

Throat irritation 10 (12%) 10 (13%) 7 (9%)Sinusitis 8 (10%) 10 (13%) 6 (8%)

Non-site specificCandidiasis unspecified site 6 (7%) 8 (10%) 16 (21%)Fever 10 (12%) 10 (13%) 1 (1%)

GastrointestinalCandidiasis mouth/throat 10 (12%) 10 (13%) 8 (10%)

NeurologyHeadaches 11 (13%) 7 (9%) 5 (6%)

Abbreviations: FP � fluticasone propionate; BID � twice-daily; and QD � once-daily.

Table 5. Morning Plasma Cortisol Concentrations (�g/dL) During the Double-Blind Phase(mean � SEM)

FP, 200 �g QD FP, 100 �g BID Placebo

Baseline 9.2 � 0.6 (n�84) 8.5 � 0.4 (n�80) 9.0 � 0.5 (n�78)DB endpoint (week 12/early

termination)10.9 � 0.7 (n�81) 9.8 � 0.5 (n�80) 10.0 � 0.6 (n�74)

Change from baseline: (%)who had

Increase 47 (58%) 43 (54%) 42 (57%)Decrease 28 (35%) 27 (34%) 28 (38%)No change 6 (7%) 10 (13%) 4 (5%)

Abbreviations: FP � fluticasone propionate; BID � twice-daily; QD � once-daily; SEM �standard error of the mean; and DB � double-blind.Note: 1.0 �g/dL � 27.6 nmol/L.

412 ANNALS OF ALLERGY, ASTHMA, & IMMUNOLOGY

patients who received the QD regimenin both phases, showed improvementat the end of a year generally compa-rable to patients who received BIDdosing during both phases. Factors thatmay have influenced the efficacy ob-served with both regimens during theopen-label phase include the naturalgrowth of the patients (with corre-sponding increases in lung volume)and the use of concurrent asthma med-ications.

The results of studies evaluating theefficacy of once-daily dosing of in-haled corticosteroids have beenmixed.17–22 Factors that may confoundstudy results include the degree ofasthma control at baseline, asthma se-verity, the time of day that the once-daily dosage was administered,23 andstudy duration. Generally, studies con-ducted with adults have shown thatonce-daily dosing of inhaled cortico-steroids is effective in most patients

with mild-to-moderate asthma, partic-ularly patients whose asthma is alreadywell controlled.24 Results from an8-week, double-blind trial in 328 chil-dren with stable, mild-to-moderateasthma demonstrated that a once-dailydose of 100 �g FP powder adminis-tered via Accuhaler (ie, FP Diskus) inthe evening was as effective as a twice-daily dose of 50 �g.25 In this study, weassessed the efficacy of once-daily(morning) and twice-daily dosing ofthe maximum pediatric dosage (200�g per day) of FP Diskus for up to 64weeks. The results with the BID regi-men confirm other studies in childrenas young as 4 that demonstrate thesafety and efficacy of twice-daily dos-ing with FP powder via Diskhaler11,12and FP powder via the new Diskusdevice.11 The positive results with theQD regimen in this study suggest thelong-term effectiveness of a reduceddosing schedule of FP Diskus in chil-dren once asthma is well controlledwith twice-daily doses of inhaled cor-ticosteroids. As with all inhaled corti-costeroids, FP Diskus doses should bereduced to the minimally effectivedose once asthma control is estab-lished.

CONCLUSIONSIn an effort to increase adherence toprescribed asthma medications, aneasy-to-use multidose powder inhalersuch as the Diskus and the option ofonce-daily dosing provide advantagesin the delivery of drugs to childrenwith asthma. This study has demon-strated the efficacy and safety of bothonce-daily and twice-daily dosing withFP Diskus (200 �g per day) for up to64 weeks in children 4 to 11-years-oldwith persistent asthma. Although theBID regimen provided greater initialimprovement, patients switched fromtwice daily dosing to once-daily dosingfor a year continued to improve pul-monary function generally comparableto those who continued BID treatment.In summary, once-daily dosing of FPDiskus 200 �g may be an effective andconvenient alternative to twice-dailydosing in some children with persistent

Figure 3. Mean change from double-blind endpoint in morning predose FEV1 using last observationcarried forward method. Figure 3a (upper panel) shows change in FEV1 for patients receiving FP, 200�g QD, during the open-label phase, and Figure 3b (lower panel) shows change in FEV1 for patientsreceiving FP, 100 �g BID during the open-label phase. FP � fluticasone propionate, BID � twice-daily,and QD � once-daily.

VOLUME 85, NOVEMBER, 2000 413

asthma, particularly those already wellcontrolled on their current regimen.

ACKNOWLEDGMENTSWe thank the following for their con-tributions to this study: P Boggs, MD,Shreveport, LA; J D Bray, MD, Mid-land, TX; P Chervinsky, MD, NorthDartmouth, MA; S P Galant, MD, Or-ange, CA; M F Goldstein, MD, Phila-delphia, PA; B Q Lanier, MD, FortWorth, TX; L Mendelson, MD, WestHartford, CT; R Menendez, MD, ElPaso, TX; R Morris, MD, Minneapolis,MN; S Pollard, MD, Louisville, KY; RH Schwartz, MD, Rochester, NY; MVandewalker, MD, Rolla, MO; and NWhitcomb, MD, Carmichael, CA.We would also like to thank Cheryl

Beale and Shehnaz Gangjee, PhD forassistance in writing and revising themanuscript and Abbas Hamedani forassistance in statistical analysis.

REFERENCES1. Highlights of the Expert Panel Report2: Guidelines for the diagnosis andmanagement of asthma. National Insti-tutes of Health (National Heart, Lung,and Blood Institute). NIH PublicationNo. 97-4051A. May 1997.

2. Djukanovic R, Roche WR, Wilson JW,

et al. Mucosal inflammation in asthma.Am Rev Respir Dis 1990;142:434–457.

3. Laitinen LA, Laitinen A, Altraja A, etal. Section I. Inflammatory determi-nants of asthma severity. Bronchial bi-opsy findings in intermittent or “early”asthma. J Allergy Clin Immunol 1996;98(5, Pt. 2):S3–6.

4. Martinez FD, Wright AL, Taussig LM,et al. Asthma and wheezing in the firstsix years of life. N Engl J Med 1995;332:133–138.

5. Agertoft L, Pedersen S. Effects oflong-term treatment with an inhaledcorticosteroid on growth and pulmo-nary function in asthmatic children.Respir Med 1994;88(5):373–381.

6. Mawhinney H, Spector SL, Kinsman

RA, et al. Compliance in clinical trialsof nonbronchodilator, anti-asthmamedications. Ann Allergy 1991;66(4):294–299.

7. Apter AJ, Reisine ST, Affleck G, et al.Adherence with twice-daily dosing ofinhaled steroids. Am J Respir Crit CareMed 1998;157(6):1810–1817.

8. Milgrom H, Bender B, Ackerson L, etal. Noncompliance and treatment fail-ure in children with asthma. J AllergyClin Immunol 1996;98(6 Pt 1):1051–1057.

9. Tashkin DP. Multiple dose regimens.Impact on compliance. Chest 1995;107(5) Suppl):176S–182S.

10. Lawrence M, Wolfe J, Webb Dr, et al.Efficacy of inhaled fluticasone propi-onate in asthma results from topical

Table 7. Morning Plasma Cortisol Concentrations (�g/dL) for Patients Treated During theOpen-Label (OL) Extension (mean � SEM)*

FP, 200 �g QD FP, 100 �g BID

Baseline (visit 2) 8.7 � 0.4 (n�91) 8.1 � 0.4 (n�93)OL endpoint (week 52

or early termination)9.5 � 0.5 (n�90) 9.4 � 0.5 (n�91)

Abbreviations: FP � fluticasone propionate; BID � twice-daily; QD � once daily; SEM �standard error of the mean; and OL � open-label.* Patients received up to 12 weeks of double-blind treatment (FP, 200 �g QD; FP, 100 �g BID;or placebo) before being re-randomized to the QD or BID regimen for 52 weeks of OLtreatment.Note: 1.0 �g/dL � 27.6 nmol/L.

Table 6. Change in Lung Function After 52 Weeks of Open-Label Treatment with FP Diskus QD or BID (200 �g per day)*†

Switched fromFP, 200QD, to:

Switched fromFP, 100 BID, to:

Switched fromPlacebo to

OL group 200 QD 100 BID 200 QD 100 BID 200 QD 100 BIDRandomized n � 33 n � 33 n � 32 n � 34 n � 29 n � 31

Week 52 n � 29 n � 26 n � 23 n � 25 n � 24 n � 19

FEV1, LDB endpoint 1.54 (0.09) 1.52 (0.09) 1.56 (0.08) 1.60 (0.09) 1.40 (0.11) 1.52 (0.07)Mean change 0.26 (0.07) 0.37 (0.07) 0.16 (0.09) 0.27 (0.08) 0.42 (0.07) 0.61 (0.11)% change 17 24 10 17 30 40

Clinic AM PEF, L/minDB endpoint 240 (15) 233 (16) 246 (15) 244 (13) 218 (17) 241 (13)Mean change 56 (12) 58 (11) 49 (12) 50 (9) 66 (17) 89 (17)% change 23 25 20 20 30 37

Diary AM PEF, L/minDB endpoint 242 (15) 237 (15) 248 (14) 249 (14) 227 (14) 252 (12)Mean change 47 (10) 46 (10) 29 (14) 34 (7) 59 (19) 78 (16)% change 19 19 14 14 26 31

Abbreviations: FP � fluticasone propionate; BID � twice-daily; QD � once-daily; OL � open label; DB � double-blind.* Standard error for each value is represented in parentheses.† Values for DB endpoint are an average of values recorded at each patient’s last visit in the double-blind phase. Mean change and % changevalues represent changes from DB endpoint to OL week 52.

414 ANNALS OF ALLERGY, ASTHMA, & IMMUNOLOGY

and not from systemic activity. Am JRespir Crit Care Med 1997;156:744–751.

11. Peden DB, Berger WE, Noonan MJ, etal. Inhaled fluticasone propionate de-livered by means of two different pow-der inhalers is effective and safe in alarge pediatric population with persis-tent asthma. J Allergy Clin Immunol1998;102(1):32–38.

12. Allen DB, Bronsky EA, LaForce CF,et al. Growth in asthmatic childrentreated with fluticasone propionate.J Pediatr 1998;132(3, Part 1):472–477.

13. American Thoracic Society. Standardsfor the diagnosis and care of patientswith chronic obstructive pulmonarydisease (COPD) and asthma. Am RevRespir Dis 1987;136:225–244.

14. Polgar G, Promadhat V. Pulmonaryfunction testing in children: techniquesand standards. Philadelphia: WB Saun-ders, 1971:12–17.

15. Tietz, NW (ed). Clinical guide to lab-oratory tests. 3rd ed. Philadelphia: WB Saunders Company, 1995:174.

16. Van Elteren PH. On the combinationof independent two sample tests of

Wilcoxon. Bull Inst Intern Statist 37:351–361.

17. Munch EP, Laursen LC, Dirksen A, etal. Dose frequency in the treatment ofasthmatics with inhaled topical ste-roids. Comparison between a twice-daily and a once-daily dosing regimen.Eur J Respir Dis 1985;67(4):254–260.

18. McGivern DVC, Ward M, MacFarlaneJT, et al. Failure of once daily inhaledcorticosteroid treatment to controlchronic asthma. Thorax 1984;39:933–934.

19. ZuWallach RL, Rosen JP, Cohen L, etal. The effectiveness of once-dailydosing of inhaled flunisolide in main-taining asthma control. J Allergy ClinImmunol 1997;99(3):278–285.

20. Gagnon M, Cote J, Milot J, et al. Com-parative safety and efficacy of singleor twice-daily administration of in-haled beclomethasone in moderateasthma. Chest 1994;105:1732–1737.

21. Stiksa G, Glennow C. Once daily in-halation of budesonide in the treatmentof chronic asthma: a clinical compari-son. Ann Allergy 1985;55:49–51.

22. Weiner P, Weiner M, Azgad Y. Long

term clinical comparison of single ver-sus twice daily administration of in-haled budesonide in moderate asthma.Thorax 1995;50(12):1270–1273.

23. Pincus DJ, Szefler SJ, Ackerson LM,Martin RJ. Chronotherapy of asthmawith inhaled steroids: the effect of dos-age timing on drug efficacy. J AllergyClin Immunol 1995;95(6):1172–1178.

24. McFadden ER, Casale TB, EdwardsTB, Kemp JP, et al. Administration ofbudesonide once daily by means ofTurbuhaler to subjects with stableasthma. J Allergy Clin Immunol 1995;104:46–52.

25. Hodges I, Netherway T. Once dailyinhaled fluticasone propionate is as ef-fective as twice daily in children withstable mild to moderate asthma. [Ab-stract] Thorax 1998;53(Suppl 4).

Request for reprints should be addressed to:Craig F LaForce, MDNorth Carolina Clinical ResearchRexwood Office Center, Suite 309A4301 Lake Boone TrailRaleigh, NC 27607email: claforce@NCCR.com

VOLUME 85, NOVEMBER, 2000 415