- Email: [email protected]
Multicenter study of flunisolide aerosol in adult patients with steroid-dependent asthma
Multicenter study of flunisolide aerosol in adult patients with steroid-dependent asthma Raymond G. Slavin, M.D., Allen E. Izu, M.S., I. Leonard Bernstein, M.D., Malcolm N. Blumenthal, M.D., Joseph F. Bolin, M.D., John J. Ouellette, M.D., Charles E. Reed, M.D., and Joseph Oren, M.D. St. Louis, MO., Cincinnati, Ohio, Minneapolis, Minn., Madison, Wk., and Palo Alto, Calif.
Seventy-three adult, steroid-dependent asthmatic patients pctrticiputed in a 16wk, double-blind study testing the eficacy of Jlunisolide aerosol. Forty receit ed j!unisolide, and 33 received placebo. The mean daily prednisone requirement of patients receiving Junisolide jell 59.2% during the testing period, and that of the patients receiving placebo fell 19.770. The median daily prednisone dose dropped 74.4 % in the flunisobde group and 4.2 %I in rhe placebo group (p = 0.006). In the flunisolide group 75% tapered use ~j’orol steroids 50% or more, und 27.5% stopped taking oral steroids completely. In the placebo group 36% tapered use of oral steroids 50% or more, and only 12% stopped taking them completely. Despite their reduction in systemic steroids, those patients receiving Junisolide achieved significantly greater reduction in the daily severity of wheezing (p = 0.014) andfrequency of asthma attacks (p = 0.049) than did those receiving placebo. In thejinal evaluation qf therapeutic response, 70% of patients receiving jlunisolide were rated as having a very good or good response, and 3070 were rated as having a jair or poor response. In contrast 33% of patients receiving placebo were rated as very good or good, and 67% were rated as fair or poor (p = 0.0009). No serious reactions were reported. Plasma cortisols showed an average increase of 42.9% in the jhrnisohde group but no change in the placebo group. Flunisolide uerosol is a well-tolerated and ejjective agent in the treatment of steroid-dependent asthma.
Systemic corticosteroids have been shown to be extremely effective in the managementof bronchial asthma. However, their usefulness is temperedby a high incidence of undesirable side effects. To avoid these problems research has centered on finding a meansof administering corticosteroidstopically to the lung and utilizing compounds that are rapidly inactivated, thereby minimizing systemic absorption. Early studies with aerosolized dexamethasone, however, were disappointing. The hypothalamic-
Fromthe Departments of InternalMedicine,
School of Medicine of St. Louis University, University of Cincinnati, University of Minnesota, University of Wisconsin, and Syntex Research, Palo Alto. Received for publication Jan. 28, 1980. Accepted for publication May 23, 1980. Reprint requests to: Raymond G. Slavin, M.D., Section of Allergy and Immunology, St. Louis University Medical Center, 1402 S. Grand Blvd., St. Louis, MO 63104. 0091-6749/80/110379+07$00.70/0
pituitary-adrenal (HPA) suppressionseenwith inhaled dexamethasonewas comparableto that seenwith systemic administration. r Later studies, however, with two other inhaled corticosteroids,beclomethasonediproprionate and triamcinolone acetonide, have suggestedthat these agentscause significantly less HPA suppression than do oral corticosteroids while still providing good control of asthma symptoms.2-X Plunisolide is a new synthetic steroid analogue. It is closely related to fluocinolone acetonide (Synalar), which is a highly potent topical corticosteroid, as demonstrated in dermal vasoconstrictor assays.gIn humans flunisolide is metabolized in the first pass through the liver into a relatively inactive metabolite. This rapid metabolism of flunisolide may contribute to its apparentlack of systemic effects. Previous studies of intranasal flunisolide have shown it to be effective and well tolerated in patients with perennial or seasonalrhinitis. lo-l2 Thesepatients have shown no signs of adrenal suppressionas mea-
0 1980 The C. V. Mosby Co.
Vol. 66, No. 5, pp. 379-395
J. ALLERGY
380 Slavin et al
0
, 0
I 2
I 4
I 6
I 10
I 8
I 12
CLIN. IMMUNOL. NOVEMBER 1980
I 14
I 16
WEEK
FIG. 1. Change in median prednisone dose during the study. (Bars represent tl about the median.)
sured by plasma cortisol levels. Further studies of flunisolide as an aerosol for the treatment of patients with asthma have similarly documented its efficacy in these patients and have demonstrated no suppression of cortisol levels.‘“-l6 The purpose of this multicenter trial was to determine the efficacy and safety of flunisolide in patients dependent on daily or alternate-day corticosteroids to control their asthma. The main criterion of efficacy was the ability to reduce the patients’ systemic corticosteroids without causing clinical deterioration of their asthma. MATERIALS AND METHODS Patients A total of 78 patients between the ages of 15 and 77 yr from four centers met the following criteria: (1) had obstructive airway disease known to be reversible within 6 mo of enrollment in the study; (2) required maintenance prednisone therapy of at least 10 mg but not more than 40 mg every other day for at least 3 mo prior to enrollment in the study; (3) were post pubertal; (4) were not taking any other investigational drugs; (5) were not pregnant, nursing, or planning to become pregnant during the study; and (6) did not have uncontrolled congestive heart failure, severe hypertension. diabetes mellitus, active tuberculosis, or any other coexisting or recent illness that might increase risks or confound study results.
standard error
Of these 78 patients 41 were randomly assigned to treatment with flunisolide, and 37 were assigned to treatment with placebo. Five patients (one receiving flunisolide and four receiving placebo) withdrew before the end of the study, and efficacy was therefore evaluated based on data from 73 patients The two groups were comparable in demographic and disease characteristics. All patients had chronic asthma; the mean duration of disease was 15 yr in the flunisolide group and 14 yr in the placebo group. Extrinsic (allergic) and intrinsic (nonallergic) subjects were the same, and the incidence of other allergic conditions (rhinitis, eczema, and drug allergies) was comparable in the two groups. Subjects were classified as extrinsic or intrinsic by the usual criteria of history and immediate skin test reactivity. Corticosteroids had been used orally continuously for an average of more than 4 yr in both groups.
Conduct
of study
This randomized, double-blind parallel design study lasted 16 wk. At the first visit (prebaseline) patients underwent a complete history and physical examination, with emphasis on allergic and asthmatic history, medication history, and respiratory tract examination. Spirometric measurements (forced vital capacity [FVC], forced expiratory volume in 1 set [FEV,], forced expiratory flow in 25 to 75 set [FEFZs-,J) were performed before and after inhalation of a standard dose of isoproteronol. Patients were instructed
VOLUME NUMBER
Flunisolide aerosol in asthma
66 5
381
i?Lee%
25.49% 1.24%
p=o.#6
FIG. 2. Percent reduction of oral corticosteroid dose from baseline to end of study. in how to keep detailed diaries of medication usage and symptom scores and told to return every 2 wk for reexamination. Bronchodilator usage remained constant throughout the study. The first 2 wk served as a baseline assessment phase. Dosage of oral corticosteroids remained constant. This period was used to stabilize each patient’s bronchodilator regimen, to assess lung function, and to accustom the patient to keeping his daily “diary” of medication usage and symptom scores. Patients then entered the randomized, double-blind treatment phase of the trial. During this 14-wk period patients received either flunisolide aerosol or placebo. Both were supplied as menthol and saccharine-flavored micronized powder propelled by volatile halogenated hydrocarbons (Freons). Inert sucrose octa-acetate replaced flunisolide in the placebo canisters. Dosage for both groups of patients was four puffs of aerosol twice daily (morning and evening). Each puff of the flunisolide canister delivered a metered dose of 0.125 mg; thus the total daily dose of flunisolide was 1 mg. Prednisone and bronchodilators continued to be supplied as needed. The purpose of this phase of the trial was to achieve maximum reduction of corticosteroid without appreciably reducing ventilatory function or causing a significant increase in asthma symptoms. To minimize symptoms of steroid withdrawal the rate at which prednisone could be reduced was limited by a predetermined schedule. (If the starting altenate-day dose was 20 to 40, 6 to 19, or 2.5 to 5 mg, dosage could be reduced by 10, 5, or 2.5 mg, respectively, every 2 wk.) Throughout this 14-wk period patients returned for biweekly visits, at which their interim history, drug usage and side effects, medication diary, complaints, and physical examination results were reviewed. Asthma frequency and severity were scored, and spirometry was repeated. Oral steroid tapering proceeded as symptomatology permitted.
At the patients’ final visit (visit 9) the physical examination was repeated, and a final evaluation including pulmonary function testing was done to assess their therapeutic response to the test drug. This assessment took into account all the clinical information that had accrued during the study, including adverse reactions but not including steroid tapering results. In addition to the evaluations mentioned above, ophthalmologic and chest x-ray examinations were carried out at visits 2 and 9; hemograms, blood chemistries, urinalyses, and cultures of the oropharynx for bacteria and fungi were done at visits 2, 5, and 9; and early morning plasma cortisol levels were determined at visits 2, 3, 8, and 9 (Murphy’s method was used to determine cortisol values”).
Statistical
analysis
Student’s t test and Fisher’s exact test were used for statistical analysis of differences between flunisolide and placebo patient groups, using demographic data obtained at admission and efficacy and safety data obtained during baseline. Repeated-measures analysis of variance incorporating the method of orthogonal contrasts was used to compare the treatments with respect to average changes from baseline over the study and with respect to differences in linear trends during the study. The nonparametric Mann-Whitney U test was used to compare the two treatments with respect to percent differences from baseline to the end of 14 wk of treatment. The Mann-Whitney U test was also used in instances where changes in an efficacy parameter for a few patients might lead to artifactual significances using the parametric repeated-measures analysis.
RESULTS Corticosteroid
reduction
The mean daily prednisone dose required by patients receiving flunisolide decreased by 9 mglday
382
Slavin
et al.
FIG. 3. Comparison placebo groups.
J. ALLERGY
of overall
evaluation
of therapeutic
(59.2%), from a baseline of 15.2 to a final dose of 6.2 mg/day. The mean prednisone requirement for patients receiving placebo decreased only slightly by 1.5 mg/day (19.7%), from 12.7 at baseline to 10.2 mg/ day. The differences between groups is even more striking when the decreases in the median prednisone dosages are compared (Fig. 1). For the flunisolide group the median prednisone requirement decreased 74.4%, from 12.5 mg at baseline to only 3.2 mg/day at the final visit. The comparable placebo group dosage decreased from 10 to 9.6 mg/day, or only 4.2%. This difference is highly statistically significant in favor of flunisolide (p = 0.001, by repeated-measures test using average changes from baseline over the entire study). The percentages of patients achieving various degrees of systemic corticosteroid reduction are presented in Fig. 2. Complete withdrawal from oral steroids was achieved by 11 of 40 patients receiving flunisolide (27.5%) but only four of 33 patients receiving placebo (12.1%). Reduction of 50% or more was achieved by an additional 47.5% of patients receiving flunisolide versus 24.2% of those receiving placebo. Conversely eight of 40 patients receiving flunisolide (20%) were refractory to tapering, compared with 14 of 33 receiving placebo (42.4%). These differences in steroid reduction are highly significant in favor of flunisolide (p = 0.006). Analyses of patient subgroups utilized the Mann-
response
between
flunisolide
CLIN. IMMUNOL NOVEMBER 1980
and
Whitney U test to evaluate percentage changes from baseline to the end of 14 wk of treatment. These analyses showed that, regardless of baseline steroid dose or type of asthma, those patients receiving flunisolide were able to achieve a greater reduction in oral corticosteroids than those receiving placebo. The difference in tapering was highly significant in the subgroup of patients taking daily oral corticosteroids (p = 0.029), and this difference was even greater for those whose daily baseline dose was even greater than 15 mg/day (p = 0.009). Patients whose baseline dose was less than 15 mglday and those who were on an alternate-day schedule were also able to achieve greater reduction with flunisolide; however, these differences were not statistically significant (p = 0.102 and p = 0.145, respectively). This may have been a consequence of less severe disease in these patients. For the subgroup of patients with intrinsic asthma the advantage of flunisolide treatment was highly significant (p = 0.017). The number of patients with extrinsic asthma was too small to draw statistical conclusions, but even here prednisone tapering was more successful in patients receiving flunisolide. Clinical
evaluation
Analysis of FEV, and FEFZj-rj revealed that the two treatment groups did not differ significantly in either spirometric measurement at baseline, and no significant change was noted during treatment. It is noteworthy that there was no ventilatory deterioration
VOLUME NUMBER
Flunisolide
66 5
in the flunisolide group, despite significant reduction of steroid doses. Clinical symptomatology, as exemplified by wheezing severity, did vary with treatment. Despite steroid tapering, patients receiving flunisolide improved from baseline during therapy, whereas those receiving placebo did not. The overall change is highly significant in favor of the flunisolide group (p = 0.014). There was also a significant difference favoring flunisolide in the average number of asthma attacks (p = 0.049), which decreased by 41% in patients receiving flunisolide compared with 12% in patients receiving placebo. (Differences were analyzed utilizing a repeated-measures test of average changes from baseline to the end of 14 wk of treatment.) For other variables recorded in patient daily diaries (chest tightness, cough, and shortness of breath) the treatment group means follow a pattern similar to that seen for wheezing severity. There was a consistent trend toward symptom improvement with flunisolide treatment, although differences between the flunisolide and placebo groups did not reach statistical significance. At each biweekly visit the physician rated the patient’s asthma severity on a four-point scale (1, asymptomatic . . . 4, constant). Analyses of these ratings demonstrated a significant difference between the flunisolide and placebo groups, with the flunisolide group experiencing less frequent symptoms of asthma (p = 0.017, repeated-measures test using average changes from baseline to end of 14 wk of treatment), Of particular interest was the finding that the number of asymptomatic patients in the flunisolide group increased from one at baseline to 12 at the end of the study. In the placebo group the number of asymptomatic patients increased from two to three and never exceeded six during the study. Asthma severity was rated on a six-point scale (1, asymptomatic . . 6, very severe). Severity tended to decrease in both treatment groups, and although this trend was greater in patients receiving flunisolide, the difference did not reach statistical significance. At the final study visit the patients and investigators made a joint overall evaluation of the therapeutic response to the 14-wk double-blind test drug period. The patient’s response was scored as “very good,” “good, ” “fair,” or “poor.” The results are illustrated in Fig. 3. These results are highly statistically significant (p = 0.0009, Mann-Whitney U test) in favor of flunisolide. Although the response of 70% of patients receiving flunisolide was rated as “very good ” or “good,’ ’ only 33% of those receiving placebo were so rated.
aerosol
in asthma
383
Safety In all, 39% of the patients receiving flunisolide and 32% of those receiving placebo reported some side effects, none of which were considered serious. Complaints recorded most frequently involved the respiratory tract and were consistent with the symptoms of underlying allergic and asthmatic conditions and with the irritative properties of inhaled powders. Those thought to be related to the test drug were more frequent in the placebo group and included respiratory complaints and nausea. Upper respiratory tract infection and nasal congestion were reported more frequently by a greater number of patients receiving flunisolide, whereas asthma-related complaints were more frequent in the placebo group. Other organ system complaints were comparably distributed between the treatment groups. There were no ophthalmologic abnormalities related to the 14 wk of treatment when pre- and poststudy eye examinations were compared. At baseline patients receiving flunisolide had a greater incidence of positive cultures for Candida albicans (41.5% versus 27%), although only one patient in each group had an oropharyngeal abnormality as determined by clinical evaluation. There were some increases in both positive cultures and positive clinical findings at interim evaluations, but the incidence was similar in the two groups at the end of the study. Of greatest interest was the association of positive clinical findings with positive cultures. One patient receiving flunisolide and none receiving placebo had such an association at the beginning of the study. During the study seven receiving flunisolide (17%) and three receiving placebo (8%) had both positive findings and cultures. Yet by the final evaluation there was a net increase from baseline of only one patient in each treatment group. Thus the incidence of confirmed clinical moniliasis was generally low. There were more conversions among patients receiving flunisolide, but they were apparently not persistent in most cases despite continued therapy. Comparative mean values for plasma cortisol determinations done at baseline and at the end of treatment are available for 30 patients receiving flunisolide and 29 receiving placebo. Values were not obtained for 19 patients, because of missing specimens, loss of patients to follow-up, or because samples were drawn in some cases after the specified time. Values less than 5 pgldl are considered below the laboratory normal range and are referred to as “suppressed.” After 14 wk of treatment the mean plasma cortisol level of the flunisolide group increased 2.7 pgldl (42.9%), from 6.3 pug/d1 at baseline to 8.9 pgldl,
384
Slavin et al.
whereas the mean level of the placebo group was unchanged at 8.2 pgldl. This difference approached statistical significance (p = 0.075). It is interesting to note that the flunisolide group achieved an increase in plasma cortisol level even though the mean baseline value was lower than that of the placebo group (6.3 pg/dl compared with 8.2 pg/dl). Of 23 patients suppressed at baseline, nine of 13 receiving flunisolide and eight of 10 receiving placebo were no longer suppressed at the end of treatment. Conversely, of patients with normal cortisol levels at baseline, 14 of 17 receiving flunisolide and 14 of 19 receiving placebo had normal levels at the end of the study. There was no significant difference between groups. There were no clinically meaningful abnormalities in any of the other laboratory values monitored during the study. DISCUSSION The treatment of patients with asthma often creates a series of difficult decisions in which the physician must weigh the risks of corticosteroid therapy against the known benefits. Minimizing patients’ exposure to the systemic complications of corticosteroids is a constant therapeutic goal. In this 14-wk double-blind study steroid-dependent asthmatic patients treated with flunisolide aerosol were able to significantly reduce their median dose of oral corticosteroids by 9.3 mglday (74%). By comparison steroid reduction in the parallel placebo group was minimal, suggesting that these patients were indeed steroid-dependent and were receiving essentially irreducible prednisone doses at baseline. Results were analyzed to determine whether any patient characteristics could be used to identify those most likely to be able to reduce their oral corticosteroid dose. However, there did not appear to be any significant relationship between the amount of steroid tapering achieved and either the baseline corticosteroid dose or the type of asthma. Those patients on a dose of greater than 15 mglday did achieve greater reduction of oral corticosteroids than those receiving less than 15 mglday or those who were on any alternate-day regimen. However, even those on lower baseline doses were able to achieve greater steroid tapering with flunisolide than with placebo. Similarly, both those patients with intrinsic and with extrinsic asthma -achieved significantly greater reduction in their corticosteroid dose when receiving flunisolide. Despite reduction of the oral corticosteroid dose during flunisolide treatment, these patients did not suffer symptomatic deterioration; rather, they tended to improve. Objectively their FEV, and FEFZsmT5remained unchanged. Subjectively, however, patients
J. ALLERGY
CLIN. IMMUNOL. NOVEMBER 1980
treated with flunisolide aerosol had a far superior overall clinical response to treatment than did patients given placebo. The clinical response was rated as “very good” or “good” in 70% of the flunisolide group, compared with only 30% in the placebo group. It is difficult to determine dose equivalents of inhaled and oral corticosteroids on the basis of our results, especially in view of the large variation in response among individual asthmatic patients. However, it does appear that in this group of patients with severe asthma, flunisolide aerosol (1 mg/day) provided superior symptomatic disease control while replacing an average of 9 mg/day of oral prednisone. This was accomplished with no serious local or systemic adverse effects during the 14 wk of treatment. There has been some recent debate about whether inhaled steroids do in fact reduce the risk of HPA suppression.lx Although it is clear that not all patients will show complete recovery of adrenal function,“, lg. 20a significant number of studies, including ours, have shown that plasma cortisol levels increase in most patients when inhaled corticosteroids are used to reduce the dose of oral corticosteroids.*-x, “* ” Not only do plasma cortisol levels increase, but other side effects of systemic therapy (e.g., moon facies, bruising, acne) decrease. In addition signs of steroid withdrawal such as rhinitis and eczema are evident in many of these patients. This evidence strongly suggests that inhaled cotticosteroid therapy minimizes systemic absorption and presents less of a risk of side effects (including HPA suppression) than does oral therapy. In this study flunisolide aerosol was found to be safe and effective in the treatment of patients with asthma who require corticosteroids for disease control. Because it need only be administered twice a day, it offers some advantage over other available inhaled corticosteroids that require a four-times-a-day regimen. Patient compliance, particularly among children, has been shown to improve with less frequent dosing. Flunisolide aerosol therefore appears to be an effective and well-tolerated alternative to oral corticosteroids in patients with asthma. We thank Diane Feldman for assistancein writing the manuscript. REFERENCES 1. Sigel SC, Heimlich EM, Richards W, Kelly VC: Adrenal function in allergy. IV. Effect of dexamethasone aerosols in asthmatic children. Pediatrics 33245, 1964. 2. Sly RM: Treatment of asthma in children with triamcinolone acetonide aerosol. J ALLERGYCLIN IMMUNOL62:76, 1978. 3. Godfrey S: Beclomethasone dipropionate aerosol in childhood asthma. Arch Dis Child 48:665, 1963. 4. Kass I, Nair SV, Patil KD: Beclomethasone dipropionate
VOLUME NUMBER
5.
6. 7.
8.
9.
10. 1I.
12.
13.
66 5
aerosol in the treatment of steroid-dependent asthmatic patients. Chest 71:703, 1977. Mabcrly DJ, Gibson GJ, Butler AG: Recovery of adrenal function after substitution of beclomethasone dipropionate for oral corticosteroids. Br Med J 1:778, 1973. Chervinsky P: Treatment of steroid-dependent asthma with triamcinolone acetonide aerosol. Ann Allergy 38: 192, 1977. British Thoracic and Tuberculosis Association: Inhaled corticosteroids compared with oral prednisone in patients starting long-term corticosteroid therapy for asthma. Lancet 2:469, 1975. Davies G, Thomas P, Broder I, Mintz S, Silverman F, Leznoff A, Trotman C: Steroid-dependent asthma treated with inhaled beclomethasone dipropionate. Ann Intern Med 86:549, 1977. Strand LJ, Flatman CJ, Northway CS, Evans CD: Comparison of Synemol cream and other corticosteroid creams using the vasoconstrictor bioassay. Cutis 19:689, 1977. Horan JD, Johnson JD: Flunisolide nasal spray in the treatment of perennial rhinitis. Can Med Assoc J 119:334, 1978. Schulz JI, Johnson JD, Freedman SO: Double-blind trial comparing Runisolide and placebo for the treatment of perennial rhinitis. Clin Allergy 8:313, 1979. Turkeltaub PC, Norman PS, Crepea S: Treatment of ragweed hay fever with an intranasal spray containing flunisolide, a new synthetic corticosteroid. J ALLERGYCLIN IMMUNOL58:597, 1976. Lowell FC, Ohman JL, Williams M: Double-blind trial of inhaled flunisolide in bronchial asthma. J ALLERGYCLIN IMMUNOL57~257. 1976.
Flunisolide
aerosol
in asthma
385
14. Zenz H, Lushkin A, Kentor P: Efficacy and safety of flunislalide aerosol in bronchial asthma. Ann Allergy 39:70, 1977. 15. Spangler DL, Bloom FL, Brestel EP, Wittig HJ: A one-year trial of aerosolized flunisolide in severe steroid-dependent asthmatics. Ann Allergy 3970, 1977. 16. Webb DR. Mullarkey MF, Freeman Ml: Flunisolide in chronic bronchial asthma. Ann Allergy 42:80, 1979. 17. Murphy BEP: Some studies of the protein-binding of steroids and their application to the routine micro and ultra-micromeasurement of various steroids in body fluids by competitive protein-binding radioassay. J Clin Endocrinol Metab 27:973, 1967. 18. Wyatt R, Waschek J, Weinberger M, Scherman B: Effects of inhaled beclomethasone depropionate and alternate-day prednisone on pituitary-adrenal function in children with chronic asthma. N Engl J Med 229:1387, 1978. 19. Kriz P: A one-year trial of triamcinolone acetonide aerosol in severe steroid-dependent asthma. Chest 72:36, 1977. 20. Vogt F, Chervinsky P, Dwek J, Grieco M: Beclomethasone dipropionate aerosol in the treatment of chronic bronchial asthma. J ALLERGYCLIN IMMUNOL58:316, 1976. 21. Falliers C: Triamcinolone acetonide aerosols for asthma. I. Effective replacement of systemic corticosteroid therapy. J ALLERGYCLIN IMMUNOL57:1, 1975. 22. British Thoracic and Tuberculosis Association: A controlled trial of inhaled corticosteroids in patients receiving prednisone tablets for asthma. Br J Dis Chest 70:95, 1976.
Seventy-three adult, steroid-dependent asthmatic patients pctrticiputed in a 16wk, double-blind study testing the eficacy of Jlunisolide aerosol. Forty receit ed j!unisolide, and 33 received placebo. The mean daily prednisone requirement of patients receiving Junisolide jell 59.2% during the testing period, and that of the patients receiving placebo fell 19.770. The median daily prednisone dose dropped 74.4 % in the flunisobde group and 4.2 %I in rhe placebo group (p = 0.006). In the flunisolide group 75% tapered use ~j’orol steroids 50% or more, und 27.5% stopped taking oral steroids completely. In the placebo group 36% tapered use of oral steroids 50% or more, and only 12% stopped taking them completely. Despite their reduction in systemic steroids, those patients receiving Junisolide achieved significantly greater reduction in the daily severity of wheezing (p = 0.014) andfrequency of asthma attacks (p = 0.049) than did those receiving placebo. In thejinal evaluation qf therapeutic response, 70% of patients receiving jlunisolide were rated as having a very good or good response, and 3070 were rated as having a jair or poor response. In contrast 33% of patients receiving placebo were rated as very good or good, and 67% were rated as fair or poor (p = 0.0009). No serious reactions were reported. Plasma cortisols showed an average increase of 42.9% in the jhrnisohde group but no change in the placebo group. Flunisolide uerosol is a well-tolerated and ejjective agent in the treatment of steroid-dependent asthma.
Systemic corticosteroids have been shown to be extremely effective in the managementof bronchial asthma. However, their usefulness is temperedby a high incidence of undesirable side effects. To avoid these problems research has centered on finding a meansof administering corticosteroidstopically to the lung and utilizing compounds that are rapidly inactivated, thereby minimizing systemic absorption. Early studies with aerosolized dexamethasone, however, were disappointing. The hypothalamic-
Fromthe Departments of InternalMedicine,
School of Medicine of St. Louis University, University of Cincinnati, University of Minnesota, University of Wisconsin, and Syntex Research, Palo Alto. Received for publication Jan. 28, 1980. Accepted for publication May 23, 1980. Reprint requests to: Raymond G. Slavin, M.D., Section of Allergy and Immunology, St. Louis University Medical Center, 1402 S. Grand Blvd., St. Louis, MO 63104. 0091-6749/80/110379+07$00.70/0
pituitary-adrenal (HPA) suppressionseenwith inhaled dexamethasonewas comparableto that seenwith systemic administration. r Later studies, however, with two other inhaled corticosteroids,beclomethasonediproprionate and triamcinolone acetonide, have suggestedthat these agentscause significantly less HPA suppression than do oral corticosteroids while still providing good control of asthma symptoms.2-X Plunisolide is a new synthetic steroid analogue. It is closely related to fluocinolone acetonide (Synalar), which is a highly potent topical corticosteroid, as demonstrated in dermal vasoconstrictor assays.gIn humans flunisolide is metabolized in the first pass through the liver into a relatively inactive metabolite. This rapid metabolism of flunisolide may contribute to its apparentlack of systemic effects. Previous studies of intranasal flunisolide have shown it to be effective and well tolerated in patients with perennial or seasonalrhinitis. lo-l2 Thesepatients have shown no signs of adrenal suppressionas mea-
0 1980 The C. V. Mosby Co.
Vol. 66, No. 5, pp. 379-395
J. ALLERGY
380 Slavin et al
0
, 0
I 2
I 4
I 6
I 10
I 8
I 12
CLIN. IMMUNOL. NOVEMBER 1980
I 14
I 16
WEEK
FIG. 1. Change in median prednisone dose during the study. (Bars represent tl about the median.)
sured by plasma cortisol levels. Further studies of flunisolide as an aerosol for the treatment of patients with asthma have similarly documented its efficacy in these patients and have demonstrated no suppression of cortisol levels.‘“-l6 The purpose of this multicenter trial was to determine the efficacy and safety of flunisolide in patients dependent on daily or alternate-day corticosteroids to control their asthma. The main criterion of efficacy was the ability to reduce the patients’ systemic corticosteroids without causing clinical deterioration of their asthma. MATERIALS AND METHODS Patients A total of 78 patients between the ages of 15 and 77 yr from four centers met the following criteria: (1) had obstructive airway disease known to be reversible within 6 mo of enrollment in the study; (2) required maintenance prednisone therapy of at least 10 mg but not more than 40 mg every other day for at least 3 mo prior to enrollment in the study; (3) were post pubertal; (4) were not taking any other investigational drugs; (5) were not pregnant, nursing, or planning to become pregnant during the study; and (6) did not have uncontrolled congestive heart failure, severe hypertension. diabetes mellitus, active tuberculosis, or any other coexisting or recent illness that might increase risks or confound study results.
standard error
Of these 78 patients 41 were randomly assigned to treatment with flunisolide, and 37 were assigned to treatment with placebo. Five patients (one receiving flunisolide and four receiving placebo) withdrew before the end of the study, and efficacy was therefore evaluated based on data from 73 patients The two groups were comparable in demographic and disease characteristics. All patients had chronic asthma; the mean duration of disease was 15 yr in the flunisolide group and 14 yr in the placebo group. Extrinsic (allergic) and intrinsic (nonallergic) subjects were the same, and the incidence of other allergic conditions (rhinitis, eczema, and drug allergies) was comparable in the two groups. Subjects were classified as extrinsic or intrinsic by the usual criteria of history and immediate skin test reactivity. Corticosteroids had been used orally continuously for an average of more than 4 yr in both groups.
Conduct
of study
This randomized, double-blind parallel design study lasted 16 wk. At the first visit (prebaseline) patients underwent a complete history and physical examination, with emphasis on allergic and asthmatic history, medication history, and respiratory tract examination. Spirometric measurements (forced vital capacity [FVC], forced expiratory volume in 1 set [FEV,], forced expiratory flow in 25 to 75 set [FEFZs-,J) were performed before and after inhalation of a standard dose of isoproteronol. Patients were instructed
VOLUME NUMBER
Flunisolide aerosol in asthma
66 5
381
i?Lee%
25.49% 1.24%
p=o.#6
FIG. 2. Percent reduction of oral corticosteroid dose from baseline to end of study. in how to keep detailed diaries of medication usage and symptom scores and told to return every 2 wk for reexamination. Bronchodilator usage remained constant throughout the study. The first 2 wk served as a baseline assessment phase. Dosage of oral corticosteroids remained constant. This period was used to stabilize each patient’s bronchodilator regimen, to assess lung function, and to accustom the patient to keeping his daily “diary” of medication usage and symptom scores. Patients then entered the randomized, double-blind treatment phase of the trial. During this 14-wk period patients received either flunisolide aerosol or placebo. Both were supplied as menthol and saccharine-flavored micronized powder propelled by volatile halogenated hydrocarbons (Freons). Inert sucrose octa-acetate replaced flunisolide in the placebo canisters. Dosage for both groups of patients was four puffs of aerosol twice daily (morning and evening). Each puff of the flunisolide canister delivered a metered dose of 0.125 mg; thus the total daily dose of flunisolide was 1 mg. Prednisone and bronchodilators continued to be supplied as needed. The purpose of this phase of the trial was to achieve maximum reduction of corticosteroid without appreciably reducing ventilatory function or causing a significant increase in asthma symptoms. To minimize symptoms of steroid withdrawal the rate at which prednisone could be reduced was limited by a predetermined schedule. (If the starting altenate-day dose was 20 to 40, 6 to 19, or 2.5 to 5 mg, dosage could be reduced by 10, 5, or 2.5 mg, respectively, every 2 wk.) Throughout this 14-wk period patients returned for biweekly visits, at which their interim history, drug usage and side effects, medication diary, complaints, and physical examination results were reviewed. Asthma frequency and severity were scored, and spirometry was repeated. Oral steroid tapering proceeded as symptomatology permitted.
At the patients’ final visit (visit 9) the physical examination was repeated, and a final evaluation including pulmonary function testing was done to assess their therapeutic response to the test drug. This assessment took into account all the clinical information that had accrued during the study, including adverse reactions but not including steroid tapering results. In addition to the evaluations mentioned above, ophthalmologic and chest x-ray examinations were carried out at visits 2 and 9; hemograms, blood chemistries, urinalyses, and cultures of the oropharynx for bacteria and fungi were done at visits 2, 5, and 9; and early morning plasma cortisol levels were determined at visits 2, 3, 8, and 9 (Murphy’s method was used to determine cortisol values”).
Statistical
analysis
Student’s t test and Fisher’s exact test were used for statistical analysis of differences between flunisolide and placebo patient groups, using demographic data obtained at admission and efficacy and safety data obtained during baseline. Repeated-measures analysis of variance incorporating the method of orthogonal contrasts was used to compare the treatments with respect to average changes from baseline over the study and with respect to differences in linear trends during the study. The nonparametric Mann-Whitney U test was used to compare the two treatments with respect to percent differences from baseline to the end of 14 wk of treatment. The Mann-Whitney U test was also used in instances where changes in an efficacy parameter for a few patients might lead to artifactual significances using the parametric repeated-measures analysis.
RESULTS Corticosteroid
reduction
The mean daily prednisone dose required by patients receiving flunisolide decreased by 9 mglday
382
Slavin
et al.
FIG. 3. Comparison placebo groups.
J. ALLERGY
of overall
evaluation
of therapeutic
(59.2%), from a baseline of 15.2 to a final dose of 6.2 mg/day. The mean prednisone requirement for patients receiving placebo decreased only slightly by 1.5 mg/day (19.7%), from 12.7 at baseline to 10.2 mg/ day. The differences between groups is even more striking when the decreases in the median prednisone dosages are compared (Fig. 1). For the flunisolide group the median prednisone requirement decreased 74.4%, from 12.5 mg at baseline to only 3.2 mg/day at the final visit. The comparable placebo group dosage decreased from 10 to 9.6 mg/day, or only 4.2%. This difference is highly statistically significant in favor of flunisolide (p = 0.001, by repeated-measures test using average changes from baseline over the entire study). The percentages of patients achieving various degrees of systemic corticosteroid reduction are presented in Fig. 2. Complete withdrawal from oral steroids was achieved by 11 of 40 patients receiving flunisolide (27.5%) but only four of 33 patients receiving placebo (12.1%). Reduction of 50% or more was achieved by an additional 47.5% of patients receiving flunisolide versus 24.2% of those receiving placebo. Conversely eight of 40 patients receiving flunisolide (20%) were refractory to tapering, compared with 14 of 33 receiving placebo (42.4%). These differences in steroid reduction are highly significant in favor of flunisolide (p = 0.006). Analyses of patient subgroups utilized the Mann-
response
between
flunisolide
CLIN. IMMUNOL NOVEMBER 1980
and
Whitney U test to evaluate percentage changes from baseline to the end of 14 wk of treatment. These analyses showed that, regardless of baseline steroid dose or type of asthma, those patients receiving flunisolide were able to achieve a greater reduction in oral corticosteroids than those receiving placebo. The difference in tapering was highly significant in the subgroup of patients taking daily oral corticosteroids (p = 0.029), and this difference was even greater for those whose daily baseline dose was even greater than 15 mg/day (p = 0.009). Patients whose baseline dose was less than 15 mglday and those who were on an alternate-day schedule were also able to achieve greater reduction with flunisolide; however, these differences were not statistically significant (p = 0.102 and p = 0.145, respectively). This may have been a consequence of less severe disease in these patients. For the subgroup of patients with intrinsic asthma the advantage of flunisolide treatment was highly significant (p = 0.017). The number of patients with extrinsic asthma was too small to draw statistical conclusions, but even here prednisone tapering was more successful in patients receiving flunisolide. Clinical
evaluation
Analysis of FEV, and FEFZj-rj revealed that the two treatment groups did not differ significantly in either spirometric measurement at baseline, and no significant change was noted during treatment. It is noteworthy that there was no ventilatory deterioration
VOLUME NUMBER
Flunisolide
66 5
in the flunisolide group, despite significant reduction of steroid doses. Clinical symptomatology, as exemplified by wheezing severity, did vary with treatment. Despite steroid tapering, patients receiving flunisolide improved from baseline during therapy, whereas those receiving placebo did not. The overall change is highly significant in favor of the flunisolide group (p = 0.014). There was also a significant difference favoring flunisolide in the average number of asthma attacks (p = 0.049), which decreased by 41% in patients receiving flunisolide compared with 12% in patients receiving placebo. (Differences were analyzed utilizing a repeated-measures test of average changes from baseline to the end of 14 wk of treatment.) For other variables recorded in patient daily diaries (chest tightness, cough, and shortness of breath) the treatment group means follow a pattern similar to that seen for wheezing severity. There was a consistent trend toward symptom improvement with flunisolide treatment, although differences between the flunisolide and placebo groups did not reach statistical significance. At each biweekly visit the physician rated the patient’s asthma severity on a four-point scale (1, asymptomatic . . . 4, constant). Analyses of these ratings demonstrated a significant difference between the flunisolide and placebo groups, with the flunisolide group experiencing less frequent symptoms of asthma (p = 0.017, repeated-measures test using average changes from baseline to end of 14 wk of treatment), Of particular interest was the finding that the number of asymptomatic patients in the flunisolide group increased from one at baseline to 12 at the end of the study. In the placebo group the number of asymptomatic patients increased from two to three and never exceeded six during the study. Asthma severity was rated on a six-point scale (1, asymptomatic . . 6, very severe). Severity tended to decrease in both treatment groups, and although this trend was greater in patients receiving flunisolide, the difference did not reach statistical significance. At the final study visit the patients and investigators made a joint overall evaluation of the therapeutic response to the 14-wk double-blind test drug period. The patient’s response was scored as “very good,” “good, ” “fair,” or “poor.” The results are illustrated in Fig. 3. These results are highly statistically significant (p = 0.0009, Mann-Whitney U test) in favor of flunisolide. Although the response of 70% of patients receiving flunisolide was rated as “very good ” or “good,’ ’ only 33% of those receiving placebo were so rated.
aerosol
in asthma
383
Safety In all, 39% of the patients receiving flunisolide and 32% of those receiving placebo reported some side effects, none of which were considered serious. Complaints recorded most frequently involved the respiratory tract and were consistent with the symptoms of underlying allergic and asthmatic conditions and with the irritative properties of inhaled powders. Those thought to be related to the test drug were more frequent in the placebo group and included respiratory complaints and nausea. Upper respiratory tract infection and nasal congestion were reported more frequently by a greater number of patients receiving flunisolide, whereas asthma-related complaints were more frequent in the placebo group. Other organ system complaints were comparably distributed between the treatment groups. There were no ophthalmologic abnormalities related to the 14 wk of treatment when pre- and poststudy eye examinations were compared. At baseline patients receiving flunisolide had a greater incidence of positive cultures for Candida albicans (41.5% versus 27%), although only one patient in each group had an oropharyngeal abnormality as determined by clinical evaluation. There were some increases in both positive cultures and positive clinical findings at interim evaluations, but the incidence was similar in the two groups at the end of the study. Of greatest interest was the association of positive clinical findings with positive cultures. One patient receiving flunisolide and none receiving placebo had such an association at the beginning of the study. During the study seven receiving flunisolide (17%) and three receiving placebo (8%) had both positive findings and cultures. Yet by the final evaluation there was a net increase from baseline of only one patient in each treatment group. Thus the incidence of confirmed clinical moniliasis was generally low. There were more conversions among patients receiving flunisolide, but they were apparently not persistent in most cases despite continued therapy. Comparative mean values for plasma cortisol determinations done at baseline and at the end of treatment are available for 30 patients receiving flunisolide and 29 receiving placebo. Values were not obtained for 19 patients, because of missing specimens, loss of patients to follow-up, or because samples were drawn in some cases after the specified time. Values less than 5 pgldl are considered below the laboratory normal range and are referred to as “suppressed.” After 14 wk of treatment the mean plasma cortisol level of the flunisolide group increased 2.7 pgldl (42.9%), from 6.3 pug/d1 at baseline to 8.9 pgldl,
384
Slavin et al.
whereas the mean level of the placebo group was unchanged at 8.2 pgldl. This difference approached statistical significance (p = 0.075). It is interesting to note that the flunisolide group achieved an increase in plasma cortisol level even though the mean baseline value was lower than that of the placebo group (6.3 pg/dl compared with 8.2 pg/dl). Of 23 patients suppressed at baseline, nine of 13 receiving flunisolide and eight of 10 receiving placebo were no longer suppressed at the end of treatment. Conversely, of patients with normal cortisol levels at baseline, 14 of 17 receiving flunisolide and 14 of 19 receiving placebo had normal levels at the end of the study. There was no significant difference between groups. There were no clinically meaningful abnormalities in any of the other laboratory values monitored during the study. DISCUSSION The treatment of patients with asthma often creates a series of difficult decisions in which the physician must weigh the risks of corticosteroid therapy against the known benefits. Minimizing patients’ exposure to the systemic complications of corticosteroids is a constant therapeutic goal. In this 14-wk double-blind study steroid-dependent asthmatic patients treated with flunisolide aerosol were able to significantly reduce their median dose of oral corticosteroids by 9.3 mglday (74%). By comparison steroid reduction in the parallel placebo group was minimal, suggesting that these patients were indeed steroid-dependent and were receiving essentially irreducible prednisone doses at baseline. Results were analyzed to determine whether any patient characteristics could be used to identify those most likely to be able to reduce their oral corticosteroid dose. However, there did not appear to be any significant relationship between the amount of steroid tapering achieved and either the baseline corticosteroid dose or the type of asthma. Those patients on a dose of greater than 15 mglday did achieve greater reduction of oral corticosteroids than those receiving less than 15 mglday or those who were on any alternate-day regimen. However, even those on lower baseline doses were able to achieve greater steroid tapering with flunisolide than with placebo. Similarly, both those patients with intrinsic and with extrinsic asthma -achieved significantly greater reduction in their corticosteroid dose when receiving flunisolide. Despite reduction of the oral corticosteroid dose during flunisolide treatment, these patients did not suffer symptomatic deterioration; rather, they tended to improve. Objectively their FEV, and FEFZsmT5remained unchanged. Subjectively, however, patients
J. ALLERGY
CLIN. IMMUNOL. NOVEMBER 1980
treated with flunisolide aerosol had a far superior overall clinical response to treatment than did patients given placebo. The clinical response was rated as “very good” or “good” in 70% of the flunisolide group, compared with only 30% in the placebo group. It is difficult to determine dose equivalents of inhaled and oral corticosteroids on the basis of our results, especially in view of the large variation in response among individual asthmatic patients. However, it does appear that in this group of patients with severe asthma, flunisolide aerosol (1 mg/day) provided superior symptomatic disease control while replacing an average of 9 mg/day of oral prednisone. This was accomplished with no serious local or systemic adverse effects during the 14 wk of treatment. There has been some recent debate about whether inhaled steroids do in fact reduce the risk of HPA suppression.lx Although it is clear that not all patients will show complete recovery of adrenal function,“, lg. 20a significant number of studies, including ours, have shown that plasma cortisol levels increase in most patients when inhaled corticosteroids are used to reduce the dose of oral corticosteroids.*-x, “* ” Not only do plasma cortisol levels increase, but other side effects of systemic therapy (e.g., moon facies, bruising, acne) decrease. In addition signs of steroid withdrawal such as rhinitis and eczema are evident in many of these patients. This evidence strongly suggests that inhaled cotticosteroid therapy minimizes systemic absorption and presents less of a risk of side effects (including HPA suppression) than does oral therapy. In this study flunisolide aerosol was found to be safe and effective in the treatment of patients with asthma who require corticosteroids for disease control. Because it need only be administered twice a day, it offers some advantage over other available inhaled corticosteroids that require a four-times-a-day regimen. Patient compliance, particularly among children, has been shown to improve with less frequent dosing. Flunisolide aerosol therefore appears to be an effective and well-tolerated alternative to oral corticosteroids in patients with asthma. We thank Diane Feldman for assistancein writing the manuscript. REFERENCES 1. Sigel SC, Heimlich EM, Richards W, Kelly VC: Adrenal function in allergy. IV. Effect of dexamethasone aerosols in asthmatic children. Pediatrics 33245, 1964. 2. Sly RM: Treatment of asthma in children with triamcinolone acetonide aerosol. J ALLERGYCLIN IMMUNOL62:76, 1978. 3. Godfrey S: Beclomethasone dipropionate aerosol in childhood asthma. Arch Dis Child 48:665, 1963. 4. Kass I, Nair SV, Patil KD: Beclomethasone dipropionate
VOLUME NUMBER
5.
6. 7.
8.
9.
10. 1I.
12.
13.
66 5
aerosol in the treatment of steroid-dependent asthmatic patients. Chest 71:703, 1977. Mabcrly DJ, Gibson GJ, Butler AG: Recovery of adrenal function after substitution of beclomethasone dipropionate for oral corticosteroids. Br Med J 1:778, 1973. Chervinsky P: Treatment of steroid-dependent asthma with triamcinolone acetonide aerosol. Ann Allergy 38: 192, 1977. British Thoracic and Tuberculosis Association: Inhaled corticosteroids compared with oral prednisone in patients starting long-term corticosteroid therapy for asthma. Lancet 2:469, 1975. Davies G, Thomas P, Broder I, Mintz S, Silverman F, Leznoff A, Trotman C: Steroid-dependent asthma treated with inhaled beclomethasone dipropionate. Ann Intern Med 86:549, 1977. Strand LJ, Flatman CJ, Northway CS, Evans CD: Comparison of Synemol cream and other corticosteroid creams using the vasoconstrictor bioassay. Cutis 19:689, 1977. Horan JD, Johnson JD: Flunisolide nasal spray in the treatment of perennial rhinitis. Can Med Assoc J 119:334, 1978. Schulz JI, Johnson JD, Freedman SO: Double-blind trial comparing Runisolide and placebo for the treatment of perennial rhinitis. Clin Allergy 8:313, 1979. Turkeltaub PC, Norman PS, Crepea S: Treatment of ragweed hay fever with an intranasal spray containing flunisolide, a new synthetic corticosteroid. J ALLERGYCLIN IMMUNOL58:597, 1976. Lowell FC, Ohman JL, Williams M: Double-blind trial of inhaled flunisolide in bronchial asthma. J ALLERGYCLIN IMMUNOL57~257. 1976.
Flunisolide
aerosol
in asthma
385
14. Zenz H, Lushkin A, Kentor P: Efficacy and safety of flunislalide aerosol in bronchial asthma. Ann Allergy 39:70, 1977. 15. Spangler DL, Bloom FL, Brestel EP, Wittig HJ: A one-year trial of aerosolized flunisolide in severe steroid-dependent asthmatics. Ann Allergy 3970, 1977. 16. Webb DR. Mullarkey MF, Freeman Ml: Flunisolide in chronic bronchial asthma. Ann Allergy 42:80, 1979. 17. Murphy BEP: Some studies of the protein-binding of steroids and their application to the routine micro and ultra-micromeasurement of various steroids in body fluids by competitive protein-binding radioassay. J Clin Endocrinol Metab 27:973, 1967. 18. Wyatt R, Waschek J, Weinberger M, Scherman B: Effects of inhaled beclomethasone depropionate and alternate-day prednisone on pituitary-adrenal function in children with chronic asthma. N Engl J Med 229:1387, 1978. 19. Kriz P: A one-year trial of triamcinolone acetonide aerosol in severe steroid-dependent asthma. Chest 72:36, 1977. 20. Vogt F, Chervinsky P, Dwek J, Grieco M: Beclomethasone dipropionate aerosol in the treatment of chronic bronchial asthma. J ALLERGYCLIN IMMUNOL58:316, 1976. 21. Falliers C: Triamcinolone acetonide aerosols for asthma. I. Effective replacement of systemic corticosteroid therapy. J ALLERGYCLIN IMMUNOL57:1, 1975. 22. British Thoracic and Tuberculosis Association: A controlled trial of inhaled corticosteroids in patients receiving prednisone tablets for asthma. Br J Dis Chest 70:95, 1976.