Investigation of Dose-Response of Exhaled Nitric Oxide After Oral Inhalation of Fluticasone

Supplementary Materials (SM)

Investigation of dose-response of exhaled nitric oxide after oral inhalation of fluticasone propionate

Objective: To investigate the effect of varying doses (44 mcg, 88 mcg, and 352 mcg BID) of fluticasone propionate hydrofluoroalkane (HFA) and treatment withdrawal on fractional exhaled nitric oxide (FeNO) levels in a crossover study design.

Materials and Methods

Materials

Flovent HFA 44 mcg Inhalation Aerosol was used for all active treatments. Placebo canisters (Exemplar Pharmaceuticals, Fall River, MA) were prepared with the same propellant, metering volume, and net weight as the active drug product. At the time of the study, Flovent HFA canisters were supplied with dose counters. Placebo canisters were manufactured without dose counters. In order to maintain subject blinding to the canisters, dose counters were removed from the active canisters. For blinding and in vitro performance considerations, Flovent HFA actuators were used for both the active and placebo inhalers. Removal of the dose counter from active canisters would allow airflow through the actuator window, potentially changing the airflow pattern during inhalation. To prevent this, the window in each actuator was covered with tape prior to use with the active or placebo canisters.

Study designs

This study is a single-blinded, 4-way crossover study in asthmatic patients. The study consisted of two phases, I and II. Subjects were recruited from the Weinberg Clinical Research Unit’s database through the National Jewish Health adult outpatient clinic.

Inclusion and exclusion criteria for Phase I:

Inclusion criteria: (1) mild to moderate asthma based on NAEPP guidelines for symptoms, (2) 12% and/or 200 mL improvement in FEV1(forced expiratory volume in 1 second) with albuterol, (3) FeNO ≥45 ppb (to ensure a high level of starting FeNO to assess decrease with dose), (4) methacholine challenge as a one-time measurement (no quantitative level required for study eligibility), (5) FEV1≥60% of predicted value (within the previous 3 months), and 6) age between 18-65 years.

Exclusion criteria: (1) severe asthma symptoms, (2) upper respiratory tract infection in the past 3 months, (3) pregnancy, (4) smoking within the last year or >10 pack-year smoking history, and (5) use of oral corticosteroids in the previous 3 months or nasal corticosteroids in the previous 2 weeks.

Phase I included a 16-day run-in period followed by a treatment period (for example, 88 mcg BID, i.e., 176 mcg/day) for 14 days, and concluded with a 14-day washout period. As FeNO values decrease over several days during treatment, a 14-day treatment period was chosen to allow sufficient time for FeNO levels to plateau. FeNO monitoring occurred on 3 separate days each week (generally Monday, Wednesday and Friday) through the run-in, treatment, and washout periods (SM-Figure 1). Patients with FeNO ≥45 ppb at the end of the run-in period were allowed to continue in the study. In addition to washing out any pre-study ICS, the purpose of the run-in period was to enrich the subject population and to allow subjects to become familiar with use of the inhaler device.

Inclusion criteria for Phase II

Subjects meeting the following criteria during Phase I were advanced to Phase II: (1) FeNO ≥45 ppb at the end of the run-in period, 2) at least a 25% reduction of baseline FeNO during the treatment period in Phase I, and 3) return to baseline (defined as within 50% of the observed maximum percentage drop from baseline to nadir) on two consecutive visits during the 14-day washout period of Phase I.

Subjects qualifying for Phase II were randomized to one of 12 sequences in a 4-way crossover design involving 4 treatment and 3 washout periods. The 176 mcg/day dose was replicated to estimate the intrasubject variability of FeNO. Both Phase I and II were conducted single-blinded (subject-blinded). A protocol deviation occurred in the study, in which Phase II was conducted subject-blinded rather than the protocol-specified double-blinded (i.e., subject- and investigator-blinded). Subjects were not permitted to take any other corticosteroids or long-acting beta-agonists during the course of the study.

Study treatments

There was one treatment in Phase I: Flovent HFA 176 mcg/day (two inhalations twice daily of Flovent HFA 88 mcg), placebo was given during run-in and washout periods.There were four treatments in Phase II: (1) 88 mcg/day; (2) and (3) 176 mcg/day; (4) 704 mcg/day in this study. In Phase II, subjects received a total of 10 inhalations BID from two inhalers, one containing active drug and one placebo. The 88 mcg/day dose was administered with 1 inhalation BID of active drug and 9 inhalations BID of placebo. The 176 mcg/day dose was administered with 2 inhalations BID of active drug and 8 inhalations BID of placebo. The 704 mcg/day dose was administered with 8 inhalations BID of active drug and 2 inhalations BID of placebo. During the washout periods, all subjects used 2 placebo inhalers to receive either 1 or 2 inhalations from one inhaler and 8 or 9 inhalations from the other to maintain blinding. All doses were administered from primed canisters with the actuator placed in the mouth.

Safety Monitoring

Subjects were provided with albuterol MDI as rescue medication for both Phases I and II. For the majority of the study, subjects were scheduled for regular study visits to the clinic on three separate days each week (generally Monday, Wednesday and Friday). Subjects were monitored for adverse events on a continual basis, and were given access to a 24-hour pager in case of need for any assistance.

Exhaled nitric oxide measurement

Measurements of FeNO were taken on three separate days each week through all treatment and washout periods in the clinic. The NIOX® (Aerocrine, New Providence, NJ) analyzer was used to obtain FeNO measurements, according to the ATS/ERS recommendations (1). Repeated measurements were taken at 50 mL/s expiratory flow rate until a minimum of two efforts which were within 10% or 2.5 ppb of each other were obtained. The average value from the measurements was used for statistical analysis.

Statistical analysis

All nine subjects who completed Phase II were included for data analyses. FeNO values were log-transformed prior to performing statistical analyses. Data from Days 3-16 were included in the analyses for the run-in period, and data from Days 3-14 were included in the analyses for all other periods. Data from the first two days were excluded to allow subjects time to respond to the drug. The last measurement taken during the run-in period of Phase I was considered as the baseline value of each subject for Phase II. A linear mixed-effects model was used to obtain estimates of response for each dose. A random intercept was included in the model for subjects and repeated measures over time were modeled using a spatial exponential covariance structure, as FeNO measurements were not always taken on consecutive days. Statistical analyses were performed using SAS software (version 9.2; SAS Institute; Cary, NC).

Results

Ninety-six subjects were screened, and 50 subjects gave consentto participation in the study. Of these, forty-five subjects signed the informed consent form and were enrolled in the study. The disposition of subjects is given in SM-Figure 2. Fourteen subjects (31%) did not meet the FeNO criteria (≥45 ppb), and 12 subjects did not meet other criteria for inclusion in Phase I. Nineteen subjects began Phase I run-in but only 9 subjects completed the study (through the end of Phase II). Demographic information, screening FEV1 and baseline FeNO for the 9 subjects who completed both phases of the study are shown in SM-Table 1.

An example subject profile of FeNO levels during Phase I and II is shown in SM-Figure 3. During periods of treatment with active drug, FeNO levels declined over time for all treatment doses and in all subjects. During the 14-day washout periods, FeNO levels generally increased over time. Overlay plots of individual FeNO-time profiles grouped by treatment dose are shown in SM-Figure 4A-D. A general trend toward a decline in FeNO values is seen with time but a plateau is not clearly evident for any of the treatment doses.

SM-Figure 5A-C shows overlay plots of individual FeNO-time profiles grouped by various placebo periods. FeNO levels varied widely among subjects throughout the run-in period (SM-Figure 5A). During washout periods in Phase I and II (SM-Figures 5B and C), a gradual rise in FeNO level was seen in most subjects through the 14-day period.

SM-Figure 6 shows a plot of the last FeNO measurement (expressed as % of the baseline value) taken during the washout period prior to each treatment period in Phase II. Most subjects did not return to FeNO values within 20% of baseline at the end of each washout period. In addition, the majority of subjects had FeNO values below baseline (<100%) at the end of each washout period, indicating that subjects did not return to near-baseline values of FeNO during the 14-day washout periods following the 14-day treatment periods of Phases I and II.

The least squares (LS) means and 95% confidence intervals of the FeNO values from Day 3-14 of all placebo periods (except for the Phase I run-in period which was from Day 3-16) are given in SM-Table 2. The LS means decreased over time. Differences between LS means were significant (p<0.05), when comparing the run-in period to each of the washout periods in Phase II, and when comparing the washout period in Phase I to each of the washout periods in Phase II. The LS means of the run-in and washout period of Phase I were not significantly different, and the LS means of the three washout periods in Phase II were not significantly different from each other.

Dose-response of FeNO was evaluated by comparing the LS means of the FeNO values from Days 3-14 at three doses and the baseline. The LS means decreased with increasing dose (SM-Table 3). All doses were significantly different (p<0.05) from the baseline (0 mcg dose), but were not significantly different from each other. The lack of dose-response at the subject population level was reflected in the lack of dose-response at the individual subject level (SM-Figure 7), as only 3 of 9 subjects showed a consistent decrease in the geometric mean of FeNO with increasing dose (Subjects 6, 23, 24). It should also be noted that even in these 3 subjects, the decrease in FeNO from 88 to 704 mcg was generally less pronounced than that from 0 to 88 mcg.

The intersubject variability of FeNO was calculated using the ln-transformed geometric means from Days 3-14 for each dose. The intrasubject variability was calculated using FeNO results from the two 176 mcg/day replicate arms (SM-Table 4).

Discussion

For this study, obtaining a sufficient number of subjects to complete the study was challenging. Only nine subjects completed Phase II study. Despite efforts to enrich the population (requiring a minimum baseline FeNO value, not producing less than a 25% reduction in FeNO at the intermediate fluticasone dose, and having a specified level of return to baseline following a two-week washout), subjects that completed Phase II failed to show dose-response with FeNO. Several reasons could be cited for the lack of dose-response seen in this study. Fluticasone propionate is considered to have a higher potency than other marketed ICS, such as beclomethasone dipropionate and budesonide (2). A study conducted with inhaled budesonide observed a maximum reduction in FeNO at an intermediate dose (3). Therefore, a dose-response of FeNO in a pharmacodynamic bioequivalence study may not be observed using marketed strengths of the more potent fluticasone as even the lowest labeled dose (88 mcg BID for Flovent HFA) may produce FeNO levels on or near the plateau of the dose-response curve.

Additionally, a 2 week washout period may not have been sufficient for fluticasone in this study. A previous study (4) noted that some subjects on inhaled budesonide required more than 6 weeks for FeNO levels to return to within 15% of the mean FeNO recorded during dosing with placebo. A washout period longer than 2 weeks’ duration may then be required to reach FeNO levels within a prespecified range of the baseline value, considering the relatively higher potency of fluticasone. The increase in mean FeNO levels at each placebo period over the course of the study shown in SM-Table 2 also suggests that the 14-day washout periods were not sufficient for FeNO levels to return to the proximity of the baseline value for BID dosing of fluticasone. Therefore, there may be a possibility of carryover effects of FeNO in the crossover design utilized in this investigation. Disease progression and drug treatment effects may also explain the incomplete return to baseline values.

It should be noted that although a dose-response of FeNO was not evident from this study, FeNO measurements were reproducible. The intersubject variability of the average FeNO response from Day 3-14 was between 13.46-21.17% and the intrasubject variability calculated from the replicated 176 mcg dose was 26.3%. These values are not considerably higher than the variability seen in pharmacokinetic parameters from in vivo bioequivalence studies (AUC and Cmax). The reproducibility of FeNO measurements has also been demonstrated by other investigators (5) and (6).

Statistical analyses to determine dose-response were conducted using FeNO values taken from days 3-14. The choice of the endpoint based on the FeNO data from 3-14 days is reasonable considering that previous studies (4) and (7) have indicated that levels of FeNO decrease noticeably after 3 days of ICS treatment. A separate analysis using FeNO values taken from Days 7-14 was performed but also failed to demonstrate dose-response (results not shown).

Dose-response analysis was also performed using baseline-adjusted values. FeNO values were baseline-adjusted by subtracting the last value obtained from the preceding washout period. A clear dose-response was not evident as the difference in LS means between doses were not significantly different (p<0.05).

Conclusion

A pharmacodynamic study with ICS and subject enrichment was performed to investigate the dose-response of FeNO in subjects with asthma. A clear dose-response relationship for FeNO was not evident for doses consisting of the lowest strength and multiples thereof of fluticasone propionate HFA administered twice daily, suggesting that these dose levels are near or on the plateau region of the dose-response curve. Furthermore, the crossover design of this investigation may not be optimal for FeNO dose-response studies, as most subjects did not return to near-baseline levels during washout periods. Study designs based on FeNO as a biomarker that require a dose-response relationship for ICS such as fluticasone need further evaluation before use in bioequivalence studies.

References:

1.ATS/ERS recommendations for standardized procedures for the online and offline measurement of exhaled lower respiratory nitric oxide and nasal nitric oxide, 2005. Am J Respir Crit Care Med. 2005;171(8):912-30.

2.Staresinic AG, Sorkness CA. Fluticasone propionate: a potent inhaled corticosteroid for the treatment of asthma. Expert Opin Pharmacother. 2000;1(6):1227-44.

3.Jatakanon A, Kharitonov S, Lim S, Barnes PJ. Effect of differing doses of inhaled budesonide on markers of airway inflammation in patients with mild asthma. Thorax. 1999;54(2):108-14.

4.Silkoff PE, McClean P, Spino M, Erlich L, Slutsky AS, Zamel N. Dose-response relationship and reproducibility of the fall in exhaled nitric oxide after inhaled beclomethasone dipropionate therapy in asthma patients. Chest. 2001;119(5):1322-8.

5.Kharitonov SA, Gonio F, Kelly C, Meah S, Barnes PJ. Reproducibility of exhaled nitric oxide measurements in healthy and asthmatic adults and children. Eur Respir J. 2003;21(3):433-8.

6.Bohadana A, Michaely JP, Teculescu D, Wild P. Reproducibility of exhaled nitric oxide in smokers and non-smokers: relevance for longitudinal studies. BMC Pulm Med. 2008;8:4-11.

7.O'Connor BJ, Collarini S, Poli G, Brindicci C, Spinola M, Acerbi D, et al. Rapid effects of extrafine beclomethasone dipropionate/formoterol fixed combination inhaler on airway inflammation and bronchoconstriction in asthma: a randomised controlled trial. BMC Pulm Med. 2011;11:60.