| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From Inhale Therapeutic Systems Inc. (Drs. Newhouse, Duddu, Clark, and Weers, Ms. Walter, and Mr. Tarara), San Carlos, CA; and Pharmaceutical Profiles, Ltd. (Dr. Hirst), Nottingham, UK.
Correspondence to: Jeffry G. Weers, PhD, Inhale Therapeutic Systems Inc., 150 Industrial Rd, San Carlos, CA 94070; e-mail: jweers{at}inhale.com
Abstract
Study objectives: To evaluate the efficiency and reproducibility of pulmonary delivery of an investigational tobramycin PulmoSphere formulation (PStob) [Inhale Therapeutic Systems; San Carlos, CA] by a passive dry powder inhaler, and to compare serum concentrations and whole-lung deposition with a commercial nebulized tobramycin product (TOBI; Chiron Corporation; Seattle, WA).
Design: A five-period, open-label, nonrandomized crossover study.
Participants: Fourteen healthy volunteers were studied, and 12 completed the study.
Interventions: PStob powder was manufactured using lipid-based PulmoSphere technology, producing highly dispersible porous particles. PStob was radiolabeled with 99mTc, and in vitro experiments confirmed it as a valid drug marker. To identify whole-lung distribution via scintigraphy, subjects inhaled contents of a single capsule (72 L/min) containing 25 mg of 99mTc-labeled PStob (13.5 mg of tobramycin free base) in periods 1 to 3. In period 4, subjects received 99mTc nebulized tobramycin, approximately 2.5 mL of 300 mg/5 mL. Deposition and blood samples were obtained. In period 5, six 25-mg doses of unlabeled PStob (81 mg of tobramycin base) were inhaled and blood samples were collected.
Measurements and results: Mean whole-lung deposition of PStob was 34 ± 6% and nebulized tobramycin was 5 ± 2%. Peak tobramycin concentration in serum (Cmax) values were 0.6 µg/mL with PStob and 0.28 µg/mL after nebulized tobramycin. Serum area under the curve was 4.4 µg x h/mL vs 2.1 µg x h/mL for nebulized tobramycin. Median time to Cmax for PStob was comparable to nebulized tobramycin.
Conclusions: The aerosol doses of PStob (25 mg and 150 mg) were well dispersed and tolerated. Serum drug concentrations matched scintigraphy data and were roughly twice that of the comparator. Intrasubject dose variability for three equivalent periods did not exceed 18% relative SD. PStob Cmax (0.6 µg/mL) was well below the toxic threshold (2 µg/mL).
Key Words: anti-infectives • dry powder inhalation • inhaled tobramycin • nebulization • PulmoSphere • Turbospin
For > 2 decades, inhaled antibiotics (and antifungal agents) have been used effectively for ameliorating chronic pulmonary infections in conditions such as cystic fibrosis (CF), and non-CF bronchiectasis. Various aerosol delivery devices, such as dry powder inhalers (DPIs), pressurized metered-dose inhalers, and nebulizers are available for pulmonary drug administration, each with well-recognized advantages and disadvantages. Nebulization has many well-documented disadvantages, including extended administration time, high cost, low efficiency and poor reproducibility, risk of bacterial contamination, and the need for bulky compressors or gas cylinders. Pressurized metered-dose inhalers usually provide much smaller unit doses than would be considered practical for antibiotic therapy. Administration of antibiotic dry powder aerosols to the lung has been attempted, but studies1 2 were limited by inefficient delivery devices and/or poorly dispersible lactose formulations.
Recent particle-engineering technologies, such as the PulmoSphere process (Inhale Therapeutic Systems; San Carlos, CA), have led to more easily dispersible powders with improved aerodynamic properties.3 These powders have been shown to deliver drugs efficiently to the lower respiratory tract using relatively simple and inexpensive DPIs.4 The objectives of the current study were as follows: (1) to determine the efficiency and reproducibility of pulmonary delivery of an engineered tobramycin powder aerosol (tobramycin PulmoSphere formulation [PStob]; Inhale Therapuetic Systems) from a simple, portable DPI using pharmacoscintigraphy; and (2) to compare pharmacokinetics and whole-lung deposition with a clinically available nebulized tobramycin product (TOBI; Chiron Corporation; Seattle, WA).
Materials and Methods
PStob was manufactured using an emulsion-based, spray-drying process described previously.5 The emulsion feedstock contains a volatile blowing agent (perfluorooctyl bromide), which aids in the creation of ultralow density sponge-like particles. PStob was inhaled from a Turbospin DPI (PH&T; Milan, Italy). Capsules were filled with 25 mg of PStob powder that contained nominally 90% weight/weight tobramycin sulfate, corresponding to a potency of active tobramycin free base of 13.5 mg per actuation.
TOBI was obtained from a local pharmacy by the clinical investigator; it contains 300 mg of tobramycin free base in 5 mL of sodium chloride (2.25 mg/mL) at pH 6.0. Nebulized tobramycin was administered as directed using a hand-held PARI LC Plus Nebulizer (PARI GmbH; Starnberg, Germany) in combination with the PARI Master compressor over a period of 15 min, the time recommended by the manufacturer. The PARI Master compressor has been demonstrated to have comparable performance to the DeVilbiss Pulmoaide (Sunrise Medical; Somerset, PA), the compressor recommended by the manufacturer of TOBI.6
Study Design and Subjects
A nonrandomized, five-period, crossover design (Table 1
) was conducted in 14 healthy subjects (10 men and 4 women; mean ± SD age, 34 ± 8 years). The subjects were healthy male and female volunteers previously known to the study center, or recruited from the local population via advertising. Subjects were nonsmokers, with pulmonary function at least 80% of predicted values and no history of chronic respiratory disorders. The clinical protocol was approved by the Quorn Research Review Committee (Leicestershire, UK). The study was conducted according to the Declaration of Helsinki on biomedical research in human subjects, and written informed consent was obtained from all subjects prior to recruitment.
|
Manufacture and Validation of Radiolabeled PStob Powder
PStob was radiolabeled with 99mTc at Pharmaceutical Profiles, Ltd. (Nottingham, UK) using a modification of a previously described method.7
8
The radiolabeled PStob powder was hand filled into size 2 hydroxypropylmethylcellulose capsules for aerosol performance and clinical testing. The mass of powder in the capsule was determined using a calibrated analytical balance with a precision of ± 0.01 mg (Hana Balance; Seoul, South Korea). The fill-mass specification was set at 25 ± 2 mg, and this target was met for all hand-filled capsules. Nebulized tobramycin was radiolabeled by simply dissolving an appropriate amount of 99mTc-pertechnetate in the aqueous nebulized tobramycin solution.
Before initiation of the clinical phase of the study, in vitro validation experiments were conducted to demonstrate the following: (1) significant alteration of the particle size distribution (PSD) did not occur during the labeling process, and (2) the PSD of the radiolabel matched the PSD of the drug. Particle size measurements were performed using a multistage liquid impinger fitted with a US Pharmacopeia induction port at 60 L/min (Copley Instruments; Nottingham, UK).9 Quantitation of tobramycin was done by the stability-indicating US Pharmacopeia method that involves derivatization of the tobramycin with 2,4-dinitrofluorobenzene, and quantitation by high-performance liquid chromatography. This method was also used for purity assessment.
Gamma Scintigraphy Analysis
Prior to dosing of PStob, subjects were trained to inhale from the Turbospin device at the recommended peak inspiratory flow rate of 60 L/min using a Vitalograph spirometer equipped with a visual feedback monitor (Vitalograph; North Buckinghamshire, UK). Wearing noseclips, subjects were instructed to exhale to approximately residual volume, and then inhale rapidly to total lung capacity. Subjects were then instructed to hold their breath for 5 s, and then exhale. The spirometer was used for training purposes only, and no data were collected from subjects during the clinical study.
Prior to dosing of radiolabeled TOBI, subjects were trained in tidal breathing using nebulized saline solution. The nebulizer system was configured with an exhalation filter to capture any exhaled radiolabeled aerosol.
Immediately following administration of radiolabeled aerosols, two-dimensional scintigraphic images of the anterior and posterior chest, lateral oropharynx, aerosol delivery devices, and exhalation filter were recorded using a gamma camera (Maxicamera; General Electric; Milwaukee, WI). All images were recorded on a Park Medical Micas Xplus computer system (Park Medical; Farnborough, UK) and were stored on Digital Audio Tape (Seagate; Amsterdam, Netherlands). Scintigraphic images were compared with whole-lung outlines generated from an 81mKr ventilation scan obtained from each volunteer. Regions of interest were drawn around the oropharynx, esophagus, stomach, and whole lung. The counts within these regions were corrected for background radioactivity, radioactive decay, and tissue attenuation of gamma rays.10 In regions where both anterior and posterior images were recorded, the geometric mean of counts in both images was calculated. Determination of the percentage of the dose deposited in the oropharynx included activity adhering to the mouth and pharynx together with any swallowed activity detected in the esophagus, stomach, and intestine. Counts for each area were expressed as a percentage of the nominal dose, which was determined from the sum of the total body counts in addition to those deposited on the Turbospin and capsule, or nebulizer components, and the exhalation filter.
The lungs were divided into central, intermediate, and peripheral regions of interest.11 12 The ratio of peripheral to central lung deposition (P/C ratio) was calculated as an index of regional lung deposition.
Pharmacokinetic Analysis
In period 5, subjects received six 25-mg inhalations of unlabeled PStob powder (150 mg of powder; 80 mg of tobramycin free base) from the Turbospin device. Successive inhalations of 25-mg each were performed at 3-min intervals over 15 min. The inhalation intervals were designed to match the time period for inhalation of nebulized tobramycin in period 4 (ie, 15 min). Serum samples were analyzed at MDS Pharma Services (Lincoln, NE) using a validated fluorescence polarization immunoassay according to the Abbott TDx/TDxFLx System (Abbott Laboratories; Abbott Park, IL). The standard curve ranged from 0.05 to 1.0 µg/mL, with a limit of quantitation of 0.05 µg/mL.
Pharmacokinetic parameters (peak tobramycin concentration in serum [Cmax], time to Cmax [Tmax], half-life of removal of tobramycin from serum, and area under the curve [AUC]) were generated from serum tobramycin concentrations with WinNonlin software (PharSight; Mountain View, CA) using standard noncompartmental modeling techniques. Concentrations below the lower limit of quantitation were regarded as zero. AUC was calculated as the AUC for each treatment, calculated by the log-linear trapezoidal method, from 0 to the first time point below the level of quantitation. The efficiency of PStob treatment relative to nebulized tobramycin was calculated for each subject as shown:
 |
Individual AUCs were dose normalized for each treatment based on the actual lung dose determined from scintigraphic assessments after one inhalation of 99mTc-labeled PStob powder or nebulized tobramycin, as follows:
 |
Safety Assessment
Pulmonary function (FVC, FEV1) and vital signs were recorded before and 30 min after each dose and prior to discharging the subject from the study center at each visit. Adverse events were monitored throughout the study. In addition, each subject underwent a physical examination, ECG, routine clinical chemistry, hematology and urinalysis, and pulmonary function testing at the end of the study.
Data Analysis
Scintigraphic data reported below are from the 13 subjects who completed periods 1 to 4. Serum tobramycin concentrations in periods 4 and 5 were only available for the 12 subjects who completed the study, since one subject withdrew prior to period 5. Correlations between lung dose and serum concentrations were only performed for the 12 subjects who completed the entire study.
Descriptive statistics were calculated for all variables. When applicable, statistical comparisons between treatments were made using a paired, two-tailed Student t test. All data are mean ± SD unless noted otherwise.
Results
Radiolabel Validation Studies
Before initiation of the clinical phase of the study, in vitro validation experiments were conducted with the dry powder formulation to demonstrate the following: (1) significant alteration of the PSD did not occur during the labeling process, and (2) the PSD of the radiolabel matched the PSD of the drug. These validation experiments demonstrated that the PSD of the radiolabel and the PStob formulation were well matched, with no alteration of aerosol properties of the formulations after radiolabeling. The PSD of the radiolabel in each clinical batch was also confirmed by multistage liquid impinger prior to dosing on each study day. For nebulized tobramycin, the probe and drug are both dissolved in the aqueous phase, and hence the concentration of drug and probe are uniformly distributed in a given nebulized droplet.
Scintigraphy: Deposition of Tobramycin Aerosols
The percentage of the radiolabeled PStob dose deposited in the whole lung in periods 1, 2, and 3, as determined by gamma scintigraphy, is shown in Table 2
. Subject 103 coughed during inhalation of the first dose, and powder was observed coming out of the inhaler; therefore, period 1 deposition data for this subject was not included in the descriptive statistics. The mean peak inspiratory flow rate across the subjects was 72 L/min. Whole-lung deposition was relatively constant for the first two replicate inhalations, ranging from 21 to 51% in period 1 and 21 to 48% in period 2, with an average of 32% overall. Mean lung deposition increased to 38% with the period 3 inhalation (range, 28 to 48%). Mean deposition for all subjects across all three inhalations was 34.3 ± 5.8%, for an overall coefficient of variation (CV) of 17%. The mean within-subject CV for the three identical inhalations was 17%, ranging from as low as 6.5 to 35%. Eight subjects had a CV of 
20%.
|
|
1.5. A P/C ratio of 1.5 to 1.6 corresponds to approximately 60 to 63% alveolar deposition.12
Pharmacokinetics
Serum tobramycin concentrations rose at comparable rates after inhalation of both aerosols, reaching maximum observed concentrations between 1 h and 3 h after inhalation (Tmax) was completed (Fig 1 ). Mean pharmacokinetic parameters are shown in Table 4
. Observed Cmax values of tobramycin after inhalation of the PStob were approximately twice as high as after nebulized tobramycin (0.60 ± 0.18 µg/mL vs 0.28 ± 0.09 µg/mL, p < 0.001).
|
|
Consistent with the difference in Cmax between the two treatments, the AUC after inhalation of the PStob was double that after nebulized tobramycin (4.4 µg x h/mL vs 2.1 µg x h/mL); however, when corrected for the actual lung dose (based on scintigraphic deposition values), the AUCs for both the PStob and nebulized tobramycin treatments are comparable, indicating that the same proportion of the drug delivered to the lung reached the systemic circulation for either aerosol.
Due to the low efficiency of lower respiratory tract deposition observed following nebulization, the relative bioavailability of the PStob inhaled from the Turbospin was nine times more efficient than the nebulized tobramycin system. The mean mass of tobramycin deposited in the lung as determined by gamma scintigraphy (27 mg for PStob and 15 mg for nebulized tobramycin) correlates well with the pharmacokinetics data above.
Safety
No clinically significant changes in pulmonary function test results or vital signs were observed. The largest mean decreases in FEV1 and FVC at 30-min postdose following administration of 25 mg of PStob were 1.6 ± 4% and 3.2 ± 4.1%, respectively. The mean decreases in FEV1 and FVC following six inhalations of 25 mg of PStob powder were 0.25 ± 3.0% and 0.0 ± 3.8%, respectively, indicating that multiple inhalations had no impact on pulmonary function. No adverse events of note occurred. Cough was the most common adverse event reported (11 times by a total of six subjects). Most of these incidences were described as a minor clearing of the throat, with an immediate cough observed in only 2 of 111 total inhalations (1.8%). Slight wheezing was observed in one subject. Cmax values for PStob (0.6 µg/mL) were well below the reported threshold for nephrotoxicity and ototoxicity (2.0 µg/mL). None of the subjects reported an unpleasant taste following inhalation of 150 mg of powder over the 15-min tobramycin administration period.
Discussion
Little comparative data have been published on the pharmacokinetics of tobramycin after inhalation. Touw et al13 reported Cmax values of 1.2 µg/mL (range, 0.19 to 2.57 µg/mL) after 600 mg of nebulized tobramycin in patients with CF, describing a mean systemic bioavailability of 17.5% (range, 6 to 24%) based on urinary tobramycin recovery. Cooney et al14 reported a mean systemic availability of 9.13 ± 3.8% after inhalation. Both of these bioavailability numbers are consistent with the approximate 10% fractional deposition of aerosolized drug that is deposited in the lungs following nebulization.
We found lower deposition efficiency in our nebulization study (5%), but the corresponding serum concentrations were consistent with that lung dose. Furthermore, had the nebulizer been run to dryness (ie, for approximately 30 min), the efficiency would likely have been similar to that previously reported for this nebulizer/compressor combination.
PathoGenesis Corporation15 reported median 1-h postdose Cmax values in patients with CF of 0.94 µg/mL (range, 0.18 to 3.62 µg/mL) following 300 mg of nebulized tobramycin over a 15-min period. Touw et al13 and Cooney et al14 reported a similar range of serum concentrations. However, in the study by Touw et al,13 the nebulization volume was 10 mL, twice that marketed by the manufacturer, and approximately 50% of the dose was left in the nebulizer after dosing (nebulization time not stated). In addition, the nebulizers used in the two studies were different and, hence, may not have generated the same aerosol PSDs.16
Since we were able, in the present study, to quantitate the actual lung dose by gamma scintigraphy in addition to measuring serum concentrations, direct comparison between nebulization and dry powder delivery is possible. The present study demonstrates that reproducible pulmonary delivery of relatively high doses of tobramycin is feasible using a DPI. We were able to deliver approximately 5 mg of tobramycin (free base) to the lungs per inhalation, for a total of 25 to 30 mg following six inhalations. The nebulized tobramycin system in the present study deposited only 5% of the nominal dose, or 15 mg of tobramycin, in the lungs. However, at a customary 10% nebulizer efficiency, a 30-mg dose would have been administered, equivalent to the six capsules of PStob delivered in period 5.
Consistent with the pulmonary deposition data, twice as much tobramycin reached the systemic circulation using the PStob powder compared to nebulized tobramycin. Since systemic exposure (Cmax and AUC) was proportional to the lung dose, bioavailability of tobramycin from the lung was equivalent for both formulations and methods of aerosol administration; however, under the specific conditions of this study, the PStob/Turbospin combination was nine times more efficient than the nebulized tobramycin system, since the DPI was loaded with about fourfold less drug (80 mg vs 300 mg). As well, aerosol administration via nebulization takes much longer, approximately 30 min if set-up, drug administration, and cleaning are taken into account. The six inhalations of dry powder were administered over 15 min in the present study to match the nebulization time and to facilitate pharmacokinetic comparisons, but there is no reason why multiple inhalations could not be performed within approximately 2 min. The portability of the Turbospin device and the greatly decreased administration time will likely lead to improved compliance for patients with CF receiving PStob therapy relative to nebulization. Furthermore, with dry powders, there would not be the long-term risk of bacterial contamination that may occur with nebulizers.
Currently, all marketed dry powder products contain fine micronized drug particles. Generally, these particles exhibit poor powder flow characteristics, and must be blended with large lactose carrier particles in order for the powder to be effectively metered into and dispersed from capsules. PStob exhibits excellent flowability (without blending), thereby allowing for effective metering of powder on standard capsule-filling equipment (data not shown). The absence of carrier particles enables a much greater drug dose to be loaded into capsules, allowing for doses from a few milligrams to tens of milligrams, thereby making delivery of less potent drugs (eg, anti-infectives) practical. It also likely decreases the incidence of cough for a given delivered dose, since the powder load will be less (especially for lower dose delivery). In the present study, delivery of 150 mg of PStob powder (six 25-mg inhalations) was well tolerated with little tendency for cough and no complaints of unpleasant taste.
Conclusions
Pulmonary delivery of tobramycin from a passive DPI is feasible for PStob due to the high drug loading (90% weight/weight tobramycin sulfate), and ninefold improved pulmonary deposition compared to the nebulizer over a 15-min administration period. A < 20% CV in pulmonary delivery of PStob was observed (intersubject and intrasubject CV, 18%).
A comparable dose can be delivered to the lung and/or the systemic circulation with PStob powder as with the nebulized product, but in approximately one tenth the time and with considerably less effort. This has the potential to improve patient acceptance and adherence. The relatively large mass of powder administered over 15 min (150 mg, 52 mg deposited in the lung, and 65 mg deposited in the oropharynx) was well tolerated.
Acknowledgements
The authors thank Benjamin English, Howard Gill, Vathana Tep, Keith Ung, Christi Garcia, Christine Suen, Betty-Wong Lee, Hilary Chen, Daisy Leung, Kevin O’Sullivan, Steve Sisk, Kevin Trimble, Michael Eldon, Matthew Pickford, Rebecca Elton, and Steve Newman for their significant contributions to this work.
Footnotes
Abbreviations: AUC = area under the curve; CF = cystic fibrosis; Cmax = peak tobramycin concentration in serum; CV = coefficient of variation; DPI = dry powder inhaler; P/C ratio = ratio of peripheral to central lung deposition; PSD = particle size distribution; PStob = tobramycin PulmoSphere formulation; Tmax = time to peak tobramycin concentration in serum
This work was fully funded by Inhale Therapeutic Systems Inc., which is currently developing a dry powder formulation of tobramycin for treating patients with cystic fibrosis in conjunction with Chiron Corporation (Emeryville, CA).
All of the Inhale Therapeutic Systems Inc. authors have received stock options and other compensation.
Pharmaceutical Profiles, Ltd. is a contract clinical research organization that specializes in conducting gamma scintigraphy studies of inhaled drug products. Their participation in the study was also funded by Inhale Therapeutic Systems Inc. As such, Dr. Hirst was compensated by Inhale Therapeutic Systems Inc. for his efforts.
Turbospin DPIs were supplied by PH&T (Milan, Italy).
Received for publication April 24, 2002. Accepted for publication November 12, 2002.
References
This article has been cited by other articles:
|
|
 |
|
  J. S. Patton, C. S. Fishburn, and J. G. Weers The Lungs as a Portal of Entry for Systemic Drug Delivery Proceedings of the ATS, December 1, 2004; 1(4): 338 - 344. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
