RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
R Tuning Aerosol Performance Using the Multibreath Orbital Dry Powder Inhaler Device: Controlling Delivery Parameters and Aerosol Performance via Modification of Puck Orifice Geometry BING ZHU,1 PAUL M. YOUNG,1,2 HUI XIN ONG,1,2 JOHN CRAPPER,3 CARINA FLODIN,3 ERIN LIN QIAO,3 GARY PHILLIPS,3 DANIELA TRAINI1,2 1
Respiratory Technology, Woolcock Institute of Medical Research, Glebe, New South Wales 2037, Australia Discipline of Pharmacology, Sydney Medical School, The University of Sydney, New South Wales 2006, Australia 3 Pharmaxis Ltd., Frenchs Forest, Sydney, New South Wales 2086, Australia 2
Received 23 February 2015; revised 25 March 2015; accepted 31 March 2015 Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jps.24458 R ABSTRACT: The current study presents a new approach to tackle high-dose lung delivery using a prototype multibreath Orbital dry powder inhaler (DPI). One of the key device components is the “puck” (aerosol sample chamber) with precision-engineered outlet orifice(s) that control the dosing rate. The influence of puck orifice geometry and number of orifices on the performance of mannitol aerosols were R DPI prototype. studied. Pucks with different orifice configurations were filled with 400 mg of spray-dried mannitol and tested in the Orbital The emitted dose and overall aerodynamic performance across a number of “breaths” were studied using a multistage liquid impinger. The aerosol performances of the individual actuations were investigated using in-line laser diffraction. The emptying rate of all pucks was linear between 20% and 80% cumulative drug released (R2 > 0.98), and the amount of formulation released per breath could be controlled such that the device was empty after 2 to 11 breath maneuvers. The puck-emptying rate linearly related to the orifice hole length (R2 > 0.95). Mass median aerodynamic diameters of the emitted aerosol ranged from 4.03 to 4.62 m and fine particle fraction (6.4 m) were 50%–66%. Laser diffraction suggested that the aerosol performance and emptying rates were not dependent on breath number, showing C 2015 Wiley Periodicals, Inc. and the American Pharmacists Association J Pharm Sci consistent size distribution profiles. Keywords: aerosols; pulmonary drug delivery; orbital; high dose dry powder inhaler; aerosol performance; formulation; powder technology; particle size; puck orifice geometry
INTRODUCTION Dry powder inhalers (DPIs), nebulizers, and pressurized metered dose inhalers have been developed for over 30 years for the treatment of various pulmonary diseases.1,2 The principle of these inhalation devices has focused on delivery of inhalable medications for the treatment of asthma and chronic obstructive disease that usually require a small quantity of active pharmaceutical ingredient (API) per delivered dose.3–5 Various low-dose DPI inhaler devices have been successfully developed and marketed over the years and a number of commercial products contain doses ranging from tens of micrograms (e.g., formoterol 6–12 :g; tiotropium 18 :g; salbutamol 100 :g; beclomethasone 250 :g) up to a few milligrams (e.g., nedocromil and sodium cromoglycate 2–5 mg). Suitable to deliver medicaments within the microgram range, such as corticosteroids6 and $2 agonists,7 these devices lack the ability to deliver APIs requiring higher milligram quantities (i.e., 10–1000 mg), such as antibiotics8 and mucociliary clearance R is the reagents, such as mannitol.9 Currently, TOBI Podhaler only commercially available antibiotic (tobramycin) DPI with a R , the only highhigh-delivery dose (112 mg) and Bronchitol dose mucociliary clearance therapy containing mannitol as the active ingredient (400 mg). However, the delivered dose for
Correspondence to: Paul M. Young (Telephone: +61-2-9114-0350; Fax: +61-29351-4391; E-mail: [email protected]) Journal of Pharmaceutical Sciences C 2015 Wiley Periodicals, Inc. and the American Pharmacists Association
both these products is achieved by individually dispersing multiple capsules (4 × 28 mg in the case of tobramycin and 10 × 40 mg in the case of mannitol) through a multiple “loadprime-inhale” routine. This type of aerosol delivery strategy is time-consuming and can potentially result in poor therapy adherence.10 Mannitol, used as an inhaled dry-powder osmotic reagent, has been proven to be effective in the treatment of diseases where hyper-mucus production is prevalent, such as cystic fibrosis and bronchiectasis.11,12 For example, patients with bronchiectasis showed marked improvement in mucus clearance in the central and intermediate lung regions (>20%) 75 min postinhaled mannitol intervention.9 The solid content in sputum of CF patients was significantly reduced by approximately 2%13 and forced expiratory volume increased by approximately 100 mL14 with 2-week inhaled mannitol therapy. Although this unique therapy provides a means of enhancing clearance, as with other high-dose therapies, the delivery vehicle is based on a DPI that was originally designed for lower dose therapeutics, thus requiring a patient to load, actuate, and inhale from multiple capsules. Recently, Pharmaxis developed a disposable, multibreath DPI device able to deliver large doses of active ingredients (500 mg) through a number of inhalation maneuvers.15,16 The R device (Fig. 1) consists of four major components— Orbital (1) a mouthpiece, (2) dispersion grid, (3) dispersion chamber with multiple tangential airstream inlets, and (4) a “puck” or aerosol sample compartment. During inhalation, the puck rotates within the dispersion chamber and releases the Zhu et al., JOURNAL OF PHARMACEUTICAL SCIENCES
1
2
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
HPLC grade and purchased from Chemsupply (Port Adelaide, Australia). Scanning Electron Microscopy The morphology of mannitol was visualized using a JEOL-6000 bench-top scanning electron microscope (SEM; JEOL, Tokyo, Japan). Prior to imaging, mannitol samples were deposited onto carbon tape mounted on an aluminum stub and coated with gold using a sputter coater (DII-29010SCTR; JEOL) at a coating thickness of approximately 15 nm. Imaging was conducted at an accelerating voltage of 5 keV. Particle Sizing
R Figure 1. Schematic of the prototype Orbital device.
formulation through precision-engineered orifice(s). Previous studies15,16 have demonstrated that the device can successfully deliver high payloads of cohesive ciprofloxacin/mannitol and azithromycin/mannitol spray-dried formulations (200 mg) via approximately 10 inhalation maneuvers (equivalent to the number of capsules that would be required in conventional DPIs), while obtaining respiratory fractions greater than 60%. In these previous studies, the authors focused on engineering and characterization of novel spray-dried powders for infection R to be verand bronchiectasis. They demonstrated the Orbital satile for the delivery of a number of molecules when using standard prototype device configuration. However, features of the device such as resistance, dispersion grid, and puck dimensions may be modified to “tune” both the delivery profile and aerosol performance. Here, we investigate the emptying properties and aerosolizaR device as a function of puck tion performance of the Orbital hole number and geometry using mannitol as the model API, R for mucus clearance. similar to that used in Bronchitol
MATERIALS AND METHODS Materials Spray-dried mannitol (Batch ID: EXP 280) was supplied by Pharmaxis Ltd. (Sydney, Australia) and used as received. R DPI devices, including pucks with differPrototype Orbital ent orifice configurations (Table 1), were supplied by Pharmaxis Ltd. Water was purified using reverse osmosis (Merck Millipore, Bayswater, VIC, Australia). All other solvents were
The volume median size (D0.5 ) of the spray-dried mannitol was measured by laser diffraction using a Mastersizer 3000 (Malvern Instruments, Malvern, UK) equipped with hydrodispersion unit. Samples were prepared by suspending approximately 40 mg of spray-dried mannitol in 5 mL of cyclohexane, which was then added dropwise to the hydrodispersion unit, operating at a stirring speed of 2000 rpm, until 5%–15% obscuration was achieved. Samples were measured with a density setting of 1.51 g/cm3 and refractive index of 1.33. Each sample was analyzed in triplicate. Multibreath-Delivered Dose Study The puck and puck orifice is a key parameter that influences the emptying and aerosolization performance of the Orbital. By modifying the puck hole geometry, number of holes, and hole size, it becomes possible to modify the rate at which the powder is dispersed and thus the number of breaths required to achieve a target dose. In the current study, different types of pucks (Table 1) were loaded with 400 ± 0.8 mg (n = 3) spray-dried mannitol and assembled in the Orbital inhaler device, to simulate standard treatment.12 Because of the high payload delivered, a multistage liquid impinger (MSLI; Westech, Bedfordshire, UK) equipped with a United States Pharmacopeia (USP) induction port was used to collect the emitted aerosol to avoid filter clogging. The airflow rate through the assembly was calibrated at 60 ± 2 L/min using a flowmeter (TSI 4040; TSI Instruments Ltd., Shoreview, Minnesota). The assembled Orbital device was inserted into a silicon mouthpiece adaptor, mounted onto the USP induction port, and powder dispersed at 60 L/min over a 4-s period to simulate one breath.17 The weights of the assembled inhaler were recorded before/after each actuation/breath and shot weight of each actuation deemed as the weight difference (mg) of the two consecutive dispersions. The process was repeated with intermediate airflow calibration (60 L/min) for each shot until the difference between two actuations were less than 0.2 mg. The experiment was conducted in triplicate for each puck type.
Table 1. Puck Orifice Configurations Puck ID Orifice Shape Orifice Number 1 P-002 2 P-003 3 P-013a 4 P-028 5 P-024 6 P-021 a
Rectangular Rectangular Spherical Spherical Spherical Spherical
1 1 1 1 2a 2a
Orifice Geometry (mm) 0.5 × 0.77 (width × length) 0.5 × 0.9 (width × length) 0.6 (diameter) 0.9 (diameter) 0.4 (diameter) 0.55 (diameter)
Orifices at opposite sides of puck.
Zhu et al., JOURNAL OF PHARMACEUTICAL SCIENCES
Aerosol Particle Size Distribution Measured by Cascade Impaction In addition to delivered dose, puck geometry may also have an impact on the aerosol performance of the emitted aerosol cloud. Therefore, the aerosolization efficiency of the Orbital inhaler device with the different pucks was investigated using a MSLI. In brief, pucks (Table 1) were filled with 400 ± 0.5 mg of spray-dried mannitol (n = 3) and assembled in the Orbital device. The formulation was dispersed into the MSLI at 60 ± 2 L/min for 4 s per actuation (n = 3) until the puck was DOI 10.1002/jps.24458
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
3
empty. The weight of the inhaler was recorded before/after each dispersion, and the puck was considered empty when the weight difference between two consecutive actuations was 0.2 mg. The puck, Orbital device, and Orbital mouthpiece were weighed before and after the test, and the amount of mannitol deposited on each part was calculated by weight difference. The MSLI USP induction port (Throat), stages, and filter were thoroughly rinsed with deionized water (recovery solution). Samples were collected for chemical quantification using high-performance liquid chromatography (HPLC). All experiments were conducted in triplicate.
dex of 1.33. Post-throat aerosol characteristics of each shot were determined by the median volume diameter (D0.5 , :m) and particle size distribution and temporal laser obscuration (%) profiles reported over the measurement period.
Chromatographic Methodology
Mannitol Particle Characterization
Mannitol concentrations of the samples collected from the MSLI were quantified using a validated HPLC method. The system consisted of a Shimadzu HPLC equipped with an LC20AD pump, RID 10A refractive index detector, SIL 20A autosampler, DGU 20A degassing unit, and CTO 20A column oven (Shimadzu, Rydalmere, Australia). Samples (100 :L) were R RCU-USP Sugar Alcohols LC column injected into a Rezex 2 (8 :m, 250 × 4 mm ; Phenomenex, Lane Cove, Sydney, Australia) and analyzed at a flow rate of 0.4 mL/min with a column oven temperature of 85°C, using degassed HPLC grade water as mobile phase with a retention time of approximately 6.5 min. The linearity of mannitol standard solutions was confirmed in between 0.1 and 3.0 mg/mL, with a regression value 0.999. The limit of quantification was 0.1 mg/mL and accuracy reported to be 99.5%–99.8% at target levels of 0.1, 1.0, and 3.0 mg/mL.
Figures 2a and 2b show scanning electron micrographs of the mannitol particles. In general, the spray-dried mannitol had a spherical morphology with crystalline-like surface characteristics. The cumulative volume size distribution is shown in Figure 2c. The sample presented a distribution profile with a size range of approximately 1–7 :m, with reported D0.1 , D0.5 , and D0.9 values of 1.7 ± 0.01, 3.1 ± 0.05, and 5.4 ± 0.13 :m, respectively (n = 3 ± SD).
Statistical Analysis Data were analyzed using one-way analysis of variance (ANOVA) followed by the Tukey’s post-hoc test. Differences were considered statistically significant at a level of p < 0.05.
RESULTS AND DISCUSSION
Multibreath-Delivered Dose Study Compared with conventional capsule-based DPI devices, the Orbital achieves high doses through multiple breaths from the puck loaded with a premetered mass of active ingredient.
Shot by Shot Evaluation Using In-Line Laser Diffraction Cascade impaction/impinger studies can only realistically provide information regarding the overall aerodynamic performance of the emitted aerosol clouds after multiple inhalation maneuvers as the laborious nature of this method makes it impractical to conduct shot by shot studies. Although the ultimate cumulative aerosol particle size is the key attribute for the Orbital device, it is worthwhile understanding how aerosolization performance varies for each inhalation maneuver. SubseR quently, an in-line laser diffraction method using a Spraytec particle sizer (Malvern Instruments) was developed and used to test the aerosol performance of the device with each puck type. Other authors have previously reported this method as a fast-screening tool for aerosol performance analysis.18 In brief, a USP induction port was fitted onto the inlet of R inhalation cell and outlet connected to an MSLI, the Spraytec with the configuration used for shot weight and cascade impaction studies. The inspiratory airflow through the entire assembly was set to 60 ± 2 L/min using a GAST pump and solenoid flow controller. The MSLI, in this case, was only used for aerosol collection purpose to avoid filter clogging and impactor device saturation, as previously mentioned. Pucks described in Table 1 were filled with 400 mg of spraydried mannitol and assembled in the Orbital inhaler device. The assembly was set up at the above-mentioned airflow rate and the solenoid was actuated for 4 s. The process was repeated until the weight difference between two consecutive actuations were below 0.2 mg. The setup was disassembled and cleaned after each actuation to ensure a good signal-to-noise ratio. Measurements were conducted at a sampling frequency of 2.5 kHz, with particle density of 1.51 g/cm3 and refractive inDOI 10.1002/jps.24458
Figure 2. (a) Low magnification, (b) high magnification SEM images, and (c) volume particle size distribution data for the mannitol inhalation powder. Zhu et al., JOURNAL OF PHARMACEUTICAL SCIENCES
4
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Figure 3. Cumulative emitted dose from the Orbital DPI at 60 L.min−1 (4 L) as a function of shot number for each puck hole geometry. n = 3 ± SD.
Patients receive high doses from sequential inhalation maneuvers and dispose of the device when the puck is empty. Consequently, it is important to understand the influence of puck hole number and geometry on the delivery efficiency of Orbital when operated under standard conditions (4 L inhalation volume– 60 L/min over a 4-s period17 ). Figure 3 shows the cumulative emitted dose of mannitol released from Orbital DPI with pucks containing different hole numbers and geometries. The dose released per actuation for all puck types was linear until approximately 300 mg was emitted, after which the emitted dose plateaued. Importantly, the release profile was puck-type dependent. For example, pucks with a 0.5 × 0.9 mm rectangular orifice geometry (Puck ID #P003) delivered 321.8 ± 1.5 mg (n = 3) of mannitol within the first two inhalation maneuvers, accounting for 80.5 ± 0.9% of the loaded mass. In comparison, 80% of the loaded dose was delivered over 11 shots with the puck containing two 0.4 mm spherical orifices (Puck ID #P-024). In general, high cumulative emitted dose (>300 mg) could be achieved within three inhalation maneuvers, using pucks with large orifice geometries (i.e., 0.9 mm diameter single spherical, 0.5 × 0.9 and 0.5 × 0.77 mm rectangular), whereas emitted dose of pucks with small orifice diameters were significantly lower (approximately 50 mg per shot), having prolonged dose release characteristics. In order to understand the puck-emptying profiles, the cumulative emitted dose as a percentage of mass loaded into the puck (400 mg) was plotted (Fig. 4a) and linear regression conducted between 20% and 80%. In general, regression analysis produced R2 values 0.99 and the slope ranged from 7.5 ± 0.3% emitted/shot for the 0.4-mm hole puck with two holes to 40.2 ± 0.2% emitted/shot for the 0.5 × 0.9 mm hole puck. This equates to a “dose per breath” range of approximately 7.5%–40% of the loaded dose (in this case 400 mg). Importantly, the mass release and emptying rate of mannitol per actuation was consistent for each puck type over the dose range. This is reflected by the small standard deviations between replicate measurements for any given puck at any specific shot number. Such consistency could be qualitatively explained by the “mass dependency theory.”16 In brief, the dose released from the puck is controlled by the hole diameter and Zhu et al., JOURNAL OF PHARMACEUTICAL SCIENCES
Figure 4. (a) Cumulative percent powder released from puck (n = 3 ± SD). (b) Relationship between emptying rate (% of 400 mg loaded dose) and puck hole diameter (n = 3). Y error bars are standard deviation in slope and X error bars are tolerance in puck orifice geometry (±0.025 mm).
remaining mass in the puck, which determines the puck rotation speed. When the remaining mass in the puck is high, a large amount of powder exits the puck orifice at a relatively slow puck rotation speed. As the dispersion continues, the formulation in the puck is released through the orifice at a higher rotation frequency, compensating for the reduced release mass per puck rotation. It is also worth noting that the emptying rate was dependent on the primary orifice diameter rather than the geometry or number of holes (i.e., hole length for rectangular orifices). A plot of emptying rate versus orifice diameter for all puck types showed a linear relationship (R2 of 0.95) with increased emptying rate being observed with increased orifice diameter. The puck hole diameters used in the analysis included the diameter of the single orifice (if dual orifice) and length (i.e., the greater of the two dimensions of a rectangle) for the rectangular orifices. Furthermore, the emptying rates of pucks with dual orifices were primarily determined by the diameter of the single holes instead of the number of orifices, which is supported by the similar dose release data of pucks P-013a (single orifice, 0.6 mm in diameter) and P-021 (dual orifice, 0.55 mm in diameter) (Table 1). This phenomenon is possibly because of the powder dynamic in the puck during the dispersion. As mentioned previously, the puck rotates in the dispersion chamber during the dispersion; therefore, the powder may subsequently rotate within the puck in the same manner. A certain aliquot of powder can be released because of centrifugal force only when the formulation reaches the exiting orifice. As orifices were DOI 10.1002/jps.24458
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
5
diagonally configured on the puck, it is likely that the majority of the formulation is only released from one orifice on one side per puck rotation, resulting in dose dependency in orifice diameter rather than number of holes on the puck.
Aerodynamic Performance and Delivery Efficiency In addition to the emitted dose, the aerodynamic characteristics of the aerosol cloud also play a key role in determining local drug concentration within the lung19 and thus the bioavailability of administered medication.7 As mannitol is used as the osmotic reagent to increase the water efflux to the airway lumen,12 an appropriate concentration at the target site is required for the desirable therapeutic effects. The overall aerosolization performance of mannitol particles delivered via the Orbital device from multiple inhalation manoeuvres were tested using different puck configurations (Table 1). Mass deposition on each MSLI stage was expressed as the percentages of the total drug mass recovered as shown in Figure 5. All the testing conditions showed low mass retention in the pucks (
R Tuning Aerosol Performance Using the Multibreath Orbital Dry Powder Inhaler Device: Controlling Delivery Parameters and Aerosol Performance via Modification of Puck Orifice Geometry BING ZHU,1 PAUL M. YOUNG,1,2 HUI XIN ONG,1,2 JOHN CRAPPER,3 CARINA FLODIN,3 ERIN LIN QIAO,3 GARY PHILLIPS,3 DANIELA TRAINI1,2 1
Respiratory Technology, Woolcock Institute of Medical Research, Glebe, New South Wales 2037, Australia Discipline of Pharmacology, Sydney Medical School, The University of Sydney, New South Wales 2006, Australia 3 Pharmaxis Ltd., Frenchs Forest, Sydney, New South Wales 2086, Australia 2
Received 23 February 2015; revised 25 March 2015; accepted 31 March 2015 Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jps.24458 R ABSTRACT: The current study presents a new approach to tackle high-dose lung delivery using a prototype multibreath Orbital dry powder inhaler (DPI). One of the key device components is the “puck” (aerosol sample chamber) with precision-engineered outlet orifice(s) that control the dosing rate. The influence of puck orifice geometry and number of orifices on the performance of mannitol aerosols were R DPI prototype. studied. Pucks with different orifice configurations were filled with 400 mg of spray-dried mannitol and tested in the Orbital The emitted dose and overall aerodynamic performance across a number of “breaths” were studied using a multistage liquid impinger. The aerosol performances of the individual actuations were investigated using in-line laser diffraction. The emptying rate of all pucks was linear between 20% and 80% cumulative drug released (R2 > 0.98), and the amount of formulation released per breath could be controlled such that the device was empty after 2 to 11 breath maneuvers. The puck-emptying rate linearly related to the orifice hole length (R2 > 0.95). Mass median aerodynamic diameters of the emitted aerosol ranged from 4.03 to 4.62 m and fine particle fraction (6.4 m) were 50%–66%. Laser diffraction suggested that the aerosol performance and emptying rates were not dependent on breath number, showing C 2015 Wiley Periodicals, Inc. and the American Pharmacists Association J Pharm Sci consistent size distribution profiles. Keywords: aerosols; pulmonary drug delivery; orbital; high dose dry powder inhaler; aerosol performance; formulation; powder technology; particle size; puck orifice geometry
INTRODUCTION Dry powder inhalers (DPIs), nebulizers, and pressurized metered dose inhalers have been developed for over 30 years for the treatment of various pulmonary diseases.1,2 The principle of these inhalation devices has focused on delivery of inhalable medications for the treatment of asthma and chronic obstructive disease that usually require a small quantity of active pharmaceutical ingredient (API) per delivered dose.3–5 Various low-dose DPI inhaler devices have been successfully developed and marketed over the years and a number of commercial products contain doses ranging from tens of micrograms (e.g., formoterol 6–12 :g; tiotropium 18 :g; salbutamol 100 :g; beclomethasone 250 :g) up to a few milligrams (e.g., nedocromil and sodium cromoglycate 2–5 mg). Suitable to deliver medicaments within the microgram range, such as corticosteroids6 and $2 agonists,7 these devices lack the ability to deliver APIs requiring higher milligram quantities (i.e., 10–1000 mg), such as antibiotics8 and mucociliary clearance R is the reagents, such as mannitol.9 Currently, TOBI Podhaler only commercially available antibiotic (tobramycin) DPI with a R , the only highhigh-delivery dose (112 mg) and Bronchitol dose mucociliary clearance therapy containing mannitol as the active ingredient (400 mg). However, the delivered dose for
Correspondence to: Paul M. Young (Telephone: +61-2-9114-0350; Fax: +61-29351-4391; E-mail: [email protected]) Journal of Pharmaceutical Sciences C 2015 Wiley Periodicals, Inc. and the American Pharmacists Association
both these products is achieved by individually dispersing multiple capsules (4 × 28 mg in the case of tobramycin and 10 × 40 mg in the case of mannitol) through a multiple “loadprime-inhale” routine. This type of aerosol delivery strategy is time-consuming and can potentially result in poor therapy adherence.10 Mannitol, used as an inhaled dry-powder osmotic reagent, has been proven to be effective in the treatment of diseases where hyper-mucus production is prevalent, such as cystic fibrosis and bronchiectasis.11,12 For example, patients with bronchiectasis showed marked improvement in mucus clearance in the central and intermediate lung regions (>20%) 75 min postinhaled mannitol intervention.9 The solid content in sputum of CF patients was significantly reduced by approximately 2%13 and forced expiratory volume increased by approximately 100 mL14 with 2-week inhaled mannitol therapy. Although this unique therapy provides a means of enhancing clearance, as with other high-dose therapies, the delivery vehicle is based on a DPI that was originally designed for lower dose therapeutics, thus requiring a patient to load, actuate, and inhale from multiple capsules. Recently, Pharmaxis developed a disposable, multibreath DPI device able to deliver large doses of active ingredients (500 mg) through a number of inhalation maneuvers.15,16 The R device (Fig. 1) consists of four major components— Orbital (1) a mouthpiece, (2) dispersion grid, (3) dispersion chamber with multiple tangential airstream inlets, and (4) a “puck” or aerosol sample compartment. During inhalation, the puck rotates within the dispersion chamber and releases the Zhu et al., JOURNAL OF PHARMACEUTICAL SCIENCES
1
2
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
HPLC grade and purchased from Chemsupply (Port Adelaide, Australia). Scanning Electron Microscopy The morphology of mannitol was visualized using a JEOL-6000 bench-top scanning electron microscope (SEM; JEOL, Tokyo, Japan). Prior to imaging, mannitol samples were deposited onto carbon tape mounted on an aluminum stub and coated with gold using a sputter coater (DII-29010SCTR; JEOL) at a coating thickness of approximately 15 nm. Imaging was conducted at an accelerating voltage of 5 keV. Particle Sizing
R Figure 1. Schematic of the prototype Orbital device.
formulation through precision-engineered orifice(s). Previous studies15,16 have demonstrated that the device can successfully deliver high payloads of cohesive ciprofloxacin/mannitol and azithromycin/mannitol spray-dried formulations (200 mg) via approximately 10 inhalation maneuvers (equivalent to the number of capsules that would be required in conventional DPIs), while obtaining respiratory fractions greater than 60%. In these previous studies, the authors focused on engineering and characterization of novel spray-dried powders for infection R to be verand bronchiectasis. They demonstrated the Orbital satile for the delivery of a number of molecules when using standard prototype device configuration. However, features of the device such as resistance, dispersion grid, and puck dimensions may be modified to “tune” both the delivery profile and aerosol performance. Here, we investigate the emptying properties and aerosolizaR device as a function of puck tion performance of the Orbital hole number and geometry using mannitol as the model API, R for mucus clearance. similar to that used in Bronchitol
MATERIALS AND METHODS Materials Spray-dried mannitol (Batch ID: EXP 280) was supplied by Pharmaxis Ltd. (Sydney, Australia) and used as received. R DPI devices, including pucks with differPrototype Orbital ent orifice configurations (Table 1), were supplied by Pharmaxis Ltd. Water was purified using reverse osmosis (Merck Millipore, Bayswater, VIC, Australia). All other solvents were
The volume median size (D0.5 ) of the spray-dried mannitol was measured by laser diffraction using a Mastersizer 3000 (Malvern Instruments, Malvern, UK) equipped with hydrodispersion unit. Samples were prepared by suspending approximately 40 mg of spray-dried mannitol in 5 mL of cyclohexane, which was then added dropwise to the hydrodispersion unit, operating at a stirring speed of 2000 rpm, until 5%–15% obscuration was achieved. Samples were measured with a density setting of 1.51 g/cm3 and refractive index of 1.33. Each sample was analyzed in triplicate. Multibreath-Delivered Dose Study The puck and puck orifice is a key parameter that influences the emptying and aerosolization performance of the Orbital. By modifying the puck hole geometry, number of holes, and hole size, it becomes possible to modify the rate at which the powder is dispersed and thus the number of breaths required to achieve a target dose. In the current study, different types of pucks (Table 1) were loaded with 400 ± 0.8 mg (n = 3) spray-dried mannitol and assembled in the Orbital inhaler device, to simulate standard treatment.12 Because of the high payload delivered, a multistage liquid impinger (MSLI; Westech, Bedfordshire, UK) equipped with a United States Pharmacopeia (USP) induction port was used to collect the emitted aerosol to avoid filter clogging. The airflow rate through the assembly was calibrated at 60 ± 2 L/min using a flowmeter (TSI 4040; TSI Instruments Ltd., Shoreview, Minnesota). The assembled Orbital device was inserted into a silicon mouthpiece adaptor, mounted onto the USP induction port, and powder dispersed at 60 L/min over a 4-s period to simulate one breath.17 The weights of the assembled inhaler were recorded before/after each actuation/breath and shot weight of each actuation deemed as the weight difference (mg) of the two consecutive dispersions. The process was repeated with intermediate airflow calibration (60 L/min) for each shot until the difference between two actuations were less than 0.2 mg. The experiment was conducted in triplicate for each puck type.
Table 1. Puck Orifice Configurations Puck ID Orifice Shape Orifice Number 1 P-002 2 P-003 3 P-013a 4 P-028 5 P-024 6 P-021 a
Rectangular Rectangular Spherical Spherical Spherical Spherical
1 1 1 1 2a 2a
Orifice Geometry (mm) 0.5 × 0.77 (width × length) 0.5 × 0.9 (width × length) 0.6 (diameter) 0.9 (diameter) 0.4 (diameter) 0.55 (diameter)
Orifices at opposite sides of puck.
Zhu et al., JOURNAL OF PHARMACEUTICAL SCIENCES
Aerosol Particle Size Distribution Measured by Cascade Impaction In addition to delivered dose, puck geometry may also have an impact on the aerosol performance of the emitted aerosol cloud. Therefore, the aerosolization efficiency of the Orbital inhaler device with the different pucks was investigated using a MSLI. In brief, pucks (Table 1) were filled with 400 ± 0.5 mg of spray-dried mannitol (n = 3) and assembled in the Orbital device. The formulation was dispersed into the MSLI at 60 ± 2 L/min for 4 s per actuation (n = 3) until the puck was DOI 10.1002/jps.24458
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
3
empty. The weight of the inhaler was recorded before/after each dispersion, and the puck was considered empty when the weight difference between two consecutive actuations was 0.2 mg. The puck, Orbital device, and Orbital mouthpiece were weighed before and after the test, and the amount of mannitol deposited on each part was calculated by weight difference. The MSLI USP induction port (Throat), stages, and filter were thoroughly rinsed with deionized water (recovery solution). Samples were collected for chemical quantification using high-performance liquid chromatography (HPLC). All experiments were conducted in triplicate.
dex of 1.33. Post-throat aerosol characteristics of each shot were determined by the median volume diameter (D0.5 , :m) and particle size distribution and temporal laser obscuration (%) profiles reported over the measurement period.
Chromatographic Methodology
Mannitol Particle Characterization
Mannitol concentrations of the samples collected from the MSLI were quantified using a validated HPLC method. The system consisted of a Shimadzu HPLC equipped with an LC20AD pump, RID 10A refractive index detector, SIL 20A autosampler, DGU 20A degassing unit, and CTO 20A column oven (Shimadzu, Rydalmere, Australia). Samples (100 :L) were R RCU-USP Sugar Alcohols LC column injected into a Rezex 2 (8 :m, 250 × 4 mm ; Phenomenex, Lane Cove, Sydney, Australia) and analyzed at a flow rate of 0.4 mL/min with a column oven temperature of 85°C, using degassed HPLC grade water as mobile phase with a retention time of approximately 6.5 min. The linearity of mannitol standard solutions was confirmed in between 0.1 and 3.0 mg/mL, with a regression value 0.999. The limit of quantification was 0.1 mg/mL and accuracy reported to be 99.5%–99.8% at target levels of 0.1, 1.0, and 3.0 mg/mL.
Figures 2a and 2b show scanning electron micrographs of the mannitol particles. In general, the spray-dried mannitol had a spherical morphology with crystalline-like surface characteristics. The cumulative volume size distribution is shown in Figure 2c. The sample presented a distribution profile with a size range of approximately 1–7 :m, with reported D0.1 , D0.5 , and D0.9 values of 1.7 ± 0.01, 3.1 ± 0.05, and 5.4 ± 0.13 :m, respectively (n = 3 ± SD).
Statistical Analysis Data were analyzed using one-way analysis of variance (ANOVA) followed by the Tukey’s post-hoc test. Differences were considered statistically significant at a level of p < 0.05.
RESULTS AND DISCUSSION
Multibreath-Delivered Dose Study Compared with conventional capsule-based DPI devices, the Orbital achieves high doses through multiple breaths from the puck loaded with a premetered mass of active ingredient.
Shot by Shot Evaluation Using In-Line Laser Diffraction Cascade impaction/impinger studies can only realistically provide information regarding the overall aerodynamic performance of the emitted aerosol clouds after multiple inhalation maneuvers as the laborious nature of this method makes it impractical to conduct shot by shot studies. Although the ultimate cumulative aerosol particle size is the key attribute for the Orbital device, it is worthwhile understanding how aerosolization performance varies for each inhalation maneuver. SubseR quently, an in-line laser diffraction method using a Spraytec particle sizer (Malvern Instruments) was developed and used to test the aerosol performance of the device with each puck type. Other authors have previously reported this method as a fast-screening tool for aerosol performance analysis.18 In brief, a USP induction port was fitted onto the inlet of R inhalation cell and outlet connected to an MSLI, the Spraytec with the configuration used for shot weight and cascade impaction studies. The inspiratory airflow through the entire assembly was set to 60 ± 2 L/min using a GAST pump and solenoid flow controller. The MSLI, in this case, was only used for aerosol collection purpose to avoid filter clogging and impactor device saturation, as previously mentioned. Pucks described in Table 1 were filled with 400 mg of spraydried mannitol and assembled in the Orbital inhaler device. The assembly was set up at the above-mentioned airflow rate and the solenoid was actuated for 4 s. The process was repeated until the weight difference between two consecutive actuations were below 0.2 mg. The setup was disassembled and cleaned after each actuation to ensure a good signal-to-noise ratio. Measurements were conducted at a sampling frequency of 2.5 kHz, with particle density of 1.51 g/cm3 and refractive inDOI 10.1002/jps.24458
Figure 2. (a) Low magnification, (b) high magnification SEM images, and (c) volume particle size distribution data for the mannitol inhalation powder. Zhu et al., JOURNAL OF PHARMACEUTICAL SCIENCES
4
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Figure 3. Cumulative emitted dose from the Orbital DPI at 60 L.min−1 (4 L) as a function of shot number for each puck hole geometry. n = 3 ± SD.
Patients receive high doses from sequential inhalation maneuvers and dispose of the device when the puck is empty. Consequently, it is important to understand the influence of puck hole number and geometry on the delivery efficiency of Orbital when operated under standard conditions (4 L inhalation volume– 60 L/min over a 4-s period17 ). Figure 3 shows the cumulative emitted dose of mannitol released from Orbital DPI with pucks containing different hole numbers and geometries. The dose released per actuation for all puck types was linear until approximately 300 mg was emitted, after which the emitted dose plateaued. Importantly, the release profile was puck-type dependent. For example, pucks with a 0.5 × 0.9 mm rectangular orifice geometry (Puck ID #P003) delivered 321.8 ± 1.5 mg (n = 3) of mannitol within the first two inhalation maneuvers, accounting for 80.5 ± 0.9% of the loaded mass. In comparison, 80% of the loaded dose was delivered over 11 shots with the puck containing two 0.4 mm spherical orifices (Puck ID #P-024). In general, high cumulative emitted dose (>300 mg) could be achieved within three inhalation maneuvers, using pucks with large orifice geometries (i.e., 0.9 mm diameter single spherical, 0.5 × 0.9 and 0.5 × 0.77 mm rectangular), whereas emitted dose of pucks with small orifice diameters were significantly lower (approximately 50 mg per shot), having prolonged dose release characteristics. In order to understand the puck-emptying profiles, the cumulative emitted dose as a percentage of mass loaded into the puck (400 mg) was plotted (Fig. 4a) and linear regression conducted between 20% and 80%. In general, regression analysis produced R2 values 0.99 and the slope ranged from 7.5 ± 0.3% emitted/shot for the 0.4-mm hole puck with two holes to 40.2 ± 0.2% emitted/shot for the 0.5 × 0.9 mm hole puck. This equates to a “dose per breath” range of approximately 7.5%–40% of the loaded dose (in this case 400 mg). Importantly, the mass release and emptying rate of mannitol per actuation was consistent for each puck type over the dose range. This is reflected by the small standard deviations between replicate measurements for any given puck at any specific shot number. Such consistency could be qualitatively explained by the “mass dependency theory.”16 In brief, the dose released from the puck is controlled by the hole diameter and Zhu et al., JOURNAL OF PHARMACEUTICAL SCIENCES
Figure 4. (a) Cumulative percent powder released from puck (n = 3 ± SD). (b) Relationship between emptying rate (% of 400 mg loaded dose) and puck hole diameter (n = 3). Y error bars are standard deviation in slope and X error bars are tolerance in puck orifice geometry (±0.025 mm).
remaining mass in the puck, which determines the puck rotation speed. When the remaining mass in the puck is high, a large amount of powder exits the puck orifice at a relatively slow puck rotation speed. As the dispersion continues, the formulation in the puck is released through the orifice at a higher rotation frequency, compensating for the reduced release mass per puck rotation. It is also worth noting that the emptying rate was dependent on the primary orifice diameter rather than the geometry or number of holes (i.e., hole length for rectangular orifices). A plot of emptying rate versus orifice diameter for all puck types showed a linear relationship (R2 of 0.95) with increased emptying rate being observed with increased orifice diameter. The puck hole diameters used in the analysis included the diameter of the single orifice (if dual orifice) and length (i.e., the greater of the two dimensions of a rectangle) for the rectangular orifices. Furthermore, the emptying rates of pucks with dual orifices were primarily determined by the diameter of the single holes instead of the number of orifices, which is supported by the similar dose release data of pucks P-013a (single orifice, 0.6 mm in diameter) and P-021 (dual orifice, 0.55 mm in diameter) (Table 1). This phenomenon is possibly because of the powder dynamic in the puck during the dispersion. As mentioned previously, the puck rotates in the dispersion chamber during the dispersion; therefore, the powder may subsequently rotate within the puck in the same manner. A certain aliquot of powder can be released because of centrifugal force only when the formulation reaches the exiting orifice. As orifices were DOI 10.1002/jps.24458
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
5
diagonally configured on the puck, it is likely that the majority of the formulation is only released from one orifice on one side per puck rotation, resulting in dose dependency in orifice diameter rather than number of holes on the puck.
Aerodynamic Performance and Delivery Efficiency In addition to the emitted dose, the aerodynamic characteristics of the aerosol cloud also play a key role in determining local drug concentration within the lung19 and thus the bioavailability of administered medication.7 As mannitol is used as the osmotic reagent to increase the water efflux to the airway lumen,12 an appropriate concentration at the target site is required for the desirable therapeutic effects. The overall aerosolization performance of mannitol particles delivered via the Orbital device from multiple inhalation manoeuvres were tested using different puck configurations (Table 1). Mass deposition on each MSLI stage was expressed as the percentages of the total drug mass recovered as shown in Figure 5. All the testing conditions showed low mass retention in the pucks (
