Dose administration maneuvers and patient care in tobramycin dry powder inhalation therapy

ABSTRACT The purpose of this work was to study a new dry powder inhaler (DPI) of tobramycin capable to simplify the dose administration maneuvers to maximize the cystic fibrosis (CF) patient care in antibiotic inhalation therapy. For the purpose, tobramycin/sodium stearate powder (TobraPS) having a high drug content, was produced by spray drying, characterized and the aerodynamic behavior was investigated in vitro using different RS01 DPI inhalers. The aerosols produced with 28, 56 or 112mg of tobramycin in TobraPS powder using capsules size #3, #2 or #0 showed that there was quasi linear relationship between the amount loaded in the device and the FPD. An in vivo study in healthy human volunteers showed that 3–6 inhalation acts were requested by the volunteers to inhale 120mg of TobraPS powder loaded in a size #0 capsule aerosolized with a prototype RS01 device, according to their capability to inhale. The amount of powder emitted at 4kPa pressure drop at constant air flow well correlated with the in vivo emission at dynamic flow, when the same volume of air passed through the device. The novel approach for the administration of 112mg of tobramycin in one capsule could improve the convenience and adherence of the CF patient to the antibiotic therapy.


Introduction
Cystic fibrosis (CF) is a genetic rare disease caused by mutations in the gene coding the CF transmembrane conductance regulator protein leading to viscous mucus presence in airways (Moskowitz et al., 2005;Sheppard and Nicholson, 2002). Thus, CF patients are susceptible to pulmonary infections caused mainly by Pseudomonas aeruginosa (PA). Three approaches to the management of infections in CF patients are suggested by the European consensus guidelines (Flume et al., 2007): (i) prophylactic therapy for the prevention of infection and colonization; (ii) intravenous therapy for acute pulmonary infections and (iii) maintenance therapy by inhalation to prolong the interval between exacerbations. Thus, inhalation antibiotic therapy is recommended in exacerbation prevention of PA infection (Doring et al., 2000). Despite different antibiotics are available on the market (aztreonam lysine, colistimethate sodium and levofloxacin) (Buttini et al., 2016;Hewer, 2012), the aminoglycoside tobramycin remains standard to manage PA infection in order to delay exacerbation infection episodes (Cheer et al., 2003;Ratjen et al., 2009). Local concentrations of tobramycin in the lung of patients higher than intravenous administration is the objective of inhalation administration. The antibiotic deposition in the lung does not elevate the plasma drug concentration and, consequently, drug toxicity is limited (Weers, 2015). A tobramycin dry powder inhaler (TOBI TM Podhaler TM , Novartis) has been authorized for the use in CF patients.
This tobramycin dry powder inhaler introduced a significant advantage for patient compliance and quality of life, compared with nebulization. This method has been demonstrated satisfactory for PA infection management (Geller et al., 2007), since the inhalation of 112 mg of tobramycin was clinically equivalent to 300 mg of drug administered by nebulization.
From the technological point of view, the large amount of powder (200 mg), carrying 112 mg of tobramycin, could not be loaded in a single hard capsule reservoir of the device and inhaled in one inhalation act. Consequently, in TOBI TM Podhaler, the powder amount was shared in four hard capsules size #2, each one loading ~50 mg of PulmoSphere TM powder (Galeva et al., 2013). The inhalation manoeuvres, i.e., to load one capsule into the inhaler, pierce it and inhale twice, have to be repeated four times to administer the entire dose. The therapy has to be done two times per day over 28 days. Despite the advantage for patient towards nebulization, the administration procedure significantly affected the patient adherence to the therapy (Boerner et al., 2014).
These facts evidence that there is a need of novel powders and devices for better managing the lung administration of high dose antibiotics.
Regarding the devices, Twincer®, a disposable inhaler containing two air classifier technology, was developed with the aim to release up to 25 mg of unprocessed micronised particles or soft spherical agglomerates without special particle engineering processes (De Boer et al., 2006). Based on the same principle, Cyclops, a single classifier version of Twincer was developed to aerosolize up to 50 mg of pure spray-dried amorphous aminoglycosides (Hoppentocht et al., 2015). FPFs (at 34 L/min) of tobramycin, amikacin or kanamycin with the Cyclops ranged 78-90%. The Orbital device is another single-use, disposable unit containing a ''puck'' that holds up to 400 mg of drug powder. A key innovative step in the Orbital design is the puck orifice that acts as the rate-limit step for amount of released powder. This system showed to be capable releasing efficiently fixed dose over a series of inhalation acts of both amorphous, crystalline and co-spray dried powder systems (Young et al., 2014a). These include 400 mg of mannitol, 100-400 mg doses of spray-dried ciprofloxacin/mannitol, 200 mg combination co spray-dried azitrohromycin/mannitol (Young et al., 2014b), 200 mg tranexamic acid (Haghi et al., 2015)and 200 mg of micronized crystalline tobramycin (Zhu et al., 2016). Finally, the fluidized bed DPI is another novel multi-breath high dose DPI which consists of a formulation reservoir (dosing sphere) that contains up to 100 mg of powder which is released during the inhalation act via 2 to 6 dosing holes into a fluidized device bed. Using this design, the FPF of co-spray dried mannitol and ciprofloxacin reached 93% (Farkas et al., 2015).
On the formulation perspective, a novel tobramycin inhalation powder formulated with a lipophilic adjuvant has been described (Buttini et al., 2010). The tobramycin powder microparticles, constructed by spray drying, using a minimal amount of sodium stearate, exhibited a very small aerodynamic diameter, low density and favourable shape for aerosolization. Sodium stearate, a surface-active substance, molecularly coated the tobramycin microparticles, due to the preferential accumulation at the droplet air interface during drying. Microparticles with sodium stearate on surface showed a great aerosol performance. Moreover, the excipient provided protection from the environmental humidity determining a superior stability (Parlati et al., 2009), as also recently confirmed by (Yu et al., 2018).
The high drug content in tobramycin/sodium stearate microparticulate powder helps in capturing the patient adherence during PA infection management, if formulation/device combination provides a patient centric product. The driving hypothesis of this research was that the time for the daily antibiotic administration would be minimized by using a highly respirable and high drug content formulation, provided to have a device capable to regulate the amount of powder emitted. Hence, the aim of this work was to study a new dry powder inhaler of tobramycin capable to simplify the dose administration manoeuvres and to maximize the patient care in antibiotic inhalation therapy.

Tobramycin microparticles and agglomerates manufacturing
Tobramycin dissolved with 1% w/w of sodium stearate was spray dried accordingly to (Parlati et al., 2009). Briefly, 4.95 g of tobramycin was dissolved in 350 ml of purified water at 30°C whereas 0.05 g of sodium stearate was dissolved in 150 ml ethanol 95° at 30 °C. The two solutions were mixed and the final solution with a solid content of 1% w/v, kept at 30°C, was spray-dried using a Buchi B-290 (Buchi, Flawil, Switzerland). The spray-drier conditions were: feed rate 3 ml min −1 , aspiration rate 100%; air flow rate 600 l h −1 , inlet and outlet temperatures 125°C and 75-78°C, respectively. The spraydried powder, coded TobraPS, was used as it is after the production or transformed in agglomerates by mechanical vibration as previously described (Belotti et al., 2014). In detail, 5 g of powder were placed on a stack of two sieves of 600 and 106 μm size and vibrated using a sieve shaker for size analysis (Fritsch GmbH, Oberstein Deutschland) at the amplitude 3 for 5 minutes in a cabinet under nitrogen atmosphere. The agglomerates in the size range 106-610 µm were collected, stored in sealed glass vial at 25°C-60% RH. The yield of the process was higher than 70%.

Powder characterization 2.3 Assay of tobramycin
The assay of tobramycin was performed by the USP 37 HPLC method. System suitability was performed according to USP 37. The method precision (Relative Standard Deviation calculated following six injections of a 0.5 mg/mL standard solution) was 0.85% and the linearity was in a range from 0.1 to 1.5 mg/mL (R 2 = 0.9945). LOD and LOQ values were 0.02 mg/ mL and 0.06 mg/mL respectively.

Particle size distribution
Particle size distribution of TobraPS powders was measured using the laser light scattering apparatus (SprayTec, Malvern Instruments Ltd, UK). Approximately 10 mg of the sample was dispersed in 20 mL solution of 0.1% (w/v) Span 80 in cyclohexane and sonicated for 5 min. The particle size distribution was measured in triplicate with an obscuration threshold of 10%. Data were expressed in terms of median volume diameter and percentiles, D(v,0.1), D(v,0.5), D(v,0.9).

Loss on drying
The residual solvent of powders was determined as loss on drying with thermo-gravimetric analysis. The instrument used was the Mettler Toledo Thermogravimetric Analyzer (Mettler Toledo, Switzerland). About 5 mg of powder was introduced in a 70 µL alumina pans with a pierced cover and analysed from 25°C to 250°C at a heating rate of 10°C/min under a nitrogen stream flowing at 80 ml/min. The weight loss was measured in the range between 25 -125 °C.

Crystallinity
The study the crystal form of tobramycin powders was performed by X Ray Powder Diffraction analysis. The instrument employed was a MiniFlex X-Ray Diffractometer (Rigaku, Japan). About 200 mg of powder were loaded on the sample older and then analysed from a start angle of 2 θ to an end angle of 35 θ with 0.5 θ steps.

Scanning Electron Microscopy
TobraPS powder morphology was assessed by Scanning Electron Microscopy (SEM) (JSM-6400, JEOL Ltd., Japan) at high magnifications (10000x -30000x) with a EHT of 1.60 kV. The samples were placed on a double-sided adhesive tape pre-mounted on an aluminium stub and analysed after a 30 min depressurization.

Dissolution profile
The dissolution rate of tobramycin powders was measured using a Franz cell constituted by a donor compartment separated from receptor compartment by a disc of high pure filter paper (Albet LabScience 84 g/m -2 ). The receptor compartment has a sampling port that allows to collect the samples for tobramycin HPLC determination.
The Franz cell receptor was filled with 20 ml of degassed purified water by reverse osmosis and the filter paper disc was wet with 0.5 ml of purified degassed water to create a thin film above it. The Franz cell was thermostated at 37°C; a stir-bar rotating at 200 rpm was introduced in the receptor compartment and the absence of air bubble in the cell under the membrane was verified.
Approximately, 10 mg of powder TobraPS accurately weighed, were distributed on the filter paper.
In the case of TOBI TM Podhaler 10 mg of powder were sampled from the capsule content.
At time zero, through the sampling port, 2 ml of receptor solution was taken with a syringe and analysed by HPLC. After each sampling, the cell was refilled with an equivalent volume of purified degassed water. The sampling were performed at prefixed time of 5, 10, 15 minutes. Each powder was tested in triplicate. Plates were read, after incubation at 37°C for 18-24 hours in moistened air, with the aid of a magnifying mirror. The test was considered valid when acceptable growth has occurred in the antibiotic-free growth control well and no contaminant growth was present in the medium sterility control well.

Activity of tobramycin formulation on
The MIC was defined as the lowest concentration of antimicrobial at which there is no visible growth of PA ATCC 27853 after overnight incubation. The MBC was defined as the lowest concentration of antibacterial agent that reduced the viability of the initial bacterial inoculum by >99.9%.

In vitro aerodynamic drug deposition
The dispensing and dispersion performance assessments were performed using Next Generation Impactors (NGI) (Copley Scientific, Nottingham, UK). The methodology employed followed USP38 General Chapters, Physical tests and determinations for dry powder inhalers (Apparatus 5).
The collection stages were coated with Span 85 in cyclohexane solution (1% w/v) to prevent particle bounce. Powder formulations were aerosolised by the inhaler(s) while attached to the NGI and the tobramycin retained in the capsule, inhaler, and impactor was collected using a H2SO4 The DPI devices, capsule size, intrinsic resistance and flow rate adopted is reported in Table 1.  years old respectively) in order to assess the number of inhalation acts required to inhale 120 mg of TobraPS powder loaded in a size #0 capsule and aerosolized by RS01 prototype device. The amount of powder extracted after each inhalation was also measured. The work has been carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki) and a signed informed consent was obtained by each volunteer after carefully explanation of the purpose of the study and its execution.
Preliminary, the capsule was inserted in the device but not pierced, and the volunteers were instructed to inhale twice the capsule with a deep single breath with an inspiratory flow sufficient to hear the capsule spinning. In such a way, subjects become confident with the inspiratory effort required to lift and rotate the capsule in the device. The minimum airflow suitable for this was in vitro experimentally measured to be > than 20 L/min. The effective capsule rotation was easy to verify by listening to the noise produced during capsule spinning.
After the training session, volunteers were requested to inhaled TobraPS powder and an apparatus suitable for the measurement of airflow rate produced by the inhalation act during the dose extraction was an in-house developed (Fig 1). The latter consists in a 1.25 L chamber in which the RS01 was tightly inserted, leaving the mouthpiece out. The chamber air was supplied through a and the air volume inhaled. Then, the RS01 prototype device was loaded with a size #0 capsule containing 120 mg of TobraPS agglomerated corresponding to 112 mg of tobramycin. The capsule was pierced and the device was weighted before to be introduced in the inhaler and sealed in the chamber connected to flowmeter, leaving out the mouthpiece. After the first inhalation act, the amount of powder emitted was measured by weighting the device recovered from the apparatus. Then, the system was reassembled and the procedure was repeated until the capsule was emptied.
The waiting time between two inhalation acts of the volunteers was not less than 1 min.

Statistic
Statistical calculations were performed with the software KaleidaGraph (Synergy Software, U.S).
For statistically significant differences, t-test analysis was performed. Significance level of 5% was set.

Results and Discussion
Powder characterization Using the lab scale spray-drier, several batches up to 10 g of tobramycin solution containing 1% of sodium stearate were dried and the process yields was in the range 68-72%, similar to (Parlati et al., 2009). The powder recovered (TobraPS) from the spray drier collector appeared as irregular large clusters of micronized particles; the drug content was of 91.1% w/w. The microparticle 10%, 50% and 90% size distribution percentiles, measured as D v by laser diffraction, were 1.03 ± 0.21 μm, 2.40 ± 0.36 μm and 5.29 ± 0.91 μm, respectively: the particle size distribution was log normal.
However, the scanning electron images (Fig. 2) showed numerous particles larger than 5 μm, spherical and with smooth surfaces. Apparently, the particles were empty since many of them were inflated and exploded. Their fragments were visible as flat flakes in the particle population.
Curiously, frequently smaller particles were observed inside the cavity of the exploded larger ones.
Together with large particles, there are small ones with less smooth surface; in general, these particles are grouped together or adherent to the largest or to their fragments.
This morphology was recently observed also in spray dried amikacin powders for inhalation prepared in the same conditions and it was attributed to the amikacin solubility in the water-alcohol solutions used for spray drying. Since the alcohol solubility of aminoglycoside is lower than in water, during the drying of sprayed droplet, an early precipitation of drug took place at droplet surface creating a crust that obstructs the solvent evaporation. Thus, the increase of vapor pressure inflated the particle during drying causing its explosion, more frequent when alcohol was present in the drug solution (Belotti et al., 2014;2015). Morphology and volume diameter of tobramycin spray-dried microparticles contributed to the powders favourable characteristics for pulmonary deposition.
X-ray diffraction analysis of TobraPS powder evidenced the amorphous state of spray dried tobramycin powder (see supplementary data). TOBI have an amorphous state (Geller et al., 2011) and to be stable at low relative humidity storage conditions. TobraPS packed in a sealed glass vial showed a chemical stability unmodified at 25°C for 9 months; the physical stability, in particular the aerodynamic performance, showed a significant decrease of 15% during this storage time, hence requesting a specific air tight store conditions. The DSC comparison of the powder stored during this time period showed no variation indicating recrystallization (see supplementary data).
TobraPS powder batches presented an average loss of weight during TGA analysis of 7.93 ± 1.35%, confirmed as water by Karl Fisher titration.
The administration of antibiotic solid particles requires for activity a quick local availability of dissolved drug molecules. For this purpose, the dissolution profiles of TobraPS powders, compared to tobramycin formulation of TOBI TM Podhaler TM , are reproduced in Fig. 3.
Both powders dissolved rapidly and more than 85% of drug was in solution in 15 minutes.
However, in the first ten minutes, tobramycin spray dried powders with sodium stearate reached almost complete dissolution (92.2%), whereas TOBI TM powder dissolution was slightly slower (67.8%). Moreover, in the case of TobraPS, all the powder was dissolved without any visible residue on the diffusion cell membrane where the powder was deposited; differently, with TOBI TM powder left a white solid residue at the end of the dissolution. Likely, some excipients from tobramycin Pulmosphere TM remained on the filter of the cell donor compartment. In summary, the availability at the lung deposition site from TobraPS powder is expected to be prompt and complete.
The microbiological activity of the TobraPS powder was assessed in an experiment in which the powder was dissolved before to contact the microorganisms. The results demonstrated a microbiological activity not different from the non-processed raw material: the MIC value obtained for TobraPS (0.5 µg/ml) was consistent with EUCAST data (Leclercq et al., 2013)   It's well know that the reduction of operation steps and manoeuvres decrease the probability of serious errors with DPI (Voshaar et al., 2014). As well as, it was reported that the patient adherence to the antibiotic therapy in cystic fibrosis PA management was negatively affected by sharing the dose to administer: only 52% of 288 patients enrolled in the study declared to inhale 4 capsules (Boerner et al., 2014). Hence, with the goal to attempt a reduction of the number of capsules constituting the dose, TobraPS was loaded in increasing doses in capsules of augmented size and aerosolized using the appropriate RS01 device.
The set of experiments was started by testing the performance of TobraPS powder with the commercial high resistance (0.033 kPa 0.5 /LPM) device RS01 using a capsule size #3, loaded with 32 mg of (corresponding to 28 mg of tobramycin, i.e., the same dose of TOBI TM ). Due to the TobraPS bulk density (0.17 g/ml), this size of capsule size could easily accommodate the mass of powder. In RS01 device, the capsule, pierced in correspondence of the cap and body extremities, due to the inhalation air flow rate rotated on the main axis inside the device, so centrifuging out the powder.
Conversely in the medium resistance T-326 inhaler (0.025 kPa 0.5 /LPM), has the same powder emission mechanism of Turbospin ® device (PH&T, Milan, Italy), i.e. the capsule pierced at the bottom, rattles and swirls for powder emission (Martinelli et al., 2015). Thus, Table 3  However, it has to be again stressed that, in the case of the TobraPS a consistent lower amount of powder has to be aerosolized to deliver the same dose of TOBI TM , due to the high drug content.
Aiming at the reduction of the number of capsules, the amount of powder correspondent to the entire dose (112 mg) had to be increased in the capsule. Capsules #3 could not accommodate an amount of TobraPS powder larger than 40 mg without compacting the powder, thus affect the aerodynamic performance. Hence, the switch to the use of larger capsule was a forced choice.
Capsule #2 and #0 were adopted and the content aerosolized using novel RS01 inhalers, having the same emission mechanism, but capable of accommodate the larger capsule size.
Thus, 56 mg of formulation, that would correspond to the entire dose shared in two capsules, were loaded in capsule #2. The results of the aerodynamic assessment of this capsule are illustrated in Table 3. The fine particle dose obtained after inhalation of 56 mg tobramycin was approximately doubled, resulting 32.0 mg. These results could make possible to propose the halving the entire dose from four to two capsules to be inhaled by the patients. Table 3. Aerodynamic assessment of TobraPS using RS01 devices and capsules of different size (4 L of air through the device at flow rate corresponding to the pressure drop of 4 kPa).
* powder agglomerates; # determined by weighing The therapy could reach a more relevant benefit in term of convenience and adherence by the patient if the dose could be administered employing only one capsule for the total dose delivery. Therefore, a RS01 device capable to accommodate a capsule size 0 or 00 was tested for delivery of high payload formulations (Parumasivam et al., 2017). A similar prototype with a resistance of 0.027 kPa 0.5 /LPM was employed in this work in order to administer 112 mg of tobramycin (as 120 mg of TobraPS), the equivalent of 4 capsules 28 mg each.
It was not possible to load the entire dose of microparticles in one capsule size #0, unless an agglomeration process were applied in order to increase the powder bulk density. Agglomerates are soft pellets in which the microparticles are hold together by weak interaction forces (Russo et al., 2004); they can be destroyed into the device by the air turbulence produced by patient inhalation.
This technology permitted to introduce 120 mg of formulation inside the capsule #0. The agglomerates or pellets are free flowing and the homogeneity of the powder was substantially increased, so facilitating the loading of the capsule to inhale.
In details, approximately 120 mg of agglomerated TobraPS were loaded in the capsule #0 and the entire content was aerosolized in one shot of 4 L of air at pressure drop of 4 kPa. The FPD value obtained was 57.6 ± 1.6 mg (Table 3). This value was not far away from the FPD accumulated from four capsules containing 28 mg of TobraPS each aerosolized with RS01 size #3 that was 66 mg (16.5 mg x 4). More interestingly, the value was similar to TOBI TM Podhaler TM FPD that, following the emission of four capsules, was 58.8 mg. The FPD from one capsule with RS01 size #0 corresponded to 97.8% of the expected dose from four capsules with Podhaler TM .
The inclusive comparison of the fine particle dose of the aerosols produced with 28, 56 or 112 mg of tobramycin in TobraPS powder and agglomerates with the RS01 device using capsules size #3, #2 or #0 (Table 3) shows that the FPD at the same 4 kPa pressure drop slightly decreased by increasing the amount of drug in the capsule. The FPD of TobraPS aerosolized with the RS01 devices was plotted versus the different amounts of powder and compared with the FPD of TOBI TM Podhaler TM (Fig. 4). There is a quasi linear relationship (R 2 =0.99442) between the amount loaded in RS01 devices and the FPD. The "position" in the graph of TOBI TM Podhaler TM illustrates the difference existing between this formulation and TobraPS.
The comparison among the three different doses has been tested also by using the Fast Screening Impactor that could allow to determine the amount of drug remaining in the device, in the IP, emitted as coarse fraction and as fine fraction (less than 5 µm). It was confirmed that the fine particles approximately doubled by doubling the dose, such as the coarse fraction. The amount of formulation remaining in the device was also correlated with the amount loaded, but not the amount deposited in the IP (Fig. 5).

In vivo powder emission after multiple inhalation acts
TobraPS cumulated in one capsule size #0 could be extracted by the device in vitro with one single actuation at 72 L/min per 3.3 seconds. However, in vivo this amount of powder should be delivered to the patient in successive inhalation acts. Recently, the antibiotic colistimethate sodium powder for inhalation at the dose of 125 mg was introduced in a single capsule in order to be extracted by the user in 6-7 successive inhalation acts (Tappenden et al., 2013). The tobramycin deposition data reported until now have been collected in vitro. In vivo, with the high tobramycin dose, all the powder extracted in one shot would be unfeasible and unsafe. A graduation of the emission could be done by performing successive inhalation acts through the same loaded RS01 device. Therefore, for the in vivo test measurements, the air flow profiles of three healthy volunteers during powder inhalation were recorded, using the device RS01 capsule size #0 loaded with 120 mg of TobraPS.
These obtained profiles of healthy volunteers were paralleled to the amount of drug extracted or emitted and could be compared to CF patient profiles published by Haynes (Haynes et al., 2016). performing the inhalation acts (Fig. 7, blue symbols), was able to extract the drug dose in three-four acts. Volunteer 2 employed four acts (Fig. 6, red symbols), whereas volunteer 3 (Fig. 7, green symbols), inhaling slowly due to personal apprehension, required from five to six inhalations in order to extract the entire dose. Any evident disconfort has been observed by the three voluteers enrolled in the test.
In order to identify the determinant of the dose emission in correspondence on each inhalation act, the emitted dose after inhalation from the different volunteers and successive replicas were plotted versus the PIF value of the inhalation profile (Fig. 8a). For assuring the comparison in the same conditions, only the first inhalation act of each volunteer was considered. There was a quite expected dependence of the emitted amount from PIF, albeit the significance of the linear relationship was low (R 2 = 0.36857). Despite the scattering of the values, this result underlines the relevance of PIF for the dose extraction. The PIF activated the rotation of the capsule in RS01 device, but PIF could not be the only determinant for powder emission from device. The flow profile during powder emission allows the calculation of the area under the curve, i.e., the volume of air passed through the device during the inhalation act. The measured volumes were plotted versus the emitted dose (Fig. 8b). This plot shows a more significant linear relationship between the powder amount emitted and the volume of air passed through the device, at PIF values higher than 20 L/min. This aspect, noticed by other authors (Sosnowski, 2018;Weers and Clark, 2017), supports the importance of the air volume for the quantitative emission of powder. In fact, the linearity of the relationship between powder emitted at first inhalation versus air volume inhaled, supported the relevant role of air volume, together with the PIF presented.
In summary, using RS01 device, that has a mechanism of powder emission dictated by the rotation of the pierced capsule in a spinning chamber, the air flow through the device guarantees the lift of the capsule and its rotation at threshold values of flow rate higher than 20 L/min. Increasing the PIF value, the amount emitted increased, but the relationship was less predictable. There was a stricter correlation between the amount emitted and the volume of air passed through the device. This meant that the device emission was not only dependent on the air PIF, but the dose extracted at low PIF depended also by the duration of the inhalation act. Repetition of the inhalation act on the same capsule make more reliable the amount of drug inhaled.

Conclusions
The increase of tobramycin content in TobraPS inhalation dry powder contributed to a significant reduction of the mass of powder entering the lung of patient for administering the prescribed dose of tobramycin. In the TobraPS tobramycin strength 112 mg/capsule, the amount of formulation to inhale is around 120 mg, approximately 40% less than the reference approved product (TOBI Podhaler).
The RS01 emission mechanism, based on the spinning of the reservoir capsule, demonstrated to be able to control the amount of powder emitted during the inhalation act. The number of acts to perform in order to inhale the entire high dose of powder is dependent on the capability of the patient to inhale. A minimum air flow rate value of 20 L/min through the inhaler was requested to make the capsule lifting and rotating. The amount of powder emitted at 4 kPa pressure drop, at constant air flow rate, well correlated with the in vivo emission in dynamic flow conditions, when the same volume of air passed through the device.

Figure Captions
Fig. 1. System in-house assembled capable to register the inhalation profile during the powder extraction by the volunteer. TobraPS powder was loaded in a capsule, inserted in RS01 size 0 device that was sealed inside a volumetric chamber leaving only the mouthpiece outside. When the volunteer was inhaling the supplied air passed throw a flowmeter capable to register the air velocity ant to transfer the data to a software.         Tobramycin powder production and characterisation

In vivo dose extraction by volunteers
In vitro / in vivo delivered dose correlation TobraPS spray-dried powder (tobramycin:sodium stearate 99:1) RS01 mouthpiece Flowmeter Volumetric Chamber