AU2007202942B2 - Methods and unit dose formulations for the inhalation administration of aminoglycoside antibiotics - Google Patents

Description

AUSTRALIA Patents Act 1990 CHIRON CORPORATION COMPLETE SPECIFICATION STANDARD PATENT Invention Title: Methods and unit dose formulations for the inhalation administration of aminoglycoside antibiotics The following statement is a full description of this invention including the best method of performing it known to us:- ]A METHODS AND UNIT DOSE FORMULATIONS FOR THE INHALATION ADMINISTRATION OF AMINOGLYCOSIDE ANTIBIOTICS 5 RELATED APPLICATIONS DATA This application is a divisional application of Australian Patent Application No. 2002339338 which is the Australian national phase filing of International Application 10 No. PCT/US2002/015999 (publication No. WO 2002/094217) filed on 17 May 2002, which claims priority from USSN 60/292,234 filed on 18 May 2001, the contents of which are incorporated herein in their entirety by way of reference. 15 FIELD OF THE INVENTION The present invention relates to new and improved unit dose containers of aminoglycoside antibiotics, such as tobramycin, for delivery by aerosol inhalation, and 20 to improved methods of treatment of susceptible acute or chronic endobronchial infections. 25 BACKGROUND OF THE INVENTION Progressive pulmonary disease is the cause of death in over 90% of cystic fibrosis (CF) patients (Koch, C. et al., "Pathogenesis of cystic fibrosis," Lancet 341 (8852):1065-9 (1993); Konstan M. W. et al., "Infection and inflammation of the lung in cystic fibrosis," Davis PB, ed., Lung Biology in Health and Disease, Vol. 64. 30 New York,- NY: Dekker: 219-76 (1993)). Pseudoinonas aeruginosa is the most significant pathogen in CF lung disease. Over 80% of CF patients eventually become colonized with P. aeruginosa (Fitzsimmons S.C., "The changing epidemiology of cystic fibrosis," J Pediatr 122(1):1-9 (1993)). The standard therapy 5 for P. aeruginosa endobronchial infections is 14 to 21 days of parenteral antipseudomonal antibiotics, typically including an aminoglycoside. However, parenteral aminoglycosides, as highly polar agents, penetrate poorly into the endobronchial space. To obtain adequate drug concentrations at the site of infection with parenteral administration, serum levels approaching those associated with .10 nephro-, vestibulo-, and oto-toxicity are required ("American Academy of Otolaryngology. Guide for the evaluation of hearing handicap," JAMA 241(19):2055 9 (1979); Brummett R.E., "Drug-induced ototoxicity," Drugs 19:412-28 (1980)). Aerosolized administration of aminoglycosides offers an attractive alternative, delivering high concentrations of antibiotic directly to the site of 15 infection in the endobronchial space while minimizing systemic bioavailability (Touw D.J. et al., "Inhalation of antibiotics in cystic fibrosis," Eur Respir J 8:1594 604 (1995); Rosenfeld M. et al., "Aerosolized antibiotics for bacterial lower airway infections: principles, efficacy, and pitfalls," Clinical Pulmonary Medicine 4(2): 101 12 (1997)). 20 Tobramycin is commonly prescribed for the treatment of serious P. aeruginosa infections. It is an aminoglycoside antibiotic produced by the actinomycete, Streptomyces tenebrarius. Low concentrations of tobramycin (< 4 ig/mL) are effective in inhibiting the growth of many Gram-negative bacteria and under certain conditions may be bactericidal (Neu, H.C., "Tobramycin: an overview," 25 J Infect Dis 134, Suppl: S3-19 (1976)). Tobramycin is poorly absorbed across mucosal -surfaces, conventionally necessitating parenteral administration. Tobramycin activity is inhibited by purulent sputum: high concentrations of divalent cations, acidic conditions, increased ionic strength and macromolecules that bind the drug all inhibit tobramycin in this environment. It is estimated that 5-to -10 times 30 higher concentrations of tobramycin are required in the sputum to overcome these inhibitory effects (Levy J. et al., "Bioactivity of gentamicin in purulent sputum from patients with cystic fibrosis or bronchiectasis: comparison with activity in serum," J Infect Dis 148(6):1069-76 (1983)). Delivery of the poorly absorbed antibiotic tobramycin to the airway by the 5 aerosol route of cystic fibrosis (CF) patients has been documented using the aerosol route. Much of this work has been done toward treatment of chronic lung infections with P. aeruginosa in cystic fibrosis (CF) patients. A multicenter, double blind, placebo-controlled, crossover trial of 600 mg tid of aerosolized tobramycin for endobronchial infections due to P. aeruginosa in 71 CF patients demonstrated a 10 significant reduction in sputum density of this pathogen as well as improved spirometry in the treatment group. Emergence of P. aeruginosa strains highly resistant to tobramycin (defined as MIC > 128 jig/mL) was comparable in the placebo and treatment groups. The presence in the sputum of Gram-negative organisms other than P. aeruginosa intrinsically resistant to tobramycin occurred 15 with equal frequency during administration of tobramycin or placebo (Ramse'y B. -et al., "Response to Letter to the Editor: Aerosolized tobramycin in patients with cystic fibrosis," NEngi IMed 329:1660 (1993)). Although this regimen was found to be both safe and efficacious, it is costly and inconvenient. A* survey of the MICs for P. aeruginosa isolates from initial 20 sputum cultures for patients at the Children's - Hospital CF Center, Seattle, Washington, in 1993 found that 90% of isolates had MICs 16 pg/mL and 98% of all isolates had MICs 128 tig/mL. This survey suggested that achieving a sputum tobramycin concentration of 128 pg/mL should treat the endobronchial infection in CF patients (Levy J. et al., "Bioactivity of gentamicin in purulent sputum from 25 patients with cystic fibrosis or bronchiectasis: comparison with activity in serum," J Infect Dis 148(6):1069-76 (1983)). A randomized, crossover study compared the ability of several nebulizers to deliver tobramycin by measuring peak sputum tobramycin concentrations in samples collected ten minutes after completion of the aerosol dose. This study administered 30 TOBI@ tobramycin solution for inhalation, PathoGenesis Corporation, Seattle, Washington (now Chiron Corporation, Emeryville, California), containing 60 mg/mL tobramycin in 5 mL one quarter (1/4) normal saline, using the Pari@ LC jet nebulizer, Pari Respiratory Equipment, Inc., Richmond, Virginia. This delivery system was shown to deliver a mean peak sputum tobramycin concentration of 678.8 pg/g (s.d. 5 661.0 pg/g), and a median peak sputum concentration of 433.0 pig/g. Only 13% of patients had sputum levels 128 pig/g; 87% of patients achieved sputum levels of > 128 pg/g (Eisenberg, J. et al., "A Comparison of Peak Sputum Tobramycin Concentration in Patients With Cystic Fibrosis Using Jet and Ultrasonic Nebulizer Systems. Aerosolized Tobramycin Study Group," Chest 111(4):955-962 (1997)). 10 Recently, the Pari@ LC jet nebulizer has been modified with the addition of one-way flow valves, and renamed the Pari@ LC PLUS. The one-way valves in the Pari@ LC, PLUS have been described as permitting the delivery of more drug than the Pari@ LC jet nebulizer, while decreasing the potential for accidental spillage and allowing for the use of an expiratory filter. Experience has shown that mean peak sputum 15 tobramycin concentrations achieved using the Pari LC PLUS jet nebulizer are significantly higher than those using .the Pari@ LC jet nebulizer as described in Eisenberg et al. (1997), supra. Two placebo-controlled, multicenter, randomized, double blind clinical trials of intermittent administration of inhaled tobramycin in cystic fibrosis patients with 20 P. aeruginosa infection were reported in Ramsey, B.W. et al., "Intermittent Administration of Inhaled Tobramycin in Patients with Cystic Fibrosis. Cystic Fibrosis Inhaled Tobramycin Study Group." N. Engl. J. Med. 340(l):23-30 (1999). In these studies, five hundred twenty subjects were randomized to receive either* 300 mg inhaled tobramycin or placebo twice daily for 28 days followed by 28 days 25 off study drug. Subjects continued on treatment or placebo for three "on-off' cycles for a total of 24 weeks. Efficacy variables included sputum P. aeruginosa density. Tobramycin-treated patients had an average 0.8 logio decrease in P. aeruginosa density from Week 0 to Week 20, compared with a 0.3 logo increase in placebo treated patients (P<0.001). Tobramycin-treated patients had an average-1.9 logio decrease in P. aeruginosa density from Week 0 to Week 4, compared with no change in placebo-treated patients (P<0.001). A preservative-free, stable, and convenient formulation of tobramycin (TOBI* tobramycin solution for inhalation; 60 mg/mL tobramycin in 5 mL of 1/4 5 normal saline) for administration via jet nebulizer was developed by PathoGenesis Corporation, Seattle, Washington (now Chiron Corporation, Emeryville, California). The combination of a 5 mL BID TOBI dose (300 mg tobramycin) and the PARI LC PLUS/PulmoAide compressor delivery system was approved under NDA 50-753, December 1997, for the management of P. aeruginosa in CF patients, and remains 10 the industry standard for this purpose. The aerosol administration of a 5m] dose of a formulation containing 300 mg of tobramycin in quarter normal saline for the suppression of P. aeruginosa in the endobronchial space of a patient is disclosed in U.S. Patent No. 5,508,269, the disclosure of which is incorporated herein in its entirety by this reference. 15 Although the current conventional delivery systems have been shown to be clinically efficacious, they typically suffer from relatively low efficiency levels in delivering antibiotic solutions to the endobronchial space of a patient, thereby wasting a substantial portion of the nebulized antibiotic formulations and substantially increasing drug delivery costs. The low efficiency of current 20 conventional delivery systems requires patients to devote relatively long time periods to receive an effective dose of the nebulized antibiotic formulations, which can lead to decreased patient compliance. Accordingly, there is a need for new and improved methods and devices for the delivery of aminoglycoside antibiotic compounds to a patient by inhalation to reduce administration costs, increase patient compliance and 25 enhance overall effectiveness of the inhalation therapy. SUMMARY OF THE INVENTION It has now been discovered that patients suffering from an endobronchial infection can be effectively and efficiently treated by administering to the patient for inhalation a dose of less than about 4.0 ml of a nebulized aerosol - formulation 30 comprising from about 60 to about 200 mg/mrl of an aminoglycoside antibiotic, such -6 as tobramycin, in a physiologically acceptable carrier in a time period of less than about 10 minutes, more preferably less than about 8 minutes, and even more preferably less than about 6 minutes. In other aspects, the administered dose may be less than about 3.75 ml or 3.5 ml or less, and the aminoglycoside antibiotic 5 formulation may comprise from about 80 to about 180 mg/ml of aminoglycoside antibiotic or more preferably from about 90 to about 150 mg/ml of aminoglycoside antibiotic. In other aspects, the present invention provides unit dose formulations and devices adapted for use in connection with a high efficiency inhalation system, the 10 unit dose device comprising a container designed to hold and store the relatively small volumes of the aminoglycoside antibiotic fannulationsnf the inventionand to deliver the formulations to an inhalation device for delivery to a patient in aerosol form. In one aspect, a unit dose device of the invention comprises a sealed container, such as an ampoule, containing less than about 4.0 ml of an aminoglycoside 15 antibiotic formulation comprising from about 60 to about 200 mg/ml of an aminoglycoside antibiotic in a physiologically acceptable carrier. The sealed container is preferably adapted to deliver the aminoglycoside antibiotic formulation to a high efficiency inhalation device for aerosolization and inhalation by a patient. In other aspects, the container of the unit dose device. may contain less than about 20 3.75 ml, or 3.5 ml or less, of the aminoglycoside antibiotic formulation, and the aminoglycoside antibiotic formulation may comprise from about 80 to about 180 mg/ml, or from about 90 to about 120 mg/ml, of aminoglycoside antibiotic. I yet other aspects, the present invention relates to a system for delivering an aminoglycoside antibiotic formulation to a patient in need of such treatment, 25 comprising a unit dose device as described in detail above, comprising a container containing less than about 4.0 ml of an aminoglycoside antibiotic formulation comprising from about 60 to about 200 mg/ml of an aminoglycoside antibiotic in a physiologically acceptable carrier, and means for delivering the aminoglycoside antibiotic formulation from the unit dose device for inhalation by the-patient-in 30 aerosolized form in less that 10 about minutes.

7 BRIEF DESCRIPTION OF THE DRAWINGS The foregoing aspects and many of the attendant advantages of this invention will become more readily appreciated as the same become better understood by reference to the following detailed description, when taken in conjunction with the 5 accompanying drawings, wherein: FIGURE 1 is a graphical representation illustrating the mean relative changes in FEV, %.predicted from before to 30 minutes after dosing with 300 mg tobramycin with a PARI LC PLUS jet nebulizer/PulmoAide compressor delivery system, or with 30, 60, or 90 mg tobramycin with an Aerodose breath actuated nebulizer, as 10 described in Example 1; FIGURE-2 - is-a- graphical representation -showing- -sputum -tobramyciconcentrations by time from dosing by the tobramycin formulations of FIGURE 1, as described in Example 1; FIGURE 3 is a graphical representation showing sputum maximum plasma 15 concentrations (Cm.) following dosing by the tobramycin formulations 'of FIGURE 1, as described in Example 1; FIGURE 4 is a graphical representation showing sputum area under the plasma concentration time profile (AUCo-s) following dosing by the tobramycin formulations of FIGURE 1, as described in Example 1; 20 FIGURE 5 is a graphical representation showing serum tobramycin concentrations by time following dosing by the tobramycin formulations of FIGURE 1, as described in Example 1; FIGURE 6 is a graphical representation showing serum maximum plasma concentrations (Cm) following dosing by the tobramycin formulations of 25 FIGURE 1, as described in Example 1; FIGURE 7 is a graphical representation showing serum area under the plasma concentration time profile (AUCo-s) following dosing by the tobramycin formulations of FIGURE 1, as described in Example 1; -8 FIGURE 8 is a graphical representation showing the mean recovery of tobramycin from urine 0-8, 8-24 and 0-24 hours post dosing with the formulations of FIGURE 1, as described in Example 1; and 5 FIGURE 9 is a graphical representation showing the mean nebulization time in minutes for dosing with the formulations of FIGURE 1, as described in Example 1. FIGURE 10 is a graphical representation showing the average serum-time profiles of tobramycin after administration of 300 mg tobramycin (TOBI) and 420 mg 10 tobramycin solution for inhalation (TSI), as described in Example 3. FIGURE II is a graphical representation showing the average sputum-time profiles of tobramycin after administration of 300 mg tobramycin (TOBI) and 420 mg tobramycin solution for inhalation (TSI), as described in Example 3. 15 DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS In one example, the present invention provides methods for the treatment of a patient in need of treatment, such as a patient suffering from an endobronchial P. aeruginosa infection, comprising administering to the patient for inhalation a relatively 20 small volume of an aminoglycoside antibiotic formulation over a relatively short period of time. This aspect of the invention is particularly suitable for formulation of concentrated aminoglycosides, such as tobramycin, for aerosolization by small volume, breath actuated, high output rate and high efficiency inhalers to produce a aminoglycoside aerosol particle size between I and 5 pm desirable for efficacious 25 delivery of the aminoglycoside into the endobronchial space to treat susceptible microbial infections, such as Pseudomonas aeruginosa infections. The formulations preferably contains minimal yet efficacious amount of aminoglycoside formulated in smallest practical volume of a physiologically acceptable solution, for example an aqueous solution having a salinity adjusted to permit generation of aminoglycoside 30 aerosol particles that are well-tolerated by patients but preventing the development of secondary undesirable side effects such as bronchospasm and cough. By the more efficient administration of the aminoglycoside formulation provided by the present invention, substantially smaller 9 volumes of aminoglycoside than the conventional administration regime are administered in substantially shorter periods of time, thereby reducing the costs of administration and drug waste, and significantly enhancing the likelihood of patient compliance. 5 Thus, in accordance with one aspect of the present invention, methods are provided for the treatment of a patient in need of treatment, sudh as a patient suffering from an endobronchial P. aeruginosa infection, comprising administering to the patient for inhalation a dose of less than about 4.0 ml of a nebulized aerosol formulation comprising from about 60 to about 200 mg/nl of an aminoglycoside 10 antibiotic in a time period of less than about 10 minutes. In other aspects, the dose of the-aerosol formulation-is -administered to thepatienrin-le-ssthan -b-65d B ininntes. In yet other aspects, the dose of the aerosol formulation is administered to the patient in less than about 6 minutes. The aerosol formulations administered in the practice of the invention may 15 comprise from about 60 to about 200 mg/ml of aminoglycoside antibiotic. In other aspects of the invention, the aerosol formulations administered in the practice of the invention may comprise from about 80 to about 180 mg/ml of aminoglycoside antibiotic. In yet other aspects of the invention, the aerosol formulations administered in the practice of the invention may comprise from about 90 to about 20 150 mg/ml of aminoglycoside antibiotic. In the practice of the methods of the invention, substantially smaller volumes of aerosol formulation are administered to the patient, as compared with the conventional administration processes. Thus, in one aspect a dose of less than about 4.0 ml of a nebulized aerosol formulation is administered to the patient. In another 25 aspect, a dose of less than about 3.75 ml of a nebulized aerosol formulation.is administered to the patient. In yet another aspect, a dose of 3.5 ml or less of a nebulized aerosol formulation is administered to the patient. In yet other aspects, the present invention relates to a system for delivering an aminoglycoside antibiotic formulation to a patient in need of such .treatment, 30 comprising a unit dose device as described in detail herein, comprising a container 10 containing less than about 4.0 ml of an aminoglycoside antibiotic formulation comprising from about 60 to about 200 mg/ml of an aminoglycoside antibiotic in a physiologically acceptable carrier, and means for delivering the aminoglycoside antibiotic formulation from the unit dose device for inhalation by the patient in 5 aerosolized form in less that 10 about minutes. In order to deliver the relatively small volumes of the relatively high concentration aminoglycoside antibiotic formulations to the patient for inhalation in the relatively short dosing periods of the invention, the antibiotic formulations are preferably administered with the use of an inhalation device having a relatively high 10 rate of aerosol output. Useful devices may also exhibit high emitted dose efficiency (i.e., low residual volume in the device). In order to increase the overall efficiency of the system, emission may additionally be limited to periods of actual inhalation by the patient (i.e., breath actuated). Thus, while conventional air-jet nebulizers exhibit a rate of aerosol output on the order of 3 sl/sec, inhalation devices useful for use in 15 the practice of the present invention will typically exhibit a rate of aerosol output of not less that about 4 pl/sec. In some cases, inhalation devices useful for use in the practice of the present invention will exhibit a rate of aerosol output of not less than about 5 d/sec or even not less than about 8 pl/sec. In additiork, while conventional air-jet nebulizers have a relatively low emitted dose efficiency and typically release 20 about 55% (or less) of the nominal dose as -aerosol, inhalation devices useful for use in the practice of the present invention may release at least about 75%, more preferably at least about 80% and most preferably at least about 85% of the loaded, dose as aerosol for inhalation by the patient In other aspects; conventional air-jet nebulizers typically continually release aerosolized- drug throughout the delivery 25 period, without regard to whether the patient is inhaling, exhaling or in a static portion of the breathing cycle, thereby wasting a substantial portion of the loaded drug dose. In some embodiments, inhalation devices for use in the present invention will be breath actuated, and restricted to delivery of aerosolized particles of the aminoglycoside formulation to the period of actual inhalation by the patient. - One 30 representative inhalation device meeting the above criteria and suitable for use in the practice of the invention is the Aerodose" inhaler, available from Aerogen,. Inc., Sunnyvale, California. The Aerodosem inhaler generates an aerosol using a porous membrane driven by a piezoelectric oscillator. Aerosol delivery is breath actuated, and restricted to the inhalation phase of the breath cycle, i.e., aerosolization does not 5 occur during the exhalation phase of the breath cycle. The airflow path design allows normal inhale-exhale breathing, compared to breath-hold inhalers. Additionally, the AerodoseT' inhaler is a hand-held, self-contained, and easily transported inhaler. Although piezoelectric oscillator aerosol generators, such as the Aerodosel m inhaler, represent one embodiment for use in the practice of the invention, other inhaler or 10 nebulizer devices may be employed that meet the above performance criteria and are capable of delivering the small dosage volumes of the invention with a relative high effective deposition rate in a comparatively short period of time. In other embodiments of the invention devices useful for delivering the concentrated aminoglycoside formulations of the invention include conventional air-jet nebulizers 15 coupled with a compressor capable of higher than conventional output pressures. Enhanced compressor output pressures useful in the practice of the invention will be readily determinable to those skilled in the.art in view of the disclosure contained herein. As one representative example, the PARI LC PLUSTm jet nebulizer, PARI GmbH, Stamberg, Germany, driven by a. Invacare MOBILAIRETM compressor, 20 Invacare Corporation, Elyria, Ohio, set for an output pressure of about 35 psi has been found to be .capable of delivering 3.5 ml of the concentrated aerosolized aminoglycoside formulations of the invention (such as tobramycin) in 10 minutes or less, as is hereinafter described in detail in Example 3. Aminoglycoside antibiotics useful in the practice of the invention include, for 25 example, gentamicin, amikacin, kanamycin, streptomycin, neomycin, netilmicin and tobramycin. A presently particularly preferred aminoglycoside antibiotic for this purpose is tobramycin. Formulations according to the invention typically contain from about 60 to about 200 mg, more preferably from about 80 to about 180, and most preferably from about 90 to about 120 mg of aminoglycoside per'ii of solution. 30 The aminoglycoside antibiotic of the invention may be incorporated into sterile water 12 or physiologically acceptable solution. Other components may be included in the formulation, as desired. In order to facilitate administration and compatibility with the endobronchial space, the aminoglycoside antibiotic of the invention is preferably formulated in a diluted physiological saline solution, such as in one quarter strength 5 of normal saline, having a salinity adjusted to permit generation of tobramycin aerosol well-tolerated by patients but to prevent the development of secondary undesirable side effects such as bronchospasm and cough. Typically, about 90 to about 120 mg of aninoglycoside antibiotic is dissolved in I ml solution of a diluted, typically quarter normal saline containing about 0.225% NaCI. Quarter normal 10 saline, that is 0.225% of sodium chloride, is a presently preferred vehicle for delivery of aminoglycoside into endobronchial space. By way of illustration, high concentrations of tobramycin administered to the lungs by aerosolization result in maximization of sputum levels of tobramycin and in minimization of tobramycin serum levels. Thus, administration of tobramycin by 15 aerosolization has the advantage of reducing systemic toxicity while providing efficacious concentrations of tobramycin in the sputum. . The bronchial barrier restricts the movement of acrosolized tobramycin and prevents it from reaching high systemic levels. In other aspects of the present invention, unit dose formulations and devices 20 are provided for administration of an aminoglycoside antibiotic formulation to a patient with an inhaler, in accordance with the methods of the invention as described supra. Preferred unit dose devices comprise a container designed to hold and store the relatively small volumes of the aminoglycoside antibiotic formulations of the invention, and to deliver the formulations to an inhalation device for delivery to a 25 patient in aerosol form. In one aspect, unit dose containers of the invention comprise a plastic ampoule filled with an aminoglycoside antibiotic formulation of the invention, and sealed under sterile conditions. Preferably, the unit dose ampoule is provided with a twist-off tab or other easy opening device for opening of the ampoule and delivery of the aminoglycoside antibiotic formulation to the-inhalation 30 device. Ampoules for containing drug formulations are well known to those skilled - 13 in the art (see, for example, U.S. Patent Nos. 5,409,125, 5,379,898, 5,213,860, 5,046,627, 4,995,519, 4,979,630, 4,951,822, 4,502,616 and 3,993,223, the disclosures of which are incorporated herein by this reference). The unit dose containers of the invention may be designed to be inserted directly into an inhalation device of the invention for delivery of the 5 contained aminoglycoside antibiotic formulation to the inhalation device and ultimately to the patient. In accordance with this aspect of the invention, a unit dose device is provided comprising a sealed container containing less than about 4.0 ml of an aminoglycoside antibiotic formulation comprising from about 60 to about 200 mg/ml of an aminoglycoside antibiotic in a 10 physiologically acceptable carrier, the sealed container being adapted to deliver the aminoglycoside antibiotic formulation to an inhalation device for aerosolization. Suitable aminoglycoside antibiotics for use in connection with this aspect of the invention include those aminoglycoside antibiotics described in detail, supra. In a presently preferred embodiment, the aminoglycoside antibiotic employed in the unit dose devices of the invention is tobramycin. In 15 other aspects, the unit dose devices of the invention contain less than about 3.75 ml of the aminoglycoside solution. In other aspects, the unit dose devices of the invention contain 3.5 ml or less of the aminoglycoside solution. In other aspects of the invention, the unit dose devices of the invention may contain an amino glycoside antibiotic formulation comprising from about 80 to about 180 mg/mI of 20 aminoglycoside antibiotic. In yet other aspects of the invention, the unit dose devices of the invention may contain an aminoglycoside antibiotic formulation comprising from about 90 to about 150 mg/ml of aminoglycoside antibiotic. In preferred unit dose formulations of the invention, the physiologically acceptable carrier may comprise a physiological saline solution, such as a solution of one quarter strength 25 of normal saline, having a salinity adjusted to permit generation of a tobramycin aerosol that is well-tolerated by patients, but that prevents the development of secondary undesirable side effects such as bronchospasm and cough. Any discussion of documents, acts, materials, devices, articles or the like which has been included in the present specification is solely for the purpose of providing a context for the 30 present invention. It is not to be taken as an admission that any or all of these matters form part of the prior art base or were common general knowledge in the field relevant to the present invention as it existed before the priority date of each claim of this application. Throughout this specification, unless the context requires otherwise, the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the 35 inclusion of a stated step or element or integer or group of steps or elements or integers but not the exclusion of any other step or element or integer or group of elements or integers.

14 These and other aspects of the inventive concepts may be better understood in connection with the following non-limiting examples. EXAMPLES EXAMPLE I 5 IN VIVO STUDY I A comparison was made of the safety, pharmacokinetics, aerosol delivery characteristics, and nebulization time of the conventional dose and inhalation delivery system (5 mL ampoule containing 300 mg tobramycin and 11.25 mg sodium chloride in sterile water for injection (TOBI* tobramycin solution for inhalation, 10 Chiron Corporation, Seattle, Washington), pH 6.0; administered with a PARI LC PLUS Mjet nebulizer with a PulmoAide compressor) with 3 doses of TOBI (30 mg tobramycin in 0.5 mL solution, 60 mg in 1.0 mL, and 90 mg in 1.5 mL) using a AeroDose" inhaler device. The study was designed as an open label, randomized, multicenter, single 15 dose, unbalanced, four treatment, three period crossover trial. Each patient was to receive three single doses of aerosolized antibiotic: the active drug control treatment during one treatment period and two of three experimental treatments during two additional treatment periods. Single dose administration during the three treatment periods was to occur at one-week intervals. 20 In accordance with the study design, forty eight eligible male and female patients 12 years of age or older with a confirmed diagnosis of cystic fibrosis Were to be enrolled in the study and randomly assigned to one of 12 treatment sequences of three treatments each (one active control and two experimental treatments) with the constraint that the active control treatment was to be administered in either the first 25 or the second of the three treatment periods. Experimental treatments were administered during all three treatment periods. Each patient inhaled a single dose of aerosolized control and two of three experimental treatments in accordance with the present invention as follows.

15 - control delivery treatment (PARI LC PLUS iet nebulizer + PulmoAide compressor): * TOBI 300 mg in 5 mL solution. * experimental delivery treatments (AeroDose' inhaler breath actuated nebulizer): 5 - TOBI 30 mg in 0.5 mL solution; * TOBI 60 mg in 1.0 mL solution; e TOBI 90 ng in 1.5 mL solution. The duration of study participation for each patient was to be approximately five weeks including a brief (2 days to one week) screening period, three one-week 10 treatment periods, and a one-week telephone follow-up period. Control and Experimental Treatments Each patient was to self-administer under research staff supervision a total of three single doses of aerosolized tobramycin during the study, one dose per crossover treatment period. Patients were to receive a single dose of the control delivery 15 treatment during period I or period 2 of the three treatment periods. In addition, each patient was to receive single doses of two of the three experimental delivery . treatments during the remaining two treatment periods. Control and experimental delivery treatments were specified as follows. Control Delivey Treatment: 20 PARI LC PLUS jet nebulizer with PulmoAide compressor: preservative free tobramycin 60 mg/mL (excipient 5 mL of 1/4 normal saline adjusted to a pH of 6.0 ± 0.5); 3 0 0 mg in 5 mL. Experimental Delivery Treatments: * Aerodose with a 3-4 pm mass medium diameter (MMD) aerosol particle size: 25 preservative free tobramycin 60 mg/mL (excipient 0.5 mL of 1/4 normal saline adjusted to a pH of 6.0 0.5); 3 0 mg in 0.5 mL; 16 - Aerodose with a 3-4 pim MMD: preservative free tobramycin 60 mg/mL (excipient 1.0 mL of 1/4'normal saline adjusted to a pl of 6.0 0.5); 60 mg in - 1.0 mL; e Aerodose with a 3-4 lim MMD: preservative free tobramycin 60 mg/mL 5 (excipient 1.5 mL of 1/4 normal saline adjusted to a pH of 6.0 i 0.5); 90 mg in 1.5 mL. Patients were placed upright in a sitting or standing position to promote normal breathing and were instructed to place the nose clips over the nostrils and to -breath normally through the .mouth until there was no longer any mist produced by 10 the nebulizer. Aerosol delivery was anticipated to take 15 minutes to complete. A pharmacist or coordinator prepared the 30 mg dose of TOBI by drawing 0.5 mL of the 60 mg/mL TOBI formulation into a one-mL syringe. Each syringe was labeled with the patient identification number. Study drug was dispensed into the medication reservoir as indicated in the Aerodose directions for use. TOBI 60 mg 15 and 90 mg doses were similarly prepared by drawing two and three 0.5 mL aliquots, respectively, from the TOBI ampoule into two and three one-mL syringes. Aerosol Delivery Systems The control delivery system (PARI LC PLUS jet nebulizer) was used once per patient during the study for administration of TOBI 300 mg (control treatment). 20 The experimental delivery system (Aerodose inhaler) was used to deliver only one dose of study treatments. The control nebulizer, the PARI LC PLUS jet nebulizer with DeVilbiss PulmoAide compressor, generates aerosol by air-jet shear. A detailed comparison of experimental and control devices is provided in Table 1.

17 TABLE 1. DEVICE COMPARISON PART LC PLUS Nebulizer Device Characteristic Aerodose Nebulizer and DeVilbiss PulnoAide ________________Compressor Aerosol generating principle Piezoelectric vibration Air-jet shear Aerosol characteristics with TOBI - Mass median diameter (MMD) 4.0 pn 4.8 pm - Output rate 8.0 RUsec 3.6 pLsec - Emitted dose 85% 57% Dose actuation Breath-actuated by user On/off switch; when on, inhalation medication aerosolized continuously Control of aerosol generation Breath actuated. An Continuous aerosol output during airflow sensor system is both inhalation and exhalation used to limit aerosol generation to inhalation ______________ User indicator lights Green LED flashing for "device ready" and solid Nn for "nerosolization" Nn __________________Red LED for "low battery" Physical characteristics 3.3" x2.6" x 1.1 7.5" x 7.5" x3.0* (nebulizer) - Size 10.1" x 10.5" x 6.5" (compressor) - Weight 140 g 68 gm (nebulizer) 3200 gm (compressor) Power source Four AAA alkafinc 115 VAC, 60Hz batteries _ _ _ _ _ _ Power consumption 2.5 watts 90 watts( (ax. Where used Fully portable Restricted to power outlets Ssui 115 VAC, 60 Hz Selection ofB Doses in the Stud Commercial TOBI 60 mg/mL in 5 mL solution administered by PAR! LC PLUS jet nebulizer and powered by the PulmoAide compressor was the active drug 5 control delivery system against which potential improvements in aerosol delivery technology by the Aerodose breath actuated nebulizer were compared in this example. The selection of doses of experimental treatments (TOBI 30 rsg in 0.5 L solution, 60 mug in 1.0 mL, and 90 mg in 1.5 mL) was based on empirical daita on the 10 comparative predicted efficiency of the Aerodose inhaler relative to the PARI LC 18 PLUS nebulizer. The selection of doses was also based on the assumption that TOBI delivered via the PARI LC PLUS jet nebulizer leads to the systemic absorption of approximately 11.7% of the administered dose (Pitlick, Nardella, et al., 1999}. Furthermore, the mean and standard deviation of the serum concentration one hour 5 after inhalation was 1.0 pg/mL ± 0.58, suggesting a wide range of deposition (Table 5.2 C, Clinical Pharmacology, PathoGenesis NDA, #50,753). Due to design features of the Aerodose inhaler, it was estimated that between 50-70% of the drug would be delivered to the lung. This assumption is based on the predicted efficiency of a nebulized dose. 10 Patients were randomized to treatment sequence groups, and predose procedures were completed including a physical examination (only if abnormal during screening), recheck of inclusion and exclusion criteria, interim history review, spirometry, clinical evaluation, and blood and urine specimens for laboratory tests (only if abnormal during screening). A bronchodilator was to be administered before 15 dosing if regularly used by the patient. Spirometry was completed 15-60 minutes after the bronchodilator, if applicable. Patients received a single dose of study treatments during each of three treatment periods separated by an interval of 7 days between treatments. At the time of single dose administration during each period, patients were instructed to sit 20 upright and use nose clips during aerosol dose administration. Patients remained at the clinic through completion of 8-hour post treatment procedures (nebulization time, spirometry, and sputum, serum and urine specimens for tobramycin determinations). Patients were then discharged from the clinic and were expected to collect and return their 8-24 hour urine-collection at the next visit, 25 no later than 7 days after their previous visit. Patients were to refrigerate urine collections at all times except during transport. Safety Variables Safety was assessed by monitoring the incidence of bronchospasm and by the quantitative change in pulmonary function (measured as change in'-FEVI ~%/ 30 predicted), the incidence of treatment emergent adverse events, and the incidence of 19 unusually high serum tobramycin results (2 4 pg/mL), the significance of clinical laboratory test results, and the significance of change in clinical evaluation results. Bronchospasm (Airway Reactivity) One objective of the study was to compare the rate of occurrence of 5 bronchospasm (airway reactivity) between control and experimental delivery systems. Bronchospasm was measured by the change in forced expiratory volume in 1 second [FEVI (liters)) from before dosing to 30.minutes after dosing during periods 1, 2, and 3. The number and percent of patients who experienced predose to postdose decreases in FEVI (liters) that were , 10 % and those that were 2 20 % were 10 determined to.assess the comparative incidence of bronchospasm among control and experimental treatments. Decreases in FEVI (liters) that were a 20 % were considered clinically significant for the purposes-of the study. Additionally, an acute decrease in FEVI (liters) ; 30% from before to after treatment was considered a symptom of respiratory distress. In this event, continuation of the patient in the study 15 was at the discretion of the investigator. Norms have been developed for FEVI. These norms are commonly used in studies of pulmonary patients. This study employed the Knudson equations that use age, gender, and height to predict a patient's FEVI values as if the patient was free of pulmonary function disease. The actual FEVI value is divided by the normative 20 value, and the resulting ratio is multiplied by 100 to produce a measure that represents percentage of predicted normal function, commonly called percent predicted. The transformation is: FEVI %predicted= (FEVI al value /FEV, xoativevaiuc) x 100 Relative change in FEVI % predicted is defined as the percent change from 25 predose to 30 minutes postdose in FEVI % predicted and is calculated as: relative change in FEVI % predicted = [(FEVI (%predied a 30 mines posey - FEVI t/ prediced a pt y) / FEVI (% predicted at preloas)] X 100 20 Clinical Laboratory Tests Serum creatinine, blood urea nitrogen (BUN), and dipstick urine protein results were obtained from specimens drawn during screening aid before dosing during treatment period 3. Urine dipstick testing was always performed on fresh 5 specimens. Serum and urine specimens that needed to be retained at the site (e.g., those drawn after shipping pick-up hours or on Friday or Saturday) were frozen until shipment at the next earliest shipping time. Specimens were covered with dry ice for shipping. All out of range laboratory results were evaluated for clinical significance and 10 drug relationship by the investigator using the following classification scheme: * clinically insignificant; & possible study medication relationship; e probable study medication relationship; * unrelated to study medication, related to concurrent illness; 15 e unrelated to study medication, related to other concurrent medication; e other (investigator commentary). Aerosol Delivery Variables Evaluation of the aerosol delivery characteristics of the Aerodose breath actuated nebulizer, compared to characteristics of the FDA-approved PARI LC 20 PLUS jet nebulizer with PulmoAide compressor, was based on determination of sputum, urine, and serum tobramycin concentrations, calculation of certain sputum and serum pharmacokinetic parameters, and measurement of nebilization time. Sputum Tobramycin Concentrations Before study treatments were administered, patients expectorated sputum 25 produced from a deep cough into an individual specimen container. Immediately after treatment, patients rinsed their mouths three times with 30 mL of normal saline, gargled for 5-10 seconds, and expectorated the rinse. ..

21 Post treatment sputum specimens were collected following the normal saline gargle at 10 minutes and at 1, 2, 4, and 8 hours after completion of the aerosol drug administration for determination of sputum tobramycin concentrations. Sputum specimens were judged to be acceptable if collected within ± 2 minutes of the 5 scheduled 10-minute posttreatment collection time and within ± 10 minutes of the scheduled 1-, 2-, 4-, and 8-hour collection times. After collection, specimens were immediately frozen for later determination of tobramycin concentrations in sputum. A minimum of 1 gram of sputum was required for analysis. Tobramycin concentrations in sputum (sputum LOQ = 20.0 pg/gm) were measured by using 10 HPLC. Serum Tobramycin Concentrations Whole blood was drawn by venipuncture, an indwelling heparin/saline lock, or a permanent venous access port at 10 minutes and at 1, 2, 4, and 8 hours after completion of dosing. Blqod specimens were judged to be acceptable if collected 15 within ± 2 minutes of the scheduled 10-minute posttreatment collection time and within 10 minutes of the scheduled 1-, 2-, 4-, and 8-hour collection times. Blood specimens were allowed to clot for 30 minutes and were then centrifuged at 1500 x g for 10 minutes until clot and serum separated. Serum samples (3 mL) were pipetted into plastic vials and frozen immediately for later determination of serum tobramycin 20 concentrations. Tobramycin concentrations in serum were measured by Abbott TDxFLx* assay (Abbott Laboratories, Abbott Park, Illinois) [serum lower limit of quantitation (LOQ)= 0.18 pg/mnL]. Urine Tobramycin Recovery 25 Urine specimens were collected and combined in a 24-hour collection container during the 12 hours before treatment (-12-0 hour period) and during 0-8 hour and 8-24 hour collection periods after treatment according to instructions provided in the Study Manual. Total urine volume for the collection period was recorded, and a 10 mL aliquot from each urine collection was retained and frozen-for 30 later analysis of urihe tobramycin concentration.

22 The recovery of tobramycin in urine (in milligrams) during 0-8 hour and 8-24 hour collection periods was calculated as follows. urine tobramycin recovery (pg) = urine volume (mL) - urine tobramycin concentaion (pg/mL) 5 Urine tobramycin recovery was normalized for each collection period according to TOBI dose as follows. dose-normalized urine tobramycin recovery (pg/mg) [urine tobramycin recovery (jpg) + TOBI dose (mg)] The percent of the TOBI dose excreted .in urine in the 24-hour period 10 following treatment was calculated as follows. % tobramycin excreted in urine =[(urinary recovery in pg + 1000 pg/mg) + TOBI dose in mg] e 100%. If either the urine volume or the urine tobramycin concentration for a collection interval was missing, then the urine tobramycin recovery was not 15 calculable for that interval. If calculated urine tobramycin recovery was missing for either the 0-8 hour or the 8-24 hour collection interval, then the 0-24 hour urine tobiamycin recovery was not calculated. Missing urine tobramycin recovery values were not replaced by estimated values for analysis purposes. Tobramycin concentrations in urine were measured by Abbott TDxFLx* 20 assay [urine lower limit of quantitation (LOQ) = 1.0 pg/mL]. Pharmacokinetic Parameters The maximum tobramycin concentrations (C.) in sputum and serum during the 8-hour posttreatment sampling period were identified for each patient during each treatment period, and the time at which C.na was observed (T.) was recorded. 25 Area under the concentration-time curve through 8 hours postdose (AUCo.:) was calculated from sputum and serum tobramycin concentrations using the linear 23 trapezoidal method. Nebulization time (excluding time for refilling) was added to the time between predose and 10 minutes postdose for AUCo. calculations. Area under the concentration-time curve extrapolated to infinity (AUCo.) was calculated for sputum and serum as follows. 5 AUCo.= AUCo.t + Cot kei where: AUCo., is area under the curve from predose through the last non-BQL time C(k) is the last non-BQL posttreatment concentration result ka is the elimination rate constant (terminal phase slope) 10 and kI log 2+ Tm where T 1 / is the elimination half-life for the patient. Relative systemic bioavailability was calculated based on serum AUCO-s values for control (TOBI 300 mg delivered by PARI LC PLUS nebulizer) and experimental (TOBI 30 mg, 60 mg, and 90 mg delivered by Acrodose inhaler) groups 15 as follows. relative bioavailability (%) experimental group serum AUCo- 8 + control group serum AUCO- 8 Missing tobramycin concentrations and those reported as zero or below quantifiable limits (BQL) were not to be replaced with any estimated value. C,, and 20 AUC 0 o. were always determinable except in the event that all posttreatment tobramycin concentrations were missing, zero, or BQL. There was no missing sputum C,, and AUC.

4 values among the 49 patients who completed the study (refer to report section 9.3.1 for details). Four completing patients had indeterminate serum Cm. and AUCo. values due to BQL serum results for each posttreatment 25 sampling time (refer to report section 9.4.1 for details). Nebulization Time The timing (duration) of nebulization began with the patient's first idal breath after the device was in place and continued until the device aerosolized no more 24 TOBI solution. Nebulization time did not include any interruptions or time needed for instillation of drug into the nebulizer between the repeat filling of the AeroDoseO inhaler. The length of any interruption in nebulization and the reason for interruption were recorded. 5 Safety Analyses Reductions in FEV, % predicted 2 10/6 and 2 20% were used as indicators of the occurrence of bronchospasm (airway reactivity). McNemar's test for paired comparisons (replacing the .Cochran-Mantel-Haenszel (CMH) test) was used for control vs. experimental treatment comparisons of the incidence of patients with 10 predose to 30-minute postdose decreases in FEVI % predicted that were 2 10% and ; 20%. In addition, pairwise t-tests were used to compare mean relative change in spirometry FEVI % predicted from predose to postdose between each experimental treatment and the control treatment. All statistical analyses were performed using two-sided tests conducted at a 0.05 significance level (i.e., a = 0.05). Since all 15 statistical tests were exploratory in nature, no adjustment of p-values was made for multiple testing. Changes from predose to postdose in vital signs, body weight, and the incidence of abnormal and/or clinically significant laboratory and physical examination results were summarized and evaluated descriptively. Individual patient serum tobramycin results were monitored for unusually 20 high values (24 pg/mL) that might potentially indicate the occurrence of systemic toxicity. Aerosol Delivery Analyses The natural logarithms of AUCO.8, AUC., and C. based on sputum and serum tobramycin concentrations were to be statistically analyzed using a rnixed 25 effect repeated-measure analysis of variance model containing treatment, sequence, period, and carryover as fixed effects and patient as a random effect. In the planned analysis of variance model, sequence and carryover (treatment by period interaction) effects were confounded. The actual model used for the analysis was therefore modified by dropping the sequence term so that the assessment of carryover (i-e., 30 treatment by period interaction) could proceed. When AUCo. values were 25 calculated, large outlier values were noted, and the analysis for this parameter was dropped. Three hypotheses regarding whether the experimental delivery treatment of 30 mg, 60 mg, or 90 mg TOBI was equivalent to the control delivery treatment of 5 300 mg TOBI were to be tested in the model. The experimental treatment to control ratio for each of the log AUC and Cm. parameters was estimated with 90 percent confidence intervals (Cis). Upper and lower limits for the Cis were then obtained by back transformation (i.e., by exponentiating the log values of the upper and lower limits) to the original scale of the parameter. If the Cis for the ratio of experimental 10 and control treatments contained the value of 1.0, it was concluded that the treatments were not significantly different at the a = 0.1 for the 90% CIs. If demographic or baseline characteristics showed important apparent differences between the three experimental AeroDose- groups compared to all patients, then the discrepant factor and its interaction with the delivery treatment 15 factor were to be added to the mixed-effect model. Exploratory evaluations of age, gender, body weight, and baseline pulmonary function (FEVI percent predicted) demonstrated no important effects on pharmacokinetic results. Disposition of Patients A total of 56 patients were screened for the study by the nine investigators. 20 Fifty-three patients met entrance criteria, were enrolled in the study, and were randomized to one of the 12 sequences of treatment administration identified in the randomization code. A total of 3 patients failed to meet entrance criteria and were not enrolled in the study: 2 patients had screening FEVI % predicted results that were below the 40 % criterion required for entry, and one patient exhibited 25 disqualifying serum creatinine, BUN, and/or proteinuria. Accrual of the 53 randomized patients at 9 sites was as follows: 3 sites randomized 8 patients each, 2 sites randomized 7 patients, 3 sites randomized 4 patients, and one site randomized 3 patients. Fifty two patients received at least one dose of study treatments, and one patient was enrolled and randomized butwithdrew 30 from the study before the first study treatment due to increased productive cough 26 with a significant decline in forced expiratory volume (FEV) since screening (both events and associated hyperventilation were considered SAEs due to hospitalization of the patient: included in study database). Of the 52 patients who received study treatments, 49 patients completed the 5 study, and 3 patients withdrew after having received one dose of study treatment. Two of the withdrawn patients discontinued the study during the control treatment period (TOBI 300 mg administered by PARI LC PLUS nebulizer), and one patient withdrew during the TOBI 90 mg by AeroDose T inhaler treatment period. Baseline Characteristics 10 Enrolled patients had documented laboratory (sweat chloride 2: 60 mEq/L by quantitative pilocarpine iontophoresis test (QPIT) and/or genotype with 2 identifiable mutations) and clinical evidence consistent with a diagnosis of cystic fibrosis. Patients met all inclusion and exclusion criteria except for one patient whose pulmonary function entrance requirement (FEVI 2: 40% of predicted based on 15 gender, age, and height) was waived (the patient's screening FEVI % predicted was 39.87%). The average FEVI % predicted of all randomized patients was 66.4% at screening with a range from approximately 40% to 116%. Patients reported no known local or systemic hypersensitivity to aminoglycosides. Patients had taken no loop diuretics, no form of aminoglycoside 20 within 7 days before study treatments, and no investigational medications within 2 weeks before study treatments. Female patients had a negative pregnancy test before study treatments, and all patients had serum creatinine 2.0 mg/dL, BUN < 40 mg/dL, and < 2+ proteinuria at visit I screening, as required by the protocol. Screening or repeat serum creatinine 25 and BUN results were within the normal ranges for these tests before study treatments. Screening or repeat urine protein results were positive 1+ in 3 patients, but this result did not preclude participation of these patients in the study. No disqualifying medical history or physical examination findings were noted at visit I screening. Screening and visit 1 predose vital signs were unremarkable for 30 nearly all patients. One patient exhibited low systolic and diastolic blood pressures 27 at (72/49 mmHg), but these results did not preclude participation of the patient in the study. Safety Evaluation Extent of Exposure 5 Forty-nine patients received all 3 single doses of study treatments according to the randomization code, and 3 patients who withdrew from the study received one dose of study treatment. These 52 patients were included in the safety evaluation. Fifty-one of the 52 patients received a single dose of TOBI 300 mg, and 34, 32, and 33 of the -52 patients received a single dose of TOBI 30 mg, 60 mg, and 90 mg, 10 respectively. Three of the 49 completing patients had to stop treatment due to inhaler malfunction and .subsequently repeated the treatment period at a later date. As a result, these 3 patients received a partial dose of TOBI during the period in which the malfunction occurred (the amount of the partial dose was not recorded) and a full dose of TOBI during the repeated period. 15 Pulmonary Function Results Brgnchospasm In one aspect, the study compared the rate of occurrence of bronchospasm (airway reactivity) between control and experimental delivery systems. The occurrence of bronchospasm was determined quantitatively based on the percent 20 change in FEV, (liters) from before dosing to 30 minutes after dosing in each of the 3 treatment periods. For the purposes of the study, predose to postdose reductions in FEV (liters) 10% and 20% were defied as bronchospasm; reductions in FEV (liters) 20% were considered clinically significant. Fifteen patients (9 male and 6 female) experienced 24 instances of 25 bronchospasm during the study. Two instances of clinically significant bronchospasm were observed (decline in FEVI (liters) : 20%: patient 105-1034 after TOBI 300 mg and patient 102-1040 after TOBI 60 mg). No statistically significant pairwise differences in the overall incidence of bronchospasm were noted between control and experimental treatments. No clear relationship appeared to exist between 28 the incidence of bronchospasm and TOBI dose or delivery system (see Table 2 below). Table 2. Incidence of Acute Bronchospasm by Treatment TOBI 300 mg TOB 30 mg TOBI 60 mg TOBI 90 mg Bronchospas PARI LC PLUS' Aerodose Aerodose Aerodose m Parameter . _inhaler 2 inhaler 2 inhaler 2 (N=51) (N =34) (N=32) (N=33) FEV, Decrease 9(17.6%) 5(14.7%) 6 (18.8%) 4(12.1%) > 10% FEVi Decrease 1 (2.0%) 0 (0.0%) 1 (3.1%) 0 (0.0%) > 20% Bronchospasm was defined by protocol as a decrease in FEV (liters) >10% and > 20% from predose to 30 minutes postdose. Declines > 20% were considered clinically significant. 1 -Control ( C ) treatment = TOBI 300 mg delivered by PARI LC PLUS nebulizer. 2 Experimental ( E) treatments = TOBI 30, 60, or 90 mg delivered by Aerodose inhaler. 5 One patient 34 experienced clinically significant bronchospasm at 30 minutes after completing the TOBI 300 mg dose during treatment period 1 (visit 2). This 32 year old male patient's FEV, was 2.55 L before dosing and 1.98 L (decline in FEV, (liters) > 20%) at 30 minutes after dosing. He experienced moderate chest tightness that resolved spontaneously. This patient also experienced a second episode of 10 bronchospasm 30 minutes after TOBI 60 mg during period 2. The FEV, was 2.47 L before dosing and 2.14 L (decline in FEV, (liters) > 10% but < 20%) at 30 minutes after dosing. No symptomatology was reported at the time of this event. No prestudy aminoglycoside use was noted for this patient. One patient experienced one instance of clinically significant bronchospasm 15 30 minutes after TOBI 60 mg during period 3 (visit 4) of the crossover. This 36-year old male patient's FEV, was 2.26 L before dosing and 1.75 L (decline in FEV, (liters) > 20%) at 30 minutes after dosing (Archival Listing 3), but he reported no other symptomatology at this time. No prestudy aminoglycoside use was noted for thiis patient. This episode of bronchospasm appeared due in part to . an 29 uncharacteristically high prose FEVI value. The 30-minute posttreatment value was similar to that obtained during period 2 when the change in FEVI did not meet the definition of bronchospasm. Among the 13 patients who experienced clinically non-significant 5 bronchospasm, one patient experienced a decline in FEV, (liters) 2 10% but < 20% after all three stUdy doses were administered, 6 patients experienced a decline in FEV (liters) 10% after two doses of study medication, and 6 patients experienced a single instance of bronchospasm. Table 3 below lists instances of bronchospasm by patient, treatment period, and TOBI dose.

30 Table 3. Patient Dosing Regimen and Acute B.ronchospasm Site- Period 1 (Visit 2) Period 2 (Visit 3) Period 3 (Visit 4) Patient ID/Gender TOBI Dose Received TOBI Dose Received TOBI Dose Received 108-1048b/Female 300' 30' 60 109-1015 b/Male 300 30 ' 60 107-1027 /Male 300 30* 90 * 103-1038 b/ Female 300 c 60 30 105-1034 / Male 300 'c 60 c 30 107-1026/ Female 300 60 ' 90 102-1009 /Female 300' 90' 30 102-1040 /Male 300 90 60 ' 106-1050 /Female 30' 300 90 102-1007 /Male 60' 300' 30 104-1021 /Male 60 ' 300 ' 30 108-1044 /Male 60 300' 30' 105-1047 /Female 60 300 ' 90 106-1022b/Male 90' 300 30 106-1041 b/Male 90 '' 300 ' 60 c Bronchospasm is defined as a decrease in FEV, (liters) >10% and > 20% from predose to 30 minutes postdose. Declines > 20% were considered clinically significant. b The patient used a bronchodilator before dosing with study medication. Bronchospasm (not clinically significant): decrease in FEV, (liters) > 10% but <20%. d Bonchospasm (clinically significant): decrease in FEVI (liters) > 20%. 'The patient also reported *lung function decrease" (COSTART term) as an AE during the designated treatment period. Three of the 15 patients with bronchospasm reported treatment-related symptoms at the same time. One patient 15 experienced moderate wheezing (coded as asthma) after TOBI 30 mg during period 2, one patient 4 experienced moderate 5 chest tightness (coded as chest pain as reported previously) after TOBI 300 mg during period 1, and one patient 41 experienced increased cough after TOBI 60 mg during period 3. All events resolved either spontaneously (chest tightness), with treatment (wheezing), or by holding and restarting therapy (increased cough). None of the adverse events led to a serious outcome..-.- 31 Four of the 15 patients with bronchospasm (and one patient without bronchospasm) reported "lung function decreased" (COSTART term) as an adverse event. In addition to the 4 patients with bronchospasm identified in Table 3 above, one patient who experienced no bronchospasm, reported lung function decreased 5 once after TOBI 60 mg and once after TOBI 90 mg delivered by the AeroDose M inhaler. Initial instances of bronchospasm occurred more frequently during period I than during periods 2 or 3 of the crossover. Nine of the 15 patients first experienced bronchospasm during the first treatment period (visit 2), five patients during the 10 second treatment period, and one patient during the third treatment period. Patients who routinely used a bronchodilator were penitted to continue to do so during the study. Bronchodilator doses were to be administered 15 to 60 minutes prior to study treatments. Nine of the 15 patients who experienced bronchospasm during the study used a bronchodilator prior to administration of study treatment. 15 Relative Change in FEV, % Predicted The magnitude of the relative change in FEVI % predicted was calculated as a quantitative measure of the effect of TOBI treatments on pulmonary function during the study. There were no statistically significant differences among the 4 treatments and no evidence of the presence of period or carryover (treatment by 20 period interaction) effects. Results of pairwise comparisons between control and experimental treatments are summarized in Table 4. Since the overall treatment difference was not statistically significant, the significant p-value for the TOBI 300 mg vs. TOBI 30 mg comparison in Table 4 below (p = 0.019) should not be interpreted as conclusive evidence of a difference. Figure 1 graphically illustrates the 25 mean relative changes in FEVI % predicted from before to 30 minutes after dosing for each of the treatments.

32 TABLE 4. MEAN (SD) RELATIVE CHANGE IN FEVI % PREDICTED FEV, % TOBI 300 mg TOBI 30 mg TOBI 60 mg TOBI 90 mg Predicted (%). PARI LC PLUS' Aerodose inhaler 2 Aerodose inhaler2 Aerodose inhale. Parameter (n = 51) (n =34) (n =32) .(n=33) Predose 67.8 (18.4) 65.5 (17.1) 65.4 (16.8) 71.3 (20.0) n=SI n=34 n=32 n=33 30 minutes 637 (17.6) 63.0(16.7) 62.5 (15.7) 68.7 (19.1) postdose n-5I n=34 n=32 n=32 Relative change -6.1 (5.2) .3.8 (5.4) -4.2 (6.2) -3.2 (7.4) from predose' n=51 n = 34 n=32 n=32 P-value for Treatment: 0.141 Period: 0.199 Canyover NC crossover Pairwise contrasts: 0.019 0.058 0.083 C vs. E p-value (paired t-test): I Control ( C ) treatment= TOBI 300 mg delivered by PARI LC PLUS nebulizer. 2 Experimental ( E ) treatments = TOBI 30, 60, or 90 mg delivered by Acrodosc inhaler. I Relative change from predose - 100 % e ((30 minute postdose value - predose value)predose value). NC = carryover (treatment by period interaction) effect not statistically significant and was dropped from final model. Safety Conclusions . Nine males and six females experienced treatment-induced bronchospasm during the study. There was no difference in the rate of occurrence of TOBI induced 5 bronchospasm between control and experimental delivery systems regardless of dose. The occurrence of bronchospasm was rarely associated with patient symptoms. All but four of the patients experiencing drug-induced bronchospasm had been prescribed bronchodilators prior to the study suggesting that they had a history of airway reactivity. The disproportionate number of males versus females 10 experiencing airway reactivity is unusual in light of the fact that enrollment was approximately 60% female and 40% male. The pivotal trials showed that gender had 33 no influence on drug induced airway reactivity. However, it would be difficult to base any conclusions on this finding due to the small patient numbers in this study. Treatment-emergent adverse events occurred in all treatment groups regardless of causality. The most common treatment-emergent experiences were 5 associated with Respiratory and Body as a Whole systems. The most common individual events were cough increased, rhinitis, sputum increased, asthma, chest pain, and headache. These events were also common to the patient's pretreatment symptoms reflecting the patients underlying disease. For the majority of treatment emergent adverse events, there were no meaningful differences between TOBI doses 10 or between the PAR] LC PLUS nebulizer and the AeroDose m inhaler. The serious adverse events (SAEs) reported were primarily associated with an exacerbation of the patients underlying disease states. The one treatment-related SAE involved a possible sensitivity reaction that, if documented, would have occurred regardless of device or dose. 15 Review of the clinical chemistry, vital signs, and physical findings did not reveal any clinically significant safety issues associated with the dose or delivery system used to administer TOBI. All the patients were on multiple concurrent medications appropriate to their disease state (cystic fibrosis), other underlying illnesses, and age throughout the 20 study. The concurrent medications did not appear to have any influence on the safety profile of the study drug or either device during the study. Overall, no clinically significant or unexpected safety issues for TOBI were identified in the study. The study showed that there were no meaningful differences in the safety profiles of administering TOBI via the PARI LC PLUS delivery system in comparison with the 25 Aerodose delivery system regardless of dose. Aerosol Delivery Results Data Analysis Forty-nine of the 52 dosed patients completed the study and were evaluable for pharmacokinetics by reason of having completed at least 2 doses. of study 30 treatments. These 49 patients also constituted the "completers" subset of patients 34 referred to in the summary tables. Three of the 52 dosed patients discontinued the study before completing 2 doses of study treatments and were not evaluable for pharmacokinetics. All 52 patients were evaluable for the aerosol delivery objective (nebulization time) of the study. 5 Sputum Tobramycin Concentrations and Pharmacokinetic Parameters Compliance with Specimen Collection Requirements Six of 49 completing patients had a total of 11 missing sputum specimens. No more than one sputum sample was missed per treatment-time (e.g., for TOBI 300 mg at one hour postdose). Two patients missed 2 or more sputum samples during.the 10 study, and four patients missed a single sputum sample. A single completing patient provided no sputum pharmacokinetic data for the TOBI 60 mg treatment. One patient had missing sputum samples from 10 minutes through 8 hours after TOBI 60 mg treatment. After the database was locked, the missing sputum concentration results were located. Sputum tobramycin 15 concentrations at 10 minutes and 1, 2, 4, and 8 hours were 0.82 tg/gm, BQL, 0.0, 0.0, and 0.0, respectively. The database was not subsequently unlocked to add these data, since the inclusion of these values would have had minimal impact on estimation and analyses of pharmacokinetic parameters. As a result, only C,. (0.82 gg/gm) and Tm. (10 minutes = 0.17 hour) values were excluded from TOBI 60 mg 20 PK estimates and analyses; AUC values were incalculable due to BQL tobramycin concentrations from one through 8 hours after TOBI 60 mg treatment. Sputum Tobramycin Concentrations Pretreatment sputum tobramycin concentrations for all completing patients were below the limit of quantifiability (LOQ) throughout the study. 25 After dosing, sputum concentrations increased rapidly, reaching maximum concentrations within 10 minutes (see Figure 2), and declined thereafter with median half-life values ranging from approximately 1.6 to 2.1 hours during the four treatments. The sputum concentrations were highly variable among patients, as coefficients of variation (standard deviation divided by the mean times 100%) 30 approached or exceeded 100% for each treatment at all time points.

35 For the AeroDose'" inhaler, mean sputum tobramycin concentrations increased with increasing TOBI dose at'each measurement time during the 8-hour postdose period. Mean sputum concentrations for the TOBI 90 rmg treatment with the AeroDose inhaler were similar throughout the 8-hour period to those obtained 5 for the TOBI 300 mg treatment with the PARI LC PLUS nebulizer. By 2 hours after the end of TOBI 30 mg and by 8 hours after TOBI 60 mg, 90 mg, and 300 mg treatments, sputum concentrations were below LOQ in at least half of the patients. Period effects on sputum tobramycin concentrations were not observed. 10 . After TOBI administration using the AeroDose m inhaler, maximum plasma concentrations (Cm) and area under the plasma concentration time profile (AUCo-s) increased linearly with dose (Table 5 below and Figures 3 and 4), suggesting linear pharmacokinetics. Dose normalized C,,. and AUC values were comparable among AeroDose"' dose levels, indicating dose proportionality (based on AUC values). 15 Comparing devices, mean C.. and AUCo.S for the TOBI 90 mg treatment delivered by the AeroDose" inhaler achieved similar levels as those obtained by the TOBI 300 mg treatment delivered by the PARI LC PLUS nebulizer. The dose normalized Cm. and AUCo.& results were higher during AeroDose" treatments than during the PARI LC PLUS treatment, indicating that the AeroDose" inhaler 20 exhibited higher efficiency. The bioavailability of the AeroDose"" device was about 3-fold higher than that of the PARI LC PLUS nebulizer. Exploratory analyses suggested that sputum pharmacokinetic results were unaffected by characteristics present before treatments began (age, gender, body weight, FEVI % predicted at screening) and were unaffected by events noted after the 25 start of treatments (device failure, occurrence ofbronchospasm defined as a decrease > 10% in FEV1, and relative change in FEVI % predicted).

36 TABLE 5. MEAN (SD) SPUTUM TOBRAMYCIN PHARMACOKINETIC PARAMETERS TOBI 30 mg TOBI 60 mg TOBI 90 mg Sputum TOBI 300 mg Aerodose Aerodose Aerodose Pharmacokinetic PARI LC PLUS' inhaler" inhaler' inhaler' Parameters (n=49) (n34) (n=32) (n=32) C. (g/gm) 985.65 (839.34) 329.05 (311.30) 577.83 (538.42) 958.00 (952.30) -No. pts with data: 49 34 31 32 -- E vs C p-value': <0.001 0.002 0.856 - E/C (90/6 CIs)': (0.23, 0.41) (0.43, 0.75) (0.72, 1.30) Dose-normalized Cmax (pg/gm)/mg 3.29(2.80) 10.97 (10.38) 9.63 (8.97) 10.64 (10.58) - No. pts with data: 49 34 31 32 - EIC (90% CIs)': (2.82, 5.13) T, (hr) 0.26 (0.38) 0.24 (0.24) 0.38 (0.76) 0.33 (0.41) -No. pts with data: 49 34 31 32 Tin (hr) 6.41(24.09) 2.04 (1.31). 12.89 (42.61) 13.02 (36.91) - Median Tin (br) 1.71 1.78 2.06 1.60 - No. pts with data: 41 Is 21 24 AUC..3(hrepg/gm) 1471.16(1278.22) 360.79 (422.23) 804.78 (722.83) 1275.23 (1358.52) - No. pts with data: 49 34 31 32 - E vs C p-value*: <0.001 <0.001 0.465 - E/C (90% CIs)': (0.19,0.28) (0.45, 0.69) (0.72, 1.14) Dose-normalized AUC... (hropg/gm)/mg 1.90(4.26) 12.03 (14.07) 13.41(12.05) 14.17 (15.10) - No. pts with data: 49 34 31 32 - FJC (90% CIs)': (2.78, 4.12) AUCo..(hrgjig/gnm) 1996.36 (2013.70) 638.68 (586.85) 1661.66 (2334.89) 5544.88 (14831.0) - No. pts with data: 41. 15 21 24 . a Control (C) treatment= TOBI 300 mg delivered by PARI LC PLUS nebulizer. b Experimental ( E ) treatments = TOBI 30, 60, or 90 mg delivered by Aerodose inhaler. c Pairwise contrast: TOBI 300 mg PARI LC PLUS group vs TOBI (30,60,90 mg) Aerodose groups' d Back-transformed 90% confidence intervals around the mean of the log ratio of E and C treatments. Sputum limit of quantifiability (LOQ): 20 pg/gm. Differences among the treatment groups in Cmx and AUCo.

8 (Table 5 above; Figures 3 and 4) were statistically significant (p < 0.001) with no evidence of period 37 or carryover (treatment by period interaction) effects. In pairwise comparisons, C,, and AUCo.- were significantly greater for TOBI 300 mg than for TOBI 30 mg and for TOBI 60 mg but not for TOBI 90 mg (90% CIs for C.= (0.72, 1.30); for AUCo.g= (0.72, 1.14)). 5 The AeroDose m inhaler was more efficient, regardless of TOBI dose, than the PARI LC PLUS nebulizer based on dose normalized sputum C. and AUC. results. Dose normalized means for these pharniacokinetic parameters were similar among AeroDose treatments but approximately 3-fold higher than the dose normalized results after TOBI 300 mg delivered by the PARI LC PLUS nebulizer (see Table 5). 10 The time to maximum sputum tobramycin concentrations (T. in Table 5 above) was similar for all treatment groups and averaged between 0.24 and 0.38 hours for AeroDose" doses compared to 0.26 hours for the TOBI 300 mng treatment using the PARI LC PLUS. Elimination half-life (median T 1 1 in Table 5) was also similar among AeroDose" treatments, averaging 1.60 to 2.06 hours, compared to 15 1.71 hours for TOBI 300 mg. Exploratory analyses revealed no substantial association between sputum pharmacokinetic results and patient characteristics present before treatments (age, gender, body weight, pulmonary function [FEV, % predicted] at screening) or emergent events after the start of treatments (device failure, occurrence of 20 bronchospasm [decrease 2 10% in FEV from predose to 30 minutes postdose], relative change in FEVI,. Serum Tobramycin Concentrations and Pharmacokinetic Parameters Forty-four (44) of 49 completing patients had no measurable serum tobramycin concentrations before dosing in any of the 3 treatment periods, and five 25 patients exhibited measurable predose serum tobramycin above the lower LOQ in the periods indicated in Table 6 below.

38 TABLE 6. MEASURABLE TOBRAMYCIN IN PREDOSE SERUM SPECIMENS Previous Treatment Period Measurableb Predose 8-hour Serum Tobramycin during Tobramycin Serum Period LAsted TOBI Dose Concentration Tin Tobramycin Treatment' (pg/mL) Patient Sequence (ng) (br) Concentration (pg/mL) 107-1030 C-1-2 prestudy na" na' Per 1 - 0.70 107-1027 C-1-3 300 <0.20 1.68 Per 2 - 0.29 105-1034 C-2-1 prestudy nac nac Per 1 - 0.28 300 1.00 -7.75 Per 2 - 0.23 103-1019 1.C-2 30 0.35. 10.85 Per 2 - 0.20 102-1007 2-C-1 prestudy na" Ba' Per 1 - 0.77 60 0.75 7.71 Per 2 - 1.36 r-300 0.96 10.62 Per 3 - 0.60 a Treatments: C=Control TOBI 300 mg using PARI LC PLUS; I=TOBI 30 mg using Acrodose inhaler; 2=TOBI 60 mg using Aerodose inhaler; 3=TOBI 90 mg using Aerodose inhaler. b Measurable tobramycin in serum: tobramycin concentration> LOQ (0.2 pg/mL). c na - not available before the start of Pcriod 1. Table 6 also identifies predose serum specimens for periods 2, 3, or both that had measurable tobramycin in 4 of the 5 patients. These findings are also reflected in 5 non-zero mean amounts of predose tobramycin concentrations in periods 2 and 3. Three of the 5 patients exhibited measurable serum tobramycin after having received TOBI 300 mg during.the immediately preceding study period. These measurable predose results may represent carryover from previous treatment or non-specific assay interference, but the low frequency and magnitude of 10 the results suggests that a substantial effect on posttreatment analyses was unlikely. After each of the four TOBI treatments, serum tobramycin concentrations gradually increased, reaching a maximum at one hour after dosing (Figure 5), and declined thereafter with median half-lives ranging from 2.73 to 4.27 hours (Table 7 below).

39 For the Aerodose inhaler, mean serum tobramycin concentrations increased with increasing TOBI dose at each tinie during the 8-hour posttreatment period, but mean values for TOBI 90 mg were less at each posttreatnent time than those seen for TOBI 300 mg using the PARI LC PLUS nebulizer. 5 By 4 hours after the end of TOBI 30 mg and by 8 hours after TOBI 60 mg and 90 mg treatments, serum concentrations were below LOQ in at least half of the patients [median (50h percentile) serum concentrations = 0.0 pg/mL]. More than half of the TOBI 300 mg patients remained above the serum LOQ at 8 hours posttreatment. There was no apparent pattern of change in posttreatment serum 10 tobramycin concentrations from period to period for any of the 4 treatments, and there was no clear indication of the presence of a carryover (treatment by period interaction) effect in posttreatment results. Serum Pharmacokinetic Parameters Aftcr TOBI administration using the Aerodose inhaler, mean C,= and AUC 15 results increased linearly with dose after the administration of the 30, 60, and 90 mg doses (Table 7), suggesting linear pharmacokinetics. Dose normalized AUC results were similar among the Aerodose dose levels, suggesting dose proportionality. Comparing devices, C., and AUCo. for the TOBI 90 mg dose using the Aerodose inhaler were not as high as results achieved by the TOBI 300 mg dose 20 using the PARI LC PLUS nebulizer. However, the dose-normalized parameters were higher for the Aerodose inhaler at all three TOBI dose levels, indicating better efficiency of the new device. Similar to the sputum data, the relative bioavailability was approximately 3-fold higher for the Aerodose inhaler as compared to the PARI nebulizer. The variability based on AUCs was similar for both devices. 25 Exploratory analyses suggested that serum pharmacokinetic results were unaffected by characteristics present before treatments began (age, gender, body weight, FEVI % predicted at screening) and were unaffected by events noted after the start of treatments (device failure, occurrence of bronchospasm defined as a decrease 2 10% in FEVI, and relative change in FEVI % predicted).

40 TABLE 7. MEAN (SD) SERUM TOBRAMYCIN CONCENTRATIONS BY TIME AND PHARMACOKJNETIC PARAMETERS TOBI 300 mg Serum PARI LC TOBI 30 mg TOBI 60 mg TOBI 90 mg Pharmacokinetic PLUS' Aerodose inhalerb Acrodose inhaler Aerodose inhalerb Parameters (n = 49) (a = 34) (n = 32) (n =32) C.. (pgmL) 1.12 (0.44) 0.38 (0.17) 0.69 (0.34) 0.96(0.40) -No. pts with data: 49 30 32 32 - E vs C p-value': <0.001 <0.001 0.027 - E/C (90% Cls)d: (0.29, 0.36) (0.53, 0.66) (0.75,0.96) Dose-normalized C. (pg/mL)/mg 0.0037 (0.0015) 0.0127 (0.0058) 0.0116 (0.0056) 0.0106 (0.0045) - No. pts with data- 49 30 32 32 - E/C (90% CIs)': (2.52, 325) T.. (hr) . 1.05 (0.38) 1.14(0.42) 0.98(0.28) 1.14 (0.64) - No. pts with data: 49 30 32 32 Tm (hr) 3.42(1.63) 6.75(5.31) 4.16(2.34) 3.10(1.10) - Median Tia (hr.) 3.14 4.27 3.42 2.73 - No. pts with data: 49 )1 28 31 AUC..(bregg/mL) 4.96 (2.24) 1.43 (1.43) 2.98 (1.92) 3.94(1.52) -No. pts with data: 49 30 32 32 -E vs C p-value": <0.001 <0.001 0.165 - E/C (90/ CIs)': (0.18, 0.25) (0.46, 0.62) (075, 1.03) Dose-normalized AUCo. (hropg/mL)/mg 0.0166 (0.0075) 0.0478 (0.0477) 0.0496 (0.0319) 0.0438(0.0169) - No. pts with data: 49 30 32 32 - E/C (90% CIs)d: (2.51, 3.21) AUCo.(hrepg/mL) 6.66(4.32) 6.49(7.71) 5.11(4.62) 5.02(1.63) -No. pts with data: 49 11 . 28 31 a Control ( C ) treatment - TOBI 300 mg delivered by PARI LC PLUS nebulizer. b Experimental ( E ) treatments = TOBI 30, 60, or 90 mg delivered by Aerodose inhaler. c Pairwise contrast: TOBI 300 mg PARI LC PLUS group vs TOBI (30,60,90 mg) Aerodose groups. d Back-transformed 90% confidence intervals around the mean of the log ratio of E and C treatments. Serum limit of quantifiability (LOQ): 0.2 pg/mL Differences among treatment groups in serum Cma and AUCuG (Table 7 above; Figures 6 and 7) were statistically significant (p < 0.001) with no period or 41 carryover effects in the overall analyses. In pairwise comparisons, Cm and AUCo.

were significantly greater for TOBI 300 mg using the PARI LC PLUS than for TOBI 30 mg and TOBI 60 mg using the Aerodose inhaler (p < 0.001 in each comparison). C. was statistically significantly higher (p 0.027) for TOBI 300 mg compared to 5 the TOBI 90 mg dose, and AUCo. was slightly but not significantly (p = 0.165) greater for TOBI 300 mg than for TOBI 90 mg. The Aerodose inhaler was more efficient, regardless of TOBI dose, than the PARI LC PLUS nebulizer based on dose normalized sputum C. and AUCo 0 s results. Dose normalized means for these pharmacokinetic parameters were similar among 10 Aerodose treatments but approximately .1-fold higher than the dose normalized results after TOBI 300 mg delivered by the PARI LC PLUS nebulizer (Table 7). T.. (Table 7) was similar for the four treatments, averaging between 0.98 and 1.14 hours for Aerodose treatments and 1.05 hours for the TOBI 300 mg treatment using the PARI LC PLUS. Median TI/ ranged from 2.73 to 4.27 hours among the 15 Aerodose dose levels, compared to 3.14 hours for TOBI 300 mg using the PARI LC PLUS nebulizer. Median T]/ results using the Aerodose inhaler appeared to decrease with increasing TOBI. dose, but this was considered an artifact related to greater frequency of missing T 1 1 values (due to more BQL results)- at lower TOBI dose levels. 20 Exploratory analyses revealed no substantial association between serum pharmacokinetic results and patient characteristics present before treatments (age, gender, body weight, pulmonary function [FEVI % predicted] at screening) or emergent events after the start of treatments (device failure, occurrence of bronchospasm [decrease 10% in FEVI from predose to 30 minutes postdose], 25 relative change in FEVj. Urinary Recovery of Tobramycin Thirty-nine (39) of 49 completing patients had no measurable urine tobramycin concentrations before dosing in any of the 3 treatment periods, and 10 patients exhibited measurable predose urine tobramycin above the lower fIOQ in the 30 periods indicated in Table 8 below.

42 TABLE 8. MEASURABLE TOBRAMYCIN IN PREDOSE URINE SPECIMENS Previous Period Measurableb Predose 8-24 hour Urine Tobramycin during Treatment' Tobramycin Serum . Period Listed - Urine Patient Sequence TOBI dose Concentration T/ Tobratycin Concentration (mg) (pg/mL) (hr) W/ML) 103-1005 C-1-2 prestudy nad nad Per 1 - 3.80 300 3.92 4.80 Per 2 - 2.06 30 2.48 not estimable Per 3 - 1.20 103-1039 C-1-3 prestudy nd na- Per 1 - 1.82 300 6.76 1.87 Per 2 - 2.58 104-1024 C-1-3 300 5.14 3.16 Per 2 - 1.48 107-1027 C-1-3 prestudy na' na' Per I - 3.14 300 6.04 1.68 Per 2 - 1.58 104-1020 C-2-1 prestudy nad 'ad Per I - 1.74 300 13.40 2.93 Per 2 - 2.28 60 5.80 12.96 Per 3 - 1.30 109-1014 C-2-3 60 <1.0 4.06 Per 3 - 13.22 106-1025 1-C-2 300 5.14 3.80 Per 3 - 2.70 103-1012 2-C-3 300' 2.26 3.63 Per 3 - 1.16 101-1002 3-C-1 300 7.82 3.37 Per 3' - 1.12 103-1006 3-C-2 prestudy na ma' Per - 2.72 90 10.10 .3.14 Per 2 - 3.10 1-300 8.06 4.48 Per 3 - 2.08 a Treatments: C=Control TOBI 300 mg using PARI LCPLUS; I=TOBI 30 mg using Aerodose inhaler; 2=TOBI 60 mg using Aerodose inbaler; 3=TOBI 90 mg using Aerodose inhaler. b Measurable tobramycin in urine: tobramycin concentration > LOQ (1.0 ig/mL). c Dosing interrupted by inhaler malfunction. d na = not applicable; previous urine specimens were not collected. Table 8 shows that measurable urine tobramycin was recovered before dosing in periods 2, 3, or both for all 10 patients. Nine of the 10 patients had measurable 5 predose urine tobramycin after TOBI 300 mg treatment during the preceding study period. One patient exhibited measurable tobramycin in both predose serum and 43 predose urine, and these events both followed TOBI 300 mg administration during the. previous period. Although carryover effect cannot be ruled out, the overall results suggest that such an effect is unlikely. The elimination half-life in sputum ranged from 1.60 to 5 2.06 hours, and in serum ranged from 2.73 to 4.27 hours, with no substantial differences between the four treatments. Additionally, the amount of tobramycin excreted in urine was larger during the 0-8 hour period compared to the 8-24 hour period, consistent with the short Tj/ of tobramycin. More importantly, in clinical Phase L studies in patients, multiple daily administrations did not result in any 10 accumulation. Therefore it can be concluded that such carryover effect is most likely duc to nonspecificity of the assay. Consistent with the serum data, the amount of tobramycin excreted in urine was higher for TOBI 300 mg compared to TOBI 90 mg (Table 9 below). However, the percent of dose excreted in urine was 3-fold higher for the Aerodose inhaler at all 15 dose levels (16 to 18%) as compared to the PAR] LC PLUS nebulizer.

44 TABLE 9. MEAN (SD) URINARY RECOVERY OF TOBRAMYCIN BY TIME Urine TOBI 300 mg TOBI 30 mg TOBI 60 mg TOBI 90 mg Tobramycin PARI LC PLUS' Aerodose inhalerb Aerodose inhalerb Aerodose inhaler Recovery (n = 49) (n =34) (n = 32) (n = 32) Collection Interval Before and After Dosing: -12-0 hr predose 305.1 122.8 67.9 615.5 (pg) (1412.0) (340.7) (192.8) (3202.5) -No. pts with data 48 33 32 31 0 --8 hr postdose 15003.0 4835.6 8490.3 12304.8 (pg) (7116.2) (2649.6) (3159.6) (5352.7) -- No. pts with data 48 34 32 32 Dose-normalized 50.0 161.2 141.5 136.7 (P.O)IM g (23.7) (88.3) (52.7) (59.5) - No. pts with data 48 34 32 32 - E/C (90% Cls)': (2.50, 3.62) 8 - 24 hr postdose 3072.1 794.1 1367.4 2095.2 (pg) (2271.2) (853.1) (1118.8) (1818.7) - No. pts with data 47 34 31 31 Dose-normalized 10.2 26.5 22.8 23.3 (pg)/mg (7.6) (28.4) (18.6) (20.2) - No. pts with data 47 34 31 31 - E/C (90% Cls)': (2.44, 3.48) Total 0 -24 hour 18113.2 5629.7 9802.7 14588.1 (pg) . (8303.4) (2993.6) (3771.0) (6044.9) -No. pts with data 46 34 31 31 Dose-normalized 60.4 187.7 163.4 162.1 (pg)/mg (27.7) (99.8) (62.8) (67.2) - No. pts with data 46 34 31 31 - E/C (90% CIs)d: (2.23,3.27) Percent of Dose 6.0 18.8 16.3 16.2 Excreted (%)* 45 a Control (C) treatment = TOBI 300 mg delivered by PARI LC PLUS nebulizer. b Experimental ( E ) treatments = TOBI 30, 60, or 90 mg delivered by Aerodose inhaler. c % excreted = ((urinary recovery in pg + 1000 pg/mg) + Dose in mg] a 100%. Urine limit of quantifiability (LOQ): 1.0 pg/mL urine. For the Aerodose inhaler, mean 24-hour recovery of tobramycin from the urine increased with increasing TOBI dose during the study (Table 9 above; Figure 8). Tobramycin recovery appeared to be dose proportional for the Aerodose inhaler, as mean 24-hour recovery normalized for dose was similar among Aerodose 5 treatments. Comparing devices, mean recovery for the TOBI 90 mg .treatment was less than that seen for TOBI 300 mg using the PARI LC PLUS nebulizer. However, a greater percentage of the administered TOBI dose was recovered in the urine of patients who were dosed with the Aerodose inhaler (18.8%, 16.3%, and 16.2%, 10 respectively), irrespective of TOBI dose, than was recovered from patients who were dosed with the PARI LC PLUS nebulizer (6.0% of the administered TOBI 300 mg dose). The largest amount of tobramycin was recovered during the first 8 hours after dosing. There was no apparent pattern of pcriod-to-period change in posttreatment 15 urine tobramycin recovery for any of the 4 treatments. Although a potential carryover could not be ruled out in approximately 20% of the patients due to recovery of measurable tobramycin in predose urine, there was no clear indication of the presence of a carryover (treatment by period interaction) effect in posttreatment results. 20 The percent of administered dose recovered in urine over 24 hours postdose does not represent the delivered dose in the lung or absolute bioavailability. It is understood that a substantial amount of lung deposited dose still remains in the body at 24 hours postdose. Nebulization Time 25 Mean total nebulization time increased with increasing TOBI dose (Table 10 below; Figure 9) and was substantially less when the Aerodose inhaler was used at each TOBI dose level (mean ± SD for TOBI 30 mg = 2.8 ± 1.0 min; TOBI 60 mg 46 5.4 ± 2.1 min; TOBI 90 mg = 8.0 ± 2.5 min) than when the PARI LC PLUS nebulizer was used (TOBI 300 mg = 17.7 ± 4.7 min). TABLE 10. MEAN (SD) NEBULIZATION TIME TOBI 300 mg TOBI 30 mg , TOBI 60 mg TOBI 90 mg Parameter PARI LC PLUS' Aerodose inhaler Aerodose Inhaler Aerodose inhaler? (n ==51) (n =34) (n =32) (n =33) Nebulization 17.7(4.7) 2.8 (1.0) 5.2 (2.1) 8.0(2.5) Time 3 (min) -No. pts with data 51 34 32 32 I Control ( C ) treatment = TOBI 300 mg delivered by PARI LC PLUS nebulizer. I Experimental ( E ) treatments = TOBI 30, 60, or 90 mg delivered by Aerodose inhaler. 2 Total duration of nebulization excluding fill time. Conclusions 5 The Aerodose inhaler substantially reduced the amount of time required to nebulize the administered TOBI dose, compared to the approved PARI LC PLUS nebulizer, and nebulization time increased with increasing TOBI dose (TOBI 300 mg delivered by PARI LC PLUS mean = 17.7 minutes vs. 2.8 minutes, 5.4 minutes, and 8.0 minutes for TOBI 30 mg, 60 mg, and 90 mg, respectively). 10 Sputum tobramycin concentrations throughout the 8-hour sampling period after dosing increased with increasing TOBI dose through 90 mg delivered by the Aerodose inhaler, but results for TOBI 90 mg and TOBI 300 mg delivered by the PARI LC PLUS nebulizer did not differ substantially or consistently. Sputum tobramycin results were highly variable, with coefficients of variation approaching or 15 exceeding 100% for each treatment at all time points. On average, sputum concentrations reached their maximum at 10 minutes after each of the 4 treatments. By 2 hours after TOBI 30 mg and by 8 hours after TOBI 60 mg, 90 mg, and 300 mg, sputum concentrations were below the lower limit of quantifiability (LOQ) in at least half of the patients. 20 The mean of the maximum sputum concentration was significantly greater after TOBI 300 mg (mean = 985.65 pg/gm) than after TOBI 30 mg (329.05 pg/gm: 47 p < 0.001) and TOBI 60 mg (577.83 pg/gm: p = 0.002) but not TOBI 90 mg (958.00 pg/gm: p = 0.856; 90% CIs for the ratio of TOBI 90 mg / TOBI 300 mg C. = 0.72, 1.30). The Aerodose inhaler was more efficient than the PARI LC PLUS nebulizer based on sputum Cm. results adjusted for TOBI dose administered (TOBI 300 mg 5 with. PARI LC PLUS: dose-normalized mean C.= 3.29 (jig/gm)/mg; TOBI 30, 60, and 90 rmg with Aerodose = 10.97, 9.63, and 10.64 (pg/gm)/mg, respectively). Mean sputum T.., was virtually identical for TOBI 300 mg (mean - 0.26 hr) and TOBI 30 mg (0.24 hr) but was slightly less than T. for TOBI 60 mg (0.38 hr) and TOBI 90 mg (0.33 hr). 10 Mean sputum AUCo.s was significantly greater after TOBI 300 mg (mean = 1471.16 hrepg/gm) than after TOBI 30 mg (360.79 hr-Ig/gm: p <0.001) and TOBI 60 mg (804.78 hrepg/gm: p < 0.001) but not TOBI 90 mg (1275.23 hropg/gm: p = 0.465; 90% CIs for the ratio of TOBI 90 mg / TOBI 300 mg AUC4= 0.72, 1.14). The greater efficiency of the Aerodose inhaler was also seen in dose-normalized 15 AUCo results (TOBI 300 mg with PARI LC PLUS = 4.90 [hropg/gm]jmg; TOBI 30, 60, and 90 mg with Aerodose = 12.03, 13.41, and 14.17 [hrepg/gmJ/mg, respectively). No inferential analyses of sputum AUCo.. were performed due to high variability that increased with increasing TOBI dose. 20 Serum tobramycin concentrations also increased with increasing TOBI dose at each time during the 8-hour posttreatment observation period. Mean serum tobramycin concentrations reached their maximum at one hour after each treatment. By 4 hours after TOBI 30 mg and by 8 hours after TOBI 60 mg and TOBI 90 mg, serum concentrations were below LOQ in at least half of the patients. More than half 25 of the TOBI 300 mg patients remained above the serum LOQ at 8 hours posttreatment. Mean serum C.. was significantly greater after TOBI 300 mg (mean = 1.12 pg/mL) than after the other 3 treatments (TOBI 30 mg = 0.38 Jpg/mL, p < 0.001; TOBI 60 mg = 0.69 pg/mL, p <0.001; TOBI 90 mg = 0.96 ig/mL, p = 0.027). The 30 Aerodose inhaler was also more efficient than the PARI LC PLUS nebulizer based 48 .on serum C. results adjusted for TOBI dose administered (TOBI 300 mg with PARI LC PLUS: dose-normalized mean C.. = 0.0037 (pg/mL)/mg; TOBI 30, 60, and 90 mg with Aerodose = 0.0127, 0.0116, and 0.0 106 (pg/mL)/mg, respectively. Mean serum T. was similar for the 4 treatments (mean = 1.05 br, 1.02 hr, 5 0.98 hr, and 1.14 hr for TOBI 300 mg, 30 mg, 60 mg, and 90 mg, respectively). Mean serum AUCo 0 4 was significantly greater after TOBI 300 mg (mean 4.96 hr*pg/mL) than after TOBI 30 mg (1.43 hreyg/mL, p <0.001) and TOBI 60 mg (2.98 hrepg/mL, p < 0.001) but not TOBI 90 mg (3.94 hrepg/mL, p = 0.165; 90% Cis for the ratio of TOBI 90 mg / TOBI 300 mg AUCoG = 0.75, 1.03). The greater 10 efficiency of the Aerodose inhaler was also seen in dose-normalized AUCo.: results (TOBI 300 mg with PARI LC PLUS = 0.0166 [hropg/mL]/mg; TOBI 30, 60, and 90 mg with Aerodose = 0.0478, 0.0496, and 0.0438 [hre tg/mL]/mg, respectively). Serum AUC ) was not analyzed statistically due to high variability but generally appeared to increase as the TOBI dose increased. 15 Recovery of tobramycin from the urine within 24 hours after dosing increased with increasing TOBI dose during the study (expressed in mg [mg = pig/1000], mean urine tobramycin recovery =18.1 mg, 5.6 mg, 9.8 mg, and 14.6 mg after TOBI 300 mg, TOBI 30 mg, TOBI 60 mg, and TOBI 90 mg doses, respectively). Most of the tobramycin was recovered within the first 8 hours after dosing. Normalized for dose, 20 urine tobramycin recovery within 24 hours was 6.0%, 18.8%, 16.3%, and 16.2% of the administered TOBI 300 mg, TOBI 30 mg, TOBI 60 mg, and TOBI 90 mg doses, respectively. Results of the present study showed that TOBI 300 mg delivered by the PARI LC PLUS nebulizer (the control delivery system) and TOBI 30 mg, 60 mg, and 90 25 mg delivered by the Aerodose inhaler (the experimental delivery system) were safe and .well-tolerated by male and female cystic fibrosis patients. Fifteen patients (9 male and 6 female) experienced 24 instances of bronchospasm (decline in FEVI (liters) 2 10%). There were no statistically significant differences between control and any experimental treatment in the incidence of bronchospasm. Therm were no 49 overall treatment differences in quantitative change in FEVI from predose to 30 minute postdose measurement times. The study found no evidence that CF patients were at increased risk by reason of inhaling single TOBI doses of 30 mg, 60 mg or 90 mg compared to the single 5 TOBI 300 mg dose delivered by the PARI LC PLUS jet nebulizer. -The most frequently reported treatment emergent adverse events (cough increased, rhinitis, sputum increased, chest pain, asthma, and headache) and the SAEs reported by 4 of the patients were primarily associated with patients' underlying CF disease and related medical conditions. The incidence of these events before and after study 10 treatments was substantially similar, suggesting that neither TOBI dose levels.nor control and experimental inhalers altered ongoing symptomatology associated with patients' underlying medical conditions. There were also no clinically significant safety issues reflected in clinical laboratory test results, vital signs, or physical findings. 15 In this example, the Aerodose inhaler substantially reduced the time required for nebulization of all three dose levels (30 mg, 60 mg, and 90 mg) of TOBI compared to the nebulization time for the approved TOBI 300 mg delivery system using the PARI LC PLUS jet nebulizer. Average nebulization times were 2.8, 5.4, and 8.0 minutes using the Aerodose inhaler to deliver TOBI 30 mg, 60 mg, and 90 20 mg, respectively vs. 17.7 minutes using the PARI LC PLUS nebulizer to deliver TOBI 300 mg. The Aerodose inhaler therefore cut nebulization time of the TOBI 90 mg dose by more than 50% compared to the PARJLC PLUS nebulizer in the present study, and nebulization times for lower TOBI doses were reduced by even greater amounts. Present nebulization time results in CF patients 12 years of age with 25 baseline FEVI % predicted 40% were consistent with those obtained after single doses of TOBI 60 mg using the Aerodose inhaler (mean = 5.7 minutes) but slightly less than TOBI 300 mg results using the PARI LC PLUS nebulizer (mean = 20.4 minutes) in the TOBI gamma scintigraphy study of tobramycin deposition in the lungs of healthy adult male and female volunteers of Example 2, infra. - 50 This example demonstrates that TOBI 90 mg (but not TOBI 60 mg.or TOBI 30 mg) delivered by the Aerodose inhaler achieved similar actual pulmonary deposition, systemic absorption, and urinary recovery of tobramycin as that achieved by administration of the TOBI 300 mg dose delivered by the PAR LC. PLUS 5 nebulizer. Normalized for TOBI dose, the Aerodose inhaler was substantially more efficient than the PARI LC PLUS nebulizer in the delivery of aerosolized tobramycin. to the lungs and to the systemic circulation. Pulmonary deposition of tobramycin was measured by determination of sputum tobramycin concentrations and by calculation of sputum pharmacokinetic 10 parameters. Maximum sputum tobramycin concentrations were reached by 10 minutes after administration of each treatment, and concentrations were below the LOQ in half or more of the patients at 2 hours after TOBI 30 mg and at 8 hours after TOBI 60 mg, 90 mg, and 300 mg. The extent of pulmonary deposition of tobramycin, as measured by maximum sputum concentrations and sputum AUCo 4 15 results, increased with increasing TOBI dose through 90 mg, but TOBI 90 mg and TOBI 300 mg did not differ statistically (mean sputum C. - 958.00 and 985.65 pg/gm; mean sputum AUC.= 1275.23 and 1471.16 hregg/gm, respectively). Mean sputum C, results after TOBI 30 mg and 60 mg doses were significantly less than that of the TOBI 300 mg dose. Present sputum C. results achieved after the single 20 TOBI 300 mg dose were slightly less than sputum tobramycin concentrations achieved 10 minutes after a single TOBI 300 mg dose (mean sputum tobramycin concentration = 1237 pg/gm, median = 1090 pg/gm) in two large previously conducted Phase III pivotal trials. The results of this example demonstrate that at least one of the three TOBI 25 doses (TOBI 90 mg) delivered by the experimental Aerodose inhaler achieved similar actual sputum tobramycin concentrations and that these results in turn were similar to sputum results obtained in the prior pivotal studies supporting the' commercial TOBI product. It is also important that present sputum results demonstrated that the experimental Aerodose inhaler was substantially more 30 efficient, regardless of TOBI dose, in delivery of aerosolized tobramycin to the lung 51 than the PAIU LC PLUS jet nebulizer. Dose-nornalized sputum Cm,= was 10.97, 9.63, and 10.64 (j g/gm)/mg for TOBI 30 mg, 60 mg, and 90 mg delivered by Aerodose inhaler, respectively, compared to 3.29 (pg/gm)/mg for TOBI 300 mg delivered by PARI LC PLUS. Similarly, dose-normalized sputum AUCO..

2 was 5 12.03, 13,41, and 14.17 [hregg/gm]/mg for TOBI 30-mg, 60 mg, and 90 mg delivered by Aerodose inhaler, respectively, compared to 4.90 [hrOpfg/gm]/mg for TOBI 300 mg delivered by PARI LC PLUS. Systemic absorption of tobramycin was measured by determination of serum tobramycin concentrations and by calculation of serum pharmacokinetic parameters. 10 Maximum serum tobramycin concentrations were reached at one hour after each of the four TOBI treatments, and concentrations were below LOQ in half or more of the patients by 4 hours after TOBI 30 mg and by 8 hours after TOBI 60 mg and 90 mg. More than half of the patients at TOBI 300 mg had measurable serum tobramycin at 8 hours postdose. The extent of absorption of tobramycin, as measured by serum 15 C,, results, increased with increasing TOBI dose, as C. was significantly greater after TOBI 300 mug (mean = 1.12 jig/mL) than after each of the lower TOBI doses (mean = 0.38, 0.69, and 0.96 pg/mL for TOBI 30 mg, 60 mg, and 90 mg doses, respectively). Serum C. for TOBI 300 mg in the present study was slightly higher (mean SD = 1.10 0.44 pg/mL with a mean T. of 1.05 hr) than the mean serum 20 tobramycin concentration reported at one hour after TOBI 300 mg in the TOBI NDA (0.95 ± 0.50 pg/mL). Serum C. achieved by the Aerodose inhaler at the TOBI 90 mg dose level in the current study was virtually identical to the NDA serum concentrations one hour after TOBI 300 mg (mean = 0.96 ± 0.37 pg/mL), although it was significantly (p = 0.027) less than the current TOBI 300 mg. 25 Thus, present serum tobramycin results demonstrated that TOBI 90 mg delivered by the Aerodose inhaler were similar (AUCo) or nearly similar (C.) to those obtained after TOBI 300 mg delivered by the PARI LC PLUS nebulizer in the present study and in the prior pivotal studies supporting the TOBI commercial. product. Present serum results also demonstrated that the experimental -Aerodose 30 inhaler was substantially more efficient, regardless of TOBI dose, in delivery of 52 aerosolized tobramycin to the systemic circulation than the PARI LC PLUS jet nebulizer. Dose-normalized serum Cm, was 0.0127, 0.0116, and 0.0106 (pg/mL)/mg for TOBI 30 mg, 60 mg, and 90 mg delivered by Aerodose inhaler, respectively, compared to 0.0037 (pg/mL)/mg for TOBI 300 mg delivered by PARI 5 LC PLUS. Similarly, dose-normalized serum AUCo.- was 0.0478, 0.0496, and 0.0438 [hre Lg/mL]/mg for TOBI 30 mg, 60 mg, and 90 mg delivered by Aerodose inhaler, respectively, compared to 0.0166 [hropg/mL]/mg for TOBI 300 mg delivered by PARI LC PLUS. The greater efficiency of the Aerodose inhaler observed in present serum tobramycin results is consistent with greater efficiency and 10 less wastage of the tobramycin dose observed in earlier studies. Urinary recovery of tobramycin was measured by determining the cumulative amount of tobramycin recovered in urine collected for 24 hours after dosing. The amount of urinary tobramycin recovered within 24 hours postdose increased with increasing TOBI dose (expressed in mg [mg = pg/1000], mean urine tobramycin 15 recovery = 5.6 mg, 9.8 mg, 14.6 mg, and 18.1 mg tobramycin after TOBI 30 mg, 60 mg, 90 mg, and 300 mg). The results were not tested statistically, and it was not possible to determine whether TOBI 90 mg and TOBI 300 mg results for 24-hour recovery of urine tobramycin were similar or different. Normalized for dose by dividing the mean amount of tobramycin recovered 20 by the nominal amount of TOBI administered, urinary recovery of tobramycin was approximately 18.8 %, 16.3 %, 16.2 %, and 6.0 % of the administered TOBI 30 mg, 60 mg, 90 mg, and 300 mg doses, respectively. During the study, measurable tobramycin (i.e., above the lower limit of . quantifiability [LOQ] of the assay) was detected in 12-hour predose urine collections 25 in a total of 10 patients, including 5 patients before the first dose of study treatments in period one and all 10 patients before the second or third doses in periods 2 and 3 or both. Similarly, measurable tobramycin was detected in predose serum specimens in a total of 5 patients, including 3 patients before the first dose of study treatments in period one and 4 patients before dosing in periods 2, or 3, or both. A single patient 30 bad measurable tobramycin in both urine and serum.

53 Substantial variability is known to occur among patients in the rate and extent of uptake, renal accumulation, and elimination of aminoglycoside antibiotics, even in patients with normal renal function. Each of these factors may lengthen the amount of time that measurable concentrations of aminoglycoside antibiotics may be 5 detected in serum and urine. The present study -employed a prestudy washout interval of 7 days from previous prescription aminoglycoside antibiotic use and a 7 day interval between the 3 single doses of TOBI, an aminoglycoside antibiotic, during the crossover treatment periods. It is plausible that prestudy and on-study washout intervals in the study may have been too short for complete elimination of 10 residual tobramycin previously administered, if any. Measurable amounts of tobramycin for these patients would have had little effect on study results, since the amounts and concentrations detected were very small in nearly all cases, and no unusually high serum or urine tobramycin results were noted during the study. The Aerodose inhaler was a safe and efficient aerosolization and delivery 15 device for TOBI during the study. EXAMPLE 2 SCINTJGRAPHY STUDY In order to assess the in vivo lung deposition of 300 mg tobramycin (TOB*) inhaled using the PARI LC PLUST jet nebulizer / DeVilbiss PulmoAide* 20 compressor delivery system (current commercial delivery system) compared with the deposition of 60 mg tobramycin (TOBI*) using the AeroDoseT' inhaler in accordance with the present invention, a gamma scintigraphy study was performed. The imaging technique of gamma scintigraphy is a well-established method'' 2 that provides precise quantification of drug delivered to the lungs' 3 . It also provides an 25 assessment of the distribution of deposited drug in different lung regions (peripheral, intermediate and central lung regions corresponding to small airways, medium sized airways and large airways, respectively1 4 . Gamma scintigraphy is the only non invasive method currently available for obtaining this type of information. The study of this example was designed as an open label, randomized, single 30 center, single dose, two period crossover Phase I study of aerosol delivery 54 characteristics and safety of two inhalation devices in healthy adult volunteers. A maximum of 14 healthy male or non-pregnant, non breast-feeding female volunteers aged 18 to 65 years of age were to receive in random order two single doses of aerosolized antibiotic mixed with a sterile radiotracer (technetium bound to 5 diethylenetrianiinepentaacetic acid: 99 Tc DTPA) separated by a washout'interval of a minimum of 44 hours between doses. Radiolabeled aerosols consisted of a single 300 mg dose in a 5 mL solution of TOBI delivered by the control delivery system (PARI LC PLUS jet nebulizer with a PuloAide compressor) and a single 60 mg dose in a I mL solution of TOBI delivered by the experimental delivery system 10 (Aerodose inhaler). Aerosol delivery characteristics of control and experimental delivery systems were compared on .the basis of lung deposition of radiolabeled tobramycin determined by gamma scintigraphy, time to complete nebulization of aerosolized doses, serum concentrations of tobramycin determined by Abbott TDxFLx assays, 15 and serum tobramycin pharmacokinetic parameters. The safety of control and experimental TOBI delivery systems was compared on the basis of changes in pulmonary function, the incidence of treatment emergent adverse events, and the occurrence of clinically significant laboratory and clinical evaluations and of unusually high serum tobramycin concentrations. 20 The duration of study participation for each subject was to be approximately five weeks including a screening period of up to 3 weeks in duration, two treatment periods of approximately 9 hours each separated by a minimum 44-hour washout interval, and a follow-up period through 2 weeks after the end of dosing. Treatments 25 TOBI* was administered by inhalation as a single 300 mg dose and as a single 60 mg dose to each subject during the study. The 300 mg dose was supplied as a commercial ampoule of TOBI. The 60 mg dose of tobramycin solution was prepared by study site personnel by withdrawing 1.0 mL of solution from the 300 mg /5 mL commercial ampoule of TOBI into two unit dose syringes containing 0.5 inL 30 each.

55 Sterile "Tc DTPA was added as a radiotracer to both 300 mg and 60 mg solutions at the study site prior to instillation into the nebulizer. Sufficient ""Tc DTPA was added to both the 300 mg and the 60 mg dose so that no more than 10 MBq ""Tc DTPA was delivered to the subject with each single dose adniistered. 5 Using control and experimental aerosol delivery systems, each subject was to self-administer two single aerosolized doses of radiolabeled ("Tc DTPA) TOBI, one dose in each of two crossover treatment periods, according to the randomization scheme for the study. Subjects were instructed to use nose clips and breathe in a normal breathing pattern while inhaling the medication according to the instructions 10 for use for each inhaler. Control and experimental treatment delivery systems were specified as follows. Control Treatment Delivery System: PARI LC PLUS jet nebulizer with DeVilbiss PulmoAide compressor delivering 300 mg (5 mL) of TOBI. 15 Experimental Treatment Delivery System: Aerodose inhaler delivering 60 mg (1 mL) of TOBI. When the PARI LC PLUS nebulizer was used, S mL radiolabeled TOBI was added to the drug reservoir and nebulized without interruption until the nebulizer reservoir was dry. The PARI system was configured such that exhalation by the 20 subject did not result in escape of radioactive aerosol into the surrounding atmosphere. Exhaled droplets were collected using a filter attached to the side of the inhaler by a T-piece. In addition, a scavenger filter was placed above the inhaler, which was in turn connected to a vacuum pump. The scavenger system was used to collect any radiolabeled droplets escaping from the inhaler. 25 When the Aerodose inhaler was used, one 0.5 mL aliquot of radiolabeled TOBI was added to the drug reservoir and nebulized to dryness. A second 0.5 mL dose was then added to the reservoir and nebulized to dryness. The inhaler was surrounded with an exhaled air collection.box. Air was drawn through a filter at the back of the box using a vacuum pump.

56 Start and stop times of nebulization for both the Aerodose and PARI LC PLUS nebulizers were to be recorded in CRFs. Nebulization time for the Aerodose inhaler was not to include the time needed to refill the drug reservoir according to the protocol. 5 Enrolled volunteers were randomly assigned to two treatment sequence groups as illustrated below according to a randomization scheme. PARI 300 mg- / Aerodose 60 mg: e period 1: PARI LC PLUS with TOBI 300 mg e period 2: Aerodose with TOBI 60 mg 10 Aerodose 60 mg / PARI 300 mg: a period 1: Aerodose with TOBI 60 mg e period 2: PAR] LC PLUS with TOBI 300 mg All subjects randomly assigned to a single treatment sequence group received control and experimental treatments in the same order during the study, while 15 subjects assigned to the other treatment sequence group received treatments in the reverse order. Table 11. below shows the two sequences of treatment administration employed during the study via the randomization process. TABLE 11. TREATMENT SEQUENCE GROUPS AND SEQUENCE OF TREATMENTS IN THE STUDY Treatment Sequence Treatment Period I Treatment Period 2 Group_

C-E

2 C E E-C E C 'Subjects were randomly assigned to the two treatment sequence groups. 2 C and E refer to control and experimental treatments administered during the study as follows: C PARI LC PLUS jet nebulizer (60 mg/mL; 300 mg in 5 mL) E Aerodose inhaler (60 mg/mL; 60 mg in 1.0 mL) 20 Before dosing, TOBI formulations were radiolabeled with "'"Tc-DTPA in preparation for gamma scintigraphy to determine posttreatment iobranycin deposition in the lungs. Subjects practiced the inhalation procedure with both control 57 and experimental devices filled with normal saline. When the investigator was satisfied that the subject could reproducibly perform the correct inhalation technique, the inhaler was filled with the radiolabeled formulation, and the subject inhaled the radiolabeled dose until the nebulizer was dry and nebulization was stopped. 5 Immediately following inhalation of radiolabeled, aerosolized tobramycin, scintigraphic images were recorded to determine radioactivity associated with lung and oropharyngeal tobramycin deposition and with external items such as nebulizer parts, mouthpieces, filters, and tissues used by subjects. If not previously done within the last 5 years, a posterior lung ventilation scan was also performed during 10 the study after subjects inhaled the radioactive inert gas "'Kr to determine the lung outlines and facilitate the determination of regional deposition of radiolabeled tobramycin. Deposition of Tobramycin Assessment and comparison of tobramycin deposition patterns between PART 15- LC PLUS and Aerodose delivery systems was a primary objective of the study. Deposition patterns of inhaled, radiolabeled tobramycin were determined using scintigraphic imaging methodology. Lung, oropharyngeal, and (if necessary) abdominal radioactivity was measured from images obtained immediately after inhalation of each single dose of radiolabeled tobramycin using a gamma camera 20 (General Electric Maxicamera) with a 40 cm field of view and fitted with a low energy parallel hole collimator. Images were obtained as described below: * posterior view of the chest; * anterior view of the chest; e right lateral view of the oropharynx; 25 e anterior and posterior abdominal views if necessary, i.e., if activity had spread through the intestine, beyond the field of view in either of the chest images; 0 items external to the body of the subject as follows: * for the PARI LC PLUS system: nebulizer cup 58 " mouthpiece " exhalation filter and T-piece " scavenger filter " any tissues used by the subject 5 * for the Aerodose system: e Aerodose inhaler * exhaled air containment box and filter * any tissues used by the volunteer Additionally, a posterior lung ventilation scan was performed using the 10 radioactive inert gas, krypton (8"Kr), to determine the lung outlines. The lung outlines were used to divide lung images of each subject into central, intermediate, and peripheral lung zones in order to determine the amount of aerosolized tobramycin deposited in each of these zones 17 . Lung ventilation scans taken for subjects who participated in earlier studies were acceptable for use for this study 15 provided the scan was obtained within the last five years and the subject had no record of serious lung disease in the intervening period. Deposition zones of interest on scintigraphic images were additionally drawn around the oropharynx, esophagus, and stomach (including any activity in the small intestine). The counts obtained within all regions of interest were corrected for 20 background radioactivity, radioactive decay, and for tissue attenuation'. In regions where both anterior and posterior images were recorded, the geometric mean of counts in both images was calculated prior to correction for tissue attenuation. Determination of the percentage of the dose deposited in the oropharynx included activity adhering to the mouth and oropharynx together with any swallowed activity 25 detected in the esophagus, stomach, and intestine. All images were recorded using Micas X plus software installed on a UNIX based computer system. Images were stored on digital audio tape (DAT) for subsequent analysis and archiving. Scintigraphic data were analyzed by Pharmaceutical Profiles Ltd. (PPL) in accordance with the PPL Standard Operating 59 Procedure N 1013 "Lung Quantitative Data Analysis". The data were summarized to obtain the following parameters: e whole lung deposition (% of metered dose); * central lung zone deposition (% of metered dose); 5 e intermediate lung zone deposition (% of metered dose); e peripheral lung zone deposition (% of metered dose); e ratio of peripheral to central zone deposition (lung penetration index); e oropharyngeal deposition (including esophagus and stomach) (% of metered dose); 10 e inhaler deposition (PARI LC PLUS or AeroDose) (% of metered dose); * radioaerosol in exhaled air (filters) (% of metered dose); e radioaerosol on PART LC PLUS mouthpiece, T-piece, scavenger filter and subject tissues (% of metered dose); e radioaerosol on Aerodose exhaled air collection box and subject tissues (% of 15 metered dose). The counts in each area were expressed as a percentage of the metered dose that was determined from the sum of the total body counts in addition to those deposited on the inhaler and the exhalation filter. Since the volume of TOBI placed into each of the two inhalers was different, direct comparisons of the percentage 20 deposition values was problematic. To aid interpretation of the data, the percentage deposition values were multiplied by the nominal metered dose (300 mg for the PARI LC PLUS and 60 mg for the Aerodose inhaler) to obtain amounts of drug deposited in milligrams for each of the deposition parameters listed above. Nebulization Time 25 Assessment and comparison of nebulization time between PARI LC PLUS and Aerodose delivery systems was another objective of the study. Elapsed time from the start of nebulization (defined as the subject's first tidal breath after the inhaler was in place) until no more TOBI solution was aerosolized by the inhaler was 60 measured by staff at the site using a stopwatch. Nebulization time was not to include time needed for instillation of drug into the nebulizer between the repeat filling of the Aerodose inhaler. The length of any interruption in nebulization and the reason for interruption were recorded. 5 Serum tobramycin concentrations were determined for the present study, and pharmacokinetic parameters were calculated, to provide preliminary estimates of the bioavailability of 60 mg TOBI delivered by the Aerodose system in comparison with that of the marketed 300 mg TOBI formulation: Additionally, unusually high serum tobramycin results (> 4 pg/mL) were considered an important measure of safety 10 during the study. Venous blood samples (8 mL) for the determination of serum tobramycin concentrations were collected by intravenous cannula or by venipuncture before each single dose of TOBI and at 30 minutes and 1, 2, 4, and 8 hours after completion of dosing. The first one mL of blood withdrawn from the cannula was discarded, and 15 the subsequent 7 mL was withdrawn into serum sampling tubes. Cannulae were frequently flushed with saline during the course of the treatment day. Blood samples were centrifuged at approximately 1600 g for 10 minutes at 4*C. The resulting serum fraction was split into two aliquots by pipetting into two prelabeled polypropylene screw cap tubes. Tubes were stored at -20*C for each 20 study period and were then transferred to a -70*C freezer. The maximum tobramycin concentration (C,,,.) and the time to reach C.. (T,.) were the observed values. The elimination rate constant (kd: used to calculate AUCo. see next paragraph) was calculated as the negative slope of the log plasma concentration vs. time plot using the last two measurable concentrations.. Use of 25 more than two concentrations at or after Tm, is preferred for calculation of the elimination rate constant; however, several subjects had only two measurable tobramycin concentrations at the terminal phase after TOBI 60 mg using the Aerodose inhaler. The alternate method of calculating kI using the last two measurable concentrations was employed for all subjects for both period 1 and 30 period 2.

61 Area under the curve through 8 hours postdose (AUCos) and extrapolated to infinity (AUCo4.) were calculated for serum tobramycin concentrations using the linear trapezoid rule. Actual nebulization time was added to the time between predose and 30 minutes after the end of inhalation when calculating AUCo.. AUCo, 5 was extrapolated from the last measurable concentration to infinite time by adding the quantity equal to the last measurable concentration divided by the elimination rate constant (kc). Statistical Methods Planned in the Protocol Scintigraphic data were analyzed in accordance with the current version of 10 the PPL Standard Operating Procedure N 1013 "Lung Quantitative Data Analysis". Manipulation and calculation of -radioactivity counts were accomplished using a custom written region of interest program, where regions of interest were central, intermediate, peripheral, and stomach/mtestines if necessary. Numerical data were downloaded automatically from the Park Medical computer into a customized 15 spreadsheet. Due to the small number of subjects in the study, statistical analysis was performed only on whole lung deposition data and on selected pharmacokinetic parameters. All other study data were summarized descriptively. Descriptive summaries for quantitative data included sample size, mean, standard deviation, 20 median, minimum, maximum, and/or range values as appropriate. Descriptive summaries for qualitative or categorical data included number and percent of subjects with the characteristic. All clinical data manipulations, analyses, summaries, and transformations employed SAS version 6.12 zo-n Aerosol Delivery Analyses 25 Whole lung deposition was the primary endpoint for the analysis. The Wilcoxon one-sample, matched-pairs, signed ranks test was used to determine whether differences between the whole lung deposition patterns (percent and amount of metered dose deposited) for the two inhalers were significant. The significance level was set at a =0.05.

62 Serum pharmacokinetic parameters (C,., AUCO-s, and AUCo..) were analyzed for differences between delivery systems using a repeated measures analysis of variance. The statistical model included study period and delivery systems as fixed effects and subject as the random effect. The carryover effect from 5 treatment period 1 to 2 was also investigated. The significance level was set at a = 0.05, and tests of significance were two-sided. Additional deposition measures of interest, nebulization time, serum tobramycin concentrations and pharmacokinetic parameters were summarized and evaluated descriptively for apparent differences between aerosol delivery systems. 10 Study Drug Administration All subjects were successfully dosed according to the randomization schedule for the study, and all subjects received and completed both inhalation administrations. All subjects received single doses of TOBI 300 mg and TOBI 60 mg during the study. 15 Deposition of Radiolabeled Tobramycin Tobramycin deposition results indicated that the Aerodose system was more efficient than the PARI LC PLUS system. The Aerodose system with TOBI 60 mg delivered a greater percentage of the dose to the lungs (mean ± SD = 34.8 ± 10.1 %) than the PARI system with TOBI 300 mg (8.2 ± 3.6 %), and the difference was 20 statistically significant (p = 0.005) (see Table 12 below). Results from the analysis (n = 9) that excluded data from one patient were similar (means = 35.4% vs. 9.1% for Aerodose and PARI systems, respectively; p= 0.008). The actual arnount of drug delivered to the lungs (Table 13 below) was lightly but not significantly less (p = 0.202) using the Aerodose inhaler (20.9 ± 6.0 25 mg) than using the PARI inhaler (24.5 ± 10.7 mg). Excluding subject 1007, the analysis showed significantly less (p = 0.04) deposition of the Aerodose 60 mg dose (21.2 mg) than the PARI 300 mg dose (27.2 mg).

63 TABLE 12. MEAN (SD) PERCENTAGE DEPOSITION OF THE METERED TOBIDOSE Intent to Treat Excluding Subject 1007 (n 10) (n=9) TOBI TOBI TOBI TOBI Zone of 3 00mg 60mg .300mg 60mg Deposition PARI LC PARI LC PLUS AeroDose PLUS AcmDose p.value* Whole lung 8.2 (3.6)* 34.8 (10.1)* 9.1(2.2) 35.4(10.5) 0.005 central 2.4(12) 10.1 (4.0) 2.7(0.9) 102(42) intermediate 2.7(12) 11.6(3.6) 3.0(0.8) 11.8(3.7) peripheral 3.1 (1.3) 13.2 (3.4) 3.5(0.7) 13.4(3.5) ratio: peripheral / central 1.2 (0.5) 1.4 (0.4) 1.4(0.3) 1.4(0.4) Oropharynx(including 14.4(6.7) 31.5(11.6) 16.0(4.7) 31.5(123) esophagus and stomach) Inhaler 42.6(6.7) 15.2(8.4) 43.5(6.4) 15.1 (8.9) Exhalation filter 31.6(10.9) 16.9 (5.6) 28.3(2.7) 16.3(5.6) PARI-specific:P mou9p2ece 1.0(0.5) .1.0(0.5) T-piece 2.0(0.6) 2.0(0.5) t3sue 0.0(0.1) . )0.0(0.1) scavengcrfilter 0.1(0.1) _0.1(0.1) 1.4 (0.4) AeroIose-speciic: box 1.7(.5) 1.6(1.6) tissue 00.0(0.1) 0.0(0.1) * Wilcoxon matched-pairs signed ranks test on intent to treat dataset. Excluding Subject 1007: p = 0.008. Statistical significance: p.5 0.05. The Aerodose inhaler deposited proportionally more (Table 12 above) tobramycin in the lungs than in the oropharynx (mean 34.8 % vs. 31.5 % of the 60 5 mg dose), while the PARI LC PLUS nebulizer deposited less tobramycin in the lungs than in the oropharynx (mean 8.2 % vs. 14.4 % of the 300 mg dose). The ratio of lung to oropharyngeal deposition (whole lung deposition divided by oropharynx deposition in Table 12 above) was approximately 1.1 for the Aerodose inhaler compared to approximately 0.6 for the PARI LC PLUS nebulizer.

64 Regional deposition within the lung was predominantly peripheral and very similar for both inhalers (ratio of radioactivity in peripheral to central zones: Aerodose = 1.4 0.4; PARI LC PLUS = 1.2 ± 0.5). Substantially less tobramycin was deposited on the Aerodose inhaler (15.2 ± 5 8.4 %; 9.1 ± 5.1 mg; Tables 4 and 5, respectively) and exhalation filter (16.9 ± 5.6 %; 10.1 ± 3.3 mg) than on the PARI LC PLUS nebulizer (42.6 ± 6.7 %; 127.8 ± 20.0 mg) and filter (31.6 ± 10.9 %; 94.8 i 32.7 mg). No more than 2% of the metered dose was deposited on inhaler-specific surfaces or tissue paper used by subjects.

65 TABLE 13. MEAN (SD) AMOUNT (MG) OF DEPOSITION OF THE METERED TOBI DOSE Intent to Treat Excluding Subject 1007 (n=10) (a9) TOBI300mg TOBI60mg T01300mg TOB160mg Zone of PARI LC PARILC Deposition PLUS AeroDose PLUS AeroDose p-value Whole lung 24.5 (10.7)* .20.9 (6.0)* 27.2(6.7) 21.2(6.3) 0-202 central 7.3 (3.6) 6.0 (2.4) 8.0(2.8) 6.1(2.5) intermediate 8.0 (3.7) 6.9 (2.1) 9.9(25) 7.1 (2.2) peripheral 9.3 (3.8) 7.9 (2.1) 10.4(2.0) 8.1(2.1) Oropharynx (including 43.3 (20.2) 18.9 (6.9) 48.1(14.0) 18.9(7.4) esophagus and stomach) Inhaler 127.8 (20.0) 9.1 (5.1) 130.5 (19.2) 9.0(5.4) Erhalation filter 94.8(32.7) 10.1 (3.3) 84.8(8.1) 9.8(3.4) PARI-specific: mouthpiece 3.0(1.4)3.1 (1.5) T-piece 6.1(1.7) 5.9(0.6) tissue 0.1 (0.2) 0.1(0.2) - scavenger filter 0.4 (0.4) 0.4(0.4) AeroDose-specific: box 1.0 (0.9) 1.0(0.9) tissue . 0.0(0.1) 0.0(0.1) Wilcoxon matched-pairs signed ranks test on intent to treat dataset. Excluding Subject 1007: p i 0.04. SItatistical signiicance: p!0.05. Nebulization ime The nebulization time (i.e., time required from first tidal breath until the 5 nebulizer ran dry) was significantly shorter (p = 0.005) *for the Aerodose delivery system (mean E SD 5.70 ± 1.16 minutes) than for the PARJ LC PLUS system (20.40 ±3.47 minutes) (Table 14 below).

66 TABLE 14. MEAN (SD) NEBULIZATION TIME Intent to Treat (n 10) Nebulization Time* TOBI 300mg TOBI 60mg Parameter PARI LC PLUS AeroDose p-value Nebulization Time (minutes): Mean 20.40 5.70 0.005 SD 3.47 1.16 Minimum 17.0 4.0 Maximum 29.0 8.0 no. subjects . 10 10 Serum Tobramycin Concentrations and Pharmacokinetic Parameters Administration of TOBI 300 mg using the PARI LC PLUS delivery system produced higher mean serum tobramycin concentrations, a higher mean Cn, and a 5 greater AUCo.8) than administration of TOBI 60 mg using the Aerodose delivery system. The mean time to maximum tobramycin concentration (Tm.) wa's similar for the two delivery systems. Serum tobramycin concentrations for all subjects were below quantifiable limits before dosing in both period I and period 2. Figures I through 20 graphically .10 illustrate serum tobramycin concentrations before and after period 1 and period 2 dosing for all individual subjects. After dosing, two subjects had serum tobramycin concentrations that could not be measured (i.e., results were below the quantifiable limit of 0.20 pg/mL) during one of the two treatment periods. These two subjects were inevaluable for 15 pharmacokinetic analysis during the period indicated but provided evaluable results for the alternate period. Consistent with the high efficiency of the Aerodose system, mean serum tdbramycin concentrations were slightly lower throughout the 8-hour postdose observation period after delivery of TOBI 60 mg using the Aerodose system than 20 after delivery of TOBI 300 mg using the PARI LC PLUS system (Table 15 below).

67 Maximum plasma concentrations for both regimens were reached within 2 hours after the end of inhalation (TOBI 300 mg and PAR inhaler: I hr and 2 hr means = 0.63 pg/rnL; TOBI 60 rig and Aerodose inhaler: 2 hr mean = 0.48 pg/mL). By 8 hours after the end of inhalation, the plasma concentrations were below the limit of 5 quantitation in 5 subjects after the Aerodose inhaler and in two subjects after the PARI LC PLUS nebulizer.

68 TABLE 15. SERUM TOBRAMYCIN CONCENTRATIONS AND PHARMACOK[NETIC PARAMETERS Intent to Treat (a -10) TOBI 300mg TOBI 60mg Parameter* PARI LC PLUS* AeroDoseb Serum Tobramycin (pg/mL): Time Before and After DosinE: Predose 0.00(0.00) 9 0.00 (0.00) 9 30 minutes 0.42(0.24) 9 0.22(0.23) 9 1 hour 0.63 (0.29) 9 0.41 (0.22) 9 2 hours 0.63 (0.25) 9 0.48 (0.20) 9 4 hours 0.50(0.16) 9 0.38 (0.10) 9 8 hours 0.22(0.14) 9 0.13 (0.12)9 Pharmacokinetic Parameters: C..(pg/nIL) 0.677 (0.279) 9 0.482 (0.201) 9 T.(hr) 2.213 (0.923) 9 2.207 (0.788) 9 TW (hr) 4.269(1.058) 9 6.071 (3.357)9 AUCro.,>(gg/mL.hr) 3.622 (1.319) 9 2.553 (0.989) 9 AUCyo.) (ig/mLehr) 5.273 (1.699) 9 4.630 (0.967) 9 Pharmacokinetic Parameters Normalized to Dose: C.. (pg/mL)/mg 0.002 (0.001) 9 0.008 (0.003) 9 AUC(o) (pg/mLehr)/mg 0.012 (0.004) 9 0.043 (0.016) 9 AUqo..,)(pg/mLehr)/mg 0.018 (0.006) 9 0.077 (0.016) 9 * Cell entries are mean, (SD), no. of subjects. a TOBI 300 mg summary statistics exclude BQL results for Subject 1007 throughout period 2. b TOBI 60 mg summary statistics exclude.BQL results for Subject 1006 throughout period 1. Pharmacokinetic Results The mean of the maximum tobramycin concentrations for all subjects (Cm.- in 5 Table 15 above) was greater after TOBI 300 mg delivered by the PARI LC PLUS system (mean ± SD = 0.677 ± 0.279 jpg/mL) than after TOBI 60 mg delivered by the Aerodose system (0.482 ± 0.201 pg/mL). This mean difference in log,.C m was statistically significant (p = 0.0018), and there was no evidence to suggest the presence of a carryover effect in Cm (p = 0.6400). The Aerodose inhaler was more 69 efficient than the PARI LC PLUS nebulizer based on C.,c results adjusted for TOBI dose administered (TOBI 300 mg~ with PARI LC PLUS = 0.002 ± 0.001 (pg/mL)/mg; TOBI 60 mg with Aerodose =0.008 ± 0.003 (pg/mL)/mg). The time to maximum tobramycin concentrations (Ta,) was virtually 5 identical for the two delivery systems (mean = 2.213 hours for PARI LC PLUS and 2.207 hours for Aerodose systems in Table 15 above). T. results in the present study were consistent with observations in a previous study1 5 that peak serum tobramycin concentrations occurred at I to 2 hours after inhalation. The mean elimination half-life (T 1

,

2 ) was 4.269 hours for the PAR! LC PLUS 10 system and 6.071 hours for the Aerodose system (Table 7). The mean area under the serum concentration-time curve through 8 hours postdose (AUC(o.s)) was significantly greater (p = 0.0002 on log AUC(o.s)) after TOBI 300 mg delivered by the PARI LC PLUS system (3.622 ± 1.319 pg/mL.hr) than after TOBI 60 mg delivered by the Aerodose system (2.553 ± 0.989 g/mL.hr). There 15 was no evidence (p = 0.7858) to suggest the presence of carryover effect in AUCo.s). The greater efficiency of the Aerodose inhaler was also seen in dose-normalized AUCo.s) results (TOBI 300 mg with PARI LC PLUS = 0.012 ± 0.004 [pg/mL.hr]/mg; TOBI 60 mg with Aerodose = 0.043 ± 0.16 [.ig/mL-hr]/Mg). The mean area under the serum concentration by time curve extrapolated to 20 infinity (AUC(o..) in Table 7 above) was not significantly different (Jp = 0.5477 on log AUC(o.)) after administration of TOBI 300 mg using the PARI system (5.273 ± 1.699 pg/mL-hr) than after administration of TOBI 6 0 mg using the Aerodose system (4.630 ± 0.967 pg/mL.hr). No carryover effect was detected (p = 0.6006). The greater efficiency of the Aerodose inhaler was similarly seen in dose-normalized 25 AUC(o.) results (TOBI 300 mg with PARI LC PLUS = 0.018 ± 0.006 [pg/mL.hr]/mg; TOBI 60 mg with Aerodose = 0.077 ± 0.16 [pg/mL.hr]/mg). Unplanned, exploratory analyses suggested that female subjects achieved slightly higher Cn., AUC(o 4 s) and AUC((..) results than male subjects after both TOBI 300 mg and TOBI 60 mg treatments.

70 Extent of Exposure The duration of exposure to study drug and the dose of study drug were not varied in this study. All 10 subjects received a single 300 mg (5 mL) TOBI dose using the PARI LC PLUS jet nebulizer with the DeVilbiss PulmoAide compressor 5 delivery system (control treatment) on one occasion and a single 60 mg (I mL) TOBI dose using the Aerodose inhaler (experimental treatment) on a second o ccasion. Each dose was radiolabeled with up to I OMBq '"Tc-DTPA and administered in a randomized two-way crossover fashion separated by a 44-hour minimum washout period. 10 The mean whole lung deposition using the PARI LC PLUS nebulizei was 8.2% (24.5 mg) of the 300 mg TOBI dose. The mean whole lung deposition using the Aerodose inhaler was 34.8% (20.9 mg) of the 60 mg TOBI dose. A mean of 14.4 % (43.3 mg) and 31.5% (18.9 mg) of the corresponding doses were deposited in the oropharynx using the PARI LC PLUS and Aerodose inhalers, respectively. Both 15 inhaler systems were configured such that each subject's exhaled material was collected and did not escape with radioactive aerosol into the surrounding atmosphere. The PARI LC PLUS nebulizer also included a system to collect any radiolabeled droplets escaping from the nebulizer. Bronchospasm 20 In this study, decreases in the relative FEV % predicted 10% (not clinically significant if < 20%) and 2 20% (clinically significant) from predose measurements to 3 0-ninutes postdose measurements with each delivery system were used as indicators of bronchospasm (airway reactivity). Reductions in FEVI % predicted 20% were considered clinically significant for the purposes of the study. No 25 subject had a drop in FEVI % predicted 10% from predose to postdose regardless of delivery system during this study.

Discussion and Overall Conclusions The study of this example demonstrates that the AeroDose" inhaler was more efficient in delivery of aerosolized tobramycin to the lungs of heIlthy adult 30 volunteers than the approved PARI LC PLUS jet nebulizer with DeVilbiss 71 PulmoAide compressor. Since the Aerodose inhaler is breath-actuated and generates aerosol only during inhalation, proportionally more of the Aerodose dose should be delivered to the lungs than is delivered by the PARI LC PLUS, and there should be minimal wastage of drug by aerosolization during exhalation or by incomplete 5 aerosolization of the contents of the drug reservoir. During the. study, the Aerodose inhaler delivered a significantly greater percentage of the dose to the lungs than the PAR] LC PLUS nebulizer (mean 34.8% vs. 8.2%: p = 0.005). The actual amount of the dose deposited in the lungs was slightly but not significantly less using the Aerodose inhaler than using the PARI LC 10 PLUS nebulizer (20.9 mg vs. 24.5 mg: p = 0.202). These data demonstrate that the Aerodose inhaler delivered nearly as much tobramycin to the lungs as the PARI LC PLUS nebulizer despite nebulizing one-fifth the amount of tobramycin. Approximately 32% of the Aerodose dose was wasted on the inhaler and exhalation filter combined. By contrast, more than 74% of the PARI LC PLUS dose 15 was wasted by deposition on the inhaler and exhalation filter. When the Aerodose inhaler was used, 15.2% (9.1 mg) of the 60 mg TOBI dose remained deposited on the inhaler, and 16.9% (10.1 mg) was deposited on the exhalation filter. Since no aerosolization occurred during exhalation when the Aerodose was used, the observed deposition could have been due only to seepage 20 through the mouth-inhaler seal or to residual radiolabeled tobramycin inhaled but immediately exhaled and not deposited in either the lungs or the oropharynx (including esophagus and stomach). Four subjects were noted to have either experienced problems maintaining a seal around the mouthpiece of the Aerodose inhaler or reported that the inhaler failed to nebulize one of the two aliquots of the 25 dose solution. These subjects had approximately 47%, 19%, 53%, and 26%, respectively, of the 60 mg dose deposited on the inhaler and exhalation filter combined. The highest two of these figures were above the range noted for the rest of the subjects (ranging from 17% to 40% deposited on inhaler and exhalation filter combined). Problems with incomplete nebulization or wide variation -in subject 72 inhalation effectiveness may have contributed to the amount of wastage of drug during Aerodose usage in the present study. By comparison, when the PARI LC PLUS jet nebulizer was used, 42.6% (127.8 mg) of the 300 mg TOBI dose remained deposited on the inhaler, and 31.6% 5 (94.8 mg) was deposited on the exhalation filter. Presumably, most or-all of the exhalation filter deposition was due to continued aerosolization and consequent loss of drug while subjects exhaled. Thus, both the Aerodose inhalers and PARI LC PLUS nebulizers wasted drug product in the present study by reason of retention of radiolabeled drug on or in the 10 inhaler or deposition of drug on the exhalation filter (an average of approximately 19 of 60 mg wasted when the Aerodose inhaler was used and approximately 223 of 300 mg wasted when the PARI LC PLUS nebulizer was used). The proportion of the total dose wasted using the Aerodose inhaler was less than half of that wasted using the approved PARI LC PLUS nebulizer. 15 The Aerodose inhaler also appeared to exhibit better "targeting" or delivery of the dose to the lungs, the target site of the usual P. aeruginosa infection in cystic fibrosis patients, than the PARI LC PLUS nebulizer. The Aerodose inhaler deposited slightly more tobramycin in the lungs than in the oropharynx, esophagus, and stomach (lungs 34.8% vs. 31.5% of the 60 mg dose). By comparison, the PARI LC 20 PLUS nebulizer deposited proportionally less of the dose in the lungs than in oropharynx, esophagus, and stomach (lungs 8.2% vs. 14.4% of the 300 mg dose). The ratio of lung to oropharyngeal, esophagus, and stomach combined was approximately 1.1 for the Aerodose inhaler and 0.6 for the PAR LC PLUS nebulizer. 25 In addition to greater efficiency by greater delivery of drug to the lungs and proportionally greater targeting of the lungs, the Aerodose inhaler was also anticipated to be more efficient by reason of proportionally greater delivery of tobramycin to peripheral rather than central lung regions. The Aerodose particle MMD is smaller (mean MMvD = 4.0 pm) than that produced by the PAR- LC PLUS 30 nebulizer (mean MMW = 4.8 pm), so the expectation was that the Aerodose inhaler 73 would deposit a greater proportion of aerosol generated during inhalation in the peripheral airways than the PARI LC PLUS. During the study, the Aerodose inhaler deposited 13.2% (7.9 mg) of the 60 mg dose in the peripheral airways, while the PARI LC PLUS nebulizer deposited 3.1% (9.3 mg) in peripheral airways. Although 5 -the Aerodose inhaler achieved proportionally greater peripheral deposition than the PARI LC PLUS nebulizer, both inhalers fell short of amounts predicted for peripheral deposition based on theoretical considerations (Aerodose estimated to peripherally deposit 60% (36 mg) of the 60 mg dose = 1.0 mL fill volume * 0.95 aerosolization a 0.62 respirable particles; PARI LC PLUS estimated to peripherally 10 deposit 16% (48 mg) of the 3 00 mg dose = 5.0 mL fill volume 9 0.64 aerosolization . 0.44 respirable particles). Results of the study also showed that the Aerodose inhaler required significantly less nebulization time than the PARI LC PLUS nebulizer (mean 20.4 vs. 5.7 minutes, respectively). The 5.7 minute average nebulization time for the 15 Aerodose inhaler did not include the amount of time needed to fill the drug reservoir before nebulization of the second aliquot. Based on nebulization time results and other inhaler features including portability, ease of use, and lack of a need for a compressor, it is anticipated that the Aerodose inhaler would improve patient compliance. 20 Serum tobramycin concentrations, maximum concentrations, and extent of absorption were greater after administration of TOBI 300 mg using the PARJ LC PLUS nebulizer than after administration of TOBI 60 mg using the Aerodose inhaler. These results appeared to be consistent with amounts of tobramycin deposited in lungs and oropharynx (including esophagus and stomach) combined where systemic 25 absorption occurred (mean tobramycin deposited in lungs and oropharynx combined = 67.8 mg after TOBI 300 mg; mean = 39.8 mg after TOBI 60 mg). Mean serum tobramycin concentrations were higher throughout the 8-hour observation period after administration of TOBT 300 rag using the PARI LC PLUS nebulizer than after administration of TOBI 60 mg using the Aerodose inhaler. Mean C,,, values were 30 0.677 and 0.482 ptg/mL for TOBI 300 mg and TOBI 60 mg, respectively (statistically 74 significant: p = 0.0018). Mean Tm results for both inhalers were virtually identical (2.213 and 2.207 hours, respectively). Apparent absorption of tobramycin was significantly greater during the 8-hour postdose period after TOBI 300 mg than after TOBI 60 mg (mean AUCos = 3.622 and 2.553 pg/mL-hr, respectively; statistically 5 significant: p = 0.0002), but no treatment differences were noted in AUC.

1 (TOBI 300 mg and TOBI 60 mg means = 5.273 and 4.630 pg/mL.hr, respectively; p 0.5499). Current results suggested that the 60 mg TOBI dose aerosolized using the Aerodose inhaler produced ' tobramycin deposition and serum tobramycin 10 concentration results that were significantly or substantially less than results obtained after aerosolization of the approved TOBI 300 mg dose using the PARI LC PLUS nebulizer. Normalized for administered dose,'the Aerodose inhaler was substantially more efficient on a per milligram basis in delivery of tobramycin to the systemic circulation than the PARI LC PLUS nebulizer. These results are consistent with the 15 higher deposition (on a milligram basis) in the lung. Results of the study also showed that single doses of TOBI 300 mg delivered using the PARI LC PLUS jet nebulizer and of TOBI 60 mg delivered using the Aerodose breath actuated nebulizer were safe and well-tolerated by healthy adult male and female volunteers. No instances of bronchospasm were observed, and no 20 notable quantitative changes in pulmonary function were seen. No notable adverse events (AEs) were reported by subjects, and there were no apparent differences between treatment groups in incidence of any AE. Six treatment emergent A~s were reported by 4 subjects, but all events were mild in intensity. Two instances of headache were considered possibly or definitely related to treatment. No clinically 25 significant laboratory results or changes in results were observed. No adverse vital signs, body weights, physical findings, or electrocardiogram results were observed. No evidence of systemic toxicity, as measured by unusually'high serum tobramycin concentrations, was observed.

75 EXAMPLE 3 INVIVO STUDY2 A comparison was made of the safety, pharmacokinetics, aerosol delivery characteristics, and nebulization time of the conventional dose and inhalation 5 delivery system (5 mL ampoule containing 300 mg tobramycin and 11.25 mg sodium chloride in sterile water for injection (TOBI* tobramycin solution for inhalation, Chiron Corporation, Seattle, Washington), pH 6.0; administered with a PARI LC PLUS" jet nebulizer with a DeVilbiss PulmoAideTm compressor set to deliver an output pressure of 20 psi - the "control delivery system") with a dose of 420 mg 10 Tobramycin Solution for Inhalation at 120 mg/mL (excipient 3.5 mL of 1/4 normal saline adjusted to a pH of 6.0 0.5; 420 mg in 3.5 mL) delivered by the PAR LC

PLUS

T

N jet nebulizer with a Invacare MOBILAIRETH compressor set to deliver an output pressure of 35 psi (the "experimental delivery system"). The study was designed as an open label, randomized, single-dose, 15 multicenter, two treatment, active-control, and parallel trial. Each patient was administered a single aerosolized dose of study drug with either the control delivery system or the experimental delivery system. In accordance with the study design, a total of 36 eligible male and female patients 12 years of age or older with -a confirmed diagnosis of cystic fibrosis were enrolled with a minimum of 4 patients at 20 each site. A 2:1 randomization ratio was employed for assignment of patients to the treatment groups. In the presence of the investigator or study coordinator, each patient was to self-administer either a single dose of 300 mg TOB1 with the control delivery treatment or a single dose of 420 mg Tobramycin Solution for Inhalation with the experimental delivery treatment as listed below. 25 Control Treatment: Aerosolized 300 mg TOBI was delivered by PARI LC PLUS jet nebulizer/DeVilbiss PulmoAide compressor: Preservative free tobramycin for inhalation 60 mg/mL (excipient 5 mL of 1/4 normal saline adjusted to a pH of 6.0 ± 0.5); 300 mg in 5 mL; lot number 03K1 C (TOBI* at 60 mg/mL).

-

76 Experimental Treatment ( 4 20mg Tobramycin Solution for Inhalation or Aerosolized 420 mg Tobramycin Solution for Inhalation (TSI) was delivered by PARI LC PLUS jet nebulizer/Invacare MOBILAIRE compressor: Preservative 5 free tobramycin 120 mg/mL (excipient 3.5 mL of 1/4 normal saline adjusted to a pH of 6.0 ± 0.5); 420 mg in 3.5 it. Both 300 mg TOBI and 420 mg Tobramycin Solution for Inhalation are sterile, non-pyrogenic, preservative-free antibiotics prepared for aerosolization. Each mL of TOBI* contains 60 mig tobramycin and 2.25 mg sodium chloride in sterile 10 water for injection, pH 6.0 ± 0.5 (control treatment). Each mL of TSI contains 120 mg tobramycin and 2.25 mg sodium chloride in sterile water for injection, pH 6.0 0.5 (experimental treatment). Drug supplies for this study were manufactured by Automated Liquid Packaging (ALP), Woodstock, IL. -All repackaging, labeling, and distribution for clinical use was provided by Packaging Coordinators, Inc. (PCI), 15 Philadelphia, PA. Study drug and device supplies were shipped from Chiron Corporation, Emeryville, CA for each patient upon enrollment in the study. The duration of study participation for each patient was approximately two weeks including a brief (one day one week before treatment) screening period, one day treatment period, and a follow-up one-week after treatment. Study treatments 20 were evaluated for safety and aerosol delivery characteristics up to eight hours post dose on the day of the single dose treatment administration. The patient was to return to the clinic for a seven day post-treatment follow-up assessment of safety. There were no planned interim safety analyses. Criteria for evaluation: 25 Safety: - Incidence of bronchospasm defined as FEVI decrease of ; 10% and FEV, decrease of2 20% from predose to 30 minutes postdose; - Relative change and absolute change in airway response (FEVy) after sjigle dose of study drug; 30 e Laboratory measures of safety (clinical lab tests, spirometry testing); 77 - Incidence of treatment emergent adverse event: Aerosol Delivery: * Pharmacokinetic assessment of sputum and serum tobramycin concentrations; - Sputum was collected at pre-dose and 15 minutes, 1, 2, 4, and 8 hours 5 after dosing; - Serum was collected at pre-dose and 10 minutes, 1, 2, 4, 6, and 8 hours after dosing; e Nebulization time. Statistical methods: All patients who received a dose of study treatment were 10 evaluated for safety and aerosol delivery characteristics. Rate of bronchospasm measured by the percent of patients with 10% and 220% relative decrease in FEVI% from pre-dose to 30 minutes post-dose was summarized and compared between treatments using the Fisher's exact test. A two sample t-test was used to compare the relative change in FEV 1 % from 15 predose to 30 minutes postdose between experimental and control treatments. Summary statistics for relative and absolute change in FEVI were tabulated by treatment. Sputum and serum area under curve (A UCo.i) and maximum concentrations (C,,r) were summarized and analyzed for treatment differences using a general linear 20 model analysis of variance (ANOVA). Pharmacokinetic parameters were calculated using a non-compartmental model. Sputum and serum concentrations were summarized and graphically illustrated by treatment. Laboratory measures of safety and incidence of treatment-emergent adverse events were summarized and descriptively compared between treatments. 25 Nebulization time was recorded and summarized for each of the two delivery treatments. Safety Variables Aerosol delivery variables were tobramycin concentrations in sputum and serum, sputum and serum tobramycin pharmacokinetic parameters, and aerosol 78 nebulization time. Safety variables were the incidence and severity of bronchospasm, measured as the number of patients experiencing a 2 10% and a 20% decrease in forced expiratory volume in one second (FEV:) within 30 minutes after dosing (a 20% decrease in FEVI was considered clinically significant), the 5 incidence of treatment emergent -adverse events (AEs), clinical laboratory test results, the number of patients with serum tobramycin concentrations 2 4 pg/mL, physical examination findings, and vital signs results. Primary Aerosol Delivery Variables Evaluation of the aerosol delivery characteristics of 420 mg Tobramycin 10 Solution for Inhalation at 120mg/mL delivered by the PARI LC PLUS

T

M/Invacare MOBILAIREu delivery system compared to 300 mg TOBI* at 60 mg/mL delivered. by the FDA-approved PARI LC PLUSmh/DeVilbiss PulmoAide? 4 delivery system was based on determination of sputum and serum tobramycin concentrations, calculation of certain sputum and serum pharmacokinetic parameters, and 15 measurement of nebulization time. Sputum Tobramycin Concentrations: Sputum samples were expectorated by patients from a deep cough and collected before day I dosing (predose) and at 0.25, .1, 2, 4, and 8 h after the end of the nebulization period. Sputum samples were collected as close as possible to specified times and were considered to have been 20 drawn on time within ± 2 minutes for the 15-minute posttreatment collection and within 10 minutes for the 1-, 2-, 4-, and 8-hour posttreatment collections. Samples collected outside these intervals were considered protocol deviations. A minimum 100 mg sputum (not saliva) sample was collected before the single dose of study treatment to determine the baseline tobramycin concentration. Immediately after 25 dosing, patients rinsed their mouths with 30 mL of normal saline, gargled for 5-10 seconds, and expectorated the rinse.. This sequence of post-treatment rinsing was repeated for a total of three rinses. Sputum samples were stored at -70*C or below until analysis. The concentration of tobramycin was analyzed using reversed-phase high-performance liquid chromatography (HPLC) with ultraviolet detecti6i:,i Patient 30 sputum samples were first liquefied with 0.2 N NaOH and diluted with Tris buffer 79 (20.0 g Trizma base/L). Sputum standard samples were prepared by spiking diluted pooled sputum from CF patients with tobramycin to final concentrations of 0, 20, 40, 100, 200, 400, and 1000 pg/g of sputum. Assay quality control samples were prepared by spiking diluted pooled sputum to contain 40, 300, and 800 pg/g. The 5 internal standard sisomycin (100 pL, 0.15 mg/mL in Tris buffer) was added to 100 pL of each standard, control, and subject sample, followed by 400 pL of acetonitrije and 50 pL of 2 ,4-dinitrofluorobenzene (0.17 g/mL). The sample reaction mixtures were heated in a dry-block heater for I h at 80"C. After addition of 600 pL of 60140 acetonitrile/water (v/v), 50 pL was analyzed by HPLC. Samples were injected onto a 10 Waters Nova-Pak C-18, 3.9 x 150 mm, 4 pm column connected to a Waters HPLC with 600E pump, 486 or 2487 ultraviolet detector (,. = 360 nm) and 717 Plus autosampler. The mobile phase consisted of 0.2% acetic acid in acetonitrile (39/61, v/v), pumped at a rate of 1.5 mUmin for 5 min, 2.0 mUmin-for an additional 9 or 10 min, depending on the length of the run. Waters Millennium-32 C/S LC Software 15 (version 3.20) was used to operate the Waters HPLC instruments as well as acquire raw data, process, compute, and report the analytical results. The ratio of the peak height of tobramycin to the internal standard sisomycin (PHR) was calculated. The assay was completed in 8 runs. Retention time ranges of 4.2 to 4.4 min, and 10.8 to 11.8 min were observed for tobramycin and sisomycin, respectively. A linear 20 relationship existed between PHR and concentration from 20 to 1000 pg/g for sputum. The regression model was PHR = Bx + A (x = tobramycin concentration), weighted 1/x. The lower limit of quantitation was 20 pg/g. The concentrations of the standard samples were within 97 to 105% of the nominal concentration, with coefficients of variation not higher than 3.4%. The precision of the assay, as 25 reflected by the CV of the quality control samples, was 2.3%, 2.2% and 2.6%, for the 40, 300, and 800 pig/g samples, respectively. The accuracy of the method, reflected by the interassay recoveries of the quality control samples, was 103%, 99%, and 98% for the 40, 300, and 800 pg/g quality control samples, respectively. Overall, this method exhibited suitable accuracy and precision for pharmacokinetic analysis. - 80 Serum Tobramycin Concentrations: Blood samples were collected at predose and at 0.167, 1, 2, 4, 6, and 8 h after the end of the nebulization period. Samples were collected as close as possible to specified times and were considered to have been drawn on time within ± 2 minutes for the 10-minute posttreatment collection 5 and within ± 10 minutes for the 1-, 2-, 4-, 6-, and 8-hour posttreatment collections. Samples collected outside these intervals were considered protocol deviations. Serum was harvested and stored at -70*C or below until analysis. Concentrations of tobramycin in serum were analyzed with a modified fluorescence polarization immunoassay (FPIA) method using the Abbott TDx*/TDxFLx* System. Samples 10 were added directly to the dilution well of the sample cartridge. The net polarization was acquired by the TDx*/TDxFLx* apparatus and manually entered into an Oracle database. A weighted four parameter logistic equation was used to calculate the concentrations of tobramycin. The concentrations of tobramycin were reported in terms of free base equivalents. For assaying the subject samples of the study, 15 calibration standards (0.050, 0.100, 0.200, 0.400, 0.600, 0.800, 1.000 pg/mL) and quality control samples (0.150, 0.400, and 0.750 tg/mL) were prepared in house. The assay was completed in 8 runs. A linear relationship existed between polarization response and concentration from 0.050 pg/mL to 1.00 tg/mL. The lower limit of quantitation was 0.050 pg/mL. The precision of the assay, as reflected 20 by the CV of the quality control samples, was 3.3%, 4.9%, and 4.9% for the 0.150, 0.400, and 0.750 pg/mL samples, respectively. The accuracy of the method, reflected by the mean interassay recoveries of the quality control samples, was 101%, 103%, and 104% for the 0.150, 0.400, and 0.750 pig/mL samples, respectively. Overall, this method exhibited suitable accuracy and precision for pharmacokinetic 25 analysis. Nebulization Time: Nebulization time was defined as the length of time from the start of the patient's first tidal breath to completion of aerosol administration. Aerosol administration was complete when the nebulizer began to sputter. - If aerosol 30 administration was interrupted for any reason, the time of interruption and start and 81 stop times of continued aerosol administration were recorded. If dosing was interrupted, nebulization time was considered to be not calculable. Residual Tobramycin in the Nebulizer: The amount of residual tobramycin solution remaining in the nebulizer after completion of aerosol administration was 5 determined by recording pretreatment and posttreatment weight of the nebulizer system including nebulizer, filter valve, and study drug. The research coordinator collected residual study drug remaining in the nebulizer after aerosol administration into a vial labeled with patient information. The vial was returned for measurement of the amount of drug output from the nebulizer and for determination of the extent 10 of the concentration of study drug left in the nebulizer. Safety Variables Bronchospasm: The study protocol prospectively identified bronchospasm as an adverse airway response to inhalation of aerosolized antibiotic of particular relevance to patients with cystic fibrosis. In order to determine whether current study 15 treatments produced bronchospasm, patients performed spirometry (pulmonary function) tests to measure FEVi before and 30 minutes following completion of study treatment administration according to the method described in the protocol. Airway response to the study drug was assessed by evaluating the relative percent change in FEVI from predose to 30 minutes after the end of treatment using the 20 following formula. relative FEVI % change = 30 min postdose FEVi- predose FEVI x 100% predose FEVI Bronchospasm was defined as a decrease in FEVI of 10% at 30 minutes after dosing, relative to the predose result. A decrease in FEVI of 2 20% was 25 considered to represent clinically significant bronchospasm. Moreover, if there was a posttreatment decrease in FEVI of 2 30%, spirometry was to be repeated until the FEV decrease was < 10% below the predose result An FEVI % decrease > 30%, and all symptoms associated with the change in pulmonary function,-were to-be recorded as adverse events. The protocol defined the severity of decrease in FEVI 82 based in part on the National Cancer Institute (NCI) Common Toxicity Criteria Adverse Events Grading Scale. However, slight inconsistencies in the protocol definitions of bronchospasm and of the severity of FEVI changes were noted during preparation of the analyses and report. To resolve the differences, the actual. system 5 used during the analysis to classify the severity of FEVI changes relative to the predose result is listed below. TABLE 16. AIRWAY RESPONSE (FEVI) (BRONCHOSPASM) FEV % DECREASE BELOW PREDOSE VALUE 10 Severity Protocol Classification Analysis Classification Mild: . 210%.-: s20% -10% - < 20% Moderate: > 20% - 5 30% 20% - < 30% Severe: > 30% 30% Clinical Laboratory Tests 15 At screening, laboratory tests were performed to measure serum creatinine, blood urea nitrogen (BUN), urine protein (proteinuria by dipstick), and to detect pregnancy in females of childbearing potential. If abnormal at screening, serum creatinine, BUN, and urine protein tests were to be repeated before the time of dosing. Final test results were obtained based on specimens drawn at the follow-up 20 visit on day 8 of the study. After the mean body weight difference between treatment groups became known by Chiron personnel, estimated creatinine clearance was calculated for patients using the Cockroft-Gault equation below to evaluate renal clearance characteristics of the two groups and to clarify the pharmuacokinetic results of the 25 study. Male patients: estimated creatinine clearance (mt/min) (1 4 0-age[yr])(body weight[kg])/72*(serum creatinine 30 [mg/dL]) 83 Female patients: estimated creatinine clearance (mUmin)= 0.85*((140-age[yr])(body weight[kg])/72*(seun creatinine [mg/dL]) 5 All abnormal laboratory test results, whether present on entry into the study or arising during the study, were evaluated by the study investigator for clinical significance and relationship to study drug. If the abnormal result was considered unrelated to study drug, the investigator was to identify the probable cause of the result. Laboratory results considered markedly abnormal and clinically significant 10 were BUN > 16 mmole/l (> 45 mg/dl), serum creatinine > 177 pmole/1 (> 2 mg/di), and proteinurea 3+. Other Safety Variables Serum assay results were screened for tobramycin concentrations 4 pg/mL from specimens collected from 10 minutes through 8 hours after completion of study 15 treatments. In parallel, patient records and CRFs were examined for evidence of systemic toxicity potentially related to elevated tobramycin levels. Assay results were not available until after patients' discharge from the study, so screening for unusually high serum tobramycin concentrations and evidence of systemic toxicity was undertaken when all pertinent results were received. 20 Pharmacokinetics Pharmacokinetic parameters for both sputum and serum tobramycin were derived to characterize aerosol delivery capabilities of control and experimental treatments. The concentration (C) versus time (t) data (Listings 16.2.5.2 and 16.2.5.3) were analyzed by model-independent methods to obtain the pharmacokinetic 25 parameters. The areas under the plasma concentration-time curve from time zero (predose) to infinity (A UC) and under the first moment of the plasma concentration time curve (A UMC) were obtained by the trapezoidal rule, extrapolated to infinity. The terminal rate constant (X,) was determined by log-linear regression of the terminal phase. The maximum concentration (Cm) and the time to maximum after 84 the end of the nebulization period (tm.) were obtained by inspection. In addition, the following parameters were calculated: t112 = ln(2)/A, CLIF=DAUC V / F = CL,, /A, S where t112 is the terminal half-life, CL/F is the total body clearance, and Vz/F is the terminal volume of distribution-i Since the absolute bioavailability of tobramycin (F) in the two formulations used in this study is not known, the calculated clearance and volume of distribution are hybrid parameters that do not account for differences in bioavailability between the two formulations. All parameters were calculated for 10 serum; only A UC, Cm., tma, A,. and to were calculated for sputum. Concentrations below the lower limit of quantitation were treated as zero for all calculations. Since there was an insufficient volume of matrix to assay tobramycin in the following time points, they were excluded from the pharmacokinetic analysis: 85 TABLE 17. EXCLUSIONS FROM PHARMACOKINETIC ANALYSIS Matrix Subject Time Serum 01-110 6 02-116 1 03-102 0.167, 1,2 03-105 0,0.167,1,2,4,6,8 03-131 0.167 05-125 4, 8 06-120 2 Sputum 08-127 2 Data Handling 5 Case report form data were entered in duplicate into a ClintrialT' database by the department of Biostatistics and Clinical Data Management (BCDM) at Chiron Corporation. Data quality assurance was performed using PUSQL and SAST" 6.12 or higher software (SAS Institute, Cary, NC). Analysis was performed by Chiron Corporation, using SAS version 6.12 or higher software, based on a predefined 10 analysis plan developed by Chiron Corporation. The estimated overall database error rate was 0.xx% with an upper 95% confidence limit of 0.xx%. This upper confidence limit is below the departmental standard of 0.5%. Statistical Methods and Determination of Sample Size Statistical and Analytical Plans: Serum and sputum pbannacokinetic 15 parameters, the incidence of bronchospasm, and the relative change from predose in 86 30-minute postdose FEVI % predicted were analyzed statistically to assess the significance of any apparent differences between test and reference treatments. All statistical tests described in following sections were two-tailed tests of significance, and the criterion for statistical significance was set at a = 0.05 unless otherwise 5 noted. Primary Aerosol Delivery Analyses: All patients who received the single dose of test or reference treatment were included in the analysis and evaluation of aerosol delivery characteristics. Aerosol delivery was characterized on the basis - of serum and sputum tobramycin 10 concentrations, derived serum and sputum pharmacokinetic parameters, and nebulization time. The effect of treatment (300 mg TOBI vs 420 mg TSI), gender, and age group (less than 18, 18 years or older) on the A UC, C,,., Xz, CUF, and z/F of tobramycin in serum, and on the AUC, C.., and X, of tobramycin in sputum, was analyzed by a three-way analysis of variance. Furthermore, the relationship between 15 body weight and A UC, C.m, CLIF, and VzIF of tobramycin in serum, and between body weight and AUC and C.. of tobramycin in sputum were analyzed by regression analysis. All tests employed a significance level a= 0.05. All parameters are expressed as the mean i SD. A harmonic half-life was estimated as: 11 /2 = ln(2) /A, 20 in which AZ is the arithmetic mean of the terminal rate constants at each dose. The standard deviation of the harmonic half-life, SD(t 11 ), was obtained as: - ln(2) SD(A,) SD(11= . 2 where SD(A.) is the standard error of the mean terminal rate constant at each dose. Safety Analyses 25 Analysis of Airway Response: The primary safety variable was the rate of bronchospasm, defined as a a 10% decrease in FEVI from predose to 30 minutes after treatment on day 1 of the study. Secondary safety variables were (a) the rate of clinically significant 87 bronchospasm, defined as a 2 20% decrease in FEV, from predose to 30 minutes after treatment on day 1, and (b) the relative change in FEVI from predose to 30 minutes after treatment on day 1. The rates of occurrence of all instances of bronchospasm (FEVI % decrease 10%) and of all instances of clinically significant 5 bronchospasm (FEVI % decrease 20%) were analyzed to assess the statistical significance of test vs. reference treatment differences using the Fisher's Exact test. The protocol specified that the treatment difference in the incidence of bronchospasm would be tested for statistical significance using the Cochran-Mantel-.Haenszel test. Due to the low incidence of bronchospasm in the enrolled patients, the Fisher's exact 10 test was used for this analysis since it makes no assumptions regarding the minimum expected cell frequencies. The test vs. reference treatment difference in mean relative change from predose in 30-minute postdose FEV % predicted was tested for statistical significance using the'two-sample t-test. Adverse Events: The total incidence of individual treatment emergent 15 adverse events (percent of patients who experienced the event at least once during or after study treatment) was evaluated descriptively for any noteworthy differences between test and reference treatments. AEs were also summarized by severity (mild, moderate, severe) and drug relationship (unrelated, possibly related) for test and reference treatments. 20 Disposition of Subiects A total of 40 patients were screened for the study by the eight investigators. Thirty-eight of the 40 screened patients met entrance criteria, were enrolled in the study (Table 18), and were randomized to one of the two treatments. Enrollment and randomization of the 38 patients at the eight sites was as summarized in Table 18 25 below: 88 TABLE 18. ENROLLMENT AND RANDOMIZATION BY SITE AND TREATMENT 300 mg TOB" 420 mg TSI PARI LC PLUST/DeVilbiss PARI LC PLUST"Ilnvacare PulmoAide"m Delivery System MOBILAIREhm Delivery System Site (no. patients (no. patients enrolled and randomized) enrolled and randomized) 01 2 2 02 1 3 03 2 6 04 0 2 05 2 4 06 .2 3 07 2 2 08. 3 2 Total enroUed 14 24 and randomized 5 Two of the 40 screened patients failed to meet entrance criteria and were not enrolled in the study: one patient did not meet the protocol inclusion criterion requiring patients to have screening FEVI % predicted results that were 2 25 %; and one patient did not meet the exclusion criterion requiring patients to have not taken inhaled or intravenous aminoglycosides within seven days before study treatment 10 administration. Thirty-eight patients -met all study entry criteria and were randomized to treatments. Thirty-seven of the 38 randomized patients received one dose of study treatment (Table 18). One patient was enrolled and randomized but was withdrawn from the study before dosing due to staff inability to establish venous access for predose day I (visit 2) blood draws. The 37 randomized and dosed 15 patients constituted the intent to treat (ITT) population. All 37 patients who received study treatments completed the study.

89 AEROSOL DELIVERY EVALUATION Data Sets Analyzed All 37 patients in the ITT population (i.e., those who were randomized and received a dose of study treatment) were evaluable for the aerosol delivery objective 5 of the protocol. Twenty four patients received a dose of 420 mg TSI using the PARI LC PLUS'/Invacare MOBILAIRE7 Delivery System, and 13 patients received a dose of 300 mg TOBI* using the PARI LC PLUSTN/DeVilbiss PulmoAideT Delivery System. Patient 08/137 was excluded from all aerosol delivery evaluations due to withdrawal from the study before dosing. 10 Demographic and Other Baseline Characteristics Demogaphic Characteristics:. Nineteen male and 18 female patients, 12 to 44 years of age and diagnosed with cystic fibrosis, constituted the ITT population. Thirty-one patients were Caucasian, four patients were Hispanic, and two patieiits were black. Gender and 15 race distributions were similar between the 420 mg TSI and 300 mg TOBI* treatment groups. On the average, IT patients in the 300 mg 1OBI* group were approximately 2.7 years older, 4.9 centimeters taller, and 10.7 kilograms heavier at screening (visit 1) than ITT patients in the 420 mg TSI group. A similar treatment difference in mean body weight was apparent before day I (visit 2) dosing, and no 20 noteworthy change in mean weight was noted between screening and day 1. Analysis of Aerosol Delivery Primary Aerosol Delivery Analysis: Examination of the mean plasma concentration-time plot for both formulations in serum (Figure 10) indicates that tobramycin is rapidly absorbed: all subjects achieved maximum concentrations in the 25 time span of 10 min to 4 h. An elimination phase was also observed in the concentration-time profiles, with individual estimates of half-life ranging from 1.1 to 6.8 h. In sputum (Figure 11), maximum concentrations were achieved between 15 rnin and 2 h, and individual estimates of half-life ranged from 0.48 to 9.47 h. These estimates are consistent with previous studies. .

90 Serum and sputum pharmacokinetic parameters are surunarized in Tables 19 and 20 as follows. TABLE 19 SERUM PFARMACOKINETIC PARAMETERS 5 (MEAN i SD) OF TOBRAMYCIN AFTER ADMINISTRATION OF 300 MG TOBI AND 420 MG TSI Parameter 300 mg TOBI 420 mg TSI A UC (pg hmL) 4.38 ± 1.97 4.41 ± 1.69 C.i (pg/mL) 0.861 ± 0.344 0.906 ± 0.542 Median :. (h) 1 (1-4) 1(0.17-2) (h-') 0.250 ± 0.052 0.243 ± 0.098 ti (h) 2.78 ± 0.58 2.86± 1.15 CL/F(L/h) 88 ±62 114±59 V/F (L) 379 ± 325 511±278 TABLE 20 SPUTUM PHARMACOKINETIC 10 PARAMETERS (MEAN ± SD) OF TOBRAMYCIN AFTER ADMINISTRATION OF 300 MG TOBI AND 420 MG TSI Parameter 300 mg TOBI 420 mg TSI AUC(pg h/g) 1521 t845 1176±686 C., (sg/g) ' 930 ±795 935 i 1040 Median I., (h) 0.25 (0.25-2)* 0.25 (0.25-0.25) 0.59 ± 031 0.52 t 0.37 Tm, (h) 170-98 -. 33 ± 0.95 The serum and sputum concentration-time curves for both treatments were 15 virtually superimposable (Figures 10 and I; Tables 19 and 20). Serum parameters (Cm., ta, A UC, CL/F, Vr/F) showed no statistically significant differences between the treatment groups (Table 19). Sputum parameters (A UC, Cm., and X) also showed no statistically significant treatment differences (Table 20). Neither age nor 91 body weight had *a statistically significant effect on these pharmacokinetic parameters. In addition, there were no statistically significant correlations between serum and sputum A UC, and between serum and sputum C,,. 'Tbe variability of the pharmacokinetic parameters in serum and in sputum was similar to previous trials. 5 In summary, these findings indicate that it is possible to achieve comparable serum and sputum levels of tobramycin to the 300 mg TOBI formulation by using the 420 mg TSI formulation. Secondary Aerosol Delivery Analyses Nebulization' Time: Nebulization time was substantially reduced during 10 administration of the test 420 mg TSI formulation below that observed during administration of the marketed 300 mg TOBI* formulation. Mean ± SD total nebulization time was 9.7 ± 3.0 minutes during 420 mg TSI administration compared to 18.1 ± 3.6 minutes during 300 mg TOBI* administration (Table 21). These findings indicate that the reduced nebulization times used in the 420 mg TSI 15 treatment did not change the pharmacokinetics of tobramycin relative to the marketed 300 mg TOBI formulation. TABLE 21 MEAN (SD) NEBULIZATION TIME 300 mg TOBI 420 mg TSI -Parameter PARI LC PLUS" PARI LC PLUSb [mean (SD)] PulmoAide Compressor MOBILAIRE Compressor (n =13) (n = 2 4 ) Nebulization Time (min) 18.1 (3.6) 9.7 (3.0) -- No. pts with data 12 23 Source: Table 14.2.2.1. Notes: a Reference treatment - TOBI 300 mg delivered by PARI LC PLUS nebulizer with PulmoAide con*ipressor. Nebulization time for patient 07/132 indeterminate due to inten-uption in dosing and unrecorded stop/start times. b Test treatment = TSI 420 mg delivered by PARI LC PLUS nebulizer with MOBILAIRE compressor. Nebulization time for patient 07/126 indeterminate due to interruption in dosing and unrecorded stop/start times.

92 Nebulizer Weight: Nebulizer weight changes from before to after dosing indicated that the test 420 mg TSI formulation delivered less product to patients than the marketed 300 mg TOBI* formulation. Mean ± SD amounts of product delivered to patients was 1.86 ± 0.53 gm during 420 mg TSI administration and 2.74 ± 1.64 gin 5 during 300 mg TOBI" administration (Table 14.2.2.2), as summarized in Table 11.4 4 below. These findings likely reflect the smaller 3.5 mL volume of TSI formulation in the nebulizer compared to the 5 mL volume of the TOBI* formulation. TABLE 22 MEAN (SD) NEBULIZER WEIGHT AND CHANGE IN WEIGHT 10 300 mg TOBIS 420 mg TSIG Parameter PARI LC PLUS PARILCPLUS [mean (SD)] PulmoAide Compressor MOBILAIRE Compressor (n = 3) (n =24) Nebulizer Weight (gm) Predose 68.25 (7.30) 69.17 (0.61) -No. patients with data 13 24 Postdose 65.51 (6.89) 67.30 (0.80) - No. patients with data 13 23 Change in weight -2.74 (1.64)' -1.86 (0.53) - No. patients with data 113 2 Notes: a Reference treatment = TOBI 300 mg delivered by PARI LC PLUS nebulizer with PuMmoAide compressor. b Test treatment = TSI 420 mng delivered by PARI LC PLUS nebulizer with MOBILAIRE compressor. Nebulizer weight for patient 02/116 not recorded after dosing. c The posttreatient nebulizer weight for patient 07/132 included the weight of the filter, and the pretreatment to posttreatment change in nebulizer weight was an increase by 2.20 g. Excluding this erroneous value yields a mean (SD) change of -3.16 gm. Discussion Aerosol delivery findings indicate that it is possible to achieve comparable serum and sputum levels of tobraiycin to the 300 mg TOBIer formulation by using 15 the 420 pg TSI formulation. Present findings also indicate that ittreduced nebulization times and reduced amount of product delivered to patients during 93 administration of the 420 mg TSI treatment did not change the pharmacokinetics of tobramycin relative to the marketed 300 mg TOBI formulation. Mean serum tobramycin concentration-time plots for both formulations indicate that tobramycin is rapidly absorbed: all subjects achieved maximum 5 concentrations in the time span of 10 min to 4 h. An elimination phase was also observed in the concentration-time profiles, with individual estimates of half-life ranging from 1.1 to 6.8 h. In sputum, maximum concentrations were achieved between 15 min and 2 h, and individual estimates of half-life ranged from 0.48 to 9.47 h. 10 The serum and sputum concentration-time curves for both treatments in the present study were virtually superimposable. Serum parameters (C., t., AUC, CL/F, Vz/F) showed no statistically significant differences between the treatment groups. Mean (± SD) serum C. results for both the 420 mg TSI and the 300 mg TOBI* groups (0.906 ± 0.542 gg/mL vs. 0.861 i 0.344 pg/mL, respectively) were 15 consistent with results from previous studies. 1'40-41 The average serum concentration of tobramycin one hour after inhalation of a single. 300 mg dose of TOBI* by CF patients was 0.95 pg/mL.

5 After 20 weeks of therapy on the TOBI* regimen, the average serum tobramycin concentration one hour after dosing was 1.05 pg/mL. Sputum parameters (A UC, C., and X) also showed no statistically 20 significant treatment. differences in the present study. Mean (± SD) sputum C. results for both the 420 mg TSI and the 300 mg TOB* groups (935 ± 1040 pg/g vs. 930 ± 795 pg/g, respectively) were consistent with results from previous studies. -4- ' Sputum results in the present study were highly variable. By comparison, high variability of tobramycin concentration in sputum was also observed in both Phase 3 25 trials.2 30 Ten minutes after inhalation Pf the first 300 mg dose of TOBI* in the Phase 3 trials, the average concentration of tobramycin in sputum was 1237 ig/g (ranging from 35 to 7414 pg/g). Tobramycin does not accumulate in sputum; after 20 weeks of therapy with the TOBI* regimen, the average concentration of tobramycin at ten minutes after inhalation was 1154 pog/g (ranging from.39. to 8085 94 pg/g). Two hours after inhalation, sputum concentrations declined to approximately 14% of the tobramycin levels measured at ten minutes after inhalation. Neither age nor body weight had a statistically significant effect on serum and sputum pharmacokinetic parameters. In addition, there were no statistically 5 significant correlations between serum and sputum A UC and between serum and sputum C., Nebulization time for the test 420 mg TSI formulation was substantially reduced below that observed during administration of the marketed 300 mg TOBI* formulation -(mean ± SD 9.7 ± 3.0 min vs. 18.1 ± 3.6 min, respectively). 10 Nebulization times for the marketed 300 mg TOBI* formulation were consistent with previous studies. 4"' Therefore, the study achieved a key benchmark with the demonstration that the alternative delivery system, consisting of 3.5 mL of a 120 mg/mL (total 420 mg tobramycin) Tobramycin Solution for Inhalation (TSI) delivered using a PARI LC PLUSTm jet- nebulizer driven by an Invacare 15 MOBILAIRE

T

" compressor, reduced nebulization time below -10 minutes on the average. Finally, present findings indicate that the reduced nebulization times during administration of the 420 mg TSI treatment did not change the pharmacokinetics of tobramycin relative to the marketed 300 mg TOBI* formulation. 20 Safety findings indicate that both a single dose of the 420 mg TSI formulation and a single dose of the marketed 300 mg TOBI* formulation were well-tolerated by patients with cystic fibrosis. The incidence of bronchospasm ( 10% relative decrease -in FEVI) was approximately 8% for each treatment (two 420 mg TSI and one 300 mg TOBI* patients); a single patient in the 300 mg TOBI@ group had 25 clinically significant bronchospasm ( 10% relative decrease in FEVI). The treatment mean relative decrease in FEVI was -3.36 5.47% for 420 mg TSI and 2.14 ± 9.62% for 300 mg TOBI*. By comparison, in the Phase IlI trials of TOBI*, ihe median change in FEV 30 minutes after the first dose of study drug had been administered was 1:8% -in the 30 tobramycin group. At Week 20, the median change in FEVI was -2.0% in the 95 tobramycin group. Because up to 95% of .CF patients have bronchodilator responsive airflow obstruction, and the within-subject variability for pulmonary function tests in CF patients has been documented to be greater than in normal patients, a 20% decrease in FEVI was considered clinically significant." Twelve 5 of 258 TOB1 patients (4.7%) had a 20% decrease in FEVI with TOBI* administration. Only two of these patients documented acute symptoms, and no patients had a 20% decrease in FEVI more than once with TOBI*. The present study also showed that the incidence of other treatment-related adverse events was very low (2 of 24 TSI patients and 1 of 13 T1B1 patients = 8%) 10 and did not differ between treatments. All three patients reported mild to moderate decreased pulmonary function test results, and one of the three patients also reported severe cough. Among all treatment-emergent AEs, events reported most frequently by 420 mg TSI patients were cough (4 patients = 17%), crepitations and sore throat (13%), and pyrexia, nasal congestion, rhinorrhoea, and sputum increased (8%). AEs 15 reported most frequently by 300 mug TOB* patients were cough (3 patients 23%) and sore throat, dyspnoea, and rhinorrhoea (15%). These events were mostly mild to moderate in intensity (two instances of severe cough), were most likely related to patients underlying cystic fibrosis and other medical conditions, and were consistent with previous large Phase 3 study results.

2 9 '3 A single patient experienced serious 20 non-drug-related symptoms (SAEs) indicative of an exacerbation of CF. None of the patients in the study were withdrawn due to AEs, and no other clinically significant findings. were noted in physical examinations, vital signs, or other safety measurements that represented an increase in risk to patients by reason of administration of study treatments. 25 Conclusions The findings of the present study indicate that it is .possible to achieve comparable serunim and sputum levels of tobramycin to the 300 mg TOBI* formulation by using 420 mg TSI formulation. Current findings also indicate that the reduced nebulization times used in the 420 mg TSI treatment did not- change the 30 pharmacokinetics of tobramycin relative to the marketed 300 mg TOBI* formulation.

96 Mean plasma concentration-time plots for both formulations in serum indicate that tobramycin is rapidly absorbed: all subjects achieved maximum concentrations in the time span of 10 min to 4. h. An elimination phase was also observed in the concentration-time profiles, with individual estimates of half-life ranging from 1.1 to 5 6.8 h. In sputum, maximum concentrations were achieved between 15 mii and 2 h, and individual estimates of half-life ranged from 0.48 to 9.47 h. Thes6 estimates are consistent with previous studies. The serum and sputum concentration-time curves for both treatments were virtually superimposable. Serum parameters (C..x rx, A UC, CL/F, Vz/F) showed 10 no statistically significant differences between the treatment groups. Sputum parameters (A UC, C.a, and XJ also showed no statistically significant treatment differences. Neither age nor body weight had a statistically significant effect on these pharmacokinetic parameters. In addition, there were no statistically significant correlations between serum and sputum A UC, and between serum and sputum C.,x. 15 During administration of the test 420 mg TSI formulation, nebulization time was substantially reduced below that observed during administration of the marketed 300 rug TOBI formulation (mean ± SD = 9.7 t 3.0 min vs. 18.1 ± 3.6 min, respectively). The apparent treatment difference in change in nebulizer weight likely reflected the different starting volumes of TSI and TOBI* formulations in the 20 nebulizer (mean ± SD = 1.86 ± 0.53 g vs. 2.74 ± 1.64 g, respectively). Aerosol delivery findings indicate that it is possible to achieve comparable serum and sputum levels of tobramycin to the 300 mg TOBIe formulation by using the 420 mg TSI formulation. Current findings also indicate that the reduced nebulization times during administration of the 420 mg TSI treatment did not change 25 the pharmacokinetics of tobramycin relative to the marketed 300 mg TOB1* formulation. While the preferred embodiments of the invention have been illustrated and described, it will be appreciated that various changes can be made therein without departing from the spirit and scope of the invention..-

Claims (20)

1. 2. The method of claim I wherein the unit dose is administered to the patient as nebulized aerosol particles having particle sizes between about I gm to about 5 gm.
2. 3. The method of claim I or 2 wherein the unit dose comprises 3.75 ml or less of the aqueous solution.
3. 4. The method according to any one of claims I to 3 wherein the unit dose comprises 3.5 ml or less of the aqueous solution.
4. 5. The method according to any one of claims I to 4, wherein the aqueous solution comprises from about 80 to about 180 mg/ml of tobramycin.
5. 6. The method of claim 5 wherein the aqueous solution comprises from about 90 to about 150 mg/ml of tobramycin.
6. 7. The method according to any one of claims I to 6, wherein the duration of nebulization is less than about 8 minutes.
7. 8. The method of claim 7, wherein the duration of nebulization is less than about 6 minutes.
8. 9. The method according to any one of claims 1 to 8, wherein the inhalation device is breath actuated.
9. 10. The method according to any one of claims I to 9, wherein the inhalation device has a rate of aerosol output of not less than about 8 il/sec. -98
10. 11. The method according to any one of claims I to 10, wherein the inhalation device releases at least about 80 percent of the unit dose.
11. 12. The method according to any one of claims 1 to 11, wherein the inhalation device releases at least about 85 percent of the unit dose.
12. 13. The method according to any one of claims I to 12, wherein the endobronchial infection is a Pseudomonas aeruginosa infection.
13. 14. A method of treatment of a cystic fibrosis patient having a Pseudomonas aeruginosa infection comprising administering to the patient a unit dose of 3.5 ml or less of an aqueous solution comprising from about 90 to 150 mg/mi of tobramycin in a physiologically acceptable carrier, wherein the unit dose is nebulized using an inhalation device having a rate of aerosol output of not less than about 5 pl/sec, that releases at least about 75% of the unit dose, and that produces aerosol particles having a particle size of about I jim to about 5 jim, wherein the duration of nebulization is less than about 6 minutes.
14. 15. A system for delivering a tobramycin formulation to a cystic fibrosis patient having an endobronchial infection, comprising a unit dose device comprising a container containing a unit dose of 4.0 ml or less of a tobramycin formulation comprising from about 60 to 200 mg/ml of tobramycin in a physiologically acceptable carrier, and an inhalation device having a rate of aerosol output of not less than 5 ptl/sec and which releases at least 75% of the unit dose for delivering the tobramycin formulation from the unit dose device to the patient in aerosolized form in less than about 10 minutes.
15. 16. The system of claim 15, wherein the endobronchial infection is a Pseudomonas aeruginosa infection.
16. 17. The system of claim 15 or 16, wherein the unit dose comprises 3.75 ml or less of the formulation.
17. 18. The system according to any one of claims 15 to 17, wherein the formulation comprises from 80 to 180 mg/ml of tobramycin. -99 19. The system according to any one of claims 15 to 18, wherein the inhalation device has a rate of aerosol output of not less than about 5 [I/sec.
18. 20. The system according to any one of claims 15 to 19 when used to deliver a tobramycin formulation to a cystic fibrosis patient having an endobronchial infection.
19. 21. The method according to any one of claims I to 14, or the system according to any one of claims 15 to 20, wherein the inhalation device is a piezoelectric oscillator inhalation device.
20. 22. The method according to any one of claims I to 14 or 21, or the system according to any one of claims 15 to 21, substantially as herein before described with reference to the accompanying Examples and/or Drawings. Dated this TWENTY NINTH day of JANUARY 2010 Novartis Vaccines and Diagnostics, Inc., By Patent Attorneys for the Applicant: F B RICE & CO