Citation
Investigation on Pulmonary Targeting of Inhaled Corticosteroids

Material Information

Title:
Investigation on Pulmonary Targeting of Inhaled Corticosteroids
Creator:
Shao, Jie
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (115 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Pharmaceutical Sciences
Pharmaceutics
Committee Chair:
Hochhaus,Guenther
Committee Co-Chair:
Bulitta,Jurgen Bernd
Committee Members:
Song,Sihong
Winner,Lawrence Herman
Graduation Date:
12/13/2019

Subjects

Subjects / Keywords:
corticosteroids -- pharmacokinetics -- pulmonary
Pharmaceutics -- Dissertations, Academic -- UF
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Pharmaceutical Sciences thesis, Ph.D.

Notes

Abstract:
Inhaled corticosteroids (ICSs) are currently prescribed as a first line treatment for asthma controlling the underlying inflammation. The goal of inhalation therapy is to achieve pulmonary targeting by attaining the desired effect without significant systemic side effects. Drug properties that are important for achieving pulmonary targeting include high systemic clearance, prolonged pulmonary residence time, and low oral bioavailability. These properties have been identified through simulation; however, pulmonary targeting of commercially available inhaled corticosteroids have not yet been directly compared in a suitable animal model. The present work using an ex-vivo receptor binding model in rats was interested in assessing pulmonary targeting for a series of commercially available corticosteroids by monitoring receptor occupancies in lung and systemic organs (spleen, brain, kidney and liver) after intravenous (IV) injection or intratracheal (IT) instillation of a dry powder at a dose of 100 microgram per kilogram. Pulmonary targeting, defined as the difference in cumulative receptor occupancies (AUCE) between lung and kidney after pulmonary delivery, differed across the investigated corticosteroids with the highest degree of pulmonary targeting found for corticosteroids with high systemic clearance and pronounced lipophilicity (presumably allowing a long pulmonary residence time). In addition, this study demonstrated differences in the receptor occupancies across systemic organs. Using kidney receptor occupancies as the comparator, liver receptor occupancies were reduced after IV and IT administration for corticosteroids with high intrinsic clearance, while they were increased for corticosteroid prodrugs due to hepatic activation. Spleen receptor occupancies were increased after intratracheal, but not after IV administration. This was especially true for slowly dissolving drugs. Reduced brain uptake was also observed for ciclesonide and des-ciclesonide, two compounds previously not investigated. To further investigate pulmonary targeting within physiologically condition, we develop a physiologically based PK/PD model to link drug properties with its plasma and tissue concentrations, and receptor occupancies as pharmacodynamic marker. Both beclomethasone dipropionate (BDP) and its active metabolite, beclomethasone 17-monopropionate (BMP) were given to rats via intravenous (IV) injections of a solution and intratracheal (IT) instillation of a dry powder. A physiologically based PK/PD model was developed to describe plasma concentrations of BDP and BMP, tissue concentrations of BMP (lung, liver, and kidney), and receptor occupancies in tissues (lung, liver, and kidney) after IV and IT administration in rats. After IV administration of BDP, 95.4% of BDP was converted to BMP. After pulmonary delivery of BDP, 51.2% of BDP was absorbed directly and 46.6% of inhaled BDP was absorbed as BMP. Based on the parameter estimation using the proposed model, BDP and BMP dissolved in epithelium lining fluid at rates of 0.47/h and 3.01/h, respectively. The permeabilities in central lung were estimated as 15.0 and 2.92 * 106 cm/s for BDP and BMP, respectively. Subsequently, the permeabilities in the peripheral lung were calculated as 509 and 98.9 * 106 cm/s for BDP and BMP, respectively, by adjusting the permeabilities in the central region for differences in the epithelium thicknesses. An Emax model was used to link free BMP concentrations with receptor occupancies in the tissues. Emax and EC50 were estimated as 85.5% and 0.0017 ng/mL, respectively. The PBPK/PD model was able to describe the PK/PD data sufficiently well as the predicted plasma concentrations and tissue receptor occupancies agreed reasonably well with the observations. The present PK/PD analysis has shown that BDP is targeting the lung. Simulations based on the developed PBPK model demonstrated that low dissolution rate and permeability can improve pulmonary targeting. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ph.D.)--University of Florida, 2019.
Local:
Adviser: Hochhaus,Guenther.
Local:
Co-adviser: Bulitta,Jurgen Bernd.
Statement of Responsibility:
by Jie Shao.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
LD1780 2019 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

INVESTIGATION ON PULMONARY TARGETING OF INHALED CORTICOSTEROIDS By JIE SHAO A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2019

PAGE 2

© 2019 Jie Shao

PAGE 3

To my family

PAGE 4

4 ACKNOWLEDGMENTS I would like to extend my appreciation to my researc h advisor Dr. G ü nther Hochhaus for accepting me as his stude nt , giving me the opportunity to work on this project , and guiding me through the five years. I would like to thank the members of my supervisory committee, Dr . J ü r gen Bullita, Dr. Sihon g Song, and Dr. Laurence Winner, for their valuable and kind advice th roughout my Ph.D. study. I would like to express my gratitude to Yufei Tang for her technical support of analytical method development. I am grateful to Dr. M ongjen Chen and all members in my lab, Sharvari Bhagwat , Uta Schilling, Stefanie Drescher, Abhina v Kurumaddali , Elham Amini, and Simon Berger, for their inspiration discussion. I would like to thank the professors, staff, and graduate students in Pharmaceutics and the College of Pharmacy. Special thanks to Dr. Guohua An, Dr. Anthony Palmieri , Dr. Hart mut Derendorf , Dr. Lawrence Lesko, Dr . Stephan Schmidt, Dr. William Mobley, Dr. Sihem Bihorel , Vivian Lantow, Kimberly Mahoney, Leslie E. McKenna , Ann Ross, Milena Ozimek, Ye Yuan, Ahmed Elshikha, Yichao Yu, Sibo Jiang, Xun Tao, Y u li Qian, Yiting Lien, Bre tt Fleisher . Special thanks to Dr. Er ica Levitt, in the Department of Pharmaco logy & Therapeutics, and Dr. Huiping Yang, in School of Forest Resources & Conservation, for providing the osmometer for me. Finally, many thanks to my family members, my parent s, m y parents in law, my husband, and my kids, for their love and support . This work would not have been possible without them.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ .......... 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 11 ABSTRACT ................................ ................................ ................................ ................... 14 CHA PTER 1 INTRODUCTION ................................ ................................ ................................ .... 17 Asthma ................................ ................................ ................................ .................... 17 PK/PD Activity after Inhalatio n of Corticosteroids ................................ ................... 19 Lung Anatomy and Physiology ................................ ................................ ......... 19 Pharmacokinetic Processes after Inhalation ................................ ..................... 21 Deposition ................................ ................................ ................................ .. 21 Dissolution ................................ ................................ ................................ . 22 Mucociliary clearance and macrophage clearance ................................ .... 23 Absorption ................................ ................................ ................................ .. 23 Pharmacodynamic of Corticosteroids ................................ ............................... 24 PK/PD Factors Important for Pulmonary Targeting ................................ .......... 25 P ulmonary residence time ................................ ................................ .......... 25 Systemic clearance ................................ ................................ .................... 27 Oral bioavailability ................................ ................................ ...................... 27 Other factors ................................ ................................ .............................. 28 Objectives ................................ ................................ ................................ ............... 28 2 EVALUATING PULMONARY TARGEING OF INHALED CORTICOSEROIDS USING NONCOMPARTMENTAL ANALYSIS ................................ ......................... 34 Background ................................ ................................ ................................ ............. 34 Method ................................ ................................ ................................ .................... 35 Materials ................................ ................................ ................................ ........... 35 Animal Procedures ................................ ................................ ........................... 35 Ex Vivo Receptor Binding Assay ................................ ................................ ...... 37 FP Stability in Liver Homogenate ................................ ................................ ..... 38 Data Analysis ................................ ................................ ................................ ... 38 Results ................................ ................................ ................................ .................... 40 Receptor Occupancy in Systemically Exposed Organs ................................ .... 40 Pulmonary Targeting after IT Administration ................................ .................... 41 Relationship Between Drug Properties and AUC E ................................ ......... 41

PAGE 6

6 FP Stability Test ................................ ................................ ............................... 42 Discussion ................................ ................................ ................................ .............. 42 Summary ................................ ................................ ................................ ................ 48 3 PHYSIOLOGICALLY BASED PK/PD MODEL TO ASSESS PULMONARY TARGETING OF BECLOMETHASONE DIPROPIONATE AND ITS ACTIVE METABOLITE ................................ ................................ ................................ ......... 57 Backgrou nd ................................ ................................ ................................ ............. 57 Methods ................................ ................................ ................................ .................. 58 Chemicals ................................ ................................ ................................ ......... 58 Animal Procedures ................................ ................................ ........................... 58 Sample Preparation and LC/MS Analysis ................................ ......................... 60 Ex Vivo Receptor Occ upancy Assay ................................ ................................ 61 Noncompartmental Analysis ................................ ................................ ............. 63 PBPK Model Development ................................ ................................ ............... 63 Global PBPK model structure ................................ ................................ .... 63 Sub compa rtmental model of lun g ................................ ............................. 65 Dissolution after pulmonary delivery ................................ .......................... 65 Pulmonary absorption ................................ ................................ ................ 66 E max Model for Receptor Binding ................................ ................................ ...... 6 8 Sensitivity Analysis ................................ ................................ ........................... 69 Model Validation ................................ ................................ ............................... 70 Assessment of Pulmonary Targeting ................................ ................................ 70 Results ................................ ................................ ................................ .................... 71 PK/PD Data for BDP and BMP in Rats ................................ ............................. 71 PBPK/PD Model in Rats ................................ ................................ ................... 71 Sensitivity Analysis ................................ ................................ ........................... 72 Model Verification ................................ ................................ ............................. 72 Evaluation of Pulmonary Targeting ................................ ................................ .. 73 Discussion ................................ ................................ ................................ .............. 73 Summary ................................ ................................ ................................ ................ 79 4 CONCLUSION ................................ ................................ ................................ ........ 92 APPENDIX A TABLES AND FIGURES FOR RECETPR OCCUPANCY ANALY SIS .................... 93 Receptor Occup ancy after Intravenous (IV) Administration of Tested Inhaled Corticosteroids in Rats. The Mean ± SD Is Given. ................................ .............. 93 Receptor Occupancy after Intratracheal (IT) Administration of Tested Inhaled Cort icosteroids in Rats. The Mean ± SD Is Given. ................................ .............. 94 Receptor Occupancy Time Profiles for BDP after Systemic (IV) and Pulmonary (IT) Administration. The Mean ± SD Is Given. ................................ ... 95 Receptor Occupancy Time Profiles for BMP after Systemic (IV) and Pulmonary (IT) Administration. The Mean ± SD Is Given. ................................ ... 96

PAGE 7

7 Receptor Occupancy Time Pr ofiles for BUD after Systemic (IV) and Pulmonary (IT) Administration. The Mean ± SD Is Given. ................................ ... 97 Receptor Occupancy Time Profiles for CIC after Systemic (IV) and Pulmonary (IT) Administration. The Mean ± SD Is Given. ................................ ..................... 98 Receptor Occupancy Time Profiles for desCIC after Systemic (IV) and Pulmonary (IT) Administration. The Mean ± SD Is Given. ................................ ... 99 Receptor Occupancy Time Profiles for TA after Systemic (IV) and Pulmonary (IT) Administration. The Mean ± SD Is Given. ................................ ................... 100 B MONOLIX CODE FOR PBPK MODEL AFTER IT ADMINI STRATION OF BDP .. 101 LIST OF REFERENCES ................................ ................................ ............................. 106 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 115

PAGE 8

8 LIST OF TABLES Table page 1 1 Phy siologic al values of human lung ................................ ................................ .... 30 1 2 Physiological values of rat lung ................................ ................................ .......... 31 2 1 E , mean ± SD) between non kidney organs (lung, spleen, liver, and brain) and kidne y after IV administration of seven ICSs in rats. ................................ ................................ .. 49 2 2 Differences in E , mean ± SD) between non kidney organs (lung, spleen, liver, and brain) and kidney after IT administration of the seven ICSs in rats. ................................ ............................ 50 2 3 Summa ry of biopharmaceutic al and PK/PD properties of tested corticosteroids in rats. ................................ ................................ ......................... 51 3 1 Physiological parameters for modeling in rats ................................ .................... 81 3 2 Tissue plasma partition coefficients for tissue s in the model .............................. 82 3 3 The estimates of PK parameters for BDP and BMP in rats using noncompartmental analysis ................................ ................................ ................ 83 3 4 The estimates of PK/PD parameters for BDP and BMP in rats using the proposed model ................................ ................................ ................................ .. 84

PAGE 9

9 LIST OF FIGURES Figure page 1 1 A schematic o f airwa y branching in the human lung ................................ ........... 32 1 2 Fate of inhaled corticosteroids after inhalation ................................ ................... 33 2 1 Receptor oc cupied time profile s for FP after systemic (IV) and pulmonary (I T) administration. . ................................ ................................ ............................. 52 2 2 Receptor occupancies of lung, spleen, liver and brain comparing with kidney receptor occupancies afte r IV administration o f tested compounds in rats. ........ 53 2 3 Receptor occupancies of lung, spleen, liver and brain comparing with kidney receptor occupancies after IT administration of tested compoun ds in rats. ........ 54 2 4 E in non pulmonary organs in rats. ................................ ................................ ................................ ................. 55 2 5 Pulmonary targeting of tes ted compounds and its relationship with drug properties for seven corticosteroids in rats. . ................................ ....................... 56 3 1 Schematic representation of a physiologically based pharmacokinetic model for BDP and BMP after intravenous (IV) or intratracheal (IT) administration (A), compartmental representation of lung disposition (B), and PD model to describe the relationship between recept or occupancy and free drug concentration in tissues (C). ................................ ................................ ............... 85 3 2 BDP and BMP concentration in plasma , BMP concentration in tissues , and receptor occupancy in tis sues after IV administration of BDP , IT administration of BDP , IV administration of BMP , and IT adm inistration of BMP . ................................ ................................ ................................ ................... 86 3 3 Observed and predicted concentrations and receptor occupancies in plasma and various tissues in rats. ................................ ................................ ................. 87 3 4 Sensitivity anal ysis. The total effec t on lung receptor occupancy (A) and kidney receptor occupancy (B) with ten different parameters given as a proportion of variance when IT administration of BDP was given. ...................... 88 3 5 Observed and PB PK model pred icted plasma concentrations of BDP and BMP after IV and IT administration of BDP in rats. ................................ ............. 89 3 6 Observed and PBPK/PD model predicted receptor occupancies after IV administration of BDP , IT administration of BDP , and IT adm inistration of BMP . ................................ ................................ ................................ .................. 90

PAGE 10

10 3 7 Simulation of receptor occupancy in the lun g, and kidney with different dissolution rate and permeabil ity after IT of BDP i n rats. ................................ .... 91

PAGE 11

11 LIST OF ABBREVIATIONS AIC AM alveolar macrophages AUC a re a un der the curve AUC E accumulative receptor occupancy BDP beclomethasone dipropionate BIC Baye sian information crit erion BMP 17 beclomethasone monopropionate BUD budesonide CBP CREB binding protein CIC ciclesonide CL systemic clearance CL i nt h epatic intrinsic clearance DCs dendritic cells desCIC des Ciclesonide EDTA ethylenediaminet etraacetic acid DPI d ry powder inhaler eFAST extended Fourier amplitude sensitivity test ELF epithelial lining fluid ESI electrospray ionization FP f luti casone propionate FPF f ine particle fraction GM CSF granulocyte macrophage colony stimulation factors

PAGE 12

12 GR glucoco rticoid receptors GREs glucocorticoid responsive elements HPLC high performance liquid chromatography ICSs i nhaled corticosteroids I gE i mmunoglobulin E IP intraperitoneal IT intratracheal IV intravenous LC MS/MS liquid chr omatography tandem ma ss spectroscopy LLOQ lower limits of quantification logP log octanol water partition coefficients MMAD mass median aerodynamic di amet er MRM multiple reactions monitoring NCA noncompartmental analysis NS non specific binding OIDPs o rally inhaled drug products PBPK physiologically based PK PK/PD p harmacokinetic and pharmacodynamics PMSF phenylmethylsulphonyl fluoride P gp P glyc oprotein RRA relative receptor binding affinity RSE r elative standard errors Sol solubilit y in water

PAGE 13

13 TA triam cinolone acetonide TB total binding Th2 T helper 2 cells

PAGE 14

14 Abstract of Dissertation Presented to the Graduate School of the Uni vers ity of Florida in Partial Fulfillment of the Requirements for the Degree of Doct or of Philosophy I N V ESTIGATION ON PULMONARY TARGETING OF INHALED CORTICOSTEROIDS By Jie Shao December 2019 Chair: G ü nther Hochhaus Major: Phar maceutic al Sci ences Inhaled corti costeroids (ICSs) are currently prescribed as a first line treatment for ast hma controlling the u nderlying inflammation . The goal of inhalation therapy is to achieve pulmonary targeting by attaining the desired effect wit hout signi fica nt systemic side effects . Drug properties that are important for achieving pulmonary targeting in clude high systemic c learance, prolonged pulmonary residence time, and low oral bioavailability. These properties have been identified through simulation; howe ver, pulmonary targeting of commercially available inhaled corticosteroids have not yet been dire ctly compared in a su itable animal model. The present work using an ex vivo receptor binding model in rats was interested in assessing pulmonary targeting for a series of commercially available corticosteroid s by monitoring receptor occupancies in lung an d systemic organs (sp leen, brain, kidney and liver) after intravenous (IV) injection or intratracheal (IT) instillation of a dry powder at a dose of 100 µg /kg. Pulmonary targeting , defined as the difference in cumulative receptor occupancies (AUC E ) between lung and kidney afte r pulmonary delivery, E range: 33 ± 46 to 143 ± 52 % *h) wit h the highest degree of pulmonary targeting found for corticosteroids with high

PAGE 15

15 systemic clearance and pronounced lipophilicity (p resumably al lowing a long pulmonary residence time). In addition, this study demonstrated differences in the receptor occupanc ies across systemic organs. Using kidney receptor occupancies as the E ra nge: 124 ± 59 to 122 ± 51 % *h) after IV and IT administration for corticosteroids with high intrinsic clearance, while they were increased for corticosteroid prodrugs due to hepatic activation. Spleen receptor occupancies were increased after intratrache E ra nge: 33 ± 35 to 105 ± 47 % *h), but not after IV administration. This was especially true for slowly dissolving dr ugs. Reduced brain uptake was also observed for ciclesonide and des c iclesonide , two compounds previously not investigated. To further inv estigate pulmonary targeting within physiologically condition, we develop a physiologically based PK/PD model to l ink drug properties with its plasma and tissue concentrations, and receptor occupancies as pharmacodynamic marker . Both beclometha sone dipropi onate (BDP) and its active metabolite, beclomethasone 17 monopropionate (BMP) were given to rats via intravenous ( IV) injections of a solution and intratracheal (IT) instillation of a dry powder. A physiologically based PK/PD model was develope d to describ e plasma concentrations of BDP and BMP, tissue concentrations of BMP (lung, liver, and kidney), and receptor occup ancies in tissues (lung, liver, and kidney) after IV and IT administration in rats. After IV administration of BDP, 95.4% of BDP w as converted to BMP. After pulmonary delivery of BDP, 51.2% of BDP was absorbed directly and 46.6% of inhaled BDP was absorbed as BMP. Based on the parameter estimation using the proposed model, BDP and BMP dissolved in epithelium lining fluid at rates of 0.47/h and 3 .01/h, respectively. The permeabilities in

PAGE 16

16 central lung were estimated as 15.0 and 2.92 × 10 6 cm/s for BDP and BMP, respectively. Subsequently , the permeabilit ies in the peripheral lung were calculated as 509 and 98.9 × 10 6 cm/s for BDP and BMP , respective ly, by adjusting the permeabilit ies in the central region for differences in the epithelium t hicknesses. An E max model was used to link free BMP concentrations with receptor occupancies in the tissues. E max and EC 50 were estimated as 85.5% and 0.0017 ng/mL , respectively. The PBPK/PD model was able to describe the PK/PD data sufficiently well as th e predicted plasma concentrations and tissue receptor occupancies agreed reasonably well with the observations. The present PK/PD analysis has shown th at BDP is ta rgeting the lung . Simulations based on the developed PBPK model demonstrate d that low dissolu tion rate and permeability can improve pulmonary targeting. Thus, our model can help to assess pulmonary targeting of an inhaled corticosteroid prodrug , and furthe r facilitate drug development.

PAGE 17

17 CHAPTER 1 INTR O DUCTIO N Asthma is a chronic inflammatory disease of the airways. Inhaled corticosteroids have been used as the first line medication to suppress the inflammation in the asthmatic airways for years (1) . T h rough the inhalation route glucocorticoids are directly delivere d to the target organ to obtain a high drug concentration in the lung while limiting the concentrations in the system. The goal of inhaled corticosteroids is to achieve pulmonary targeting by producing a high pulmonary effect while minimizing systemic side effects . To clearly understand the concept of pulmonary targeting of inhaled corticosteroids for asthma therapy, the PK/PD factors that affect pulmonary targeting must be discussed. In this chapter, w e will first review the curr ent literature on asthma pathology and the role of corticosteroids in the treatment of asthma . Then, the fate of inhaled corticosteroids after inhalation is studied. Finally, the PK/PD factors that are relevant to pul m onary targ eting are discussed . Asthma According to the report from the Global Asthma N etwork, asthma kills about one thousand people every day and affects more than 300 million people worldwide (2) . In US, 1 in 13 people have asthma and more than 25 million p eople w ere diagnosed to have asthma (3) . Asthma is considered as a socioeconomic issue and a global health burden. Asthma is a chronic inflammation airway disorder characterized by airflow obs truction and airway hyperrespon siveness (4) . Symptoms that induced by t he inflamed airway includ e wheezing, coughing, chest tightness, and shortness of breath

PAGE 18

18 (ref) . T hese symptoms can be mild or severe due to the disease stage. And so far, it is not clear how to prevent asthma from developing and there is no cure. T he fact that inflammation plays a central role in the pathology of asthma has been known for years, however, t he relationship between the inflammatory process and the airw ay obstruc tion remains unclear (5) . The airway inflammation involves an interaction of many cell types and multiple mediator s with in the airways. Inflammatory cells include lymphocytes, mast cell s, eosinophils, neutrophils, dendritic cells, macrophages, and epithelial cells . The inflammatory mediators in clude chemokines, cytokines, cysteinyl leukotrienes, nitric oxide, and immu noglobulin E (IgE) . The inflammatory cells and mediators inte rvened eac h other to induce the inflammation and exacerbate the s ymptoms. One generally acknowledged signaling pathway for the inflammation in asthma involves the activation of T helper 2 cells ( Th2), which is the subpopulations of lymphocytes (6) . It has been sho wn that de ndritic cells are activated by allergens from the airway surface and migrate to lymph nodes and stimulate Th2 cells (7) . And Th2 cells initiate inflammation by releasing Th2 affiliated cytokines such as IL5, IL4, and IL3 (8) . Cytokine IL5 is need ed for eos inophil differentiation and survival (ref) , IL4 is important for Th2 cell differentiation and IL3 is important for I gE formation. S ome chemokines can also recruit Th2 cells, such as thymus and activation regulated chemokines (TARCs) and macrophag e derived chemokines (MDCs). In addition, the constantly active Th2 cells are also found in the asthmatic airways. Thus, asthm a is considered as a disease driven by Th2 (9) .

PAGE 19

19 In addition to the extensive evaluation of inflammatory ce lls and me diators, there is increasing recognition that airway remodeling may contribute to asthma progression . The c onsistent inflammation in the airway can cause the thickening of the airway wall (10) , as a result, the airway is narrow ed. And this narrowed a irway can further exacerbate the i nflammation in the airway (11) . Me dications for the treatment of asthma depending on the age and disease stage. Corticosteroids are the cornerstone for the long term asthma control because of their anti inflammatory effects . And c orticosteroids given through inhalation is preferred, becaus e lower do se s can be used than those necessary after systemic administ ration , thereby, system ic side effects could be reduced. PK/PD A ctivity after I nhalation of C orticosteroids The fate of inhaled corticosteroids is determined by drug formulation propert ies and th e physiological condition of the lung . The interplay between these two aspects play an important role on pharmacokinetic and pharmacodynamic of the compound. Lung A natomy and P hysiology T he lungs are a complex organ aimed to exchange gases betwee n the bloo d stream and the respiratory airspace. The complete cardiac output flow ing through the alveolar vascular region help s the oxygen change efficiently (12) . To perform the function of efficient oxygen exchange and protection from exogenous compounds, t he airway branching has shown a repeated bifurcate pattern up to 23 divisi ons from trachea to alveoli r egion ( Figure 1 1 ) . Along w ith the tracheobronchial tree , the diameter of the bronchi is reduced , length is shortened , but the surface area is increase d (13) . To simplify the lung structure, the lung can be considered into two district zones : conducting zone (0 16 generations) and respiratory zone (17 23 generations). The

PAGE 20

20 conducting zone contains trachea, bronchi, and bron chioles. A nd t he respiratory zone contains respiratory bronchioles, and alveol ar ducts and sacs . In addition to gas exchange, molecul es can be absorbed from the lu ng into the blood stream . The molecules must pass across a series of barriers in the following order: su rfactant, surface lining fluid, epithelium, interstitium and basement membrane (14) . Surfactant is covering b oth airway and alveolar surface with a thin layer. In contrast, the amount of epithelial sur face fluid s between airway and alveolar region is quite differen t . There is a relatively thick mucus layer covering the airway region, in which cil ium are present to move particles and pathogens constantly towards the trachea . Instead of the thick mucus la yer, there is a thin fluid layer covering the alveolar region, in which there is no pushed cilia (15) . Therefore, the thickness of lining fluid be tween these two regions is different (Table 1 1). The differences in the thickness and the constitution can affect drug dissolution. But the overall l ining flui d is either isotonic or slightly hypotonic and slightly acidic (pH 6.9) relative to plasma (16) . Not only the epithelium lung fluid, t he epithelium layers are also quite different between these two regions . The main types of cells in the airways are basal cell, the ciliated c ells, the goblet cell and the Clara cell, while the types that in the alveolar regions are Type I and Type II cells . These distinct differences in the compositi on and thickness could play a very important role in drug permeability. The fourth potential bar rier for absorption that mentioned above is the interstitium, which is made by the extracellular and extravascular space between cells in the tissue. We can als o assume the structure is different between the airway and alveolar regions, but the clear influence on a bsorption through interstitium need to be further studied.

PAGE 21

21 Because of the complexity of lung anatomy, there are a lot of unknowns about pulm onary disp osition. Thus, preclinica l studies in animals, such as rats and mice, are very useful. Knowing of the differences between rodent and human is critical to interpret the results. One of the main differences is the branching pattern, human lung is n ot perfect ly symmetrical, but is ir regularly dichotomous. In rat and mouse, the branching pattern has shown as a monopodial structure (17) . In addition, t he physiol ogical parameters in diff erent regions are different ( Table 1 2). Pharmacokinetic P rocesses after I nhalation Upon the inhalation of a glucocorticoid, depending on the deposition efficiency, 10% to 60% of the inhaled dose is deposited in the lung . The rema inder of the dose is depo sited in the mouth and throat region, swallowed and subject to oral absorption (Fig ure 1 2 adapted from (18) ). The drug particles deposited in the lung will dissolve with a rate depending on its physicochemical properties and formulation factors. Undissolved drug is subject to mucociliary removal from the upper parts of the lu ng, swallo wed and potentially absor bed from the GI tract. Only the dissolved drug interacts with the glucocorticoid receptors present in the lung to produce the anti inflammatory effects and further absorbed into the systemic circulation. The swallowed dos e is avail able for gastrointestinal absorption and is absorbed into the systemic circulation after undergoing first pass inactivation by the liver. The free drug of the corticosteroid in the systemic circulation binds to the systemic glucocorticoid recepto rs to prod uce the systemic side eff ects. Deposition Deposition is the first process after inhalation. Drug particles leave the device, travel through the airway and deposited in the respiratory system. Deposited regions

PAGE 22

22 can be roughly characterized as pha rynx thro at region, conducting airway and alveolar regions (19) . Factors that affect particle deposition include device factors (e . g. devi ce deposit ion efficie ncy), formulation factors (e . g. particle size distribution) and patient factors (e . g. inhalation flow rate and disease status ) (20,21) . One essential propert y that determines the deposition pattern is the particle size distribution (22) characte rized by mass median aerodynamic diameter (MMAD) . Drug particles with MMAD < 5µm can be deposited into the deep lung with high possibility . Particle deposition ba sically follow s t hree mechanism s : inertial impaction, gravit ational sedimentation, and diffus ion or Brow nian motion (23) . Gravity and diffusion play important role for small particles depositin g into the deep lung (24) . Because of this, holding breath after inhalation can help the sedimentation for small particles. ( Small particles are affected more by the sedimentation by gravity and raised when hold ing the breath after inhalation. Diffusion is also relevant. Both mechanisms affect the s mall particles and deposited in the alveolar region ). Dissolution After deposition, the particles land on the lining fluid of the surface and must dissolve to be further absorbed. Factors that affect dissoluti on rate include drug physicochemical properties and subject physiology (25) . As mentioned above , the thickness and fluid comp onent are different between central and peripheral regions, therefore, the drug dissolution rate could be different across regions. Slow dissolution rate is favorable because it increases the lung residence time and ef fect duration.

PAGE 23

23 Mucociliary clearance and macro phage clearance After deposition, particles that deposited in the central region of the lung are subject to muc o cocili ary clearance while in deep lung are subject to alveolar macrophage which engulf the particles by phytocytosis . As mentioned i n lung ana tomy, thick mucus layer is covering the airway region . As a protection mechanism, cilia are present to beat the pathogens and particles towa rds the phary nx (26) . It has been shown that mucociliary clearance increases when airway gets wider, thus, drug particles that deposited in the upper airways are cleared more quickly (27) . Since the mucus layer is thicker in asthmatic patient airways, the mucociliary clearance could be decreased (28) . In the peripheral region of the lung, particle s that not readily dissolved could be clear ed by alveolar macrophages . Macrophages engulf the particles and transfer to the lymph n odes (29,30) . Because corticosteroids are considered to diffuse freely across membranes due to their lipophilicity, and m acro phage clearance much slower comparing with mucociliary clearance, it is assumed to be negligible for inhal ed drugs (31) . Absorption Pulmonary absorption of dissolv ed drug is considered as a passive process that is driven by a concentration gradient with rates dependent on pulmonary anatomical and physiological parameters (e . g. surface area, thickness and volume of mucus and pulmonary cells) and drug dependent parame ters (e . g. permeability) for the adjacent compartm ent in the lung. The dissolved drug starts to pass across membrane barriers from epithelium to the cells, eventually enters blood stream. Compounds like corticosteroids are

PAGE 24

24 consider ed to diffuse across me mbranes ra pidly because of their l ipophili city. So concentration gradient is the driven force for passive (14) . For most long acting beta agonist bronchodilators, they are thought to be absorbed through paracellular because of their hydrophilicity. They pass across the epithelium via the aqueous pores in intercellular gap junctions (32) . Physiologic conditions of the pulmonary environment would absorption. As mentioned above , the thickness and surface area of epithelium between conducting region and alveola r region are very different (Figure 1 1) . Therefore, the absorption rates could be different across sections. These include parameters such as surface area of defined regions of the lung, pulmonary distribution volumes (lung lining fluid volume, aqueous volume of d efined section s of the lung intracellular space, airway thickness) for which variable values have been reported. The v ariability in these parameters will significantly affect prediction of the dissolution rate and permeability estimates and, consequently, the resulting free drug concentrations in the lung. Eventually , the drug gets absorbed into the blood circulation heavily rel ies on perfusion. The lung is the highest perfused organ in the body , and the cardiac output perfused in the alveolar region . Phar macodynamic of C orticosteroids Inhaled corticosteroids (ICSs) have been used as first line medication to suppress the inflammation in the asthmatic airways. They suppress inflammation mainly by activating anti inflammatory genes, suppressing multiple activ ated inflammatory gene expressi on, inhibiting inflammatory cells, after binding to glucocorticoid receptors (GR) (33) .

PAGE 25

25 GRs are predominantly expressed in the cytoplasm of almost all t ype s of cells. There are two su bgroups of GRs, GR and . GR binds to corticosteroids to produce anti inflammatory effect, while the interaction with are reported to be involved in steroid resistance in asthma treatment. Prior to binding corticoster oids, GRs are present as a mult i protein complex which is mainly associated with heat shock proteins (hsp90) (34) . Following activation by a ligand, a corticostero id, the receptor undergoes a se ries of conformational alterations, leading to its dissociation from the cytoplasmic chaperons and translocated into nucleus (35,36) . The receptor ligand complex forms receptor homodimers and binds to glucocorticoid responsive elements (GREs) in the promoter region of the target gene, which induc e anti inflammatory protein expression. This process is called transactivation. GR homodimers can also interact with negative GREs to suppress genes, particularly those linked to side effects. This process is called transrepr ession . Nuclear GR also interac t with coactivator molecules, such as CREB binding protein (CBP), which activated by proinflammatory transcription factors. The major action of corticosteroids is to suppress multiple activated inflammatory genes that encode for cytokines, chemokines, infl ammatory enzymes and receptors. As a result, reduce inflammation in the asthmatic airways. PK/PD F actors I mportant for P ulmonary T argeting Pulmonary residence time As discussed above, particles undergo several pharmacokineti c processes after inhalation . T he overall pulmonary residence time of a molecule can be characterized by the time required for dissolution and the time required for passing across membranes.

PAGE 26

26 The l ung is a highly perfused organ and studies have shown that the absorption tim e for drug solution is very short (37) . In order to prolong the time that drug stays in the lung, drug properties that affect dissolution rates are desired . Factors tha t can affect disso lution rate include physicochemical properties (e.g. l ipophilicity) and formulation factors (e.g. surface area). Although diffusion is very fast once drug dissolved , permeability could prolong pulmonary retention for some hydrophilic comp ounds. F actors tha t can affect absorption of a dissolved molecule include drug properties (e.g. protein binding) and physiological environment (e.g. partition coefficient, membrane thickness, and membrane surface area). As we have known that, dissolution is a rate limiting step for lipophilic compounds, such as corticosteroids (38) . Simulations have shown that an optimal dissolution rate can increase pulmonary targeting because slowly dissolved drug can elevate drug concentration in the lung over systemic circulation for a period of time. As a result, the more d rug can bind to glucocorticoid receptors to elicit anti inflammatory effect in the lung. However, if dissolution is too slow , drug deposited in the centr al region of the lung can be cleared via mucocil iary clearance. As a result, the cleared drug removed b y mucociliary cl earance gets swallowed and absorbed into systemic circulation . Thus, the systemic side effects increase. Therefore, there is an optimal dissolution rate for a given corticosteroid. In addition, prodrug strategy has the potential to improve drug retention in the lung (39) . Prodru g , such as ciclesonide, can reduce the local side effects on the throat after in halation and increase pulmonary effect, if complete conversion to des CIC is

PAGE 27

27 achieved. Other strateg ies , such as ester ification , intend to enable the compound stay in the lung for a relative ly long time (40) . Overall, pulmonary residence time is determined by the interplay between dissolution, mucociliary clearance and permeability (18) . Both drug properties and physiological environment play important roles in this process. Systemic clearance Once the drug enters systemic cir culation, high s ystemic clearance can remove the drug from the body as efficient ly as possible, therefore, high systemic clearance can improve pulmonary targeting by decreasing receptor occupancy in the kidney (41) . Results from our analysis using experimental data have shown that corticosteroids with high systemic clearance tend t o have high pulmonary targeting. Compounds, such as FP and des CIC, which have a higher degree of pulmonary targeting, possesses favorable drug properties (e.g. high systemic clearance). Oral bioavailability Depending on the device deposit ion efficiency, a significant portion of the inhaled dose is deposited in the pharynx region and swallowed. Eventually get absorbed into systemic circulation. The first pass effect in the liver is the process th at determines how much drug eventually enters sys temic circulation after absorption from GI tract. Simulations have shown that compounds with low oral bioavailability will have high pulmonary targeting (41) . This is consistent with our analysis in the next chapter that corticosteroids with high intrinsic clearance tend to have less hepatic receptor occupancy.

PAGE 28

28 Othe r factors Depending on the deposition efficiency, part of the inhaled dose is directly deposited into the lung and the remaining are going to absorb from GI tract. High deposition efficiency determines how much dose initially deposited in the lung. Fac tors that affect deposition efficiency include particle size distribution, velocity of the droplets and particle density (42) . Inhalation device s with higher pulmonary deposition will improve pulmonary targeting. This is because the more efficient device can increase the amount of drug in the lung, meanwhile reduc ing the amount of drug that absorbed into GI tract . Boger et al (43) indicated slow dissociation of the receptor complex can improve pulmonary targeting. This might be true in the case of drug with high pulmonary improving receptor binding affinity can increase drug effect in the lung as well as in the system. As a result, pulmonary targeting is not impr oved . Objectives The overall goal of this work is to assess pulmonary targeting of inhaled corticosteroids in rats and provide quantitative understanding of PK/PD relationship using a physiologically based model. The first objective is to compare the pu lmonary targeting of a series of commercially available corticosteroids in rats using an ex vivo receptor binding assay. Further, pulmonary targeting of these inhaled corticosteroids was quantified and correlated with their PK/PD properties. In add ition, a physiologically based model with lung disposition was developed for beclomethasone dipropionate (BDP) and its active metabolite in rats. The model

PAGE 29

29 integrated drug properties with physiological parameters. Thus, simulations that were performed in a physio logical relevant condition can provide an overall understanding of the relationship between drug properties and pulmonary targeting.

PAGE 30

30 Table 1 1. Physiological values of human lung Parameters Central Peripheral Ref. Epitheli a l lining fluid thickness ( µ m) 10 0.05 0.08 (14,44) Epithelial lining fluid volume ( mL ) 4 25 7 20 (45) Epithelium thickness ( µ m) --0.07 (14) Surface area (m 2 ) 2.8 102 (45)

PAGE 31

31 Table 1 2. Physiological values of rat lung Parameters Central Peripheral Ref. Epithelial lining fluid thickness ( µ m) 5 0.2 (46) Epithelial lining fluid volume ( mL ) 4 25 7 20 (47) Epithelium thickness ( µ m) 13 0.38 (48) Surface area (m 2 ) 2.8 102 (49)

PAGE 32

32 Figure 1 1 . A schematic of airway branching in the human lung .

PAGE 33

33 Figure 1 2 . Fate of inhaled corticosteroids after inhalation . Adapted from (18) .

PAGE 34

34 CHAPTER 2 E VALUATING P ULM ONARY TARGEING OF INHALED CORTICOSEROIDS USING NONCOMPARTMENTAL ANAL YSIS Background Pulmonary diseases are often treated with drugs given via inhalation as the local delivery provides distinct pulmonary effects with reduced systemic side eff ects. It has been s hown previously, that the degree of pulmonary selectivity upon inhalation, often also referred to as pulmonary targeting (18) , is determined by the interplay of several pharmacokinetic and pharmacodynamic (PK/PD) factors: prolonged pulmonary res idence time, high systemic clearance, and low oral bioavailability (41) . While these simulations showed the general importance of these parameters, a direct com parison of the degr ee of targeting achieved by clinically employed inhale d corticosteroids in a suitable animal model has not yet been reported. In order t o assess pulmonary targeting, receptor occupancy was used as a biomarker to link free drug concentr ations at the site of action with its pharmacological effects. Because co rticosteroids are required to bind the glucocorticoid receptors (GR) to elicit their pharmacological effects, the degree of receptor occupancy and the extent of induced biological res ponse are closely c orrelated (50) . Therefore, recept or occupancy can be used for quantifying pharmacologic ally relevant concentrations in the tissues. The assessment of the drug exposure in the target site and systemic organs can help to understand the degree of pulmonary selectivity as b asis for therapeuti c and adverse effect quantitatively. In the present study, receptor occupancies of a series of commercially available corticosteroids were measured in rats using a previously described ex vivo receptor

PAGE 35

35 binding assay (37) . Pu lmonary targeting of these corticosteroids was quantified and correlations to drug properties were at tempted. Method Materials The following corticosteroids in micronized form were obtained from t he indicated source s: beclomethasone dipropionate (BDP) from 3M ((St Paul, MN, USA), b udesonide (BUD) from Sicor (Milan, Italy) , c iclesonide (CIC) and des Ci clesonide (des CIC) from BYK Gulden (Konstanz, Germany) , fluticasone propionate (FP) from Glaxo Wel lcome (Resear ch Tri angle Park, NC, USA ) , triamcinolone acetonide (TA) from Sigma (St Louis, MO, USA)). 17 beclomethasone monopropionate (BMP) was purchased f rom European Directorate for the Quality of Medicine ( https://w ww.edqm.eu/ ). 3 H l abeled t riamcinolone a cetonide (38 Ci/mmol) was obtained from New England Nuclear (Wilmington, DE, USA). Extra fine lactose monohydrate wa s purchased from EM industries (Hawthorne, NY). All other reagents were obtained from Sigma Chemica l Co. (St. Louis, M O, USA) or equivalent sources. Analytical HPLC grade methanol, formic acid, ethyl acetate, and other chemicals were purchased from Fisher Scientific ( Springfield, NJ , USA ) or equivalent source . Animal P rocedures The animal procedures wer e approved by the I nstitutional Animal Ca re and Use Committee, University of Florid a. M ale F 344 rats, weighing 200 250 grams, were obt ained from Harlan Labo ratories (IN , USA) and were housed in a constant temperature environment with a 12 h light/dark cyc le. On the day of experiment, the rats were handled gently to produce minimum stress. Each rat was weighed and anesthetized via intraperitoneal (IP) injecti on of a

PAGE 36

36 freshly prepared mixture at a dose of 1mL/kg (the mixture contained 1.5 mL of ketamine (100 mg/m l), 1.5 mL of xylazine (20 mg/ml), and 0.5 mL of acepromazine (10 mg/ml))). Then, rats received either dry powder mixtures (tested compounds mixed with extra fine lactose) via intratracheal (IT) administration at 100 µg/kg, or solutions via intravenou s (IV ) administrati on. IV solutions were prepared by dissolving the glucocorticoids in a mixture of PEG 300 and saline (2:1 v/v) (3:1 v/v for BMP) to obtain a final concentration of 200 µg/mL (180 µg/mL for BMP). Then either 100 µL of IV solution or the so lvent (for the plac ebo group) was injected into the tail vein. IT administration of glucocorticoid powders (BDP, BMP, BUD, CIC, desCIC, FP, and TA) was perf ormed as previously described (Arya et al. 2005). Briefly, the neck of the animal was shaved and as eptic ally cleaned, and the trachea was exposed. A tracheotomy was performed between the third and fourth tracheal rings. One inch of a 14 gauge Novalon cathe ter sheath attached to a delivery device (Philadelphia, PA) for IT administration. A dry powder mix ture (tested compou nds mixed with extra fine lactose) of 5 ± 0.5 mg was placed in the chamber of the device and instilled in the lungs with insufflation of 3 mL of air. A control rat, which received 5 mg of the vehicle (lactose), was included in each set o f exp eriments. Afte r IV or IT administration, the rats were decapitated with a guillotine at 0.5, 1, 2, 4, 6, or 1 2h (0.5, 1, 2, 4, 6, 7 , and 12h for BDP and BMP) . After decapitation, lungs, livers, kidneys, spleens (not for BDP and BMP), and brains were i mmedi ately removed and processed for receptor binding studies. Each animal represented a single time point with grouped data for selected organs. Three to ni ne animals were used per time point for every form (IV or IT) of administration .

PAGE 37

37 Ex V ivo R eceptor B indin g A ssay Recept or binding in different tissues (lung, liver, kidney , spleen and brain ) after IV and IT administration of the seven corticosteroids were m easured in an ex vivo receptor binding assay as described previously (37) . Briefly, immediately after decapitation, the lung without trachea, the liver, the two kidney beans , spleen , and brain were resected and placed on ice (kidney, spleen and brain were not collected after IV o f TA; spleen was no t collected after IV and IT of BDP and BMP) . To the weighed tissue an appropriate amount of ice cooled incubation buffer (10 mL of buffer for 1 g of liver and 4mL of buffer for 1g of lung and kidney tissue; buffer composition: 10mM Tris/ HCL, 10 mM sodium m olybdate, 2mM 1,4 dithiothreitol and 2mM ph enylmethylsulphonyl fluoride, PMSF) was added and the tissue s were homoge nized in a Virtis 45 homogenizer for three periods of 5 seconds at 40% of full speed, followed by intermittent cooling p eriod of 30second s . One mL of the homogenate was transferred i nto a centrifuge tube to which 100 µ L of 5% activated charcoal was added and mixed. After a 5 m in incubation on ice, the suspension was centrifuged (20 min s at 20 , 000 g, 4 ° C) to obtain a clear s upernatant. Aliquot s of the supernatant (150 µ L ) were transfer red into microcentrifuge tubes. Then, 50 µ L of 40 nM 3 H TA in incubation buffer was added to de termine total binding (TB), while 50 µ L of incubation buffer containing 40 nM 3 H TA and 40 µM unlab eled TA was added t o determine the non specific binding (NS). After a 16 24h s incubation at 4 °C, the unbound glucocorticoid was removed by adding 200 µL of 5% activated charcoal suspension. The mixture was incubated for 5 min s on ice and then centrifuged at 10,000 rpm for 5 min s . The radioactivity (dpm) in 300 µL of

PAGE 38

38 supernatant was determined using a liquid scintillation counter (Beckman model LS 5000 TD, Pal o Alto, CA). All determinations were performed in triplicate. FP S tability in L iver H omogenate Stoc k solution of FP we re prepared as 1mg/mL by dissolving the com pound in methanol. Rat liver cytosol was obtained as described above. FP solution was added to freshly prepared liver cytosol to obtain final concentration of 3 and 10 µg/mL and stored at 4°C u p to 24h. At 1.5, 3 , 15, and 24h, 100 µL of samples were remov ed, mixed with 300 µL of acetonitrile/water 50/50 for protein precipitation. After 10 min incub ation at 4°C, tubes were centrifuged at 13 000 rpm for 5 min using an Eppendorf Centrifuge (NY, USA ). 200 µL of the su pernatant was injected in HPLC. Isocratic H PLC analysis was performed at ambient temperature using a reversed phase C18 3.5µm column (Symm etry, 4.6×100mm) on a Waters HPLC (Milford, MA, USA). UV detection was set at 254 nm. The mobile p hase contained acet onitrile/water 70:30 and flow rate was 1mL/ min. A calibration curve covering a concentration range of 0.5 µg /mL to 10 µg/mL (r2 > 0.999) and the lower limit of quantification was 0.5 µg/mL . Data A nalysis The specific binding, determined in the ex vivo rec eptor binding assay for the rats in the placebo group, was used to quantify the overall available binding sites (binding sites present in the absence of tested corticosteroids). The ratio of the specific binding for drug treated rats to that for placeb o ra ts served as the estimate of percent of available binding sites. The percentage of occupied receptors, or receptor occupancy (RO) , was cal culated from this ratio as follows:

PAGE 39

39 ( 2 1) The accumulative recept or occupancy (AUC E ) for each organ was defined as area under the receptor occupanc y time curve over a period of 6 hs. In order to estimate mean and standard deviation of AUC E , a resampling technique (51) was used. Briefly, bootstrap samples were tak en from the recept or occupancy measurements at each time point. Then construct pse udo profile of occupancy over time. The two steps were repeated 100 times. As a result, 100 occupancy time pseudo profiles were constructe d. In order to investigate the relat ionship between re ceptor occupancy and drug properties, the AUCE differences (aver age receptor occupancy in non kidney tissues minus average receptor occupancy in kidney) were calculated. Then pulmonary targeting was def ined as ( 2 2 ) The tested drug properties, including log octanol water partition coefficients (logP), receptor binding affinity (RRA), solubility in water (Sol), systemic cleara nce (CL), and plasma protein binding, were obtained from literature. Hepatic intrinsic clearances (CL int ) were calculated with the assumption of the well stirred hepatic distribution model, which is related to liver blood flow, systemic clearance, and frac tion unbound (52) . Data analysis was performed in R (version 3. 5.1). The receptor occupancy differences across organs for a given corticosteroid were analyzed by one way ANO VA,

PAGE 40

40 followed by Tukey post hoc test for multiple comparisons. When p values were less than 0.05, diffe rences were considered statistically signifi cant. Results Typical recept or occupancy time profiles, using data for fluticasone propionate (FP) as an examp le are shown in Figure 2 1. The cumulative receptor occupancies (AUC E ) for a given tissue (lung, liver , kidney, spleen, and brain) are shown for a ll compounds in Figure 2 2 a nd Figure 2 3 after IV and IT administration, respectively. The kidney receptor oc cupancy was used for all compounds as a reference organ for quantifying systemic receptor occupancy (n o distinct metabolism and active transporter s relevant for corticosteroi ds) and displayed throughout Figures 2 1, 2 2 and 2 3. Receptor occupancy differen ces between non kidney organs and kidney are listed in Table 2 1 and 2 2 for all compounds after IV an d IT administration, respectively. The organ s that received drug from th e bloodstream were defined as systemically exposed organs, which include all organ s after IV administration and non pulmonary organs after IT administration. Receptor O ccupanc y in S ystemically E xposed O rgans For a given drug , receptor occupancies in lu ng, spleen, and kidney were very similar for most tested compounds after IV admini stration (Figure 2 2A B). Significantly higher splenic receptor occupancies were observed after IT (bu t not IV administration of all investigated drugs) (Figure 2 3B), sugges ting that drug does not enter the spleen only through passive events from the bloo d. After IV administration, hepatic receptor occupancies of BMP, BUD, and CIC were similar to those i n the kidney (Figure 2 2C), while significan tly lower hepatic receptor o ccupancies were observed for desCIC and FP; a higher hepatic occupancy

PAGE 41

41 was observe d for BDP when compared to the kidney. After IT administration, hepatic receptor occupancies of BMP, B UD, CIC, and desCIC were not significantly d ifferent from those in the k idney (Figure 2 3C), while significantly lower hepatic receptor occupancies were o bserved for FP; a higher hepatic occupancy was observed for BDP. The pattern of the receptor occupancy difference between liver and kidney were ve ry similar between the admin istration routes. This similarity was expected for organs receiving drugs from blo odstream. As previously reported (Arya et al. 2005) and shown here only for completeness, brain rece ptors of BDP, BMP, BUD, FP, and TA were gene rally occupied to a smaller degree than those in the kidney (Fig ure 2 2D & 2 3D). This was also true for CIC a nd desCIC which were not included in the previous publication. Pulmonary T argeting after IT A dministration As expected, the pulmonary receptor o ccupancies were significantl y higher than kidney receptor occupancies after IT administration for all investig ated compounds (Fig ure 2 3A). The positive differences in receptor occupan cies between lung and kidney after pulmonary delivery (Fig ure 2 3A) indi cated that the tested compou nds have achieved pulmonary targeting, but at different levels (Fig ure 2 5A). Re lationship B etween D rug P roperties and AUC E To further investigate the degree of pulmonary targeting, we correlated the drug properties with the occupancy difference between non kidney tissues (lung, liver, spleen, and brain) and kidney. The receptor occupancy differences between non kidney or gans a nd kidney are listed in Table 2 1 and 2 2 for all compounds after IV and IT administration, respectiv ely. Selected drug propertie s of the tested compounds are summarized in Table 2 3.

PAGE 42

42 Occupancy differences between spleen and kidney (AUC E, spleen A UC E, k idney ) after IT administration is negatively correlated with the solubility of the tested compounds ( Fig ure 2 4A). Mean differenc e between liver and kidney (AUC E, liver AUC E, kidney ) for the two administration routes is negatively correlated to the drug hepatic intrinsic clearance (CL int ) (Fig ure 2 4B). Consistent with previous results (Arya et al. 2005 ), with the new data for CIC and desCIC, there is positive correlation between mean difference between brain and kidney (AUC E, brain AUC E, kidney ) for th e two administration routes and lipophilicity of the tested compounds (Fi gure 2 4C). More important ly, the investigated cortico steroids have shown different levels of pulmonary targeting after IT administration (Fig ure 2 5A). And the degree of pulm onary targeting (AUC E, lung AUC E, kidney ) is positively correlated with total systemic clearance of the c ompounds, and negatively cor related with their solubilities (Fig ure 2 5B & C). FP S tability T est After 24h incubation in the buffer that was used i n the ex vivo receptor occupancy assay, 88.3% and 91.7% of the original concentration remained for concentr ation 3 and 10 µg/mL, respec tively. Discussion Ex vivo receptor binding studies have been employed for a number of drug classes including corticoster oids (37,53) , opiates (47) and histamine (54) . The present study assessed the receptor occupancies in lung and a range of systemic organs for s even corticosteroids after t wo administration routes (IV and IT) in rats. Because the abundance of glucocorticoid receptor is similar for most organ s ( 55 57) , the degree of receptor occupancy should be closely correlated with free concentrations in the tissues. Thus, the major advantage of this approach is that the

PAGE 43

43 receptor occupancies measured in tiss ues can serve as a bioma rker for the pharmacodynamic ally relevant drug concentrations in the tissues. Thus, this biomarker measured using ex vivo (37) or in vivo (58) methodology can be used to assess drug exposure in the lung and other organs in animals with high temp oral resolution, while o ther assays (e.g. Sephadex i nduced inflammation) do not have time resolution (59) . It should be mentioned that the animal experiments in this study were performed not simultaneously, because of the large number of corticosteroids included in the study. While, the ex vivo receptor occupancy metho dologies were identical for all investigated compoun ds some of the variabilities within this study might be related to slight differences of the experimental conditions. In our study, corticosteroids were given via both IV and IT administration. Differenti al receptor occupancies in a range of tissues after IV administration can provide more understanding of the drug distribution and metabolism. In addition, comparison between systemic and pulmonary administration allows to quantify the benefits of local pul monary delivery. It is w ell established that cortico steroids diffuse freely across membranes because of their lipophilicity. It was reasonable to expect that free drug exposure in organs that received drug from the bloodstream (all organs after IV administ ration and non pulmonary organs after IT administrat ion) without transporter or enzyme involvement will be similar and as consequence also the receptor occupancies. This was the case for lung, spleen (with one exception) and kidney after IV administration (Fig ure 2 2A B) indicati ng that the drug distributio n process is similar for these organs. This was not the case for brain (as reported for most of the corticosteroids previously (60) ) and liver (Fig ure 2 2C D, 2 3C D). Other processes, e.g.

PAGE 44

44 transporters in case of the brain (60) , or meta bolic events occurring in the liver, might be relevant for these observations. As mentioned above, after IV administration, the receptor occupancies in lung, spleen, and kidney were very similar for a giv en corticosteroid (Fig ur e 2 2A B), with the one exce ption being the pro drug CIC. The significantly higher spleen receptor occupancy for CIC might be due to a high degree of activation of CIC in the spleen after IV administration. Unfortunately, the experi ments for BDP, being als o a pro drug, did not monito r spleen receptor occupancy; thus, not allowing support of this hypothesis. An in vitro experiment showed that CIC and desCIC had the same potency in inhibiting the concanavalin A stimulated rat spleen ce lls (61) . This result suggests that CIC can easily be converted to desCIC in these cells, which is in line with our findings of spleen activation of CIC. Except CIC, the good agreement for lung, spleen and kidney, reinforces the hypot hesis, which is drug exp osure of these organs is mai nly driven by passive diffusion. This also suggests that kidney might be a good organ to serve as the reference organ for specifying pulmonary targeting after IT administration of drugs. Reasons for not u sing spleen will be disc ussed below. In contrast, re ceptor occupancies in the brain for all the compounds (Fig ure 2 2D and 2 3D) and in the liver for some compounds (e.g. FP and desCIC) (Fig ure 2 2C and 2 3C) were different from those in the kidney. The re asons for the brain resu lts, as previously reported (60) , are related to the effects of transporters while distinct met abolism (activation or d e activation of the drug und er investigation) affecting free drug concentrations in the given organ might be relevant for observations in liver. As previously published (60) , the presence of efflux transporter P glycoprotein (P gp) at the

PAGE 45

45 blood brain barrier decreased the free drug concentrations in the brain, and consequently reduced receptor occup ancies in the brain. In agreement with these results , a recent study has demonstrated that P gp transporters played also a role in affecting free concentrations of budesonide in fetuses by modulating transplacental distribution in mice (62) . For FP and desCIC, but not CIC, lower rec e ptor occupancies in the liver than those in the kidney were observed after IV administration (Fig ure 2 2C) and after IT administration (Fig ure 2 3C). Results for FP and desCIC can be explained by the high intrinsic clearance of FP and desCIC in the liver r esultin g in free concentrations in the liver being lower than those in the kidney. It might be argued that lower receptor occupancies observed for desCIC and FP in the liver could be due to metabolism of drug during incubation in the ex vivo receptor occup ancy as s ay. However, we are able to demonstrate that FP was stable under the non physiological conditions of the receptor binding ( e.g. FP sta bility test ). Thus, the lower liver receptor occupancies for FP and des CIC are likely due to the high intrinsic c learance, a s at any given time point both free drug concentrations and liver receptor occupancies should be lower than those in other organs. This is further supported by a negative correlation between the occupancy difference (AUC E, liver AUC E, kidney ) and hepati c intrinsic clearance (Fig ure 2 4B), as compounds with high intrinsic clearance tend to have low hepatic receptor occupancies while receptor occupancies for lower intrinsic clearance drugs are similar to those in kidney. Because of missing TA kid ney data, l iver receptor occupancies were used in Figure 2 2A for expressing pulmonary targeting. In this case liver could be used

PAGE 46

46 as a reference organ (37) , because the intrinsic cleara nce for TA is relatively small resulting in liver receptor occupancies being similar to those in the kidney. In the case of BDP, hepatic receptor occupancies were higher than those in the kidney (Fig ure 2 2C and 2 3C). This is likely due to the activation of BDP in t he liver to BMP resulting in a higher local concentration of BMP than those in other tissues. Since CIC is also a prodrug, higher hepatic receptor occupancies could be expected from the activation of CIC to desCIC. However, we believe that the e ffects of h epatic activation of CIC and the very efficient metabolism of desCIC counteract each other, resulting in receptor occupancies similar to those observed in the kidney. The above discussion about receptor occupancies in the liver reinforces that liver rece p tor occupancy should not serve as overall systemic receptor occupancies, especially for the compounds with high intrinsic clearance. As discussed above, we found for non prodrugs, similar receptor occupancies in lung, spleen, and kidney after I V administ r ation (Fig ure 2 2A B). This agreed with our hypothesis that receptor occupancies are similar in organs that receive drugs from the bloodstream, as long as spleen does not represent a metabolizing organ. If the assumption that spleen is receiving drug only from the bloodstream is also true after IT administration, one might expect similar spleen and kidney receptor occupancies also after IT administration. However, spleen receptor occupancies after IT administration were significantly higher than those in t h e kidney (Fig ure 2 3B) for all of the drugs tested, suggesting that free concentrations realized in the spleen after IT administration are higher than those in the kidney. This effect was more pronounced for slowly dissolving compounds (Fig ure 2 4A). Lite r ature demonstrated dendritic cells (DCs) and

PAGE 47

47 alveolar macrophages (AMs) are phagocytic cells that can internalize particles and pathogens from the lung (63,64) , move out of the lung and into draining lymph modes, and migrate to other organs (65) , such as the spleen (66) . If this is the case, the difference of receptor occupancy between spleen and kidney should be more pronounced for slowly dissolving drugs. The positive correlation between occupancy differences betw e en spleen and kidney and solubility indicates that compounds with low solubility tend to have higher splenic receptor occupancies (Fig ure 2 4A). Therefore, spleen should not be an organ to assess the extent of systemic exposure of drugs, especially for th e compounds with low solubility (43) . The above discussions suggest, that kidney is a proper organ to serve as a reference systemic organ for describing the overall systemic exposure, because it is a highly perfused organ and has relative low enzyme and active t ransporters that are relevant for corticosteroids. Receptor occupancies in kidney differ across tested compounds. These results are due to the additive effects of the differences in their drug properties, such as receptor binding affinity, protein binding , volume of distribution, and systemic clearance. Using kidney as the systemic organ, all of the investigated corticosteroids demonstrated pulmonary targeting after IT administration of the dry powder formulations (Fig ure 2 5A) with the highest targeting o bserved for the second or third generation corticosteroids. Previous PK/PD simulations (41) have shown that high systemic clearance and low solubility is beneficial for pulmonary targeting (Fig ure 2 5B C). Our ex vivo receptor binding results seem to confirm results of these simulations. The top three compounds, FP , CIC, and desCIC, possess favorable drug properties, such as

PAGE 48

48 low solubility and high systemic clearance. Although BUD has been shown to be retained in airways as esters, which can potentially prolong pulmonary residence time (67) , beneficial effects of esterification on pulmonary selectivity seems not to be visibl e in our experiment. As the pulmonary fate and subsequent degree of pulmonary selectivity depends not only on drug but also formulation factors (e.g. particle size distribution, particle shape, effects of adjuvants on dissolution), also our DPI form ulatio ns were custom made and representing commercially available formulations, translating the results of our animal experiments to the clinical situation needs to be done with care. Summary Our study investigating receptor occupancies after IV and IT adm inistr ation of commercially available corticosteroids indicated that receptor occupancies in tissues differ and are affected by transporters, organ specific metabolic events, and cell trafficking. The present study provides a quantitative assessment of pul monary targeting of inhaled corticosteroids and demonstrates differences across commercially available corticosteroids for both pulmonary as well as systemic drug exposure. Monitoring the receptor occupancies in lung and systemic organs is a viable method to ass ess the properties of inhaled corticosteroids.

PAGE 49

49 Table 2 1 . E , mean ± SD) between non kidney organs (lung, spleen, liver, and brain) and kidney after IV administration of seven ICSs (87 µg/kg for BDP, 78 µg/kg for DP, 100 µg/kg for other ICSs) in rats . ICSs Receptor occupanc y difference E ) (kidney as reference organ) (% *h) AUC E,lu AUC E,ki AUC E,sp AUC E,ki AUC E,li AUC E,ki AUC E,br AUC E,ki BDP (n=3) 30 ± 53 --66 ± 30 216 ± 56 BMP (n=3) 1 ± 25 --31 ± 15 355 ± 38 BUD (n=3) 23 ± 34 20 ± 49 32 ± 46 242 ± 64 CIC (n=6) 63 ± 34 158 ± 31 19 ± 16 48 ± 36 Des CIC (n=6) 43 ± 55 54 ± 55 32 ± 19 97 ± 44 FP (n=6) 49 ± 82 52 ± 17 157 ± 43 215 ± 57 TA (n=3) 19 ± 28 * ------Note: *AUC E,lu AUC E,li , because kidney receptor o ccupancy for TA after IV administration was not present The collected organs are lung (lu), kidney (ki), spleen (sp), liver (li), and brain (br). The tested compounds are Beclomethasone dipropionate (BDP), Beclomethasone 17 monpropionate (BMP), budesonide (BUD), Ciclesonide (CIC), Des ciclesonide (desCIC), fluticasone propionate (FP), and triamcinolone acetonide (TA) .

PAGE 50

50 Table 2 2 . Differences in receptor E , mean ± SD) between non kidney organs (lung, spleen, liver, and brain) an d kidn ey af ter IT administration (100 µg/kg) of the seven ICSs in rats. ICSs Receptor occupancy difference E ) (kidney as reference organ) (% *h) AUC E,lu AUC E,ki AUC E,sp AUC E,ki AUC E,li AUC E,ki AUC E,br AUC E,ki BDP (n=3) 80 ± 67 --178 ± 4 2 83 ± 70 BMP (n=3) 101 ± 31 --45 ± 29 278 ± 23 BUD (n=6) 33 ± 46 33 ± 35 16 ± 30 102 ± 42 CIC (n=6) 141 ± 39 135 ± 28 24 ± 28 59 ± 28 Des CIC (n=6) 103 ± 85 6 9 ± 80 57 ± 79 48 ± 67 FP (n=9) 143 ± 52 105 ± 47 90 ± 43 183 ± 111 TA (n =9) 82 ± 41 66 ± 26 35 ± 28 339 ± 72 Note: The collected organs are lung (lu), kidney (ki), spleen (sp), liver (li), and brain (br). The tested compounds are Beclomethasone dipropionate (BDP), Beclomethasone 17 monpropionate (BMP), budes onide (BUD), Ciclesonide (CIC), Des ciclesonide (desCIC), fluticasone propionate (FP), and triamcinol one acetonide (TA).

PAGE 51

51 Table 2 3 . Summary of biopharmaceutical and PK/PD properties of tested corticosteroids in rats . Corticosteroids Lo gP a RRA b Solubility (mg/L) CL (L/h) CL int (L/h) n f u BDP 4.1 53 0.16 c 4.30 i --0.13 o BM P 3.14 1345 14 d 0.38 i 44.34 0.016 i BUD 2.8 935 23 e 0.33 j 7.59 0.074 l CIC 4.3 12 0.09 f 2.08 k --0.006 p desCIC 3.89 1212 --0.70 k 311.11 0.018 q FP 3.8 1800 0.4 g 0.67 l 179.26 0.023 i TA 2.2 223 25.5 h 0.23 m 3.34 0.09 9 m Note: The tested compounds are Beclomethasone dipropionate (BDP), Beclomethasone 17 monpropionate (BMP), budesonide (BUD), Ciclesonide (CIC), Des cicl esonide (desCIC), fluticasone propionate (FP), and triamcinolone acetonide (TA). a reference (60) ; b reference (68) ; c reference (69) ; d reference (70) ; e reference (71) ; f reference (72) ; g reference (73) ; h calculated using General Solubility Equations (74) ; i reference (75) ; j reference (76) ; k reference (77) ; l refere nce (78) ; m reference (79) ; n reference (80) ; o reference (81) ; p CL int was calculat ed; q BDP fraction unbound in human (82 ) ; r CIC fraction unbound in human (83) ; s reference (84) .

PAGE 52

52 Figure 2 1 . % Receptor occupied time profiles for F P after systemic (IV) and pulmonary (IT) administration . Receptor occupancy in kidney was used as a r eference and displayed in each figure. The mean ± SD is given .

PAGE 53

53 Figure 2 2 . Receptor occupancies of lung, spleen, liver and brain comparing with kidney receptor occupancies after IV administration of tested compounds in rats. The mean ± SD is gi ven. Note: *hepatic receptor occupancy was used because kidney receptor occupancy for TA after IV administration was not present. The tested compounds are Beclomethasone dipropionate (BDP), Beclomethasone 17 mon propionate (BMP), budesonide (BUD), Ciclesoni de (CIC), Des ciclesonide (desCIC), fluticasone propionate (FP), and triamcinolone acetonide (TA) .

PAGE 54

54 Figure 2 3 . Receptor occupancies of lung, spleen , liver and brain comparing with kidney receptor occupancies after IT administration of tested compou nds in rats. The mean ± SD is given. Note: The tested compounds are Beclomethasone dipropionate (BDP), Beclomethasone 1 7 monpropionate (BMP), budesonide (B UD), Ciclesonide (CIC), Des ciclesonide (desCIC), fluticasone propionate (FP), and triamcinolone acet onide (TA) .

PAGE 55

55 Figure 2 4 . E in non pulmonary organs in rats. The mean E between spleen and kidney vs. solubility , E between liver and kidney vs. hepatic intrinsic clearance (CL int ) , E between brain and kidney vs. logP Note: The tested compounds are Beclomethasone dipropionate (BDP), Beclomethasone 17 monpropionate (BMP), budesonide (BUD), Ciclesonide (CIC), Des ciclesonide (desCIC), fluticasone propionate (FP), and triamcinolone acetonide (TA) .

PAGE 56

56 Figu re 2 5 . Pulmonary targeting of tested compounds and its relationship with drug properties for seven corticosteroids in rats. The mean ± SD is given . A) Pulmonary targeting of seven corticosteroids , AUC E between lung and kidney (pulmonary targeting) vs. systemic clearance , AUC E between lung and kidney (pulmonary targeting) vs. solubility . Note: The tested compounds are Beclomethasone dipropionate (BDP), Beclomethasone 17 monpropionate (BMP), b udesonide (BUD), Ciclesonide (CIC), Des ciclesonide (desCIC), fluticasone propionate (FP), and triamcinolone acetonide (TA) .

PAGE 57

57 CHAPTER 3 PHYSIOLOGICALLY BASED PK/PD MODEL TO ASSESS PULMONARY TARGETING OF BECLOMETHASONE DIPROPIONATE AND ITS ACTIVE METABOLIT E Background Orally inhaled drug products (OIDP s ) for treatment of pulmonary diseases have been designed to deliver the drug directly to the site of action, thereby achieving distinct pulmonary effects with reduced systemic side effects . Although metered dose inhalers have been introduced more than 60 years (1) , the pulmonary fate of inhaled medicines, as determined by the biopharmaceutical characteristics of the drug/formu lation/device combination and their interaction with the body, is far from being well unde rstood. Challenges are mainly related to the complex interplay between anatomical and physiological aspects of the lung, including drug and formulation properties whi ch determine together the pulmonary deposited dose, regional deposition and post depositio n events (dissolution/ diffusion/receptor interaction). Recently, the use of computational fluid dynamic approaches (85) or algebraic models (Emmace, MPPD (86) ) resulted in a better understanding and prediction of deposition events. To further understand post deposition events and to relate these to the overall fate of deposited drug in the lung and systemic circulation, physiologically based approaches have been applied very recently to bronchodilators and corticosteroids with the overall goal to predict pulmo nary and systemic concentration time profiles (43,87,88) . However, for a better understanding of efficacy and safe ty, studies should not be limited to total drug concentrations alone but should either be able to predict free drug concentrations or link tot al drug to markers of pulmonary effects and systemic side effects. For corticosteroids, monitoring the receptor oc cupancy in the lung

PAGE 58

58 and systemic organs through ex vivo (37) or in vivo r eceptor binding (58) assays have been suggested as a suitable approach to quantify pulmonary targeting. In this study we were interested in developing a physiologically based PK/PD (PBPK/PD) model for the glucocorticoid prodrug beclomethasone dipropionate (BDP) after pulmonary and intravenous administration. The model attempted to predi ct drug conc entration time profiles of the prodrug BDP and its active metabolite beclomethasone 17 monopropionate (BMP) an d linked pulmonary and systemic concentrations to the measured pulmonary and systemic receptor occupancy time profiles. Intravenous an d pulmonary administration of BDP and BMP allowed to more precisely evaluate events important for pulmonary target ing. Methods Chemicals Micron ized BDP was kindly provided by 3M (St Paul, MN, USA). BMP was purchased from European Directorate f or the Qual ity of Medicine ( https://www.edqm.eu/ ). F luticasone propionate (FP), used as the internal standard in LC/MS assay , was received from Gla xo Group Research (Herts, UK). 3 H labeled t riamcinolone acetonide (TA) (38 Ci/mmol) was obtained from New England Nuclear (Wilmington, DE, USA). Analytical HPLC grade methanol, formic acid, ethy l acetate, and other chemicals were purchased from Fisher Scientific ( Springfield, NJ , USA ) or equivalent source . The solid phase LC18 (3 m L ) car tridges for sample extraction were obtained from Supelco (Bellefonte, PA , USA ). Animal P rocedures The animal pr ocedures were approved by the Institutional Animal Care and Use Committee, University of Florid a. M ale F 344 rats, weighing 200 250 grams, were

PAGE 59

59 o bt ained from Harlan Laboratories (IN USA) and were housed in a constant temperature environment with a 12 h lig ht/dark cycle. On the day of experiment, e ach rat was weighed and anesthetized via intraperitoneal (IP) injection of a freshly prepared anesthes ia cocktail at a dose of 1mL/kg (prepared by mixing 1.5 mL of ketamine (100 mg/mL), 1.5 mL of xylazin e (20 mg/m L), and 0.5 mL of acepromazine (10 mg/mL)) . Subsequently, the rats received either dry powder formulations (BDP or BMP mixed with extra fine lact ose) via intratracheal (IT) administration of 100 µg/kg or 100 µl of solution (200 µg/ml of BDP or 180 µg/ml of BMP) via intravenous (IV) administration. Dry powder formulation was prepared by geometrically diluting the tested compounds (BDP and BMP) wit h extra fine lactose . Intratracheal (IT) administration was performed by introducing a special round ripped can nula to the trachea through the mouth. The cannula was attached to a delivery device (Penn Century, Philadelphia, PA) for administration of dry p owder. A mixture of lactose and drug (100µg/kg of drug) or lactose alone was placed in the chamber of the device and instilled in the lungs by insufflating 3 mL of air. For the intravenous (IV) administration , either 100 µ L of IV solution (200 µg/m L for BDP or 180 µg/mL for BMP in a mixture of PEG 300 and saline 3:1 v/v) or 100µ L of the solvent (for the plac ebo group ) was injected into the tail vein. The rats were subsequently decapitated with a guillotine at 0.5h, 1h, 2h, 4h, 7h, 12h after drug administ ration and 6 hours after lactose instillation or saline /PEG 300 injection . Blood was collected in tubes co ntaining heparin and suitable enzyme inhibitors ( ethylenediaminetetraacetic acid sodium salt ( EDTANa2 ) and sodium fluoride ( NaF) )

PAGE 60

60 and centrifuged at 3 , 000 rpm for 10 min s . The plasma was separated and stored at approxim ately . After preparation of plasma, lungs , livers, and kidneys were removed and immediately processed for receptor binding studies. Three animals were used per tim e point for every form of administration (IV and IT) and for BDP and BMP . Sample P reparation and LC/MS A n alysis The concentrations of BDP and BMP in the plasma and tissue samples were analyzed using a liquid chromatography tandem mass spectroscopy (LC MS/ MS) assay published by Wang et al. (87) . Briefly, a 1mL of plasma samples were thawed at room temperat ure. Then 50 µL of internal standard solution was added and mixed well. After adding 1mL of 30% (v/v) ethanol, the mixed solution was incubated for 15 and centrifuged at 4000rpm (3148g) for 15 min (Jouan Centrifuge Model CR422, Winchester, VA) to remove the protein precipitate. The supernatant was then transferred onto the solid phase extraction column. After drained under necessary vacuum, th e column was then washed with three different elution. Finally, the sample was eluted with 3mL of 35:65 (v/v) ethyl acetate/n heptane mixture. For determining BDP and BMP concentration in tissue samples (lung, liver, kidney), the compounds were extracted from rat tissue homogenates that were prepared as described in the ex vivo receptor binding assay section (see below). One mL of the homogenate was mixed with 4mL of ethyl acetate. After centrifugation, the upper organic layer was transferred into a new tu be and evaporated in a vacuum centrifuge. The residue was subse quently reconstituted in 50 µL of mobile phase. For both plasma and tissue samples, a volume of 30 µL was injected into the LC/MS/MS system.

PAGE 61

61 LC/MS/MS was performed using a Micromass Quattro LC Z triple quadrupole mass spectrometer (Beverly, MA, US) linked to a high performance liquid chromatography. A Waters C18 3.5µm column (Symmetry, 2.1×50mm i.d., Milford, MA) and a Whatman 3.5 µm ODS C18 guard column cartridge (Clifton, NJ) were used. The an alytes were eluted isocratically with a mobile phase of water c ontaining 33mM ammonium acetate (A) and methanol containing 0.33% formic acid buffer (pH 3.44) (B) ( A:B = 30:70, v/v) at a flow rate of 0.3 m L/ min. Conditions for MS analysis included a source temperature of 120° C , the desolvation temperature of 450° C , c apillary and cone voltages of 3.0 kV and 30 V , respectively. The MS was performed in a p ositive mode under multiple reactions monitoring (MRM) in electrospray ionization ( ESI ). The m/z ratios of molecular ion and product ion for BDP, BMP, and FP (interna l standard) were 521 > 319, 465 > 279, and 501 > 313, respectively. Data analysis was performed using Masslynx software. The lower limits of quantification (LLOQ) were 0.05 ng/mL for both BDP and BMP within an accuracy range of 80 120%. Ex V ivo R eceptor O ccupancy A ssay Receptor binding in the selected tissues (lung, liver, and kidney) after IV or IT administration of BDP and BMP was measured in a previously developed ex vivo receptor binding assay (37) . Briefly, immediately after decapitation, the lung, without trachea, the liver, and the two kidney beans were resected and placed on ice. To the weighed tissue an appropriate a mount of ice cooled incubation buffer (10 mL of buffer for 1 g of liver and 4mL of buffer for 1g of lung and kidney tissue; buffer composition: 10mM Tris/HCL, 10 mM sodium molybdate, 2mM 1,4 dithiothreitol and 2mM phenylmethylsulphonyl fluoride, PMSF) was added and homogenized in a Virtis 45 homogenizer at 40% of ful l speed, for three periods of 5 seconds each with a 30 second

PAGE 62

62 cooling period between each step. A portion of this homogenate was used for determining the tissue drug concentrations by LC MS/MS ( see above) after storage of the homogenate at 70 ° C. For the receptor binding assay, one mL of the unfrozen, directly processed homogenate was transferred into a centrifuge tube to which 100 µL of 5% activated charcoal was added and mixed. After a 5 min s incubation on ice, the suspension was centrifuged to obtain a clear supernatant. Aliquots of the supernatant (150 µL) were transferred into microcentrifuge tubes. Then, 50 µL of 40 nM 3H TA in incubation buffer was added to determine total binding (TB) w hile 50 µL of incubation buffer containing with 40 nM 3H TA an d 40 µM unlabeled TA was added to determine the non specific binding (NS). After a 16 24h s incubation at 4 °C, the unbound glucocorticoid was removed by adding 200 µL of 5% activated charcoal su spension. The mixture was incubated for 5 min s on ice and the n centrifuged at 10,000 rpm for 5 min s . The radioactivity (dpm) in 300 µL of supernatant was determined using a liquid scintillation counter (Beckman model LS 5000 TD, Palo Alto, CA). All dete rminations were performed in triplicate. The specific binding for rats in the placebo group was used to quantify the overall available binding sites (binding sites present in the absence of BDP treatment). The ratio of the specific binding for drug treated rats to that for placebo rats served as the estimate of percen t of available binding sites . The % of occupied receptors were calculated from this ratio as follows ( 3 1)

PAGE 63

63 Non c ompartmental A nalysis Because both parent drug and metabolite were given to rats via IV and IT administration, noncompartmental analysis (NCA) was used to calculate some metabolite related pharmacokinetic parameters from BDP and BMP concentration time profiles and resulting ar ea under the concentration time profiles (AUC ; calculated by the trapezoidal rule ) . The degree of BDP conversion was calculated by comparing the AUC 0 of BMP after IV administration of BDP ( AUC 0 , BMP ( IV BDP) ) to that after IV administration of an equimolar dose of BMP ( AUC 0 , BMP (IV B M P) ): ( 3 2 ) The absolute systemic bioavailability of IT administered BDP and BMP was calculated as follows: ( 3 3 ) where j represents BDP or BMP; AUC 0 , BDP (I V BDP) , AUC 0 , B D P (I TBD P) , AUC 0 , B M P (I VBM P) , and AUC 0 , B M P (ITB M P) are AUCs of BDP or BMP after IV or IT administration of BDP or BMP; Dose BDP (IVBDP) , Dose BDP (ITBDP) , Dose BMP (IVBMP) and Dose BMP (ITBMP) are 87, 100, 78, and 100 µg/kg, respectively. The clearance of BMP after IV administr ation was also calculated as dos e / AUC 0 , and used as the initial estimate for the parameter estimation in the PBPK model. PBP K M odel D evelopment Global PBPK model structure A physiologically based pharmacokinetic model (PBPK) incorporating lung disposit ion and conversion of BDP to BMP was developed within Monolix ( version

PAGE 64

64 2018R2, LIXOFT, France ) (Figure 3 1A). The model incorporated lung, gut, spleen, liver, kidney, and adipo se tissues, while other tissues were combined to reduce the model complexity and dimensionality according to Nestorov et al. (88) . Perfusion rate limited distribution was assumed to apply for all the tissues. The following equation was used to de scribe the change in the amount of drug in a given organ: ( 3 4 ) where j represents the parent drug BDP or the metabolite BMP; t i represents a given tissue compartment: gut, spleen, liver, kidney, adipose, rapidly perfused, or slowly perfus ed tissues; V is the tissue volume; Q is the tissue blood flow; K p is the tissue partition coefficient; C ti,j and C p,j are the concentrations of compound j in a given tissue t i and plasma, respectively. The blood flows and volumes of the tissues were taken from the literature (Table 3 1) (89,90) . The tissue partition coefficients for both BDP and BMP (except for lung, liver, and kidney for BMP) were calculated from dr ug physicochemical properties and protein binding according to Rodger & Rowland method (91) . K ps of BMP for lung, liver and kidney were estimated by fitting the proposed model (Figure 3 1) to plasma and tissue (lung, liver and kidney) concentrations after IV administr ation of BMP. The unbound fractions of BDP and BMP in plasma (f up ) were taken as 0.13 and 0.016, respectively (77,82) . Th e unbound fractions in tissues (f ut ) were estimated from the volume of distributi on at steady state (V dss = V p + V t * f up / f ut ) after IV injection of BDP and BMP, respectively. BDP, as a prodrug, has been shown to be enzymatically hydrolyzed to BMP via esterase (92) . This process was assumed to occur in the blo od and all tissue compartments with the same metabolic clearance (CL met ) following the equation :

PAGE 65

65 ( 3 5 ) where f up is the unbound fraction of BDP in the plasma. CL met was esti mated via fitting the plasma and tissue (lung, liver, and kidney) concentrations after IV of BDP when other parameters were fixed. Sub compartmental model of lung Lung was divided into the tracheobronchial (central) and alveolar (peripheral) compartments based on its anatomical and physiological differences (42) (Figure 3 1A). The central region was perfused by the arterial blood via the bronchial circulation (1% o f the cardiac output (93) ), while blood was reaching the alveolar region via the pulmonary circulation with a rate identical to the cardiac output. After pulmonary de livery of BDP or BMP, BDP or BMP particles deposited in the lung and s ubsequently dissolved in the epithelial lining fluid (ELF) according to their dissolution rates, the dissolved drug was subsequently absorbed into the lung tissue via passive diffusion ( Figure 3 1B). Because information on the BDP and BMP dry powders were not available, deposition and dissolution process could not be predicted from suitable deposition software or Nernst Brunner relationships and while knowing the pulmonary deposited dose the central to peripheral deposition ratio was predicted during the fi tting process. Dissolution after pulmonary delivery The change in undissolved drug was consequently described by a simple first order process: ( 3 6 )

PAGE 66

66 w here i represents centra l or peripheral region of the lung; j represents BDP or BMP after pulmonary delivery of BDP or BMP; A represents the amount of undissolved drug; k dis is the first order dissolution rat e constant. Pulmonary absorption When drug is dissolved, the primary f orce driving its transport into the blood stream is the concentration gradient between adjacent compartments. The change of the dissolved drug in the lining fluid was defined as: ( 3 7 ) w here i represents central or peripheral re gion of the lung; j represents BDP or BMP; V fluid is the volume of rat lung lining fluid, which can be calculated from the corresponding surface area and the thickness of the lining fluid layer. The surface area S and th e thickness for central and peripher al lung was taken as 75 cm 2 and 6357 cm 2 (49,94) , and 5 µm and 0.2 µm (46,95) , respectively. P app is the permeability; f u,fluid is the unbound fraction in the lining fluid and assumed to be 1 (43) . Metabolism was not allowed in th e lining fluid, as data on the ester ase activity were not available . Consequently, the drug was allowed to enter the lung tissue. In the case of BDP, the rate with which the drug amount changes in the lung tissue compartments (central or peripheral) was de termined by the rates BDP is enterin g and leaving the specific compartment and in addition by the rate with which BDP is activated in the lung to BMP. This is described by the following equati on :

PAGE 67

67 ( 3 8 ) w here i represents central or periphera l lung; j represents BDP or BMP. For describing the rate change in BMP after pulmonary delivery of BMP, the above equation was adapted. The last term in the above equation was removed, but dissolution and absorption into the lung tissue was modelled in the same way as described for BDP. The rate of change in BMP after pulmonary delivery of BDP was derived from the metabolic input rate and the absorptio n into the blood stream. Parameters were estimated in a stepwise approach: K ps in lung, liver and kidney we re estimated using Equation 3 4 using data after IV administration of BMP (K ps of BDP in lung, liver, and kidney, and K ps of BDP and BMP in gut, sple en, adipose tissue, rapidly perfused and slowly perfused tissues were calculated as described by Rodgers an d Rowland (91) ; the metabolic rate for BDP in all tissues and system conversion to BMP was estimated using Equation 3 5 after IV administration of BDP; and pulmonary related parameters were estimated using E quation 3 6 to 3 8 after IT adminis tration of BDP and BMP. In d etails, the dissolution rate constants and permeability from ELF into lung tissue for BDP were estimated via fitting the plasma and tissue (lung, liver and kidney) concentrations after I T administration of BDP ; the dissolution r ate constants and permeabili ty from ELF into lung tissue for B M P in the lung were estimated via fitting the plasma and tissue (liver and kidney) concentrations after IT administration of B M P , when other parameters (systemic specific parameters and paramet ers obtained by

PAGE 68

68 fitting IV d ata) were fixed . Since the overall drug permeability and membrane thickness are correlated, the alveoli permeability was obtained from the estimates of the central lung permeability by correcting for the differences in thickness ( 96,97) . The cell layer thic knesses of central and peripheral lung was taken as 13 µm and 0.38 µm, respectively (48) . E max Model for Receptor Binding An E max model was used to relate th e receptor occupancies to th e free drug concentrations in the corresponding organs (lung, liver, and kidney). The equation is defined as follows: ( 3 9 ) w here t i represents lung, liver, and kidney; E is the receptor; E max is the maximum receptor occupancy estimated using the developed model. EC 50 is the free drug concentration corresponding to the half maximal effect. E max and EC 50 were obtained by sequen tially fi tting the tissue concentrations (lung, liver and kidney) and receptor occupancies after I V administration of BMP. Since the lung has been divided int o two regions, the whole lung receptor occupancy was considered as a weighted average receptor occ upancy of the two regions. The whole lung receptor occupancy was calculated considering the volume of each region according to Boger et at. (43) .

PAGE 69

69 ( 3 1 0 ) The accumulative receptor occupancy (AUC E ) was calculated by the trapezoidal rule from the receptor occupancy time profiles. Then pulmonary targeting is defined as the difference of AUC E between lung and kidney (AUC E,lung AUC E, kidney ). The param eter estimation for PBPK/PD model was implemented in MONOLIX ( version 2018R2, LIXOFT, France ) and MONOLIX code is in Appendix. The random effect s among subjects were fixes as 30% by assessing the data variability. The final models were evaluated based on g oodness of fit plots, precision of parameter estimation, the Akaike information criterion (AIC). Sensitivity A nalysis The extended Fourier amplitude sensitivity test (eFAST) method as described by McNally et al. (98) was used to evaluate the influence of each model parameter on the simulated pulmonary and renal rec eptor occupancy after IT administration of BDP. A first order sensitivity index, S i , of a given parameter i , was calculated a s a fraction of total variance: ( 3 11 ) Where was calculated from the Fourier coefficients at the f requency of interest; i is the tested parameters: k dis , P app , F c , CL, CL met , body weight, K p in the lung an d kidney for BDP and BMP . The total order sensitivity index, S Ti , of a given parameter

PAGE 70

70 i was calculated as the remaining variance after the contribut ion of the complementary set . ( 3 12 ) w here Sci is the summed sensitivity index of the entire complementary set of parameters using their identification frequencies. A small index would indicate that the model output , receptor occupancy of lung or kidney would be insensitive to the parameter. I ndexes close to 1 would suggest that there may be amplified parameter error s associated with model structure. T he sensitivity analysis and all graphical presentations were performed in R s oftware (version 3.4.3). Model V alidation To validate the PBPK model structure and parameter estimates for bot h IV and IT routes, plasma concentrations of BDP and BMP were simulated after IV and IT administration of BDP. The predicted concentrations were subsequently compared with plasma PK reported after IV and IT administration of BDP (2.9 µmol/kg and 0.33 µmol /kg) (99) . The receptor occupancy time profiles were also predicted with the developed model and compared with r eceptor occupancy data of our study after IT administration of BMP, IV and IT delivery of BDP (see Method secti on). Assessment of Pulmonary Targeting Pulmonary targeting was defined as the cumulative receptor occupancy difference between lung and kidney a fter pulmonary administration. To assess the effect of drug properties on pulmonary targeting of BDP, dissoluti on rate (e.g. 0.4/h and 4/h) and permeability (e.g. 15 and 0.15 ×10 6 cm/s) were tested. Receptor occupancy time

PAGE 71

71 profiles in lung, central lung a nd kidney were predicted with the developed PBPK/PD model. Results PK/PD D ata for BDP and BMP in R ats BDP and BMP plasma concentration time profiles after IV and IT administration of BDP, plasma and tissue concentration time profiles of BMP, receptor occu pancy time profiles after IV and IT administration of BDP and BMP are shown in Figure 3 2. Results of the nonco mpartmental analysis are shown in Table 3 3. Based on AUC 0 estimates of BDP and BMP plasma concentration time profiles following a single IV injections of equimolar dose of BDP and BMP (Table 3 3), 95 .4 % of BDP was converted to BMP, indicatin g that BMP is the main metabolite of BDP. After IT administration of BDP, 5 1.2 % of intact BDP was directly absorbed from the lung into systemic c irculation , while 46.6% of pulmonary deposited BDP was converted to BMP in the lung . Therefore, almost all (97.9%) the BDP dry powder was ab sorbed after IT administration as either BDP or ted to be 71.5%. The total clearances were calculated as 3.6 and 0.35 L/h, and volumes of distribution at the steady stat e (V dss ) were calculated as 4.4 and 0.36 L for BDP and BMP, respectively. PBPK/PD M odel in R ats The proposed PBPK model was able to ca pture both BDP and BMP concentration time profiles in plasma and various tissues after both IV and IT administration route s, with model prediction in close agreement with the experimental observations (Figure 3 3) . The PBPK model based estimated PK paramet ers for BDP

PAGE 72

72 and BMP are shown in Table 3 4 with clearance values close to those derived from NCA analysis (Table 3 3). The dissolution rates for BDP and BMP were estimated as 0.4 7 ± 0.1/h and 3.01±1.4/h, respectively. The permeabilities of the central lung for BDP and BMP were estimated as 15 ± 8.3 and 2.92 ± 0.8 ×10 6 cm/s, respectively with those for the peripheral lung (ob tained through scaling according to membrane thickness differences) being 509 and 98.9 ×10 6 cm/s, respectively. With exception of two parameters (K diss of BMP and P app of BDP) r elative standard error s (RSE ) for most PK parameters were less than 20%. Re ceptor occupancy time profiles observed after IV administration of BMP for lung, liver, and kidney w ere fitted to the proposed E max model (Figure 3 3) . The es timates of E max and EC 50 were 85.5% and 0.0017 ng/mL, respectively (Table 3 4) . The RSE for all PD parameters were less or around 30%. Sensitivity A nalysis The eFAST sensitivity indices for t he rat PBPK /PD model regarding receptor occupancy in the lung and kidney are shown in Figure 3 4 . All tested parameters had sensitivity indices smaller than 0.4, in dicating that the model output is insensitive to the variation of the given parameter . Mode l V erification To assess the PBPK model structure and parameter estimates, BDP and BMP concentrations were predicted and compared with experimental BDP and BMP conc entrations in plasma after IV and IT administration of BDP (99) . The model predicted plasma concentrations of BDP and BMP agreed reasonably well with external observations (Figure 3 5).

PAGE 73

73 To validate the pharmacodynamic part of the model (E max and EC 50 ) derived parameters from data after IV administration of BMP, the receptor occupancy time profiles in the tested tissues (lung, liver, and kidney) after IV and IT administration of BDP, and IT administration of BMP (lung is n ot included as lung free BMP concentrations could not be determined)) were predicted (Figure 3 2C, 3 2F, and 3 2L). Considering the experimental variability and model complexity, the predicted receptor occupancy time profiles agreed reasonably well with ob served receptor occupancies in lung, liver, and kidney after IT of BMP, IV and IT administra tion of BDP (Figure 3 6). Evaluation of P ulmonary T argeting To investigate the effect of drug properties on pulmonary targeting, lung and kidney receptor occupanc ies were simulated with different dissolution rates and permeability (Figure 3 7). When diss olution was slow, receptor occupancies in the lung were higher than those in kidney receptor occupancy, indicating pulmonary targeting. However, receptor occupancie s in these three tissue compartments were very similar when dissolution was fast (Figure 3 7 A). Simulations within Monolix ( version 2018R2, LIXOFT, France ) preformed using different permeability values are shown in Figure 7B. Receptor occupancy in the lun g was higher than that in kidney when permeability is lower, indicating low permeability can improve pulmonary targeting. Discussion A better understanding of factors relevant for achieving pulmonary targeting is desirable in facilitating the drug developm ent (18) . Our previous simulations (41) have shown that high systemic clearance, low oral bioavailability and prolonged residence time can improve pulmonary targeting. Our recent study has assessed pulmonary

PAGE 74

74 targeting of a series clinically used corticosteroids and evaluated the importance of clearan ce and factors determining the pulmonary fate for explaining differences across these corticosteroids. A further step in un derstanding the pulmonary fate and performance of inhalation drugs is it to link it to physicochemical properties and performance in subsequent events (deposition, dissolution, metabolism, pharmacodynamic interactions) within the lung. Recently, PBPK appr oaches integrating drug specific properties with lung physiologic characteristics have been used to improve the understanding of pulm onary disposition. Errikson et al (100) have proposed a mechanistic model to investigate epithelial permeability in isolated perf used rat lung. Boger et al (43) h ave developed a physiologically based model that incorporated drug formulation factors within physiological conditions. Those studies have shown that employing experimental data in conjunction with suitable in silico methodologie s would provide insightful understanding of pulmonary targeting. In the present study, a physiologically based PK/PD model was developed to describe prodrug disposition in the lung and link it to pulmonary targeting using receptor occupancy as a biomarker. The receptor occupancy of corticosteroids reflects the pharmacodynamically relevant drug concentrations in the tissues. We had previously developed an ex vivo rece ptor occupancy assay suitable to monitor pharmacodynamic relevant events in a rang e of tissu es, including the lung afte r pulmonary and systemic administration (37) . Using this model and incorporating pharmacokinetics, we could assess the relationship between drug properties and receptor occupancies in dif ferent tissues and further evaluate pulmonary targeting. We

PAGE 75

75 selected BDP, although it was developed more than 50 years ago to learn more about the fate of an inhaled pro drug. BDP, as a prodrug needs to be activated into its ac tive metabolite before it c an initiate the desired effects. This activation can occur in the lung or in the system after absorption from the lung. Thus, for a better understanding, one needs to understand the fate of BDP and BMP both in the lung and the sy stemic circulation. To dete rmine pulmonary activation of BDP, IV administration of BDP and BMP and subsequent analysis of the systemic metabolic events were included in this study as a platform to investigate the pulmonary fate in more detail. NCA analysi s (Table 3 3) showed that t he majority of BDP (95.4%) was converted to BMP after IV administration of BDP. This high conversion percentage is due to its rapid hydrolysis to BMP (101) . After IT administration of BDP, 51.2% of intact BDP was absorbed from the lung into systemic circulation (Table 3 3), indicating that about half of pulmonary deposited BDP was directly absorbed into syste mic circulation before acti vation in the lung. This is in line with a rat study which suggested the extent of biotransformation of BDP in the lung was very limited (99) . However, a clinical study suggests that only 2% of th e inhaled BDP was directly absorbed into system (92) . The observed difference might be to species differences or the higher doses ( 100 µg/kg of bodyweight) administered in the rat experiments. Based on NCA analysis, clearances of BD P and BMP were 3.6 and 0.35 L/h, respectively (Table 3 3). The two values are close to the rate of cardiac output (89) and hepatic bl ood flow (102) , respectively. T h e V dss of BDP and BMP , which were significantly greater t han the volume of total water in the rat (89) , indicate both

PAGE 76

76 compounds were extensively distributed in the body. The half life of BDP and BMP was calculated as 0.52 h and 1.1 h, respectively. The BMP half life was close to the results reported (1.5h) in the literature (99) . The BDP half life was longer than the reported value (0.17h) (99) , probably due to variability of the terminal slope. BDP was eliminated very fast after IV administration and terminal concentr ations were close to limit of quantification of the analytical assay. As a result, the accuracy of BDP half life e stimates were compromised. The developed PB/PK model is the first trying to capture the fate of an inhaled pro drug. During model developmen t, several lung models , differing in the number of compartments but each capturing dissolution, metabolism and dif fusion processes , were tested. The current model structure (Figure 3 1) was selected based on model selection criteria (e.g. AIC, BIC values, goodness of fit), and lung physiology considerations . Two pulmonary compartments with differing absorption rates w ere necessary to describe the plasma and tissue concentrations of BMP (25% were deposited peripherally), while the majority of BDP (99%) was d eposited centrally within one lung compartment (Table 3 4). The terminal slopes of BDP after IV and IT administrat ion differed (Figure 3 2), indicating a flip flop situation due to a slower pulmonary absorption of BDP. Thus, a combined fast and slow absorp tion process was considered, especially for BMP in the model development. Similar model structures were used befor e (97,103) as they refle ct lung physiological features (42) . As expected, most of the drug powder was deposited centrally because drug was introduced directly into the tracheal region regardless of particle size distribution. The permeability estimates f or BDP and BMP were 15 and 2.9 cm/s ×10 6 , respectively

PAGE 77

77 (Table 3 4). The estimates were in line with the permeabilities that were reported for corticosteroids in cell culture (0.6 25 cm/s ×10 6 ) (104) . The higher permeability of BDP when compared with BMP agrees with its higher lipophilicity. The estimated dissolution rates of BDP and BMP were 0.47/h and 3.01/h, respectively (Table 3 4). The results are close to the absorption rates (0.34 and 2.86 /h) that were reported in Sharvari et al study (105) i n which dissolution is assumed to be the rate limiting step. I n contrast, clinical studies have reported a fast absorption of BDP in the human lung (92) . The fast absorption could be due to the commercially available MDI formulations dissolving faster. The current model does not include K muc , be cause it was not identifiable when K muc and k dis were estimated at the same time without additional information, such as particle size distribution and the use of the Noyes Whitney equation inst ead of the simple first order process which were previously us ed to model the dissolution process (105) . Furthermore, the ciliary function was inhibited in anesthetized animal (106,107) . The present PBPK model considered six orga n s: lung, liver, gut, spleen, kidney, and adipose tissues (Figure 3 1). These six organs were included because either they are structurally necessary, or their concentrations are available. Brain was not included because the tissue concentration was not av a ilable. Although receptor occupancies in brain were measured, the data is limited to describe drug distribution in brain when transporters are involved. To validate the PBPK model structure and parameter estimates, plasma concentrations of BDP and BMP aft e r IV and IT administration in rats were predicted. The prediction agreed reasonably well with the external experimental data (99) . To further evaluate the model, AUCs that were calculated using experimental

PAGE 78

78 data were compar e d with the AUCs that were calculated based on simulated plasma concentration time profiles using the proposed PBPK model. Based on the proposed model, 94.0% of BDP was converted to BMP after IV administration, which agrees with the NCA calculation (95.4%) using experimental data (Table 3 3). After IT administration of BDP, 47.1% of BDP was directly absorbed from the lung and 49.8% of pulmonary deposited BDP was converted to BMP. These results are also similar as the values (51.2% and 46.6%) that were repor t ed in Table 3 3. Overall, the developed model structure (global model and pulmonary disposition) was able to describe PK data well. To describe receptor occupancy in tissues, an E max model was used because fast equilibrium between plasma and tissues was a s sumed. E max was estimated as 85.5% (Table 3 4), which is slightly underestimated from 100%, likely to be related to the limited dissociation of BMP from the receptor during the ex vivo incubation step. The EC 50 was estimated as 1.7pg/mL (Table 3 4), which is in the same range of K d values (7.1 pg/mL) reported for fluticasone propionate (FP) in an in vivo receptor binding assay (58) . While we may expect BMP having slightly greater EC 50 value than FP, because the relative receptor binding affinity (RRA) of BMP is smaller than that of FP (RRA BMP = 1345 and RRA FP = 1800 (101,108) ) . D binding might explain this result . To validate our PD model, receptor occupancy time profiles for all scenarios (IV and IT administration of BDP and BMP ) were predicted from PD parameters resulting from fitting only data related to the IV administration of BMP (Figure 3 6). While a good agreement between predicted and observed receptor b inding data was obtained, we compared also the AUC E (area under the receptor occupancy time curves) that were

PAGE 79

79 calculated from NCA a nalysis using experimental data (AUC E,lung = 332 %*h and AUC E,kidney = 252 %*h) with those calculated from predicted occupanc y time profiles (AUC E,lung = 389 %*h and AUC E,kidney = 336 %*h). Pulmonary targeting that was estimated using the proposed PBPK/PD model was only about 30% less than the estimates obtained directly from experimental data. Considering the complexity of the model, the variability and sample size of the data, the model is able to capture the key features in the observations and well expl ained the PK/PD relationship . Simulations have shown that a concentration gradient, achieved by slow dissolution or low perm eability, was obtained to achieve pulmonary targeting (Figure 3 7). Our previous results show that prolonged pulmonary residence ti me could improve pulmonary targeting (41) . As expected, the s low dissolution and low permeability result in a relative higher concentration in the lung than in systemic circulation, as a result, pulmonary targeting was improved. The simulation exercise on low permeability was theoretically explored to investigate th e impact of drug properties on pulmonary targeting. Overall, this study demonstrated the value of mechanistically based appro ach to describe pulmonary disposition within physiologically relevant condition and reinforced the importance of dissolution and pe rmeability for the permeance of an inhaled corticosteroid. Summary We reported for the first time a physiologically based PK/PD model with lung disposition for a corticosteroid prodrug and its active metabolite. The model can describe plasma concentrations , tissue concentrations and receptor occupancies reasonably well after two administration routes of the prodrug and its active metabolite. The simulated PK/PD profiles agreed well with the observed external concentrations

PAGE 80

80 and experimental receptor occupanc y data . Simulations have revealed the relationship between drug properties (e.g. dissolution and permeability) and p ulmonary targeting. Thus, the model can facilitate drug development of inhaled corticosteroids , also have applications in the assessment of pulmonary targeting.

PAGE 81

81 Table 3 1 . Physiological parameters for modeling in rats Tissue Volume (fraction of BW) Blood flow (fraction of Q co ) Lung 0.005 a 0.021/1 a, e Gut 0.027 b 0.14 b Spleen 0.002 a 0.0715 d Liver 0.04 0 a 0.183 a Kidney 0 .0073 a 0.14 a Adipose tissue 0.16 a 0.09 c Rapidly perfused tissue 0.039 c 0.3096 c Slowly perfused tissue 1 rem ai nder 1 rem ai nder Arterial blood 0.02 a NA Venous blood 0.04 a NA Note: NA, not available ; Qco = cardiac output, 26.5 L/h/kg (89) , ref c. (109) ; a: ref (89) ; b: ref (43) ; d: ref (90) Rapidly perfused tissues include richly perfused + heart ; Slowly perfused tissues = 1 remainder ; e blood flow of central and peripheral region

PAGE 82

82 Table 3 2 . Tissue plasma partition coefficients for tissues in the model a Tissue / Compound BDP BMP Lung 3.88 1.87 d Gut 4.65 1.7 1 Spleen 1.86 1.18 Liver 3.41 3. 51 d Kidney 3.1 0 1.45 d Adipose tissue 38.9 4.27 R apidly perfused tissue b 5.30 2.03 Slowly perfused tissue c 3.02 1.2 0 Note: a Kp was calculated by Rodger & Rowland method (91) except for lung, l iver, and kidney for BMP ; b kp of richly perfused tissue is the kp in heart ; c kp of poorly perfused tissu e is the mean kp of muscle and bone marrow ; d kp was estimated by fitting plasma and concentration after IV of BMP

PAGE 83

83 Table 3 3 . The estimates of PK parameters for BDP and BMP in rats using noncompartmental analysis --95.4 a 51.2 b 46.6 c 100 Note: a 95.4% of BDP was converted to BMP after IV administration of BDP; b 51.2% of BDP was directly absorbed and c 46.6% of BDP was abso rbed as BMP after IT administration of BDP.

PAGE 84

84 Table 3 4 . The estimates of PK/PD parameters for BDP and BMP in rats using the proposed model PK Parameter Units Values RSE (%) K p, lung, BMP 1.87 17.8 K p, l iver , BMP 3. 51 1 7.7 K p, kidney , BMP 1.45 17.4 BMP clearance L/h 0. 4 4 17.8 Metabolic c learance (BDP > BMP) L/h 1.6 18.0 P app, central lung, BDP cm/s ( x 10 6 ) 15 55.4 P app, peripheral lung, BDP cm/s ( x 10 6 ) 509 38.4 P app, central lung, B MP cm/s ( x 10 6 ) 2.92 27.3 P app, peripheral lung, BDP cm/s ( x 10 6 ) 9 8 .8 26.3 Dissolution rate of BDP 1/h 0. 47 1 9.1 Dissolution rate of BMP 1/h 3.01 45.1 Fraction deposition to central region for BDP 0.99 3.6 Fraction deposition to central region for BM P 0.76 5.99 PD Parameters Units Value RSE (%) E max % 85.5 17.7 EC 50 ng/mL 0.0017 33.5 Note: Kp is the tissue partition coefficient ; P app is the permeability, calculated based on permeability of the central lung

PAGE 85

85 Figure 3 1 . Schematic r epresentation of a physiologically based pharmacokinetic model for BDP and BMP after intravenous (IV) or intratracheal (IT) administration (A), compartmental representation of lung disposition (B), and PD model to describe the relationship between receptor occupancy and free drug concentration in tissues (C). Note: Fc: fraction of t he pulmonary dose that is deposited in the central region of the lung; Q br and Q co : bronchial blood flow and cardiac output, respectively; Q g , Q li , Q sp , Q kid , Q adi : blood flow i n gut, liver, spleen, kidney, adipose tissues, respectively; Q ra : blood flow in rapidly perfused tissues (mainly brain and heart); Q sl : blood flow in slowly perfused tissues (mainly skin, muscle, and bone); CL met : intrinsic clearance for BDP conversion to BMP; CL total clearance of BMP; Solid: BDP or BMP undissolved particles that de posited in the lung; K dis : first order dissolution rate ELF: epithelial lining fluid; P app : permeability; S: surface area; C: concentration between ELF and lung tissue; R max : t he maximum observed receptor occupancy; E 0 : receptor occupancy when no tested drug present; EC 50 : free drug concentration corresponding to the half maximal difference between R max and E 0 ; C u : free drug concentration at the site of action. Process 1 repre sents the dissolution of solid BDP; process 2 represents the diffusion of dissolved BDP between ELF and lung tissue; process 3 presents perfusion between lung tissue and bloodstream .

PAGE 86

86 Figure 3 2 . BDP and BMP concentration in plasma (A, D, G, J), BMP c oncentration in tissues (B,E, H, K), and receptor occupancy in tissues (C, F, I, L) after IV administration of BDP (A C), IT administration of BDP (D F), IV administration of BMP (G I), and IT administration of BMP (J L). Each point and vertical bar rep resent the mean and SD of the observations at a given time poin t (n=3) , respectively. *data were not included in the model development but used for model validation.

PAGE 87

87 Figure 3 3 . Observed (dots) and predicted (lines) concentrations (Conc.) and receptor occupancies (Occ.) in plasma and various tissues in rats. Note: Each dot represents the mean data of three independent animal s . Solid lines are the fit predictions based on the pro posed PBPK/PD model.

PAGE 88

88 Figure 3 4. Sensitivity analysis . The total effect on lung receptor occupancy (A) and kidney receptor occupancy (B) with ten different parameters given as a proportion of variance when IT administration of BDP was given. Note: The following parameters were included in the analysis: Kp_lu n g_BMP: lung to plasma partition coefficient of BMP; CL: total clearance of BMP; K dis : dissolution rate of BDP; BW: body weight; CL met : metabolic clearance of BDP conversion to BMP; P app : permeab ility of BDP; Kp_kid_BMP: kidney to plasma partition coeffici e nt of BMP; Kp_lung_BDP: lung to plasma partition coefficient of BDP; Kp_kid_BDP: kidney to plasma partition coefficient of BDP; Fc: fraction of the pulmonary dose that is deposited in the centra l region of the lung.

PAGE 89

89 Figure 3 5 . O bserved (dot s) and PBPK model predicted (lines) plasma concentrations of BDP and BMP after IV (A) and IT (B) administration of BDP in rats. Measured BDP and BMP concentrations were digitized from Chanoine et al. (99)

PAGE 90

90 Figure 3 6 . O bserved (dots) and PBPK/PD model predicted (lines) receptor occupancies (lines) after IV administration of BDP (A C), IT administration of BDP (D F), and IT administration of BMP (G I). Observed receptor occupancies in the tissu es were measured in this study (se e Method section).

PAGE 91

91 Figure 3 7 . Simulation of receptor occupancy in the lung and kidney with different dissolution rate (0.4 and 4 /h) and permeability (0. 15 and 15 *10 6 cm/s) after I T of BDP in rats .

PAGE 92

92 CHAPTER 4 CONCLUSION The goal of inhalation therapy is to achieve distinct pulmonary effects without inducing systemic side effects (pulmonary targeting) . Assessing pulmonary targeting is therefore vital duri ng drug candidate selection in early drug development. Within the first part of this thesis, w e evaluated pulmonary target ing of a series of commerc ia lly available corticosteroid s : BDP, BMP, BUD, CIC, desCIC, FP, and TA . A fter IV and IT administration th e receptor occupan cies in tissues differ and are affected by transporter s , organ specific metabolic events , and cell trafficking. The present study provides a quantitative assessment of pulmonary targeting of inhaled corticosteroids and demonstrates differ ences across comme rcially available corticosteroids for both pulmonary as well as systemic drug exposure. Furthermor e, w e reported for the first time a physiologically based PK/PD model with lung disposition for a corticosteroid prodrug and its active met abolite. The model can describe plasma concentrations, tissue concentrations and receptor occupancies reasonably well after two administration routes of the prodrug and its active metabolite. The simulated PK/PD profiles agreed well with the observed exter nal concentrations and experimental receptor occupanc y data . Simulations have revealed the relationship between drug properties (e.g. dissolution and permeability) and pulmonary targeting. Overall, this work reveals favorable drug properties of inhaled co rticosteroids. And the developed model can be used for drug design and assessment of pulmonary targeting.

PAGE 93

93 APPENDIX A TABLES AND FIGURES FOR RECETPR OCCUPANCY ANALYSIS Receptor O ccupancy after I ntravenous (IV) A dministration of T ested I nhale d C orticoste roids in R ats. The M ean ± SD I s G iven. ICSs Receptor occupancy (% *h) AUC E,lung AUC E,kidney AUC E,liver AUC E,spleen AUC E,brain BDP (n=3) 309 ± 77 339 ± 35 405 ± 58 --123 ± 89 BMP (n=3) 444 ± 15 443 ± 25 474 ± 13 --88 ± 32 BUD (n=3) 293 ± 86 316 ± 9 6 284 ± 101 296 ± 103 74 ± 61 CIC (n=6) 174 ± 128 111 ± 105 92 ± 95 269 ± 105 63 ± 74 Des CIC (n=6) 209 ± 133 166 ± 97 134 ± 97 220 ± 112 69 ± 66 FP (n=6) 339 ± 126 290 ± 60 133 ± 94 342 ± 51 75 ± 28 TA (n=3) 272 ± 107 --253 ± 126 ----

PAGE 94

94 Receptor O ccupancy after I ntratracheal (IT) A dministration of T ested I nhaled C orticosteroids in R ats . The M ean ± SD I s G iven . ICSs Receptor occupancy (% *h) AUC E,lung AUC E,kidney AUC E,liver AUC E,spleen AUC E,brain BDP (n =3) 332 ± 99 252 ± 62 43 0 ± 22 --169 ±81 BMP (n=3) 401 ± 49 300 ± 54 345 ± 25 --22 ± 38 BUD (n=6) 203 ± 105 170 ± 83 154 ± 84 203 ± 72 68 ± 65 CIC (n=6) 248 ± 97 107 ± 89 83 ± 79 242 ± 88 48 ± 69 Des CIC (n=6) 250 ± 112 147 ± 96 90 ± 94 21 6 ± 104 9 9 ± 79 FP (n=9) 364 ± 8 7 221 ± 72 131 ± 4 6 326 ± 65 38 ± 61 TA (n=9) 455 ± 69 373 ± 52 408 ± 65 439 ± 50 34 ± 50

PAGE 95

95 Receptor O ccupancy T ime P rofiles for BDP after S ystemic (IV) and P ulmonary (IT) A dministration. The M ean ± SD I s G iven.

PAGE 96

96 Receptor Occupancy Time Profiles for B M P after Systemic (IV) and Pulmonary (IT) Administration. The Mean ± SD Is Given.

PAGE 97

97 Receptor Occupancy Time Profiles for B UD after Systemic (IV) and Pulmonary (IT) Administration. The Mean ± SD Is Given.

PAGE 98

98 Receptor O ccupancy T ime P rofiles f or CIC after S ystemic (IV) and P ulmonary (IT) A dministration. The M ean ± SD I s G iven.

PAGE 99

99 Receptor O ccupancy T ime P rofiles for desCIC after S ystemic (IV) and P ulmonary (IT) A dministration. The M ean ± SD I s G iven.

PAGE 100

100 Receptor O cc upancy T ime P rof iles for TA after S ystemic (IV) and P ulmonary (IT) A dministration. The M ean ± SD I s G iven.

PAGE 101

101 APPENDIX B MONOLIX CODE FOR PBPK MODEL AFTER IT ADMINISTRATION OF BDP DESCRIPTION: The administration is I T of BDP. The PK m odel contains l ung, liver, kidney, spleen, gut, and adipose. [LONGITUDINAL] input = {Pc, kdiss, Fc} PK: depot(adm=1, target = APsclung,p = Fc) depot(adm=2, target = APsplung,p = 1 Fc) EQUATION: odeType = stiff ;===== Input parameters ===== BW = 0.23 ;kg or L, average rat 230g, assume desity 1g/cm3 Qco = 26.5 * BW ;cardiac output CLM = 0.439 ;L/hr CLmet= 1.6 ;L/hr, BDP conv er t to BMP R = 1 ;blood/plasma ratio fupp = 0.13 ;BDP, fraction unbound in plasma fupm = 0.016 ;BMP, fraction unbound in plasma ffu =1 ;fraction unbound in ELF ;===== organ (compartment) volume ===== Vlung = 0.005 * BW ; fraction of BW, lung, L Vlive = 0.04 * BW ; fraction of BW, liver Vkid =0.0073 * BW ; fr acti on of BW, kidney Vspl = 0.002 * BW ; fraction of BW, spleen Vgut = 0.027 * BW ; fraction of BW Vadip = 0.16 * BW ; fraction of BW Vart = 0.02 * BW ; fraction of BW, arterial blood Vven = 0.04 * BW ; fraction of BW, venous blood Vrich = 0.039 * BW Vall = 0.005 + 0.04 + 0.0073 + 0.002 + 0.027 + 0.16 + 0.02 + 0.04 + 0.039 ; fraction of the rest Vpoor = (1 Vall) * BW ; 1 the r est, poorly perfused ;===== blood flow between organs (compartments) ===== Qclung= 0.021 * Qco ; fraction of Q_CO Qplung= 1 * Qco Qlive = 0.183 * Qco ; fraction of Q_CO, liver Qkid = 0.1 4 * Qco ; fraction of Q_CO, kidney Qspl = 0.0715 * Qco ; fraction of Q_CO, spleen Qgut = 0.14 * Qco ; fraction of Q_CO Qadip = 0.09 * Qco ; fraction of Q_CO

PAGE 102

102 Qrich = 0.30 96 * Qco Qall = 0.0 21 + 0.183 + 0.14 + 0.0715 + 0.14 + 0.09 + 0.3096 ; fraction of all Qpoor =(1 Qall) * Qco ; 1 the rest, poorly perfused ;===== tissue to plasma partition co efficients ===== KPPlung = 3.88 KPPlive = 3.41 KPPkid = 3.1 KPPspl = 1. 86 KPPgut = 4.65 KPPadip = 38.9 KPPrich = 5.3 KPPpoor = 3.02 KPMlung = 1.87 KPMl ive = 3.51 KPMkid = 1.45 KPMspl = 1.18 KPMgut = 1.71 KPMadip = 4.27 KPMrich = 2.03 KPMpoor = 1.2 ;===== rat lung related parameters ===== ;Papp dm/hr = 10cm/36 00s ;1L = 1dm3 Sclung = 3.27 * BW ; surface area, dm2/kg Splung = 276.4 * BW ; surface area, dm2/kg Vfplung= 1.636e 4 * BW ; L, 163.6 uL/kg Vfclung= 1.935e 4 * BW ; L, 193.5 uL/kg Vclung = 0.19 * Vlung ; fr action of tissu e vo lume Vplung = 0.81 * Vlung ; fraction of tissue volume hclung = 13 hplung = 0.38 ;um epithelim thickness s = hclung/hplung Pp = P c * hclung/hplung ;Papp in peripheral ;===== definition of variables in c ompartments === == ; Conc in ng/mL CPfplung = APfplung/Vfplung CPfclung = APfclung/Vfclung CPtplung = APtplu ng/ Vplung CPtclung = APtclung/ Vclung CPkid = APkid / Vkid CPadip = APadip / Vadip CPspl = APspl / Vspl CPgut = APgut / Vgut

PAGE 103

103 CPlive = APlive / Vliv e CPart = APart / V art CPven = APven / Vven CPrich = APrich / Vrich CPpoor = APpoor / Vpoor CMtplung= AM tplung/ Vplung CMtclung= AMtclung/ Vclung CMlung = (AMtclung + AMtplung) / Vlung CMkid = AMkid / Vkid CMadip = AMadip / Vadip CMspl = AMspl / Vspl CMgut = AMgut / Vgu t CMlive = AMlive / Vlive CMart = AMart / Vart CMven = AMven / Vven CMrich = AMrich / Vrich CMpoor = AMpoor / Vpoor ;===== ODEs initial values ===== t_0 = 0 APsplung_0= 0 ;AP parent drug APsclung_0= 0 APfplung_0= 0 APfclung_0= 0 APtplu ng_0= 0 APtc lung _0= 0 APart_0 = 0 APkid_0 = 0 APadip_0 = 0 APrich_0 = 0 APpoor_0 = 0 APven_0 = 0 APspl_ 0 = 0 APgut_0 = 0 APlive_0 = 0 AMplung_0 = 0 ;AM metabolite AMclung_0 = 0 AMart_0 = 0 AMkid_0 = 0 AMadip_0 = 0 AMrich_0 = 0 AMpoor_0 = 0

PAGE 104

104 AMven_0 = 0 AMspl_0 = 0 AMg ut_0 = 0 AMlive_0 = 0 ;===== ODEs ===== ddt_APsclung= kdiss * APsclung ddt_APsplung= kd iss * APsplung ddt_APfclung= kdi ss * APsclung Pc * Sclung *(CPfclung*ffu CPtclung*fupp/KPPlung) ddt_APfplung= kdiss * APsplung Pp * Splung *(CPf plung*ffu CPt plun g*fupp/KPPlung) ddt_APtclung= Qclung* CPart Qclung*R* CPtclung/KPPlung CLmet * fupp * CPtclung/KPPlung + Pc * Sclun g *(CPfclung*ffu CPtclung*fupp/KPPlung) ddt_APtplung= Qplung* CPven Qplung*R* CPtplung/KPPlung CLmet * fupp * CP tplung/KPPlung + Pp * Splung *(CPfplung*ffu CPtplung*fupp/KPPlung) ddt_APart = Qplung* CPart + Qplung*R* CPtplung/KPPlung CLmet * f upp * CPart ddt_APkid = Qkid * CPart Qkid *R* CPkid / KPPkid CLmet * fupp * CPkid / KPPkid ddt_APadip = Qadip * CP art Qadip *R* CPa dip / KPPadip CLmet * fupp * CPadip / KPPadip ddt _APrich = Qrich * CPart Qrich *R* CPrich / KPPrich CLmet * fupp * CPrich / KPPrich ddt_APpoor = Qpoor * CPart Qpoor *R* CPpoor / KPPpoor CLmet * fupp * CPpoor / KPPpoor ddt_APven = Qkid * CPve n + Qkid *R* CPkid / KPPkid Qadip * CPven + Qadip *R* CPadip / KPPadip Qrich * CPven + Qrich *R* CPrich / KPPrich Qpo or * CPven + Qpoor *R* CPpoor / KPPpoor Qclung * CPven + Qclung*R* CPtclung/KPPlung Qspl*CPven Qgut*CPven Qlive*CPven +(Qlive +Qsp l+Qgut)*R*CPlive/KPPlive CLmet * fupp * CPven ddt _APspl = Qspl * CPart Qspl *R* CPspl / KPPspl CLmet * fupp * CPs pl/ KPPspl ddt_APgut = Qgut * CPart Qgut *R* CPgut / KPPgut CLmet * fupp * CPgut/ KPPgut ddt_APlive = Qlive * CPart + Qgut *R* CP gut / KPPgut + Qspl *R* CPspl / KPPspl (Qlive+Qspl+Qgu t)*R*CPlive/KPPlive CLmet * fupp * CPlive/ KPPlive ddt_AMtplung= Qplung* CMven Qplung*R* CMtplung/KPMlung + CLmet * fupp * CPtplung/KPPlung ddt_AMtclung= Qclung* CMart Qclung*R* CMtclung/KPMlun g + CLmet * fupp * CPtclung/KPPlung ddt_AMart = Qplu ng* CMart + Qplung*R* CMtplung/KPMlung + CLmet * fupp * CPart ddt_AM kid = Qkid * CMart Qkid *R* CMkid / KPMkid + CLmet * fupp * CPkid / KPPkid d dt_AMadip = Qadip * CMart Qadip *R* CMadip / KPMadip + CLmet * fupp * CPadip / KPPadip ddt_AMrich = Qrich * CMart Qrich *R* CMrich / KPMrich + CLmet * fupp * CPr ich / KPPrich ddt_AMpoor = Qpoor * CMart Qpoor *R* CMpoor / KPMpoor + CLmet * fupp * CPpo or / KPPpoor ddt_AMven = Qkid * CMven + Qkid *R* CMkid / KPMkid Qadip * CMven + Qadip *R* CMadip / KPMadip Qrich * CMven + Qrich *R* CMrich / KPMrich Qpoor * CMven +

PAGE 105

105 Qpoor *R* CMpoor / KPMpoor Qclung* CMven + Qclung*R* CMtclung/KPMlung Qspl*CMve n Qgut*CMven Qlive*CMven +(Qlive+Qspl+Qgut)*R*CMlive/KPMlive + CLmet * fupp * CPven CLM*CMven ddt_AMspl = Qspl * CMart Qspl *R* CMspl / KPMspl + CLmet * fup p * CPspl/ KP Pspl ddt_AMgut = Qgut * CMart Qgut *R* CMgut / KPMgut + CLmet * fupp * CPgut / KPPgut ddt_AMlive = Qlive * CMart + Qgut *R* CMgut / KPMgut + Qspl *R* CMspl / KPMspl (Qlive+Qspl+Qgut)*R*C Mlive/KPMlive + CLmet * fupp * CPlive/ KPPlive OUTPU T: output = { CPven, CMven, CMlung, CMlive, CMkid}

PAGE 106

106 LIST OF REFERENCES 1. Stein SW, Thiel CG. The History of Therapeutic Aerosols: A Chronological Review. J Aerosol Med Pulm Drug Deliv. 2017 Feb;30(1):20 41. 2. The global asthma rep ort 2018 [Internet]. Available from: http://www.globalasthmareport.org/ 3. Asthma [Internet]. Centers for Disease Control and Prevention; Available from: https://www.cdc.gov/asthma/default.htm 4. Expert Panel Report3: Guidelines for the Diagnosis and Man agement of Asthma. National Heart, Lung, and Blood Institute; 5. Kudo M, Ishigatsubo Y, Aoki I. Pathology of asthma. Front Microbiol. 2013 Sep 10;4:263. 6. Cohn L, Elias JA, Chupp GL. Asthma: mechanisms of disease persistence and p rogression. Annu Rev Immunol. 2004;22:789 815. 7. Kuipers H, Lambrecht BN. The interplay of dendritic cells, Th2 cells and regulatory T cells in asthma. Curr Opin Immunol. 2004 Dec;16(6):702 8. 8. Walker JA, McKenzie ANJ. TH2 cell development and funct ion. Nat Rev Immunol. 2018;18(2):121 33. 9. Fahy JV. Type 2 inflammation in asthma -present in most, absent in many. Nat Rev Immunol. 2015 Jan;15(1):57 65. 10. Zhang S, Smartt H, Holgate ST, Roche WR. Growth factors secreted by bronchial epithelial cel ls control myofibrobl ast proliferation: an in vitro co culture model of airway remodeling in asthma. Lab Investig J Tech Methods Pathol. 1999 Apr;79(4):395 405. 11. Chiappara G, Gagliardo R, Siena A, Bonsignore MR, Bousquet J, Bonsignore G, et al. Airway remodelling in the p athogenesis of asthma. Curr Opin Allergy Clin Immunol. 2001 Feb;1(1):85 93. 12. Charles L. Schauf, David Moffett, Stacie Moffett. Human physiology: foundations & frontiers. William C Brown Pub; 1993. 851 p. 13. Weibel ER. Morphomet ry of the Human Lung [Internet]. Berlin Heidelberg: Springer Verlag; 1963 [cited 2019 Jul 20]. Available from: https://www.springer.com/gp/book/9783642875557 14. Patton JS. Mechanisms of macromolecule absorption by the lungs. Adv Drug Deliv Rev. 1996 Apr 30;19(1):3 36.

PAGE 107

107 15. Ng AW, Bidani A, Heming TA. Innate host defense of the lung: effects of lung lining fluid pH. Lung. 2004;182(5):297 317. 16. Valeyre D, Soler P, Basset G, Loiseau P, Pre J, Turbie P, et al. Glucose, K+, and albumin concentrations in the alveolar milieu o f normal humans and pulmonary sarcoidosis patients. Am Rev Respir Dis. 1991 May;143(5 Pt 1):1096 101. 17. Miller FJ, Mercer RR, Crapo JD. Lower Respiratory Tract Structure of Laboratory Animals and Humans: Dosimetry Implications. Aer osol Sci Technol. 199 3 Jan 1;18(3):257 71. 18. Stefanie K Drescher, Mong Jen Chen, Jürgen B Bulitta, Günther Hochhaus. Pharmacokinetics and Pharmacodynamics of Drugs Delivered to the Lung. In: anthony J. Hickey, Sandro R. da Rocha, editor. Pharmaceutical Inhalation Aerosol T echnology, Third Edition. CRC Press; 2019. 19. Chrystyn H. Methods to identify drug deposition in the lungs following inhalation. Br J Clin Pharmacol. 2001 Apr;51(4):289 99. 20. Labiris NR, Dolovich MB. Pulmonary drug delivery. Par t I: physiological fa ctors affecting therapeutic effectiveness of aerosolized medications. Br J Clin Pharmacol. 2003 Dec;56(6):588 99. 21. Thorsson L, Geller D. Factors guiding the choice of delivery device for inhaled corticosteroids in the long term ma nagement of stable as thma and COPD: focus on budesonide. Respir Med. 2005 Jul;99(7):836 49. 22. Edsbäcker S, Johansson C J. Airway selectivity: an update of pharmacokinetic factors affecting local and systemic disposition of inhaled steroids. Basic Clin Pharmacol Toxicol. 20 06 Jun;98(6):523 36. 23. Darquenne C, Prisk GK. Aerosol deposition in the human respiratory tract breathing air and 80:20 heliox. J Aerosol Med Off J Int Soc Aerosols Med. 2004;17(3):278 85. 24. Thompson PJ. Drug delivery to the sm all airways. Am J Res pir Crit Care Med. 1998 May;157(5 Pt 2):S199 202. 25. Borghardt JM, Weber B, Staab A, Kloft C. Pharmacometric Models for Characterizing the Pharmacokinetics of Orally Inhaled Drugs. AAPS J. 2015 Jul;17(4):853 70. 26. Bustamante Mar in XM, Ostrowski LE. Cilia and Mucociliary Clearance. Cold Spring Harb Perspect Biol. 2017 Apr 3;9(4). 27. Foster WM, Langenback E, Bergofsky EH. Measurement of tracheal and bronchial mucus velocities in man: relation to lung clearance. J Appl Physiol. 1 980 Jun;48(6):965 71.

PAGE 108

108 28. 7. 29. inhaled drug discovery and development: Induced alveolar macrop hage responses. Adv D rug Deliv Rev. 2014 May;71:15 33. 30. Kirby AC, Coles MC, Kaye PM. Alveolar macrophages transport pathogens to lung draining lymph nodes. J Immunol Baltim Md 1950. 2009 Aug 1;183(3):1983 9. 31. Camberlein E, Cohen JM, José R, Hyams CJ, Callard R, Chima lapati S, et al. Importance of bacterial replication and alveolar macrophage independent c learance mechanisms during early lung infection with Streptococcus pneumoniae. Infect Immun. 2015 Mar;83(3):1181 9. 32. Patton JS, Fishburn CS, Weers JG. The lungs as a portal of entry for systemic drug delivery. Proc Am Thorac Soc. 2004;1(4):338 44. 33 . Barnes PJ. Inhaled Corticosteroids. Pharm Basel Switz. 2010 Mar 8;3(3):514 40. 34. Smith DF, Toft DO. Steroid receptors and their associated proteins. Mol Endocr inol Baltim Md. 1993 Jan;7(1):4 11. 35. Okamoto K, Hirano H, Isohashi F. Molecular cloni ng of rat liver glucocorticoid receptor translocation promoter. Biochem Biophys Res Commun. 1993 Jun 30;193(3):848 54. 36. Isohashi F, Okamoto K. ATP stimulated tra nslocation promoter that enhances the nuclear binding of activated glucocorticoid receptor complex. Biochemical properties and its function (mini review). Receptor. 1993;3(2):113 24. 37. Hochhaus G, Gonzalez Rothi RJ, Lukyanov A, Derendorf H, Schreier H, Dalla Costa T. Assessment of glucocorticoid lung targeting by ex vivo receptor binding st udies in rats. Pharm Res. 1995 Jan;12(1):134 7. 38. Burton JA, Schanker LS. Absorption of antibiotics from the rat lung. Proc Soc Exp Biol Med Soc Exp Biol Med N Y N. 1974 Mar;145(3):752 6. 39. Borghardt JM, Kloft C, Sharma A. Inhaled Therapy in Respir atory Disease: The Complex Interplay of Pulmonary Kinetic Processes. Can Respir J. 2018;2018:2732017. 40. van den Brink KIM, Boorsma M, Staal van den Brekel AJ, Eds bäcker S, Wouters EF, Thorsson L. Evidence of the in vivo esterification of budesonide in human airways. Br J Clin Pharmacol. 2008 Jul;66(1):27 35.

PAGE 109

109 41. Hochhaus G, Möllmann H, Derendorf H, Gonzalez Rothi RJ. Pharmacokinetic/Pharmacodynamic Aspects of Aer osol Therapy using Glucocorticoids as a Model. J Clin Pharmacol. 1997 Oct;37(10):881 92. 42. Hastedt JE, Bäckman P, Clark AR, Doub W, Hickey A, Hochhaus G, et al. Scope and relevance of a pulmonary biopharmaceutical classification system AAPS/FDA/USP Wor kshop March 16 17th, 2015 in Baltimore, MD. AAPS Open. 2016 Jan 27;2(1):1. 43. Boger E, Evans N, Chappell M, Lundqvist A, Ewing P, Wigenborg A, et al. Systems Pharmacology Approach for Prediction of Pulmonary and Systemic Pharmacokinetics and Receptor Oc cupancy of Inhaled Drugs. CPT Pharmacomet Syst Pharmacol. 2016;5(4):201 10. 44. Boucher RC. Human airway ion transport. Part one. Am J Respir Crit Care Med. 1994 Jul;150(1):271 81. 45. Mercer RR, Russell ML, Roggli VL, Crapo JD. Cell number and distrib ution in human and rat airways. Am J Respir Cell Mol Biol. 1994 Jun;10(6):613 24. 46. Yo neda K. Mucous blanket of rat bronchus: an ultrastructural study. Am Rev Respir Dis. 1976 Nov;114(5):837 42. 47. Richards ML, Sadée W. In vivo opiate receptor bindi ng of oripavines to mu, delta and kappa sites in rat brain as determined by an ex vivo lab eling method. Eur J Pharmacol. 1985 Aug 27;114(3):343 53. 48. Maynard RL. Comparative Biology of the Normal Lung (treatise on pulmonary toxicology, volume 1). Occup Environ Med. 1995 Nov;52(11):783. 49. Yeh HC, Schum GM, Duggan MT. Anatomic models of t he tracheobronchial and pulmonary regions of the rat. Anat Rec. 1979 Nov;195(3):483 92. 50. Beato M, Kalimi M, Feigelson P. Correlation between glucocorticoid bindi ng to specific liver cytosol receptors and enzyme induction in vivo. Biochem Biophys Res C ommun. 1972 Jun 28;47(6):1464 72. 51. Mager H, Göller G. Resampling methods in sparse sampling situations in preclinical pharmacokinetic studies. J Pharm Sci. 1998 Mar;87(3):372 8. 52. Bi Y, Deng J, Murry DJ, An G. A Whole Body Physiologically Based Ph armacokinetic Model of Gefitinib in Mice and Scale Up to Humans. AAPS J. 2016 Jan;18(1):228 38. 53. mediated pharmacod ynamics of prednisolone in the rat. J Pharmacokinet Biopharm. 1986 Oct 1;14(5):469 93.

PAGE 110

110 54 . Le S, Gruner JA, Mathiasen JR, Marino MJ, Schaffhauser . Correlation between ex vivo receptor occupancy and wakepromoting activity of selective H3 receptor antag onists. J Pharmacol Exp Ther. 2008;902 909. 55. Hochhaus G, Rohdewald P, Möllmann H, Gre schuchna D. Identification of glucocorticoid receptors in normal and neoplastic adult human lung. Res Exp Med Z Gesamte Exp Med Einschl Exp Chir. 1983;182(1):71 8. 5 6. Ballard PL, Baxter JD, Higgins SJ, Rousseau GG, Tomkins GM. General presence of glucoc orticoid receptors in mammalian tissues. Endocrinology. 1974 Apr;94(4):998 1002. 57. Pujols L, Mullol J, Roca Ferrer J, Torrego A, Xaubet A, Cidlowski JA, et al. Ex isoforms in human cells and tissues. Am J Ph ysiol Cell Physiol. 2002 Oct 1;283(4):C1324 31. 58. Boger E, Ewing P, Eriksson UG, Fihn B M, Chappell M, Evans N, et al. A novel in vivo receptor occupancy methodol ogy for the glucocorticoid receptor: toward an improved understanding of lung pharmacokinetic/pharmacodynamic relationships. J Pharmacol Exp The r. 2015 May;353(2):279 87. 59. Uller L, Persson CG, Källström L, Erjefält JS. Lung tissue eosinophils may be c leared through luminal entry rather than apoptosis: effects of steroid treatment. Am J Respir Crit Care Med. 2001 Nov 15;164(10 Pt 1):1948 56. 60. Arya V, Issar M, Wang Y, Talton JD, Hochhaus G. Brain permeability of inhaled corticosteroids. J Pharm Phar macol. 2005 Sep;57(9):1159 67. 61. Stoeck M, Riedel R, Hochhaus G, Häfner D, Masso JM, Schmidt B, et al. In vitro and in vivo anti inflammator y activity of the new glucocorticoid ciclesonide. J Pharmacol Exp Ther. 2004 Apr;309(1):249 58. 62. Zaidi S, C hen M J, Lee DT, Neubart E, Ewing P, Miller Larsson A, et al. Fetal Concentrations of Budesonide and Fluticasone Propionate: a Study in Mice. AA PS J. 2019 Apr 16;21(4):53. 63. Arredouani MS, Palecanda A, Koziel H, Huang Y C, Imrich A, Sulahian TH, et al. MARCO is the major binding receptor for unopsonized particles and bacteria on human alveolar macrophages. J Immunol Baltim Md 1950. 2005 Nov 1; 175(9):6058 64. 64. Patel VI, Metcalf JP. Airway Macrophage and Dendritic Cell Subsets in the Resting Human Lu ng. Crit Rev Immunol. 2018;38(4):303 31. 65. Vogel DYS, Heijnen PDAM, Breur M, de Vries HE, Tool ATJ, Amor S, et al. Macrophages migrate in an activation dependent manner to chemokines involved in neuroinflammation. J Neuroinflammation. 2014 Feb 1;11:23.

PAGE 111

111 66. Grayson MH, Ramos MS, Rohlfing MM, Kitchens R, Wang HD, Gould A, et al. Controls for Lung Dendritic Cell Maturation and Migration during Respiratory Viral Infection. J Immunol. 2007 Aug 1;179(3):1438 48. 67. Edsbäcker S, Brattsand R. Budesonide fa tty acid esterification: a novel mechanism prolonging binding to airway tissue. Review of available data. Ann Allergy Asthma Immunol Off Publ Am Coll Allergy Asthma Immunol. 2002 Jun;88(6):609 16. 68. Choy YB, Prausnitz MR. The rule of five for non oral routes of drug delivery: ophthalmic, inhalation and transdermal. Pharm Res. 2011 May;28(5):943 8. 69. Derendorf H, Hochhaus G, Meibohm B, Möll mann H, Barth J. Pharmacokinetics and pharmacodynamics of inhaled corticosteroids. J Allergy Clin Immunol. 1998 Apr;101(4 Pt 2):S440 446. 70. Sakagami M, Kinoshita W, Sakon K, Sato J, Makino Y. Mucoadhesive beclomethasone microspheres for powder inhalati on: their pharmacokinetics and pharmacodynamics evaluation. J Control Release Off J Control Release Soc. 2002 Ap r 23;80(1 3):207 18. 71. Boobis AR. Comparative physicochemical and pharmacokinetic profiles of inhaled beclomethasone dipropionate and budeso nide. Respir Med. 1998 Jul;92 Suppl B:2 6. 72. van Amerongen IA, De Ronde HAG, Klooster NTM. Physical chemical characterization of semisolid topical dosage form using a new dissolution system. Int J Pharm. 1992 Oct 10;86(1):9 15. 73. Feth MP, Volz J, H ess U, Sturm E, Hummel R P. Physicochemical, crystallographic, thermal, and spectroscopic behavior of crystallin e and X ray amorphous ciclesonide. J Pharm Sci. 2008 Sep;97(9):3765 80. 74. Jain N, Yalkowsky SH. Estimation of the aqueous solubility I: appl ication to organic nonelectrolytes. J Pharm Sci. 2001 Feb;90(2):234 52. 75. Högger P, Rohdewald P. Binding kin etics of fluticasone propionate to the human glucocorticoid receptor. Steroids. 1994 Oct;59(10):597 602. 76. Baumann D, Bachert C, Högger P. D issolution in nasal fluid, retention and anti inflammatory activity of fluticasone furoate in human nasal tissue ex vivo. Clin Exp Allergy J Br Soc Allergy Clin Immunol. 2009 Oct;39(10):1540 50. 77. Yaning Wang. PHARMACOKINETIC AND PHARMACODYNAMIC EVALUA TION OF BECLOMETHASONE DIPROPIONATE. University of Florida; 2003. 78. James David Talton. Pulmonary targeting of inhaled glucocorticoid dry powders. University of Florida; 1999.

PAGE 112

112 79. Guo Z, Gu Z, Howell SR, Chen K, Rohatagi S, Cai L, et al. Ciclesonide disposition and metabolism: pharmacokinetics, metabolism, and excretion in the mouse, rat, rabbit, and dog. Am J Ther. 2006 Dec;13(6):490 501. 80. Jones RM, Harrison A. A new methodology for predicting human pharmacokinetics for inhaled drugs from oratra cheal pharmacokinetic data in rats. Xenobiotica Fate Foreign Compd Biol Syst. 2012 Jan;42(1):75 85. 81. Rojas C, Nagaraja NV, Webb AI, Derendorf H. Microdialysis of triamcinolone acetonide in rat muscle. J Pharm Sci. 2003 Feb;92(2):394 7. 82. Martin LE , Harrison C, Tanner RJ. Metabolism of beclomethasone dipropionate by animals and man. Postgrad Med J. 1975;51 S uppl 4:11 20. 83. Rohatagi S, Luo Y, Shen L, Guo Z, Schemm C, Huang Y, et al. Protein binding and its potential for eliciting minimal systemic side effects with a novel inhaled corticosteroid, ciclesonide. Am J Ther. 2005 Jun;12(3):201 9. 84. Wu K, Blo mgren AL, Ekholm K, Weber B, Edsbaecker S, Hochhaus G. Budesonide and ciclesonide: effect of tissue binding on pulmonary receptor binding. Drug Metab Dispos Biol Fate Chem. 2009 Jul;37(7):1421 6. 85. Flynn SJ, Tong ZB, Yang RY, Kamiya H, Yu AB, Chan HK. Computational fluid dynamics (CFD) investigation of the gas solid flow and performance of Andersen cascade impactor. Powder Technol. 2015 Nov 1; 285:128 37. 86. Miller FJ, Asgharian B, Schroeter JD, Price O. Improvements and additions to the Multiple Path Particle Dosimetry model. J Aerosol Sci. 2016 Sep 1;99:14 26. 87. Wang Y, Hochhaus G. Simultaneous quantification of beclomethasone dipropionate and its metabo lite, beclomethasone 17 monopropionate in rat and human plasma and different rat tissues by li quid chromatography positive electrospray ionization tandem mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci. 2004 Jun 15;805(2):203 10. 88. Nes torov IA, Aarons LJ, Arundel PA, Rowland M. Lumping of whole body physiologically based pharma cokinetic models. J Pharmacokinet Biopharm. 1998 Feb;26(1):21 46. 89. Brown RP, Delp MD, Lindstedt SL, Rhomberg LR, Beliles RP. Physiological parameter values f or physiologically based pharmacokinetic models. Toxicol Ind Health. 1997 Aug;13(4):407 84. 9 0. Chen A, Kaufman S. Splenic blood flow and fluid efflux from the intravascular space in the rat. J Physiol. 1996 Jan 15;490 ( Pt 2):493 9.

PAGE 113

113 91. Rodgers T, Row land M. Physiologically based pharmacokinetic modelling 2: predicting the tissue distribution of acids, very weak bases, neutrals and zwitterions. J Pharm Sci. 2006 Jun;95(6):1238 57. 92. Daley Yates PT, Price AC, Sisson JR, Pereira A, Dallow N. Beclomet hasone dipropionate: absolute bioavailability, pharmacokinetics and metabolism following intra venous, oral, intranasal and inhaled administration in man. Br J Clin Pharmacol. 2001 May;51(5):400 9. 93. Woo MS, Szmuszkovicz JR. Chapter 4 Pulmonary Manife stations of Cardiac Diseases. In: Turcios NL, Fink RJ, editors. Pulmonary Manifestations of Pe diatric Diseases [Internet]. Philadelphia: W.B. Saunders; 2009 [cited 2019 Aug 2]. p. 79 97. Available from: http://www.sciencedirect.com/science/article/pii/B978 1416030317000048 94. Maynard RL. Comparative Biology of the Normal Lung (treatise on pulmonar y toxicology, volume 1). Occup Environ Med. 1995 Nov;52(11):783. 95. Bastacky J, Lee CY, Goerke J, Koushafar H, Yager D, Kenaga L, et al. Alveolar lining layer is thin and continuous: low temperature scanning electron microscopy of rat lung. J Appl Physi ol Bethesda Md 1985. 1995 Nov;79(5):1615 28. 96. Salar Behzadi S, Wu S, Mercuri A, Meindl C, Stranzinger S, Fröhlich E. Effect of the pulmonary deposition and i n vitro permeability on the prediction of plasma levels of inhaled budesonide formulation. Int J Pharm. 2017 Oct 30;532(1):337 44. 97. Shao J, Chen M J, Kuehl PJ, Hochhaus G. Pharmacokinetic and pharmacodynamic modeling of gut hormone peptide YY(3 36) af ter pulmonary delivery. Drug Dev Ind Pharm. 2019 Jul 3;45(7):1101 10. 98. McNally K, Cotton R, Loizou GD. A Workflow for Global Sensitivity Analysis of PBPK Models. Front Pharmacol. 2011;2:31. 99. Chanoine F, Grenot C, Heidmann P, Junien JL. Pharmacoki netics of butixocort 21 propionate, budesonide, and beclomethasone dipropionate in the rat aft er intratracheal, intravenous, and oral treatments. Drug Metab Dispos Biol Fate Chem. 1991 Apr;19(2):546 53. 100. Eriksson J, Sjögren E, Thörn H, Rubin K, Bäckm an P, Lennernäs H. Pulmonary absorption estimation of effective pulmonary permeability and t issue retention of ten drugs using an ex vivo rat model and computational analysis. Eur J Pharm Biopharm Off J Arbeitsgemeinschaft Pharm Verfahrenstechnik EV. 201 8 Mar;124:1 12. 101. Würthwein G, Rohdewald P. Activation of beclomethasone dipropionate by hydrolysis to beclomethasone 17 monopropionate. Biopharm Drug Dispos. 1990 Jul;11(5):381 94.

PAGE 114

114 102. Davies B, Morris T. Physiological parameters in laboratory ani mals and humans. Pharm Res. 1993 Jul;10(7):1093 5. 103. Weber B, Hochhaus G. A Pharmacokinet ic Simulation Tool for Inhaled Corticosteroids. AAPS J. 2012 Nov 10;15(1):159 71. 104. Crowe A, Tan AM. Oral and inhaled corticosteroids: differences in P glyco protein (ABCB1) mediated efflux. Toxicol Appl Pharmacol. 2012 May 1;260(3):294 302. 105. Bha gwat S, Schilling U, Chen M J, Wei X, Delvadia R, Absar M, et al. Predicting Pulmonary Pharmacokinetics from In Vitro Properties of Dry Powder Inhalers. Pharm Res . 2017 Dec;34(12):2541 56. 106. Patrick G, Stirling C. Measurement of mucociliary clearance from the trachea of conscious and anesthetized rats. J Appl Physiol. 1977 Mar;42(3):451 5. 107. Christopher AB, Ochoa S, Krushansky E, Francis R, Tian X, Zahid M, et al. The effects of temperature and anesthetic agents on ciliary function in murine respi ratory epithelia. Front Pediatr. 2014;2:111. 108. Valotis A, Högger P. Human receptor kinetics and lung tissue retention of the enhanced affinity glucocorticoid fluticasone furoate. Respir Res. 2007;8(1):54. 109. Gearhart JM, Jepson GW, Clewell HJ, And ersen ME, Conolly RB. Physiologically based pharmacokinetic and pharmacodynamic model for the inhibition of acetylcholinesterase by diisopropylfluorophosphate. To xicol Appl Pharmacol. 1990 Nov;106(2):295 310.

PAGE 115

115 BIOGRAPHICAL SKETCH Jie Shao received a degree in p harmaceutical e ngineering from Beijing University of Chinese Medicine in China . Then, she got a degree in p harmaceutical s ciences at SUNY, University at Buffalo in 2013. She completed her t of PEITC on NNMT and STAT3 expression in HT29 Human under the supervision of Dr. Marilyn Morris . Thereafter, she joined Center for Pharmacometrics & Systems pharmacology at the University of Florida to start her Ph.D. study under supe rvision of Dr. Guohua An . Then, she joined Dr. G ü nther Hochhaus lab in September 2014 to continue her Ph . D . study. Her research interest is in PK/PD ana lysis of pulmonary delivered drugs. Meanwhile, she obtained a minor in s tatistics in 2016. In 2019, she received her Ph. D. degree in p harmaceutic al sciences .