Polymer Therapies for Treating Pediatric Osteosarcoma

MISSING IMAGE

Material Information

Title:
Polymer Therapies for Treating Pediatric Osteosarcoma
Physical Description:
1 online resource (155 p.)
Language:
english
Creator:
Popwell, Sam
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Materials Science and Engineering
Committee Chair:
Batich, Christopher D
Committee Co-Chair:
Wagener, Kenneth B
Committee Members:
Gower, Laurie B
Baney, Ronald H
Sinnott, Susan B
Milner, Rowan J
Bolch, Wesley E

Subjects

Subjects / Keywords:
cancer -- osteosarcoma -- polymer -- radiopharmaceutical
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre:
Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
A multidisciplinary approach to advance targeted radiotherapy for treating metastatic bone cancer is described. The design, synthesis, characterization, and preclinical evaluation of two polymer therapeutics designed to target osteosarcoma are presented in detail. The polymer, polyethyleneimine methylenephosphonate (PEI-MP), was first introduced by Milner et al. and was reproduced here for further study. Successful synthesis of PEI-MP was supported by 1H NMR, 13C NMR, 31P NMR, FT-IR, and elemental analysis. PEI-MP was fractionated to provide particles for drug delivery with an average diameter of 5.54 nm. The zeta potential of PEI-MP was measured to be -52.43, indicating a stable dispersion. Biodistribution studies in the osteosarcoma canine model (n=6) indicated good tumor targeting of PEI-MP with little uptake in regions associated with dose limiting organs. Renal clearance rates of 35-40% of excess radioactivity were achieved, indicating the need for improvement in particle size distribution in order to limit radiation exposure to non-target organs. Biodistribution studies conducted in dogs with non-osseous tumor types (n=3) show no retention of the radiopharmaceutical. Poor binding of therapeutic radiolanthanides by PEI-MP in vivo was addressed through the design of a new polymer ligand, polyethyleneimino ethylenetetramine dimethylenephosphonate (PEI-EDTMP). Successful synthesis of PEI-EDTMP was supported by 1H NMR, 13C NMR, 31P NMR, FT-IR, and elemental analysis. The polymer was fractionated to provide particles for drug delivery with an average diameter of 4.27 nm. The zeta potential of PEI-EDTMP was measured to be -28.63, indicating a moderately stable dispersion. Biodistribution studies conducted in healthy canines (n=4) indicated low uptake regions associated with dose limiting organs of healthy bone and associated bone marrow with a renal clearance rate of 65%. An additional study in a canine osteosarcoma model (n=1) indicated good uptake within the tumor region. Incorporation of the multidentate chelating functionality into PEI-EDTMP has shown to increase the binding stability with the lanthanide samarium relative to the polymer, PEI-MP, as indicated by a two order of magnitude increase in the thermodynamic binding constant as determined by isothermal titration calorimetry.  Based on this work, phosphonate functionalized polymers demonstrate great potential as a systemic therapy for treating osteosarcoma.
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.
Statement of Responsibility:
by Sam Popwell.
Thesis:
Thesis (Ph.D.)--University of Florida, 2012.
Local:
Adviser: Batich, Christopher D.
Local:
Co-adviser: Wagener, Kenneth B.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2014-08-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2012
System ID:
UFE0044424:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 POLYMER THERAPIES FOR TREATING PEDIATRIC OSTEOSARCOMA By SAM POPWELL A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHI LOSOPHY UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2012 Sam Popwell

PAGE 3

3 To my loving and supportive wife Christy

PAGE 4

4 ACKNOWLEDGMENTS There are a number of people to whom I am grateful for their support, encouragement, and guidance. First I would like to thank my wife Christy for her continued love, support and sacrifices as I have pursued my interest s and goals. She has selflessly provided her support during the challenges of graduate school. I thank my parents for their example of hard work and service to o thers. They have taught me by example how to step outside of my comfort zone to pursue new challenges. Thank you to my twin sister, Susan for always being there I am lucky to have been born with a best friend to grow up with. I am grateful for my two sis ters, Wendy and Julie, for valuing family and always putting others first. Duri ng my undergraduate studies at T he University of Southern Mississippi, Dr. John Pojman sparked my interest in chemistry and polymers. I would not have achie ved all that I have w ithout the opportunities that he provided I thank Dr. Kenneth Wagener for bringing me to the University of Florida I appreciate how he always looks for the potential in others, and I have spent countless hours in his office learning as much about life as chemistry. I cannot express in words my gratefulness to my graduate school advisors, Drs. Christopher Batich and Kenneth Wagener, for their unconditional support and for giving me the opportunity and resources to complete my degree. The support from fel low Mississippian, George Butler and hi s lovely wife Josephine, have made it possible for m yself and many others to receive a superior education. I have had the opportunity to work very closely with several of my committee members during my dissertation work I am especially grateful to Dr. Rowan Milner and Dr. Wesley Bolch for their support. They have gone above and beyond with their willingness to assist me as I have stepped into areas of research that were foreign to

PAGE 5

5 me. Dr. Kip Berry was an essential asset to my work, and I am grateful for the time he dedicated to helping me. I thank Drs. Joanne Lagmay and Robert Zlotecki for p roviding valuable direction for my project. Additionally, I thank Dr. Susan Sinnott, Dr. Ronald Baney, and Dr. Laurie Gower fo r taking the time to serve on my committee. I would like to thank Marcel Wissel, Dr. Markus Klapper, and Dr. Klaus Mullen for the opportunity to experience science outside of the United States. My time spent with them conducting research at the Max Plank I nstitute for Polymer Research in Mianz, Germany has provided me with the a broadened perspective. I will always look at my short time in Germany as one of my b est educational experiences I am additionally thankful to the Burroughs Wellcome Fund and Dr. K enneth Wagener for making that experience possible. I have had the opportunity to work with many amazing people at the University of Florida. Many friendships have been forged that will live well beyond our time in Gainesville. James Leonard went from bein g my chemistry mentor to being a great friend, and I look forward to our annual two man golf tournament s Thanks to previous Wagener Group members E ric Berda and Kate Opper for being good labmates and showing me the way during my first years in graduate sc hool. I will always remember Paula Delgado for her thoughtfulness and kind ways. I thank bowling team captain, Brian Aitken, for teaching me to catch a redfish; I always enjoyed our lengthy debates on every topic imaginable I than k Bora Inci for his kindness and advice along the way ; I look f orward to future visits with him hopefully someday in Turkey. I am grateful for my friendship and time spent grilling, golfing, and taking road trips with Lu ke Fisher. Pascale Atallah has prov ided me with great conversation and

PAGE 6

6 encouragement; I look forward to our next adventure race. I appreciate Chet Simocko for le tting me bounce ideas off of him as we shared hood space. I enjoyed my time in the lab with Ashlyn Dennis and her homemade cobbler s I am thankful for Julian Wang who was a great asset to my project as I finished up my final semester. I leave my project in good hands with Michael Schulz, with whom I have enjoyed discussing ideas for new cancer therapies. I enjoyed my morning breakfas ts with Chip Few and weekly Friday night carrot cake with Taylor Gaines discussing chemistry and life I have enjoyed getting to know the next gen eration of Wagener group member s including Donovan Thompson, Nicolas Sauty, and Lucas Caire da Silva. I wish t hem all best of luck.

PAGE 7

7 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ .......... 10 LIST OF FIGURES ................................ ................................ ................................ ........ 12 LIST OF ABBREVIATIONS ................................ ................................ ........................... 17 ABSTRACT ................................ ................................ ................................ ................... 18 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 20 Pediatric Osteosarcoma ................................ ................................ .......................... 20 Current Cancer Therapies ................................ ................................ ...................... 21 Surgery ................................ ................................ ................................ ............. 21 Chemotherapy ................................ ................................ ................................ .. 21 Radiation Therapy ................................ ................................ ............................ 22 Radionuclide therapy ................................ ................................ ................. 23 Current radionuclide therapies ................................ ................................ ... 24 Cancer Biology: Tumor Vascularization ................................ ................................ .. 25 Hypoxia ................................ ................................ ................................ ............ 25 Angiogenesis ................................ ................................ ................................ .... 26 Polymer Therapy: Enhanced Permeability and Retention Effect ............................ 26 Polymer Cancer Therapeutics ................................ ................................ ................ 28 Particle Size ................................ ................................ ................................ ..... 28 Binding Stability ................................ ................................ ................................ 30 Biological Evaluation ................................ ................................ ........................ 31 Rad iopharmaceutical Design ................................ ................................ .................. 31 Polymer Selection ................................ ................................ ............................ 32 Radionuclide Selection ................................ ................................ ..................... 32 Chelating Ligand Selection ................................ ................................ ............... 34 2 SYNTHESIS, CHARACTERIZATION, AND PRE CLINICAL EVALUATION OF A POLYMER BASED RADIOPHARMACEUTICAL FOR INTERAL RADIONUCLIDE THERAPY ................................ ................................ ................... 36 History of Polyethyleneimine Methylenephosohonate (PEI MP) ............................. 36 Experimental ................................ ................................ ................................ ........... 39 Materials ................................ ................................ ................................ ........... 39 Instrumentation and Analysis ................................ ................................ ........... 39 trimethylenephosphonate Polyethyleneimine (2 1B). ...... 40 Particle Size Separation by Ultrafiltration ................................ ......................... 40

PAGE 8

8 Endotoxin Testing ................................ ................................ ............................. 41 99m Tc PEI MP Kit Prepar ation ................................ ................................ ........... 41 Biodistribution Studies ................................ ................................ ...................... 42 Particle Size Determination ................................ ................................ .............. 42 Zeta Potential Measurements ................................ ................................ ........... 43 Results and Discussion ................................ ................................ ........................... 43 Synthesis and Structural Analysis ................................ ................................ .... 43 Polymer Filtration ................................ ................................ ............................. 47 Endotoxin Testing ................................ ................................ ............................. 48 Preparation of 99m Tc PEI MP Radiopharmaceutical Kit ................................ .... 49 Particle Size Analysis ................................ ................................ ....................... 51 Particle Stability ................................ ................................ ................................ 57 Animal Care and Procedur es ................................ ................................ ........... 60 Biodistribution Studies ................................ ................................ ...................... 60 Summary ................................ ................................ ................................ ................ 8 0 3 DESIGN OF A POLYMER RAD IOPHARMACEUTICAL FOR IMPROVED BINDING WITH RADIOLANTHANIDES ................................ ................................ 82 Polyethyleneimine Ethylenediamine Tetramethylene Phosphonate ....................... 82 Exp erimental ................................ ................................ ................................ ........... 83 Materials ................................ ................................ ................................ ........... 83 Instrumentation and Analysis ................................ ................................ ........... 84 Synthesis of 2,3 bis((tert butoxycarbonyl)amino)propionic acid (3 4B) ............ 84 Coupling of Polyethylenimine and 2,3 bis((tert butoxycarbonyl)amino)propionic acid (3 6C). ................................ ................ 85 Deprotection of Polyethylenimine 2,3 bis((tert butoxycarbonyl)amino)propionic acid (3 7B) ................................ ................. 85 Synthesis of Polyethyleneimine ethylenediaminetetramethylene Phosphonate (3 8B) ................................ ................................ ...................... 86 Particle Size Fractionation by Ultrafiltration ................................ ...................... 86 Particle Size Determination by Dynamic Light Scattering ................................ 87 Zeta Potential Measurements ................................ ................................ ........... 87 99m Tc PEI EDTMP Pharmaceutical Kit Preparation ................................ .......... 87 Biodistribution Studies ................................ ................................ ...................... 88 Results and Discussion ................................ ................................ ........................... 88 Design and Synthesis ................................ ................................ ....................... 88 Particle Size Analysis ................................ ................................ ....................... 94 Particle Stability ................................ ................................ ................................ 96 Animal Care and Procedures ................................ ................................ ........... 98 Biodistribution ................................ ................................ ................................ ... 98 Summary ................................ ................................ ................................ .............. 108 4 IN VITRO CHARACTERIZATION OF BINDING STABILITIES OF POLYMER RADIOPHARMACEUT ICALS ................................ ................................ ............... 109 In Vitro Investigation of Polymer Metal Systems ................................ ................... 109

PAGE 9

9 Binding Properties of Lanthanides ................................ ................................ .. 110 Thermodynamic Stability and Kinetic Inertness ................................ .............. 110 Measuring Complex Stability ................................ ................................ .......... 111 Experimental ................................ ................................ ................................ ......... 112 Materials ................................ ................................ ................................ ......... 112 Methods ................................ ................................ ................................ .......... 112 Description of ITC Binding Experiment ................................ ........................... 113 Description of Dialysis Experiment ................................ ................................ 114 Results and Discussions ................................ ................................ ....................... 114 Thermody namic Binding Parameters ................................ ............................. 114 Binding constants ................................ ................................ ..................... 123 Binding free energy ................................ ................................ .................. 125 Enthalpy contribution to binding ................................ ............................... 127 Entropy contribution to binding ................................ ................................ 128 Binding stoichiometry ................................ ................................ ............... 128 Summary of Thermodynamic Binding Parameters ................................ ......... 129 Ligand to Metal Ratios ................................ ................................ .................... 129 Kinetic Stability of Li gand Complexes ................................ ............................ 129 Summary ................................ ................................ ................................ .............. 138 5 CONCLUSIONS ................................ ................................ ................................ ... 140 Summary ................................ ................................ ................................ .............. 140 Outlook ................................ ................................ ................................ ................. 143 Polymer ................................ ................................ ................................ .......... 144 Ligand ................................ ................................ ................................ ............. 145 Radionuclide ................................ ................................ ................................ ... 146 APPENDIX CALCULATION OF DEGREE OF METHYLENE PHOSPHONATE FUNCTIONALIZATION OF PEI MP ................................ ................................ ...... 147 LIST OF REFERENCES ................................ ................................ ............................. 149 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 155

PAGE 10

10 LIST OF TABLES Table page 1 1 Different types of par ticle emissions and energies ................................ ............. 33 1 2 Nuclear properties of lanthanides ................................ ................................ ....... 34 2 1 Endotoxin assay results for PEI MP ................................ ................................ ... 49 2 2 Average particle sizes and particle size distributions for PEI MP fractions ......... 56 2 3 Tumor locations for osteosarcoma biodistribution studies w ith PEI MP ............. 62 2 4 Percent injected activity excreted through the kidneys for 99m Tc PEI MP administered to dogs with osteosarcoma ................................ ........................... 73 2 5 Percent organ activity distribution at 3 hours for 99m Tc PEI MP administered to osteosarcoma bearing canines ................................ ................................ ....... 74 2 6 Percent injected activity organ distribution of 99m Tc PEI MP at three h ours ....... 74 2 7 Tumor types and locations for non osteosarcoma studies ................................ 75 2 8 The percent injected activity excreted through the kidne ys for 99m Tc PEI MP administered to dogs with non osteosarcoma tumors ................................ ........ 78 2 9 Percent injected activity organ distribution of 99m Tc PEI MP at three hours ....... 79 2 10 Percent organ activity distribution at 3 hours for 99mTc PEI MP administered to non osseous tumor bearing canines ................................ ............................... 79 3 1 Average particle sizes and particle size distributions for PEI EDTMP ................ 96 3 2 Percent injected activity of 99m Tc PEI EDTMP excreted from healthy canines through the urine over a three hour time period ................................ ................ 104 3 3 Percent organ activity distribution at three hours in four healthy dogs administered 99m Tc PEI EDTMP ................................ ................................ ....... 104 3 4 Percent injected activity organ distribution of 99 m Tc PEI EDTMP at three hours in healthy canines ................................ ................................ ................... 105 3 5 Percent organ activity distribution at 3 hours for 99m Tc PEI EDTMP administered to an osteosarcoma bearing canine ................................ ............ 107 3 6 Percent injected activity organ distribution of 99m Tc PEI EDTMP at three hours in an osteosarcoma bearing canine ................................ ........................ 107

PAGE 11

11 4 1 Concentrations of reagents that provided appropriate ITC curves .................... 119 4 2 Thermodynamic binding parameters as determined by ITC for three phosphonate ligands ................................ ................................ ........................ 126 5 1 Comparison of radiopharmaceuticals ................................ ............................... 140 A 1 Number of carbons associated with each amine in the PEI repeat unit ............ 147 A 2 Number o f carbons associated with each amine in the PEI MP repeat unit ..... 148

PAGE 12

12 LIST OF FIGURES Figure page 1 1 External versus internal radiation thera py ................................ .......................... 24 1 2 Structure of 153 Sm EDTMP (Quadramet ) ................................ .......................... 24 1 3 Enhanced Permeability and Retention (EPR) method of macromolecular drug del ivery ................................ ................................ ................................ ............... 27 1 4 Relative sizes of multiple molecular weight polymers and various pores in the body ................................ ................................ ................................ .................... 30 1 5 Features of a polymer radio pharmaceutical ................................ ....................... 31 1 6 Polyethyleneimine, repeat unit structure and branched morphology .................. 32 1 7 Structure of polyamino phosphonate l igands ................................ ...................... 35 1 8 Structure of polyamino carboxylate ligands ................................ ........................ 35 2 1 The small molecule radiopharmaceutical, Quadramet and polymer rad iopharmaceutical, PEI MP ................................ ................................ ............. 36 2 2 Organ activity distribution for 99m Tc PEI MP versus 153 Sm EDTMP ................... 37 2 3 Organ activity distribution for 99m Tc PEI MP versus 153 Sm PEI MP .................... 38 2 4 Synthesis of branched polyethyleneimine ................................ .......................... 44 2 5 Synthesis of polyethyleneimine methyle nephosphonate (PEI MP) ..................... 45 2 6 Infrared spectrum of starting material, polyethylenimine ................................ .... 46 2 7 Infrared spectrum of methylene phosp honate functionalized polyethyleneimine ................................ ................................ ............................... 46 2 8 Reduction/oxidation reaction of stannous chloride ................................ ............. 50 2 9 Image of lyophilized radio pharmaceutical kits containing PEI MP and SnCl 2 ..... 51 2 10 Particle size distribution for polyethyleneimine (Lupasol WF) ............................. 53 2 11 Part icle size distribution for polyethyleneimine methylene phosphonate (PEI MP) in 0.15 M NaCl ................................ ................................ ............................ 54 2 12 C omparison plot for PEI MP particle sizes in the pres ence of the ions calcium, magnesium and samarium ................................ ................................ .. 55

PAGE 13

13 2 13 Particle size distribution of PEI MP retained during filtration through a 30 kDa MW cut off filter ................................ ................................ ................................ .. 55 2 14 Partic le size distribution of PEI MP retained after sequential filtration through 30 and 10 kDa MW cut off filters ................................ ................................ ......... 55 2 15 Particle size distribution of PEI MP that passed through a 10 kDa MW cut off filter ................................ ................................ ................................ ..................... 56 2 16 Hindered reputation of a branched polymer through pores as described by Frechet et al. ................................ ................................ ................................ ....... 57 2 17 Particle size distr ibution of aggregated PEI MP in solution ................................ 57 2 18 Zwitterionic form of polyethyleneimine methylenephosphonate likely to exist under physiological conditions ................................ ................................ ............ 58 2 19 Zeta potential for PEI MP in 0.15 M NaCl at pH 6 8 ................................ ........... 59 2 20 Positioning of dog for scintigraphic imaging ................................ ....................... 61 2 21 Scintigraphic image and time activity curve for organs of interest for 99mTc PEI MP administered to a dog with osteosarcoma in the left distal radius ......... 62 2 22 Scintigraphic image and time activity curve for tumor region for 99mTc PEI MP administered to a dog with osteosarcoma in the left distal radius ................ 63 2 23 Scintigraphic image and time activity curve for organs of i nterest for 99m Tc PEI MP administered to a dog with osteosarcoma in the right distal radius ....... 64 2 24 Scintigraphic image and time activity curve for tumor region for 99m Tc PEI MP administered to a dog with osteosarcoma in the right distal radius ..................... 65 2 25 Scintigraphic image and time activity curve for organs of interest for 99mTc PEI MP administered to a dog with osteosarcoma in the ri ght proximal humerus ................................ ................................ ................................ ............. 66 2 26 Scintigraphic image and time activity curve for tumor region for 99m Tc PEI MP administered to a dog with osteosarcoma in the right proximal humerus ........... 67 2 27 Scintigraphic image and time activity curve for organs of interest for 99m Tc PEI MP administered to a dog with osteosarcoma in the left proximal humerus ................................ ................................ ................................ ............. 68 2 28 Scintigraphic image and time activity curve for tumor region for 99m Tc PEI MP administered to a dog with osteosarcoma in the left proximal humerus ............. 69 2 29 Scintigraphic i mage and time activity curve for organs of interest for 99m Tc PEI MP administered to a dog with osteosarcoma in the right distal radius ....... 70

PAGE 14

14 2 30 Scintigraphic image and time activity curve for t umor region for 99m Tc PEI MP administered to a dog with osteosarcoma in the right distal radius ..................... 71 2 31 Scintigraphic images and time activity curves for 99m Tc PEI MP administered to a dog with osteosarcoma in the right distal radius ................................ .......... 72 2 32 Scintigraphic image and TAC curve for 99m Tc PEI MP administered to a dog with soft tissue carcinoma in the left pelvic limb ................................ ................. 75 2 33 Scintigraphic image and TAC curve for organs of interest for 99m Tc PEI MP administered to a dog with histi ocytic sarcoma of the perineum ......................... 76 2 34 Scintigraphic image and TAC curve for tumor region for 99m Tc PEI MP administered to a dog with histiocytic sarcoma of the perineum ......................... 77 2 35 Scintigraphic image and TAC curve for 99m Tc PEI MP administered to a dog with adenocarcinoma of the mammary gland ................................ ..................... 78 3 1 Structure of ethylenediamine tetramethylenephosphonate (EDTMP) ................. 89 3 2 Design of the polymer radiopharmaceutical PEI EDTMP for improved binding with therapeutic radionuclides ................................ ................................ ............ 89 3 3 Structure of polyethyleneimine ethylenediamine tetramethylene phosphonat e (PEI EDTMP) ................................ ................................ ................................ ...... 90 3 4 Protection of primary amines in D,L 2,3 diamino propionic acid hydrochloride .. 90 3 5 1 H NMR spectrum showing t he presence of the boc protecting group at 1.4 ppm ................................ ................................ ................................ .................... 91 3 6 Synthetic scheme for the coupling of polyethyleneimine and boc protected diamino propionic acid ................................ ................................ ........................ 92 3 7 Deprotection of boc protected diaminopropionic amines ................................ .... 92 3 8 Synthetic scheme for the methyl phosphonation of the primary amines ............. 93 3 9 Synthetic scheme for the formation of the sodium salt of PEI EDTMP ............... 93 3 10 Infrared spectra of polyethyleneimine, PEI Boc Dap and PEI EDTMP ............ 94 3 11 Particle size distribution of PEI EDTMP (>30 kDa) ................................ ............. 95 3 12 Particle size distribution of 10 30 kDa PEI EDTMP ................................ ............ 95 3 13 Particle size distribution of PEI EDTMP in the absence of plasma ions and in the presence of calcium ................................ ................................ ...................... 96

PAGE 15

15 3 14 Zeta potential values for PEI EDTMP ................................ ................................ 97 3 15 Scintigraphy images showing the biodistribution of 99m Tc PEI EDTMP (10 30 kDa) in a healthy canine ................................ ................................ ..................... 99 3 16 Decay corrected tim e activity curves for 99mTc PEI EDTMP (10 30 kDa) in a healthy canine ................................ ................................ ................................ .. 100 3 17 Scintigraphy images showing the biodistribution of 99m Tc PEI EDTMP (10 30 kDa) in a healthy canine ................................ ................................ ................... 100 3 18 Decay corrected time activity curves for 99m Tc PEI EDTMP (10 30 kDa) in a healthy canine ................................ ................................ ................................ .. 101 3 19 Scintigraphic images showing the biodist ribution of 99m Tc PEI EDTMP (10 30 kDa) in a healthy canine ................................ ................................ ................... 101 3 20 Decay corrected time activity curves for 99m Tc PEI EDTMP (10 30 kDa) in a healthy canine ................................ ................................ ................................ .. 102 3 21 Scintigraphic images showing the biodistribution of 99m Tc PEI EDTMP (10 30 kDa) in a healthy canine ................................ ................................ ................... 102 3 22 Decay corrected time activity curves for 99m Tc PEI EDTMP (10 30 kDa) in a healthy canine ................................ ................................ ................................ .. 103 3 23 Decay corrected time activity curves for kidney uptake of 99mTc PEI EDTMP (10 30 kDa) in four separate healthy canines ................................ ................... 106 3 24 Mean and SD of liver ROIs for 99m Tc PEI EDTMP (10 30 kDa) administered to four separate healthy canines ................................ ................................ ....... 106 3 25 Time activity curves for 99mTc PEI EDTMP administered to an osteosarcoma bearing canine ................................ ................................ .......... 107 4 1 In vivo reactions of a polymer radiopharmaceutical in the bloodstream ........... 109 4 2 Potential energy diagram for a binding reaction ................................ ............... 111 4 3 Three bisphosphonate ligands: ethylenediamine tetramethylene phosphonate (EDTMP), polyethyleneimine methylene phosphanote (PEI MP), polyethelyneimine etylediamine tetramethylene phosphonate (PEI EDTMP) ... 115 4 4 Schematic of isothermal titration calorimetry instrument ................................ .. 117 4 5 Isothermal titration calorimetry raw data and integrated data output ................ 117 4 6 The shape of ITC curves at different concentrations of reactants ..................... 118

PAGE 16

16 4 7 ITC data output for PEI MP titrated with samarium ................................ .......... 120 4 8 ITC data output for EDTMP titrated with samarium ................................ .......... 121 4 9 ITC data output for PEI EDTMP titrated with samarium ................................ ... 122 4 10 Possible mechanism for two step binding process with bidentate chalators .... 123 4 11 Binding constants as measured by ITC ................................ ............................ 124 4 12 Monodentate ligands and bidentate ligands binding with nickel ....................... 124 4 13 Thermodynamic parameters for three bisphosphonate ligands ........................ 126 4 14 Relationship between a chemical binding event and the equilibrium constant 130 4 15 Dialysis experimental set up ................................ ................................ ............. 131 4 16 Dialysis of a 15 ppm solution of samarium in DI water ................................ ..... 132 4 17 Th e effect of membrane soaking time on dialysis ................................ ............. 132 4 18 The effect of membrane on dialysis ................................ ................................ .. 133 4 19 Dialysis of samarium in the prese nce of PEI MP ................................ .............. 134 4 20 The effect of calcium on dialysis of samarium in the presence of PEI MP ....... 135 4 21 The binding of samarium by P EI MP in blood serum ................................ ........ 136 4 22 The dialysis of samarium in the presence of PEI EDTMP ................................ 137 4 23 The effect of calcium on the binding of samarium by PEI EDTMP ................... 138 A 1 Differentiation of carbons within the PEI repeat unit ................................ ......... 147 A 2 Differentiation of carbons within the P EI MP repeat unit ................................ .. 147

PAGE 17

17 LIST OF ABBREVIATIONS ATR Attenu ated total r eflectance DLS Dynamic light s cattering DNA Deoxyribonucleic a cid EDTMP Ethylenediamine tetramethylene phosponate EPR Enhanced permeability and r etention FD A Food and Drug Administration FGF Fibroblast growth factor FT IR Fourier transform infrared spectroscopy H R Hydrodynamic radius HRMS High resolution mass spectrometry IA Injected activity kDa Kilodaltons LAL Limulus amebocyte l ysate MBq Megabecquerel NMR Nuclear magnetic resonance O A Organ activity PDI Polydispersity index PEI Poly(ethyleneimine) PEI EDTMP Polyethyleneimine ethylenediamine tetramethylenephosphonate PEI MP Polyethyleneimine methylene phosphonate RES Reticuloendothelial system ROI Region of interest VEG F Vascular endothelial growth factor

PAGE 18

18 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy POLYMER THERAPIES FO R TREATING PEDIATRIC OSTEOSARCOMA By Sam Popwell August 2012 Chair: Christopher Batich Co chair: Kenneth B. Wagener Major: Materials Science and Engineering A multidisciplinary approach to advance targeted radiotherapy for treating metastatic bone can cer is described The design, synthesis, characterization, and preclinical evaluation of two polymer therapeutics designed to target osteosarcoma are presented in detail The polymer, polyethyleneimine methylenephosphonate (PEI MP), was first introduced b y Milner et al and was reproduced here for further study. Successful synthesis of PEI MP was supported by 1 H NMR, 13 C NMR, 31 P NMR, FT IR, and elemental analysis. PEI MP was fractionated to provide particles for drug delivery with an average diameter of 5 .54 nm The zeta potential of PEI MP was measured to be 52.43, indicating a stable dispersion. Biodistribution studies in the osteosarcoma canine model (n=6) indicated good tumor targeting of PEI MP with little uptake in regions associated with dose limit ing organs. Renal clearance rates of 35 40% of excess radioactivity were achieved, indicating the need for improvement in particle size distribution in order to limit radiation exposure to non target organs. Biodistribution

PAGE 19

19 studies conducted in dogs with n on osseous tumor types (n=3) show no retention of the radiopharmaceutical. P oor binding of therapeutic radiolanthanides by PEI MP in vivo was addressed through the design of a new polymer ligand, polyethyleneimino ethylenetetramine dimethylenephosphonate (PEI EDTMP). Successful synthesis of PEI EDTMP was supported by 1 H NMR, 13 C NMR, 31 P NMR, FT IR, and elemental analysis. The polymer was fractionated to provide particles for drug delivery with an average diameter of 4.27 nm. The zeta potential of PEI EDTM P was measured to be 28.63 indicating a moderately stable dispersion. Biodistribution studies conducted in healthy canines (n=4) indicated low uptake regions associated with dose limiting organs of healthy bone and associated bone marrow with a renal cle arance rate of 65% An additional study in a canine osteosarcoma model (n=1) indicated good uptake within the tumor region. Incorporation of the multidentate chelating functionality into PEI EDTMP has shown to incre ase the binding stability with the lantha nide samarium relative to the polymer, PEI MP as indicated by a two order of magnitude increase in the thermodynamic binding constant as determined by isothermal titration calorimetry Based on this work phosphonate functionalized polymers demonstrate great potential as a systemic therapy for treating osteosarcoma.

PAGE 20

20 CHAPTER 1 INTRODUCTION Pediatric Osteosarcoma Osteosarcoma accounts for 9 % of all cancer related deaths in children. 1 The societal impact of this disease is disproportionate to the number of cases diagnosed every year d ue to the low average age of the patients The World Health Organization has estimated that an average of 17 life years ar e lost for every osteosarcoma patient. This is compared to 6.5 years for cancer s of the bo w el, lung, and breasts 2 Prior to 1970, the main treatment for osteosarco ma was amputatio n of the a ffected area, with the resultant survival rate s being as low as 10 20%. 3 4 The survival rate quickly climbed to 60 7 0% w ith the emergence of chemotherapy drugs However, t his number has plateaued with little improvement in survival rate s being observed over the last 30 years. Of all patients with localized osteosarcoma, 30 40% will develop local recurrenc e with 5 year survival rates being repor ted in the range of 20 29% 5 6 Addition ally, approximately 20% of patients will present clinically with metastatic disease, where cancer has spread to other organs, with overall survival rates b eing reported as low as 10% for these patients 7 Al though survival rate s have improved with the ad vent of chemotherapy, there is great need for new treatments for osteosarcoma as children who have inaccessible primaries and metastases still ha ve dismal outco mes despite best known systemic t reatments. New agents and strategies to overcome malignant osteosarcoma are required. Current therapies and new strategies for systemic therapy of pediatric osteosarcoma are discussed in more detail in the balance of Chapte r 1

PAGE 21

21 Current Cancer Therapies The objective of cancer treatment is to reduce the tumor cell population to zero cells. The three traditional modes of cancer therapy are surgery, chemotherapy and radiation therapy. Surgery Limb salvaging surgery with adju vant therapy remains the first line of defense for treating osteosarcoma. Curative surgery is possible when the cancer is found in only one location, allowing all of it to be removed. Surgery is most often coupled with chemotherapy as either an adjuvant or neoadjuvant therapy to re duce the risk of recurrence or tumor bulk prior to surgery. Chemotherapy About half of all cancer patients present clinically with advanced disease. In cases with distant metastases, systemic chemotherapy is often considered one of the only viable methods to distribute effective therapy to all lesions in disseminated disease. The goal of using traditional cytotoxic chemotherapy is to achieve a differential response between the tumor and affected host tissue by exploiting differen ces in their ability to repair damage, since malignant cells have impaired DNA damage repair processes. Normal tissues recover at different rates, with the most clinically relevant tissue being the bone marrow. 8 As a result, bone marrow is the dose limiting organ in the treatment of osteosarcoma. The most popular chemotherapy for treating osteosarcoma is a regimen of the four dr ugs cisplatin, m ethotrexate, doxorubicin, and ifosfamide. 9

PAGE 22

22 Radiation T herapy Nearly half of all cancer patients receive some form of radiation therapy. 10 Ionizing r treatment. In radiation therapy, ionizing radiation is used to damage DNA severely enough to kill tumor cells. DNA damage from radiation therapy generally kills cells in the process of dividing in mitosis (mitotic cell death). Cells that divide frequently, such as cancer cells, are therefore more susceptible to mitotic cell death than cells that divide infrequently. Similar to chemotherapy, radiation therapy can be admin istered as an adjuvant or neoadjuvant therapy when combined with surgery. The goal of radiation therapy is the eradication of tumors while sparing healthy tissue with in the treatment field. This ideal goal is generally not achieved because radiation react s with matter (tissue) in a non specific and random fashion Tolerance of normal tissue in the treatment field therefore limits the tumor dose of radiation Radiation therapy is underutilized as a primary treatment for os teosarcoma except for lesion s localized to inaccessible sites. 11 This underutilization is partly due to the relative radio resistant nature of osteosarcoma, 12 although some studies have shown external beam radiation to have a positive effect on the treatment of this disease. 13 15 In vitro studies of osteosarcoma cell lines have shown them to be as radiosensitive as non sarcoma cell lines that are known to be radiocurable. 16 17 This indicates that other radiobiological factors such as tumor hypoxia, size and reoxygenation are responsible for the observed clinical radiation resist ance. 18 19 To overcome this radiation resistance, a higher dose must be l ocalized at the tumor site. A possible means of increasing the radiocurabi lity of osteosarcoma by increasing the local tumor dose is to develop tumor

PAGE 23

23 targeting radiopharmaceuticals. Factors affecting the design and delivery of tumor targeting radiopharmaceuticals are described in greater detail throughout Chapter 1 Radionuclid e t herapy Techniques for radiation dose delivery include external beam (traditional) therapy and internal radionuclide therapy. External beam therapy can be effective when the cancer is localized to one or few anatomic sites. For disseminated metastatic disease, internal emitter radiotherapy has the greater potential for therapeutic outcome. However, past efforts via antibody targeting have only seen success with diffuse malignancies, such as lymphoma and leukemia, with poor clinical effectiveness observe d in the targeting of solid tumors. New and innovative approaches to radionuclide therapy are thus warranted. The major challenge of radiation treatment lies in the need for site specific targeting, while preventing damage to healthy tissues. 20 Targeting cancer through internal radiotherapy holds great potential for a breakthrough in this regard. Targeted radiotherapy is different from external beam radiation, which i s commonly used today (Figure 1 1 ). Issues limiting the efficacy of external beam radiation include the use of high energy particles which must pass through layers of healthy tissue to reach the tumor site. This places major constraints on the therapeutic dose of radiation that can b e administered, especially for metastatic cancer with multiple loci Internal radiotherapy avoids many of these problems, since a tumoricidal dose can be delivered in vivo to multiple, site specific locations. This field of internal radiotherapy research i s at an exciting and innovative stage. The design and engineering of new macromolecular polymer therapies which is the subject of this dissertation offers a promising approach to overcome present day challenges.

PAGE 24

24 Fig ure 1 1 External (left) versus i nternal (right) radiation therapy Current r adionucl ide t herapies A variety of small molecule radiopharmaceuticals are found on the market today. One example is 153 Sm complexed with ethylenediaminetetramethylene phosphonic acid ( 153 Sm EDTMP), sold under the trade name Quadramet (Figure 1 2 ). The clinical indication for Quadramet relates to palliative treatment of bone pain associated with multiple osteoblastic skeletal metastases. 21 22 This drug offers relief to an estimated 200,000 patients suffering from debilitating pain each year. Fig ure 1 2. Structure of 153 Sm EDTMP (Quadramet ) Systemic radionuclide therapies have s h own to be as equally effect ive as wide field external beam therapy with less associated bone marrow toxicity. 23 As a result, toxicity is low er compared with other systemic therapies and radionuclide therapy is well

PAGE 25

25 tolerate d. 24 26 However, the therapeutic efficacy of 153 Sm EDTMP (Quadramet ) is in part limited by the radioactive dose that can be delivered to a patient without inflict ing damage to healthy tissue. This is especially problematic for pediatric patients who have areas of high bone turnover in normal, non cancer regions of the skeleton Quadramet has high potential for accumulation in growth plates in patients under the age of 16 years resulting in subsequent growth arrest. In order to improve the therapeutic value of Quadramet and other radiopharmaceuticals authorized in the same therapeutic context ( 32 P and 89 Sr), our work aims to limit the uptake in healthy tissue t hrough a targeted radionuclide therapy approach Cancer Biology: Tumor Vascularization Cancer is characterized by uncontrolled growth, proliferation, and dissemination of cells with in the body. Two basic characteristics of cancer are: 1) uncontrolled ce ll growth due to lack of response to normal regulatory mechanisms and 2) spread of cells throughout the body with conti nued growth and proliferation at the new location. Hypoxia Two properties of cancer that play an important role in polymer drug delive ry are tumor hypoxia and angiogenesis. These two events are tied together in a vicious cycle. Tissue oxygenation depends on cellular oxygen consumption rate and oxygen supply to respiring cells. In tumor cells, the oxygen consumption rate is high due to the large number of proliferating cells (cellular reproduction is energy demanding). 27 The oxygen supply to the tumor is poor due to abnormal tumor microvasculature. As a result, most tumors are hypoxic to some degree the oxygen supply is inadequate to meet the demand and tumor oxygenation is low. 28 The resulting hypoxia is both diffusional and ischemic, resulting in hypoxic areas being heterogeneously distributed throughout the

PAGE 26

26 tumor mass. 29 Biologically, hypoxia can either impair tumor cell growth or can be a factor in malignant progression. 30 Hypoxia within a tumor often triggers angiogenesis. Angiogenesis Angiogenesis (new blood vessel development) is a normal biological process. Non pathological angiogenesis takes place in a sequence of steps that are tightly controlled and regulated by molecular factors. 31 Angiogenesis is regulated by a balance between pro and anti angio genic factors. Some of these pro angiogenic factors include: vascular endothelial growth factor (VEGF), angioproteins, platelet derived growth factor (PDF), and fibroblast growth factor (FGF). In the case of tumor angiogenesis, deregulation occurs due to an i mproper balance of pro and anti angiogenic factors during the angiogenesis process. Almost all so lid tumors express some amount o f VEGF. The result is that tumor microvasculature is structurally and functionally a bnormal. These microvasculature abn ormalities include poor to no differentiation into proper arterioles and venules, chaotic and disorganized branching, gaps in endothelial lining and basement membrane, large inter vessel distances, larger than normal diameters, and increased vessel tortuos ity. The end result is that the vasculature supply triggered by the tumor is abnormal and thus while it enables the tumor to grow it also causes more hypoxia This induced h y poxia triggers more angiogenesis whi ch triggers more hypoxia and so on. Polymer Therapy: Enhanced Permeability and Retention Effect Polymer therapeutics fight against cancer based on their ability to target abnormal vasculature associated with solid tumors. Research has shown that for a tumor to grow beyo nd 1 2 mm 3 in size it must develop new blood vessels, referred to as neovasculature. 32

PAGE 27

27 between normal and healthy vasculatu re using scanning electron microscopy. 33 This was achieved by injecting water soluble acrylic monomer into the blood vessels of cancer infected mice. The acrylic mono mer was polymerized to form a cast of the blood vessels within both the tumor and healthy tissue. Examination of the casts shows that plastic resin is found extensively outside the blood vessels of the metastatic tumor nodule, due to the leaky nature of t his vasculature Moreover, this effect is not observed in the case of blood vessels within healthy tissue In addition to this leaky vasculature, tumor characteristics include malfunctioning lymphatics and high interstitial pressures. 34 These features allow for passive delivery and subsequent retention of polymer drugs to vascularized tumors. This experimentally verified p henomenon has been termed the Enhanced Permeability and Retention (EPR) effect. 35 Whereas low molecular weight drugs are free to traverse through blood vessels of healthy tissue, the EPR effect allows for macromolecules to be sequestered in the blood supply until reaching the site of the tumor (Figure 1 3 ). Figure 1 3. Enhanced Permeab ility and Retention (EPR) method of macromolecular drug delivery The EPR effect is applicable to almost all solid tumo rs and is a relatively recent but very general approach to cancer targeting drug design. 34 Clinical trials have already

PAGE 28

28 established the efficacy of this approach for delivery of common chemotherapy agents such as paclitaxel. 36 Few advances have been made in employing the EPR effect to systemic radionuclide therapy The gap between the development of radio diagnostic agents and radio therapeutic agents is substantial. Underutilization of this therapeutic modality is evi dent in the case of somatostatin analogue based radiotherapy of neuronendocrine tumors. 35 No somatostatin based radiotherapy has advanced beyond clinical studies, despite the fact that the first human clinical trials were conducted over a decade ago. 37 The lack of progre ss is attributed to challenges associated with delivery of radionuclides compared to chemotherapy counterparts. These challenges include the binding of radionuclides, accurate assessment of pharmacokinetics, and prevention of renal damage. In addition, th e requirement of a multifaceted approach has hindered the development of new macromolecular therapies. The fact is that most researchers lack the resources to facilitate collaboration with clinicians. As a result, little has been done with regards to addr essing these issues through macromolecular drug design, especially when involving radiotherapy. Polymer Cancer Therapeutics There are several characteristics of polymer cancer therapeutics that need to be considered in the development of a new radiopharma ceutical, including particle size, binding stabilities, and in vivo biological behavior. Particle Size Effective delivery of polymer therapeutics to solid tumors involve s a balance between three events: (1) elimination of the polymer drug from the bloodst ream via the kid neys, liver, spleen, and urinary bladder, (2) extravasation of the polymer drug from

PAGE 29

29 the bloodstream into the tumor interstitium and (3) minimizing uptake in healthy, non target tissue. Based on these three events, the characteristics of a polymer that will determine the success of polymer drug delivery are hydrodynamic volume (or molecular weight), molecular conformation, chain flexibility, branching, and location of the attached drug. 38 Since the main route of elimination of a polymer from the body is through kidney clearance, glomerular pore size (in the kidneys) tends to be the deciding factor for the biological h alf life of polymer drugs in the bloodstream. Polymers with hydrodynamic radii smaller than glomerular pores, which are typically around 5 nm in size, are eliminated from the body through urine production 39 This pore size roughly corresponds to a polymer molecular weight in the range of 30 to 50 kDa, depending on the polymer chemistry, molecular conformation and flexibility. 40 Another method of polymer e limination from the body is through the liver and reticuloendothelial (RES) system. Particles above 200 nm in diameter are more readily removed from the blood stream via this route of elimination. 41 Clearance of polymers from the bloodstream into tissue occurs primarily by diffusion through small gaps in the endothelial wall. The size of these gaps in healthy tissue is on the order of 2 6 nm. 42 Pores of the tumor vasculature are significantly larger than those in kidneys and healthy tissue, ranging in diameter from 40 to 80 nm. 43 The variation in pore size is illustrated in Figure 1 4. 38 Based on the pore siz es, it is clear that hydrodynamic volume (V h ) and therefore molecular weight are key factors in the pharmacokinetics and biodistribution of polymer drugs when employing the EPR effect. Once a critical hydrodynamic radius (h r ) equal to the renal threshold is reached, the glomerular elimination of polymers decreases with increasing hydrodynamic radius. This increase in polymer biological half life with

PAGE 30

30 increasing molecular weight holds true for polymers with the same chemistry, architecture and molecular c onformation up to the point where polymers are eliminated by the reticuloendothelial system (h r > 200 nm). Fig ure 1 4. Re lative sizes of multiple molecular weight polymers and various pores in the body Binding Stability The stability of a metal liga nd complex is a key factor in internal radionuclide therapy. Upon injection into the blood stream, the radiometal is likely to dissociate from the chelating ligand due to the low concentration. Such dissociation would result in accumulation of radionuclid e in healthy tissue. In vivo dissociation of the radionuclide from the chelate invariably results in non target organ uptake, leading to irradiation of these non target tissues. 44 45 There fore, the thermodynamic stability of the metal chelate is important. However, the solution stability of the radiopharmaceutical in the blood is ultimately determined by the kinetic inertness, or the rate of dissociation of the radionuclide from the metal chelate. The primary goal in selecting a chelating ligand is

PAGE 31

31 to minimize the in vivo dissociation. Techniques for determining the complex stability of a radiopharmaceutical prior to clinical evaluation are explored in this work Biological Evaluation Pr ior to clinical studies of a new pharmaceutical in humans, biological response should be measured in an appropriate animal model to establish efficacy of the new therapy. Naturally occurring cancers in dogs and humans share many common features, including histology, genetics, biological behavior, and response to conventional treatments. 46 Canine studies therefore provide valuable insight into the development and translation of new cancer therapies by defining toxicity, activity and pharmacokinetics. 46 The canine model is utilized in this work for developing new polymer radionuclide therapies for the tr eatment of primary osteosarcoma Radiopharmaceu tical Design The design of an effective therapeutic polymer radiopharmaceutical involves: (1) the selection of a polymer to deliver a radioactive element to diseased tissue, (2) the selection of a radionuclide that delivers appropriate energy for destructi on of diseased tissue, and (3) attachment of a bi functional chelating ligand for combining the rad ioisotope and polymer (F igure 1 5 ). 47 Th e design of a polymer ligand for binding of therapeutic radionuclides will be based on these three distinct features. Figure 1 5. Features of a polymer radiopharmaceutical

PAGE 32

32 Polymer Selection The polymer act s as the carrier for delivery of a radion uclide t o diseased tissue. P olymer s must posses s the requisite properties for passive delivery via the EPR effect water solubility, sufficiently high molecular weight, and biological inertness. Based on these desired properties, poly(ethyleneimine) (PEI) (Figure 1 6) manifests itself as a good model for developmen t of new radiotherapy polymers and will be the base polymer used for studies within this work Figure 1 6 Polyethyleneimine, A) repeat unit structure and B) b ranched morpho logy Radionuclide S election The choice of a radionuclide depends on the medical application of interest. For internal radionuclide therapy, the radiation s of interest are often particulate emissions (alpha particles, beta particles, and Auger electrons) because they deposit their kinetic energy over short distances through Coulombic force interactions producing only localized tissue damage. In terms of treating cancer, the mechanism of action is the destruction of nuclear DNA strands by radiation induce d ionizations, excitations, chemical transmutations, and local charge effects. 48 Alpha particles have the highest energy among the particles mentioned, followed by beta particles, and then Auger electrons. The range of particles in tissue is on the order of 5 10 cell diameters (40 100 m) for alpha particles and 1 cm for beta particles, a size comparabl e to that of larger A B

PAGE 33

33 tumors (Table 1 1 ). 49 The range of Auger electrons in tissue is on the order of nanometers, resulting in nuclear localization and sub cellular accumulation being required for high killing efficiency. 50 In order to uniformly irradiate the tumor cells, the nuclide distribution must be homogeneous with respect to the tumor volume. Even very modest levels of heterogeneity, where a small number of cells are untargeted, can be catastrophic in terms of tumor cure. 51 Beta particle emitters are the focus for this proposal because they yield a fairly homogeneous dose distribution even when they are heterogeneously distributed throughout t he target tissue. 52 The tumor size will determine the optimum energy of the beta emitter to be administered. Table 1 1. Different types of particle emissions and energie s, adapted from Kassis, A. I.; Adelstein, S. J. J. Nucl. Med. 2 005 46 4S Particles Energy E min E max Range Linear Energy Transfer Beta Particle Electrons Medium to High (0.02 2.3 MeV) 0.2 12 mm 0.2 keV/ Alpha Particle Heium Nuclei High (5 9 MeV) 40 EC/IC Auger Electrons Very Low (eV KeV) Several nm 4 Radiolanthanides are the focus of this work This choice was made with the idea that there is no single ideal radionuclid e, but rather there are a number of choices depending on the delivery vehicle and the clinical target. Lanthanides offer notable opportunities in the design of therapeutic radiopharmaceuticals. 53 Among the fifteen lanthanide elements, all have very similar chem istries, yet they posses s a wide range of

PAGE 34

34 nuclear properties ( Table 1 2 ). This affords the opportunity to match a desired set of nuclear properties with a specific clinical application. For example, several lanthanide isotopes have varying degrees of emi tted kinetic energy: low energy emitter 177 Lu, medium energy emitter 153 Sm, and high energy emitters 166 Ho and 90 Y. The low energy emitters would be suitable for smaller metastases, while high energy emitters are good candidates for large tumors. T able 1 2. Nuclear properties of lanthanides adapted from Cutler, C. S.; Smith, C. J.; Ehrhardt, G. J.; Tyler, T. T.; Jurisson, S. S.; Deutsch, E. Cancer Biotherapy & Radiopharmaceuticals 2000 15 531 Isotope Half Life Decay Days Max b (MeV) Gamma (keV ) Pm 149 2.21 1.1 286 Sm 153 1.93 0.69 103 Dy 166 3.4 0.4 82.5 Ho 166 1.12 1.8 80.6 Yb 175 4.19 0.47 396 Lu 177 6.71 0.5 208 Chelating Ligand S election A chelating ligand is used to link the radioisotope and carrier molecule for delivery to th e target area. The chelating ligand is covalently bound to the polymer and acts to strongly coordinate the radionuclide. The ligand used is highly dependent on the oxidation state and size of the metal. As a result, chelating ligands with different donor atoms and frameworks are required for different radionuclides. This necessitates an

PAGE 35

35 understanding of the chelating ligand coordination chemistry with the appropriate radionuclide. An acceptable ligand is one that forms a stable radiometal chelate with h igh thermodynamic stability and kinetic inertness. 47 Therefore, the choice of the chelating ligand will depend on the coordination chemistr y between the radionuclide and ligand. Two prominent classes of bifun ctional chelating ligands for l anthanide 3+ cations are poly(amino phosphonate)s (Figure 1 7 ) and poly(amino carboxylate) ( Figure1 8 ). These ligands have been used extensively for their ability to form thermodynamically stable and kinetically inert chelates for a wide variety of metals, due to the chelating abilities of the multidentate ligands. 44 54 Figure 1 7 Structure of polyamino phosphonate ligands Fig ure 1 8 Structure of polyamino carboxylate ligands

PAGE 36

36 CHAPTER 2 SYNTHESIS, CHARACTER IZATION, AND PRE CLINICAL EVALUATION OF A POLY M ER BASED RADIOPHARMACEU TICAL FOR INTERAL RA DIONUCLIDE THERAPY History of Polyethyleneimine Methylenephosohonate (PEI MP) The small molecule radiopharmaceutica l, ethylenediamine tetramethylene phosphonate ( 153 Sm EDTMP), trade name Quadramet was int roduced in Chapter 1 (Figure 2 1 A ). The therapeutic efficacy of Quadramet is in part limited by uptake in healthy tissu e. In an attempt to overcome this shortcoming of Quadramet Milner et al. produced the polymer based radiopharmaceutical polyethyleneimine m ethylene phosphonate (PEI MP) (Figure 2 1B) designed to target osteosarcoma with more selectivity than small molecule radiopharmaceuticals The presumable targe ting mechanis m of action of the polymer drug, PEI MP, is the enhanced permeability and reten tion (EPR) effect described in Chapter 1 Figure 2 1 A) The small molecule radiopharmaceutical, Quadramet and B) polymer radiopharma ceutical, PEI MP Extensive pre clinical work has established that the synthetic polymer, polyethyleneiminomethyl phosphonic acid (PEI MP), tagged with the diagnostic, gamma emitting radionuclide technetium 99m ( 99m Tc ) shows significantly better retention in canine osteosarcoma and lower localization in healthy, non target tissue than the c urrent FDA approved radioligand Quadramet (Figure 2 2 ). 21 55 57 Of B A

PAGE 37

37 particular significance is the reduced uptake in the dose limiting organs of healthy bone and the associated bone marrow. Figure 2 2 Organ dose distribution for 99m Tc PEI MP versus 153 Sm EDTMP adapted fro m Milner, R. J.; Dormehl, I. C.; Lowe, W. K. A.; Kilian, E. Eur J Nucl Med 1999 26 1220 Additionally, the polymer radiopharmaceutical 99m Tc PEI MP demonstrated a more consistent tumor to healthy bone uptake ratio compared to small molecule ligands. The variation in uptake is significant and is best understood by looking at the coefficient of variance (SD/Mean x 100) for uptake. The coefficient of variance for small molecule radioligands currently on the market are 39.04% for 153 Sm EDTMP (Quadramet ) and 52.57% for 99m Tc MDP compared to 6.23% for the new polymer ligand, 99m Tc PEI MP. This is a significant improvement in biodistribution and can be attributed to a more consistent uptake ratio of the polymer ligand, PEI MP, and its hypothesized mechanism of action of enhanced permeability and retention (EPR) in tumo r vasculature. It has been reported that PEI MP particle sizes obtained by ultrafiltration fractionation through polyethersulfone membranes wit h molecular weight cut offs of 2 0

PAGE 38

38 30 kDa provided the best blood clearance of the drug via the kidneys. 57 In later work, a blood plasma model was constructed to determine the viability of delivering lanthanide radionuclides with therapeutic potential to bone tumors via the polymer ligand PEI MP. 58 This blood plasma model was designed to predict metal ion speciation in plasma, giving an indication of the ability of PEI MP to survive competition for the radionuclide by blood plasma ions and oth er biological chelators. Based on this computational model, q uestionable retention of t he therapeutic radiolanthanides holmium and samarium by PEI MP was expected in vivo The poor retention of lanthanide ions was attributed to PEI the blood plasma ion calcium. It was estimated that 82.2% of PEI MP would be bound by calcium, leaving almost all of the lanthanide free to bind with the competitive chelator citrate. 58 Clini cal studies were conducted in the canine model to compare the delivery of the diagnostic radionuclide, 99m Tc to the therapeutic radionuclide, 153 Sm In this clinical study, lower bone uptake was observed for 153 Sm PEI MP compared to 99m Tc PEI MP. Figur e 2 3. Organ dose distribution for 99m Tc PEI MP versus 153 Sm PEI MP This low uptake of the lanthanide is attributed to the idea that very little sama rium remained bound to the bone seeking ligand, PEI MP. Additionally, higher uptake of

PAGE 39

39 153 Sm PEI MP was obs erved in the liver and blood pool, indicating that 153 Sm is no longer bound to the polymer and is free to be distributed thro ughout the entire body (Figure 2 3) Encouraged by previous development s we have further investigate d PEI MP as a potential polym er radiopharmaceutical for the e n hanced permeability and r etention ( EPR ) delivery of radionuclides to solid tumors We have re evaluated the osteosarcoma targeting capability of PEI MP and have further investigate d the tumor targeting mechanism of PEI MP b y studying the biodistri bution in tumor types other than osteosarcoma Experimental Materials Polyethyleneimine (Lupasol Water free) was supplied by BASF (Ludwigshafen, Germany) Serum vials, septa and aluminum caps used in preparation of drug kits were received from Voigt Global, Inc. (Lawrence, KS). Gel clot endotoxin assay kits (ToxinSensor TM ) were purchased from GenScript. Ultrafiltration centrifugal membranes (Omega TM ) were supplied by Pall Corporation. All other reagents and chemicals were used as received from Aldrich Chemical Company unless otherwise noted. Instrumentation and Analysis 1 H NMR, 13 C NMR and 31 P NMR spectra were recorded on a Varian Associates Mercury 300 spectrometer. Chemical shifts from 1 H and 13 C NMR were referenced to resid ual signals from CDCl 3 ( 1 H = 7.24 and 13 C NMR = 77.23) and to an internal standard of H 2 PO 3 ( 31 P = 0 ) for 31 P NMR. High resolution mass spectra (HRMS) were obtained on a Finnegan 4500 gas chromatograph /mass spectrometer using the chemical ionization mode

PAGE 40

40 Synthesis of N trimethylenephosphonate P olyethyleneimine (2 1B ). A mixture of phoshporous acid (18.4 g, 4.3 eq) and concentrated hydrochloric acid (51.3 mL, 11.9 eq) was stirred at 80 C under argon atmosphere until dissolution of the phosphorou s acid was complete. Formalin (23.3 mL, 1.0 eq) was added to the reaction mixture dropwise over a thirty minute period. The reaction was brought to reflux at 90 C and a solution of polyethyleneimine (8.33 g in 40 mL of water, 1.0 eq) was added slowly. The reaction was stirred under reflux for an additional hour and then cooled slowly to room temperature overnight. The product was precipitated by addition of distilled water. The crude product was dissolved in a minimal amount of saturated sodium carbona te. The final polymer was collected via lyophilization. 1 H NMR (D 2 O): broad signal 2.8 3.8 ppm (CH 2 (ethyleneimine and methylenephosphonate)). 31 P NMR (D 2 O): single broad peak centered at 8 ppm. IR (cm 1 ): 3300 3500 (st. OH, st. NH), 2850 2930 (st. CH). Carbon:Nitrogen molar ratios were determined for the starting polymer (PEI) and the functionalized polymer (PEI MP) using elemental analysis. On the basis of the empirical formula for the repeat unit of PEI MP, the degree of methyl phosphonation was deter mined to be 74%, taking into account a 1 : 2 : 3 amine ratio of 1: 1.20: 0.76. Polyethylenimine (WF, Lupasol, BASF): Calculated for C 6 H 15 N 3 : C, 55.78; H, 11.70; N, 32.52 Found: C, 53.74; H. 11.80; N, 31.01. Polyethyleneimine methylenephosphonate: Calcu lated for C 9 H 18 N 3 O 9 P 3 : C, 26.68; H, 4.48; N, 10.37 Found: C, 17.22; H, 4.96; N, 6.78. Particle S ize S eparation by U ltraf iltration Fractionation of PEI MP was achieved using polyethersulfone (Macrosep Pall Life Sciences) Centrifugal Filters. Approximate ly 0.25 g of polymer was dissolved in 20

PAGE 41

41 mL of endotoxin free water (certified < 0.1 EU/mL). Polymer solutions were passed through 30 kDa and 10 kDa cut off filters sequentially. Filtration was determined to be complete once no more polymer was eluting t hrough the filter. Endotoxin T esting Fractionated PEI MP (10 30 kDa) was tested for endotoxins prior to kit preparation using a LAL gel clot assay kit (ToxinSensor TM GenScript) with a detection limit of 0.25 EU/mL. A serial dilution with endotoxin fre e water was performed on the product to prevent any chemical inhibition of the LAL reaction. The E. coli standard (0.5 EU/mL) provided in the kit was reconstituted by addition of 1 mL LAL reagent water and vortexed for 15 minutes. The provided LAL was r econstituted by addition of 2 mL of LAL reagent w ater and stirring gently for approximately 30 seconds. Into each of the endotoxin free vials provided for testing, 0.1 mL of LAL solution was dispensed. To each vial of LAL solution, 0.1 mL of diluted prod uct was added. The solutions were placed in a 37 C water bath for one hour and then checked for clotting. Flow of the gel upon inversion of the vials indicated a negative result, whereas gelation was indicative of a positive result. 99m Tc PEI MP Kit P re paration Lyophilized kits of the polymer ligand, PEI of a solution containing 250 mg of SnCl 2 2 H 2 O crystals in 0 5 ml HCl (c) with an aqueous PEI MP solution (48 mg Na PEI MP / mL pH = 8 0) whereafter the pH was adjusted to 6. The mixture was dispensed into 5 vials and fre eze dried. The vials were fitted with septa and sealed with an aluminum cap.

PAGE 42

42 The radionuclide labeling was performed under contract by Cardinal Health Inc., in Gainesville, FL. T he final solutions for injection were prepared by adding 0.3 ml 99m Tc (111 to 185 MBq) to the above vials. Radiochemical purity was checked using thin layer chromatography on silica gel impregnated glass fiber sheets as a stationary phase and acetone and 0.9% sodium chloride solution for a mobile phase. Biodistribution Studies Bio distribution studies were achieved through collaboration with the College of V eterin arian M edicine at the University of Florida. In this study, client owned dogs with eithe r a sarcoma or carcinoma were enrolled i n the trial. Enrollment was tailored to in clude important tumor type s e.g. osteosarcoma, hemangiosarcoma, hepatic carcinomas, and pulmonary carcinomas. The dogs were given a diagnostic dose of 99m Tc PEI MP of between 333 925 MBq p er animal. Scintigraphy was performed in the Radiology Section of the Veterinary Medical Cen ter (VMC). The gamma camera was centered over the thorax and abdomen to include the heart, lungs, liver, both kidneys, a nd tumor. Both forelimbs were included to acquire data from the metaphysis of distal radius and cortical bone area. Data acquisition was performed as a dynamic study of 10 x 1 minute frames followed by single two minute frames at 15, 30, 45, 60, 120, and 180 minutes on a countdown bolus injection of the radiopharmaceutical (matrix 64x64). Urine was collected for the duration of the scan The activity and volume (total urine volume for each time point) of each sample was measured and registered. Particle Size Determination The hydrodynamic radii of PEI MP( < 1 0 kDa) PEI MP(10 30 kDa) and PEI MP( > 30 kDa) fractio ns and their complexes with Ca(II), Mg(II), Zn(II), Sm(III), and Y(III) were measured by dynamic light scattering (DLS ). Mean particle size and size distribution of

PAGE 43

43 was measured using dynamic light scattering (DLS) on a Nanotrac ULTRA with an external pro be (Microtrac, Largo, FL, USA). Each sample was analyzed in triplicate and each replicate was measured six times to yield the average particle size. This was performed using stock solutions which containe d ligand concentrations of 10 m M for all combinatio ns of PEI MP In addition, the ionic strength of the 10 mM solutions were controlled by addition of 0.15 M NaCl. Zeta Potential Measurements Polymer ligand charge stability was analyzed using a ZetaPlusTM zeta potential analyzer (Brookhaven Instruments Co rporation, Holtsville, NY). The data was processed using the ZetaPlus sizing software V2.27. Results and Discussion Synthesis and Structural Analysis The base poly mer selected for the design of the polymer radiopharmaceutical PEI MP was polyethyleneimine ( PEI). PEI is a highly water soluble polymer prepared by cationic polymeriz ation of ethyleneimine (Figure 2 4 ). Poly ( ethyleneimine ) (PEI) is an aliphatic polymer containing primary, secondary and tertiary amines in an approximate ratio of 1:2:1. It has bee n shown that PEI is relatively safe for internal use in animals and humans. 59 PEI has been explored as a model water soluble carrier for deliv ery of small molecule pharmaceuticals d ue to its properties of hydrophilicity, biocompatibility and thermal stability 59 60 Additionally, the branched structure of PEI has been s hown to increase the blood circulation half life relative to other polymer architectures with similar molecular weight and chemistry. 39 The slower rate of renal filt ration could be attributed to a hindered ability of the polymer to reptate through glomerular pores of the kidneys

PAGE 44

44 Figure 2 4. Synthesis of branched polyethyleneimine PEI MP (Figure 2 1 B) consist of a branched poly(ethylen eimine) derivative in which functional groups have been att ached to the polymer backbone. Synthesis of polymer 2 1 B PEI MP, was achieved using Mannich type reaction chemistry described by Moedritzer and Irani (Figure 2 5 ). 21 56 57 61 Reaction of polyet hyleneimine with concentrated hydrochloric acid, phosphorous acid, and formalin at reflux temperature results in the formation of aminomethylenephosphonic acid functionalized polymer 2 5B Upon cooling, 2 5B settled to the bottom of the reaction vessel as a viscous oil, which transforms to a doughy mass upon washing with distilled water. Purification is relatively simple, as repeated washing with fresh DI water removes residual starting material and reagents. Dissolution of 2 5B in a minimal amount of 1 M sodium carbonate solution results in the formation of the water soluble polymer salt 2 5C The final polymer 2 5C was collected via lyophilization as a light brown solid. The aminomethylphosphonic acid mo iety has been shown to exhibit favorable chel ating properties for polyvalent metal ions. 62 63 The incorporation of the phosphonate In itiation: Propagation: Branching:

PAGE 45

45 functionality is key to the success of this material, allowing polymer 2 5C to act as a carrier for EPR delivery of radionuclides to vascularized tumors. 64 Figure 2 5. Synthesis of polyethyleneimine methylenephosphonate (PEI MP) 31 P NMR analysis was used for functional group detection. 31 P NMR revealed the presence of a single broad peak, indicating that all of the phosphorous was bound to the polymer. Broadening of the spectral l ines is expected and is associated with the ionic interactions accompanying the polymer. The broad signal centered at 8 ppm is representative of the shifts reported for aminomethylphosphonic acids at pHs 6 8. 63 Further structural confirmation was provided by obser ved changes in the N H and P O stretching region s in the FT IR spectrum of polymer 2 5C compared to starting material 2 5A The FT IR spectrum of the starting material, PEI, shows an N H stretch of the primary amine at 2809 cm 1 and the N H bending of the primary amine at 1578 cm 1 (Figure 2 6). The strong, broad band at 693 cm 1 is representative of the N H wag of primary and secondary amines When PEI is methyleneph osphonate functionalized, we see the disappearance of the strong N H stretch of primary am ine peaks located at 2809 cm 1 with only the presence of a broad stretch associated with secondary amines being observed around 3280 cm 1. Potential P O stretching is also observed around 971 cm 1 (Figure 2 7) A B C

PAGE 46

46 Figure 2 6. Infrared spectrum of starting material, polyethylenimine Figure 2 7. Infrared spectrum of methylene phosphonate functionalized polyethyleneimine Carbon:n itrogen mola r ratios were determined for starting polymer 2 5A and functionalized polymer 2 5C using elemental analysis. On the b asis of the empirical formula for the repeat unit of structure 2 5C the degree of methyl phosphonation was determined to be approximately 74%, taking into account a 1 : 2 : 3 amine ratio of

PAGE 47

47 1: 1.20: 0.76 as provided by the supplier of starting mater ial 2 5A For this calculation, the repeat unit of PEI MP was broken down into three segments, one for each type of amine (primary, secondary, or tertiary). The calcula tions were conducted as shown in the Appendix P olymer Filtration Dormhel et al ., previo usly investigated the biodistribution and pharmacokinetics of polyethyleneiminomethylene phosphonate (PEI MP) complexed with the emitting radionuclide 99m Tc 57 The experiments were conducted by injecting the polymer radionuclide complex into the norm al chacma baboon ( papio ursinus ) and obtaining gamma ray images of the entire animal for up to 4 hours after injection. An optimal molecular size of polymer radionuclide complex was determined for maximal protection of healthy tissue and organs. It was d etermined that a 10 30 kDa fraction would be optimum due to relatively low accumulation in normal bone, liver and kidney. Based on this previous study, polyethyleneiminomethyl phosphonic acid was fraction ated through polyethersulfone u ltrafiltration membra nes prior to clinical evaluation. It should be noted here that fractionation by ultrafiltration provides fractions based on the nominal pore size of the filter rather than the molecular weight of the polymer. Although the pore size and molecular weight do not necessarily correspond, this method does provide qualitative insight into the performance of different particle sizes Fractions were investigated by dynamic light scattering to correlate particle size and nominal pore size. The filtrations were cond ucted by sequentially passing a solution of polyethyleneiminomethyl phosphonate in endotoxin free water through polyethersulfone

PAGE 48

48 ultrafiltration membranes with mole cular weight cut offs of 30 kDa and 10 kDa using either a stir cell apparatus operated at 4 psi or ultrafiltration centrifugal filters. Endotoxin Testing Any substance capable of inducing a fever is termed a py rogen. One subset of pyrogens capable of inducing fever is endotoxins. The high molecular weight complex atural lipopolysaccharide (LPS) complex found in the outer cell wall of bacteria. Endotoxins aid in the survival of bacteria by providing structural integrity and assisting in transport activity. In 19 12, Hort and Penfield showed that this endotoxin comp lex exists only in the cell wall of gram negative bacteria and that the pyrogenic activity of distilled water is related to the microbial count. 65 They also demonstrated that dead bacteria had as much pyrogenic activity as living bacteria. As little as a few nanograms of endotoxin are enough to elicit an immunological response, making endotoxin contamination control a priority in t he parenteral manufacturing of pharmaceuticals. FDA guidelines validate the use of end product Limulus Amebocyte Lysate (LAL) testing for indication of contaminants. Water for injection has an allowable endotoxi n limit of 0.25 Endotoxin units (EU)/mL. 66 There are currently four commercially available m ethods approved by the FDA for e nd product release testing: (1) gel clot, (2) spectrophotometric, (3) colorimetric (Lowry protein), and (4) chromogenic assay. It should be noted that chemical or physical factors are known to interfere with the Chemical inhibition can be caused by chelation of cations necessary for the LAL reaction by ligands such as EDTA. This type of chemical inhibition may be overcome by dilution of the product 66 66

PAGE 49

49 A serial dilution was conducted on PEI MP with endotoxin free water to prevent any chemical inhibition of the LA L reaction. As indicated in T able 2 1 below, a negative resul t was obtained for all concentrations of the product samples. Table 2 1. Endotoxin assay results for PEI MP Sample Concentration Result 1 (+) Control positive 2 ( ) Control negative 3 10 mg/mL negative 4 5 mg/mL negative 5 2.5 mg/mL negative 6 1. 25 mg/mL negative 7 0.625 mg/mL negative 8 0.3125 mg/ mL negative Preparation of 99m Tc PEI MP Radiopharmaceutical Kit Many precautions must be taken in the preparation of radiopharmaceutical kits. Care must be taken to use aseptic con ditions to avoid introducing pathogens into the injection solution after it has passed endotoxin testing After endotoxin testing, PEI MP was subsequently labeled with technetium 99m ( 99m Tc), a gamma emitting, diagnostic radionuclide. 99m Tc was chosen for its ideal nuc lear properties. 99m Tc has a half life of six hours, providing sufficient time for labeling but minimizing radiation exposure to the patient. Additionally, the main gamma emission of 99m Tc is on the order of 140 k eV which is an ideal energy for absorption by gamma cameras used for imaging. 67 These properties have led to 99m Tc being a central component in nuclear medicine, comprising nearly 80% of radiopharmaceuticals in clinical diagnostic use. 68

PAGE 50

50 99m Tc is obtained from a Mo/Tc generator. The 99m Tc eluted from the generator is in the pertechnetate form, TcO 4 which has a charge of 1 and an oxidation state of +7. Since the negative charge on the pertechnetate will have a repulsive interaction with the negatively ch arged phosphonate groups on the chelating polymer, it is necessary to use a reducing agent to convert the 99m Tc into a useful chemical form for binding. Stannous chloride is most often employed in the reduction methodology. In the reduction/oxidation reac tion below, we see that the reducing agent, Sn 2+ is oxidized to the stannic ion, Sn 4+ while TcO 4 is reduced to Tc 4+ (Figure 2 8 ) Figure 2 8 Reduction/oxidation reaction of stannous chloride Care must be taken to not intr oduce air into the radiopharmaceutical vial, as the oxygen content in 0.1 mL of air is enough to destroy the stannous ion used as a reducing agent in many commercially available cold kits. 69 Upon reduction, 99m Tc is in a favorable chemical form for the formation of a chelate by mixing with the negatively cha rg ed phosphonate polymer Quality control of radiochemical purity of 99m Tc chelates was achieved through the use of thin layer chromatography. The radiochemical impurities of 99m Tc include free pertechentate of the chemical form TcO 4 and hydrolyzed reduced Tc of the chemical form TcO(OH ) 2 H 2 O. The thin layer chromatography systems consist of paper backed silica gel as the stationary phase and acetone/saline as the mobile phase. Where hydrolyzed reduced form of Tc remains at the bottom half of the silica plate, while all 3 Sn 2+ 6 e 3 Sn 4+ 2 TcO 4 + 16 H + + 6 e 2 Tc 4+ + 8 H 2 O Overall 3 Sn 2+ + 2 TcO 4 + 16 H + 3 Sn 4+ + 2 Tc 4+ + 8 H 2 O

PAGE 51

51 other forms a re moved by saline to the top half of the plate. In addition, any free Tc will be moved to the top half of the silica plate by acetone, while bound Tc will remain at the bottom half. Lyophilized kits of the polymer ligand, PEI MP, and reducing agent, SnCl 2 were prepared in aluminum capped septa plugged vials filled with nitrogen gas and stored in the freezer until further use (Figure 2 9 ). Figure 2 9 Image of lyophilized radiopharmaceutical kits containing PEI MP and SnCl 2 Particle Size Analysis The importance of particle size for tumor targeting polymer drugs was addres sed in Chapter 1 The polymer particles must be of an appropriate size to maintain balance between three key events: 1) large enough to prevent uptake in healthy tissue, 2) small enoug h to enter the tumor volume, and 3) small enough for excess in blood to be eliminated through kidney filtration It has been established that polymer particles with average molecular weights of around 40 kDa work well for drug delivery applications employi ng the EPR effect. 35 Previous studies with PEI MP have shown that particles obtained by sequential ultrafiltration through filters with molecular weight cut offs o f 10 30 kDa demonstrate the best balance between the aforementioned key events. In an attempt to correlate particles size with nominal filtration pore sizes, dynamic light

PAGE 52

52 scattering (DLS) was conducted on fractions of PEI MP that had been filte red sequent ially through 30 kDa and 10 kDa ultrafiltration membranes. Dynamic light scattering is a useful tool for measuring particle size. The technique measures the Brownian motion of partic les in solution and correlates t he detection signal to the hydrodynamic r adius. Since Brownian motion is the movement of particles in solution due to the bombardment of solvent molecules, larger particles will move much slower in solution that smaller particles. The size of the particle is calculated using the Stoke s Einstein E quation 2 1. (2 1) temperature, is the viscosity, and D is the translational diffusion coefficient. The hydrodynamic radius that is measured equates to the diameter of a unit density sphere that has the same translational diffusion coefficient. Factors that affect the diffusion speed o f the particle include the ionic strength of the medium, the surface structure of the particle, and the deviation from a true spherical shape. Controlling ionic strength in dynamic light scattering is crucial, as the total ionic concentration in the medium affects the thickness of the electric double layer, which in turn affects the diffusion speed. If the conductivity of the medium is low, the diffusion speed will be reduced resulting in an apparent larger hydrodynamic radius. 70 To control this factor, all light scattering studies were conducted in 0.15 M NaCl solutions to best mimic physiological conditions. Any changes in the particle surface and hence conformation of the polymer due to intera ctions in solution or pH could also affect particle size. To establish the effect of ion binding on conformation and particle size, d(H) = kT 3

PAGE 53

53 DLS was conducted on PEI MP in the presence of samarium, as well as the blood plasma ions calcium, magnesium, a nd zinc. It m ust be kept in mind that working with non sp h erical particles is an inherent problem with measuring particle size by DLS, and t he measured hydrodynamic radius is a value relative to a perfect sphere with the same diffusion speed. Polyethyleneimine (PEI) i s prepared by a chain growth addition type polymerization through cationic initiation of ethylene imine (Figure 2 4) As expected for a chain growth polymer, PEI has a broad polydispersity index (PDI). This broad PDI also corresponds to a broad particle s i ze range as depicted below in F igure 2 10 The average particle size for the base polymer PEI (lupasol WF) in 0.15 M NaCl solution was measured to be 2.996 nm with a particle size range from 1.4 to 6.39 nm (Figure 2 10) Figure 2 10. Particle size distri bution for polyethyleneimine (Lupasol WF) Post polymerization modification of PEI through addition of methylene phosphonate groups to produce polyethyleneimine methylene phosphonate (PEI MP) changes the surface structure and correspondingly the hydrodynam ic radius of the

PAGE 54

54 polymer as depicted by the particle size distribution show n in F igure 2 11 The parti cle size range for PEI MP is thus larger at 2.60 to 2 7.2 nm Figure 2 11 Particle size distribution for polyethyleneimine methylene phosphonate (PEI M P) in 0.15 M NaCl The conformation of polyelectrolytes are known to change in the presence of additional ions. 71 It is expected that the hydrodynamic radius will decrease in the presence of an oppositely charged ion, as the inter chain i nteractions induce tighter packing of the chains that are otherwise solvated. This effect is expected to decrease as the ratio of metal to ligand approaches one, at which point the hydrodynamic radius will more resemble that of the free polymer ligand. In order to understand the effect of ionic binding on the particle size of PEI MP, DLS studies were conducted with PEI MP in the presence of s amarium, as well as the blood plasma ions calci um and magnesium (Figure 2 12) No significant change in particle siz e was observed by addition of blood plasmas ions. The particle sizes of PEI MP used in clinical studies is altered by a fractionation process to narrow the polydispersity index and obtain desired particle size for drug delivery via the enhanced permeabilit y and retent ion (EPR) effect The particle size distributions for PEI MP fractionated through nominal pore sizes of >30 kDa, 30 10 kDa,

PAGE 55

55 and 10 3 0 kDa are shown below (Figures 2 13, 2 14, 2 15) and are summarized in T able 2 2 Figure 2 12. Comparison plot for PEI MP (red) particle sizes in the presence of the ions calcium (green), magnesium (yellow), and samarium (blue) Figure 2 13. Particle size distribut ion of PEI MP retained during filtration through a 30 kDa MW cut off filter Figure 2 14. Particle size distribution of PEI MP retained after sequential filtration through 30 and 10 kDa MW cut off filters

PAGE 56

56 Figure 2 15. Particle size distribution of PEI MP that passed through a 10 kDa MW cut off filter Table 2 2. Average particle sizes and particle size distributions for PEI MP fractions There appears to be a significant amount of overlap in particle sizes for each fraction, although the average particle size is reduced upon the reduction of the nominal pore size of the filter (Figures 2 13, 2 14, 2 15). The overla p of particle sizes indicates insufficient filtration of the polymer and may be partly attributed to the morphology of the polymer PEI MP. The highly branched structure of the polymer allows for hindered reputation through pores as described by Frechet (F igure 2 16). 38 It is this characteristic that makes the branched structure attractive for EPR delivery applications, as it leads t o increased biological half life in the blood plasma. However, this feature also leads to difficulty in achieving a clean separation by means of ultrafiltration. Polymer Fraction (kDa) Particle Size Range (nm) Average Particle Size PEI MP < 10 2.46 13.93 3.71 PEI MP 10 30 2.46 13.93 5.54 PEI M P > 30 4.92 30.0 8.6

PAGE 57

57 Figure 2 16. Hindered reputation of a branched polymer through pores as described by Frechet et al adapted from Fox, M. E.; Szoka, F. C.; Frechet, J. M. Accounts of chemical research 2009 42 1141. Figure 2 17. Particle size distribution of aggregated PEI MP in solution One concern related to particle size in drug delivery applications is th e potential for aggregation. Particle sizes larger than 200 nm will be trapped by the liver, resulting in excessive radiation exposure to this non target ed organ. Shown below is the particle size distribution for a PEI MP solution in which aggregation was forced by sonication of the sample (Figure 2 17 ) The stability of PEI MP particles were analyzed and will be discussed in the following section Particle Stability The balance between chain chain and chain solvent interactions in a polymer system are the driving forces for polymer aggregation. Polyethylenimine is known to form strong coordination complexes. At low acidic pH values the amines in polyethyleneimine are protonated to form a positively c harged polyelectrolyte. At high basic pH values polyethy lenimine is neutrally charged. 72 It has been shown that for

PAGE 58

58 linear polyethylenimine, approximately 55% of the amine groups are protonated under phy siological conditions. 73 Therefore, the repeat unit structure of polyethyleneimine methylenephosphonate ( PEI MP) most likely exists under physiological conditions in a zwitte rio nic form as shown in F igure 2 18 Figure 2 18 Zwitterionic form of polyethyleneimine methylenephosphonate likely to exist under physiological conditions Aggregation of polyethylenei mine in the presence of negatively charged phosphonate ions has been attributed to the attraction between the positively charged polyamine and the neg atively charged phosphonate and/ or the formation of hydrogen bonding. 72 To better understand the tendency of PEI MP to aggregate upon vigorous shaking associated with the radionuclide labeling step, particle stability measurements were conducted. Zeta potential measurements were conducted to determine if the charge interactions between quaternary ammonium ions and phosphonate ions are likely to lead to aggregation of the polymer particles. The zeta potential refers to the charge that exists around a particle in solution. There are multiple lay ers of charge that surround a particle in solution. The innermost layer of ions is tightly associated with the charged particle. This dense layer of ions is called the Stern layer and has a specific electrical charge. The stern layer is surrounded by anoth er more diffuse layer called the Gauy Chapman layer, which has a specific charge all its own. Additionally, the bulk of the solvent will have a separate charge from

PAGE 59

59 that of the Stern and Gauy Chapman layer. The difference in charge between the Stern layer and the bulk of the solvent is the zeta potential. When polyethyleneimine is functionalized with phosphonate groups, the potential difference between the polymer and the bulk solvent is changed. The stability of the polymer is dependent on the zeta potent ial of the system. Generally, as the zeta potential becomes more negative, the system becomes more stable or less susceptible to agglomeration. It is generally accepted that systems with zeta potentials more negative than 20 to 30 millivolts have suffici ent repulsion to be stable, with systems being considered extr emely stable at values greater than 45 millivolts. 74 Figure 2 19. Zeta potential for PEI MP in 0.15 M NaCl at pH 6 8 The zeta potential of PEI MP in 0.15 M NaCl was measured to determin e if aggregation due to particle charge interactions is likely. This system was chosen because it is believed that aggregation is more likely to occur in the more concentrated conditions encountered in the pharmaceutical kit rather than under the dilute co nditions

PAGE 60

60 found in the blood stream. The average zeta potential for 10 runs was measured to be 52.43 millivolts, indicating that the PEI MP particles are expected to be stable in solution (Figure 2 19 ). Animal Care and Procedures Biodistribution studies were conducted with approval from the Institutional Animal Care and Use Committee (IACUC) at the University of Florida. IACUC approved facilities were used for testing and subsequent housi ng of animals. Biodistribution Studies Biodistribution studies are required to establish the in vivo performance of a radio pharmaceutical. As discussed in Chapter 1 it is necessary to balance three key events in polymer based tumor therapy: 1) uptake with in the tumor volume, 2) clearance of excess drug from the blood via kidney filtration to urine and 3) minimizing uptake in healthy, non target ed organs. The an imal model used in evaluation of the b iodistribution of PEI MP was naturally occurring osteosar coma of the dog Naturally occurring cancers in canines and humans share many common features, including histology, genetics, biological behavior, and response to conventional treatments. 46 Canine stud ies therefore provide valuable insight into the development and translation of new cancer therapies by defining toxicity, activity and pharmacokinetics. 46 The goal of PEI MP biodistribution studies i n the canine osteosarcoma model was to replicate the data previously reported by Milner et al. For this study six client owned dogs were enrolled in the trial. Ethical approval and client consent was obtained prior to enrollment in the study. All normal d ogs were subjected to the same experimental conditions. Sedative was administered via an indwelling intraven ous catheter The

PAGE 61

61 animals were placed in either right lateral or left lateral position over the gamma camera (Figure 2 20 ) Figure 2 20. Position ing of dog for scintigraphic imaging 99m Tc PEI MP was injected intravenously with a bolus of 370 5 5 5 MBq All 99m Tc PEI MP kit s were prepared as described previous ly in C hapter 2 with ultrafiltration being used to control the particle size. Data acquisit ion was p erformed as a dynamic study of 3 0 x 1 minute fram es followed by single one and a half minute frames at 30, 60, 90, 120, and 180 minutes on a countdown bolus injection of the radiopharmaceutical Urine samples were collected via a urinary catheter for the purpose of determining percent total body retention of the injected radiopharmaceutical. From the images collected, regions of interest (ROI) were drawn over the organs of interest: heart, lungs, liver, kidneys, trabecular spongiosa, and cortical b one A decay correctio n for physical decay was used to account for the exponential decrease in activity of the radionuclide activity curve s were created for e ach study. Percent organ activity (%O A ) distributi on and the per cent injected activity (%IA ) distribution at three hours was calculated from the dose corrected ROI data and percent activity excreted in the urine

PAGE 62

62 Table 2 3 lists the tumor location and administered dose for each osteosarcoma study. The scintigraphic images and time activity curves (TAC) for six osteosarcoma bearing dogs administered 99m Tc PEI MP are shown below in Figures 2 21 through 2 31 Table 2 3. Tumor locations for osteosarcoma biodistribution studies with PEI MP F igure 2 21. Scintigraphic image and time activity curve for organs of interest for 99mTc PEI MP administered to a dog with osteosarc oma in the left distal radius (C ase 1, Table 2 3) Tu mor Location Administered Activity (MBq) Case 1 Left distal radius 227 Case 2 Right distal radius 157 Case 3 Right proximal humerus 231 Case 4 Left proximal humerus 328 Case 5 Right distal radius 316 Case 6 Right distal radius 258

PAGE 63

63 Figure 2 22 Scintigraphic image and time activity curve for tumor region for 99mTc PEI MP administered to a dog with osteosarcoma in the left dista l radius (C ase 1, Table 2 3) Tumor

PAGE 64

64 Figure 2 23 Scintigraphi c image and ti me activity curve for organs of interest for 99m Tc PEI MP administered to a dog with osteosarco ma in the right distal radius (C ase 2, Table 2 3) Blood pool and Liver

PAGE 65

65 Figure 2 24 Scintigraphic image and time activity curve for tumo r region for 99m Tc PEI MP administered to a dog with osteosarco ma in the right distal radius (C ase 2, Table 2 3) Tum or z zzzzzzzzz

PAGE 66

66 Figure 2 25. Scintigraphic image and time activity curve for organs of interest for 99mTc PEI MP administered t o a dog with osteosarcoma in the right proximal humerus (C ase 3, Table 2 3) Liver Kidneys Blood pool

PAGE 67

67 Figure 2 26 Sc intigraphic image and time activity curve for tumor region for 99m Tc PEI MP administered to a dog with osteosarcoma in the right prox imal humerus (C ase 3 Table 2 3 ) Tumor

PAGE 68

68 Figure 2 27 Scintigraphic image and time activity curve for organs of interest for 99m Tc PEI MP administered to a dog with osteosarcoma in the left proximal humerus (C ase 4, Table 2 3) Liver Tumor Kidneys

PAGE 69

69 Figure 2 28 Sc intigraphic image and time activity curve for tumor region for 99m Tc PEI MP administered to a dog with osteosarcoma in the left proximal humerus (C ase 4, Table 2 3) Tumor Liver Kidneys

PAGE 70

70 Figure 2 29 Scin tigraphic image and time activity curve for organs of interest for 99m Tc PEI MP administered to a dog with osteosarco ma in the right distal radius (C ase 5, Table 2 3) Kidneys

PAGE 71

71 Figure 2 30 Scintigraphic image and time activity curve for tumor region for 99m Tc PEI MP administered to a dog with osteosarco ma in the right distal radius (C ase 5, Table 2 3) Tumor

PAGE 72

72 Figure 2 31 Sc intigraphic images and time activity curves for 9 9m Tc PEI MP administered to a dog with osteosarcoma in the right distal radius (C ase 6 Table 2 3 ) Rapid clearance from both the central blood pool and non target ed organs was observed f or all osteosarcoma bearing canines administered a dose of 99m Tc PEI M P, as seen by the exponential decay in the time activity curves of F igures 2 21 to 2 31 An additional measure of the clearance of the radiopharmaceutical from non target ed organs is the percent of the injected dose that is excreted through the urine at th ree Tumor Liver Kidneys Thyroid

PAGE 73

73 hours post injection The values for percent injected dose ex creted at 3 hours are given in Table 2 4 with a n average of 38.6% of injected activity being excreted for the four cases measured (Table 2 4 ) It is desired that any excess radiopharmaceuti cal not taken up by the tumor be excreted through the kidney filtration to minimize radiation exposure to non target ed organs. This excreted value is lower than the cleara nce rate of about 70% previously observed by Milner et al. indicating higher retenti on in the central blood pool and non target organs than desired. Table 2 4 Percent injected activity excreted through the kidneys for 99m Tc PEI MP administered to dogs with osteosarco ma T he relative percent of 99m Tc PEI MP at three hours in each organ of interest is given in Table 2 5 For C ases 5 and 6, significant retention in the kidney s and liver was observed, indicating an increase in particle size over the optimal size range es tablished in previous studies Large particle size can be due to poor filtration of the starting material or aggregation of the part icles in solution H owever, b ased on the zeta potential values for PEI MP (Figure 2 19) the particles are expected to be st able against aggregation, indicating poor filtration as a cause for high retention. This result indicate s Case Tumor Type Tumor Location %IA Excreted Case 1 Osteosarcoma Left distal radius 36.6 Case 2 Osteosarcoma Right distal radius 32.95 Case 3 Osteosarcoma Right proximal humerus 41.22 Cas e 4 Osteosarcoma Left proximal humerus 43.65 Case 5 Osteosarcoma Right distal radius N/A Case 6 Osteosarcoma Right distal radius N/A

PAGE 74

74 that future work is needed to optimize particle size for improved clearance of excess radiopharmaceutical from the blood Table 2 5 Percent organ act ivity distribution at 3 hours for 99m Tc PEI MP administered to osteosarcoma bearing canines Heart Lungs Liver Left Kidney Right Kidney Marrow Bone Tumor Case 1 15.9 8.35 15.9 0.759 0.759 0.759 42.3 Case 2 17.2 6.9 24.1 17.2 34.4 1.95 3.56 6.65 Case 3 12.27 7.42 23.65 9.05 38.02 0.41 0 .00 9.18 Case 4 6.36 6.12 23.7 24.8 14.2 2.65 2.36 19.8 Case 5 2.98 0.21 11.15 34.21 47.95 0.00 0.00 3.50 Case 6 5.80 8.44 10.27 56.31 12.66 0.00 0.00 6.53 Mean % OA 10.09 6.24 18.13 23.72 24.66 0.96 1.11 14.66 A more accurate representation of the activity distribution can be obtained by looking at the percent of the injected dose that remains in each organ at three hours. This data is obtained by multiplying the percent organ activity distribution times the perce nt injected activity that was not excreted (%OD x %ID remaining) These values are presented in T able 2 6 Table 2 6 Percent injected activity organ distribution of 99m Tc PE I MP at three hours I n all cases t umor uptake and reten tion was observed with little to no uptake in healthy bone and bone marrow as indicated in F igures 2 21 to 2 31 This result is significant and necessary for the development of an effective radiotherapy because non ta rget ed bone and bone marrow are considered to be the dose lim iting organs The Heart Lungs Liver Left Kidney Right Kidney Marrow Bone Tum or Case 1 10 1 0 5 3 0 10 1 0 0 5 0 0 5 0 0 5 0 26 8 0 Case 2 11 5 0 4 6 0 16 2 0 11 5 0 23 1 0 1 3 0 2 4 0 4 5 0 Case 3 7 2 0 4 4 0 13 9 0 5 3 0 22 4 0 0 2 0 0.00 5 4 0 Case 4 3 6 0 3 4 0 13 4 0 14 .0 0 8 0 0 1 5 0 1 3 0 11 2 0 Mean % IA 8 1 0 4 4 0 13 4 0 7 8 0 17 8 0 0 9 0 1 .0 0 12 .0 0

PAGE 75

75 ability to deliver an increased radiation dose to the tumor site while minimizing damage to the se dose limiting orga ns offers great potential for an improved therapy for treatin g pediatric osteosarcoma. Th e ability of PEI MP to target tumor types other than osteosarcoma and t he tumor ta rgeting mechanism of PEI MP was further investigate d by administering 99m Tc PEI MP to three separate client owned canines diagnosed with non osse ous tumors Table 2 7 lists the tumor types and locations for each case. The scintigraphic images and time activity curves for three dogs with various tumor types administered 99m Tc PEI MP are shown below (Figures 2 32 through 2 35 ) Table 2 7 Tumor typ es and locations for non osteosarcoma studies Case Tumor Type Tumor Location Administered Activity (MBq) Case 7 Hemangiosarcoma Left pelvic limb 206 Case 8 Histiocytic sarcoma Perineum 93 Case 9 Adenocarcinoma Mammary 284 Figure 2 32 Scintigraphic image and TAC curve for 99m Tc PEI MP administered to a dog with soft tissue carc inoma in the left pelvic limb (C ase 7) Tumor Liver and Blood pool Kidneys

PAGE 76

76 Figure 2 33 Scintigraphic image and TAC curve for organs of interest for 99m Tc PEI MP administered to a dog with histio cytic sarcoma of the perineum (C ase 8) Kidneys Liver

PAGE 77

77 Figure 2 34 Scintigraphic image and TAC curve for tumor region for 99m Tc PEI MP administ ered to a dog with histio cytic sarcoma of the perineum (C ase 8) Tumor

PAGE 78

78 Figure 2 35 Scintigraphic image and TAC curve for 99m Tc PEI MP administered to a dog with adenocarcinoma of the mammary gland (C ase 9) Table 2 8 The perce nt injected activity excreted through the kidneys for 99m Tc PEI MP administered to dogs with non osteosarcoma tumors As was obser ved for osteosarcoma C ases 1 6, a n exponential decay of radioactivity from the non target ed organs is also observed for the hemangiosarcoma histiocytic carcinoma, and adenocarcinoma ( Figures 2 32 to 2 35 ). Again the average percent of injected activity excreted via the kidney filtration for the non osteosarcoma cases is 36%, similar to that observed for t he osteosarcoma cases (Table 2 8 ). High kidney retention was observed in C ases 7 9 as well (Table 2 9 ). These results again indicate that the particle sizes are too large to clear from th e blood pool resulting in the need for further optimization of particle size prior to future studies. The percent of radionuclide activity present in each organ at three hours w as obtained by selecting regions of interest (ROI) on the scintigraphic images taken at Case Tumor Type Tumor Location %IA Excreted Case 7 Hemangiosarcoma Left pelvic limb 30.17 Case 8 Histiocytic sarcoma Perineum N/A Case 9 Adenocarcinoma Mammary 42.24 Kidneys Blood pool

PAGE 79

79 three hours. The relative distribution of activity for all organs of interest is given in T able 2 9 and the percent of the injected activity remaining in each organ at t hree hours is given in T able 2 10 Table 2 9 Percent injected activity organ distribution of 99m Tc PE I MP at three hours Heart Lungs Liver Left Kidney Right Kidney Marrow Bone Tumor Case 7 8.55 2.75 11.6 67.4 8.92 0.664 N/A Case 8 0.00 0.00 0.00 16 .78 80.00 0.00 0.00 3.22 Case 9 9.44 9 12.1 17.3 50.6 1.31 0.166 N/A Mean % O A 6.00 3.92 7.90 33.83 65.30 3.41 0.28 3.22 Table 2 10 Percent organ activity distribution at 3 hours for 99mTc PEI MP administered to non osseous tumor bearing canines Hea rt Lungs Liver Left Kidney Right Kidney Marrow Bone Tumor Case 7 6 .0 0 1 9 0 8 1 0 47 1 0 6 2 0 0 5 0 N/A Case 9 5 5 0 5 2 0 7 0 0 10 .0 0 29 2 0 0 8 0 0 1 0 N/A Mean % IA 5 7 0 3 6 0 7 5 0 28 5 0 29 2 0 3 5 0 0 3 0 N/A Virtually no uptake in healthy bone and bone marro w was observed for C ase s 7 9 ; however, in contrast to osteosarco ma C ases 1 6 tumor uptake an d retention was not observed for two of the non osseous tumor types with lower uptake and retention being observed for the third This indicates that an additiona l mechanism of action other than the enhanced permeability and retention effect is involved when targeting osteosarcoma with PEI MP. 75 In the passive mode, the natural distribution of the polymer drug conjugate in vivo is exploited to deliver the drug to diseased tissue. In the case of solid tumors, passive deliver y is made possible through the enhanced permeability and r etention effect. In contrast, an active mode of delivery is achieved by incorporation of a

PAGE 80

80 recognition moiety into the polymer carrier. This approach relies more on ligand receptor interactions fo r selective localization of the drug. Thus far, our work with PEI MP has relied heavily on the passive targeting mechanism of the EPR effect. Based on the lack of uptake in non osseous tumors, it appears that PEI MP is not sole ly a passive targeting mate rial, and t hat high affinity of the phosphonate functionality for positively charged hydroxyapatite mineral found in bone plays a key role in the active targeting of osteosarcoma with PEI MP Summary The polymer radiopharmaceutical, PEI MP, has been prepa red based on previous reports by Milner et al. PEI MP is designed to utilize the enhanced permeability and retention e ffect to target oste osarcoma with high selectivity while minimizing damage to healthy tissue. The successful synthesis of PEI MP has been confirmed through 1 H NMR, 13 C NMR, 31 P NMR, FT IR, and elemental analysis. Physical characterization including particle size determination and zet a potential measurements revealed st able particles in t he range of 2.5 to 4 nm. Diagnostic radiopharmaceutica l kits were prepared from PEI MP and b iodistribution studies were conducted in both osteosarcoma bearing canin es as well as canines with non osseous tumors. Time activity curves for all tumor types indicated rapid clearance from the central blood pool and low uptake in the dose limiting organs of healthy bone and bone marrow. This low uptake in the dose limiting organs is attributed to the enhanced permeability and retention effect of macromolecular drug delivery. The percent of injected dose cleared throu gh the kidneys was lower than expected for all cases indicating that further optimization is needed to reduce the maximum particle size prior to future pre clinical studies. Studies in canines with non osseous tumors show little to no uptake in the tumor volume, indicating that the

PAGE 81

81 phosphonate functionality of PEI MP plays a critical role in the active targeting of osteosarcoma. PEI MP offers great promise as an improved systemic therapy for treating pediatric oste osarcoma due to it s reduced uptake in heal thy tissue. Issues related to the binding of therapeutic radiolanthanides w ill be addressed in C hapter 3.

PAGE 82

82 CHAPTER 3 DESIGN OF A POLYMER RADIOPHARM ACEUTICAL FOR IMPROV ED BINDING WITH RADIOLANTHANIDE S Polyethyleneimine Ethylenediamine Tetramethylene Phospho nate In Chapter 2 the polymer radiopharmaceutical polyethyleneimine methylenephosphonate (PEI MP) was introduced. PEI MP has shown the ability to target osteosarcoma more effectively than current FDA approved radiotherapies, with lower uptake observed in healthy tissue when bound to the diagnostic radionuclide technetium 99m ( 99m Tc ) However, there is some evidence that PEI MP may not effectively deliver therapeutic radionuclide s, such as radio lanthanides, as dissociation of PEI MP and lanthanides has be en observed computationally and in vivo 58 To address the challenge associated with the delivery of radio lanthanides, such as samarium 153 ( 153 Sm), factors effecting binding stability will n eed to be considered in the re design of a polymer pharmaceutical for use in internal radionuclide therapy. The two major considerations in chelation chemistry are complementarity and constraint factors. The complementarity factors are the sum of the siz e, geometry, and electron matching between the metal ion and the chelating ligand. These complementarity factors are easier to understand than constraint factors and are either predicted from known theory or determined ex perimentally. The complementar y fa ctors for many metal ligand systems are well established, allowing for easy selection based on these criteria. However, maximizing the complementarity of a metal ligand system is only the first step in optimizing the binding stability of a complex. 76 The other factors to be considered relate to constraint factors, which are the topology (interconne ctedness) and rigidity (flexibility) of the donor atoms in a ligand framework. These constraint

PAGE 83

83 factors play a key role in the development of polymer ligands as radiopharmaceuticals. Large gains in complex stability can be achieved by controlling these con straint factors. One of t he most common and facile way s to increase the thermodynamic stability and kinetic inertness of a metal complex is to modify the constraint factors through incorporation of a polydentate chelator. 77 A polydentate chelator is one in which multiple atoms coordinate to the central atom or metal in a metal ligand complex. This denticity requirement depends largely on the size and coordinat ion geometry of the metal. The lack of multidentate chelation is a likely reason for the hindered binding of samarium with the polymer radiopharmaceutica l ligand, PEI MP, presented in C hapter 2 In an attempt to impro ve the binding stability of polymer radiopharmaceuticals with therapeutic radionuclides a new polymer ligand has been developed in which a multidentate functionality has been incorporated directly into the polymer structure The synthesis, structural characterization, and clinical evaluation of this new polymer rad iopharmaceutical are discussed in C hapter 3 A more in depth analysis of the binding stability of this new polym er ligand will be addressed in C hapter 4 Experimental Materials Polyethyleneimine (Lupasol Water free) was supplied by BASF. Serum vials, s epta and aluminum caps used in preparation of drug kits were received from Voigt Global, Inc. Gel clot endotoxin assay kits (ToxinSensor TM ) were purchased from GenScript. Ultrafiltration membranes (Omega TM ) were supplied by Pall Corporation. D,L 2,3 diam inopropionic acid monohydrochloride was purchased from Acros Organics.

PAGE 84

84 All other materials were purchased fr om Aldrich and used as received unless noted otherwise. Instrumentation and Analysis 1 H NMR, 13 C NMR and 31 P NMR spectra were recorded on a Varian Associates Mercury 300 MHz spectrometer. Chemical shifts from 1 H and 13 C NMR were referenced to residual signals from CDCl 3 ( 1 H = 7.24 and 13 C NMR = 77.23) and D 2 O ( 1 H = 4.80). For 31 P NMR an internal standard of H 3 PO 4 provided a reference for chemical shifts. Infrared spectra were obtained on a Perkin Elmer Spectrum One FT IR equipped with a LiTa 3 detector and ATR accessory. High resolution mass spectra (HRMS) were obtained on a Finnegan 4500 gas chromatograph /mass spectrometer using the chemical ion ization mode. Elemental analyses were conducted by Atlantic Microlab, Inc. (Norcross, GA). Synthesis of 2,3 bis((tert butoxycarbonyl)amino)prop ionic acid (3 4B ) 2,3 diaminopripionic acid monohydrochloride (7.14 mmol, 1 eq) was dissolved in approximat ely 20 mL of saturated sodium bi carbonate. The reaction was cooled in an ice bath and 10 mL of dioxane was added to the reaction mixture. 1 M Di tert butyl dicarbonate in THF (14.3 mmol, 2.0 eq) was added slowly by syringe The reaction was stirred for a n additional hour at 0 C and then warmed up to room temperature and stirred for an additional 12 hours. The reaction was acidified to pH 2 by addition of 1M HCl. The crude product was then extracted with ethyl acetate (x 2) and the organic phases combin ed. The organic phase was washed with brine and dried over MgSO 4 A viscous oil was obtained upon concentration of the crude product on a rotoevaporator. The oil was washed in cold hexane to yield a fluffy white powder in 65% yield. 1 H NMR (CDCl 3 ): (p pm) 1.41 (s, 18H, t Bu), 3.50 (br s, 2H, methylene), 4.83 (m, 1H, methine),

PAGE 85

85 5.35 6.26 (m, NH), 9.04 (s, OH). 13 C NMR (CDCl 3 ): 28.44 (CH 3 ), 42.35 (CH 2 ), 54.44 (CH), 156.30 (HN C=O), 173.77 (C=O). IR (cm 1 ): 3450 3300 (st. NH, st. OH), 2977 2850 (st. CH) 1735 (st. C=O acid), 1694 (st. C=O), 1514 (st. NH amide). Coupling of P olyethylenimine and 2,3 bis((tert butoxy carbonyl)amino)propionic acid (3 6C ). 2,3 bis((tert butoxycarbonyl)amino)propionic acid (2.56 mmol, 1.02 eq) was dissolved in dimethylformam ide (DMF). 2 chloro 4,6 dimethoxy 1,3,5 triazine (2.51 mmol, 1.00 eq) was then added to the stirred reaction mixture. The reaction was cooled to 0 C in an ice bath. N methyl morpholine (2.56 mmol, 1.02 eq) was added slowly by syringe to the cooled react ion. The reaction was stirred for an additional 4 hours keeping the flask in an ice bath. Polyethyleneimine (2.51 mmol, 1.00 eq) dissolved in DMF was then added by syringe along with additional N methyl morpholine (2.51 mmol, 1.00 eq). The reaction was stirred for an additional two hours in an ice bath and then warmed up slowly to room temperature. The reaction was stirred at room temperature for 48 hours. The crude product was precipitated from the reaction mixture as a light brown solid by the additio n of 0.5 M HCl. The solid product was washed 3 times with fresh distilled water. 1 H NMR (DMSO, 80 C): (ppm) 1.4 (s, 18H, t Bu), 2 4 (CH 2 polyimine), 6.2 (1H, NH, amide). IR (cm 1 ): 3500 3300 (st. NH, st. OH), 2977 2850 (st. CH), 1697 (C=O, amide), 1514 (NH, amide), 1249 (med. NH, amide). Deprotection of P olyethylenimine 2,3 bis((tert butoxy carbonyl)amino)propionic acid (3 7B ) Polyethylenimine 2,3 bis((tert butoxycarbonyl)amino)propionic acid 7 was dissolved in a 1:1 mixture of dichloromethane. The solution was stirred overnight at room temperature. The reaction was brought to dryness on a rotoevaporator.

PAGE 86

86 Essentially quantitative deprotection was indicated by the complete disappearance of the Boc group in the 1 H NMR spectrum at 1.4 ppm. 1 H NMR (D 2 O): broad signal 2 4 ppm (CH 2 (ethyleneimine, ethylenediamine). The following step was carried out without further purification. Synthesis of P olyethyleneimine ethylenediami netetramethylene P hosphonate (3 8B ) Hydrochloric acid (54 mmol, 12 eq) and phos phorous acid (18 mmol, 4 eq) were heated to 80 C until dissolution of the phosphorous acid was achieved. The formalin (4.6 mmol, 1 eq, 37% in H2O) was dripped in slowly over approximately 30 minutes upon stirring at 80 C The reaction mixture was then b rought to reflux at 90 C. Compound 3 7B (4.6mmol, 1eq) was dissolved in distilled water and slowly added to the reaction vessel. The reaction was refluxed for an additional 2 hours and then slowly cooled to room temperature overnight. The crude product was precipitated by addition of distilled water and purified by repeated washes with fresh DI water. 1 H NMR (D 2 O): broad multiplet from 2.6 3.8 ppm. 31 P NMR (D 2 O): single broad peak centered at 7 ppm. IR (cm 1 ): 3300 3500 (st. OH, st. NH), 3026 2912 ( CH), 1655 (C=O, amide), 1594 (NH, amide), 1206 and 1175 (P=O), 1070 and 1015 (PO 3 ), 981 (POH). Particle Size Fractionation by Ultrafiltration Fractionation of PEI EDTMP was achieved using polyethersulfone (Macrosep Pall Life Sciences) Centrifugal Filter s. Approximately 0.15 g of polymer was dissolved in 20 mL of endotoxin free water (certified < 0.1 EU/mL). Polymer solutions were passed through 30 kDa and 10 kDa cut off filters sequentially. Filtration was determined to be complete once polymer no lon ger eluted through the filter.

PAGE 87

87 Particle Size Determination by Dynamic Light Scattering The hydrodynamic radii of PEI EDTMP ( < 1 0 kDa) PEI EDT MP(10 30 kDa) and PEI EDT MP( > 30 kDa) fractions were measured by dynamic light scattering (DLS ). Mean particle size and size distribution of the particles was measured on a Nanotrac ULTRA with an external probe (Microtrac, Largo, FL, USA). Each sample was analyzed in triplicate and each replicate was measured six times to yield the average particle size. This was performed using stock solutions which containe d ligand concentrations of 10 m M for all combinations of PEI EDTMP The ionic strength of the 10 mM polymer solutions were adjusted by the addition of 0.15 M NaCl. Zeta Potential Measurements Polymer ligand c harge stability in 0.15 M NaCl solution was analyzed using a ZetaPlus TM zeta potential analyzer (Brookhaven Instruments Corporation, Holtsville, NY). The average of ten runs was recorded for each sample. The data was processed using the ZetaPlus sizing sof tware V2.27. 99m Tc PEI EDTMP Pharmaceutical Kit Preparation Lyophilized ki ts of the polymer ligand, PEI EDTMP, were prepared by mixing a solution containing 1 mg of SnCl 2 2 H 2 O crystals in a drop of concentrated HCl with an aqueous PEI EDTMP solution (15 mg Na PEI EDTMP pH = 8 0) where after the pH was adjusted to 6 by addition of an endotoxin free solution of sodium carbonate. The mixture was dispensed into a vial and lyophilized. The vials were fitted with septa and sealed with an aluminum cap. The v ia ls were then purged with dry N 2 gas and stored in the freezer until being labeled with a radionuclide. R adionuclid e labeling was performed by Cardinal Health Inc. in Gainesville, FL. T he final solutions for injection were prepared by addition of 99m Tc (10 15 mCi ) eluted

PAGE 88

88 from a Mo generator in saline to the above vials. Radiochemical purity was checked using thin layer chromatography on silica gel impregnated glass fiber sheets as a stationary phase and acetone and 0.9% sodium chloride solution for a mobile phase. Biodistribution Studies Biodistribution studies were conducted in the college of veterin arian medicine at the University of Florida. In this study, four healthy research beagles were used to establish biodistribution of the new polymer drug, PEI E DTMP. Additionally, one client owned dog with an osteosarcoma was enrolled i n the trial. All dogs were given a diagnostic dose of 99m Tc PEI EDTMP of between 370 and 555 MBq p er animal. Scintigraphy was performed in the Radiology Section of the Veterinary Medical Cen ter (VMC). The gamma camera was centered under the thorax and abdomen to include the heart, lungs, liver, both kidneys, a nd tumor (when applicable). Both forelimbs were included to acquire data from the metaphysis of distal radius and cortical b one area. Data acquisition was p erformed as a dynamic study of 3 0 x 1 minute frames followed by single one and a half minute frames at 30, 60, 90, 120, and 180 minutes on a countdown bolus injection of the ra diopharmaceutical Urine was collected for the duration of the scan via a urinary catheter. The activity and total urine volume of each case was registered. Results and Discussion Design and Synthesis In an attempt to improve on the binding stability of polymer radiopharmaceutical ligands with therape utic radio lanthanides, a new polymer pharmaceutical was designed in which a multidentate ligand is attached pendant to the polymer backb one. The multidentate ligand chosen was ethylenediamine t etramethylenephosphonate (EDTMP,

PAGE 89

89 Quadramet ) (Figure 3 1) whi ch is currently used as a small molecule radiopharmaceutical for treatment of metastatic bone lesions. The ligand, EDTMP, was i ntroduced in greater detail in C hapter s 1 and 2 EDTMP was chosen for incorporation into the polymer structure for its known abil ity to complex samarium 153 ( 153 Sm) with high stability in vivo Figure 3 1. Structure of ethylenediamine tetramethylenephosphonate (EDTMP) It is hypothesized that incorporation of the multidentate ligand EDTMP into a polymer structure will allow for im proved binding stabilities with a wide variety of radionuclides while maintaining the improved tumor targeting observed with the po lymer radiopharmaceutical PEI MP (Chapter 2) as a result of the EPR effect (Figure 3 2 ). Figure 3 2 Design of the polymer radiopharmaceutical PEI EDTMP for improved binding with therapeutic radionuclides The structure of the new polymer ligand, polyethyleneimine ethylenediamine tetramethylene phosphonate (PEI EDTM P), is shown in F igure 3 3

PAGE 90

90 Figure 3 3 Structure of polyeth yleneimine ethylenediamine tetramethylene phosphonate (PEI EDTMP) The synthetic strategy employed in the preparation of the new polymer radiopharmaceutical PEI EDTMP incorporates the multidentate(polyamino phosphonate) ligand 3 1 and is outlined in F igures 3 4 through 3 9 Polymer 3 3 functions as a carrier for the Quadramet ligand, EDTMP, which is known to bind large polyvalent metal ions, such as 153 Sm used in tumor therapy, with high binding stability in vivo Incorporation of tetraphosphonate moiety EDTMP into the polymer structure was achieved by first protecting diaminopropionic acid monohydrochloride with di tert butyl dicarbonate to give compound 3 4 B (Figure 3 3 ). Figure 3 4. Protection of primary amines in D,L 2,3 diamino propionic acid hydrochloride The crude product was collected as an oil, and purification of the protected diamine was achieved with multiple washes in hexane. Structural characterization with 1 H NMR, 13 C NMR, and FT IR indicates successful attachm ent of the boc protecting A B

PAGE 91

91 groups, which is vital to the success of this synthetic strategy (Figure 3 4) The protected diamine 3 4 B was produced as a white fluffy powder in 63% yield. Figure 3 5 1 H NMR spectrum showing the pre sence of the boc protecting group at 1.4 ppm T he carboxylic acid functionality of 3 4 B allows for facile coupling to primary amines using standard peptide coupling synthesis (Figure 3 6 ) 2,4,6 trichloro 1,3,5 triazine (CDMT) was chosen as the coupling agent for its ability to function in DMF, one of a few solvents capable of dissolving both the polyethyleneimine and protected diamine starting materials. The coupling reaction was complete within 14 hours to yield product 3 6 C which was collected via ex traction with ethyl acetate in a dilute acid solution for removal of weakly basic triazine byproducts. The successful coupling of 3 6 A and 3 6 B to yield 3 6 C was supported by 1 H NMR which displayed a sharp singlet of the boc protecting group at 1.4 ppm an d a broad stretch from 2 4 ppm, indicative of spectral broadening of the polyethyleneimine signal associated with hydrogen bonding. In addition, the presence of the amide linkage is observed by the strong signals in the

PAGE 92

92 FT IR spectrum at 1698 cm 1 (C=O), 1514 cm 1 (O=C N H), and 1249 cm 1 ( C N H) (Figure 3 10 ). Figure 3 6 Synthe tic scheme for the coupling of polyethyleneimine and boc protected diamino propionic acid The coupled product 3 6 C is a st icky, non water soluble polymer; deprotection was carried out prior to any further purification. Essentially quantitative removal of the Boc protecting group on structure 3 6 C was achieved by stirring in a 1:1 mixture of trifluoroacetic acid and dichlorom ethane to yield the water soluble polymer 3 7 B which was observed by the disappearance of the singlet of the boc protecting group in proton NMR. Polymer 3 7 B was converted to polymer 3 8 B using the same chemistry described ea rlier in preparation of PEI M P in C hapter 2 resulting in a methylphosphonic acid functionalized polymer, which was purified by repeated washes with distilled water. Figure 3 7 Deprotection of b oc protected diaminopropionic ami nes A B A C B

PAGE 93

93 Figure 3 8 Synthetic scheme for the methyl phosphonation of the primary amines Deprotonation of 3 9A in a minimal amount of 1 M sodium carbonate solution yielded water soluble polymer 3 9 B 31 P NMR analysis of 3 9 B revealed a broad signal centered at 8 ppm, representative of the shifts reported for aminomethylphosphonic acids at pHs 6 8. 63 Incorporation of the phosphonic acid functionality is also observed by signals in the P O stretching region of the FT IR spectrum: 1206 and 1175 cm 1 (P=O), 1015 cm 1 and 981 cm 1 (POH) (Figure 3 10 C ). Figure 3 9 Synthetic scheme for the formation of the sodium salt of PEI EDTMP A B A B

PAGE 94

94 Figure 3 10 Infrared spectra of (a) polyethyleneimine, (b) PEI Boc Dap and (c) PEI EDTMP Particle Size Analysis The importance of particle size for tumor targeting polymer drugs was addressed in greater de tail in Chapter 1 and a discussion of the dynamic light scattering (DLS) approach to measuring particle size was discussed in Chapter 2 The same approach used to measure particle size for the first polymer PEI MP (Chapter 2) via light scattering was appli ed to the new polymer ligand PEI EDTMP. Solutions of PEI EDTMP were fractionated sequentially through 30 kDa and 10 kDa molecular weight cut off polyethersulfone membranes. The 10 mM samples of each fraction were prepared in 0.15 M NaCl and the hydrodynami c radius was measured by dynamic light scattering.

PAGE 95

95 The particle size distributions for PEI EDTMP are shown in F igures 3 11 and 3 12 and the parti cle size data is summarized in T able 3 1. Figure 3 11 Particle size distribution of PEI EDTMP (>30 kDa) Figure 3 12 Particle size distribution of 10 30 kDa PEI EDTMP The amount of PE I EDTMP that passed through the 10 kDa cut off membrane filter was not great enough to obtain dynamic lig ht scattering data. This is likely due to less polymer being used for fi ltration than was reported for PEI M P in Chapter 2 based o n the low supply of PEI EDTMP. Again, as was the case for polymer ligand PEI MP considerable overlap of the fractions was observed As discussed in Chapter 2 this may be attributed to the ability of the branched structure of PEI EDTMP to reptate through

PAGE 96

96 the pores of the ultrafiltration membrane There was however a considerable shift in the particle size distributions for the two fractions. Table 3 1. Average particle sizes and particle size distr ibutions for PEI EDTMP Polymer Fraction (kDa) Particle Size Range (nm) Average Particle Size (nm) PEI EDTMP > 30 6.96 36.1 8.49 PEI EDTMP 10 30 2.93 15.2 4.27 The effect of blood plasma ions on the particle size of PEI EDTMP was determined throug h DLS measurements of the polymer in the presence of calcium and magnesium. No appreciable change in particle size was observed for PEI EDTMP in the presen ce of thes e blood plasma ions (Figure 3 13 ). Figu re 3 13 Particle size distribution of PEI EDTMP i n th e absence of plasma ions (red) and in the presence of calcium (green ) The PEI EDTMP appeared to be more sensitive to aggregation than PEI MP as aggregation was frequently observed during light scattering experiments. The particle stability was furthe r examined through zeta potential measurements. Particle Stability Zeta potential measurements were conducted to determine the relative solution stability of PEI EDTMP compared to the polymer ligand PEI MP. Generally, as the zeta

PAGE 97

97 potential becomes more neg ative, the system becomes more stable or less susceptible to agglomeration. It is generally accepted that systems with zeta potentials more negative than 20 to 30 millivolts have sufficient repulsion to be moderately stable, with systems being considered extremely stable at values exceeding 45 millivolts. 74 Figure 3 14 Zeta potential values for PEI EDTMP The average zeta potential meas ured for P EI EDTMP was 28.63 (Figure 3 14 ). This zeta potential value is considerably lower than that measured for PEI MP ( 52.43) and is on the threshold of instability. One possible reason for the decreased stability of PEI EDTMP is a lower degree of functionalization at the particle surface due the steric hindrance encountered during attachment of the bulky diaminopropionic acid functionality (Figure 3 6 ) A low density of the EDTMP functionality at the particle surface would reduce the overall repulsive negative charges of the particle, lending to the increased propensity for aggregation.

PAGE 98

98 Animal Care and Procedures Studies were conducted with approval from the Institutional Animal Care and Use Committee (IACUC) at the University of Florida. IACUC approved facilities were used for testing and subsequent housing of animals. Biodistribution The ultimate test of a new radiopharmaceutical is its biodistribution. Animal biodistribution studies are most often performed during development of new radiopharmaceuticals prior to test ing in humans. Canine in vivo imaging studies were used to provide valuable biodistribution information, which will aid in determining the potential dose limiting radiotoxicity to organs in future therapeutic studies in canines and human clinical trials. T he biodistribution of PEI EDTMP was determined using the gamma emitting diagnostic radionuclide technetium 99m ( 99m Tc). Imaging of 99m Tc PEI EDTMP in four healthy canine s was first conducted to validate the in vivo targeting ability of the polymer PEI E DTMP was synthesized and radio pharmaceutical kits were prepared a s previously described in C hapter 3 Kits were prepared from PEI EDTMP that had been sequentially filtered through 30 kDa and 10 kDa cut off ultrafiltration membranes. Complexation of PEI EDT MP with 99m Tc was achieved with good labeling efficiency and radiochemical purity as determined by thin layer chromatography (Car dinal Health, Inc, Ga inesveille, FL) All normal dogs were subjected to the same experimental conditions. Sedative was admin istered via an indwelling intraven ous catheter The animals were placed in either right lateral or left lateral position over the gamma camera. 99m Tc PEI EDTMP was injected intravenou sly with a bolus of 370 555 MBq Data acquisition was

PAGE 99

99 p erformed as a dyn amic study of 3 0 x 1 minute fram es followed by single one and a half minute frames at 30, 60, 90, 120, and 180 minutes on a countdown bolus injection of the radiopharmaceutical Urine samples were collected via a urinary catheter. From the images collected regions of interest (ROI) were dr awn over the organs of interest heart, lungs, liver, kidneys, marrow, and bone. A decay correction equation was used to take into account the exponential decrease in activity of the radionuclide with time. The decay cor activity curves were created for each study. The per cent retention, percent organ activity distribution and percent injected activity distribution at three hours was calculat ed from the decay corrected ROI data. S cint igraphic images and time activity curves for four normal dogs administered 99m Tc PEI EDTMP are shown be low in F igures 3 15 through 3 22 The percent retention, percent organ activity distribution, and percent injected activity were calculated from the imag es and time activity curves in Figures 3 15 to 3 22. Figure 3 15. Scintigraphy images showing the biodistribution of 99m Tc PEI EDTMP ( 10 30 kDa ) in a healthy canine (C ase 1) Liver Kidneys Liver Kidneys

PAGE 100

100 Figure 3 16 Decay corrected time activity curves for 99mTc PEI EDTMP (1 0 30 kDa) in a healthy canine (C ase 1) Figure 3 17 Scintigraphy images showing the biodistribution of 99m Tc PEI EDTMP (1 0 30 kDa) in a health y canine (C ase 2) Liver Kidneys Liver and Blood pool Bladder Kidneys

PAGE 101

101 Figure 3 18 Decay corrected time activity curves for 99m Tc PEI EDTMP (1 0 30 kDa) in a healthy canine (C ase 2) Figure 3 19 Scintigraphic images showing the biodistribution of 99m Tc PEI EDTMP (10 30 kDa) in a healthy ca nine (C ase 3) Liver Kidneys Liver Kidneys Blood pool

PAGE 102

102 Figure 3 20 Decay corrected time activity curves for 99m Tc PEI EDTMP (1 0 30 kDa) in a healthy canine (C ase 3) Figure 3 21 Scintigraphic images showing the biodistribution of 99m Tc PEI EDTMP (1 0 30 kDa) in a healthy canine (C ase 4) Kidneys Liver and Blood pool Kidneys Liver

PAGE 103

103 Figure 3 22 Decay corrected time activity curves for 99m Tc PEI EDTMP (10 30 kDa) in a healthy canine (C ase 4 ) T he scintigraphic images and time activity curves of four normal dogs administered 99m Tc PEI EDTMP indicate good clearance of the radiopharmaceutical from the central blood pool including the heart and lungs as indicated by the exponential decrease i n activity for these organs Additionally, from the time activity curves, low uptake is observed in the healthy bone and associated marrow. This is an important and necessary result, as the bone and marrow are the major dose limiting organs associated with radionuclide therapies. Another indicator of clearance from the central blood pool and non target organs is the amount of 99m Tc PEI EDTMP excreted through the urine. The decay corrected activity and the volume of urine excreted were used to calculate the percent injected activity that was eliminated durin g the three hour scan (Table 3 2 ). Any drug not taken up by the tumor should rapidly clear through the kidneys via the urine pathway to

PAGE 104

104 minimize radiation exposure to non target ed organs. The average per cent injected activity of 99m Tc PEI EDTMP excreted through the kidneys was approximately 60%, a value much hi gher than observed for PEI MP (38 %) and closer to the previously reported clearance rate of 70%. Table 3 2 P ercent injected activity of 99m Tc PEI EDTMP excreted from healthy canines through the urine over a three hour time period Case Condition %IA Research Dog 1 Normal PEIMP/EDTMP 65.09 Research Dog 2 Normal PEIMP/EDTMP 52.82 Research Dog 3 Normal PEIMP/EDTMP 46.45 Research Dog 4 Norm al PEIMP/EDTMP 64.91 The percent of radionuclide activity present in each organ at three hours was obtained by selecting regions of interest (ROI) on the scintigraphic images taken at three hours The relative distribution of activity for all organs of i nterest is given in T able 3 3 H igh uptake of the radiopharmaceutical in the liver and kidneys is seen in the mean organ activity percent (O A %) distribution Table 3 3. Percent organ activity distribution at three hours in four healthy dogs administered 99m Tc PEI EDTMP Heart Lungs Liver Left Kidney Right Kidney Marrow Bone Research Dog 1 3.06 1.39 15.77 0.00 78.45 0.66 0.68 Research Dog 2 4.12 2.04 18.06 27.33 48.45 0.00 0.00 Research Dog 3 0.00 7.36 51.52 8.10 29.92 0.36 2.74 Research Dog 4 2.95 1.46 46.76 19.04 29.41 0.38 0.00 Mean % O A 2.53 3.06 33.03 13.62 46.56 0.35 0.86 However, a more accurate representation of the activity distribution can be obtained by looking at the percent of the injected activity that remains in each organ at three hours. This data is obtained by multiplying the percent organ activity distribution

PAGE 105

105 times the percent injected activity that was not excreted. These v alues are presented in T able 3 4 Table 3 4 Percent injected activity organ distribution of 99m Tc PEI EDT MP at three hours in healthy canines Heart Lungs Liver Left Kidney Right Kidney Marrow Bone Research Dog 1 1 1 0 0 5 0 5 6 0 0.00 27 8 0 0 2 0 0 2 0 Research Dog 2 2 .0 0 1 .0 0 8 6 0 13 1 0 23 2 0 0.00 0.00 Research Dog 3 0.00 4 1 0 28 5 0 4 5 0 16 5 0 0 2 0 1 .0 5 R esearch Dog 4 1 .0 0 0 5 0 16 5 0 6 7 0 10 4 0 0 1 0 0.00 Mean % IA 1 .0 0 1 5 0 14 8 0 5 9 0 19 5 0 0 .00 0 4 0 In taking a closer look at the time activity curves for renal clearance of 99m Tc PEI EDTMP it can be seen that high kidney retention is observed in each c ase, with particularly high retention observed in the two cases with a lower pe rcent inje cted activity excretion (Figure 3 23 ). These results are indicative of radiopharmaceuticals with particle sizes too large for clearanc e through the glomerular pores an d are similar to the results observed by Milner et al. for PEI MP filtered sequentially through 50 and 30 kDa ultrafiltration membranes. This is higher than the optimal filtration size of 10 30 kDa reported previously. 21 57 Additionally, increased liver uptake is observed in each case, again indic ative of l arge particle sizes (Figure 3 24 ). The liver uptake results ar e representative of previously observed data for PEI MP filtered sequ entially through ultrafiltration membranes with molecular weight cut offs of 50 100 kDa. It appears then that particle sizes in the range of those for PEI MP filtered through 30 and 100 kDa cut off membranes are present in the PEI EDTMP batch. I mproved ren al clearance and reduction in liver retention can be achieved through further optimization of particle size

PAGE 106

106 In order to establish the tumor targeting capability of 99m Tc PEI EDTMP, a scan was performed on an osteosarcoma bearing dog following the same pr ocedure as was followed for the four healthy dogs previously reported. Figure 3 23 Decay corrected time activity curves for kidney uptake of 99mTc PEI EDTMP (10 30 kDa) in four separate healthy canines Figure 3 24 Mean and SD of liver ROIs for 99 m Tc PEI EDTMP (10 30 kDa) administered to four separate healthy canines

PAGE 107

107 From the time activity curve shown below, it can be seen tumor uptake and retention is observed. The percent organ activity di stribution is given in T able 3 5 Again, the percent injec ted activity distribution (%OD x %ID retained) is a better representation of the dose in the bod y and is presented in T able 3 6. The percent injected activity distribution for 99m Tc PEI EDTMP was improved for the osteosarcoma bearing canine compared to the healthy canines, with a percent injected activity for the kidneys and liver of 10.9 and 2.2%, respectively. Figure 3 25 Time activity curves for 99mTc PEI EDTMP administered to an osteosarcoma bearing canine Table 3 5 Percent organ activity distribut ion at 3 hours for 99m Tc PEI EDTMP administered to an osteosarcoma bearing canine Heart Lungs Liver Left Kidney Right Kidney Marrow Bone Tumor Case 10 8.77 4.06 53.6 25.9 0 0 7.67 Table 3 6 Percent injected activity organ distribution of 99m Tc PEI EDTMP at three hours in an osteosarcoma bearing canine Heart Lungs Liver Left Kidney Right Kidney Marrow Bone Tumor Case 10 3 7 0 1 7 0 2 26 10 9 0 0 3 2 0

PAGE 108

108 Summary A new polymer radiopharmaceutical, PEI EDTMP, has been designed for improved binding of therapeutic radionuclides through incorporation of a multidentate ligand into the polymer structure. The successful synthesis of PEI EDTMP has been confirmed through 1 H NMR, 13 C NMR and 31 P NMR FT IR, and elemental analysis. Physical characterization inc luding particle size determination and zeta potential revealed moderately stable p articles in the range of 3 to 15 nm (10 30 kDa) Biodistribution studies in healthy canines indicated good clearance from the central blood pool and low uptake in the dose li miting organs of healthy bone and bone marrow. Biodistribution studies also revealed elevated retention in the kidneys and liver for studies in healthy canines indicating potential issues related to filtration or aggregation of particles A biodistributio n study in a n osteosarcoma b earing canine demonstrated tumor targeting ability of 99m Tc PEI EDTMP with low uptake i n healthy tissue. The hypothesized improved binding of therapeutic radionuclides with PEI EDTMP will be examined further in C hapter 4

PAGE 109

109 CHAPTER 4 IN VITRO CHARACTERIZATION OF BINDING STABILITIES OF POLYMER RADIOPHARMACEUTICALS In Vitro Investigation of Polymer Metal Systems Coordination chemistry plays a significant role in the developme nt of radiopharmaceuticals. A radiopharmaceutical l igand must form metal chelates with high thermodynamic stability and kinetic inertness. 78 Dissociation of a radio metal ligand complex in vivo will result in radiation exposure to non target organs, with many radiolanthanides demons trating high affinity for accumulation in healthy bone. 79 There are several biological interactions that can occur as a radiopharmaceutical c irculates in degradation, dissociation, and blood plasma i on competition for the ligand (Figure 4 1 ). 80 These in vivo c hemical reactions are a primary source of radiation toxic ity associated with therapeutic ra d i olanthanides. 80 Figure 4 1 In vivo reactions of a polymer radiopharmaceutical in the bloodstream

PAGE 110

110 Binding Propert ies of Lanthanides Lanthanide ions do not exist as M 3+ ions in solution; instead, they are bound with surrounding water molecules. The number of water molecules bound to the metal depends on the oxidation state of the lanthanide. Lanthanide ions are genera lly found in the +3 oxidation state and because of their large size, often have coordination numbers in the range of seven to ten. Complexation of the lanthanide requires displacement of the water molecules by the chelating ligand. The interactions betwee n donor atoms of the ligand and the lanthanide ions are often ionic in nature, as the 4f electrons are inner electrons shielded from interaction with metals. Another common characteristic of lanthanide ions in solution is their tendency to precipitate upo n reaction with common anionic chelators such as carboxylates and phosphonates. 77 Precipitation in the body would alter the pharmacokinetics through undesired liver uptake and metabolis m of the lanthanide bearing radiopharmaceutical. All of these characteristics apply to y ttrium as well, as it has similar charge, ionic do ttrium ions, Ln(III) and Y(III), have been shown to have a high affinity for carbonate and phosphonate ligands. Additionally, these ions have shown high affinity for bone mineral. An injection of 90 Y in a human results in about 50% of the dose accumulating in the bone 25% in the liver, 15% excreted through the kidneys, and 10% to all other organs. 79 Thermodynamic Stability and Kinetic Inertness Dissociati on of the metal ligand chelate becomes likely once it is introduced into the bloodstream due to dilution of the ligand concentration. 47 The metal ligand complex must be thermodynamically stable In order to maintain stable complexation in the

PAGE 111

111 presence of natural competitive chelators such as transferrin or citrate. A thermodynamically stable complex requires that the metal ligand complex be lo wer in energy than the two separate entities apart from one another (the direction of the equilibrium must be favorable). The complex must also be kinetically inert (the rate of the equilibrium must be favorable). For a complex to be kinetically inert, the rate of association disassociation must be on an appropriate time scale to prevent release of the metal ion under the extremely dilute condi tions found in the blood stream. Even if the complex is thermodynamically stable but not kinetically inert, in vivo dissociation or transchelation will occur over the long time periods of time that the drug circulates in the bloodstream. 78 It is this kinetic inertness that dominates complex stability in solution. Figure 4 2 depicts the change in potential energy vs. the reaction progression, where a determines th e kinetic stability of the reaction. Figure 4 2 Potential energy diagram for a binding reaction Measuring Complex Stability The thermodynamic stability constant, K B is a strong predictor of the solution behavior of radiopharmaceuticals and is shown in Equation 4 1 77 The stability constant for metal complex formation has long been employed as an effective measure of the

PAGE 112

112 affinity of a ligand for a metal ion in solution. 81 These constants provide a quantitative indication of the success or failure of a ligand design. 82 (4 1) Although biological studies in animal models remain the most appropriate method for evaluating in vivo stabil ity of radiopharmaceuticals, better in vitro techniques are needed to establish binding properties of radio metal ligand chelates prior to clinical testing. Two approaches have been explored and are described in detail in C hapter 4 The first involves the use of Isothermal Titration Calorimetry (ITC) to measure thermodynamic binding parameters for a polymer metal system. The second approach involves the use of dialysis experiments to investigate the kinetic stabilities of polymer metal systems. Experiment al Materials Samarium chloride hexahydrate, calcium chloride, and citrate were purchased from Aldrich and used as received. Polyethyleneimine methylene phosphonate (PEI MP) and polyethyleneimine ethylenediamine tetramethylene phosphonate (PEI EDTMP) were pre pared as previously described (C hapters 2 and 3). Ethylenediamine tetra(methylene phosphonate) pentasodium salt (EDTMP) was purchased from AK Scientific, Inc. and used as received without further purification. Methods The ITC instrument used was a VP IT C microcalorimeter (Microcal, Inc., Northampton, MA), which operated at a constant temperature of 298 K. The M + L ML K B = [M L ] [M ] [L ]

PAGE 113

113 regenerated cellulose dialysis membranes (Spectra/ Por3, MWCO 3500, 18 mm) and closures (Spectra, 23 mm) were purchased from VWR. Elemental analyse s were conducted by atomic emission inductively coupled plasma spectroscopy (model Plasma 3200 ICP, Perkin Elmer, Wellesley, MA), with the mean of five repetitions being collected. Calibration of the ICP was accomplished using a 1000 ppm samarium standard solution (Aldrich) which was diluted with distilled water to produce a standard curve with points at 50, 25, and 10 ppm. Description of ITC Binding Experiment To carry out a binding experiment, the ligand (EDTMP, PEI MP, or PEI EDTMP) was placed in the s ample cell and the reference cell filled with water. Both cells were mounted in an adiabatic chamber, and the temperature was kept constant during the experiment. A long needle syringe (nominal volume fastened to its end was filled with a solution of samarium chloride (13.4 mM). The syringe was placed inside the sample cell and the entire assembly was rotated continuously to provide proper mixing of the contents of the sam ple cell. After reaching thermal equilibrium, the injection of the samarium was automatically conducted stepwise until all the binding sites on the ligand were saturated with samarium. The first L due to the possible dilution during the equilibration time preceding the measurement. The data from this first a period of 20 seconds. The chosen time interval bet ween the two consecutive injections was 300 s in order to ensure that thermodynamic equilibrium was reached prior to the next injection. The data analysis was performed using ORIGIN 7.0 based software. The thermodynamic parameters (K B d as adjustable

PAGE 114

114 parameters in the fitting procedure: one set of sites and two sets of sites, to the experimental data using a nonlinear least squares approach (Levenberg Marquardt algorithm). 83 84 Description of Dialysis Experiment Regenerated cellulose dialysis membranes (Spectra/ Por3, MWCO 3500, 18 mm) were soaked in deionized water for up to two hours. An 8 mm segment of th e tubing was filled with 2 mL of the solution to be dialyzed. The dialysis tubing was sealed with closures (Spectra, 23 mm) and placed in a container containing 98 mL of the exterior solution. The experiment was stirred with a magnetic stirrer and 5 mL ali quots were collected from the exterior solution at 30 minutes, 1 hour, 2 hours, and 3 hours. Each aliquot that was removed was replace d with the same volume of fresh exterior solution. The aliquots were then analyzed by atomic emission inductively coupled plasma spectroscopy (model Plasma 3200 ICP, Perkin Elmer, Wellesley, MA), with the mean of five repetitions being collected. Calibration of the ICP was accomplished using a 1000 ppm samarium standard solution (Aldrich) which was diluted with distilled wate r to produce a standard curve with points at 50, 25, and 10 ppm. Deionized water was used as a blank in the calibration step. Results and Discussions Thermodynamic Binding Parameters The polymer ligand polyethyleneimine methylene phosphanote (PEI MP) has demonstrated improved tumor targeting and tumor uptake ratios when delivering the diagnostic radionuclide technetium 99m ( 99m Tc) compared to the small molecule ligand ethylenediamine tetramethylene phosphonate (EDTMP, Quadramet ) (C hapter 2). However, atte mpts to deliver the therapeutic radionuclide samarium 153 resulted in

PAGE 115

115 apparent dissociation of the radionuclide PEI MP complex in vivo as indicated by high uptake in the liver. 58 In an effor t to improve the binding stability of radionuclide polymer ligand complexes, a new polymer ligand with a multidentate functionalitity, polyethelyneimine ethylediamine tetramethylene phosphonate (PEI EDTMP), was designed and prepare d as previously described (C hapter 3). Isothermal titration calorimetry experiments were conducted to determine the relative thermal binding stabilities of the bisaminophosphonate ligands ethylenediamine tetramethylene phosphonate (EDTMP), polyethyleneimine methylene phosphanote ( PEI MP), and polyethelyneimine et h ylediamine tetramethylene phosphonate (PEI EDTMP) with a stable isotope of samarium (F igu re 4 3 ). The thermodynamic binding constant, K B, was used as a measure of the stability of the metal ligand complex in solution. Figure 4 3 Three bisphosphonate ligands: A) ethylenediamine tetram ethylene phosphonate (EDTMP) B) polyethyleneimine me thylene phosphanote (PEI MP) C) polyethelyneimine etylediamine tetramethy lene phosph onate (PEI EDTMP) Isothermal Titration Calorimetry (ITC) is an analytical tool used to determine thermodynamic parameters of a binding interaction in solution. This technique is most often used in the field of biochemistry to measure binding interactions between small molecules and proteins. Mullen et al have recently demonstrated the ability to extend A B C

PAGE 116

116 this approach to study interactions in polymer inorganic hybrid systems. 85 Based on this work, ITC has been explored here as a means for investigating binding stability of polymer radiopharmaceutical systems. Isotherm al Titration Calorimetry instruments can measure heat that is generated or absorbed during a bimolecular binding event with extreme accuracy. Unlike other methods for measuring binding parameters, ITC allows for determination of all binding parameters st from a single experiment. A typical ITC experiment consists of placing a solution of the ligand into the sample cell and the solvent int o the reference cell (Figure 4 4 ). The se cond molecule to participate in the binding event is then slowly dripped into the sample cell via a syringe, with a thermodynamic profile being generated based on the heat change of the solution during the interaction. The temperature is allowed to return to equilibrium before the next injection. The heat that is produced is directly proportional to the amount of binding interactions that are occurring. As the system becomes saturated, the heat produced or absorbed will decrease to the point of dilution. A typical data output c an be seen in F igure 4 5 where the area under the peaks in the raw thermogram is integrated to provide a binding curve. The thermodynamic B ) can be taken directly from the binding curve. Isothermal titration calorimetry allows for determination of a ll of these thermodynamic parameters from a single experiment. All other thermodynamic parameters can be derived from free energy relationships.

PAGE 117

117 Figure 4 4 Schematic of isothermal titration calorimetry instrument Figure 4 5 Isothermal t itration c alorimetry raw data (left) and integrated data (right) output 93 To demonstrate the characteristic output of an ITC experiment, an ITC th er mog ram is depicted in F igure 4 5 The left portion of F igure 4 5 represents the raw data as supplied by the instrument and the progression of the experiment in time. The peaks correspond to the individual aliquots of added samarium solution, and the current a pplied for the compensation of the reaction heat was plotted in energy units against tim e. The right portion of F igure 4 5 shows the titration curve resulting from the i of samarium/ligand.

PAGE 118

118 The concentration of the macromolecule placed in the cell is a function of the binding constant for the reaction and has to be determined experimen tally. The product of the binding constant, K B and the polymer concentration, [M T ], determine the shape of the binding curve. This is described by the dimensionless parameter C, where C = K B [M T ]. It has been shown that a C value in the range of 10 500 sh ould be used. 86 When C is less than 10, a nearly horizontal plot is produced from which little information a bout the binding constant can be obtained, and when C is greater than 500, the shape of the curve becomes unresponsive to changes. Curves for various differen t C values are shown in Figure 4 6 87 For high binding constants (10 7 to 10 8 M 1 ), low concentrations of the polymer ligand must be used. Through a series of experiments, the concentrations of the ligands PEI MP, PEI EDTMP, and EDTMP that provide an appropriately shaped curve were deter mined and are given in T able 4 1 Figu re 4 6 The shape of ITC curves at different concentrations of reactants 87

PAGE 119

119 Table 4 1 Concentrations of reagents that provided useful ITC curves Ligand Concentration PEI MP 1.3 PEI EDTMP 0.1 EDTMP 1.04 Samarium 13.4 In order to determine the influenc e of the enthalpy of dissolution of samarium, a dilution experiment was first conducted. In this experiment, the ligand was injected into the sample cell containing only the solvent. The released heat caused by the solvent enthalpy was determined to be neg ligible, contributing little to the heat produced/released during the binding event. The polymer ligand PEI MP was then titrated wi th a samarium solution. Figure 4 7 shows the ITC curve for the titration of a 1.3 mM solution of PEI MP with a 13.4 mM soluti on of samarium chloride at 298 K in water. The binding observed for PEI MP with samarium was the simplest case, in which a single set of sites model could be used to fit the data. This assumes that there is only one type of binding site with a consistent a ffinity for the metal. The observed peaks indicate an endothermic process upon each injection. Evaluation of this data by integration followed a nonlinear least squares approach which leads to an entropically driven interaction with both positive entropy a nd positive enthalpy (Figure 4 7 ). A thermodynamic profile bearing both po is negative, indicating favorable formation of the metal ligand complex. A large

PAGE 120

120 contribution to the entropy term arises from the displacement of the so dium counter ions as well as the de so lvation of the samarium ions. Figure 4 7 ITC data output for PEI MP titrated with samarium The same experiment was performed with the small molecule ligand EDTMP and samarium. The binding observed for EDTMP and samarium was more complex than that observed for PEI MP and samarium, with a two binding site model pr oviding the best fit (Figure 4 8 ). This observed binding behavior arises from multiple sets of independen t binding sites or multiple sets of interacting binding sites. Similar to PEI MP the binding interaction appears to be entropically driven, with both a positive enthalpy Ligand: PE I MP Model: One Site N: 0.448 0.00999 K: 3.50 x 10 4 7.01 x 10 3 9515 288.0 52.7

PAGE 121

1 21 and positive entropy. The free energy of the binding interaction is significantly more negative than that for PEI MP, indicating a n even mo re favorable process (Figure 4 8 ). Figure 4 8 ITC data output for EDTMP titrated with samarium Titration of the polymer l igand, PEI EDTMP, provided a similar binding behavior as EDTMP, with a two set of binding sites model providing the bes t fit (Figure 4 9 ). The shape of the titration curve for PEI EDTMP strongly resembles that of the small molecule ligand afte r which it wa s modeled, EDTMP. Ligand: EDTMP Model: Two Sites N 1 : 0.550 0.00879 K 1 : 1.24 x 10 8 6.60 x 10 7 1 : 8222 99.7 1 : 64.6 N 2 : 1.74 0.0110 K 2 : 5.25 x 10 5 6.20 x 10 4 2 : 4 920 51.0 2 : 42.7

PAGE 122

122 Figure 4 9 ITC data output for PEI EDTMP titrated with samarium This two step binding process is thought to result from an initial reaction of samarium to the first donor atom of th e ligand. The energies for these binding events may vary for PEI EDTMP and EDTMP due to steric factors associated with the polymer ligand. The binding of the second donor atom of the bidentate ligand should proceed much more rapidly, as it is now tethered to the chelating ligand a nd in close proximity to B for the second binding event for both PEI EDTMP and EDTMP. This Ligand: PEI EDTMP Model: Two Sites N 1 : 0.373 0.185 K 1 : 3.47 x 10 6 7.29 x 10 5 1 : 1.678 x 10 4 6.65 x 10 3 1 : 86.2 N 2 : 8.82 0.200 K 2 : 1.55 x 10 5 1.92 x 10 4 2 : 5139 51.1 2 : 41.0

PAGE 123

123 binding event was again an entropically driven process, with a positive entropy an d positive enthalpy (Figure 4 10 ). Figure 4 10 Possible mechanism for two step binding process with bidentate chalators Binding c onstants A comparison of the binding constants for the three ligands, PEI MP, PEI EDTMP, an d EDTMP, is given in Figure 4 11 Within the context of this work, the binding constants obtained from the ITC data are meant to provide a relative comparison of binding stabilities for the three bisphosphonate systems under investigation, rather than a quantitative determination of equilibrium constants. From F igure 4 11 it can be seen that incorporation of the Quadramet functionality, EDTMP, directly into the polymer structure (PEI EDTMP) provides an increase in thermodynamic binding stability of approximately two orders of magnitude. This places the binding constant, K B for PEI EDTMP directly between that of PEI MP and EDTMP. c v c v c v c v c v c v c v c v c v c v c v

PAGE 124

124 This significant increase in binding stability can likely be a into the polymer structure. Figure 4 11 Binding constants as measured by ITC It is a firmly establishe d empirical fact that complexation resulting from coordination with multidentate ligands results in more thermodynamically stable complexes compared to multiple monodentate ligands. 88 An example of this is seen by comparing th e binding constant values for addition of two monodentate ligands co mpared to one bidentate ligand in the reaction shown below, where the replacement of 6 ammonia ions with 3 diaminomethane ligands results in an increase in the equilibrium binding constant for the 1,2 diaminoethane chelate such that this chelation complex is 10 8 times more stable than the monodentate ammonium complex (Figure 4 12) Figure 4 12 Monodentate ligands (left) and bidentate ligands (right) binding with nickel

PAGE 125

125 This difference in stability constants can be related to the entropy term, solvation changes, or ring formation. The stability of this chelate complex may be described in terms of both thermodynamics and kinetics. Thermodynamically, this can be understood by looking at t he relationship of the binding constant, K B and other thermodynamic through Equation 4 2. RT lnK e (4 2) where R is the universal gas constant and equals to 1. 987 cal K 1 mol 1 and T is the quation 4 3. T (4 3) Using the binding constants taken from the inflection points on the ITC thermograms for PEI MP, PEI EDTMP, and EDTMP, the thermodynamic parameters for each ligand binding with samarium were determined from Equations 4 2 and 4 3 (Table 4 2 ). The contributions of these individual terms will be consid ered in the following sections. Binding free e nergy The most relevant thermodynamic parameter for describing the binding stability of a compou 89 The binding free energies for the ligands PEI MP, PEI EDTMP, and EDTM P are shown in Figure 4 13 The more negative values for PEI EDTMP and EDTMP indicate a more favorable binding event for the multidentate ligands compared to the monobisphosphonate ligand PEI MP, presumably due to the previously described chelate effect.

PAGE 126

126 S econdary and tertiary structural interactions present in a polymer ligand system all contribute to the binding free energy. These interactions include hydrogen bonding, van der Waals interactions, hydration, and conformational entropy. 90 92 The changes in entropy and enthalpy can provide a better understanding of the free energy of binding. These contributions will be considered independently in the following secti ons. Table 4 2. Thermodynamic binding parameters as determined by ITC for three phosphonate ligands Ligand K B PEI MP 3.5 0 x 10 4 7.01 x 10 3 9.52 15.8 6.19 PEI EDTMP 3.47 x 10 6 7.29 x 10 5 1.55 x 10 5 1.29 x 10 4 16.8 5.14 25.6 12.2 8.92 7.08 EDTMP 1.24 x 10 8 6.6 x 10 7 5.25 x 10 5 6.20 x 10 4 8.22 4.92 19.4 12.8 11.01 7.89 Figure 4 13 Thermodynamic parameters for three bisphosphonate ligands

PAGE 127

127 Enthalpy contribution to b inding The change in enthalpy of a binding event refl ects the strength of the interaction of the components involved in the binding event relative to hydrogen bonding and van der Waals interactions with the solvent. The more negative the enthalpy, the more enthalpically favorable the binding event. Negative enthalpies require appropriate alignment of the hydrogen bonding donor and acceptor groups at the interface of the components. 93 The binding enthalpies for the ligands PEI MP, PEI EDTMP, and EDTMP were derived from the free energy Equations 4 2 and 4 3 Unfavorably positive values were obtained for each system in spite of the very favorable free energies and binding constants The positive enthalpies reported here are a result of the observed heat of the binding reaction being a global property of the entire system under investigation. 94 When a solution of metal is dripped into a so lution of ligand, the entire heat change within the calorimetric cell is recorded. measured is more than just the noncovalent interactions within the cell, it also represents the formation and dissociation of many bonds within the system. These bond ing interactions can include solvent hydrogen bonding, van der Waals interactions, or solvent reorganization at the surface of a compound. For reactions conducted in water, the main contributor to enthalpy is often the bulk hydration effect. 95 96 In addition to these hydration effects, all other noncovalent interactions at the surface contribute to noncovalent interactions between the bisphosphonate ligands and sodium counterions play a large role in the observed enthalpy of these systems. The enthalpy obtained from Isothermal Titration Calorimetry is therefore an observed or apparent quantity, obs 94

PAGE 128

128 The measured enthalpies may be unfavorable, as in the case represented here, ev en if the binding event is in itself enthalpically favorable. Entropy contribution to b inding The entropic contributions for all systems reported her ein wer e derived from Equation 4 3 A positive T term represents an entropically favored binding event. The T water molecules at the interface and the release of many counterions upon binding of metal with the ligand. Negative entropic contributions may arise from the decrease in conformational degrees of freedom resulting from the inter chain physical crosslinking that can occur in the polymer ligand systems and the coupling within the small molecule system. The large positive entropy changes for PEI MP, PEI EDTMP, and EDTMP with samarium indicate favorable binding events and entropically driven processes. Binding s toichiometry Isothermal Titration Calorimetry has become a useful tool for measuring binding stoichiometry. The predicted stoichiometries for the ligands PEI MP, PEI EDTM P, and EDTMP with samarium are 1:1 based on literature values for the stoichiometry of EDTMP samar ium complexes. 22 97 The values as measured by ITC are 0.45, 0.37, and 0.33, respectively. These stoichiometric values were determined from the molar ratios of the interacting species at the equivalence poi nt during titration. During the fitting procedure, the stoichiometry was treated as a floating parameter and was determined by iterative fitting. The cause for deviation of the measured value from predicted values is likely due to uncertainty of concentrat ions of the polymer systems. This uncertainty arises from the error associated with determining the exact degree of functionalization

PAGE 129

129 during polymer ligand synthesis Other possible sources that lead to deviation from expected stoichiometric values include s degradation of the ligand, unspecific binding, or experimental uncertainty of the data set. 89 Summ ary of Thermodynamic Binding Parameters The thermodynamic binding stability of bisphosphonate ligands appears to increase as the denticity of the ligands increases, as is evidenced by the decrease in free energy for these systems and increase in binding eq uilibrium constants. The thermodynamic profiles have been investigated and various forces that drive the binding event have been identified, including the release of water and counterion molecules. Ligand to Metal Ratios As a result of the tendency f or la nthanide ions to precipitate in vivo t he appropria te ratio of ligand to metal ion was determined. Solutions of the polymer ligand PEI MP and samarium were prepared in varying ratios and the solutions were monitored for precipitation. It was found that pol ymer samarium complexation led to precipitation at li gand to metal ratios less that 5:1 It is essential that an appropriate ratio be used to prevent this precipitation as it can lead to inflammation of the vein wall (phlebitis) or impaired pharmacokinetic s. 98 Kinetic Stability of Ligand Complexes For a typical complex formation involving a ligand, L m and a metal, M the thermodynamic equilibrium constant is equal to the ratio of two kinetic rate cons tants, k f and k r as demonstrated in Figure 4 14, where K is the ratio of the forward and reverse reactions in t he binding equilibrium reaction and provides the overall rate of the binding process.

PAGE 130

130 Figure 4 14 Relationship between a chemical binding event and the equ ilibrium constant It can be seen from Figure 4 14 that the two reaction rates can both be labile or both be inert and still have the same ratio, and hence the value for K. Therefore, even though the three bisphosphonate ligands, PEI MP, PEI EDTMP, and EDTM P appear to form thermodynamically stable complexes with samarium, the kinetics may be unfavorable to the point that samarium is released from the complexes in dilute solutions. As discussed in C hapter 2, it is thought that calcium ions in the blood may b e responsible for displacing samarium ions from the polymer ligand, PEI MP, in vivo In order to investigate this further, a set of experiments was conducted in which the amount of samarium dissociating from the polymer ligands could be measured. The exper iments involved dialyzing a solution of polymer and samarium against either a solution of water or a solution of the blood plasma ion calcium The experimental set up is depicted below in Figure 4 15 Dialysis tubing with a nominal molecular weight cut off of 3,500 kDa was used. This ensures that the polymer is retained within the dialysis tubing but that the free samarium, calcium, citrate, and other blood plasma ions are free to diffuse across the membrane into the interior or external solution. The more s amarium that dissociates

PAGE 131

131 from the polymer ligands due to labile kinetics of binding, the higher the samarium concentration is in the exterior dialysis solution. Figure 4 15 Dialysis experimental set up Prior to testing the binding of polymer and samarium, a baseline for the dialysis of samarium was created by first dialyzing a solution containing samarium against an exterior solution of deionized water. The experiment was de signed so that at equilibrium, the samarium concentration would be 15 ppm (0.1 mmol/L SmCl 3 ). The initial dialyses were conducted in glass jars with a total volume of dia lysis solution of 200 mL. 5 mL A liquots of the exterior solution were collected at 30 minutes and then hourly out to 5 hours. The aliquots were examined by inductively coupled plasma (ICP) spectroscopy to determine the amount of samarium that dialyzed through the membrane. The ICP data indicates that at equilibrium, the concentration of sam arium in the exterior solution ranged a nywhere fr om four to nine ppm (Figure 4 16 ). To determine if the soaking time of the dialysis tubing affected the performance, various soaking times were tested with no apparent improvement being observed for soaking times longer than two hours (Figure 4 17). Possible Dialysate Solutions: Water Calcium in water Serum S amarium Polymer

PAGE 132

132 Figure 4 16 Dialysis of a 15 ppm solution of samarium in DI water Figure 4 17 The effect of membrane soaking time on dialysis Since the equilibrium concentration was lower than expected, a series of experime nts was conducted to de termine if the dialysis tubing i s interacting with the

PAGE 133

133 samarium ions and affecting the equilibrium concentration. In these experiments, dialysis tubes were cut open to form a flat sheet and then placed in a 15 ppm solution of samariu m. The concentration of samarium measured for open tubing was similar to that for samarium dialy zed throug h the closed tubing (Figure 4 18 ) and the membrane was determined not to significantly affect the results Figure 4 18 The effect of membrane on dialysis Next, the ability of PEI MP to effectively bind s amarium ions in water was studied. PEI MP was mixed together with s amarium and then dissolved in deionized water The molar ratio between the PEI MP repeating unit and Sm is 5.5:1. This was the rati o that was determined to prevent precipitation of the polymer from solution. T he dialysis expe riment was carried out with deionized water as an exterior solution. Aliquots of the exterior solution taken at 30 minute intervals out to three hours were analyz ed by ICP to determine the amount of samarium that was released from the polymer. T he detection limit of samarium ion detection by inductively coupled plasma spectroscopy is 0.07 ppm,

PAGE 134

134 rendering any value below as non significant From Figure 4 19 it can b e seen that samarium is bound effectively by PEI MP in the absence of competing ions, as no detectible samarium is observed in the exterior solution in the presence of the polymer ligand PEI MP. Figure 4 19 Dialysis of samarium in the presence of PEI MP It has been theorized that the blood plasma ion calcium is acting to displace samarium from the polymer ligand, PEI MP, when delivered in vivo (Chapter 2 ). To investigate the effect of calcium on the binding of samarium by PEI MP, a dialysis experiment w as conducted in which calcium was added to the exterior solution. A calcium concentration of 40 ppm was used to mimic physiological concentrations. Analysis of aliquots taken at 30 minute time intervals out to three hours indicates that calcium does in fac t promote the release of samarium ions from the polymer ligand P EI

PAGE 135

135 MP, as seen below in Figure 4 20 wit h values of samarium above the detection limit being observed Figure 4 20 The effect of calcium on dialysis of samarium in the presence of PEI MP It has also been proposed that competitive biological chelators such as citrate may be involved in the in vivo dissociatio n of samarium from PEI MP (Chapter 2 ) in that once calcium has displaced samarium from the polymer, it will be bound by the competiti ve chelator To determine what effect other blood plasma components have on the dissociation of samarium PEI MP once the samarium has been displaced by calcium, a single experiment was run in which a solution of PEI MP and samarium was dialyzed in bovine c alf serum (Figure 4 21 ) This serum solution contains both the blood plasma ion calcium and the competitive chelator citrate in addition to other plasma ions and proteins. From this experiment, it can be seen that the amount of samarium released by the ser um is greater than that observed in dialysis against calcium solution

PAGE 136

136 The high liver uptake observed clinically with delivery of PEI MP samarium may be explained by displacement of samarium from the PEI MP samarium complex through calcium ion competition followed by binding of the samarium by citrate, which in turn accumulates in the liver within the first hour of circulation in the bloodstream. Figure 4 21 The binding of samarium by PEI MP in blood seru m Since it has been shown that PEI MP samarium co mplexes do not appear to be stable in the presence of calcium, the newly designed polymer ligand, PEI EDTMP, will need to be tested for improved binding in the presence of blood plasma ions and competitive chelators. The first experiment conducted was to test the ability of PEI EDTMP to bind samarium in the absence of any competitors. A solution of PEI EDTMP with samarium was dialyzed against deionized water. From the ICP data gathered at 30

PAGE 137

137 minute intervals out to three hours, it can be seen that no dete ctable amount of samarium is released from PEI EDTMP indicating that the PEI EDTMP samarium complex is kinetically stable Next, the effect of calcium on the stability of the PEI EDTMP samarium complex was tested by dialysis in a solution containing 40 p pm calcium. It can be seen that even in the presence of calcium, there is no significant release of samarium from PEI EDTMP, indicating improved binding of samarium by PEI EDTMP compared to PEIMP Fi gure 4 22 The dialysis of samar ium in the presence of PEI EDTMP

PAGE 138

138 Figure 4 23 The effect of calcium on the binding of samarium by PEI EDTMP Summary Two separate techniques have been introduced for evaluation of the thermodynamic and kinetic binding stabilities of the two polymer ligands, PEI MP and PEI EDTMP with the lanthanide samarium. Isothermal Titration Calorimetry provided a thermodynamic profile for the polymer metal systems and indicated an improved thermodynamic binding constant for the multidentate functionalized polymer ligand, PEI EDTMP. This inc rease in binding stability is attributed to the chelate effect and is expected to improve the in vivo complexation of samarium for delivery as a therapeutic radiolanthanide. The kinetic stability of the polymer samarium systems in solution were further inv estigated by a series of dialysis experiments, which allowed for measuring the

PAGE 139

139 amount of free samarium in the presence of polymer. It was seen that both polymers bound samarium with high stability when no other ions were present. However, introduction of t he competitive blood plasma ion calcium resulted in release of samarium from the polymer ligand PEI MP. This effect was not observed for the polymer ligand PEI EDTMP. Based on the findings in C hapter 4 introduction of a multidentate functionality into th e polymer structure has increased both the thermodynamic and kinetic binding stabilities with samarium. This improved binding is expected to allow for delivery of therapeutic lanthanides to osteosarcoma tu mors, based on the findings in C hapters 2 and 3

PAGE 140

140 C HAPTER 5 CONCLUSIONS Summary An inherent limitation of small molecule radiopharmaceuticals is their tendency to accumulate in healthy tissue, thereby limiting the dose of radiation that can be administered to the tumor site. This limitation was addre ssed in the work presented in Chapters 1 4 through synthesis, characterization, and clinical evaluation of two polymer based radiopharmaceuticals. Table 5 1 illustrates a comparison of the radiopharmaceuticals discussed in Chapters 1 4. Table 5 1. Comparis on of radiopharmaceuticals COMPARISON OF RADIOPHARMACEUTICALS EDTMP P EI MP PEI EDTMP Advantages FDA approved for treating pain associated with bone tumors Binds therapeutic radiolananides with high stabil ity in vivo Disadvantages High uptake in healthy tissue Limited therapeutic value Advantages Binds diagnostic 99M TC Targets osteosarcoma tumors Low uptake in healthy tissue Disadvantages Questionable binding of radiolanthanides Advantages Bin ds diagnostic 99M TC Targets osteosarcoma tumors Low uptake in healthy tissue Binds samarium with high stability in vivo

PAGE 141

141 Chapter 1 describes the current needs in therapies for targeting pediatric osteosarcoma, highlighting the need for new systemic the rapies to treat metastatic and inaccessible tumors. Targeted radiotherapy was presented as a promising systemic approach that offers the advantage of delivering a higher dose of radiation to the tumor site while sparing healthy tissue relative to tradition al external radiation therapies. Precedence was given for the use of polymers to passively target cancer via the enhanced permeability and retention effect, in which large molecules are less likely to penetrate healthy tissue but are capable of passing thr ough leaky vasculature known to exist around solid tumors. The key features in the design of polymer based radiopharmaceuticals polymer, chelating ligand, and radionuclide were also discussed in Chapter 1 and form the basis of the work presented in Cha pters 2 and 3. Chapter 2 outlines the synthesis, characterization, and biodistribution of polyethyleneiminemethylene phosphonate (PEI MP) originally reported by Milner et al (Figure 2 B) and reproduced in this work for further evaluation Structural char acterization confirmed successful synthesis of PEI MP. Biodistribution studies were conducted in a canine osteosarcoma model and indicate d that the PEI MP reproduced in this work demonstrates good uptake and retention in osteosarcoma tumors with little upt ake in the regions associated with the dose limiting or gans of healthy bone and marrow The clearance rate of excess radiopharmaceutical was lower than previously reported by Milner et al with 35 40% of excess radio activity being excreted through the kidn eys via the urine. This high er than desired retention of excess radiopharmaceutical increases the radiation exposure to non target organs and indicates an issue related to the particle size distribution of PEI MP. There are several variables that can lead to

PAGE 142

142 irreproducibility of the previously reported particle size distribution including filtration technique, filter manufacturing, and degree of br anching of the base polymer The issue of particle size distribution is addressed in more detail in the followi ng section on the outlook of this work. With the exception of low clearance rate, PEI MP results were positive and aligned with those previously observed by Milner et al New studies were also conducted to examine the ability of PEI MP to target tumor type s other than osteosarcoma. B iodistribution studies conducted in dogs with non osseous tum or types indicate d little to no uptake in the tumor region. These results give insight into the tumor targeting mechanism of PEI MP. It appears that the phosphonate fu nctionalized polymer prevents uptake in healthy tissue while increasing the biological half life in the blood allowing for uptake and retention in the tumor. Additionally, based on th e lack of uptake in non osseous tumors, it appears that PEI by virtue of the affinity of the negatively charged phosphonate groups for the positively charged hydroxyapatite mineral found in bone, allowing for reten tion of this bone seeking radiopharmaceutical. Chapter 3 introduces the synthesis, characterization, and pre clinical evaluation of a new polymer ligand, polyethyleneimino ethylenediamine tetramethylenephosphonate ( PEI EDTMP ) designed to improve bindin g and delivery of radiolanthanides compared to the polymer ligand, PEI MP, presented in Chapter 2 This work is based on previous reports that PEI MP may not bind radiolanthanides with high stability in vivo The new ly designed polymer, PEI EDTMP, incorpor ates a multidentate ligand known to bind radiolanthanides with high stability in vivo Biodistribution studies

PAGE 143

143 with PEI EDTMP in a healthy canine model indicate rapid clearance of the radiopharmaceutical from the central cavity and non target organs with n o uptake in the regions associated with the dose limiting organs The clearance rate of excess PEI EDTMP was much better than observed for PEI MP in Chapter 2, with approximately 65% of the injected radio activity being excreted through the kidneys via the urine. This indicates that a better particle size distribution was achieved for PEI EDTMP. An additional biodistribution study in an osteosarcoma bearing canine indicates tumor uptake and retention of the radiopharmaceutical, with very low uptake in the do se limiting regions and low retention in the liver and kidneys. Chapter 4 investigates the thermodynamic and kinetic binding stabilities of PEI MP and PEI EDTMP with the lanthanide samarium. It was concluded that the newly designed polymer, PEI EDTMP d emonstrates improved binding with samarium in the presence of the blood plasma ion calcium, presumably due to a chelation effect resulting from the increased denticity of the chelating ligand. This encouraging result offers promise for delivery of therapeu tic radiolanthanides to osteosarcoma tumors with lower uptake in healthy tissue than currently available small molecule radionuclide therapies. Outlook The two phosphonate functionalized polymers described in C hapters 2 and 3 have demonstrated the ability to target osteosarcoma tumors more selectively than similar small molecule radiopharmaceuticals. However, two challenges remain as work continues to optimize polymer radiopharmaceuticals for tumor targeting applications. The first challenge associated w ith the development of polyme r radiopharmaceuticals is achieving appropriate particle sizes for good renal clearance of

PAGE 144

144 excess radiopharmaceutical. Optimization of particle size will decrease radiation exposure to non target organs by improving renal clear ance and decreasing liver uptake. The second challenge associated with the development of polymer radiopharmaceuticals is selection of an appropriate radionuclide with sufficient nuclear properties to treat osteosarcoma. Both of these issues may be address ed through radiopharmaceutical re design based on the t hree key features discussed in Chapter 1 polymer, radionuclide, and chelating ligand. Future work associated with these three features will be discussed in greater detail in the following sections. P olymer The polymer component of the radiopharmaceutical acts as a carrier to prevent uptake of the radionuclide by healthy tissue and facilitate passive delivery of the radionuclide to solid tumors via the enhanced permeability and retention effect. Theref ore, the most important characteristics of a polymer for use in targeted radionuclide therapy are biocompatibility, solubility, and particle size. The challenges associated with the polyethyleneimine based rad iopharmaceuticals presented in C hapters 2 and 3 has been the reproduction of particle size distribution for achieving optimal pharmacokinetics. The breadth of particle size distribution is based on the chemistry used to synthesize the polymer backbone. Polyethyleneimine (PEI) is made from a chain grow th addition type polymerization, which results in a very broa d molecular weight distribution Since polymer chains of different lengths have different pharmacological responses, it is necessary to narrow down the molecular weight distribution. In th e wor k presented in Chapters 2 and 3, narrow molecular weight distribution was achieved by a post polymerization modification by filtration of the polymer through membranes with nominal pore sizes.

PAGE 145

145 Although this approach provided qualitative insight into the pa rticle size distribution required for optimal biodistribution there are several variables that can hinder reproduction of the results including membrane manufacturing, degree of base polymer branching, and experimental filtration technique. Since precise control over polymer molecular weight distribution is essential in the design of polymer radiopharmaceuticals, future work should focus on developing base polymers with narrow molecular weight distributions. Controlled polymerization techniques offer an ap proach for synthesis of polymers with nearly monodisperse molecular weights. Controlled polymerizations are found in literature, providing several options for application in targeted radionuclide therapy. 99 105 These controlled transfer and chain termination steps have been removed, allowing for polymer chains that grow at a more c onstant rate than observed in traditional polymerization techniques. Common c ontrolled polymerization techniques include living anionic, living cationic, living free radical, and living Ziegler Natta polymerizations. By incorporating a controlled synthetic approach in future work, polymers can be prepared with optimal particle sizes for tumor targeting application in a reproducible manner. Ligand The work in Chapter 3 demonstrates the ability to improve binding between polymers and radionuclides through mod ification of the che lating ligand. There are several well known ligands that can be explored for attachment to the base polymer. G ood binding with radiolanthanides appears to have been achieved through attachment of ethylenediamine tertamethylene phosphona te (EDTMP) to the polymer polyethylenimine (PEI) Use of another radionuclide in future work may require that the

PAGE 146

146 binding stabilities be reinvestigated for the new system Improvements in binding stability may be achieved through modification of the chelat ing ligand. Radionuclide There is some concern regarding the therapeutic effectiveness of beta emitting radionuclides in treating osteosarcoma. The relative radiation resistance of osteosarcoma may be overcome by using radionuclides with higher linear ener gy transfer (LET) such as alpha emitters. However, the higher LET of alpha emitters creates difficulty in handling and testing of these radionuclides in the development of new radiopharmaceuticals An alternative approach is to use Auger emitters, in whi ch a low energy beta emitter is internalized by the cell, allowing for destruction of DNA within the cancerous tumor cells This extreme localization of the radionuclide offers the ability to deliver a high dose of radiation to the tumor while sparing dama ge to bone marrow. The Auger emitter 117m Sn has been explored as a potential therapy for osteosarcoma due to its desirable nuclear properties, 106 which include g amma emission for imaging (~160 keV) and a half life of 13.6 days. Future work should parallel the use of Beta emitters with alternative radionuclides such as Auger emitting 117m Sn.

PAGE 147

147 APPENDIX CALCULATION OF DEGRE E OF METHYLENE PHOSP HONATE FUNCTIONA LIZATION OF PEI MP Figure A 1 Differentiation o f carbons within the PEI repeat unit Table A 1. Number of carbons associated with each amine in the PEI repeat unit Figure A 2. Differentiation of carbons within the PEI MP repeat unit Type of Amine Associated with Each Carbon Relative Number of Carbons Primary (0.257)(4) = 1.028 Secondary (0.405)(2) = 0.81 Tertiary (0.338)(0) = 0 (A 1) Theoretical Carbon to Ni trogen Ratio (PEI) = sum of relative amount of carbons sum of relative amount of nitrogens = 1.838 4 carbons associated with every 3 nitrogen 2 carbons associated with every 2 nitrogen 0 carbons associated with every 1 nitrogen 4 carbons associated with every 3 nitrogen 2 carbons associated with every 2 nitrogen 0 carbons associated with every 1 nitrogen

PAGE 148

148 Table A 2. Number of carbons associated with each amine in the PEI MP repeat unit Type of Amine Associated with Each Carbon Relative Number of Carbons Primary (0.257)(4) = 1.028 Secondary (0.405)(3) = 1.215 Tertiary (0.338)(2) = 0.676 (A 2) (A 3) Approximate Degree of Functionalization = = 0.74 = 74% Actual C:N ratio (PEI MP) Ac tual C:N ratio (PEI) Theoretical C:N ratio (PEI MP) Theoretical C:N ratio (PEI) Theoretical Carbon to Nitrogen Ratio (PEI MP, 100% functionalization) = = 2.919 sum of relative amount of carbons sum of relative amount of nitrogens 4 carbons associated with every 3 nitrogen 2 carbons associated with every 2 nitrogen 0 carbons associated with every 1 nitrogen 4 carbons associated with every 3 nitrogen 2 carbons associated with every 2 nitrogen 0 carbons associated with every 1 nitrogen 4 carbons associated with every 3 nitrogen 2 carbons associated with every 2 nitrogen 0 carbons associated with every 1 nitrogen 4 carbons associated with every 3 nitrogen 2 carbons associated with every 2 nitrogen 0 carbons associated with every 1 nitrogen

PAGE 149

149 LIST OF REFERENCES (1) Ottaviani, G.; Jaffe, N. Cancer Treatment and Research 2009 152 3. (2) Broadhead, M. L.; Clark, J. C. M.; Meyers, D. E.; Dass, C. R.; Choong, P. F. M. Osteosarcoma 2011 (3) Patel, S.; Stein, R.; Ong, G. L.; Goldenberg, D. M.; Mattes, M. J. J. Nucl. Med. 1999 40 13 92. (4) Halperin, C. E. Pediatric Radiation Oncology 2005 291. (5) Ferrari, S.; Bricolli, A.; Mercuri, M. J. Clin. Oncol. 2003 21 710. (6) Kempf Bielack, B.; Bielack, S. S.; Jurgens, H. J. Clin. Oncol. 2005 20 559. (7) Harris, M. B.; Gieser, P.; G oorin, A. M. J. Clin. Oncol. 1998 16 3641. (8) Camidge, D. R.; Jodrell, D. I. In Introduction to the Cellular and Molecular Biology of Cancer ; Knowles, M., Selby, P., Eds.; Oxford University Press: Oxford, New York, 2005 (9) Ferrari, S.; Palmerini, E. Current Opinion in Oncology 2007 19 341. (10) National Cancer Institute Fact Sheet, 2009. (11) Ta, T. A.; Dass, C. R.; Choong, P. F. M.; Dunstan, D. E. Cancer Metastasis Rev. 2009 28 247. (12) Federman, N.; Bernthal, N.; Eilber, F. C.; Tap, W. D. C urrent Treatment Opinions in Oncology 2009 10 82. (13) Phillips, T. L.; Sheline, G. E. Radiology 1969 92 1537. (14) Ozaki, T.; Fledge, S.; Liljenqvist, U.; Hillman, A.; Delling, G. S. R.; Salzer Kuntschil, M. Cancer 2002 94 1069. (15) Ozaki, T.; F lege, S.; Kevric, M.; Lindner, N.; Masas, R.; Delling, G. S. R. J. Clin. Oncol. 2003 21 334. (16) Larsen, R. H.; Bruland, O. H.; Hoff, P.; Alstad, E.; Rofstad, K. Br. J. Cancer 1994 69 1000. (17) Larsen, H. R.; Bruland, O. S.; Hoff, P.; Alstad, J.; L indmo, T.; Rofstad, E. K. Radiation Research 1994 139 178. (18) van Putten, L. M. Eur. J. Cancer 1968 4 173.

PAGE 150

150 (19) Rofstad, E. K.; Brustad, T. Int. J. Radiat. Oncol. Biol. Phys. 1980 6 1163. (20) Belting, M.; Wittrup, A. Molecular biotechnology 2009 43 89. (21) MIlner, R. J.; Dormehl, I. C.; Louw, W. K. A.; Kilian, E. In EANM Congress Barcelona, Spain, 1999 (22) Serafini, A. N. J. Nucl. Med. 2001 45 91. (23) Hoskin, P. J. Cancer Treat. Rev. 2003 29 321. (24) Reale, C.; Turkiewicz, A. M.; R eale, C. A. Crit. Rev. Oncol. Hematol. 2001 37 1. (25) Saarto, T.; Janes, R.; Tenhunen, M. Eur. J. Pain 2002 6 323. (26) Lewington, V. J. Nucl. Med. Commun. 2002 23 833. (27) Bussink, J.; Kaanders, J.; van der Kogel, A. Radiotherapy and Oncology 2 003 67 3. (28) Hockel, M.; Vaupel, P. Journal of the National Cancer Institute 2001 93 266. (29) Vaupel, P.; Harrison, L. The Oncologist 2004 9 4. (30) Vaupel, P. The Oncologist 2004 9 10. (31) Jussila, L.; Alitalo, K. Physiol. Rev. 2002 82 6 73. (32) Folkman, J. Journal of the National Cancer Institute 1990 82 4. (33) Skinner, S. A.; Tutton, P. J.; O'Brien, P. E. Cancer Res. 1990 50 2411. (34) Matsumura, Y.; Maeda, H. Cancer Research 1986 46 6387. (35) Maeda, H.; Bharate, G. Y.; Daru walla, J. European journal of pharmaceutics and biopharmaceutics : official journal of Arbeitsgemeinschaft fur Pharmazeutische Verfahrenstechnik e.V 2009 71 409. (36) Frenkel, E.; Haley, B. Urilogical Oncology: Seminars and Original Investigations 2008 26 57. (37) Otte, A.; Mueller Brand, J.; Dellas, S.; Nitzsche, E. U.; Herrmenn, R.; Maecke, H. R. Lancet 1998 351 (38) Fox, M. E.; Szoka, F. C.; Frechet, J. M. Accounts of chemical research 2009 42 1141.

PAGE 151

151 (39) Chen, B.; Jerger, K.; Frechet, J. M.; S zoka, F. C., Jr. Journal of controlled release : official journal of the Controlled Release Society 2009 140 203. (40) Duncan, R. Nat. Rev. Drug Discovery 2003 2 347. (41) Owens III, D. E.; Peppas, N. A. Int. J. Pharm. 2006 307 93. (42) Maeda, H.; Wu, J.; Sawa, T.; Matsumura, Y.; Hori, K. Journal of Controlled Release 2000 65 271. (43) Jain, R. K. Cancer Metastasis Rev. 1987 6 559. (44) Brechbiel, M. W.; Gansgow, O. A. Bioconjugate chemistry 1991 2 187. (45) Ugur, O.; Kostakoglu, L.; Hiu, E. T.; Hisher, D. R.; Garmestani, K.; Gansgow, O. A.; Cheung, N. V.; Larson, S. M. J. Nucl. Med. 1996 23 1. (46) Paoloni, M.; Khanna, C. Nature Reviews Cancer 2008 8 147. (47) Liu, S. Advanced drug delivery reviews 2008 60 1347. (48) Nunez, M. I.; Villalobos, M.; Olea, N. Br. J. Cancer 1995 71 311. (49) Kassis, A. I.; Adelstein, S. J. J. Nucl. Med. 2005 46 4S. (50) Faraggi, G.; Gardin, I.; de Labriolle Vaylet, C. J. Nucl. Med. 1994 35 113. (51) E., W. T. Nucl Med Commun 1993 14 408. (52 ) Schubiger, P. A.; Alberto, R.; Smith, A. Bioconjug Chem 1996 7 (2) 165 1 79 (53) Cutler, C. S.; Smith, C. J.; Ehrhardt, G. J.; Tyler, T. T.; Jurisson, S. S.; Deutsch, E. Cancer Biotherapy & Radiopharmaceuticals 2000 15 531. (54) Brechbiel, M. W.; Ga nsow, O. A.; Atcher, R. W.; Schlom, J.; Estebam, J.; Simpson, D. E.; Colcher, D. Inorganic Chem 1986 25 2772. (55) Milner, R. J.; Dormehl, I. C.; Lowe, W. K. A.; Kilian, E. Eur J Nucl Med 1999 26 1220. (56) Botelho, M. F.; Zeevaart, J. R.; Louw, W. K A.; Dormhel, I. C.; Kolar, Z. I.; Gomez, C.; Metello, L.; Abrunhosa, A.; MIlner, R. J.; Schneeweiss, F. H. A.; De Lima, J. J. P. Eur J Nucl Med 2003 Suppl 2 S338. (57) Dormhel, I. C.; Louw, W. K. A.; MIlner, R. J.; Kilian, E.; Schneeweiss, F. H. A. Dr ug Res. 2001 258.

PAGE 152

152 (58) Jarvis, N. V.; Zeevaart, J. R.; Wagener, J. M.; Louw, W. K. A.; Dormehl, I. C.; Milner, R. J.; Killian, E. Radiochim. Acta 2002 90 237. (59) Vicennati, P.; Giuliana, A.; Ortaggi, G.; Masotti, A. Current Medicinal Chemistry 2008 15 2826. (60) Godbey, W. T.; Wu, K. K.; Mikos, A. G. J. Control Release 1999 60 149. (61) Moedritzer, K.; Irani, R. J. Org Chem 1966 31 1603. (62) Bersworth, F. C. Alkylene Polyamine Methylene Phosphonic Acids. U.S. Patent 2,599,807 June 1, 1950 (63) Hamada, Y. Z.; Harris, W. R. Inorganica Chimica Acta 2006 359 1135. (64) El Mabhouh, A.; Angelov, C.; McEwan, A.; Jia, G.; Mercer, J. Cancer Biother. Radiopharm. 2004 19 627. (65) Williams, K.; Roberts, K.; Nnalue, A. Endotoxins: pyrogens, LAL testing, and depyrogenation ; Informa Healthcare: Willmington, NC, 2001 (66) Inspections Enforcement, and Criminal Investigations http://www.f da.gov/ICECI/Inspections/InspectionGuides/InspectionTechnicalGuid es/ucm072918.htm (accessed November 10, 2009 ) (67) Steigman, J.; Eckelman, W. C. The chemistry of technetium in medicine ; National Academy Press: Washington, DC, 1992 (68) Banerjeea, S. R.; Marescad, K. P.; Franceconib, L.; Villiant, J.; Babichd, J. W.; Zubieta, J. Nuclear medicine and biology 2005 32 (69) Karesh, S. Radiopharmaceuticals in Nuclear Medicine. http: //www.nucmedtutorials.com/dwradiopharm/index.html (accessed April 20, 2012 ) (70) Park, H.; Choi, E. Polymer 1996 37 313. (71) Hoogeveen, H. G.; Cohen, S.; Fleer, G. J. Colloid Interface Sci. 1996 182 146. (72) Zhao, R.; Su, B. L. Materials Letters 2012 74 163. (73) Ziebarth, J. D.; Wang, Y. Biomacromolecules 2010 11 1. (7 4) Zeta Potential Instruments, Inc What is Zeta Potential? http://www.zeta reader.com/pages/o verview.html (accessed June 20, 2012).

PAGE 153

153 (75) Dhal, P. K.; Holmes Farley, S. R.; Huval, C. C.; Josefiak, T. H. Polymer therapeutics I: polymers as drugs, conjugates and gene delivery systems Springer: Germany, 2006 (76) Busch, D. H. Chem. Rev. 1993 847. (77) Liu, S. E., D. Scott Bioconjugate chemistry 2001 12 7. (78) Volkert, W. A.; Hoffman, T. J. Chem. Rev. 1999 99 2269. (79) Ando, A. A., I. Hiraki, T. Hisda, K. Nuclear medicine and biology 1989 16 57. (80) Shuang, L. Chem. Soc. Rev. 2004 3 3 445. (81) Xing, K.; Beauchemin, D. J. Anal. At. Spectrom. 2009 24 336. (82) Martell, A. E.; Motekaitis, R. J. Determination and use of stability constants ; Wiley: New York, 1992. (83) Jelesarov, I.; Bosshard, H. R. J. Molec. Recognit. 1999 12 3. (84) Indyk, L.; Fisher, H. F. Methods Enzymol. 1998 295 350. (85) Chiad, K.; Stelzig, H.; Gropeanu, R.; Weil, T.; Klapper, M.; Mullen, K. Macromolecules 2009 42 7545. (86) Schalley, C.; Schmidtchen, F. P. In Analytical Methods in suprmolecular chemi stry ; Schalley, C., Ed.; Wiley: Germany, 2007 p 67. (87) Feig, A. L. Biopolymers 2007 87 293. (88) Steed, J. W.; Atwood, J. L. In Supramolecular Chemistry ; Wiley: United Kingdom, 2009 17. (89) Perrozo, R.; Folkers, G.; Scapozza, L. Journal of Recept ors and Signal Transduction 2004 24 1. (90) Horton, N.; Lewis, M. Protein Sci 1992 1 169. (91) Searle, M. S.; Williams, D. H.; Gerhard, U. J Amer Chem Soc 1992 114 10697. (92) Williams, D. H.; Cox, J. P. L.; Doig, A. J.; Dardner, M.; Gerhard, U.; Kaye, P. T.; Lal, A. R.; Nicholls, I. A.; Salter, C. J.; Mitchell, R. C. J Amer Chem Soc 1991 113 7020.

PAGE 154

154 (93) GE Healthcare What is ITC? http://www.microcal.com /technology/itc.asp (accessed Ma y 3, 2006 ) (94) Cooper, A.; Johnson, C. M. Method Mol Biol 1994 22 109. (95) Chervenak, M. C.; Toone, E. J. J Amer Chem Soc 1994 116 10533. (96) Connelly, P. R.; Thomson, J. A.; Fitzgibbon, M. J.; Bruzzese, F. J. Biochemistry 1003 32 5583. (97) Serafini, A. N. Cancer supplement 2000 88 2934. (98) Williams III, R. O.; Watts, A. B.; Miller, D. A. Formulating Poorly Water Soluble Drugs ; Springer, 2012 ; Vol. 3. (99) Hayes, W. R., Steve Controlled/'Living' Polymerization Methods ; Oxford University Press, Inc.: New York, NY, 2004 (100) Johnson, A. F.; Mohsin, M. A.; Meszena, Z. G.; Graves Morris, P. Macrmolecular Chemistry and Physics 1999 3 527. (101) Benoit, D. S.; Srinivasan, S.; Shubin, A. D.; Stayton, P. S. Biomacromolecules 2011 12 2708 (102) Tong, R.; Cheng, J. Angew Chem Int Ed Engl 2008 47 4830. (103) Tong, R.; Yala, L.; Fan, T. M.; Cheng, J. Biomaterials 2010 31 3043. (104) Tong, R.; Cheng, J. Macromolecules 2012 45 2225. (105) Broyer, R. M.; Grover, G. N.; Maynard, H. D. Chem Commun (Camb) 2011 47 2212. (106) Jansen, D. R.; Rijn Zeevaart, J.; Denkova, A.; Kolar, Z. I.; Krijger, G. C. Langmuir : the ACS journal of surfaces and colloids 2009

PAGE 155

155 BIOGRAPHICAL SKETCH Sammy Jonathan Popwell was born in Honolulu, Hawaii, to p arents Sammy and Linda Joan Popwell of Tylertown, Mississippi. Sam attended the University of Southern Mississippi in Hattiesburg, Mississippi, where he conducted research under the direction of Professor John Pojman and earned a Bachelor of Science degre e in c hemistry in 2005. He began graduate studies in polymer and organic chemistry at the University of Florida, where he earned a Master of Science degree under the direction of Professor Kenneth Wagener in 2009. Sam went on to complete his D octor of Phil osophy in 2012 in m aterials s cience and e ngineering at the University of Florida under the direction of Professors Christopher Batich and Kenneth Wagener.