Copyright Warning & Restrictions

The copyright law of the United States (Title 17, United States Code) governs the making of photocopies or other reproductions of copyrighted material.

Under certain conditions specified in the law, libraries and archives are authorized to furnish a photocopy or other reproduction. One of these specified conditions is that the photocopy or reproduction is not to be “used for any purpose other than private study, scholarship, or research.” If a, user makes a request for, or later uses, a photocopy or reproduction for purposes in excess of “fair use” that user may be liable for copyright infringement,

This institution reserves the right to refuse to accept a copying order if, in its judgment, fulfillment of the order would involve violation of copyright law.

Please Note: The author retains the copyright while the New Jersey Institute of Technology reserves the right to distribute this thesis or dissertation

Printing note: If you do not wish to print this page, then select “Pages from: first page # to: last page #” on the print dialog screen

The Van Houten library has removed some of the personal information and all signatures from the approval page and biographical sketches of theses and dissertations in order to protect the identity of NJIT graduates and faculty. ABSTRACT

CHARACTERIZATION OF ELECTROSPUN DLPLGA NANOFIBERS FOR A DRUG DELIVERY SYSTEM FOR INTRACRANIAL TUMORS AND AVMs

by Kimberly A. Griswold

Eecosiig ocesses ay eecic ies o a oyme souio i oe o ouce sas o oyme i e aoscae age ese oyme ies ae maiuae o

ei oosiy ig suace aea ieess a uiomiy I oe o ceae e iea

ug eiey sysem o iacaia aeioeous maomaios (AMs a umos

oy-(acice-oy-gycoie (GA aoies wee eecosu i a

eayoua ( soe ee iee coceaio aios o GA 5/15

/ a 75/5 wee aaye o oai e ime ase o ug aeaio 13-is (- cooey-1-iosouea (CU was e coissoe wi e GA aoies

o om a omogeous souio eecosu a aaye o ug caaceiaio

e SEM aaysis sowe a ie iamee is o a ucio o ug esece

e GA gas sowe ie aiaio i mass oss i comaig e eecosu ie

o e ug comie ie e SC aaysis oie a g aue aou -53°C o

o e aw oyme a e eecosu ie is oie oo a e eecosiig ocess oes o aec e cemica aue o e oyme e esece o CU o e oyme (cose /-1w was eemie y e cage i g osee o e SC gas om a aue o -53°C o wo aues o 19°C a

19°C CHARACTERIZATION OF ELECTROSPUN DLPLGA NANOFIBERS FOR A DRUG DELIVERY SYSTEM FOR INTRACRANIAL TUMORS AND AVMs

by Kimberly A. Griswold

A Thesis Submitted to the Faculty of New Jersey Institute of Technology In Partial Fulfillment of the Requirements for the Degree of Master of Science in Biomedical Engineering

Department of Biomedical Engineering

January 2004

APPROVAL PAGE

CHARACTERIZATION OF ELECTROSPUN DLPLGA NANOFIBERS FOR A DRUG DELIVERY SYSTEM FOR INTRACRANIAL TUMORS AND AVMs

Kimberly A. Griswold

Micae ae esyAiso ae eseac oesso o iomaeias I

Caes esigiaeOmo esis Co-Aiso ae Meica oco a Assisa oesso o euoogica Sugey a aioogy UM

Micae uag Commiee Meme ae Associae oesso o Cemica Egieeig I

eea Aie Commiee Meme ae Assisa oesso o iomeica Egieeig I BIOGRAPHICAL SKETCH

Author: Kimey A Giswo

Degree: Mases o Sciece

Date: auay

Undergraduate and Graduate Education:

• Mase o Sciece i iomeica Egieeig Coceaio i oyme Sciece a iomaeias ew esey Isiue o ecoogy ewak

• aceo o Sciece i iomeica Egieeig Coceaio i Oics Uiesiy o ocese ocese Y

Major: iomeica Egieeig Coceaio i oyme Sciece a iomaeias Dedicated to the memory of my grandfather, Augustus W. Griswold, co-founder of Dynak Incorporated, engineer, innovative thinker, and inventor. Also to my devout and patient family and friends who through enlightenment allowed me to be comfortable in my own skin. Finally, to my Korean family, all inclusive, known and unknown - for the gift of life, new and old, and for anonymity.

v ACKNOWLEDGMENT

I would first like to recognize the support and guidance of Dr. Michael Jaffe and Dr.

Charles J. Prestigiacomo in not only my academic endeavors, but my personal ones as well. Dr. Michael Jaffe served as my research advisor, but was very influential in the academic path chosen and the career choices I have made. Dr. Charles J.

Prestigiacomo served as my assistant research advisor and provided constant support, encouragement, and intellectual stimulation to keep me motivated throughout the project.

Special Thanks goes to Dr. Michael Huang for providing such an amazing academic experience and guiding me to a particular field of concentration for this thesis research. It is he to whom I owe all my newfound interest and my desire to obtain greater knowledge in science.

Appreciation to my fellow graduate students in the Medical Device and

Concept Laboratory, particularly Shobana Shanmugasundaram for her assistance and knowledge in the electrospinning process, and fine-tuned SEM and computer technique. Finally, particular acknowledgment to Joseph Pickton for his judgment and knowledge of Polymer Science, his support and strong shoulder to lean on, and his keen wit for humor.

vi

AE O COES

Chptr

1 IOUCIO I

11 Oecie I

I ackgou Iomaio

I1 Iacaia ascua omaios 3

I Cue eame Meos 7

13 amakieics o CU 11

1 ioegaae oymes 13

15 Kieics o ug eease I5

1 GA Caaceisics

EECOSIIG ECIQUE 9

3 MAEIAS 33

MEOOOGY 37

1 Eecosiig ocess 37

Caaceiaio o Eecosu ies 3

3 Eecosiig a Caaceiaio o ug a oyme Come

5 ESUS I

ISCUSSIO 73

7 COCUSIO 7

AEI A CAACEISICS O ESECIE AIOS O A A GA 7

AEI IOCOMAIIIY O ACIE/GYCOIE CO- OYMES 79

vii TABLE OF CONTENTS (Continued)

Chapter Page

APPENDIX C MECHANICAL PROPERTIES OF MEDISORB® ... 81

APPENDIX D SOLUBILITY CHART OF MEDISORB® POLYMERS 82

APPENDIX E THERMAL ANALYSIS DATA OF MEDISORB® POLYMERS 83

APPENDIX F CHEMICAL PROPERTIES OF BCNU 84

APPENDIX G GMP ANALYSIS OF 75/25 DLPLGA POLYMER 87

APPENDIX H GMP ANALYSIS OF 85/15 DLPLGA POLYMER 88

APPENDIX I NON-GMP ANALYSIS OF 80/20 DLPLGA POLYMER 89

APPENDIX J POLYMER CHARACTERISTICS OF MEDISORB® POLYMERS 90

APPENDIX K ENVIRONMENTALLY SENSITIVE DRUG RELEASE SYSTEMS 91

REFERENCES 92

iii LIST OF TABLES

Table Page

31 ouc Iomaio 33

51 ie Oseaios Usig OM 3

5 aoie iamee

53 OM ie Oseaios I

5 5/15-15w ecoe iamee

55 Measue iamee o /-Iw + CU

ix LIST OF FIGURES

Figure Page

I.1 (a)Normal arteriovenous anatomy, (b) AVM, (c) advanced AVM with down stream aneurysm 4

I.2 I,3-bis (2-chloroethyl)-1-nitrosoureal 1I

I.3 Reservoir systems (a) implantable/oral, (b) transdermal 17

1.4 Environmentally responsive systems (a) reservoir, (b) matrix in swelling- controlled release systems 18

1.5 Basic changes in polymer structure for environmentally sensitive systems due to swelling 20

I.6 (a) Bulk erosion drug delivery, (b) surface degradation for drug release .... 22

I.7 Microspheres of 60 lactide:glycolide PLGA 23

1.8 75:25 lactide:glycolide PLGA microparticles by bulk hydrolysis 23

I.9 (a) surface erosion of polyorthoester rods after 9weeks and (b) l6weeks ..... 23

1.I0 Ring opening polymerization of glycolide dimer to polyglycolic acid 24

1.11 Ring opening polymerization of lactide to polylactides 25

I.12 Degradation and metabolization of PLGA 28

2.1 (a) Fluid jet, (b) Splaying, (c) Whipping motion 30

2.2 Electrospinning setup 3I

5.1 Branching morphology observed in glove samples using OM, magnification of 50 46

5.2 (a) 80/20-10wt% 11/19/03 2 gauge 48

5.2 (b) 80/20-6wt% 1I/20/03 22 gauge 49

5.2 (c) 80/20-6wt% II/20/03 22 gauge 49

5.2 (d) 80/20-10wt% 11/25/03 20 gauge 50 IS O IGUES (Cntnd

r

5 (e /-1w 1I/5/3 gauge 5

5 ( /-1w 1I/5/3 gauge 51

5 (g /-Iw 11/5/3 gauge 51

5 ( /-1w 1I/5/3 gauge 5

5 (i 5/15-w 1/1/3 gauge 5

5 ( 5/15-w 1/1/3 gauge 53

5 (k 5/I5-w 1/I/3 gauge 53

5 (1 5/I5-w 1/1/3 gauge 5

5 (m 5/15-1w 1//3 gauge 5

5 ( 5/15-1w 1//3 gauge 55

5 (o 5/15-1w 1//3 gauge 55

5 ( Goe same-mui 11/15/3 5

5 (q Goe same-mui I1/I5/3 57

5 ( 5/I5-15w 1/I7/3 gauge 57

5 (s 5/I5-15w 1/I7/3 gauge 5

5 ( /-1w + CU I/I/3 gauge 5

5 (u /-1w + CU 1/I/3 gauge 59

5 ( /-Iw + CU 1/1/3 gauge 59

53 (a GA o 1w aoie wi —3w mass oss 3

53 ( GA o 1w aoie wi —3w mass oss

5 (a SC o / GA is ea cyce

x LIST OF FIGURES (Continued)

Figure Page

5 ( SC o / GA is ea cyce 7

5 (c SC o / GA is ea cyce

5 ( SC o / GA is ea cyce 7

5 (e SC o / GA is ea cyce 71

5 ( SC o / GA is ea cyce 7

1 Icease isace iamee eceases om 5um o 333m 73

ii CAE

IOUCIO

. Objtv

The objective for this thesis research is to use the electrospinning process in order to create a D, L poly-(lactide-poly-glycolide) (DLPLGA) drug delivery system, using the drug 1,3-bis (2-chloroethyl)-1-nitrosourea (BCNU), to target intracranial vascular malformations and tumors. By creating such a system to the nanoscale, delivery of the system can be done by a guided catheter (endovascular techniques) and decreases the invasiveness of the treatment procedure in comparison to current practice regimens.

By using the electrospinning technique, a non-woven sheet of nanofibers is created by which porosity, surface area, fineness, uniformity of fiber diameter, and pattern thickness of the sheet can be manipulated. The chemotherapeutic agent, BCNU, is known to alkylate the DNA of tumor cells causing apoptosis. To conduct the research,

DLPLGA, a widely used polymer for drug release with well-characterized degradation properties in concentration ratios of 85/15 (high inherent viscosity [IV], GMP), 80/20

(high IV, not GMP), and 75/25 (high IV, GMP) was used. All three ratios were tested in order to obtain the ideal method for drug delivery. Each of the ratios had different degradation rates (different release profiles), therefore met different standards for the system.

In order to define the ideal system, the electrospinning process allows for

DLPLGA nanoscale fibers to be conjoined with BCNU and deliver the drug through a

steady degradation rate over a targeted period of time. The release rate of the drug is dependent upon the degradation rate, which is based upon the characteristics of the

2 polymer. By doing so, it is anticipated that the drug would be packaged in its most stable state and concentrated form for immediate release. The solvent, Tetrahydrofuran (THF), would allow for the insertion of the delivery of this system through a guided catheter, performed by an endovascular surgeon. The drug would be released by oxidation (by means of hydrolysis) of the nanopolymer fibers at target site, thus providing high concentrations of drug to the tumor bed while potentially greatly limiting any systemic side effects.

The research done in this thesis is an initial step in creating a more complex drug delivery system. An electrospun sheet will most certainly not be the final physical form for the delivery, but it is sufficient for the initial characterization of the system for proof of concept.

.2 rnd Infrtn

It is essential to identify the problem and create a solution that has the ability to have variance in drug choice and material choice. In neuroscience, in choosing AVM's and intracranial tumors as the problem, an extensive variety of uses for the drug delivery

system were available. The use of BCNU as the drug of choice was to narrow down a focal point for the drug delivery system model. With BCNU to create and test the model

system, it could then also be determined if other drugs could be used in place of the

BCNU with a set of data for comparison. The sheet formation used in this model is not the ideal morphology, but is compatible for the initial stages of the research. Other electrospun polymer morphology is plausible and should be investigated. 3

1.2.1 Intracranial Central Nervous System Disease

Arteriovenous Malformations (AVMs) are an abnormal collection of blood vessels that affect approximately 250,000 people (one out of I000 people are born with an AVM) in the United States [7]. Histologically, these are lesions (also known as an AVM nidus) that (typically located as a connection between the artery and the vein) provide nutrient and gas exchange to the tissue. The resultant low-resistance, high pressure and high flow state within the nidus creates a shunt. Thus, in an AVM, blood flows from arteries

straight through the shunt and into the venous system [6]. Consequently, much of the blood can be diverted from the surrounding tissue in order to flow through the AVM,

since the AVM provides less resistance to blood flow than the capillary network.

Increased blood flow through the AVM increases the pressure on the vessels, which can result in formation of aneurysms and possible hemorrhage (2- 4% chance of bursting per

year) [Figure 1.1] [7]. In short description, 1 point is given to a small tumor, the size of

approximately 3cm or an eloquent or deep AVM. Non-eloquent and superficial AVM's

are given zero points. Medium size AVM's are approximately 3-6cm with a value of two

points and large size AVM's in reference to greater than 6cms with a value of three

points. 4

(a)

(b) (c)

r . (a) Normal arteriovenous formation, (b) AVM, (c) advanced AVM with down stream aneurysm.

Cerebral AVMs represent about 80% of all intracranial vascular malformations.

Exact pathogenesis is not known though a male preponderance is reported by almost all studies [8]. Incidence of bleeding is about 70% over course of lifetime [8], and is unrelated to stress, trauma, and hypertension. There exist some studies that show the possibility of rupture during the onset of pregnancy (questionable). However, in general, the risk of bleeding is greater in children, but risk percentages decline after the age of 40

[8]. There is some evidence to suggest that because of specific hemodynamic reasons, small AVMs have a higher chance of rupture due to the build up of higher pressure in the feeding artery. 5

In addition to hemorrhage, seizures are a common presentation of AVMs

occurring in about 30% of all cases and are associated with subclinical development in

about 7% of cases with the average age of onset of 25years [8]. Malformations are so complex, because they can affect individuals during the most productive years of their

lives with such devastating consequences, and because current treatment carries

substantial risk, developing a system, which can safely and effectively treat these lesions

is vital. A drug delivery system that is compatible to all and accessible to all areas is

favored.

Primary malignant tumors of the central nervous system, despite years of study,

still carry a mean life expectancy of 13 years. There exists a four-tier system for

assessing tumor activity and progression (as per the world health organization [WHO]):

Grade 4 being rapidly growing and malignant and 1 being the slowest. The area of the brain and the size of the tumor are factors involved in determining the grade level. For

the lowest grade, grade I, treatment can be as simple as surgery. However, for higher-

grade tumors, the solution is not as simple due to the invasion of cancerous cells into

surrounding tissue, which may even be transported by spinal fluid to completely different

regions of the brain, and spinal region.

Each of the grade levels applies to each type of tumor, with slight variations in

description based on the cell characteristics and treatment methods. Astrocytomas are

tumors formed from astrocyte cells, which form part of the brain's supportive (neuroglial)

tissue [9]. Grade 2 has a 34% 5 year survival rate without treatment. With radiation

therapy, survival increases to about 70%. With grade 3, the average survival of patients

is 18months, with diagnosis by surgery or stereotactic biopsy and follow-up with 6 radiation therapy and chemotherapy [9]. In grade 4 patients, the average (mean) survival after diagnosis without treatment is 17weeks. Survival is extended up to 30weeks with biopsy followed by radiation therapy and 37weeks with surgical removal of most of the tumor tissue combined with radiation therapy. The best treatment involves stereotactic volumetric resection of the tumor tissue component followed by radiation therapy, for a patient longevity of approximately 51 weeks. Glioblastoma Multiforme (grade 4 astrocytoma) makes up 25% of all primary brain tumors. This tumor is able to infiltrate throughout the entire brain and is marked by areas of necrosis (dead tumor cells) [9].

Metastatic tumors are caused by cells from other tumors in the body transported through the blood stream to the brain. The brain tissue around the tumor begins to swell

(edema), causing focal neurological deficits, lethargy, and in extreme cases coma and death. Oligodendrogliomas originate from oligodendroglial cells that form myelin, the fatty substance that surrounds the axons of nerve cells and provides the insulation, which makes nerve cell electrical transmission faster and more efficient [9]. Tumor growth depends on the rate of mitosis and rate of apoptosis. When the mitotic rate of the oligodendroglial cells has exceeded the mitotic rate of the other cells, therefore becoming the dominant cell type, the degree of classification time changes from 6months to 30 years [9]. Most glial tumors have two growth patterns: solid tumor tissue and infiltrating tumor cells, which are judged by progressive enlargement on a computerized tomography

(CT) or magnetic resonance imaging (MRI). Solid tumor tissue is characterized by the increase in mass from the production of new blood vessels from those that supply the brain with oxygen, glucose and other nutrients. The new blood vessels function to supply the growing tumor cells with oxygen, glucose and other nutrients, directly infiltrating the source of nutrients that should supply the brain. Low grade (low blood supply needed) tumors may not be apparent on a CT or MRI due to lack of contrast enhancement

(anaplastic or malignant require large blood supply) and may be removed without neurological deficit (solid tumor tissue either displaces or replaces brain substance) depending on the lesion's location [9].

Isolated tumor cells can move between the fibers of the brain, with the entire tumor consisting of these cells which reside in "sick" but still functioning brain tissue, which makes removal difficult [9]. Most low-grade infiltrating tumor cells are visible as a hypodensity on a CT and increased Ti and T2 signal on an MRI because the tumor cells tend to draw water into the infiltrated brain tissue [9]. If isolated tumor cells are not great enough in number to change the local osmolality and thus increase the amount of water surrounding them, the CT and MI will show no changes and appear normal in the area [9]. If some tumor cells still reside in the tissue after treatment, it is guaranteed that the tumor will return. These types of tumors are more frequent in young and middle-aged adults.

.2.2 Crrnt rtnt Mthd

To locate problematic areas, computerized tomography (CT's), magnetic resonance

imaging (MRI's), and Angiogramss (x-ray movies of the blood flowing through the blood

vessels, made by injecting contrast into the arteries going into the head and taking a series

of x-ray films) are used. However, not all of these methods are completely accurate. As

mentioned in section 1.2.1 Intracranial Central Nervous System Disease, there are a

variety of lesions whose behavior is not always predictable or traceable. If the cells do 8 not produce enough of a difference in osmolality compared to normal conditions, a CT or

MRI may not detect the malformation.

Embolization is a method of plugging the blood vessels of the AVM, by an x-ray guided catheter from the femoral artery in the leg up into the area to be treated [6]. The

AVM's feeder vessels are then sealed off from within using an epoxy applied by an angiographic catheter [7]. Embolization is usually done in preparation for surgery to reduce the blood flow within the AVM or is combined with radiation therapy to obliterate the AVM. The benefits of this method are that: (1) it does not require a craniotomy; (2) eloquent deep areas can be treated by doing small areas at a time and allowing the

surrounding brain to recover; and (3) testing of the importance of the area can be done by injection of medication into the area before permanent treatment is done [6]. However, it requires multiple treatments, multiple entries into the brain, the use of new materials under "Investigation" (not yet approved by the FDA or not GMP regulated), and is used as an adjunct to another method (often combined with surgery or radiosurgery) [6].

Radiation therapy is a noninvasive procedure that uses ionizing radiation to damage the DNA in all cells (normal cells can repair damage but tumor cells cannot).

Irradiated tissue temporarily swells, that there exists a maximum volume above which the amount of swelling can cause damage. Therefore to minimize the dose given, the radiation beam is directed from several angles toward the site by a linear accelerator.

The efficacy of radiation treatment is limited by the size of the lesion, with the limit at no greater than 3cm due to drastic reduction in cure rate, with a concomitant increase in complication rate. There are different techniques being developed for radiation therapy.

Stereotactic Brachytherapy is an invasive method that uses multiple sources placed within tumor tissue to provide a dose field, with 40% of patients requiring an operation

afterwards [9]. Boron Neutron Capture Therapy (BNCT) uses the theories of atomic physics in which a boron compound is taken up by tumor cells, neutrons are absorbed

into the Boron, forming gamma irradiation and an alpha particle which kills the tumor

cell [9]. This method is usually used for tumors and not AVMs.

Surgery is an invasive procedure that involves the excision of some or the entire

lesion, with preservation of normal brain tissue. There are three steps to the procedure:

(1) establish tumor cell type and grade, (2) relieve internal pressure, and (3) reduce the

tumor burden. A craniotomy is where a portion of the skull is temporarily removed,

allowing the surgeon access to the lesion. Stereotactic procedures precisely localize areas

inside the head for biopsy, such that a stereotactic biopsy (probe directed by stereotactic

frame to tumor) is performed. Another similar method, the computer-assisted volumetric

stereotaxis gathers, stores and reformats imaging-derived, 3-D volumetric information

defining an intracranial lesion with respect to the surgical field in order to plan and

simulate the surgical procedure beforehand [9]. Volumetric stereotaxis can minimize the

incision and subsequent injury to brain tissue, thus maximizing tumor removal and

reducing postoperative complications and neurological morbidity [9]. Stereotactic

radiosurgery claims a success rate of 80% after two years of follow-up, depending on the

histologic grade of the lesion [8]. As efficient as all these methods are, surgery still is an

invasive process that involves the risk of bleeding and infection, a long recuperation time,

and is dependent on how accessible (how deep into the brain tissue) the lesion is.

Chemotherapy proves to be a promising area for advancement in approaching the

problem from a different perspective. Most agents are incorporated into and alkylate the 1

DNA of rapidly dividing tumor cells like BCNU, PCNU, Procarbazine and Carboplatin.

Vincristine poisons the mitotic process and VP-16 (Etoposide is a semisynthetic derivative of podophyllotoxin used in the treatment of certain neoplastic diseases) incorporated in tumor cell wall proteins inhibits the production of microtubules.

Topotecan enhances the effect of radiation therapy by intravenously affecting rapidly dividing tumor cells and normal cells from bone marrow and bowel. Most of these agents may result in the reduction of white blood cells (causing severe, potentially life- threatening infections), red blood cells (causing anemia), and platelets (thrombocytopenia

- resulting in blood clotting disorders and bleeding). However they are the only agents that can overcome the problems with glial primary brain tumors. Many tumor cells infiltrate surrounding brain tissue where the blood-brain barrier (which excludes large molecules) is intact. These agents can reach cells in the major mass of the tumor, which is supported by leaky tumor blood vessels, via these blood vessels. The blood-brain barrier protects tumor cells residing within intact brain tissue surrounding the tumor, but can be disrupted medically by chemotherapy.

Guilford pharmaceuticals markets polyanhydride Gliadel wafers for treating a type of brain cancer called glioblastoma multiforme. Wafers consist of new polymers impregnated with carmustine (BCNU), a toxic antitumor medication and are implanted at the tumor site, release the drug locally where it is needed while reducing systemic side effects such as liver or kidney damage. However, the wafer does have other side affects due to the drug delivery system and the means of implantation. In clinical trials, I9% of patients reported new or worsened seizure symptoms and 14% experienced healing abnormalities (such as cerebral fluid leaks, subdural fluid collections, wound 11

eakow Oe aese eacios icue iacaia iecios ai i ack a ces yeesio iaea ee ases a uiay ac iecios Comiaio o

eames is e es meo i cuig ai umos a AMs ey ca e aioe o

e seciic AM ye o umo ye a aie cosieaios u equies aiioa ses ime a cooiaio []

1.2.3 Pharmacokinetics of BCNU

CU [13-is (-cooey-I-iosouea] aso kow as camusie as e sucua omua ou i igue 1

O

C CH2-CH2-NCNHCH2-CH3-CI

NO

Figure 1.2 [1,3-bis (-cooey-1-iosouea]

e cemica aciiy o CU is o akyae e A a A o ces I as

ee kow o egae o soaeousy a meaoicay e A coss-iks

yoesie o eie om cooey caoium io [] ase o iaeous suies e aeage emia a-ie (ase o osage age om 3 o 17 mg/m was miues wi a ceaace o 5 m/mi/kg a a seay-sae oume o

isiuio o 35 /kg [] In a mg/m iaeous osage -7 was ecee oug uie a -1 eie as cao ioie oe a 9 ou ime sa 2

e emaiig -3 as o ee accoue o ye e egaaio ae occus a suc a ig ae a wii 15miues o iecio o iac ug is eecae Suies

ae sow a e eacie aue o e ug is so eesie a i ca e caie y ceea oo ow o ceeosia ui ow a oecome e oo-ai aie wi i e coies o ai issue a isace away om e imaaio sie o e eice

e oo-ai aie is oecome ecause o e comous ig ii souiiy a e eaie ack o ioiaio a ysioogica ue o is oic aue

CU may aso iii seea key eymaic ocesses y caamoyaio o amio acis i oeis og-em usage o a iosoueas ouc is associae wi

eeome o secoay maigacies I as aso ee sow o e emyooic i

as a ais a i iece i a iaaeia iacaoi oue associae wi ocua

oiciy I a suy oe o mae as gie e osage aoe o umas eiiy was imaie a caciogeic a muageic eecs wee oe Iaeousy amiisee may oe sie aecs ae cocue uo eiey ese ae umoay

oiciy (suc as iosis gasoiesia oiciy eaoiciy eooiciy a ski

yeigmeaio uo coac

e y om o e acie age ca e sae o u o wo yeas i emeaue o °C - °C (3° - ° a we susee i iue eig ous a oom

emeaue away om suig I as a ow meig oi a 35°C o 3°C (9° -

9° a ca eoessy iquey (om y om we emeaues cage 13

1.2.4 Biodegradable Polymers

The first FDA-cleared PLGA product was the Lupron Depot drug-delivery system (TAP

Pharmaceutical Products Inc.; Lake Forest, IL), a controlled release device for the treatment of advanced prostate cancer that used biodegradable microspheres of 75:25 lactide/glycolide to administer leuprolide acetate over periods as long as four months

(replacing daily injections) [I3]. "Another drug-delivery device, the Gliadel Wafer

(Guilford Pharmaceuticals Inc.; Baltimore, MD), is used to prolong the life of patients suffering from a particularly deadly form of brain cancer, glioblastoma multiforme. In this case, dime sized wafers of a biodegradable polyanhydride copolymer-poly [bis (p- carboxyphenoxy) propane: sebacic acid] in a 20:80 molar ratio - are implanted directly into the brain to deliver a powerful chemotherapeutic agent (BCNU) that has deleterious side affects when administered systemically [I3]. Release of the drug delivery in the implantable wafers was stimulated by the presence of poly (N-vinylpyrrolidone) (PVP) or sodium chloride (NaC1).

However, using different biomaterials raises considerations on biocompatibility, the reaction of the patient to the material, the effectiveness of the drug delivery system, and the cost [Appendix B]. For the purpose of this research, the amorphous form of

Polyglycolic acid/Polylactic acid copolymer (PGLA or PLAGA) (some ratio of it discussed later) was chosen for the base material. This biodegradable synthetic polymer is amorphous with a weight of 40-100kD. It has a decomposition of 100% in 60-90 days

(observed in rat stem cells). A is optimal for drug delivery systems by controlled processing so that optimal release of kinetics of the drug or active agent is achieved. is accelerated by greater hydrophilicity in the 14

ackoe o e gous geae eaciiy amog yoyic gous i e ackoe ess

cysaiiy geae oosiy a smae iise eice sie [13]

We cosieig a ioegaae oyme oe as o ik aou e

ysioogic eecs e mecaica o cemica ocess a ei oucs may ae o

e suouig aea I a suy oe i Aaa Geogia i oyme egaaio

aece ce gow e aa sowe a iec aiio o eie gycoic o acic

acis o coo ce cuues wiou AGA esue i ooioa cyooiciy [1]

e meaoic eakow oucs om AGA sigiicay owee e o e eiome a a susaie ce iaiiy o AGA memae is aiay eee o e oca eseciay i cooe imie in vitro cuues [1]

is ossiy is eeicia sice e goa is o ee ce gow Si cauio mus e use i eeig amage o iae o-caceous ces I aiio e ow o e oo sou us e esiua comoes ou o e sysem e is oay

o a issue ue o e eeme ueig caaciy o oo wic wi coiuay was away ay eease oucs om e egaaio o e GA oyme e

ieece may oy e a coce we eaig wi e ug eiey e acua ug

eomace a ay coiios e aie may ae a may e aece y e cage

GA is comise o A a GA i esecie aios o coiue o e mecaica ougess a seg ou i eie e cysaie o e semicysaie

om use o ioogica aicaio [Aei A] oymes ae ey ow

oyiseiy ie aios (PDI) a caaceie e maieace o e mecaica a sucua seg [1]

A e oymes ae ey ow oyiseiy ie aios (I o eame

e I aio o A is aou 1-19 i oe o maiai mecaica a sucua cosisecy Uiiig ig-oeig oymeiaio comies e mos commo meo o commecia oucio o A a GA wi a iseio mecaism usig a mea oie [1] A moe amoous om o e oyme ca e use o ug eiey

eices wie e cysaie om is goo o uiig scaoig a oe

ioegaae sucues GA o eame is comeey amoous so eeoe i is use oy i ug eiey eices e use o ceai ugs o eame is oiie

y e eaiey ig emeaues use i cosucig ese oymes Aoe

awack is i e cooe eease o ugs uk eosio as a somewa icosise

eease o ug eeig o e amou o ug oae oo e oyme is

yoiic o yooic oeies e iiia ae o eease ca ay [1]

.2. Knt f r l

Cooe ug eiey is iecy eee uo e aue o e ioegaae

oyme (i egaaio is e coo o eease ae e acos a aec eease ae ae cemica sucue cemica comosiio ocessig coiios mooogy a sie o imaaio Cemica sucue icues moecua-weig isiuio sae a ysica acos (sae a sie cages aiaios o iusio coeicies mecaica sesses sess- a soe-iuce cackig Cemica comosiio ioes e esece o ioic gous moecua weig e esece o ow-moecua- weig comous asoe a asoe comous (wae iis ios

ysicocemica acos (io ecage ioic seg a mecaism o yoysis 1

(eymes esus wae [] ocessig coiios wi soage isoy seiiaio

ocesses a aeaig ae a aec o e ia oea ucioaiy o e eease sysem isiuio o eea uis i muimes e esece o ueece uis o cai eecs a coiguaio sucue a a ue e caegoy o mooogy

(amoous/semicysaie micosucues esiua sesses []

e eease o e acie age may e cosa oe a og eio i may e cycic oe a og eio o e eiome o oe eea ees may igge i i oe o maiai e coec amou o osage (eeig oe- a ueosig []

ee ae ee imay mecaisms o ug eease

(I iusio; ( egaaio; (3 sweig oowe y iusio I some ug eiey sysems a comiaio o e mecaisms was use iusio ca occu o a macoscoic scae as oug oes i e oyme mai o o a moecua ee y

assig ewee oyme cais wee i o cases e mooogy o e oyme is

e cooe-eease eice [] iusio aso ca occu we e ug asses om

e oyme mai (omogeous sysem ome om e uio o oyme a acie age io e eea eiome As e eease coiues is ae omay eceases wi is ye o sysem sice e acie age as a ogessiey oge isace o

ae a eeoe equies a oge iusio ime o eease []

eseoi sysems ae esige o use e meo o iusio y ecosig a acie age (soi ug iue souio o igy coceae ug souio wi

oyme mai wi a maeia a acs as a memae aciiaig eiig o eeig moecues a a cosise ae o e eseoi sysems sow i igues I a a e

ug eiey ae ca emai aiy cosa [] e memae o e eseoi ca 17

ae a uiom oyme coaig o eie oe sie o o sies eeig o wa e simuus o e acie age is e sysem sow i igue 13(a is eeseaie o a imaae o oa eseoi eiey sysem weeas e sysem sow i igue 13( iusaes a asema ug eiey sysem i wic oy oe sie o e eice wi acuay e eieig e ug []

e eiome e acie age is eease i ca eguae coo o e ug

eiey Oce e acie age as ee eease io e eea eiome a aiioa seies o iusioa a acie aso ses ca occu a ae a iec

eaiosi wi eiomea acos

(

igue 13 eseoi sysems (a imaae/oa ( asema 18

In these systems, the combinations of polymer matrices and bioactive agents

chosen must allow for the drug to diffuse through the pores or macromolecular structure

of the polymer upon introduction of the delivery system into the biological environment

without inducing any change in the polymer itself [20]. Environmentally responsive

systems [Appendix K] when in contact with a particular biological environment, release

agent or agents in a stable way based on the kinetics of osmolality. An example of this

type of mechanism is a swelling-controlled release system. These systems operate by

absorbing water or other body fluids, which induces swelling, causing the dilution of the

active agent and the enlargement in the size of the polymer mesh. Diffusion then takes

over as the means of drug release. Figures 1.4(a) and 1.4(b) are examples of these types

of devices for reservoir and matrix systems, respectively [20]. The theory behind this

particular system is based on hydrogels, polymers that swell without dissolving in

aqueous solutions. At equilibrium, hydrogels comprise 60-90% fluid and 10-30%

polymer.

(3

Figurel.4 Environmentally responsive systems (a) reservoir, (b) matrix in swelling-controlled release systems.

The onset of swelling is not just dependent upon contact with a particular environment, but also with changes in the environment. Changes in the environment include temperature, pH, ionic strength, introduction of ionic groups into the environment, ion exchange, variations of diffusion coefficients, and enzymes or other biological elements that can induce shrinkage or swelling. Hydrophilic excipients can be used to accelerate the release of drugs, though they may also increase the burst effect

(release of active agent at one time) [18].

The simplest environmental design relies on the difference in pH values, where at high pH values, the system swells and at low values it collapses. The swelling, induction of high pH values triggers the drug release and low pH values arrests release. A number of these environmentally sensitive or "intelligent" hydrogel materials are listed in

[Appendix K]. For most of these polymers, the structural changes are reversible and repeatable upon additional changes in the external environment, as illustrated in Figure

I.5 [20]. The drug release mechanism of the system in Figure 1.5 is based upon the

swelling of the polymer. Swelling is triggered by an increase in pH value and is ideal for oral delivery where release occurs in the upper small intestine where the pH values are high and not in the stomach where pH values are low. 20

Figure 1.5 asic cages i oyme sucue o eiomeay sesiie sysems ue o sweig

Coay o e esig o e sweig sysems ae e ioegaae sysems

ioegaae sysems maiuae aua ioogica ocesses o egae e oyme

eeoe eeasig e ug a was ae i o o e oyme I suc sysems e cemica sucue o e sysem cages a e eease ae is oe goee y e

egaaio ae i some eaiosi Mos ioegaae oymes ae esige o

egae ia e yoysis o e usae oyme cais as soo as ey come io coac wi wae I e simes mecaism o cemica yoysis wae eeaes

e uk eeeiay aackig e cemica os i e amoous ase a coeig og oyme cais io soe wae-soue ioogicay comaie comous o some oyme maeias suc as oyacies oygycoies a ei cooymes e oymes wi eak ow o acic aci a gycoic aci ee e

Kes cyce a e ue oke ow io cao ioie a wae a ecee ou 2 o e oy y oma ioogica ocesses Sice egaaio iiiay occus i e amoous ase e eucio i moecua weig oes o aec ysica oeies o ucioaiy o e eice

e eice is e ogee y e cysaie egios a emais iac ui e

egaaio eaces a ciica oi omooymes (oygycoic aci a oyacic aci ae cysaie esey acke a ae o ie yoyic aack uike cooymes wic ae amoous o amoous oymes a yooic aye (a cooyme o acie gacie a cacium seaae o e suace ca e geeae

omig a asoae aee o-akig uica wic ees wae (moecues cao eac cai segmes i amoous aeas o iiiae yoysis sows asoio a imoes e eeio o esie seg [17]

Oce e eice is oke ow e agmes ae ue euce o sime comous ue o eymaic a meaoic aecs We suc a acio occus wee

ee is a ai oss o oyme mass (e ae a wic wae eeaes e eice ecees a a wic e oyme is coee io wae-soue maeias i is cae

uk eosio [1] oug uk eosio e oyme egaes i a aiy uiom mae ougou e mai o y suace egaaio as i suc maeias as

oyayies a oyooeses I suace egaaio e eease ae is

ooioa o e suace aea o e ug eiey sysem esuig i a sowe ae o coesio o e oyme io wae-soue maeias A yooic oyme wose cemica os ae igy susceie o yoysis ca eeiece a om o suace

egaaio u is oe eee o as ioeosio [1] igue I(a iusaes uk eosio a igue 1( sows suace egaaio

igue 1 (a uk eosio ug eiey ( suace egaaio o ug eease

Micoaices ae a eame o uk eosio sysems use mos commoy i oa eiey sysems a sucuaeousy iece eiey sysems Micoaices o

oy (acie-co-gycoie (GA ca e eae i a aiy uiom mae o oie esseiay ooous micosees [igue 17] a euce o agmes y uk

yoysis [igue 1] (eame ae o a 755 aciegycoie GA micoaice ae 133 ays o egaaio i wae []

Aaysis o oyooese (suace-eoig oymes os ae 9 a 1 weeks o imaaio i ais sows sigiica suace egaaio u e coe o e ug

eiey sysem emais iac [igue 19] [] 2

r . Micosees o aciegycoie GA

r .8 755 aciegycoie GA micoaices y uk yoysis

(a (

r . (a suace eosio o oyooese os ae 9weeks a ( 1weeks 24

.2.6 GA Chrtrt

Poly-glycolic acid (PGA) [Figure 1.I0] is the simplest linear aliphatic polyester, with a

melting point of 220-225°C and a glass-transition temperature of 35-40°C due to its high

crystallinity (45-55% crystalline). Such high crystallinity prevents solubility in most

organic solvents with the exception of highly fluorinated organics such as

hexafluoroisopropanol. Ring-opening polymerization yields high-molecular-weight

materials, with approximately I-3% residual monomer present and dimerization of

glycolic acid results in glycolide monomers. Unlike the slow absorption of Poly-lactic

aci (A GA is asoe wii a ew mos osimaaio ue o geae

yoyic susceiiiy [15]

11- - - I I 1 5 4 3

Gycoie - ime o oygyc oic Ac i Gycoic aci

r .0 ig oeig oymeiaio o gycoie ime o oygycoic aci

I io eeimes cocue a eymes ue (moisue aeaig

eames a gamma aiaio eace egaaio o GA eeoe ow umiiy

eyee oie gas seiiaio oceues a moisue-oo ackagig ae uiie

Gamma iaiaio ca e use we e ose o egaaio is esie a a ase ae 25

PLA (poly-) [Figure 1.II] is prepared from the cyclic diester of lactic acid

(lactide) by ring opening polymerization. Lactic acid exists as two optical isomers or enantiomers (D and L), the L-enantiomer (crystalline) occurs in nature, and a D, L racemic mixture (LPLA, a semicrystalline polymer) can be made by synthetic preparation

of lactic acid.

n

acie oyacie (ie o acic aci

Figure 1.11 Ring opening polymerization of lactide to polylactides.

Poly-L-lactide is about 37% crystalline with a melting point of 175-178°C and a

glass transition temperature of 60-65°C [I7]. Electrospun fibers from L-polylactide (mp.

I70 C) have high crystallinity, whereas poly DL-lactide electrospun fibers are amorphous

(no reported melting point [mp]). Poly L-lactide is more resistant to hydrolytic

degradation than the amorphous DL form, due to its crystalline nature, which exhibits

high tensile strength and low elongation, resulting in a high modulus (more suitable for

load bearing devices). DLPLA is an amorphous polymer exhibiting a random distribution

of both isomeric forms of lactic acid, is unable to arrange into an organized crystalline

structure, and has lower tensile strength, higher elongation, and faster degradation time

(ideal for drug delivery systems). 26

Carboxyl-ended PLGA polymers degrade slower than hydrophobic end-capped

PLGA polymers. Polylactides are more hydrophobic; therefore take a longer time to

degrade, unless plasticized with triethyl citrate (producing a less crystalline, less tensile,

faster degrading material [17]). Time required for poly-L-lactide implants to be absorbed

depends on polymer quality; processing conditions, implant site, and physical dimensions

of the implant [I5]. In vivo studies in rats showed an absorption time of about I.5 years

for 50-90 mg samples of radiolabelled poly-DL-lactide implanted in the abdominal wall

[15]. In the radiolabelled implants, metabolism of the polylactides resulted in excretion

primarily via respiration (CO2) and exposure to gamma radiation showed a decrease in

molecular weight (MW). The degradation time of LPLA takes more than two years to be

completely absorbed, much slower than DLPLA. Copolymers of L-lactide and DL-

lactide have been prepared to disrupt the crystallinity of L-lactide and accelerate the

degradation process [17]. The rate of degradation of lactide based polymers and in

general all hydrophobic degradable polymers, depends on chemical composition,

crystallinity, and hydrophilicity. Degradation by chemical composition depends on the

rate of degradation of the bonds present (Anhydrides faster than esters, faster than

amides). The higher the crystallinity and hydrophobic nature (versus hydrophilic), the

slower the degradation rate. Schwendeman and Zhu (Zhu and Schwendeman, 199 et al.,

2000) have shown that by incorporating basic salts as excipient polymeric microspheres, the stability of the incorporated protein improved, with the side effect that these basic

salts slowed degradation time.

Amorphous copolymers have a compositional range between 25-70 mole percent glycolide, where pure polyglycolide is about 50% crystalline and pure poly-L-lactide is 2 about 37% crystalline. There is a disruption of the regularity of the polymer chain by the other monomer in a racemic mixture. A copolymer of 50% glycolide and 50% DL- lactide degrades faster (in 50-60 days) than either polymer independently, or copolymer with differential percentages of either polymer. A copolymer of 90% glycolide and 10%

L-lactide was developed by Ethicon under the trade name Vicryl, which absorbs within 3-

4 months but has a slightly longer strength-retention time [17]. In general, the 65:35,

77:35, and 88:15 D, L-lactide/glycolides have progressively longer in vivo lifetimes, with the 88:15 lasting about I50 days in vivo, whereas poly (D, L-lactide) requires about 12-

16 months to biodegrade completely, and poly (L-lactide), being more crystalline and less hydrophilic, can be found in vivo in about 1-I/2 to 2 years (Cutright et al. I974,

Hausberger & De Luca 1993, Holland et al. 1986, Matsusue et al. I992, Pistner et al.

1993, Yamaguchi & Anderson I993) [19]. Factors that affect the performance of biodegradable polymers (in general) are hydrophilicity, crystallinity, melt and glass- transition temperatures, molecular weight, molecular-weight distribution, end groups, sequence distribution (random versus blocky), and presence of residual monomer or additives. The most common functional groups with these characteristics are esters, anhydrides, orthoesters, and amides. A more hydrophilic backbone, less crystallinity, more porosity, more hydrophilic endgroups, more reactive hydrolytic groups in the backbone, and smaller device size are factors that reduce degradation (accelerates the action of) time. 28

0 H 0 O 1 I I I ( (" I (C _ )2 + -+ CI _ C _ C _ O + _ C _ C _ O

Figure 1.12 egaaio a meaoiaio o GA

Meiso oymes (oymes use i is eseac Aei C E G H, I, J)

ae ioegaae oyeses a uego esoio o ysioogica coiios a

ayig aes eeig o cemica caaceisics o e oyme a aiues o e

eice egaaio occus ia yoysis o ese ikages oowe y gaua eosio

o e eice ike ei geea coueas e ia oucs o G egaaio ae

acic aci a gycoic aci wae soue o-oic oucs o oma meaoism

a ae eie o ue meaoie o cao ioie a wae [igue 11] CHAPTER 2

ELECTROSPINNING TECHNIQUE

Electrospinning is the process of charging a drop of polymer in solution with tens of thousands of volts undergoing metamorphosis from a Taylor cone into strands of polymer on the nanoscale size (by spraying). Electrostatic charging of the fluid at the tip of the nozzle results in the formation of a Taylor cone, in which multiple filaments are ejected to produce nonwoven materials that have a porosity factor, high surface area, and fineness and uniformity. The filaments are the result of the jet accelerating and thinning in the electric field, a process known as "splaying." Studies suggest that the most important element operative during electrospinning is the rapid growth of a non- axisymmetric, or "whipping," instability that causes bending and stretching of the jet

[Figure 2.1(a)] [2]. The unstable region of the jet [Fig. 2.I (b)] viewed at exposure times down to the millisecond, has the appearance of an "inverted cone," while in [Fig 2.1(c)], using high-speed photography, a single, rapidly whipping jet was observed by Shin et al.

The whipping frequency is so fast that the jet appears to be splitting into multiple filaments [2].

It was demonstrated by Hohman et al. that there are three different modes which are unstable: (I) the Rayleigh mode, which is the axisymmetric extension of the classical

Rayleigh instability when electrical effects are important: (2) the axisymmetric conducting mode; and (3) the whipping conducting mode. (The latter are dubbed

"conducting modes" because they only exist when the conductivity of the fluid is finite

[3].) Hohman et al. demonstrated that the dominant instability strongly depends on the

29 3

fluid parameters of the jet (viscosity, dielectric constant, and conductivity) and also the

static charge density on the jet.

‘igis

(

r 2. (a) Fluid jet, (b) Splaying, (c) Whipping motion.

The actual process of electrospinning involves applying a high voltage to a syringe (or some form of conduit) and pumping a polymer solution through it at a steady rate. Nanofibers of polymer are collected as a nonwoven sheet on a grounded plate below the capillary [Figure 2.2].

In the absence of an electric field, the fluid forms a drop at the exit of the conduit; its size is determined by surface tension, distance between conduit and grounded plate, and pumping rate. When an electric field is applied, a charge is introduced to the fluid that quickly relaxes the fluid surface, creating a tangential stress, resulting in the deformation of the droplet into a conical shape (Taylor cone) [5]

Polymer solution Jet

Pipette ge Metering Pump Taylor cone

High Voltage Collector screen Supply (Rotating or Stationary)

r 2.2 Electrospinning setup.

Hohman et al. demonstrated that when the electric field exceeds the critical value needed to overcome the surface tension, the apex of the cone ejects a fluid jet, where at low viscosity, the jet breaks up into droplets, while at higher viscosity the jet forms a continuous, small-diameter filament. Droplet formation depends on viscosity, therefore the solution or melt viscosity is shown to affect both the processing window for electrospinning and the diameter of the fiber [4]. Electric field parameters also affect fiber morphology.

Rutledge and Shin discovered that the small-diameter fibers generated when the filament became unstable and split into smaller filaments (splaying), was actually the main filament whipping around, stretching into a single long fiber. They demonstrated this concept with poly-ethylene oxide (PEO) solutions of varying concentrations (i.e. viscosity) in water, with KBr to adjust the fluid conductivity. The results of their experimental analysis was that the solution flow rate and the electric field strength were the key operating parameters determining the jet's stability and that the measured current 32 of the jet provided information about the surface charge density on the jet [5]. Larrado and Manley determined that doubling the applied electric field decreased the fiber diameter by roughly half. However, Baumgarten showed that the diameter of the jet reached a minimum after an initial increase in field strength and then became much larger with increasing fields, an effect caused by the pumping rate [4]. Therefore he concluded that increasing the field does increase the electrostatic stresses, creating smaller diameter fibers, but it also draw more material out of the syringe [4].

In general increasing the grounded surface distance, decreases fiber diameter; increasing electric potential (kV), decreases fiber diameter to a critical point; increasing flow rate, increases fiber diameter; increasing concentration wt%, increases fiber diameter, and decreasing the conduit size decreases fiber diameter. CHAPTER 3

MATERIALS

The DLPLGA (D-L-Poly-Lactic-Glycolic-Acid) polymer in ratios of 75/25, 80/20, and

85/15 was obtained from Alkermes in 50gram packages (per ratio). Due to their highly hydrolytic nature, the polymers were kept in airtight bags in the freezer. The samples of

75/25 and 85/15 DLPLGA are GMP approved, but the 80/20 DLPLGA was not.

Obtained from the Medisorb® Polymer Products (division of Alkermes) was the following product information [Table 3.11:

Biodegradation rates depend on: device geometry, porosity, lactide:glycolide ratio, MW/inherent viscosity (IV)/Polymer end groups, and crystalline vs. amorphous character.

Table 3.1 Product Information

85/15 DL 2A 85 Mole % DL Lactide, 15 Mole % Glycolide Acid End Group High IV 0.66-0.80 (DL/G) Degradation range of 5-6months Low IV 0.50-0.65 (DL/G)

75/25 DL COI 75 Mole % DL Lactide, 25 Mole % Glycolide Custom 01 High IV 0.66-0.80 Degradation rate 4-5 months Low IV 0.50-0.65

Standard end group is lauryl ester (capped), polymer identifier marked with either A, M, or C.

A - polymers contain a free carboxyl end group (uncapped)

M - polymers contain a methyl ester end group (capped) which lower inherent viscosity with a higher Tg

33 4

C - oymes ae cusom oymes i wic e e gou wi ay eeig o e

ac oee e C esigaio may aso icue oyme a as a agee Iee

iscosiy age a oes o comy wii e Saa Iee iscosiy

Seciicaios eaie eow

age oyme Iee iscosiy (I Seciicaio ages

Polymer Ratio Polymer Identifier IV (DL/G) End Group

A ow I 5 - 5 Ese

A ig I - Ese

e moecua weig o e 75/5 GA was II3K 13K o e 5/15 same a 1K o e / GA same e moecua weig was imoa i eemiig ow we e oyme wou go io souio a a esimae o ow e eecosiig sames wou u ou (iamee sie

Oe imoa caaceisics o e oymes ae ou i Aei C E G I

e soe use was eayaua ( ecause i was oe o ew soes

a cou issoe o e GA a e CU wiou isuig e ucioaiy o eie e oyme o e ug [Aei ]

e ug o coice was eee y Caes esigiacomo M om e

euosugey eame a e Uiesiy o Meicie a eisy o ew esey

(UM Sice CU is a commo cemoeaeuic age i e eame o

aious yes o ai umos sigiica amous o aa egaig is amacokieics a ioaaiaiiy i ieig eaaios ae eaiy aaiae o comaiso makig

is e iea coice o es e aiiy o eie cemoeaeuic ages ia aoie

ecoogy As meioe i e secio 13 amacokieics o CU CU was

e ug o coice we Giae makee ei ioegaae imaae waes o

ee umo eoccuece

e ug was oaie om e amacy a UM i owe om o

issouio wi e oyme i e soe

oieay ame CU

USA a ame Seie Camusie A aoe cemoeay age SC

99 (1977

Acie Age Camusie

Commo ames CU iCU Camusie

Cassiicaio Akyaig Age iosuea

Aeage aie osage as a sige cemoeaeuic age 15-mg/m eey

weeks

oeia Sie Eecs oe Maow Suessio Aemia iaea ow Wie oo

Cous ow aee Cous Kiey amage ug amage (umoay oiciy

Mou Soeess Swaowig iicuies [1]

Mus e soe i eigeao eoe use a ecomoses a -9 ° Use wii

ous a oom emeaue a oec om suig

Ae iaeousy amiisee CU ca coss e oo ai aie ue o is ig

ii souiiy a ack o ioiaio a is aiy egae wi o iac ug

eecae ae 5miues [1] 6

eseac aaysis sowe a e oiciies o CU ae ue o meaoies;

aoimaey -7 o a oa ose is ecee oug uie wii 9ous a

aoimaey 1 as esiaoy CO e emaiig -3 is ukow [I]

e Eecosiig Assay was oaie om muie souces a was

cosuce y Soaa Samuga Suaam o e Meica eice a Coce a a

e ew esey Isiue o ecoogy (I e oage souce Gamma ig

oage eseac (Omo eac o 5k was use o ay eecic ie a

ee oes

e syige was om ise Scieiic o sies -1 gages wee a (oue

iamee o 3" a ie iamee o 3" a gage (oue iamee o "

a ie iamee o 1" was use

e scees asic wa a gass secime sies wee use as e coecig

suace Mos o e scees a asic wa wee oaie om e oca gocey soe

o awae soe isea cooos micoscoe sies (sie 5 75 1 mm om

ise Scieiic (isug A wee use CAE 4

MEOOOGY

e eseac was si io ee isic oceues (1 e Eecosiig ocess (

Caaceiaio o e oyme as eecosu ies usig aayica eciques a isumes a (3 Eecosiig a caaceiaio o e ug a oyme come

4. Eltrpnnn r

(1 eae e oyme souio - GA was issoe i E o make -w 1-w a 15-w souios

( Souio ook 3-5miues o comeey issoe e oyme (usig a mageic siig o a miig aaaus a e was e o sa oeig

(3 e syige wi a gage o -gage iamee eee o oyme souio was eae o e Syige um

( A o-coucig asic ue was coece ewee e syige a e eee

(5 e eee was ace oioay a eeicua o e eica suace (e wie scee asic im o gass secime sie o coec e eecosu aoies e isace ewee e i o e eee a e eica suace was ewee 15-cm

( e eecoe wie was coece o e syige eee e eica suace goue

(7 Aac e syige um o coui a eake some o e souio ou oug e syige

( We e souio sae o ow ou o e syige eee someimes a ea o souio wou om a e i ockig e ow A ai o weees was use o cea e osucio

(9 Oce e ow o e souio was cosa (syige um se a 13m1/mi e owe suy was ue o a oage se a 5k

8

(1 Ae eecosiig a e sames e owe suy was ue o

(11 e syige um was ue o

(1 e eecosu same was e aee a eae o caaceiaio a micoscoy

(13 A maeias wee e ceae

4.2 Chrtrztn f th Eltrpn br

ieeia Scaig Caoimey (SC — AQ1 is use o ema aaysis y measuig saiiy oug ea ow (ca eie e a ea/coo/ea ocess o a ea

ocess

(1 eae eecosu ie same y cuig a iece o e ie ma om e gass sie

( Usig SC comue sowae aciae e cooe u o SC gas

(3 Usig a aumium a ae e a a weig ie same ace a i o e same a a cim e i oo e a

( Usig e Wia oio o e coo a se aamees o same (e emeaue scae ae o emeaue cage same weig

(5 u same ga a sae ga o ie

( Uoa same u o SC gas a eaciae e cooe

ema Gaimey Aaysis (GA - AQ5 usig weig as a measue o

emeaue Wi iceasig emeaue esiua mass is os

(1 u o GA gas

( Usig GA same oaig ock us uace uo

(3 us same uo o oai access o e aace o o ouc aace

(4) Once the weigh tray is detached from the balance, tare the weigh tray with empty aluminum pan.

(5) Load sample onto the pan. Using the TGA computer software, attach weigh tray to the balance and run sample.

(6) Graph sample and save on file.

(7) Unload sample, turn off TGA gas.

Scanning Electron Microscopy (SEM) is an electron microscope that uses a beam of focused electrons across an object to produce a high-resolution (three-dimensional) image.

1 Sign the log in book.

2 Using the SEM computer software, log in, record initial parameters, and check with log in book.

3 Prepare sample, using tweezers, gloves, and loading disk.

4 Click vent (Nitrogen gas) to get to atmospheric pressure in sample chamber to load sample.

5 Load sample (still wearing gloves) and press pump. Make sure door to sample chamber is shut. At this point the gloves can be taken off.

6 Using right joint stick, bring stage up to I1 mm (or two finger width) distance between the sample and the power piece.

7 Apply voltage. Using tool bar, use center point locator, magnification, focus, and stigmation to obtain the sample area of investigation.

8 Scan sample area under investigation and save in file.

9 Turn off voltage and write down final parameters in log in book.

10 Turn vent on, and unload sample (wear gloves).

11 Turn pump on and log off software. 40

4.3 Electrospinning and Characterization of Drug and Polymer Complex

Characterization was done with the SEM to determine if the drug had any affect on the polymer diameter or morphology.

DSC was run to see any changes in temperature based on the presence of the drug.

TGA was run to determine if the amount of mass loss in the polymer sample and at what temperature degradation might take place.

All three would determine if the process of putting the drug on the polymer worked. It was the goal of this research to see if the drug was present in the polymer.

Further analysis will have to be done on the amount, release rate, and functionality of the drug in the electrospun polymer. Also whether the electrospinning had any affect on drug performance. CHAPTER 5

RESULTS

Coosig e soe was a cucia se i e ocess I oe ai wi e eecosiig ecique meyee coie (MeC1 was use as e soe o a 75/5

GA Ow acice same is was o e soe o e use o e ug

eiey sysem (wic was cose o e eayoua [] y ausig e

ecique a oaiig aequae ies e acua eeimea sames wee mae

We usig a igy oaie soe ceai meos eee o e iouce ae goes eyegasses mask (coeig ose a mou a a ume oo

(wee a e miig occue was use Wie miig e oyme a e soe usig 1mI o soe a aious gams o oyme (1gams o Ow gams o

w I5gams o 15w a gams o w ue o e aue o e soe 5

(y oume o soe sowy issiae om a Im1 souio Oce e oyme was mie e sames souios wee seae o ee ue eaoaio o soe

(eeoe ceaig a moe coceae souio a oigiay iee Ae eig

e soe see oeig e eecosiig ocess commece

e sames mae wee a Ow o / a 5/15 a a w o /

5/15 a 75/5 oyme I e iiia sages o e eecosiig ocess a sies o

aious same weig souios wee ase o ies (oy oes wee ese we secimes wee eauae wi e oica micoscoe (OM Oe eecaio o is oseaio is a e souio was o iscous eoug a eee a oge ime o see

Ae a ew ays (3-5 ays e /-Ow emosae ies o e sie wi muie eas i e sas (usig e OM a e 5/15-Ow sowe a ew sas

41 42

but mostly atomized droplets (using the OM). After a two-week period, the 80/20-

1 Owt% was completely solid, where the THE completely evaporated. The 85/I5-10wt%

solution was had very high viscosity and therefore could not undergo the electrospinning

process. The 85/15-6wt%, 80/20-6wt%, and 75/25-6wt% still had low enough viscosity

to electrospin. At the two-week period, the 85/15-6wt% seemed really viscous (not

quantified, just a physical observation) and there was some doubt as to whether it would

pump through the syringe. Surprisingly, the solution provided a small diameter fiber

network with no beading. Under optical microscopy, there were sections of the fiber that

were larger and flatter while other sections appeared round and thinner. The polymer

samples proved to be sustained at a max time limit of two weeks, in comparison to the

MeCl, which had only a one-week longevity period, but produced fibers within a day of

mixture. This may indicate that the entanglement of the polymers can be sustained for a

certain range of time depending upon the solvent used. The eventual hardening of

solution is due in part to both entanglement properties and evaporation.

In the process of spinning nanofibers, initially a 20gauge needle was used, but a

22gauge proved useful in creating smaller diameter fibers and providing extra pressure at the point of ejection of the fluid jet. Interestingly, the smaller needle diameter provided by the 22gauge allowed for fiber formation to occur in sample solutions that were creating only droplets with the 20gauge. Smaller diameter fibers were also created by increasing the distance between the tip of the fluid jet and the screen from 15mm to

—25mm. Gradually as time increased, with the fibers undergoing entanglement, there was a higher probability for fiber formation, and an increase in density of fiber samples

(observed under the OM [Table 5.I]). 4

The 80/20 1 Owt% samples on 11/25/03, observed under the OM, defined a distinction between beading and density. The 20gauge electrospun sample looked as if there were more beads intertwined in the fiber network than the 22gauge. This was due to the dense network (more fibers per um^2) of fibers by the 20gauge than the 22gauge.

The 22gauge did not produce a thick fiber mat (sheet of fiber sample on slide), therefore beading seemed noticeably less than in the 20gauge slide. In comparison though, based on the umA2 and beading average, the two samples produce the same results.

bl . Fiber Observations Using OM

Date Polymer Wt% Needle size Description 11/19/03 80/20 lOwt% 20gauge Fibers present, but lots of beads 11/19/03 80/20 6wt% 20gauge Droplets 11/19/03 85/15 lOwt% 20gauge Droplets 11/19/03 85/15 6wt% 20gauge Few fibers, mostly droplets 11/19/03 75/25 6wt% 20gauge Droplets 11/20/03 80/20 10wt% 20gauge Lots of fiber formation with beading, formation in 3D, fiber diameter more uniform, beads encapsulated polymer/solvent I 1/20/03 80/20 lOwt% 22gauge Fibers with lots of droplets, fibers not densely packed, organized pattern appearing, uniform diameter 11/20/03 80/20 6wt% 20gauge Droplets 11/20/03 80/20 6wt% 22gauge Some fibers, some droplets 11/20/03 85/15 lOwt% 20gauge Droplets 11/20/03 85/15 lOwt% 22gauge Few fibers, mostly droplets 11/21/03 80/20 I Owt% 20gauge Large amounts of fiber, scattered beading, uniform diameter 11/21/03 80/20 lOwt% 22gauge Fibers with beading 11/21/03 80/20 6wt% 22gauge Fibers with beading 44

bl . ie Oseaios Usig OM (Coiue 11/21/03 85/15 lOwt% 20gauge Fiber formation with beading 11/21/03 85/15 1 Owt% 22gauge Fiber formation with beading, not much difference from 20gage 11/21/03 85/15 6wt% 20gauge Few fiber strands, lots of droplets 11/21/03 85/I5 6wt% 22gauge Few fibers, less droplets 11/25/03 80/20 1 Owt% 20gauge Nice fiber network, uniform diameter with beading present 11/25/03 80/20 lOwt% 22gauge Uniform fiber formation, not as dense a sample, beading still exists 11/25/03 80/20 6wt% 20gauge Droplets I1/25/03 80/20 6wt% 22gauge Droplets 11/25/03 85/15 lOwt% 22gauge Uniform small diameter, fiber formation with less beading 11/25/03 85/15 6wt% 22gauge Few fiber formation, some beading, diameter looks smaller than other samples 11/25/03 75/25 6wt% 22gauge Droplets 11/25/03 Glove Multi Multi Using the sample obtained from the gloves worn, sample on the glove showed a branching network of the fibers, larger diameter than all other samples, and densely packed (3D network) 12/01/03 80/20 6wt% 20gauge Droplets 12/01/03 80/20 6wt% 22gauge Droplets 12/01/03 85/15 6wt% 20gauge Large fiber diameter, branching network, no beading 12/01/03 85/15 6wt% 22gauge Nice fiber formation, no beading, sections where fiber diameter gets larger 12/01/03 75/25 6wt% 22gauge Droplets 12/04/03 85/15 1 Owt% 22gauge Extremely large fiber diameter, no beads, areas where diameter thick and gel-like 12/04/03 85/15 6wt% 22gauge Fiber diameter smaller, sections where diameter gets thick and gel-like 12/04/03 75/25 6wt% 22gauge Droplets 45

Incidentally, during the electrospinning process on I2/04/03 the sample 85/I5-

I Owt%, an indirect relationship was noted. After turning off the power source and allowing the syringe to be placed on the collecting screen, solution was still electrospinning out. There was a slight attraction that still existed between the needle tip and the grounded collecting surface which induced fibers to form and be pulled from the solution stream (residual pressure within the syringe still existed even when the syringe pump was turned off. The distance between the syringe needle and the grounded surface was about 0.75mm. Since the fibers formed by this method were large enough to be seen by the naked eye (the size of human hair fibers), the actual circular whipping motion of the fiber was observed. Using the OM, the fibers were large in diameter, but still created a pattern similar to what was observed when the power source was on and the distance between the needle tip and the grounded surface relatively large (in comparison). Also, the glove sample (fibers taken from latex glove collected during electrospinning process) was shown to have interesting morphology, a branching network based on all the fibers collecting on one surface (all different ratios and wt%) and interacting (Figure 5.1). 46

r . acig mooogy osee i goe sames usig OM magiicaio o 5

Uo eiew o e eimiay aa om ae 51 eemiig wic oyme

aios a weig eceages oie e mos uiom a sma iamee ies sames o /-w a 5/15-15w wee use o suseque suies as ose secimes a a e ossiiiy o oiig e mos cosise ie iamees

Iiiay as osee eoe e souios i o si ou aoies u oy oes wee ese o e sie Ae a week e /-w esue i some eecosu

ies wi a miue o oes a eaig e souios wee e oe (o eae y asic wa i oe o see u e ocess o eaoaio a ossie

oyme seig I aou oe week ae miue e moe coceae souio

(15w a ess eaoaio a e ess coceae (w o souios sae ou wi 1m1 o u ae oe week e 5/15-15w si a 1m1 o soe a e /-w a oy 5m1 o soe Wi e ecease i souio oume u 47

o esiy e /-w was o acuay w/w e 5/I5-I5w su ou e

es iamee ies om e wo souios e /-w ouce a ew sas (3-

5 u o e same oume as e 5/15-I5w eeoe e eageme ocess is

o a ucio o icease coceaio y eaoaio u y ue eageme is is oe y e eaoaio o aou 5 o souio om same /-w a e

oucio o a ew ies o sames a e same ag-ime eeoe ime is o a

aco o comaiso

e es esus o e eecosiig ocess i egas o w a oyme

aios o acie o gycoie wee e /-1w a e 5/I5 w ese wo

aios wou e use i comiaio wi 5m1 o CU (icuig iue o es e eecosiig ocess o e ug

e oma osage equie o CU amiisee iaeousy is ewee

15-mg/m ase o e kow ackgou iomaio a e eecosu

aoies ae a ige suace o aea o oume aio a ese oume oe a sma aea (eeoe ige coceaio o acie age a i o e osage (5mg/m was use Ae eig e sames es 1 ous e souios wee eecosu e e

ay wi oy e /-1w us ug oucig ies e 5/I5-w i o

e oowig wee a acos a wee ke cosa ug (5mg/m o CU a iue soe ( syige iamee u-u (-5k scee isace om syige (I5mm-5mm a souio ag-ime (oeig

I oe o oai e eecosu ies wi e smaes iamee wiou e iusio o oes iewie amog em e oowig was cosiee (1 e syige gauge sie e oage e coceaio o w sames a e isace 48 between the tip of the syringe and the vertical plate. By keeping the syringe gage small, the voltage large as possible (without having droplets form again due to too high an electric field), the smallest concentration of wt% samples, and increasing the distance between the tip of the syringe and the vertical plate, the fibers below [Figure 5.2,

Appendix L] were observed using SEM analysis.

-GA / 11/19/3 1wiw; gage

E = 1 k Siga A = MSE Mag = 3 K W= mm Siga = Ies

Figure 5.2 (a) 80/20-lOwt% 11/19/03 22gauge.

5

After 11/25/03, there was more extrusion of fibers therefore a greater diameter range was noticeable. Different fiber morphologies were observed and beading was less apparent.

E = .00 Snl A = MSE W= 4 Snl = Inn

igue 5 (h) 80/20-10wt% 11/25/03 22gauge.

Snl A = MSE M = . K Snl = Inn

igue 5 (i) 85/15-6wt% 12/01/03 22gauge.

5

igue 5 ( a (q ae eeseaios o e goe same e ae goe was use as a coecio suace o a ies o iee w a aios uig e eecosiig ocess e ies ae a oous mooogy ue o e eaoaio o soe om e suace (iceasig oe sie

0p E = .00 Snl A = MSE M = 2 W = 2 Snl = Inn

igue 5 ( goe same-mui 11/15/3

5

E = .00 Snl A = Inn W = Snl = Inn

igue 5 (s) 85/15-15wt% 12/17/03 22gauge.

a

a 1 = 13

1 = 171

9m

E = .00 Snl A = Inn M = 2.42 K W = Snl = Inn

igue 5 (t) 80/20-10wt% + BCNU 12/18/03 22gauge.

r .2 (u) 80/20-10wt% + BCNU 12/18/03 22gauge.

a 1 = 93 m

= 7E m

r .2 (v) 80/20-10wt% + BCNU 12/18/03 22gauge. 60

Using SEM analysis, the following fiber diameters were measured [Table 5.2].

The samples were also tagged by date to track the settling of the polymers in solution.

The average diameter was obtained by totaling the diameter measurements and dividing by the number of measurements taken.

bl .2 Nanofiber Diameter

Polymer sample Date Diameter measured (in microns) Average diameter description (in microns) 80/20 lOwt% 11/19/03 8.3, 4.4 6.35 20gauge Min = 4.4 Max = 8.3

80/20 6wt% 11/20/03 1.4, 0.4082, 0.2253, 0.7422, 1.4, 0.684 22gauge 0.1177, 0.2746 Min = 0.1177 Max = 4.4

80/20 lOwt% 11/25/03 4.4, 3.9, 2.7, 3.0, 2.4, 0.6026, 2.23 20gauge 0.9197, 0.3171, 1.8 Min = 0.3171 Max = 4.4 80/20 lOwt% 11/25/03 5.5, 3.4, 3.0, 5.8, 4.0, 7.6, 0.7298, 2.91 22gauge 0.9288, 1.3, 0.9288, 0.6634, 1.1 Min = 0.6634 Max = 5.8

85/15 6wt% 12/01/03 4.0, 3.5, 4.0, 5.1, 6.5, 5.1, 3.3 4.5 22gauge Min = 3.3 Max = 6.5 85/15 6wt% 12/01/03 3.7, 5.2, 4.9, 4.0, 3.4, 3.4, 4.3, 4.27 20gauge 4.3, 6.1, 4.6, 3.1 Min = 3.1 Max = 6.1

85/15 lOwt% 12/04/03 7.7, 4.4, 7.2, 6.6, 4.4, 4.6, 4.1, 5.76 22gauge 3.1, 5.0, 8.9, 5.0, 6.9, 3.6, 4.8, Min = 3.1 10.1 Max = 10.1 6

om ae 5 e /-1w a e 5/15-w ouce e es ies wiou cosieig ime as a aco Wi ime a aco i e ocess (e amou o

ime i akes o e oyme souio o see i oe o ouce sma iamee ies a same o /-w a a 5/15-I5w was mae e sames wee eecosu a osee ue e OM [ae 53]

bl . OM ie Oseaios

Date Polymer Wt% Needle size Description 12/09/03 80/20 8wt% 22gauge Droplets 12/09/03 85/15 15wt% 22gauge Droplets 12/15/03 80/20 8wt% 22gauge Droplets, encapsulation 12/15/03 85/15 15wt% 22gauge Fibers and beading 12/17/03 80/20 8wt% 22gauge Droplets and multiple fibers 12/17/03 85/15 15wt% 22gauge Good fiber formation, droplets, gel-like fibers 12/18/03 80/20 8wt% 22gauge Fibers with uniform diameter, droplets, gel-like 12/18/03 85/15 15wt% 22gauge Fibers and droplets

Ge-ike ies ook ike ies i a usae om ey ae eiie sae u ook as i e was o e ie cou coase a ge-ike susace (ig iscosiy wou ou ou Ecasuaio is we e eas o e ie seem o sow a ee-imesioa image wii e ea ise y ecasuaig e u-see oyme o souio e

5/I5-15w gae e mos cocee ie omaio i a o-woe see a SEM aaysis was couce o i e aeage ie iamee [ae 5] 62

bl .4 5/15-I5w ecoe iamee

Polymer sample Date Diameter measured (in microns) Average diameter (in description microns) 85/15 15wt% 12/17/03 5.955, 3.176, 5.161, 9.925, 7.146, 4.933 22gauge 3.1, 5.2, 2.352, 10.6, 2.823, 2.823, Min = 2.352 3.293, 4.2, 3.3 Max = 10.6

om e aaysis o a e ie iamees (aeage mi a ma e ie

omaio e /-Ow miue was comie wi e ug A 5/I5-w us

CU was mae u o esus (o ie omaio wee oaie Usig SEM aaysis o e / Ow us CU e oowig iamees [ae 55] wee measue a a aeage iamee o 77I micos was cacuae

bl . Measue iamee o /-Iw + CU

Polymer sample Date Diameter measured (in microns) Average diameter (in description microns) 80/20 1 Owt% 12/18/03 7.4, 4.666, 3.5, 2.625, 2.333, 1.5, 2.771 22gauge 4.666, 2.3, 1.458, 1.9, 1.750, 2.9, Min = 1.3 2.2, 3.421, 2.658, 3.7, 1.955, 3.1, Max = 7.4 2.932, 1.222, 1.3, 1.466

e mooogy o e /-1w us CU sysem was ey ieesig e

esig ooke ey simia o ke a a a iicae ee-imesioa sucue

[igue 5 ( (u]

Usig e GA aaysis a sames wi a wiou e ug sowe a mass

oss o ewee - [igue 53 (a (] Same / 1w OE CAAaaiGAkimgoe 1 Sie 55 mg TGA Oeao Kimey Meo am u ae 15-ec-3 19 Comme 11/5/3 gage Isume GA 5 5 ui 1

I 1 emeaue ("C Uiesa 3 7A A Isumes

Figure 5.3 a TGA of 10% wt DLPLGA nanofiber with — 3% wt mass loss. Same / Ow + ug ie C AaaGAkimgoe1 Sie 3 mg TGA Oeao Kimey MeoE am u ae 1-ec-3 155 Comme 1/1/3 gage ug = CU Ow i E Isume GA Q5 5 ui 1

9

"I" 3 5 7 do 1 emeaue CC Uiesa 37A A Isumes

Figure 5.3 b TGA. of 10% wt DLPLGA nanoftber with drug BCNU — 8% wt mass loss. 65

The DSC analysis [Figure 5.4 (a) — (d)] showed that the raw polymer and the electrospun polymer had the same Tg (glass transition temperature). A Tm (melting temperature) was not observed on the DSC graphs since the polymer is amorphous. The only change in Tg observed was subject to the presence of the drug in the polymer

[Figure 5.4 (e) and (f)]. Same 1 ie CAaaSC1kimgoe3 Sie 11 mg DSC Oeao Mais Meo ea/Coo/ea u ae 17-ec-3 173 Comme GA aw oyme Isume SC 1 7 ui

Same / Sie 11 mg Meo ea/Cooea Comme GA aw oyme is eaig cyce C e miue

-

-

-1

- 1

i 1 1 1 1 Eo U emeaue (°C Uies 7A A Isumes

Figure 5.4 a DSC of 80/20 DLPLGA polymer , First heat cycle. U;

2

Same /O Sie 11 mg Meo ea/Coo/ea Comme GA aw oyme

Seco eaig cyce C/mi

0 2

0 4

55C g -

15 -1 -5 5 loo Eo U emeaue (°C

Figure 5.4 b DSC of 80/20 DLPLGA polymer , Second heat cycle. Same / Ow ie CAAaaSCkimgoe Sie 5 mg SC Oeao Kimey Meo ea/Coo/ea u ae -ec-3 131 Comme aoie same gage a gage ae maeia 11/5/3 Isume SC 1 7 ui

Same / Ow Sie 5 mg Meo ea/Coo/ea Comme aoie same gage a gage ae maeia 11/3 is eaig cyce C/mi

1 -1 •

- --I

± 59 C g

1 1 - 1 - 1

Eo i emeaue (°C Uiesa 37A A Isumes

Figure 5.4 c DSC of 80/20 DLPLGA nanofiber , First heat cycle. 69

The DSC analysis of the nanofiber sample of 80/20-10wt% DLPLGA, second

heating cycle is seen in Figure 5.4 (d). The second heating cycle could not be isolated

due to computer programming errors. However, a Tg was still measured for comparison.

In the second heating run of the 80/20-10wt% polymer plus drug, there it is not clear as to where the actual Tg is, since due to the first heating, the drug had been degraded. The first heating cycle gives more definitive proof of the presence of the drug, before degradation [Figure 5.4 (e) and (0]. Same / 1w ie CAaaSCkimgoe Sie; 5 mg SC Oeao Kimey Meo ea/Coo/ea u ae -ec-3 131 Comme aoie same gage a gage ae maeia 11/5/3 Isume SC 1 7 ui

Same / Ow Sie; 5 mg Meo; ea/Coo/ea Comme aoie same gage a gage ae maeia 11/513 Seco eaig cyce C/mi •

71°C

• - 1 1 1 1 Eo emeaue (C Uiesa 37A A isumes Figure 5.4 d DSC of 80/20 DLPLGA nanofiber , Second heat cycle. Same / 1w + ug ie CAAaaSC‘kimgoe5 Sie 1 mg SC Oeao kimey Meo ea/Coo/ea u ae 1-ec-3 171 Comme GA / Ow ug = CU Isume SC Q1 7 ui 5

Same / Ow + ug Sie 1 mg Meo ea/Coo/ea Comme GA / Ow ug = CU is eaig cyce Cimi

U- Co Co -1 19C 17 9 19°C

1 1 1 1 emeaue (C Uiesa 37A A Isumes Figure 5.4 e .DSC of 80/20 DLPLGA nanofiber with drug BCNU , First heat cycle. Same / Ow + ug ie CAaaSCkimg oe 5 Sie 1 mg SC Oeao kimeiy Meo ea/Coo/ea u ae 1-ec-3 17 1 Comme GA / Ow ug = CU Isume SC 1 7 ui

C

U Same / 1w + ug Sie 1 mg Meo ea/Coo/ea Comme GA / Ow ug = CU Seco eaig cyce C/mi - -

t g

1

-1 - - - - Eo U emeaue (°C Uiesa 3 7A A isumes Figure 5.4 f DSC of 80/20 DL.PLGA nanofiber with drug BCNU, Second heat cycle. CAE

ISCUSSIO

A suy oe y e Scoo o iomeica Egieeig Sciece a ea Sysems

ee Uiesiy A GA was eecosu i E soe i oe o esig e iea scao o issue egieeig [] e ie iamee age was om 5-

aomees

Inrn pllr rn dtn

igue 1 Icease isace iamee eceases om 5um o 333m

e Cemica Egieeig eame a igiia ec oaie simia esus wee

e iamee age osee was om Sum o 333m [igue 1] e "os" osee

a e same mooogy as e goe same oaie i is eseac wee e

oyme ies seeme o ae ace o o oe aoe

om e SEM aaysis a e aeage iamees o sigiica cage i

iamee was osee A aeage iamee o 91 micos was cacuae o e

73 74

80/20-lOwt% alone and 2.771 microns for the 80/20 1 Owt% plus BCNU. The only difference observed was in the range of polymer diameter. The 80/20-10wt% alone had a min at 0.6634 microns and a max at 5.8 microns, where the 80/20-10wt% plus BCNU had a min at 1.3 microns and a max at 7.4 microns. The drug-polymer combination allowed for a wider range of polymer diameter, but not necessarily with smaller diameters (in the future, these numbers need to be repeated for consistency). While making the solution samples of 80/20-10wt%, a week was need to allow for entanglement of the polymer in order to electrospin fibers. The drug-polymer solution of the 80/20- lOwt% did not need as much time to settle (entangle) and spun out fibers within a few days. The presence of the drug may have affected the critical time at which entanglement occurred and needs further investigation.

Using TGA of the plain polymer fibers (no drug) over a period of time in comparison to that with the drug combined into it, no change in solvent dissipation or mass ratio was observed. Comparing the TGA of the 80/20-10wt% and the 80/20 lOwt% plus BCNU [Figure 5.3], no significant mass loss difference (difference of —4%) was observed. In both the SEM analysis and the TGA analysis, it can be said based on the observations that the presence of the drug did not affect diameter size or amount of mass loss. Using DSC analysis, the raw polymer was compared to the plain electrospun fibers and showed no change in the Tg (glass transition temperature). This provided proof that the electrospinning process does not affect the polymer's properties.

Also, in comparing the raw polymer with the 80/20 1 Owt% electrospun polymers in both the TGA and DSC, the electrospinning assay did not affect the polymer characteristics. In the DSC graphs [Figure 5.4 (a)-(d)], the 80/20 1 Owt% and the 80/20 75

aw oyme gas a e same g aou -53 °C as was eece a eoe y

Aei E Comaig e SC o e / Ow us CU o e /-Ow aoe ee was a sigiica cage i e g [igue 5 (c-(] e g was eoe a 19°C a 19°C as is cocue wi iigs ou a e eame o

amacy a amaceuica ecoogy (Mai Sai is cage i SC aowe

o e assumio a e ug was ese i e oyme

I a suy oe y e eame o amacy a amaceuica ecoogy

(Mai Sai usig CU ecasuae oy (D, -acie-cogycoie micosees

(aio - 15 ug/oyme aios e iamee o e micosees icease a e ecasuaio eiciecy ecease e CU coe o e micosees ecease

G ose o e oyme [3] As a eee eeicia caaceisic e eecosu

ug eiey sysem eeases ig coceaios o e ug i sma osage oms o

e age sie eeoe eceasig e sie eecs a/o oiciies a ae ossie we amiiseig CU Wi ski coac o e ug i e eecosu ie a aegic eacio occue eaig o e yoesis a e aoies ca suo e

ug ucioaiy muc oge a i oe ysica o cemica oms suc as I

ue aaysis o e amou o ug i e oyme e egaaio ae (eease ae o e ug a e ucioaiy o e ug ee o e eoe Aso e moecua aaysis o e ug a e oyme a e eecic ie (ui ysics eecs o e soe ee o e iesigae CHAPTER 7

CONCLUSION

Based in the data gathered, it is possible to obtain small diameter fibers by electrospinning using THF as a solvent. By combining the raw polymer, solvent, and drug in a simple mixing technique, the drug was able to reside in the polymer. This was evident when performing a DSC analysis and comparing the raw polymer to the electrospun fibers and the electrospun fibers with drug. Both the TGA analysis and DSC analysis, confirmed that the electrospinning process does not affect the polymer Tg or mass lost. The only characteristic that varied was in the morphology. The SEM analysis showed that the presence of the drug did not affect the diameter size of the fiber or the mass loss, but does affect the Tg.

Further analysis should be done to assess the amount of the drug present in the polymer, the functionality of the drug, shelf life of polymer drug delivery system, the release rate of the drug, and how to get the smaller fiber diameter. However diameter is dependent upon the problem, whether dealing with AVMs or tumors. If designing the system for tumors, the diameter should be under 8 microns, due to catheter diameter constraints. If the system's purpose is to plug up AVMs, then the diameter could be much larger, around I0-15 microns. The sample run for the electrospinning process involved the usage of MeCl, which provided fibers the day after mixing, whereas the

THF did not. Testing out different solvents may help in making small diameter fibers within an allotted time constraint. Other solvents to look at would be chloroform and ethyl acetate, both dissolve BCNU and PLGA. A different drug may be used in place of

BCNU (ex. tumor necrosis factor alpha, protein-tyrosine kinase inhibitors) and the effects

76 77 o e oyme caaceisics iesigae Aso ee was a aiey o ysica mooogy a was osee acig os kos a oes a ee se isucios o ow o maiuae a guaaee a ceai sucue APPENDIX A

CHARACTERISTICS OF RESPECTIVE RATIOS OF PLA AND PLGA

esecie aios coiue o mecaica ougess a seg ou i eie e cysaie o e semicysaie om

oyme Meig Gass as em Mouus ega (°C (°C (Gaa ime (mo GA 5-3 35- 7 -I A 173-17 -5 7 A Amoous 55- 19 I-1 GA-MC /A /A -1 5/15 Amoous 5-55 5- G 75/5 Amoous 5-55 -5 G 5/35 Amoous 5-5 3- G 5/5 Amoous 5-5 I- G a = esie o eua mouus = ime o comee mass oss ae aso ees o geomey [1]

78 APPENDIX B

BIOCOMPATIBILITY OF LACTIDE/GLYCOLIDE COPOLYMERS

Considerations on biocompatibility, the reaction of the patient to the material, and the effectiveness of the drug delivery system.

Lactide/Glycolide Copolymers: Review on Toxicity, Biocompatibility and

Clinical Applications (data up to 1999)

1 Three cyclic monomers - glycolide, lactide, and c-caprolactone

2 Biodegradation of the aliphatic polyesters occurs by bulk erosion. The

polylactide/polyglycolide polymer chains are cleaved by hydrolysis to form

monomeric acids and are eliminated from the body through the Krebs cycle, primarily

as carbon dioxide and water in urine. Because the rate of hydrolysis of the polymer

chain is dependent only on significant changes in temperature and pH or presence of

catalyst, a very little difference is observed in rate of degradation at different body

sites (Brandt et al. 1984, Cutright & Hunsuck I97I).

3 The role of enzymatic involvement in the biodegradation of the

polylactide/polyglycolide polymer has been somewhat controversial. Most early

literature conclude that bioerosion of these materials occurred strictly through

hydrolysis with no enzymatic involvement (Puleo et al. 1998). Other investigators

suggest that enzymes do ply a significant role in the breakdown of the breakdown of

the lactgide/glycolide materials (Fukuzaki et al. 1990, Reed I978, Williams I977-

79 80 19I Muc o is secuaio is ase uo e ieeces osee ewee i

io a i io egaaio aes i is suose a a ie eyme ioeme is

eece i e eay sages wi oymes i e gassy sae weeas eymes ca

ay a sigiica oe o oymes i e uey sae (oa e a 19

4 Icease yoiia a age suace aea a a ige ooio o moomes

omoe ioogica egaaio (akamua e a I99 e wae uake io e

oyme is aso iuece y e aio o cysaie o amoous egios i

geea amouous egios ae moe easiy aece y yoysis (Asao e a

199 e aio o cysaie o amoous egios i e oyme is eee

o o e ysioogica eiome a o e soiciomey o e moomes I

comaiso o -acic aci -acic aci iceases e ee o e amoous

egios (Egeeg & Ko I99I Woo 19 a cooymes om acic a

gycoic acis ae moe easiy egaae a ei esecie ue oymes (Caig

e a 1975

[19] APPENDIX C

MECHANICAL PROPERTIES OF MEDISORB® POLYMERS

Summaie mecaica oeies o oyme oaie om Akemes

Ar e Mt ariti Bolk MEW Alr frrrpfrrrrj Sn tht lvr immagomiimeiciiwis

GEEA MECA1CA OEIES

oymes 55 1 1 (Cusom 91 / (Cusom omuaio 5 oy -acie / 1 oy -acie 1 oy 1-acie 9 oy--acie / 5 gycoie 1 -acie eak sess si 9 1 733 71 sai a eak 5 5 55 5 Yie sess si 371 77 1 sai a yie 51 37 9 5 Mouus si 193 717 17 1

MECAICA OEIES O OY--ACIE / CAOACOE COOYMES

oyme - acie/caoacoe 5/5 75/5 5/15 9/1 95/5 Weig aio • esie Seg si

A ma 1 35 3 9 A 1 79 1 oe oe A 3 95 15 oe oe Eogaio o Yie 1 1 1 o aiue 1 5 1 1 Mouus Ksi 1 53 17 15 Soe -aess 5 5 7 91 95 Seciic gaiy 1 1 13 15 1 Comessio moig 73 — 13 13 +/- 15 1 +/- 1 15 +/- 5 15 +/- 5 emeaue °C

MECAICA OEIES O OY--AC1E / CAOACOE COOYMES

oyme - acie/caoacoe / 75/5 5/15 9/1 95/5 Weig aio esie Seg si

A ma 5 13 1555 53 593 A 1 5 1555 oe oe A 3 3 33 11 oe oe Eogaio o Yie - - - 5 - o aiue 5 5 7 Mouus Ksi 1 15 1 135 Soe -aess 79 95 Seciic gaiy - 1 1 1 - Comessio moig - 1 - 1 - 1 1 emeaue °C

Customer Servict Tel 513-489-0294 FM' S 1 1-9-77

81 lkenneS Enrl AScience that Delivers IMAMO8084 OYW8

SOUIIY CA

MEDISOR13® Ethyl Methylene Chloroform Acetone Dimethyl Tetrahydro- Hexafluoro- acetate chloride formamide furan (THF) Isopropanol (DMF) (HFIP) 100 L (Poly-L-Lactide) ns s s ns ns ns s 100 DL s s s s s s s 8515 DL s s s s s s s 7525 DL s s s s s s s 6535 DL s s s s s s s 5050 DL ss* s s ss* s ss* s Caprolactone (PCL) s s s s s s s 75:25 L-lactide/ PCL s s s s s s s 80:20 DL-lactide/ PCL s s s s s s s 100 PGA ns ns ns ns ns ns s

ns = not soluble ss * = slightly soluble (degree of solubility is dependent on molecular weight or IV) s = soluble

Medisorb® is not soluble in water, , methanol, silicon oil , ethyl ether and petroleum ether (heptane, hexane etc,).

Customer Service Tel 513-489-0294 nr 444 APPENDIX E

THERMAL ANALYSIS DATA OF MEDISORB® POLYMERS

Summarized thermal properties of polymer obtained from Alkermes.

Alr MEIS=6 = • Sn tht lvr MASOAE tr

MEISO® OYME g A MEIG OI AA

oyme omuaio Gass asiio em°C Meig oi °C

1 GA 1 oygycoic aci 35 - 5 - 3

1 1 oy--acie 5 - 173 - 17

91 G/ 9 gycoie / 1 I-acie 35 - 5 1 -

1 1 1-acie 5 - 55 Amoous

515 5 -acie / 15 gycoie 5 - 55 Amoous

755 75 -acie / 5 gycoie - 53 Amoous

535 5 -acie / 35 gycoie 5 - 5 Amoous

55 5 -acie / 5 gycoie 3 - Amoous

515 /C 5 -acie / 15 oycaoacoe - 5 Amoous

515 /C 5 -acie / 15 oycaoacoe - 5 Amoous

755 /C 75 -acie / 5 oycaoacoe 13 - Amoous

1 C 1 oycaoacoe (- - (-5

Amoous oymes ocess emeaue age 1 — 1 °C

Customer Service l 513-489-0294 Y 1 _9_7

83 APPENDIX F

CHEMICAL PROPERTIES OF BCNU

Summaie cemica oeies o ug CU

Poly(1,3-bis-p-carboxyphenoxypropane anhydride)

ABRAHAM 3. DOMB AND ROBERT LANGER

ACRONYMS, TRADE NAMES BIODELCPP, Poly(CPP), Poly(CPP.SA)

Polyarthydrides

snit ilY(709{: E 1 yoyci oc yei oyc I „ycoo

SAMOR A Pi:Ix:mom Biodegradable polymer fra criatmlled drug delivery in a form of implant or iiiiectible microspheres G bade! -BCNU-Loadtal wafer fur the treatment of brain tumors).

PIROPEitELS SPICEAL 1:sfrifiEST Anhydride copolymers of 1,3 6is-p- carboxyphenoxypropane (CPI') with aliphatic diacids such as sehocic acid (SA) degrade in a physiological medium to ('PP and SA. Matrices of the copolymers loadeti with ili.wilved or dispersed drugs degrade in vino and in vivo to constantly release. the dings for periods from I t.10 weeks.

P1101 UNITA CONTAITKINS .E.

Molecule weight P(CPP-SA)

10 g maul GPC-poly styrene M 3±:20, M U OX5±3 [5 standards dl g i Viscosity 258C, 0X210,9 3 dieltiorortierhane

114 (characteristic absiirrion ott NaCI pellet 1,750, 1,810 111 frequencies) PSA 1,740, 1,770, 1,810 P(CPP-SA) 1,712, 1,773 .P(CPP)

iaia Film on NaC1 pellet PSA 1,739, I,503 PcCPP-SA) 1,723, 1,765, 1,804 P(CPP) 1,712, 1,764

UV (characwristic nut P(CPP-SA), dichloromethane 265 abswptioo wavelength) CPP monomer, I N Na011 solution 265

I 1 •■•• I a 11 •I ; CI1oe

84

8

iio io amaook ai iy 119 15991y Oice iesiy ess oe igs esee

1A13-is--caoyeoyoai ayie

OA CON ios AIA EEEC

Souiiy mg m1 " e-SA (C•SA 701 (

msS C O moi CW

auoo 3 Ooae 3 1 --- -- ie iyiiiiai 3 1 Keoes 1 1 Ey aceae 1 1 Akaes a aeeS 1 1 gies 1 I Wae

Maiiiouwik C1ICI 3C K 3 (3 aums K a a m g oe a(5

;1 ea oeies CC•A SC 1C mi 1 7 ;5 IMO (3

K „ 359 339 5 511 K 3 331 3 7 39

k kg 11 Ai 157 13 119

Cysaiiy ( C SA owe -ay 1 17 5 1 (3 iacio 3 1 I

W 35 1 1

Comoie sequece (C-SA i-M CC1 9 7 591 951 (3 isiuio oaiiy u SA-SA 1 3 oaiiy o SA-C 1 3 7 9 Aeage ock eg WA 13 5 Ikgee o aomess 3 7 9 1

Saiiy i cooom souio (emase ia M (C-SA ( (ayie iecage eoywiaio 1 5 •; eoyeiaio 37C 135 1535 73 ae cosa Aciaio eegy kca mo 3 7 77

Eosio ae mg I P(CPP-SA). 14 A iX2 mm isc (5 1 I osae ue 1 7 9;51 1 i 7 37C SA 3 1 am 86

3 ayie aa iiuiook Coyig i by Oxford Univetsity Press, i A iss ee

oy(11-is--caoyeoyoige anhydridu)

OUY I$ CO-IIO5 AA KIM;

Erosion o mm ay -1 M osae ue ( 7 37C C-SA 1 s 5 (C-SA 55 11 E 1

Elimination in io e SA (7)

7 days in rat brain 95 21 days i rat brain 64 100

Drug retea.se in vitro ay 13(CI-SA (7 3 CU i isc 3 ( 5 ioeaci i isc 9

ug eease i io ay 3 CU isc imae i 1 (7 a ai

icomuiiiy Comaig wi uma wai ( Coiaie wi ai ai coea musce suseuae

Suie Guio aMisUiCai ic aimoe Maya USA

EE/ECE5

1- uo A A ei a- Saiociica Aci1 1 11991i 1335 133 oi„A a M Moa Sci 31 OM 1755 3o E ci a Macomocoies 9911 71$ Mii A ai1 K ige Maco/mks 1191391i 11711 I A a K IAs i a Aca Sci USA 91199 55 Coe A„ 17 Kwyas a aago i 1 Ma ioia 1119533 a7 1 ak e a iiaeas I 5 191 1ku A 3 S Aiseo ige a M Saia- I eige i Mg Iiomica1 Mye eie y S Saay Ca ause ei 199 i9 APPENDIX G

GMP ANALYSIS OF 75/25 DLPLGA POLYMER

iise ouc seciicaio o Meiso® oyme 75/5 GA

Alr

Crtft f Anl

Medisorb® 7525 DL High IV Finished Product Specification 1751001 t : 08042 t f Mnftr: 062800

t Spftn lt

Iee iscosiy — ig 7 ig

esiua Moome oa Maimum 1

-acie eo esu 15 317 gycoie eo esu 5

Cooyme Moe aio -acie 7 — moe 75 moe gycoie — 3 moe 5 moe

eay Meas 1 m oa Maimum 1 m

Gass asiio emeaue eo g 9° C

Moecua Weig Mw eo esu 113 K M eo esu 9 K oyisesiy eo esu 19

e ess a associae cieia seciie i is Ceiicae o Aaysis coom wi e iise ouc Seciicaios esaise y Akemes Cooe eaeuics I

e mauacuig quaiy coo a eiomea ecos associae wi e aoe eeece o o Meiso® 755 ig I ae ee eiewe a ou o e comee e Quaiy Coo esig o e o comies wi e iise ouc Seciicaio 17511 esaise y Akemes Cooe eaeuics e ac was ouce i comiace wi cue Goo Mnftrn acices as ey ay o e oucio o iise maeia o isiuio is o f maeia is eease o isiuio

Quaiy Assuace-- ___ Date: g "q-3° Meissa A. Gii Quaiy Assuace Associae

87 APPENDIX H

GMP ANALYSIS OF 85/15 DLPLGA POLYMER

Finished product specification of Medisorb® polymer 85/15 DLPLGA. Mermes

Crtft f Anl Mdrb® 8 h I S 800

t : W0602

t f Mnftr: 02280

t Spftn lt

Iee iscosiy - /g 7 /g

esiua Moomes oa maimum 1

-acie eo -acie 17

gycoie eo gycoie 5

Cooyme -acie - 9 moe 5

Cooyme gycoie 1 - moe 15

eay Meas 1 m oa maimum 1 m

Gass asiio eo g 5 C emeaue

Moecua Weig eo Mw 13 k

eo M k

eo oyisesiy 15

The manufacturing, quality control and environmental records associated with the above referenced lot of Medisorb® 8515.13L High IV have been reviewed and found to be complete. The Quality Control testing of the lot complies with the Finished Product Specification 1851001 established by Alkermes Controlled Therapeutics II. The batch was produced in compliance with currenr Good Manufacturing Practices they apply to the production of finished material for distribution. This lot of material is released for distribution,

Prepared by Date „L Kathy Miller QualityAssuace Associae

Reviewed by Date Melissa Maple Quality Assuace Associae

o W359- Page 1 of 1

88 APPENDIX I

NON-GMP ANALYSIS OF 80/20 DLPLGA POLYMER

Finished product specification of Medisorb® polymer 80/20 DLPLGA.

Alr

ANALYTICAL REPORT

Experimental Medisorb® 8020 DL High IV Batch No. 00-141-87 Date of Manufacture: 05/29/01

Description Result

Inherent Viscosity 0.62 dL/g

Reviewed by : _ _ Date :

89 0 REFERENCES

1 "Eecosiig" Department of Chemical Engineering, Virginia Tech., I

YM Si MM oma M ee GC uege "Eecosiig A wiig ui e geeaes sumico oyme ies" Applied Physics Letters, o 7 ume euay 1

3 MM oma M Si G uege M ee "Eecosiig a eecicay oce es II Aicaios" Physics of Fluids, o 13 ume Augus I

C ucko C Ce Y Se C Mai "ocessig a micosucua caaceiaio o oous iocomaie oei oyme i ims" Elsevier Polymer, 7397-77 I999

5 "Moeig eeas Seces o Eecosiig" MPC Industry Collegium Report, o 17 o 3 ue I

"Aeio eous Maomaios o e ai a Sie" The Cerebrovascular Center of the Neurological Institute, Columbia University, 1995

7 "ocaig AMs" Computer Graphics Group, Brown University.

A ice amua "Ceea AMs" Neurosurgery, Apollo Hospitals, Ceai Iia

9 A ime o ai umos Brain Tumor Foundation, ew Yok Ciy http://www.braintumorfoundation.org/

92

0.. I Musey S M Iim a K auke "eeome o Aiicia oo esses Seeig a oieaio Caaceisics o Eoeia a Smoo Musce Ces o ioegaae Memaes" Ann. .Y. Ad. S. 91 79-3 ( © ew Yok Acaemy o Scieces

11 E oo M; W Seuys M "oies i Ieeioa Caioogy (Ciica Caioogy ew oies" © 199 Ameica ea Associaio Ic Ocoe 199

1 ucs M A assei M a acai M "issue Egieeig A 1 s Ceuy Souio o Sugica ecosucio" Elvr Sn In. © 1 y e Sociey o oacic Sugeos

.. Ka "eeomes i Meica oymes o iomaeias Aicaios As maeias esiges coo e uamea caeges o meica sciece e iscoey o ew iocomaie oymes is ceaig ueceee ecieme" auay 1

1 oiso M Sai ouis A amaau "oyme i Meicie CE 35 Ioucio o oymes" prtnt f Chl Ennrn, Unvrt t fflSUY, Ai 1

15 W Geaso "ioasoae oymes" prtnt f dl Ennrn, Unvrt f Mnntt, 199

1 C Mieo A io "Syeic ioegaae oymes as Meica eices" M rhv, Mac 199

17 A ie "issoae Suues ow o Suues issoe?" Unvrt f nnlvn.

1 "oymeic ug eiey - A eiew" rdl.,

19 "Meiso® oyme oucs" Akemes Ic 3 94

ao-eas "oymes i Cooe ug eiey" oeme 1997

1 Acquaea "CU o ai umos" The Musella Foundation, 1993- 3

Guio amaceuicas "Giae® Wae" Physician's Desk Reference, 173-175 Coyig ©

3 AI oes-Suae u M E Gi-Aege A eaeu M A Camaco "Caaceiaio o ioegaae CU-oae micosees eae y e soe eaoaio-eacio meo" SciFinder ®, 3

W i C aueci E Caeso S ua K Ko "Eecosu aaoies sucue a oe scao o issue egieeig" Journal of Biomedical Materials Research, o Issue I3-I